A few updates in a pinned commend: - I solved the video capture issues but only after completing the video. There is a setting in the OSSC that remedied the darkening of the video.
Can you tell us what you had to tweak in OSSC? I was planning on using an OSSC as well, and I'd love to skip the "trial and error" phase. Also, I'm not very happy with my specific OSSC box. I'm not sure if it's a hardware problem, but I have noticed (with other sources) the left side of high-contrast images get a blurry edge on the left side, with the luminance getting darker to the right. I definitely do NOT see this problem on a VGA monitor. I'm hoping it's not a problem endemic to the OSSC overall.
@@tomwilson2112 I have not noticed such a thing on my OSSC yet ... can you tell me any particular systems & games where this behavior is more prononunced so I can check on my unit ?
Thank you for this in-detail tour and portfolio of this well designed 8-bit platform, Adrian. And please let us not be ashamed to honor your important role as one of the "fixer" guys for the Commander X16 mobsters community ... :P :) (meaning: based on your important design and usability feedback and your expertise with home computer electronics.) So thank you for the latter and the video!:) Now to my opinion to a sidenote, something unimportant: I think the Through Hole Technology approach gives this design a very appealing character. Watering this down to CHEAPNESS and SMD would be a sacrilege. But that is only my opinion. Of course this design has also to be available for a reasonable price ... so fighting reality with my opinion of "the art and love of old-scool'ness" isn't an option. Also having two (or more) designs is putting pressure onto the hardware designers and their lives (poor Kevin!). Well, its complicated, as this is (hopefully, hehehe) not the road to commercial success, but for old farts (with due respect) to show the younger generations how they could get reasonable (as in "reasoning") access to computer technology. I mean, we old farts where mostly confronted by accident back in the day with computers, some of us stayed ... learned, got in love and fell into that rabbit hole. The advantage for us (at that time ... maybe you can confirm that, Adrian?), which is pure coincidence of our birthdate, was that we as humans were still able to survey such systems in their ENTIREITY. In fact for the ECO-SYSTEM of the ATTINY "Systems-Processor" (Build tools, libraries, compiler, not even mentioning the production of the hardware, here) ... every little bit surrounding this MCU by a multi billion $ company (which gone through many hands, btw), it is too complex to print even an error free manual;) Also the optional approach, the expandability of a systems isn't a bad way to go. At least this will make me feel like back then ... dreaming about my own 1541, or amber monitor, without jealousy towards my class-mate ... Hehehe:) Thank you again for showing this off, Adrian. And the whole design folks and community around David who helped to birth this dream. Von Neumann, Zuse, Nixdorf, Ada and Chuck Peddle would be proud of you all!;)
I remember when David Murray went over it and I found myself disappointed with his ideas since all he basically wanted to do was make yet another Commodore clone more or less. Apparently he didn't follow through on that and some people are mad. My disappointment is he had an opportunity to make a cool board with the much more capable 65816 and do the same things with greater efficiency. All that RAM is hobbled by the fact that the 65C02 has to bank switch what a 65816 could access in its entirety without having to do such expensive operations.
@Adrian, thanks for the much needed technical breakdown. 8-bit guy does a wonderful job with his more simplistic overview but ur deeper dive is ALSO very much needed and appreciated. So thank you for this. No one else could have explained this better.
I would actually like to see this project sold as a DIY kit that you mostly assemble yourself. I love assembling electronic devices like this. I find it fun and relaxing and you get a much better idea on how everything comes together. Of course components like FPGAs and large surface mount packages and super tiny surface mounted components would be preinstalled but all the through hole components I would solder in myself. I love doing that kind stuff. I remember years ago my dad bought me a DIY transistor radio kit from RadioShack made by RadioShack. It was lots of fun and it worked. Unfortunately, it disintegrated due to a flood as the case was made out of cardboard.
@@wishusknight3009 Why have an all through hole device that isn't a kit. Also why aren't the plans Openhardware so that if somebody did want to make it themselves they could
It's very interesting that through-hole off-the-shelf parts which started out as a sweet spot for DIY is continuously shifting towards luxury pricing. Cars we call vintage today were once just old and cheaper than a new car. Once they get old enough they become rare and coveted. This appears to be also true for through-hole 80s-style computers.
I made a post the other day comparing GenX's love for vintage computers to be like our parents were about classic cars or mines case model trains. I got the mustang I never got as a teen but still looking for the Amigas that were lost to time/one of my mothers cleaning sprees. I kick myself for not buying them when they were cheap on ebay before the batteries killed so many. The electronics world is changing. In my line of work station owners expect microprocessor/fpga based new stuff to be repairable like the 80s audio gear my predecessors fixed. I can do some SMD work but most of it is not even a modestly equipped engineer can fix it anymore because it's all custom baked in house stuff that can't be off the shelf replaced anymore.
Millennials kinda have a similar thing as well with a bit later computers (I'm pretty sure the first computer I ever used was a 68K mac), and I guess zoomers have even later computers at least for a little bit, but later zoomers and uh, nyoomers, won't have anything of value. Phones that you can't repair with spyware and macrotransactions that ruin you don't count, sorry, I will defend this hill with my life and am willing to die on it. That's sad. I guess nostalgia (that doesn't hurt you) will be forgotten as a concept. I wonder if life is worth it or if we should mercy-destroy ourselves intentionally.
Adrian, you really know how to sell (the concept of) the Commander X16 and the Vera board. You definitely renewed my interest in the project and am sure your video will boost interest in the project in general. Well done!
So I loved the initial idea of the X16, but after seeing the cost creep up and movement towards FPGA, I'm starting to just not see the point. Why not just make it an open source MiSTer core at that point? The DE-10 Nano is pretty much the same cost (or cheaper) than the dedicated board... Maybe it's just me but I'm not really seeing the point anymore. Not to cast shade on the hard work that was put into it, it is a cool product, but one I'm not really interested in picking up any longer unfortunately
Agree... but different strokes for different folks I guess. I think it would have been really cool if the community could have coalesced around an inexpensive "modern" 8-bit retro device at a Raspberry Pi-like price but don't see that happening. The Agon Light seems to be the closest to what I was hoping for. Still will be interesting to see what comes from this, especially with the emulator being available.
I do think the price is higher than it's worth right now, but I still plan on getting one as soon as it comes out. This is a project I've been following since David's first video on it and I want to support them. If they don't make enough on the first wave, the whole thing is dead on arrival.
I don't know if you have been watching all the updates on this project but it has been announced that the boards like the one in this video are the Pro level ones for developers - as in people who are looking to plug in expansion cards and possibly design them. There is a cheaper version being produced for the average user so don't be too put off by the price on this particular board. Also, it is not exactly cheap geting a Mister system running as you need a couple of other expansion boards as well as the DE-10 Nano in order to run many system types.
The chip shortage seems to have not only affected the supply chain, but also I would imagine even design considerations. I keep looking to buy a Colour Maximite 2 or Mega65 and they are perpetually unavailable. But on the X16 it's sort of curious why they did use FPGA in key areas and then have a Yamaha chip that is no longer made.
Speaking as a grizzled veteran of many 1980s flamewars, it's clear that the Commander X16 and its competitors have succeeded in recapturing the spirit of 8-bit computing. I'm enjoying it - fights about the obvious superiority of your favorite system have always been part of the fun.
@@mopspear Yep. Never underestimate the eagerness of a bunch of people to pick a trivial attribute, tie their identity to it, and rally around it like their lives depend on it.
I never understood the 8-bit flamewars to be honest. Maybe because me and my friends with 8-bit computers realized there are already 16-bit Amiga 1000, Macintosh and 32-bit PC-AT and their graphic and sound were so much better. So any 8-bit flamewars was pointless for everyone who realized that any 8-bit was already dead at that time. 8-bit was just affordable fist step into home computing. However in any society there is always some percentage of primitives who enjoys such tribal habits as flamewars.
The Video Display Processor (VDP) used on the MSX 2 and MSX 2+ are actually off-the-shelf parts, descendants of the original TMS9918: The Yamaha V9938 and V9958. The problem with them is that, besides being old parts, they are quickly becoming VERY expensive. I bought a 9958 for my Omega MSX about 1 year ago for US$ 35, recently the same seller was asking US$ 60!
Damn! Perhaps we need a FPGA replacement, if possible.... so Omegas could still be built. I did look at those chips and they certainly seem to do a lot -- taking over a lot of functions of other circuits too. It means they have a high pin count meaning that a pretty fancy FPGA would be needed, not to mention a lot of level shifters since these FPGAs all run at something lower than 5v.
@@adriansdigitalbasement There is a drop-in FPGA replacement for the TMS9918 (called F18A), and of course V9958 cores for the One-Chip MSX (OCM) and related systems, but AFAIK no option that could be used on a little board as a drop-in replacement for the real thing. I have no idea how hard it would be do do that, either. When I built my Omega, I bought extras of almost all parts in the hope of building one of two extra boards to sell. But with the current prices of the V9958... and there's also the issue that, as a high value chip, it is VERY common to get fakes and a pain to get refunds later.
@@rigues isn't the F18A getting expensive due to its FPGA being end-of-life? There's a UA-cam video of an enthusiast who created a PCB adapter using a cheaper and more modern FPGA and then installing it in his NABU.
@@TheJeremyHolloway I don't know, I don't follow the project. The 9918 is easy to find and still plentiful, so I had no need to search for a replacement. Will try to find the video you talked about!
Thanks for this great video, Adrian! And for all your help in getting the prototype working. The X16 is a very exciting project and I can't wait to see what people will do with it once they have the hardware in their hands.
@@talon12020 As sceptical asxI am, I think there is enough interest for another revision. I'd be surprised if there wasn't at least a full FPGA version released, as that's probably the only way to make it truly affordable. Or perhaps a really pared down version for those that want one, but didn't like the idea of crowdfunding it. But the Agon Light is still the fly in the ointment.
I've been following this project from its inception. It's a noble effort but so many of their current problems were easy to foresee. When embarking on a project like this you have to decide what business you want to be in. Designing the system, writing the software, shipping the Gerber file off to Asia and getting back blank boards are all very doable and fun. But manufacturing? That's another thing all together. They have been populating these boards by hand and have been having great difficulty manufacturing even the 100 boards they need to get out to developers. When I saw they bought a wave soldering machine to speed this up I knew they were in trouble (they're warping the boards). They thought they could get thousands of Yamaha sound chips but they were lied to and now they have to source them individually at 10 times their original price. Many of those individually sourced Yamaha chips have turned out to be fakes. There's a reason that most electronics are built in Asia. Their economy (and labor costs) are very different than in the west. Also, when big companies release a computer they contract with vendors for millions of units which gives them very good prices. There are many decisions I feel they made that were well thought out. The ATX power connector, the Nintendo controller ports, etc. But the bad decisions were the difficult to obtain chips (Yamaha) but mostly the decision to incorporate the Vera plug in module which is very versatile but FPGA based board which defeats the purpose of an 80's retro machine. Why not make the whole thing based on a better FPGA (like the Mister). A project in development overseas is the R800, an FPGA based Atari 800 from a company called Revive Machines which right now is vaporware. But if it becomes real, I'm very interested. Depends how expensive it ends up being. I hope that the X16 is a success but the odds are against them. They have a scaled down version in development without expansion slots and they may have a big company that has expressed interest in manufacturing it. They call it the game system. They haven't said who that big company is. Just FYI, the current big version of the X16 is at least $500 and is referred to as the development system. I don't mean to be negative about this as it is obviously a labor of love but the days of a couple guys designing a computer system in their garage are long gone. Apple did it almost 50 years ago but even back then they needed an Angel Investor (Mike Markkula) to invest a small fortune to make it happen ($250,000 line of credit and about $90,000 of his own money, more than a million dollars in today's money). He also negotiated a 26% ownership of Apple. Mike is a billionaire today. Sorry this post is so long.
Only difference between them and apple is that the X16 isn't meant to beat any sort of competition or revolutionize anything. It's just a fun community project made by people interested in the hobby. While David is a bit over his head...which, well, we all kind of know that about him at this point, he is still capable of getting stuff done, as are the people he's working with. Given that they've even gotten this far is a good sign. Also, while I get people are frustrated by the FPGA for video, it seems like people are beating a dead horse at this point. Even they didn't want it to be an FPGA. And since you mentioned the MiSter, for all we know someone will release a core for it down the line. The system already has a complete emulator available.
David's reasons for choosing the Nintendo ports were that they were easily available in quantity, cheap and easy to interface with. Many viewers would have preferred Genesis ports but I can understand his thinking on this. Another smart decision was to make the MB standard ATX size so it could fit in almost any standard case.
Dip logic is a major selling point of this machine: if you want to shove the whole thing into an FPGA, then you could just use the emulator and be done with it; the vera board is analogous to a custom video chip in the olden days. You make a comparison with early apple, but, uh, technology moves forward?? the chips are dirt cheap, PCBs are dirt cheap, and they do have a significant amount of capital. Not a remotely reasonable comparison.
@@901aerol Many successful implementations of modern 8-bit computers have an FPGA in it. Look at ZX Spectrum Next and Mega65 as cases in point. Just incorporating an FPGA doesn’t invalidate the purpose of the system
@@skvader4187 Best I can describe it is like people who put money down on the cyber truck and ended up that the truck costs more and doesn't meet the promises it made
Oh wow - those expansion slots have the voltages mirrored in such a way that if you plugged the card in backwards it doesn't fry your whole system - NICE!
It does fry a Famicom cartridge, though. Instead of mirroring the voltages they should have keyed the socket so it was impossible to insert a cartridge backwards.
@@Throckmorton.Scribblemongerthey should have chosen a different connector. The fact that they are paying more than $5 per connector is insanity. Its one of the many pieces of evidence that nobody thought about the cost of this machine when designing it.
Geez guys... Look. Connectors cost money. That's just a universal truism. When you _actually build things_ (and don't just criticize other people's projects), you start to realize that a surprisingly large chunk of your product cost is "stuff" that isn't what you would think is the most vital. Connectors are definitely one of those things. Even pin headers aren't cheap. Especially if you need both male and female connectors. At least in this case, the sockets cost a bit of money, but making cards is basically free -- it's just edge connectors designed into the PCB. That, IMO, is a solid compromise. Edge connectors are probably also more robust than pin headers. It's a tried-and-true design. And yeah, it's not electrically benign with every card you can make fit into the slot. _Oh... well..._ Most things aren't. This isn't meant for mindless toddlers to go shoving retro carts into. It sidesteps what will most likely be the most common error, and that's already more generous than many bus connectors. Keying the connector would be great, but if you think you're paying too much for a commodity connector already, just wait until you see how much it costs to have the manufacturer create custom versions for you ... at least in the small production runs this thing will see. In summary... it's a DEV BOARD, for enthusiasts, to prove a design and test the waters to see whether it makes sense to continue cost-optimizing the design. I would say most of their choices were pretty reasonable in that regard. I question the use of the Yamaha chip here, but otherwise, it seems fairly well thought-out. Better than most people would do in their situation, at least.
@@nickwallette6201 In the beginning, they were supposed to be using no FPGAs and only non-obsolete chips. The X16 has both! If they were going to use FPGAs anyway, they should have just made the whole design FPGA-based. It's also a nightmare to assemble, and is far from affordable. It was *very* poorly thought out.
thank you for fighting for 240p! If I ever get my own Commander X16 I'll want to use it with my AppleColor RGB monitor that the IIgs and SNES are sharing atm!
Just another extra complication on top of all the other "little" complications that plague the X16. It's scope has changed so much since David's first video, nobody would recognise it. Feature creep is what gets projects like this in to so much trouble.
I've built a few WDC65C02S based computers. When you push the clock speed the components you use have to be able to deal with it. Those SRAM chips have a 55ns access time which is pretty slow for a 10MHz clock.. along with all of the latency introduced with the RAM select (and other logic) propagation delays.
@@deterdamel7380 Yeah, why not add modern, super fast DIMMs? Why not add an Nvidia RTX GPU too? And make it multi-core, each with FPU and 2MB cache per core. Oh wait... it's supposed to be an 8 bit 6502 based, retro computer. Speed is not the primary goal. 🙄
To clarify @ 5:02 Commodore introduced a two button joystick for compatible 8-bitters with the C64GS. And the Amiga line supports the 7 buttoned CD32 gamepad. Adrian probably just isn't aware because games tended to also be 1-button backwards compatible.
@@Dinnye01 why should you back it up to be impressed? most open source projects don't have backing and are usually something to be really impressed with. I think the issue with this product is that the core features and decisions were locked by what David wanted, which might not be what everyone wants.
This is such a great project, unfortunately I've lost the programming skills I once had (BBC B Basic at school & commodore 64 at home) These videos are good to watch, I really hope the trend of reviving and re-using older systems and technology continues. There's so much stuff out there waiting to be used so good luck to you all, go forth and create!
I became a computer user with a 486 DX2. I never went through the Commodore as a user. But this board? I might buy that. It's beautiful, and looks like a lot of fun.
I’d love to hear Adrian’s take on the AgonLight2 versus the Commander 16. It seems like both are trying fill the same niche but with different approaches. I could use an unbiased take of the pros/cons of each.
@@VincentGroenewold remember, some youtubers produce content for the likes/ego, others produce to monetize pushing a $600 8-bit computer tells me which side of the monetization certain channels are heading
I think they're both a bit disappointing --- both systems essentially have the 8-bit processor as a peripheral to a vastly more powerful modern CPU which does all the heavy lifting. I can see the necessity, as nobody makes ordinary video chipsets any more, but having the video chipset being capable of _emulating_ the primary CPU, faster than the CPU, somehow takes all the fun out of using an 8-bit CPU.
I like the X16, but it really deviated from the original idea of what David wanted when he first announced it.. He made to many compromises. This to me is a disappointment. But, not thinking about what it should have been, and what it became, its pretty awesome. There is a 8bit machine that is really close to what his original idea was, it is the AgonLight, alltho it is a Z80 system, it still fits in the model of what David originally define as his dream computer. On that note, since the AgonLight is available, I ordered one earlier this week to play with. I may get the X16 when it is easier to get, I just haven't decided yet.
The AgonLight isn't a hobbyist through-hole board design - not many people have the skills to build it from parts as everything is surface mount chips. And there are no expansion slots nor ability to have an "game module" slot as the X16 now accommodates. When the X16 was conceived, everyone wanted an old school through-hole board design as this maximized the ability to provide kits that could feasibly be completed by those of average soldering skills, and that would offer maximum flexibility for hobbyist tinkering (replacing chips with alternatives, etc). And that would have expansion slots for open-ended hardware hobbyist add-on board design. The AgonLight is great - and every retro geek should have one as is very affordable - but it doesn't scratch the same set of retro computing itches that the X16 does, that it originally set out as goals, and that have been met in the final design. And indeed, by now incorporating an official game cartridge standard, the X16 has exceeded the original set of project goals. The X16 project can't be blamed for the impact of global Covid pandemic, Chinese Covid lockdowns, Biden sanctions against China, The Ukraine war and Biden sanctions against Russia and the resulting super high monetary inflation that caused globally by that. Even the Raspberry Pi guys have been hit hard by these factors, and they are consummate veteran industry professionals. But to this day they're still not able to overcome impact of chip shortages as their main product units are still not widely available to the general public.
AgonLight is two 21st century microcontrollers connected over a serial channel, an SRAM, and a bunch of IO ports. That is nothing close to the original design of the X16. Just getting the one microcontroller the X16 does have was a fight but was ultimately necessary because it also acts as the keyboard controller (similar to what the original PS/2 had).
Q: So, if the expansion / cartridge slot was reduced to just one as in a cost-reduced version, and set at a right-angle so cards slot in parallel to the board, does this mean something similar to a expander cartridge on the C64 could restore the ability to add more than one card?
It's a parallel bus, so yes, that would work up until the length of signal traces and parasitic capacity of components starts to cause communication problems.
Yes. And since the expansion bus is designed from day one for 5 total I/O banks, you will have a far easier time running multiple expansion cards together. If the cards are built properly, with user selectable I/O ranges, you won't get the kinds of conflicts you do when trying to use multiple C64 expansion port devices together.
While using an ATTiny for keyboard, mouse and general system design is the most cost effective way of implementation, I would have loved this being a W65C134S (6502 based microcontroller)
If a version of the W65C134S with flash RAM program storage was available, it might have been a serious option, even if the ATTiny would likely be cheaper due to scale economies. At least the ATTiny is an 8bit MCU, similar to the ones used to upgrade from IBM-XT to the IBM-AT keyboards that were the direct predecessors of the PS/2 keyboards.
@@petermuller608 Yes. If that were Flash RAM rather than a masked ROM (READ-ONLY Memory), so that new firmware could be loaded onto it, it would be a serious candidate.
@@petermuller608 Yes, it's ROM ... not PROM, not EPROM, not Flash "ROM". "This MCU has an embedded debug monitor ROM with a library of routines that can help you reduce development time."
Hardware limits are what gave those old retro computers their unique "personalities". The varied ways that different manufacturers used to make the best of limited RAM, clock speed and functionality made each system a treasure trove for programmers to explore and take advantage of, unlike modern systems that all have the same capabilities and therefore all seem the same with no individual personalities. I look forward to a future version of the Commander X16 that heavily utilizes FPGA to vastly simplify and cost-reduce the overall design (and future-proof the motherboard because it would not rely on being able to source parts that are not made anymore). It would also mean that tweaks could be made to the functionality of the system.
Just to add more context. Tehtris is a game written in Prog8 language. That's a specialised language which compiles to 6502 assembly (it supports C64 too) by a member of the community DesertFish.
As someone who has been eagerly following news about the Commander X16, a retro-inspired computer, I have witnessed the great interest it has generated among retro computing enthusiasts. However, recent cost discussions on the 8-Bit Guy's channel have given me pause, making me realize that owning one might be beyond my reach. This realization has led me to reflect on the niche market for the X16. While it offers authenticity, learning opportunities, and a vibrant community, its price point may restrict ownership to a passionate subset of retro computing enthusiasts. Considering alternatives like emulation on existing hardware might be a more accessible and cost-effective way to experience 8-bit computing.
A lot has changed since that video. The have an assembly house in DFW area. The first batch has been assembled. Just population of DIP and testing. Cost of the 2nd run should be $350 including board, PS, mouse and keyboard. David mentions down in the comments.
@@njspencer79 I'll believe it when I see the receipts. That also doesn't address that fact that $60,000 of the startup costs were "donations", not crowdfunding, not pre-orders, literal monetary gifts. This is a for-profit business, and that money was used to buy fabrication and assembly equipment for Tex-elec, not getting boards built, so that is utterly unethical and I'm not even sure if it is legal. So regardless of the performance or design or even the horrendous cost, this project is so tainted by the absolute disregard for anyone but the X16 cabal that I wouldn't buy it even if I never had to buy another piece of retro hardware again. If he refunds all of those donations... Well, *I* still wouldn't buy it, but at least he wouldn't be a freaking crook.
@@doc_sav I have seen videos of them coming off the line. Are you arguing that is fake? Why would this video address any of that? it is a technical review nothing more. As for accusing 8bg of being a crook. Wow. Do you think it is wise to post such statements online? Let me guess we should all get an AgonLight right?
@@njspencer79 I have not doubt they are being made. I doubt the price point until units go on sale. I have not made any statement that doesn't come from 8bitguy's own videos, so I have absolutely no compunctions about being critical about them here or in any other format. As to the AgonLight, I already said specifically I am not even talking about the design or performance of the device itself here, just the business practices, so it is completely irrelevant to the discussion at hand.
@@doc_sav But in any event, the Dev Board will never be as cheap as the same system using SMB versions of most of the chips and integrating the Vera circuit onto the motherboard. The X16 console targeting the $150-$200 price point seems like the one you should be looking at.
Hey Adrian, fan of your channels (been subbed to your two main ones from my personal account for years). One minor semi-correction on the topic of sound... The YM2151 was typically paired with 16 or 32-bit systems - it's most often found in systems powered by a 68000 (Sega system16, x and y boards, Capcom CPS1, various Irem boards, the sharp x68000 and MANY more). Usually the CPUs in those systems run between 7 and 12 mhz. However... these systems also often include a z80 that runs between 3.5 and 4 mhz to specifically manage the sound. In some cases though, that second "sound" CPU doesn't actually do much of anything. I've been following the x16 from the very start. I think it's an interesting concept and enjoy seeing the videos on it and what it can do, but honestly it seems pretty impractical, especially from a cost standpoint. I know the goal has always been to reduce cost over time and shrink it down and I'd really be interested in it if it ever hits a price point of somewhere around $100. But I know we're a long way away from that. Anyway, thanks for all of the detailed videos on the x16, c64, apple II, amiga and everything else!
Not really that far from $100. There are FPGAs that have enough LUTs for the entire system. The VERA or the YM2151 ones are not the big. They are big enough to do the job. The reason the DIP one is getting the attention it is, has to do w/ the demand. A lot of folks said "$300-500 and it will have socketed DIPs. When can I get one?" Is that everyone? Nope. But it is plenty to justify the effort.
@@njspencer79 Also, there is a natural development path. If it can be done with the through hole parts and glue logic, it can be done in FPGA, but the reverse does not automatically work -- you can easily implement chip select logic in FPGA or a CPLD which will have to much gate delay if implemented with glue logic. So the X16 Dev Board is the hardware reference board, and the cost reduced "Surface mount ASIC" board for the X16 console and the X16 FPGA version can aim for compatibility with the Dev Board.
@@petermuller608 Yes, that's the thing. There are different demands out there. I can't pay an extra $150 for through hole parts, so I am waiting for the X16 console. But I want an expansion port, so the FPGA version is not for me.
Every time I hear "VERA", I think of the "Video Electronic Recording Apparatus", an early form of video recorder system that was used by the BBC that used razor-sharp thin metal "tape" at high speed (pre-dated spinning head VTRs by a few years), and produced quite a wobbly picture, and if the tape went wrong and came off the reels, it was apparently quite lethal, it was featured in a section of "The Secret Life of Machines" on the video recorder... :)
@@kargaroc386 Presumably the speed at which the metal tape was running would have been too much for wire to handle, given even audio recorders that used wire would snap too, something the size of the VERA unit would have been even more lethal if they had wire whipping about everywhere like cheesewire... :S
The cartridges are more than likely going to be how expansions are handled in the future, given the plan to revise the case with cartridge slots, with the idea being that slot 1 would be for a game and slots 2 to 4 would be like for ram, co-CPUs, drivers, memory mappers and GFX cards.
Sadly the cost of this blew up considerably and if the cost cutting continues, it will become a small board similar to raspberry pi or other mini computers, with little similarities to an old commodore with real chips.
Yeah, but what do you really want out of the board? Do you want a functioning computer or a retro art piece? If it's inside a case (which it should be) this really doesn't matter. I guess if it's something you want to be able to mod electronically it would impinge on your pleasure, but I don't know if that covers the majority use case.
@@Felice_Enellen Due to the use of TTL-logic there is no way to make proper modification the beautiful way. The ban of FPGA (or similar) in the CPU-board-design is a huge mistake. The project is more likely a holy religion driven paradigm.
@@deterdamel7380 Honestly I see hardware like this as a way to preserve methods before they're lost to neuron- and bit-rot. If you want to learn to write 6502 (or, better, 65816) there are plenty of emulators that give you all kinds of debug hooks and tracing without having to do some kind of serial debugging. I think it serves a purpose but I don't think it actually serves one for the average retro-interested programmer, more the retro-electronics-whizzes out there.
Amigas (at least AGA ones) support many more buttons on a controller (like the CD32 controller). Unfortunately very few games are coded to take advantage of it... also unfortunately, the CD32 controller is rather 'meh'.
Kinda like how criminally unsupported the Enhanced Joystick Ports were on the Atari STe. The Jaguar's JagPads - originally called the "ST Power Pads" - debuted for the STe first. Very few games supported them and mainly found some support on later Atari Falcon030 specific titles [still before the Jaguar's release].
@@TheJeremyHolloway Yeah, seems only the Atari made games ever utilized the Power Pad. I own one because they're cool, and look great on my Falcon / 1040STe.
What's really sad is that the stock Amiga joystick ports fully support 3 fire buttons... but back in the day every joystick vendor just cranked out "2600 compatible" controllers, so we poor Amiga saps still only got one fire button. It's extremely rare for any game to support more than one button, let alone the analog inputs.
@@Waccoon yeah, even the Atari 8bit computers were capable of more. For some weird reason home systems didn't really start getting more buttons until the NES / SMS / 7800 era. Classic chicken / egg thing, why develop for multi button controllers when no one was making multi button joysticks? Granted most arcade games had more than one button. I have always suspected there was some form of trying to prevent kids from just staying home and playing the games if they were as good as the arcade...
@@slaapliedje Just to be complete, I must point out that the Intellivision and ColecoVision had more than one button. Those consoles were before the NES/SMS generation.
Very interesting project, but it seems a solution in search of a problem. If you want authentic 80’s vibes, there are a lot of lovely Commodore and Atari machines out in the cold who would love to be saved and adopted. I really appreciate Adrian restoring and preserving computer history.
I think David mentioned this in one of his videos, but the problem with the old computers is the steep learning curve they have to being able to use them effectively. The X16 has a much simplified hardware programming model (similar to, but not copied from, the VIC-20). E.g. pixel addresses in a framebuffer can be trivially calculated on X16 and don't have to be in a weird pseudo-tile mode like C64. As many people writing X16 software have said - programming on the X16 is "fun".
@@Wavicle in addition, it’s a good entry point if you don’t have experience fixing older computers since it’s difficult to find reasonably priced working older computers.
I like the idea of it somehow (and have been following it since it started) but I can't think of a piece of electronics that I have less desire to buy/own. I don't think I'd even take one for free.
Nothing wrong with that. There's always the emulator. I'm a hardware man and I'm excited that there is this passion to create hardware with retro feel and repairability. Price may be an issue, but a neat homebrew like this is never going to be super cheap. If the software devs get behind it, that's what will sell it 😊
Ok, I loved that updated 8-bit dance party! I got a chuckle out of that. You've GOT to show that more often in your videos. At first, I had thought that perhaps it could be used occasionally as an outro (NOT always) for your videos, but then... I thought, why not consider using that as an outro for videos on the second channel? I think that would be pretty nifty.
As I get more and more into electronics, working with surface mount has become a necessity, though hasn't been too bad after I got a heat gun. I still haven't mastered BGA though.
That's kinda weird to use the Commander X16's PSG on the ported version of the NES Marble Madness [originally done by Tengen, aka the Atari Games Corp] on the Commander X16 when the original arcade version of [Atari Games Corp's] Marble Madness used a YM2151 for its audio. The same sound chip inside the Commander X16. Seems like it would've been better just to lift the arcade YM2151 audio code and patch the ported NES version to use that very same audio chip.
I think the audio capture was a little off, but something similar to that exact thing was done. It's a little memory expensive because it is playing back processed vgmrips which aren't exactly compact.
Not supporting PAL is a bit of a surprise considering that most of the world uses PAL or some compatible derivative, but I would imagine most people would be using the VGA port anyway so its not a deal breaker. Perhaps this would be something addressed in the future, I get that most of the development of this has been for/in the USA. Also many TVs that support NTSC and PAL are available here in Europe.
@@Throckmorton.Scribblemonger No, and I doubt the designers of this particular project are limiting their decisions just to what existed within the US. Rather.. 1) NTSC is not just US. It's a widely-deployed standard. Let's relax on the nationalism a little bit. 2) It's also very easy to downgrade from the world-wide VGA standard to NTSC -- just interlace it. Same resolution, same frame rate. 50Hz would require choosing VGA or video output, and that makes no sense at all. 3) PAL was widely deployed, but it was deployed in a lot of different and partially incompatible ways. 4) As mentioned above, many PAL TVs supported NTSC. Not so much the other way around. "Why" doesn't matter, nothing is going to retroactively modify a plethora of vintage TVs and monitors. So, it makes a ton of sense to NOT fragment the software market into stupid 50/60Hz versions with slightly different resolution. Pick one and stick with it. It's not that I'm set on the superiority of NTSC (it had its issues, clearly), but given #2 and #4, it makes the most sense if you had to choose one or the other.
it's been a little while since i used my mister, is there an x16 core available for it? seems like the perfect platform for it and i'd love to play around with it some
@Adrian's Digital Basement - Why did they choose to skip every other field to produce 240p? That halves brightness and reduces 240p 's refresh rate from 240p60 to 240p30. You're supposed to just keep pounding -electrons- pixels on the same field (even I think) even if the field number is conceptually odd. The standards docs for NTSC say that you can effect this by making sure the waveform of the burst on line 10 is always the same, instead of being inverted on alternate fields. Maybe pass this on to the Commander X16 guys?
@@tomwilson2112 I suppose it's possible that Adrian misspoke when he described what they did to make 240p work. I'm not some kind of video engineer but I've done a ton of work with NTSC video and I think it's fair to say it's easy to misunderstand how the different modes work.
I did the VERA changes for 240P support. It works by always drawing even fields and accomplishes this by drawing the entirety of the last line of the even field instead of cutting it short. It still renders at 60Hz, but has no interlace flicker.
@@Wavicle That's a bit of a hack but yeah it sounds like something most monitors would be able to deal with. Good enough, though not inverting the burst on line 10 in odd fields might be more correct if people start having trouble making it work in the field. Like, some displays might try to de-interlace it because the burst still implies that it's interlaced, which with simple de-flickerers might result in a vertical blurring effect. But whatever works.
@@Felice_Enellen do you have a link for this? I worked closely with Adrian to make sure everything was correct but do not recall seeing the inverted color burst mentioned anywhere. If that is expected and has been overlooked, I would like to correct it.
I honestly not sure who this is for. It kinda looks like a pc board but isn’t. Has quite a number of chips that could have been put onto a fpga, but isn't because IIRC correctly David didn't understand it. It has vga when most screens are like hdmi these days. I don't have a single screen that does VGA. It also seems to be expensive. One of the reasons people have got into the SBC stuff is because the boards are cheap, capable and work with almost everything off the shelf. I get that it supposed to be a modern boot into basic experience but honestly you could replicate that much more cheaply and have a software layer make it look like a 8/16 bit machine. I was kinda interested when he annouced the project. But after his video where he basically rebooted the project after like 8 months it is just has taken too long and there is too many odd things around it.
8 months? More like years. Basically for many it's about having a system you can still fully understand and control top to bottom. And play with. But even then there are better options.
@@bzuidgeest His first update video is where he basically said he kinda soft rebooted the project was about a 8 months to a year after he annouced it.
@@dave7244 yeah, but that is not really the start is it? And the last video they were talking about different directions again. And that sounds quite similar to soft reboot again
I think I just stopped watching him at some point and lost track of it. Looking at the (8bit guys) channel history and the dates of the videos it roughly lines up with what I remember. But I could be wrong. I am not too bothered either about finding out for sure.
I don't think you are IIRC correctly. Seems all the points you mention are explained pretty well. They could have not done any of this at all and just wrote an emulator that ran on an RPi and not worried about the hardware at all but... the hardware is the point.
That SID at the end sounds really weird, like it's missing some channels. :P So used to hearing your 8-bit dance party on the C64s in your videos. Thanks for the video, it makes many things about the Commander X16 much more clear.
It's good that this is a proper 8/16 bit system. Most of the 8 bit machines of early '80's had terrible graphics but this is just enough better to make it right. My benchmark is can it play the arcade games of the time convincingly.
But this is an 8 bit machine. If you wanted better graphics, sound and performance than typical 8 bit machines... build a 16 bit machine. Machines like the Amiga, Atari ST and Sharp X68000 already exist, so something like that gives you a more contemporary experience. Theoretically, you could create an 8 bit machine with a modern GPU grafted on... but it wouldn't be any better or more fun than existing machines.
@@another3997 No, you miss the point, some 8 bit machines had excellent graphics, BBC, Atari and the arcade machines. Yes they were not as good as the 16 bit consoles had but the point is that the graphics on my Dragon 32 and those on a Sinclair Spectrum were substandard and held the machine back as much as it's lack of disk drive. Modern retro computers that solve these problems in a 1980's way get my approval. The CoCoSDC provides a 1980 style disk drive using current tech for example.
14:19 one thing that i much like in these slot's design is GND separation between power lines and buses lines. yeah Apple, i am looking at you and your famous LCD connector :3
Do Adrian and David (8-Bit-Guy) know why so many home computer companies failed in the 80s? It's seems that the X16 project has also the goal to reproduce these failures and add some more on the top. I mean Adrian and the 8-Bit-Guy produces excelent content and should know about the history and misstakes in home computer market, but they don't take them into account.
The primary goal of the home computers in the 80s was to sell *lots* of computers. I don't think that's a primary goal, here. I think the goal here is to make some boards and have a small ecosystem of nostalgia fans and go off to have their fun. Not sure what the problem everyone is having with it. The simple solution for the people who don't see the point, don't like the price, etc. is to don't buy one. No one is forcing them to do that. After settling the mind on not buying one, then the next step is to not let its existence torment you. Easy-peasy.
@@shanehebert396 One of the primary goal of the X16 was a 50US$ 8-Bit computer. They did everything to miss the 50US$ price target and make the opposite of a smart design. I'm not a backer of the project, but it's interesting to see how the messed it up.
@@deterdamel7380 Technically that isnt true. The $50 was for their FPGA solution (Gen 3?) which wont be here for a while.What everyone really wanted was the dev board and the console version which are coming. For most following this, price has never been the issue. It was availability.
I wouldn't mind hearing your view on what made those companies fail back in the 1980s. There are a few factors that I can think of: component price volatility (dynamic RAM, in particular), fairly rapid evolution of the technology making barely new products uncompetitive, aggressive pricing strategies to drive competitors out of the market, changing tastes and disillusionment with home computing, economic downturns, and more. Only some of those factors are relevant now. You might argue that the X16 is not cheap enough, which is a fair point, but how does that correspond to the lessons of the 1980s? Companies like Commodore and Sinclair virtually bankrupted each other trying to shift machines at the low end of the market, worrying about undercutting, while companies like Apple floated along despite some pretty significant market failures, all thanks to much nicer margins, I suspect. And another factor from the 1980s that is usually overlooked by the average home computing enthusiast was that those companies looked to higher-revenue products in the longer term because they knew they couldn't survive at the low end forever. Quite a few companies went bankrupt before they could execute that particular strategy.
It's so hard to care about this project, in the face of trying to save as much authentic retro hardware as possible. It's cool and all, but maybe I'm just not old and/or nerdy enough to really appreciate it. Also, what happens when the chip supply runs out?
Adria, I know you're more a fixer than a builder but it would be absolutely fantastic to have a video of you creating an adapter for the vera for the apple // (or vic20) ! Will you pick up the gauntlet ? ;oD
I don't know, this thing seems like a whole MISS. If they wanted something that could be easy to program, easy to understand hardware, give something for hobbyists, tinkerers, and people to learn development by getting really close to the hardware... then having a FPGA video AND sound AND support chip setup would have been fine. Treat the FPGAs like the custom Commodore chips. Instead they went FPGA for video, have a bunch of other chips, and they foolishly thought there were 10,000 Yamaha chips still out there. AND NOW they're talking about making a gaming console version and a cost reduced version with surface mounted chips and little or no expansion cards. So now you have a game console with no games and a tinkerer/hobbyist computer that can't be tinkered with because it doesn't have slots and has hard to work on surface mounted components.
I do think, once FPGA graphics were decided, some of the other support hardware should’ve been reconsidered. Like you say, just treat it like the custom chips of yore. But of course the graphics were a reticent compromise so the rest stayed as-is. Integrating the FPGA deeper wouldn’t have had to get in the way of the low level hardware access, especially if each function still responded to the right memory addresses the right way.
This version of the X16 should have just existed as a demonstration piece to show its architecture in a discrete TTL form before moving on and implementing it in programmable logic devices for use as an actual product. Too many design revisions and bodges needed to occur after the design had started being committed to a PCB and no one should be interested in buying a prototype which still has design issues in spite of them.
@@tangentspace have you ever designed motherboard from the ground up? it's normal to have more than one revision before the final version, even the genesis was launched with bodge wires, you think few developers making it would not hav the same problems?
I think the new 8BDP should become an intro or outro for certain videos. It would be interesting to see it every so often. Glad to see more on the Commander X16.
I must congratulate David and associates for all their hard work in producing this capable 'modern' retro computer. I sincerely hope they are rewarded by a lot of sales and an ongoing active community for all their efforts. While regrettably this system won't fit any use case I have, I might grab a VERA to play with if sold separately - it seems similar in concept to the Ti9918a through YM9958 which are favourites of mine in DIY projects.
I'm interested in the VERA module since that would be really cool to get interfaced to other 8 bit systems. But the rest of it, like you, I just don't have a use for it.
I was truly indifferent because it's not my beloved Atari 800 - until you showed the 8 bit dance party. Now I'm like "Hmm, how to convince the wife I need this". Here's hoping they go GA soon.
It had some growing pains but the X16 really came together into one tight clean design, it was interesting how many fields of knowledge it took to get right as in all your work to fix the problems with the board.🙂 I adore the Vera module, such a great compromise for a lack of ideal vintage VDPs available, various AV outputs, as well as doing screen modes I wish vintage computers could do. 23:48 Thank you for being disappointed, I can't believe they left out 240p originally since that's nuts for such a platform.😉 The 480i maximum makes a lot of sense for an 8-bit computer as the era was spread between SDTV and PC monitor use and both share that common denominator of 480 scanlines. Shame about PAL but it's an awfully unique set of variables compared to NTSC in relation to modern screen specs. The color output is pretty sweet, very comparable to MSX2, SMS, CPC, and as you say nearly on par with a first generation Amiga. Same with the sound capability, very robust but also quite appropriate for the limits of an 8-bit computer, reminds me of the MSX a lot in sound expansion.
On a machine like this I would suggest that build cost is much less of an issue compared with component availability. I'm assuming this machine will work without the Yamaha devices installed? People who build this will be looking for the retro and/or DIY/learning experience. Moving to surface mount will destroy the reparability of the system for many as will adding FPGAs and the like. Most people don't have an FPGA programmer sitting around. If size and cost are a consideration then people could just stick to the emulator running on a tiny PC compatible. The ability to be able to poke around with a scope, design and build your own expansion boards, experiment and fix it when you've accidently let the magic smoke out are more critical for a system like this I would think.
If you keep the kynar wires out of the holes and below the lead knees, route them in the most direct possible x/y arrangement avoiding component footprints and leads, and stake them at corners and regular intervals, you can call them "jumper wires" instead of "bodges"; even if you're doing production assembly work. :D
i won't watch david's stuff - you're aware that many youtube viewers don't like him at all - but i'm glad to watch you review retro stuff. everyone likes adrian!
I have finished with David videos as well due to 1. His hardware incompetence (he should have stayed with software development this is a field he is good at), 2. Due to his altitude toward carrying guns in public. But I'll watch the Adrians videos about anything because he is just the opposite of David in both aspects. And he is funny too. 🙂
@@rastislavzima I do still watch some of the things, mostly software development or company history stuff, but I do always think of that video. Especially the “we’re not threatening anyone!” part filmed in (a) public business(es), when it _is_ a defacto threat. Which is the whole point: “don’t mess with me”. I’m thinking he meant “I’m not _presently planning to use it”,_ but those of course are rather different things. Especially when it’s ready to go on someone’s back or shoulder, rather than safely and securely holstered.
@@kaitlyn__L I had been unaware that the person in question (I won't use his name as I don't want to revictimize anyone reading) had been engaging in these ideologically-troubling activities - and thank you for bringing them to my attention.
@@eadweard. I mean, I wouldn’t go as far as to say names like “gun nut” or anything. He’s just a locally-centre-ground Texas open-carry advocate, yk? I’m mostly annoyed by those common arguments which conflate different things, to the point of denying that the goal is to make people (in general) more fearful of the carrier. Or focusing more on abstract rights rather than what they substantively enable. And full cards on the table my position is one of strict licensing with secure storage requirements. I grew up doing martial arts, I understand the desire for genuine self defence. But unlicensed open-carry guys take it way too far in my opinion, with increasingly stretched reasoning to justify excessive capabilities and making a big public display of it. That’s probably because martial arts drilled into me to use the minimum necessary response to get out of a situation. Which is even more of an argument for training and licensing!
The old controller ports were compatible with Sega genesis 6 button controllers from what I recall. You could have had legacy joystick and 6 button controller support
While I appreciate the extraordinary amount of passion that has gone into this system, I have never been able to get excited about it. From the beginning, I didn't like that the video solution used an FPGA, and I don't like that it has an ATTINY either. Either its retro or its not, and this is just not quite retro enough for me to get excited. The other problem is its been talking to long. I totally understand its a hobby project done in people's free time, but it's been years. I think to those who have been fully engaged in the project, its amazingly exciting, but I think for most people on the outside this isn't that interesting.
There are no other options other than FPGA for video. Every board out there, uses an FPGA for video. Nothing like the vintage video chips of old is produced these days. And you can't connect those old chips to a modern display . If you want any kind of large user group you have no other choice.
@@bzuidgeest Now the RP2040 can do it, but that wasn't an option while this was being designed. (you could probably use another RP2040 with level conversion to act as a 6502...)
Not really, as he mentions it runs a fixed 640x480 resolution which means each character would need to take approximately 4.85x11.2 pixels in that "grid" - and it would have to implement this by integer scaling scaling the fixed 8x16 pixel font. I suspect calling the result "unreadable" would be charitable! All the text-modes are oriented around this - basically toggling borders, double width and double height as needed, except the 80x60 which appears to run with all off and a 8x8 pixel font - I guess it's possible it's always an 8x8 font and even 80x30 is scaled but it doesn't look as rough as that should look. As he mention later this does mean the sides tends to get cut off in high-res mode on CRTs if the program doesn't use borders. I suspect that if anything could make them change how it works that would be it, but they seem fairly set into this, I don't see a fixed crystal so it MAY be a "small matter of programming the FPGA". But FPGA programming isn't easy.
@@Torbjorn.Lindgren yep the oscillator is fixed which is why doesn't have variable pixel clocks. It's all part of making the design as simple and inexpensive as possible. The font is actually 8x8 which is why 80x60 is actually the default font size. More text could be done with bitmap mode but it would be slow and very CPU intensive.
Ok I agree it was well thought out. I would think it should be capable of much better that c64 graphics but im skeptical how much traction this thing can get
I would buy one but... It uses a bunch of old parts that are getting harder and harder to find. And he even talks about the yamaha chip being hard to find. No fast forward 10 years, even 5 years from now when you start having chip failures... Is there gonna be a supply? Why do this when I can run the emulator just fine and not worry about part failure and have nothing at all.
@@kingforaday8725 People bought the C64 Maxi and Mini because it could run all the existing C64 games. This machine can not, anything needs to be written from scratch. That will drastically reduce impact.
Does it have a c compiler available? I like a lot of this system. It would be nice if you could use it with some flavor of 6809 as well. as the 6502. I hope that they took all the basic graphics commands and have the Vera implement them in the FPGA to make them super fast. Overall a very interesting system.
That flexible RAM/ROM mapping is very nice. The C64 would have been almost this great if they designed in a way that the cartridge port had the signals for mapping writable RAM into those 8/16K partitions (and the IO for sending mapping commands to the cartridge).
The C64 _does_ have those signals. • `R/W̅` is on pin 5 of the cartridge slot, which solves your "writable RAM" problem. _Anything_ in the cart can tell if the computer's trying to read or write it, and respond appropriately. • There are two specific address ranges, $DE00 and $DF00, with their own decode pins on the cartridge port, `I̅O̅1` and `I̅O̅2`. So that makes it dead easy to have something interpreting commands there. But that's just for ease of use and as a convention; the cartridge port sees _all_ writes the CPU does (except for addresses $00 and $01), so you can pretty much use any address you like to receive commands. • There are various modes for what the cart can take over (i.e., be guaranteed that internal devices will avoid responding to), but the broadest one is by asserting the `G̅A̅M̅E̅ ` line which maps 4K of C64 internal RAM to the bottom 4K of memory, and leaves _everything_ else to the cart.
I did not know that (or I knew it wrong!). So I can put a large SRAM on a cartridge and with some decoding logic I can choose which chunk I want to see in the 64k somewhere? Then why do we need the DMA-style ram expansion if we can just swap 8k pages in "zero" time?
@@Flashy7 I see two fairly large benefits to the REU way of doing things: 1. You can transfer arbitrary-sized units of memory from 1 to 64 KB, rather than being stuck with fixed page boundaries. So if you wanted to, say, transfer just one frame buffer worth of data (as is done by the "Globe" demo), you don't need to swap out a bunch of other data. This is particularly important when working near small areas of important stuff. For example, if your page size is 8 KB and you want to use different memory anywhere under address $2000, a bank switching system would also be switching out your zero page and stack which is almost invariably not what you'd want. 2. The REU leaves all existing address mapping working as it did before, so programs that depend on it need minimal modification, and programs unaware of it have a minimal chance of being broken by it, even if other things in memory (e.g. a TSR program) are simultaneously using it.
Yes. It is truly amazing. Also note that the CX16 has TWO banking areas for RAM. Between A000 and BFFF and between C000 and FFFF. When the memory expansion card is installed it can be loaded with extra RAM chips in the C000 and FFFF address space!!!!
@@svenvandevelde1 You can do roughly same on the C64, which has independent mappings for the following: • $E000-$FFFF: internal RAM, KERNAL ROM or cartridge (ROMH line) • $D000-$DFFF: internal RAM, CHAROM, or I/O+colour RAM • $C000-$CFFF: internal RAM or cartridge (requires decoding on cartridge) • $A000-$BFFF: internal RAM, BASIC ROM or cartridge (ROMH ine) • $8000-$AFFF: internal RAM or cartridge (ROML line) • $1000-$7FFF: internal RAM or cartridge (requires decoding on cartridge) And of course the C128 has even more powerful mappings.
A few updates in a pinned commend:
- I solved the video capture issues but only after completing the video. There is a setting in the OSSC that remedied the darkening of the video.
Can you tell us what you had to tweak in OSSC? I was planning on using an OSSC as well, and I'd love to skip the "trial and error" phase.
Also, I'm not very happy with my specific OSSC box. I'm not sure if it's a hardware problem, but I have noticed (with other sources) the left side of high-contrast images get a blurry edge on the left side, with the luminance getting darker to the right. I definitely do NOT see this problem on a VGA monitor.
I'm hoping it's not a problem endemic to the OSSC overall.
@@tomwilson2112 I have not noticed such a thing on my OSSC yet ... can you tell me any particular systems & games where this behavior is more prononunced so I can check on my unit ?
Thank you for this in-detail tour and portfolio of this well designed 8-bit platform, Adrian. And please let us not be ashamed to honor your important role as one of the "fixer" guys for the Commander X16 mobsters community ... :P :) (meaning: based on your important design and usability feedback and your expertise with home computer electronics.) So thank you for the latter and the video!:)
Now to my opinion to a sidenote, something unimportant: I think the Through Hole Technology approach gives this design a very appealing character. Watering this down to CHEAPNESS and SMD would be a sacrilege. But that is only my opinion. Of course this design has also to be available for a reasonable price ... so fighting reality with my opinion of "the art and love of old-scool'ness" isn't an option. Also having two (or more) designs is putting pressure onto the hardware designers and their lives (poor Kevin!).
Well, its complicated, as this is (hopefully, hehehe) not the road to commercial success, but for old farts (with due respect) to show the younger generations how they could get reasonable (as in "reasoning") access to computer technology. I mean, we old farts where mostly confronted by accident back in the day with computers, some of us stayed ... learned, got in love and fell into that rabbit hole. The advantage for us (at that time ... maybe you can confirm that, Adrian?), which is pure coincidence of our birthdate, was that we as humans were still able to survey such systems in their ENTIREITY. In fact for the ECO-SYSTEM of the ATTINY "Systems-Processor" (Build tools, libraries, compiler, not even mentioning the production of the hardware, here) ... every little bit surrounding this MCU by a multi billion $ company (which gone through many hands, btw), it is too complex to print even an error free manual;)
Also the optional approach, the expandability of a systems isn't a bad way to go. At least this will make me feel like back then ... dreaming about my own 1541, or amber monitor, without jealousy towards my class-mate ... Hehehe:)
Thank you again for showing this off, Adrian. And the whole design folks and community around David who helped to birth this dream. Von Neumann, Zuse, Nixdorf, Ada and Chuck Peddle would be proud of you all!;)
❤😂 oh now we are ready for a real
Moon Mission 🚀🪂😂
I remember when David Murray went over it and I found myself disappointed with his ideas since all he basically wanted to do was make yet another Commodore clone more or less. Apparently he didn't follow through on that and some people are mad.
My disappointment is he had an opportunity to make a cool board with the much more capable 65816 and do the same things with greater efficiency. All that RAM is hobbled by the fact that the 65C02 has to bank switch what a 65816 could access in its entirety without having to do such expensive operations.
@Adrian, thanks for the much needed technical breakdown. 8-bit guy does a wonderful job with his more simplistic overview but ur deeper dive is ALSO very much needed and appreciated. So thank you for this. No one else could have explained this better.
I would actually like to see this project sold as a DIY kit that you mostly assemble yourself. I love assembling electronic devices like this. I find it fun and relaxing and you get a much better idea on how everything comes together. Of course components like FPGAs and large surface mount packages and super tiny surface mounted components would be preinstalled but all the through hole components I would solder in myself. I love doing that kind stuff.
I remember years ago my dad bought me a DIY transistor radio kit from RadioShack made by RadioShack. It was lots of fun and it worked. Unfortunately, it disintegrated due to a flood as the case was made out of cardboard.
I agree. Not sure I'd be interested in a pre-built one.
This would be a pretty tough build for David's audience I think. I don't blame him for not wanting to do kits.
Dave already complains about how long it takes to box video games.
@@JG-nm9zk Dave complains about a lot of things.
@@wishusknight3009 Why have an all through hole device that isn't a kit. Also why aren't the plans Openhardware so that if somebody did want to make it themselves they could
Dave has said that the availability of the sound chip has become an issue. Seems most of the ones from Chinese suppliers are COUNTERFEIT.
And anyone with any experience would know that. It's why he was advised to use FPGA clones when he started. But he would not listen.
@@bzuidgeest So, on one side you have people salty about the FPGA graphics chip, and on the other side you have people salty about not using FPGA.
@@jonasthemovie very much so🤣.
@@bzuidgeest I wonder if the graphics chip at least has enough oomph left in it to run the FM sound as well
@@jonasthemovie Yeah for real people trying to have it both ways lol
It's very interesting that through-hole off-the-shelf parts which started out as a sweet spot for DIY is continuously shifting towards luxury pricing. Cars we call vintage today were once just old and cheaper than a new car. Once they get old enough they become rare and coveted.
This appears to be also true for through-hole 80s-style computers.
I made a post the other day comparing GenX's love for vintage computers to be like our parents were about classic cars or mines case model trains. I got the mustang I never got as a teen but still looking for the Amigas that were lost to time/one of my mothers cleaning sprees. I kick myself for not buying them when they were cheap on ebay before the batteries killed so many. The electronics world is changing. In my line of work station owners expect microprocessor/fpga based new stuff to be repairable like the 80s audio gear my predecessors fixed. I can do some SMD work but most of it is not even a modestly equipped engineer can fix it anymore because it's all custom baked in house stuff that can't be off the shelf replaced anymore.
There's lots of through-hole parts you can still buy brand new. And honestly, surface mount soldering is EASIER once you actually try it.
Millennials kinda have a similar thing as well with a bit later computers (I'm pretty sure the first computer I ever used was a 68K mac), and I guess zoomers have even later computers at least for a little bit, but later zoomers and uh, nyoomers, won't have anything of value. Phones that you can't repair with spyware and macrotransactions that ruin you don't count, sorry, I will defend this hill with my life and am willing to die on it.
That's sad. I guess nostalgia (that doesn't hurt you) will be forgotten as a concept. I wonder if life is worth it or if we should mercy-destroy ourselves intentionally.
Adrian, you really know how to sell (the concept of) the Commander X16 and the Vera board. You definitely renewed my interest in the project and am sure your video will boost interest in the project in general. Well done!
I'm happy to see that you were given a board, and mentioned ON the board since you pulled the X16 project out of the fire.
Okay, the music intro and then lead in to your own theme was awesome.
You’re NOT just another guy. Man you know sooooo much. Ty!
So I loved the initial idea of the X16, but after seeing the cost creep up and movement towards FPGA, I'm starting to just not see the point. Why not just make it an open source MiSTer core at that point? The DE-10 Nano is pretty much the same cost (or cheaper) than the dedicated board... Maybe it's just me but I'm not really seeing the point anymore. Not to cast shade on the hard work that was put into it, it is a cool product, but one I'm not really interested in picking up any longer unfortunately
I agree. I never really felt like the X16 filled any kind of niche. So, I never saw the point of it in the first place.
Agree... but different strokes for different folks I guess. I think it would have been really cool if the community could have coalesced around an inexpensive "modern" 8-bit retro device at a Raspberry Pi-like price but don't see that happening. The Agon Light seems to be the closest to what I was hoping for. Still will be interesting to see what comes from this, especially with the emulator being available.
I do think the price is higher than it's worth right now, but I still plan on getting one as soon as it comes out. This is a project I've been following since David's first video on it and I want to support them. If they don't make enough on the first wave, the whole thing is dead on arrival.
I don't know if you have been watching all the updates on this project but it has been announced that the boards like the one in this video are the Pro level ones for developers - as in people who are looking to plug in expansion cards and possibly design them. There is a cheaper version being produced for the average user so don't be too put off by the price on this particular board. Also, it is not exactly cheap geting a Mister system running as you need a couple of other expansion boards as well as the DE-10 Nano in order to run many system types.
The chip shortage seems to have not only affected the supply chain, but also I would imagine even design considerations. I keep looking to buy a Colour Maximite 2 or Mega65 and they are perpetually unavailable. But on the X16 it's sort of curious why they did use FPGA in key areas and then have a Yamaha chip that is no longer made.
45:18 - "Again, Ignore the darkness." - me to myself every morning.
Speaking as a grizzled veteran of many 1980s flamewars, it's clear that the Commander X16 and its competitors have succeeded in recapturing the spirit of 8-bit computing. I'm enjoying it - fights about the obvious superiority of your favorite system have always been part of the fun.
Or it could mean that a lot us in our 40s and 50s haven't really grown up since school. 😂
@@mopspear Yep. Never underestimate the eagerness of a bunch of people to pick a trivial attribute, tie their identity to it, and rally around it like their lives depend on it.
@@OzRetrocompyes my friend you hit the nail on the head. 😂
I never understood the 8-bit flamewars to be honest. Maybe because me and my friends with 8-bit computers realized there are already 16-bit Amiga 1000, Macintosh and 32-bit PC-AT and their graphic and sound were so much better. So any 8-bit flamewars was pointless for everyone who realized that any 8-bit was already dead at that time. 8-bit was just affordable fist step into home computing. However in any society there is always some percentage of primitives who enjoys such tribal habits as flamewars.
Did one 8-bit home computer cost 10x as much as the other one in the 80s?
57:04 The jumping head and the gummi bears, it's hillarious 😂
The Video Display Processor (VDP) used on the MSX 2 and MSX 2+ are actually off-the-shelf parts, descendants of the original TMS9918: The Yamaha V9938 and V9958.
The problem with them is that, besides being old parts, they are quickly becoming VERY expensive. I bought a 9958 for my Omega MSX about 1 year ago for US$ 35, recently the same seller was asking US$ 60!
Damn! Perhaps we need a FPGA replacement, if possible.... so Omegas could still be built.
I did look at those chips and they certainly seem to do a lot -- taking over a lot of functions of other circuits too. It means they have a high pin count meaning that a pretty fancy FPGA would be needed, not to mention a lot of level shifters since these FPGAs all run at something lower than 5v.
@@adriansdigitalbasement There is a drop-in FPGA replacement for the TMS9918 (called F18A), and of course V9958 cores for the One-Chip MSX (OCM) and related systems, but AFAIK no option that could be used on a little board as a drop-in replacement for the real thing. I have no idea how hard it would be do do that, either.
When I built my Omega, I bought extras of almost all parts in the hope of building one of two extra boards to sell. But with the current prices of the V9958... and there's also the issue that, as a high value chip, it is VERY common to get fakes and a pain to get refunds later.
@@rigues isn't the F18A getting expensive due to its FPGA being end-of-life? There's a UA-cam video of an enthusiast who created a PCB adapter using a cheaper and more modern FPGA and then installing it in his NABU.
@@TheJeremyHolloway I don't know, I don't follow the project. The 9918 is easy to find and still plentiful, so I had no need to search for a replacement. Will try to find the video you talked about!
@@TheJeremyHolloway You're referring to the tn9k_f18a. It seems like a good path forward for the F18A.
Thanks for this great video, Adrian! And for all your help in getting the prototype working. The X16 is a very exciting project and I can't wait to see what people will do with it once they have the hardware in their hands.
That's really going to depend on how much. I'm disabled and would love to have this since my C64 and 128 are now gone but i can only afford so much.
@@davidmiller9485 the early units will be more expensive, but the cost is going to go down a lot for future production runs and generations
@@slithymatt As if there will be any future production runs. The market for this thing mostly exists as backers of the Kickstarter.
@@talon12020 As sceptical asxI am, I think there is enough interest for another revision. I'd be surprised if there wasn't at least a full FPGA version released, as that's probably the only way to make it truly affordable. Or perhaps a really pared down version for those that want one, but didn't like the idea of crowdfunding it. But the Agon Light is still the fly in the ointment.
I just pre-ordered one of the next batch...
I've been following this project from its inception. It's a noble effort but so many of their current problems were easy to foresee. When embarking on a project like this you have to decide what business you want to be in. Designing the system, writing the software, shipping the Gerber file off to Asia and getting back blank boards are all very doable and fun. But manufacturing? That's another thing all together. They have been populating these boards by hand and have been having great difficulty manufacturing even the 100 boards they need to get out to developers. When I saw they bought a wave soldering machine to speed this up I knew they were in trouble (they're warping the boards).
They thought they could get thousands of Yamaha sound chips but they were lied to and now they have to source them individually at 10 times their original price. Many of those individually sourced Yamaha chips have turned out to be fakes.
There's a reason that most electronics are built in Asia. Their economy (and labor costs) are very different than in the west. Also, when big companies release a computer they contract with vendors for millions of units which gives them very good prices.
There are many decisions I feel they made that were well thought out. The ATX power connector, the Nintendo controller ports, etc. But the bad decisions were the difficult to obtain chips (Yamaha) but mostly the decision to incorporate the Vera plug in module which is very versatile but FPGA based board which defeats the purpose of an 80's retro machine. Why not make the whole thing based on a better FPGA (like the Mister).
A project in development overseas is the R800, an FPGA based Atari 800 from a company called Revive Machines which right now is vaporware. But if it becomes real, I'm very interested. Depends how expensive it ends up being.
I hope that the X16 is a success but the odds are against them. They have a scaled down version in development without expansion slots and they may have a big company that has expressed interest in manufacturing it. They call it the game system. They haven't said who that big company is.
Just FYI, the current big version of the X16 is at least $500 and is referred to as the development system.
I don't mean to be negative about this as it is obviously a labor of love but the days of a couple guys designing a computer system in their garage are long gone. Apple did it almost 50 years ago but even back then they needed an Angel Investor (Mike Markkula) to invest a small fortune to make it happen ($250,000 line of credit and about $90,000 of his own money, more than a million dollars in today's money). He also negotiated a 26% ownership of Apple. Mike is a billionaire today.
Sorry this post is so long.
Only difference between them and apple is that the X16 isn't meant to beat any sort of competition or revolutionize anything. It's just a fun community project made by people interested in the hobby. While David is a bit over his head...which, well, we all kind of know that about him at this point, he is still capable of getting stuff done, as are the people he's working with. Given that they've even gotten this far is a good sign.
Also, while I get people are frustrated by the FPGA for video, it seems like people are beating a dead horse at this point. Even they didn't want it to be an FPGA. And since you mentioned the MiSter, for all we know someone will release a core for it down the line. The system already has a complete emulator available.
The SNES controller ports were stupid too. But agreed, when you start adding fpga, you've lost the entire goal of the project.
David's reasons for choosing the Nintendo ports were that they were easily available in quantity, cheap and easy to interface with. Many viewers would have preferred Genesis ports but I can understand his thinking on this. Another smart decision was to make the MB standard ATX size so it could fit in almost any standard case.
Dip logic is a major selling point of this machine: if you want to shove the whole thing into an FPGA, then you could just use the emulator and be done with it; the vera board is analogous to a custom video chip in the olden days. You make a comparison with early apple, but, uh, technology moves forward?? the chips are dirt cheap, PCBs are dirt cheap, and they do have a significant amount of capital. Not a remotely reasonable comparison.
@@901aerol Many successful implementations of modern 8-bit computers have an FPGA in it. Look at ZX Spectrum Next and Mega65 as cases in point. Just incorporating an FPGA doesn’t invalidate the purpose of the system
The thing i love nost about the CommanderX16 is the free Community Drama DLC
Well deserved drama
Drama? Can someone explain?
@@skvader4187ignore it, he’s one of the nay-sayers
it’s a real shame. the cases were a total scam
@@skvader4187 Best I can describe it is like people who put money down on the cyber truck and ended up that the truck costs more and doesn't meet the promises it made
As someone that designs cloud computing systems all day long, I find this extremely refreshing 😂
Oh wow - those expansion slots have the voltages mirrored in such a way that if you plugged the card in backwards it doesn't fry your whole system - NICE!
It does fry a Famicom cartridge, though.
Instead of mirroring the voltages they should have keyed the socket so it was impossible to insert a cartridge backwards.
@@Throckmorton.Scribblemongerthey should have chosen a different connector.
The fact that they are paying more than $5 per connector is insanity. Its one of the many pieces of evidence that nobody thought about the cost of this machine when designing it.
@@tomcombe4813 The RC2014 just uses pin headers for its backplane and it has more expansion cards designed for it than the X16 ever will.
Geez guys... Look. Connectors cost money. That's just a universal truism. When you _actually build things_ (and don't just criticize other people's projects), you start to realize that a surprisingly large chunk of your product cost is "stuff" that isn't what you would think is the most vital. Connectors are definitely one of those things.
Even pin headers aren't cheap. Especially if you need both male and female connectors. At least in this case, the sockets cost a bit of money, but making cards is basically free -- it's just edge connectors designed into the PCB. That, IMO, is a solid compromise. Edge connectors are probably also more robust than pin headers. It's a tried-and-true design.
And yeah, it's not electrically benign with every card you can make fit into the slot. _Oh... well..._ Most things aren't. This isn't meant for mindless toddlers to go shoving retro carts into. It sidesteps what will most likely be the most common error, and that's already more generous than many bus connectors.
Keying the connector would be great, but if you think you're paying too much for a commodity connector already, just wait until you see how much it costs to have the manufacturer create custom versions for you ... at least in the small production runs this thing will see.
In summary... it's a DEV BOARD, for enthusiasts, to prove a design and test the waters to see whether it makes sense to continue cost-optimizing the design. I would say most of their choices were pretty reasonable in that regard. I question the use of the Yamaha chip here, but otherwise, it seems fairly well thought-out. Better than most people would do in their situation, at least.
@@nickwallette6201 In the beginning, they were supposed to be using no FPGAs and only non-obsolete chips. The X16 has both! If they were going to use FPGAs anyway, they should have just made the whole design FPGA-based. It's also a nightmare to assemble, and is far from affordable. It was *very* poorly thought out.
I can't wait to see how this computer progresses.
I don’t think it will because tech is already available. It’s not like in the beginning of the computer era where tech developed.
thank you for fighting for 240p! If I ever get my own Commander X16 I'll want to use it with my AppleColor RGB monitor that the IIgs and SNES are sharing atm!
Just another extra complication on top of all the other "little" complications that plague the X16. It's scope has changed so much since David's first video, nobody would recognise it. Feature creep is what gets projects like this in to so much trouble.
Yes 1h of Adrian showing cool stuff. But as always time flew by and 1h felt like 15min or so. Cool project!
I've built a few WDC65C02S based computers. When you push the clock speed the components you use have to be able to deal with it. Those SRAM chips have a 55ns access time which is pretty slow for a 10MHz clock.. along with all of the latency introduced with the RAM select (and other logic) propagation delays.
Yes, if you decide to use SRAM-chips in 600mil-DIP, you get slow and crappy SRAM. Seems to be a design goal. But the CX16 is soo great.. !?
Hence the 8MHz system clock on the Dev Board.
@@deterdamel7380 Yeah, why not add modern, super fast DIMMs? Why not add an Nvidia RTX GPU too? And make it multi-core, each with FPU and 2MB cache per core. Oh wait... it's supposed to be an 8 bit 6502 based, retro computer. Speed is not the primary goal. 🙄
@@another3997 No need to reply, if you don't know what SRAM is.
@@brucemcfarling6594 Yes, 8MHz, due to the selected fantom-soundchip.
Very, very cool! Thanks for the overview.
2:24 That project was a sub $50 computer using off-the-shelf parts.
Excellent video - very detailed and informative, with enough details for specialist and fast skip for enthusiast.
Thank you very much @Adrian!
If they think vga output is ancient, wait until they discover the cpu :D
The CPU came out in 1975...
@@michaelblair5566 The 65C02 came out in 1983...
@@Okurka. Well technically yes the 65C02 is 1983 but the 6502 was 1975.
@@Okurka. and its just as capable as the 1975 6502. 1976 if you discount the ROR instruction.
To clarify @ 5:02 Commodore introduced a two button joystick for compatible 8-bitters with the C64GS. And the Amiga line supports the 7 buttoned CD32 gamepad. Adrian probably just isn't aware because games tended to also be 1-button backwards compatible.
I'm excited to watch your take on it as I was pretty unimpressed with David's last video on it.
What did you expect? Did you back the project?
@@Dinnye01 why should you back it up to be impressed? most open source projects don't have backing and are usually something to be really impressed with. I think the issue with this product is that the core features and decisions were locked by what David wanted, which might not be what everyone wants.
@@cocusaryeah, well, that’s how projects devised by one person work. You can never provide what everyone wants, it’s just impossible
This is such a great project, unfortunately I've lost the programming skills I once had (BBC B Basic at school & commodore 64 at home) These videos are good to watch, I really hope the trend of reviving and re-using older systems and technology continues. There's so much stuff out there waiting to be used so good luck to you all, go forth and create!
I became a computer user with a 486 DX2. I never went through the Commodore as a user. But this board? I might buy that. It's beautiful, and looks like a lot of fun.
What you need is an emulator.
@@Okurka. Or a real C64 or some other retro computer. There were so many to choose from. But an emulator will give him an idea ofcwhat he missed.
I’d love to hear Adrian’s take on the AgonLight2 versus the Commander 16. It seems like both are trying fill the same niche but with different approaches. I could use an unbiased take of the pros/cons of each.
Or the f256 junior.
Unbiased may be a little thing as Adrian is printed on the board of the X16. :)
One costs $50 and the other $500.
One is available at multiple places, the other isn't.
@@VincentGroenewold remember, some youtubers produce content for the likes/ego, others produce to monetize
pushing a $600 8-bit computer tells me which side of the monetization certain channels are heading
I think they're both a bit disappointing --- both systems essentially have the 8-bit processor as a peripheral to a vastly more powerful modern CPU which does all the heavy lifting. I can see the necessity, as nobody makes ordinary video chipsets any more, but having the video chipset being capable of _emulating_ the primary CPU, faster than the CPU, somehow takes all the fun out of using an 8-bit CPU.
I like the X16, but it really deviated from the original idea of what David wanted when he first announced it.. He made to many compromises. This to me is a disappointment. But, not thinking about what it should have been, and what it became, its pretty awesome. There is a 8bit machine that is really close to what his original idea was, it is the AgonLight, alltho it is a Z80 system, it still fits in the model of what David originally define as his dream computer. On that note, since the AgonLight is available, I ordered one earlier this week to play with. I may get the X16 when it is easier to get, I just haven't decided yet.
There is also the f256 junior. It's half the x16 price and twice as capable.
What’s the FPGA status in those systems?
There will be a version 2 thats better. I don't blame David, I blame covid and the nightmare supply chain issue fall out from it.
The AgonLight isn't a hobbyist through-hole board design - not many people have the skills to build it from parts as everything is surface mount chips. And there are no expansion slots nor ability to have an "game module" slot as the X16 now accommodates. When the X16 was conceived, everyone wanted an old school through-hole board design as this maximized the ability to provide kits that could feasibly be completed by those of average soldering skills, and that would offer maximum flexibility for hobbyist tinkering (replacing chips with alternatives, etc). And that would have expansion slots for open-ended hardware hobbyist add-on board design.
The AgonLight is great - and every retro geek should have one as is very affordable - but it doesn't scratch the same set of retro computing itches that the X16 does, that it originally set out as goals, and that have been met in the final design. And indeed, by now incorporating an official game cartridge standard, the X16 has exceeded the original set of project goals.
The X16 project can't be blamed for the impact of global Covid pandemic, Chinese Covid lockdowns, Biden sanctions against China, The Ukraine war and Biden sanctions against Russia and the resulting super high monetary inflation that caused globally by that. Even the Raspberry Pi guys have been hit hard by these factors, and they are consummate veteran industry professionals. But to this day they're still not able to overcome impact of chip shortages as their main product units are still not widely available to the general public.
AgonLight is two 21st century microcontrollers connected over a serial channel, an SRAM, and a bunch of IO ports. That is nothing close to the original design of the X16. Just getting the one microcontroller the X16 does have was a fight but was ultimately necessary because it also acts as the keyboard controller (similar to what the original PS/2 had).
Now that was an extensive review and thanks for answering the questions 👍
Q: So, if the expansion / cartridge slot was reduced to just one as in a cost-reduced version, and set at a right-angle so cards slot in parallel to the board, does this mean something similar to a expander cartridge on the C64 could restore the ability to add more than one card?
It's a parallel bus, so yes, that would work up until the length of signal traces and parasitic capacity of components starts to cause communication problems.
Yes. And since the expansion bus is designed from day one for 5 total I/O banks, you will have a far easier time running multiple expansion cards together. If the cards are built properly, with user selectable I/O ranges, you won't get the kinds of conflicts you do when trying to use multiple C64 expansion port devices together.
Love your dance party, great demonstration of this system, thanks Adrian.
While using an ATTiny for keyboard, mouse and general system design is the most cost effective way of implementation, I would have loved this being a W65C134S (6502 based microcontroller)
If a version of the W65C134S with flash RAM program storage was available, it might have been a serious option, even if the ATTiny would likely be cheaper due to scale economies. At least the ATTiny is an 8bit MCU, similar to the ones used to upgrade from IBM-XT to the IBM-AT keyboards that were the direct predecessors of the PS/2 keyboards.
@@brucemcfarling6594 to me it looks like the WDC MCU has both ROM (4k) and RAM (192) on chip
@@petermuller608 Yes. If that were Flash RAM rather than a masked ROM (READ-ONLY Memory), so that new firmware could be loaded onto it, it would be a serious candidate.
@@brucemcfarling6594 oh is that really ROM as in One Time Programmable? Didn't realize, just assumed ROM as in Flash
@@petermuller608 Yes, it's ROM ... not PROM, not EPROM, not Flash "ROM". "This MCU has an embedded debug monitor ROM with a library of routines that can help you reduce development time."
Hardware limits are what gave those old retro computers their unique "personalities". The varied ways that different manufacturers used to make the best of limited RAM, clock speed and functionality made each system a treasure trove for programmers to explore and take advantage of, unlike modern systems that all have the same capabilities and therefore all seem the same with no individual personalities.
I look forward to a future version of the Commander X16 that heavily utilizes FPGA to vastly simplify and cost-reduce the overall design (and future-proof the motherboard because it would not rely on being able to source parts that are not made anymore). It would also mean that tweaks could be made to the functionality of the system.
Just to add more context. Tehtris is a game written in Prog8 language. That's a specialised language which compiles to 6502 assembly (it supports C64 too) by a member of the community DesertFish.
As someone who has been eagerly following news about the Commander X16, a retro-inspired computer, I have witnessed the great interest it has generated among retro computing enthusiasts. However, recent cost discussions on the 8-Bit Guy's channel have given me pause, making me realize that owning one might be beyond my reach. This realization has led me to reflect on the niche market for the X16. While it offers authenticity, learning opportunities, and a vibrant community, its price point may restrict ownership to a passionate subset of retro computing enthusiasts. Considering alternatives like emulation on existing hardware might be a more accessible and cost-effective way to experience 8-bit computing.
A lot has changed since that video. The have an assembly house in DFW area. The first batch has been assembled. Just population of DIP and testing. Cost of the 2nd run should be $350 including board, PS, mouse and keyboard. David mentions down in the comments.
@@njspencer79 I'll believe it when I see the receipts. That also doesn't address that fact that $60,000 of the startup costs were "donations", not crowdfunding, not pre-orders, literal monetary gifts. This is a for-profit business, and that money was used to buy fabrication and assembly equipment for Tex-elec, not getting boards built, so that is utterly unethical and I'm not even sure if it is legal. So regardless of the performance or design or even the horrendous cost, this project is so tainted by the absolute disregard for anyone but the X16 cabal that I wouldn't buy it even if I never had to buy another piece of retro hardware again. If he refunds all of those donations... Well, *I* still wouldn't buy it, but at least he wouldn't be a freaking crook.
@@doc_sav I have seen videos of them coming off the line. Are you arguing that is fake? Why would this video address any of that? it is a technical review nothing more. As for accusing 8bg of being a crook. Wow. Do you think it is wise to post such statements online? Let me guess we should all get an AgonLight right?
@@njspencer79 I have not doubt they are being made. I doubt the price point until units go on sale. I have not made any statement that doesn't come from 8bitguy's own videos, so I have absolutely no compunctions about being critical about them here or in any other format. As to the AgonLight, I already said specifically I am not even talking about the design or performance of the device itself here, just the business practices, so it is completely irrelevant to the discussion at hand.
@@doc_sav But in any event, the Dev Board will never be as cheap as the same system using SMB versions of most of the chips and integrating the Vera circuit onto the motherboard. The X16 console targeting the $150-$200 price point seems like the one you should be looking at.
Hey Adrian, fan of your channels (been subbed to your two main ones from my personal account for years). One minor semi-correction on the topic of sound...
The YM2151 was typically paired with 16 or 32-bit systems - it's most often found in systems powered by a 68000 (Sega system16, x and y boards, Capcom CPS1, various Irem boards, the sharp x68000 and MANY more). Usually the CPUs in those systems run between 7 and 12 mhz. However... these systems also often include a z80 that runs between 3.5 and 4 mhz to specifically manage the sound. In some cases though, that second "sound" CPU doesn't actually do much of anything.
I've been following the x16 from the very start. I think it's an interesting concept and enjoy seeing the videos on it and what it can do, but honestly it seems pretty impractical, especially from a cost standpoint. I know the goal has always been to reduce cost over time and shrink it down and I'd really be interested in it if it ever hits a price point of somewhere around $100. But I know we're a long way away from that.
Anyway, thanks for all of the detailed videos on the x16, c64, apple II, amiga and everything else!
Not really that far from $100. There are FPGAs that have enough LUTs for the entire system. The VERA or the YM2151 ones are not the big. They are big enough to do the job. The reason the DIP one is getting the attention it is, has to do w/ the demand. A lot of folks said "$300-500 and it will have socketed DIPs. When can I get one?" Is that everyone? Nope. But it is plenty to justify the effort.
@@njspencer79 I agree. Personally I'm only interested in a DIP version. FPGA is so close to software emulation, that I see nothing intriguing.
Thus
@@njspencer79 Also, there is a natural development path. If it can be done with the through hole parts and glue logic, it can be done in FPGA, but the reverse does not automatically work -- you can easily implement chip select logic in FPGA or a CPLD which will have to much gate delay if implemented with glue logic. So the X16 Dev Board is the hardware reference board, and the cost reduced "Surface mount ASIC" board for the X16 console and the X16 FPGA version can aim for compatibility with the Dev Board.
@@brucemcfarling6594 Yep the path being taken albeit a bit slower, is a path affords the possibility of a low cost FPGA version.
@@petermuller608 Yes, that's the thing. There are different demands out there. I can't pay an extra $150 for through hole parts, so I am waiting for the X16 console. But I want an expansion port, so the FPGA version is not for me.
This video was pure joy!
Every time I hear "VERA", I think of the "Video Electronic Recording Apparatus", an early form of video recorder system that was used by the BBC that used razor-sharp thin metal "tape" at high speed (pre-dated spinning head VTRs by a few years), and produced quite a wobbly picture, and if the tape went wrong and came off the reels, it was apparently quite lethal, it was featured in a section of "The Secret Life of Machines" on the video recorder... :)
I think of Jayne's rifle. ;)
Surprised they never tried using a wire - you could get a whole lot more thin wire onto a spool than a wide strip of tape.
@@kargaroc386 Presumably the speed at which the metal tape was running would have been too much for wire to handle, given even audio recorders that used wire would snap too, something the size of the VERA unit would have been even more lethal if they had wire whipping about everywhere like cheesewire... :S
Great video! Well worth the bandwidth. Makes me feel like it's 1983 again!
The cartridges are more than likely going to be how expansions are handled in the future, given the plan to revise the case with cartridge slots, with the idea being that slot 1 would be for a game and slots 2 to 4 would be like for ram, co-CPUs, drivers, memory mappers and GFX cards.
Sadly the cost of this blew up considerably and if the cost cutting continues, it will become a small board similar to raspberry pi or other mini computers, with little similarities to an old commodore with real chips.
I think it will look more like the A1200 with THT components largely replaced with SMT, but following the same design philosophy.
Yeah, but what do you really want out of the board? Do you want a functioning computer or a retro art piece? If it's inside a case (which it should be) this really doesn't matter. I guess if it's something you want to be able to mod electronically it would impinge on your pleasure, but I don't know if that covers the majority use case.
Thats pretty much exactly what I was thinking
@@Felice_Enellen Due to the use of TTL-logic there is no way to make proper modification the beautiful way. The ban of FPGA (or similar) in the CPU-board-design is a huge mistake. The project is more likely a holy religion driven paradigm.
@@deterdamel7380 Honestly I see hardware like this as a way to preserve methods before they're lost to neuron- and bit-rot. If you want to learn to write 6502 (or, better, 65816) there are plenty of emulators that give you all kinds of debug hooks and tracing without having to do some kind of serial debugging. I think it serves a purpose but I don't think it actually serves one for the average retro-interested programmer, more the retro-electronics-whizzes out there.
I trust Adrian here, it would be cool to program in 6502 and have new old games
That's what C64 Studio is for.
Exactly!
Amigas (at least AGA ones) support many more buttons on a controller (like the CD32 controller). Unfortunately very few games are coded to take advantage of it... also unfortunately, the CD32 controller is rather 'meh'.
Kinda like how criminally unsupported the Enhanced Joystick Ports were on the Atari STe. The Jaguar's JagPads - originally called the "ST Power Pads" - debuted for the STe first. Very few games supported them and mainly found some support on later Atari Falcon030 specific titles [still before the Jaguar's release].
@@TheJeremyHolloway Yeah, seems only the Atari made games ever utilized the Power Pad. I own one because they're cool, and look great on my Falcon / 1040STe.
What's really sad is that the stock Amiga joystick ports fully support 3 fire buttons... but back in the day every joystick vendor just cranked out "2600 compatible" controllers, so we poor Amiga saps still only got one fire button. It's extremely rare for any game to support more than one button, let alone the analog inputs.
@@Waccoon yeah, even the Atari 8bit computers were capable of more. For some weird reason home systems didn't really start getting more buttons until the NES / SMS / 7800 era. Classic chicken / egg thing, why develop for multi button controllers when no one was making multi button joysticks? Granted most arcade games had more than one button.
I have always suspected there was some form of trying to prevent kids from just staying home and playing the games if they were as good as the arcade...
@@slaapliedje Just to be complete, I must point out that the Intellivision and ColecoVision had more than one button. Those consoles were before the NES/SMS generation.
Very interesting project, but it seems a solution in search of a problem. If you want authentic 80’s vibes, there are a lot of lovely Commodore and Atari machines out in the cold who would love to be saved and adopted. I really appreciate Adrian restoring and preserving computer history.
And a lot of modern solutions to make using them easier.
I think David mentioned this in one of his videos, but the problem with the old computers is the steep learning curve they have to being able to use them effectively. The X16 has a much simplified hardware programming model (similar to, but not copied from, the VIC-20). E.g. pixel addresses in a framebuffer can be trivially calculated on X16 and don't have to be in a weird pseudo-tile mode like C64.
As many people writing X16 software have said - programming on the X16 is "fun".
@@Wavicle in addition, it’s a good entry point if you don’t have experience fixing older computers since it’s difficult to find reasonably priced working older computers.
@@Wavicle Going to be even more of a steep learning curve on a machine with no support.
No Atari's or Commodore's that's in economic range here in northern Scandinavia
I like the idea of it somehow (and have been following it since it started) but I can't think of a piece of electronics that I have less desire to buy/own. I don't think I'd even take one for free.
Nothing wrong with that. There's always the emulator. I'm a hardware man and I'm excited that there is this passion to create hardware with retro feel and repairability. Price may be an issue, but a neat homebrew like this is never going to be super cheap. If the software devs get behind it, that's what will sell it 😊
Same lol
Thinking of it as being a piece of 8-Bit Guy memorabilia instead of a useful computer and it makes more sense.
@@TheKnobCalledTone. Why would anyone spend $500 on something made by someone who is literally just some guy on UA-cam? I don’t understand that.
Ok, I loved that updated 8-bit dance party! I got a chuckle out of that. You've GOT to show that more often in your videos.
At first, I had thought that perhaps it could be used occasionally as an outro (NOT always) for your videos, but then... I thought, why not consider using that as an outro for videos on the second channel? I think that would be pretty nifty.
dunno about that updated 8 bit dance party ... I'm not feeling it ... the drums have no punch ...
The gummy bears were a nice touch! 😆
I'm so jelly! Looking forward to enjoying this Video:)
Thanks for the detailed review. I can understand your excitement about the system. It looks great! Looking forward to seeing the production units.
Great intro tune selection! 4mat is a genius
As I get more and more into electronics, working with surface mount has become a necessity, though hasn't been too bad after I got a heat gun.
I still haven't mastered BGA though.
Indeed -- I still do prefer to stick with through hole, but that's mainly because I like using stuff from the 70s and 80s :-)
You should have been a college professor, Adrian. The way you explain detailed topics in an understandable way is top-notch.
That's kinda weird to use the Commander X16's PSG on the ported version of the NES Marble Madness [originally done by Tengen, aka the Atari Games Corp] on the Commander X16 when the original arcade version of [Atari Games Corp's] Marble Madness used a YM2151 for its audio. The same sound chip inside the Commander X16. Seems like it would've been better just to lift the arcade YM2151 audio code and patch the ported NES version to use that very same audio chip.
Agreed. I gottsa fig the porter just wasnt aware.
I think the audio capture was a little off, but something similar to that exact thing was done. It's a little memory expensive because it is playing back processed vgmrips which aren't exactly compact.
I have no idea what this is, or what it does, but it's neat watching you explain it.
Not supporting PAL is a bit of a surprise considering that most of the world uses PAL or some compatible derivative, but I would imagine most people would be using the VGA port anyway so its not a deal breaker. Perhaps this would be something addressed in the future, I get that most of the development of this has been for/in the USA. Also many TVs that support NTSC and PAL are available here in Europe.
But the entire world revolves around 'Merica.
@@Throckmorton.Scribblemonger I would gladly buy the retro PC built after some 40-years old PC from your country. Are there many options? No? How so?
@@rkurbatov You mean the AgonLight?
Shouldn't you be fighting for your country instead of defending the US? Or are you one of those deserters?
@@Throckmorton.Scribblemonger No, and I doubt the designers of this particular project are limiting their decisions just to what existed within the US. Rather..
1) NTSC is not just US. It's a widely-deployed standard. Let's relax on the nationalism a little bit.
2) It's also very easy to downgrade from the world-wide VGA standard to NTSC -- just interlace it. Same resolution, same frame rate. 50Hz would require choosing VGA or video output, and that makes no sense at all.
3) PAL was widely deployed, but it was deployed in a lot of different and partially incompatible ways.
4) As mentioned above, many PAL TVs supported NTSC. Not so much the other way around. "Why" doesn't matter, nothing is going to retroactively modify a plethora of vintage TVs and monitors.
So, it makes a ton of sense to NOT fragment the software market into stupid 50/60Hz versions with slightly different resolution. Pick one and stick with it. It's not that I'm set on the superiority of NTSC (it had its issues, clearly), but given #2 and #4, it makes the most sense if you had to choose one or the other.
@@nickwallette6201 Never Twice the Same Color
it's been a little while since i used my mister, is there an x16 core available for it? seems like the perfect platform for it and i'd love to play around with it some
There’s a software emulator for it but no FPGA core
@Adrian's Digital Basement - Why did they choose to skip every other field to produce 240p? That halves brightness and reduces 240p 's refresh rate from 240p60 to 240p30. You're supposed to just keep pounding -electrons- pixels on the same field (even I think) even if the field number is conceptually odd. The standards docs for NTSC say that you can effect this by making sure the waveform of the burst on line 10 is always the same, instead of being inverted on alternate fields. Maybe pass this on to the Commander X16 guys?
That doesn't sound right. The 240P renderer is true 240P60. This might be related to the incorrect setting Adrian said he was using on his OSSC.
@@tomwilson2112 I suppose it's possible that Adrian misspoke when he described what they did to make 240p work. I'm not some kind of video engineer but I've done a ton of work with NTSC video and I think it's fair to say it's easy to misunderstand how the different modes work.
I did the VERA changes for 240P support. It works by always drawing even fields and accomplishes this by drawing the entirety of the last line of the even field instead of cutting it short. It still renders at 60Hz, but has no interlace flicker.
@@Wavicle That's a bit of a hack but yeah it sounds like something most monitors would be able to deal with. Good enough, though not inverting the burst on line 10 in odd fields might be more correct if people start having trouble making it work in the field. Like, some displays might try to de-interlace it because the burst still implies that it's interlaced, which with simple de-flickerers might result in a vertical blurring effect. But whatever works.
@@Felice_Enellen do you have a link for this? I worked closely with Adrian to make sure everything was correct but do not recall seeing the inverted color burst mentioned anywhere. If that is expected and has been overlooked, I would like to correct it.
I honestly not sure who this is for.
It kinda looks like a pc board but isn’t. Has quite a number of chips that could have been put onto a fpga, but isn't because IIRC correctly David didn't understand it.
It has vga when most screens are like hdmi these days. I don't have a single screen that does VGA. It also seems to be expensive.
One of the reasons people have got into the SBC stuff is because the boards are cheap, capable and work with almost everything off the shelf. I get that it supposed to be a modern boot into basic experience but honestly you could replicate that much more cheaply and have a software layer make it look like a 8/16 bit machine.
I was kinda interested when he annouced the project. But after his video where he basically rebooted the project after like 8 months it is just has taken too long and there is too many odd things around it.
8 months? More like years.
Basically for many it's about having a system you can still fully understand and control top to bottom. And play with. But even then there are better options.
@@bzuidgeest His first update video is where he basically said he kinda soft rebooted the project was about a 8 months to a year after he annouced it.
@@dave7244 yeah, but that is not really the start is it? And the last video they were talking about different directions again. And that sounds quite similar to soft reboot again
I think I just stopped watching him at some point and lost track of it. Looking at the (8bit guys) channel history and the dates of the videos it roughly lines up with what I remember. But I could be wrong. I am not too bothered either about finding out for sure.
I don't think you are IIRC correctly. Seems all the points you mention are explained pretty well. They could have not done any of this at all and just wrote an emulator that ran on an RPi and not worried about the hardware at all but... the hardware is the point.
That SID at the end sounds really weird, like it's missing some channels. :P
So used to hearing your 8-bit dance party on the C64s in your videos.
Thanks for the video, it makes many things about the Commander X16 much more clear.
Can't wait until David comes out with the emulator!!
github.com/X16Community/x16-emulator/releases/tag/r43
Awesome! Thanks Adrian!
thank you very much Adrian for this exploration into the commander x16! Looking forward for the final product and if they ship to europe, too 🙂
2:11 heh.. somebody is "excited" to get his hands on the X16 😜🔥
🤣🤣🤣
It's good that this is a proper 8/16 bit system. Most of the 8 bit machines of early '80's had terrible graphics but this is just enough better to make it right. My benchmark is can it play the arcade games of the time convincingly.
But this is an 8 bit machine. If you wanted better graphics, sound and performance than typical 8 bit machines... build a 16 bit machine. Machines like the Amiga, Atari ST and Sharp X68000 already exist, so something like that gives you a more contemporary experience. Theoretically, you could create an 8 bit machine with a modern GPU grafted on... but it wouldn't be any better or more fun than existing machines.
@@another3997 No, you miss the point, some 8 bit machines had excellent graphics, BBC, Atari and the arcade machines. Yes they were not as good as the 16 bit consoles had but the point is that the graphics on my Dragon 32 and those on a Sinclair Spectrum were substandard and held the machine back as much as it's lack of disk drive.
Modern retro computers that solve these problems in a 1980's way get my approval. The CoCoSDC provides a 1980 style disk drive using current tech for example.
Adrian where can I buy the t-shirt you are modeling today?
AH unfortunately, it was a viewer donation and I don't recall where to get them.
Oh no!
14:19 one thing that i much like in these slot's design is GND separation between power lines and buses lines. yeah Apple, i am looking at you and your famous LCD connector :3
Do Adrian and David (8-Bit-Guy) know why so many home computer companies failed in the 80s? It's seems that the X16 project has also the goal to reproduce these failures and add some more on the top.
I mean Adrian and the 8-Bit-Guy produces excelent content and should know about the history and misstakes in home computer market, but they don't take them into account.
The primary goal of the home computers in the 80s was to sell *lots* of computers. I don't think that's a primary goal, here. I think the goal here is to make some boards and have a small ecosystem of nostalgia fans and go off to have their fun. Not sure what the problem everyone is having with it. The simple solution for the people who don't see the point, don't like the price, etc. is to don't buy one. No one is forcing them to do that. After settling the mind on not buying one, then the next step is to not let its existence torment you. Easy-peasy.
@@shanehebert396 One of the primary goal of the X16 was a 50US$ 8-Bit computer. They did everything to miss the 50US$ price target and make the opposite of a smart design. I'm not a backer of the project, but it's interesting to see how the messed it up.
@@deterdamel7380 Technically that isnt true. The $50 was for their FPGA solution (Gen 3?) which wont be here for a while.What everyone really wanted was the dev board and the console version which are coming. For most following this, price has never been the issue. It was availability.
@@deterdamel7380 They are on the way of repeating yet another CBM mistake - fragmenting the ecosystem due to varying features within an architecture.
I wouldn't mind hearing your view on what made those companies fail back in the 1980s. There are a few factors that I can think of: component price volatility (dynamic RAM, in particular), fairly rapid evolution of the technology making barely new products uncompetitive, aggressive pricing strategies to drive competitors out of the market, changing tastes and disillusionment with home computing, economic downturns, and more.
Only some of those factors are relevant now. You might argue that the X16 is not cheap enough, which is a fair point, but how does that correspond to the lessons of the 1980s? Companies like Commodore and Sinclair virtually bankrupted each other trying to shift machines at the low end of the market, worrying about undercutting, while companies like Apple floated along despite some pretty significant market failures, all thanks to much nicer margins, I suspect.
And another factor from the 1980s that is usually overlooked by the average home computing enthusiast was that those companies looked to higher-revenue products in the longer term because they knew they couldn't survive at the low end forever. Quite a few companies went bankrupt before they could execute that particular strategy.
It's so hard to care about this project, in the face of trying to save as much authentic retro hardware as possible. It's cool and all, but maybe I'm just not old and/or nerdy enough to really appreciate it. Also, what happens when the chip supply runs out?
8 bit dance party... oh yeah. ❤
Where do I find a VGA monitor these days? Prolly need to stick an HDMI port on the VERA.. oK, NVM. Finished the video..
Adria, I know you're more a fixer than a builder but it would be absolutely fantastic to have a video of you creating an adapter for the vera for the apple // (or vic20) !
Will you pick up the gauntlet ? ;oD
56:24 - And the image isn't distorted like it was on the other systems!
I don't know, this thing seems like a whole MISS. If they wanted something that could be easy to program, easy to understand hardware, give something for hobbyists, tinkerers, and people to learn development by getting really close to the hardware... then having a FPGA video AND sound AND support chip setup would have been fine. Treat the FPGAs like the custom Commodore chips. Instead they went FPGA for video, have a bunch of other chips, and they foolishly thought there were 10,000 Yamaha chips still out there. AND NOW they're talking about making a gaming console version and a cost reduced version with surface mounted chips and little or no expansion cards. So now you have a game console with no games and a tinkerer/hobbyist computer that can't be tinkered with because it doesn't have slots and has hard to work on surface mounted components.
I think that's the lower cost version they have in planning.
I do think, once FPGA graphics were decided, some of the other support hardware should’ve been reconsidered. Like you say, just treat it like the custom chips of yore. But of course the graphics were a reticent compromise so the rest stayed as-is. Integrating the FPGA deeper wouldn’t have had to get in the way of the low level hardware access, especially if each function still responded to the right memory addresses the right way.
This version of the X16 should have just existed as a demonstration piece to show its architecture in a discrete TTL form before moving on and implementing it in programmable logic devices for use as an actual product. Too many design revisions and bodges needed to occur after the design had started being committed to a PCB and no one should be interested in buying a prototype which still has design issues in spite of them.
@@tangentspace have you ever designed motherboard from the ground up? it's normal to have more than one revision before the final version, even the genesis was launched with bodge wires, you think few developers making it would not hav the same problems?
you're being geerous to think they will sell 10,000 of these, it's just a proof of concept
Awesome intro mod from 4mat 😍 (frameskool by equinox)
I think the new 8BDP should become an intro or outro for certain videos. It would be interesting to see it every so often. Glad to see more on the Commander X16.
I'm so glad they fixed the mouse pointer being hidden in R43. Looking good!
I must congratulate David and associates for all their hard work in producing this capable 'modern' retro computer. I sincerely hope they are rewarded by a lot of sales and an ongoing active community for all their efforts.
While regrettably this system won't fit any use case I have, I might grab a VERA to play with if sold separately - it seems similar in concept to the Ti9918a through YM9958 which are favourites of mine in DIY projects.
I'm interested in the VERA module since that would be really cool to get interfaced to other 8 bit systems. But the rest of it, like you, I just don't have a use for it.
Why does this intrigue me more than any new game in the last 10 years...?
The 65C02 combined with the PSG synth and the extremely powerful graphics of the VERA reminds me a lot of the PC Engine/Turbografx 16
Yeap, this seems like a SuperGrafx on extreme steroids. I just discovered this Commander X16 project today and I'm very, very excited.
There is no PSG present on the X16
@@tangentspace There is a PSG on the VERA
I was truly indifferent because it's not my beloved Atari 800 - until you showed the 8 bit dance party. Now I'm like "Hmm, how to convince the wife I need this". Here's hoping they go GA soon.
It had some growing pains but the X16 really came together into one tight clean design, it was interesting how many fields of knowledge it took to get right as in all your work to fix the problems with the board.🙂
I adore the Vera module, such a great compromise for a lack of ideal vintage VDPs available, various AV outputs, as well as doing screen modes I wish vintage computers could do. 23:48 Thank you for being disappointed, I can't believe they left out 240p originally since that's nuts for such a platform.😉
The 480i maximum makes a lot of sense for an 8-bit computer as the era was spread between SDTV and PC monitor use and both share that common denominator of 480 scanlines. Shame about PAL but it's an awfully unique set of variables compared to NTSC in relation to modern screen specs.
The color output is pretty sweet, very comparable to MSX2, SMS, CPC, and as you say nearly on par with a first generation Amiga. Same with the sound capability, very robust but also quite appropriate for the limits of an 8-bit computer, reminds me of the MSX a lot in sound expansion.
On a machine like this I would suggest that build cost is much less of an issue compared with component availability. I'm assuming this machine will work without the Yamaha devices installed? People who build this will be looking for the retro and/or DIY/learning experience. Moving to surface mount will destroy the reparability of the system for many as will adding FPGAs and the like. Most people don't have an FPGA programmer sitting around. If size and cost are a consideration then people could just stick to the emulator running on a tiny PC compatible. The ability to be able to poke around with a scope, design and build your own expansion boards, experiment and fix it when you've accidently let the magic smoke out are more critical for a system like this I would think.
I want one, but I'll wait for the price to come down by a lot first. A bit too much money to invest in a toy at this point.
If you keep the kynar wires out of the holes and below the lead knees, route them in the most direct possible x/y arrangement avoiding component footprints and leads, and stake them at corners and regular intervals, you can call them "jumper wires" instead of "bodges"; even if you're doing production assembly work. :D
i won't watch david's stuff - you're aware that many youtube viewers don't like him at all - but i'm glad to watch you review retro stuff. everyone likes adrian!
I have finished with David videos as well due to 1. His hardware incompetence (he should have stayed with software development this is a field he is good at), 2. Due to his altitude toward carrying guns in public.
But I'll watch the Adrians videos about anything because he is just the opposite of David in both aspects. And he is funny too. 🙂
@@rastislavzima I do still watch some of the things, mostly software development or company history stuff, but I do always think of that video.
Especially the “we’re not threatening anyone!” part filmed in (a) public business(es), when it _is_ a defacto threat. Which is the whole point: “don’t mess with me”. I’m thinking he meant “I’m not _presently planning to use it”,_ but those of course are rather different things. Especially when it’s ready to go on someone’s back or shoulder, rather than safely and securely holstered.
@@kaitlyn__L I had been unaware that the person in question (I won't use his name as I don't want to revictimize anyone reading) had been engaging in these ideologically-troubling activities - and thank you for bringing them to my attention.
@@rastislavzima You prefer criminals to be the only civilians to carry weapons?
@@eadweard. I mean, I wouldn’t go as far as to say names like “gun nut” or anything. He’s just a locally-centre-ground Texas open-carry advocate, yk? I’m mostly annoyed by those common arguments which conflate different things, to the point of denying that the goal is to make people (in general) more fearful of the carrier. Or focusing more on abstract rights rather than what they substantively enable.
And full cards on the table my position is one of strict licensing with secure storage requirements. I grew up doing martial arts, I understand the desire for genuine self defence. But unlicensed open-carry guys take it way too far in my opinion, with increasingly stretched reasoning to justify excessive capabilities and making a big public display of it. That’s probably because martial arts drilled into me to use the minimum necessary response to get out of a situation. Which is even more of an argument for training and licensing!
The old controller ports were compatible with Sega genesis 6 button controllers from what I recall. You could have had legacy joystick and 6 button controller support
Ahhh the X16, the only retro computer that orders a subway armed to the teeth and terrifies the employees.
Nah, that's just Texas in general.
Judging on how the "optional memory" went with zx spectrum next, everyone wants the maximum amount.
While I appreciate the extraordinary amount of passion that has gone into this system, I have never been able to get excited about it. From the beginning, I didn't like that the video solution used an FPGA, and I don't like that it has an ATTINY either. Either its retro or its not, and this is just not quite retro enough for me to get excited. The other problem is its been talking to long. I totally understand its a hobby project done in people's free time, but it's been years. I think to those who have been fully engaged in the project, its amazingly exciting, but I think for most people on the outside this isn't that interesting.
There are no other options other than FPGA for video. Every board out there, uses an FPGA for video. Nothing like the vintage video chips of old is produced these days. And you can't connect those old chips to a modern display .
If you want any kind of large user group you have no other choice.
So, how would you have solved HDMI support?
@@bzuidgeest Now the RP2040 can do it, but that wasn't an option while this was being designed. (you could probably use another RP2040 with level conversion to act as a 6502...)
Would adding a 132 x 43 text mode be possible? That was my prefered resolution in MS-DOS for programming and worked well 😀
Not really, as he mentions it runs a fixed 640x480 resolution which means each character would need to take approximately 4.85x11.2 pixels in that "grid" - and it would have to implement this by integer scaling scaling the fixed 8x16 pixel font.
I suspect calling the result "unreadable" would be charitable! All the text-modes are oriented around this - basically toggling borders, double width and double height as needed, except the 80x60 which appears to run with all off and a 8x8 pixel font - I guess it's possible it's always an 8x8 font and even 80x30 is scaled but it doesn't look as rough as that should look.
As he mention later this does mean the sides tends to get cut off in high-res mode on CRTs if the program doesn't use borders. I suspect that if anything could make them change how it works that would be it, but they seem fairly set into this, I don't see a fixed crystal so it MAY be a "small matter of programming the FPGA". But FPGA programming isn't easy.
@@Torbjorn.Lindgren yep the oscillator is fixed which is why doesn't have variable pixel clocks. It's all part of making the design as simple and inexpensive as possible. The font is actually 8x8 which is why 80x60 is actually the default font size. More text could be done with bitmap mode but it would be slow and very CPU intensive.
Ok I agree it was well thought out. I would think it should be capable of much better that c64 graphics but im skeptical how much traction this thing can get
Hardcore Commodore fan boys!! The same folks that bought the C64 mini and maxi.
I bought both and will buy this just to have it!
I would buy one but...
It uses a bunch of old parts that are getting harder and harder to find.
And he even talks about the yamaha chip being hard to find.
No fast forward 10 years, even 5 years from now when you start having chip failures...
Is there gonna be a supply?
Why do this when I can run the emulator just fine and not worry about part failure and have nothing at all.
@@kingforaday8725 People bought the C64 Maxi and Mini because it could run all the existing C64 games. This machine can not, anything needs to be written from scratch. That will drastically reduce impact.
Commander X16 community love ❤
I think the X16 has the power of an Amiga...but... easier to program
Does it have a c compiler available?
I like a lot of this system. It would be nice if you could use it with some flavor of 6809 as well. as the 6502.
I hope that they took all the basic graphics commands and have the Vera implement them in the FPGA to make them super fast.
Overall a very interesting system.
iirc there is a c compiler for the x16
That flexible RAM/ROM mapping is very nice. The C64 would have been almost this great if they designed in a way that the cartridge port had the signals for mapping writable RAM into those 8/16K partitions (and the IO for sending mapping commands to the cartridge).
The C64 _does_ have those signals.
• `R/W̅` is on pin 5 of the cartridge slot, which solves your "writable RAM" problem. _Anything_ in the cart can tell if the computer's trying to read or write it, and respond appropriately.
• There are two specific address ranges, $DE00 and $DF00, with their own decode pins on the cartridge port, `I̅O̅1` and `I̅O̅2`. So that makes it dead easy to have something interpreting commands there. But that's just for ease of use and as a convention; the cartridge port sees _all_ writes the CPU does (except for addresses $00 and $01), so you can pretty much use any address you like to receive commands.
• There are various modes for what the cart can take over (i.e., be guaranteed that internal devices will avoid responding to), but the broadest one is by asserting the `G̅A̅M̅E̅ ` line which maps 4K of C64 internal RAM to the bottom 4K of memory, and leaves _everything_ else to the cart.
I did not know that (or I knew it wrong!). So I can put a large SRAM on a cartridge and with some decoding logic I can choose which chunk I want to see in the 64k somewhere? Then why do we need the DMA-style ram expansion if we can just swap 8k pages in "zero" time?
@@Flashy7 I see two fairly large benefits to the REU way of doing things:
1. You can transfer arbitrary-sized units of memory from 1 to 64 KB, rather than being stuck with fixed page boundaries. So if you wanted to, say, transfer just one frame buffer worth of data (as is done by the "Globe" demo), you don't need to swap out a bunch of other data.
This is particularly important when working near small areas of important stuff. For example, if your page size is 8 KB and you want to use different memory anywhere under address $2000, a bank switching system would also be switching out your zero page and stack which is almost invariably not what you'd want.
2. The REU leaves all existing address mapping working as it did before, so programs that depend on it need minimal modification, and programs unaware of it have a minimal chance of being broken by it, even if other things in memory (e.g. a TSR program) are simultaneously using it.
Yes. It is truly amazing. Also note that the CX16 has TWO banking areas for RAM. Between A000 and BFFF and between C000 and FFFF. When the memory expansion card is installed it can be loaded with extra RAM chips in the C000 and FFFF address space!!!!
@@svenvandevelde1 You can do roughly same on the C64, which has independent mappings for the following:
• $E000-$FFFF: internal RAM, KERNAL ROM or cartridge (ROMH line)
• $D000-$DFFF: internal RAM, CHAROM, or I/O+colour RAM
• $C000-$CFFF: internal RAM or cartridge (requires decoding on cartridge)
• $A000-$BFFF: internal RAM, BASIC ROM or cartridge (ROMH ine)
• $8000-$AFFF: internal RAM or cartridge (ROML line)
• $1000-$7FFF: internal RAM or cartridge (requires decoding on cartridge)
And of course the C128 has even more powerful mappings.