The big difference between a consumer PC motherboard and an SBC like the Pi Compute Module is that latter is supported and manufactured for a decade while consumer products have a much shorter lifecycle. Yes, theoretically you could replace it with the same motherboard if it’s available on the second hand market, but you can’t just drop in any ITX motherboard and expect it to work as the chipsets need to match as well. This is a point that should have been mentioned and the primary reason why their component choice is problematic.
Forgive me for my ignorance here, but what even is the purpose of the itx board. It seems to be connected via usb, so you should be able to just pop in a new computer system. with proper software
@@ToniTechBoi It is for file handling / storage and feeding the samples into the analog board. It you look closely you will see, that there is a PCB that connects the motherboards PCIE connector with the analog board. There might be some preprocessing done when you load the samples.
Seems to be running QNX, would be interesting to find out what storage the board uses and if there's a full image of it instead of the update packages to try it out on a different board but AFAIK, as long as the chipsets are register compatible and not tied to secure boot it might actually be fairly portable
People get a bug up their butts because they're anti-DAW snobs and it rubs them the wrong way to have a "computer" inside their Real Musical Instrument.
The audiophile people would have the PC in a separate box. The philosophy there is anything that causes EMI should not be anywhere near anything that does low level audio signals. This is why you have external DACs rather than Soundblaster cards. They run off separate power supplies.
A good retort. Espen admitted in a comment to me on his second video about this subject that he's doing this for the clicks and thus money. BTW, that video is the one with the noose in the thumbnail in which he claimed that Sequential was pissed at him when really all they did was try to tell him that the PSU he ignorantly suggested would not work and indeed damage the Prophet X due to a different pinout. But yes, even worse than selling any integrity he may have had for limited clicks from synth nerds are the myriad misinformed commenters who evidently have never peered into an electronic instrument before. Since the dawn of digital control of analog synths, ie the original Prophet 5 (designed by the genius, Dave Smith, who also designed the Prophet X obviously), manufacturers have used off the shelf parts to save time and money. The Prophet 5 uses a Z80 and all the standard outboard support ICs it needs to function. It's basically an 8bit computer with it's own ROM, RAM, and I/O. There's nothing custom about it except the software. Actually it goes back way before that. Guess what op-amp the original, highly regarded Oberheim SEM uses all over the place? The lowly 741. And yet it sounds great. Contrast that with the Alesis Andromeda, an instrument that uses custom designed ASICS. It's a great synth but if something goes wrong with one of those custom chips, you're screwed. Alesis hasn't made them in 20 years. The other irony is the only reason he was able to fix his Prophet X is because it IS a cheap, standard PC motherboard in there with a coin-slot battery. Ever change a battery in a DX7? In an Oberheim OB-Xa? Korg PolySix? I'm pretty sure Mr. Kraft wouldn't be able to do it.
Many moons ago, I made a deliberate decision to never monetize the channel. This was for a variety of reasons, but one was that I never wanted to be tempted to do things that would degrade the integrity of my work on this channel for financial reasons. (I do confess that my self esteem has largely become tied to watching those view counts and the subscriber count go up...)
I got curious and looked up some tidbits about the Prophet-X, and it's kind of crazy how, I wouldn't say _over_-engineered, but I guess, how complex it is for what is essentially a very expensive sampler. Like yeah sure, _part_ of it is a relatively cheap PC, but you've got what's likely completely bespoke audio software running on top of a proprietary-licensed OS kernel - they quite possibly had to write their own low-latency audio driver as QNX typically isn't used for embedded audio AFAIK, as well as code to interface with the controller and filter boards. And that's just the 'Freakin Cheap PC'! $4k is a lot of money, and honestly in spite of the engineering prowess, I would not consider the Prophet X a worthwhile purchase, but Espen's implication that it's a sloppily designed piece of kit, or that Sequential gouged their customers is dumb as rocks.
I feel very seen by that comment about VNAs! We may be using "standard" PC motherboards in the sense that they could be used to run anything (see CRD's little guys series). The main difference is supply chain. When we design around an SBC, we need assurances that the SBC will be available for as long as we will conceivably make the product, and when they do discontinue it they'll give us an opportunity to put in one last big order before their last manufacturing run. A customer as small as DSI can't get that from the likes of ASRock! When this specific motherboard goes off the market, they'll be on the hook to find something compatible or add support for something new.
There’s nothing wrong with using a pre-made computer platform in a synth. However some are better choices than others. Using a standard pc motherboard in a end product that doesn’t include a pc screen and keyboard is not a good idea because the BIOS need those peripherals to communicate errors to the user, and give the users a mean resolve those problems. In the case of Espen’s problem, the BIOS was indicating a problem with the CMOS battery and asked the user to press F1 to bypass this error and boot anyway. Since the prophet X didn’t have a standard computer screen and keyboard it looked like the synth was frozen because the boot process was interrupted and the Prophet X firmware wasn’t running. Had Sequential used a proper embedded computer platform like a Raspberry Pi or similar, this problem could have been properly handled because those platforms are designed to be embedded in applications that don’t include a standard pc screen and keyboard. This is a poor hardware choice coupled with a poor firmware that gives a sub standard user experience IMO.
But like wearing a suit at a business meeting, you should show your customers some respect. Don't let them see a low cost ASUS branded board in your product. HP once sold a low cost minitower PC at Walmart where it had a laptop mobo with an external power adapter. It was essentially a repurposed older laptop model sold as a desktop. It was shocking to see such a janky presentation come from an otherwise highly respectable company. Apple would never do that!
@@neilpatrickhairless Every high tech company wants the Burger King Rodeo Burger. Repurpose cheap hardware and freely available software, slap a label on it and call it a day.
All great points. I think you nailed it when you said 'you're paying for the interface'. These synths could just as easily be sold as a pc peripheral with some bundled software to install on your own PC... and charge roughly the same for them. I think these companies/products reach a point where the decision is made 'it's already costing thousands of dollars, we may as well integrate the pc into it so we have a controlled environment for our software'.
There's a company called SoundForce that sells custom small desktop MIDI controllers built for specific VST's like TAL U-NO-LX. Incidentally, the controllers cost only slightly less than actual small synths like the Nymphes, Pro-800 and DeepMind 6. When you buy a hardware synthesizer you're essentially buying a MIDI controller which just happens to come with a synth built in.
That's a very good take on the issue; the 224 immediately came to mind for me as well. Other examples would be the original PPG Waveterm A which was AFAIK also built around an existing SBC (not sure about the Waveterm B). When it comes to operating systems for synths or samplers it gets even more messy- the Fairlight III being built on top of OS9 (not MacOS 9!) or some Mackie stuff on top of BeOS (the d8b? or some harddisk recorder?). I believe at least some of the synths of the late 70's and early 80's were conceived with the help of microprocessor development boards (similar to the SBC in the 224); and when development was finished the hardware was redesigned using only those parts that were actually needed for the tasks at hand. This was more cost efficient and also made the devices smaller (again: the SBC in the 224 is quite large). Seeing how much the size has shrunken since then it isn't exactly a surprise for me that standard motherboards are used nowadays.
@@Lantertronics IMO, BeOS is one of the best operating systems no one knows about. It's unnatural how smooth and responsive the UI remains on ancient hardware loaded to 100% CPU utilization. The folks that designed and implemented their interrupt handling and scheduler were truly gifted.
@@stevepreskitt283 I like to imagine an alternate reality in which Apple based its new OS on BeOS instead of NeXTSTEP. (Of course that would never have happened with Jobs at the helm, and it's hard to imagine Apple surviving the 90s without Jobs at the helm).
First, you hit the main points very well. The reason you use a "standard" of anything, is because it reduces (long list), plus if enough of a certain item is made, then someone will make it for a long time as a repair part (see car parts that many manufacturers will make, not OEM, but then, some are with a different name). Whenever building something, using an "off of the shelf" (aka standard), just makes sense and is best practice. Even at the OS level. If people really knew how hard it is to make an OS. Furthermore, doing low level code is pretty F'n hard. Better to use proven APIs, if boot time is a problem, they should plan ahead. Good video, good points. Custom means, not better, just means, special (and hope it doesn't break).
This is the problem with the internet. The uninformed (which, let’s face it…that’s most people) rise above all and some stupid, baseless narrative can catch on like wildfire out of nowhere. The internet gave a voice to the previously voiceless and we had no idea how annoying that would ultimately become…
Years ago I had a Prophet 600 (one of the very first MIDI synths available) -- it ran on a bone stock Z80 CPU and several MUX/DEMUX chips sending data through digital-to-analog chips to create the control voltages to all those wonderful Curtis CEMxxxx VCOs, VCAs, and VCFs. It worked, and it worked well (until something inside went wacky and it wouldn't tune right - and I traded it in on a Casio CZ1 - another early MIDI synth, which I still have, and use). So what's the big deal?
Some people seem to be really upset that a manufacturer might choose to use an OEM PCB with a microprocessor on it designed by another manufacturer in their product vs. in-house designing their own PCB with that same microprocessor, which is seen as more honest and pure, or something... ;)
@@Lantertronics OK, but apart from maybe Yamaha (which ironically enough, made synthesizer chips to put in computers), how many companies have chip fabs to create their own chips. And how many of the old synths had those Curtis chips in them? quite a few, I would imagine...
The days of purpose-built hardware seem to be gone, which makes sense, since computing power is so abundant that you don't need to build specialized chips for everything. Even early digital synths were already using some off-the-shelf chips. I mean even my Ensoniq Mirage is really just a fancy Commodore 64 with a nice soundcard and a keyboard.
A relative of mine bought a broken Lexicon 224. He had to make a couple minor repairs to two of the internal PCBs, and then he designed his own remote for it in a new style case, and it's flawless now. Quite a cool project. I hate it when people who don't do any design work or building, or anything else at all (sometimes) go an complain about the design choices of designers. Or the price complaints. Like you, I tell people, buy it or don't, enjoy it or don't, but things cost what they cost, and design choices were made for very good reasons (most of the time). Excellent video!
Great video, Dr. L. I really don't see a problem with using OTS components if it allows a quicker/cheaper development process and allows the manufacturer to easily source parts that they'd have to have custom-made otherwise. We're not back in the 80's when Yamaha had to design the OPS/EGS chipset to be able to offer a rudimentary but usable FM/PM synthesis capability. Any one of the $30 Raspberry Pis sitting on my desk has more computational horsepower than a full $250K Synclavier or Fairlight rack system had back then, and SO MUCH of what used to have to be done with discrete hardware can now be done in software, in real time, for very little cost. Is $4000 a lot for an instrument? Sure it is, but consider that the original Prophet-5 cost the same amount of money in 1978, which is over $20K after adjusted for inflation, and the Prophet-X is a much more capable instrument overall, "vintage analog" arguments aside. I will say that I would MUCH rather my Yamaha SY-77 had a simple SBC than the stacks and stacks of bespoke boards that are in it - just changing the display in that beast is a major operation because of everything that has to be removed to get to it. 😀
Except when your synth is completely a VST with a physical interface all you are really paying for is the looks/feel/ergonomics of the synth. Like why not just get a midi controller and spend that money on a nice PC to play as many VSTs as you’d like. I’d rather have an actual working computer if I’m just going to be purchasing a computer (ala the Prophet X). The Hydrasynth is a better value proposition than the X in this regard.
@@rustyshackleford634 But that's NOT what the Prophet X is - it's a true digital/analog hybrid. The X is controlled by a regular PC board, but there's quite a bit of custom circuitry in there (primarily the analog filters and UI hardware) that the synth simply doesn't work without. I totally agree that VSTs are a better value proposition for many (myself among them), but just calling the Prophet X a VST disregards the very real amount of additional stuff that's in the audio path.
@@rustyshackleford634 omg I can't with this argument. some people want to have an actual instrument; who cares if it runs on tubes or an atx motherboard. Yes you're paying for the looks; feel; ergonomics; software dev; engineering; sound of the synth; basically yes you're paying for a synth. and again; the prophet X is a hybrid synth; with analog filters; so the furthest thing from just being a "vst in a box". and even a fully analog synth has a digital part; so basically a little computer; so in fact aren't all analog synth vst's in a box? why are you triggered by the fact this synth uses a part that you can recognizer? Do you know they're not manufacturing every single analog chip they use in their design either?
The Prophet X isn't like the Hydrasynth. It has boatloads of analog filter chips. If that's not important to you, the yes, the Prophet X doesn't have a good value proposition.
Saying "It's just a cheap ITX motherboard" makes no sense. Most of the electronics that goes into synths is cheap. You can buy the SSI filter and oscillator chips that go into the $5k OB-X8 for about $5 each (probably much less if you buy in bulk). Capacitors, op-amps etc. are also dirt cheap. You aren't paying for the individual components, you're paying so the company can recoup the R&D cost and the cost of all the tooling for setting up for mass production. Of course the more units they sell, the more people there are to share this cost, which is how Behringer are able to keep their prices so low.
All you need these days is a PC, a midi keyboard, some nice speakers connected to a decent soundcard and some virtual synth software. You could possibly build a system that sounds like an expensive synth for around a thousand quid these days
@ I just recently learned that but now it all makes sense that you could connect a monitor, mouse and keyboard to those. Which actually made editing easier. The PPG digital synths also were all based off of some standard industrial 6809 platform btw.
No, the Roland S series samplers (and W30 workstation) had nothing to do with the MSX architecture. The MSX used an 8 bit Zilog Z80 processor, whereas the S series used processors from the 16 bit Intel 809x family. Earlier Roland products did use the Z80, but nothing about the implementation reflects any influence from the MSX standard.
@ thank you, recently read it and from the looks of the screen it would have been possible. On a vintage synth page I found the info that Roland used the MSX mouse. Maybe that’s where this comes from. Edit: I checked the service manual and the CPU and it was a 16 bit architecture. That makes more sense than an 8 bit Z-80 architecture.
the idea that off-the-shelf parts are insufficient for an application like this is laughable. these components were sold into a highly-competitive market that's extremely price-sensitive and margins are very tight. every RMA represents a significant loss for the manufacturer - these parts are designed to be fumbled together by sweaty kids in their basements, and just work - so it's in their best interest to maintain a high level of reliability. if Sequential had replicated all the R&D to build their own custom board with the same level of performance and reliability, the price would have been more than double.
The fundamental philosophy behind technology choices is directly related to purpose. A pc motherboard is a PC motherboard because it facilitates and communicates best for a computing environment with a monitor, keyboard, and mouse. NOT a synth. I have owned synths since 1985, and one pf the bedrock principles of a synth is it needs to work reliably when you turn it on and need to play/perform. For that you need bedrock reliability and something that can withstand transport (banging around). Designing for that purpose delivers a board that facilitates those needs.
In this case Sequential failed to properly perform reliability testing on this particular configuration. Perhaps they didn't mount the motherboard well in a way that reduced it being subject to vibration; perhaps there's a problem with thermals; or a combination of those and other things. I don't recall Korg customers complaining about motherboards failing in the Kronos, so the use of a motherboard itself isn't inherently the problem.
This was exactly my thought when that video first came out. Like do people not realize that everything has a computer in it nowadays lol? And the DSP isnt usually even done in that part of the architecture. Espen kraft is on this kick of shaming hardware companies and users and raking in the views. It comes off to me as major projection for his own vintage synth GAS, from which he’s now repented and been born again. Its just corny as shit
Sequential isn't involved in price gouging. BUT the prophet rev2 was released at €1443 in 2018 and now costs €2485. That's a 72% price increase. If you consider inflation, the price should be around €1800.
Inflation isn't the same across the board though - some things are more affected than others. So the price increase might reflect their actual increase in operating costs - or it might be even worse price gouging. We don't really know.
I guess some people are going to be really pissed that the Korg Kronos used an off-the-shelf Intel atom motherboard running Linux, the newer Akai MPCs used an ARM 32-bit processor on a SOM module running Linux, and Korg's smaller synths such as the Wavestate, Modwave, Opsix, and now the Multi/Poly synths use Raspberry PI compute modules also running a real-time fork of Linux. Edit: Sorry for the repeated info here, I commented before hearing the rest of the video.
Just saw a "Look Mum, No Computer" video where he got to play on Zinovieff's EMS VCS4. He mentioned in that video the rest of the EMS studio was lost in a flood while in storage, presumably including the PDP8s.
I've got a kit called a "PiDP8" that I need to get around to building. Appropriately enough considering your video's subject, it's a working replica of a PDP8 control panel with a Raspberry Pi inside that runs an emulation of the DEC mini computer itself.
@@chriswareham I've seen it! I'm obsessed with the PDP-8 so I've been tempted to get it, but I have to be realistic and realize I'd probably never get around to building the kit...
Not sure how this is different from the 1980s Synclavier that came with a Mac II, apart from that the Mac II wasn't stripped down to the pcb and incorporated in the Synclavier rack.
The Mac was basically just the terminal for the Synclavier which still was running its own OS on its own processor. So the Synclavier was basically "outsourcing" a part of the user interface to the Mac (and admittedly some other functions - but sound generation and sequencing was still completely inside the Synclavier itself).
@@maro_from_germany Sure but my point is they could have done that in-house and in hardware like Fairlight. Using a Mac instead seems effectively the same as using a mass-market PC motherboard inside the keyboard. It's just that computers have become so compact that it now fits inside rather than being a separate unit.
@@jonathanhendry9759 The Synclavier used external terminals from the start (first a DEC, then another one that I don't remember right now); and back in the day NED even had a mouse that connected directly to the mainframe. Using a Mac was basically the continuation of this; in large parts it still was that "dumb" terminal that was not much more than a glorious console connecting to a serial port. When using it beyond that you needed to add a bespoke Nubus card from NED.
Wow, isn't this silly! The "computer inside" thing isn't anything new. It's not even something from this century. In fact, the first synth using a CPU is something which many synthesists would gleefully agree to unpleasant bodily alterations in order to have one. I bring you... The ARP Chroma. Notice that I put the ARP part in there, because they developed it. But what they developed it around was a logic board powered by an Intel 80186. Yep, that's a "1" there; this chip was intended for use in industrial and early embedded use. Yes, it was used to control the parameters, deal with user preferences, and I would figure it provided the VCO tuning and the like. I did get to play one of the 50-ish ARP builds, though...and the comparisons are alarming. And I mean "alarming" as in a bizarre drift problem that the CPU was supposed to be preventing...but which wouldn't be fixed even if you hooked it up to Deep Blue! No, it was the VOICE cards...and NOT the processor...that were at fault. Or more correctly, it was a bunch of people who thought they knew better at Gulbranson's old Hoopeston, Illinois factory, where the manufacturing moved post-ARP. Gulbranson used a flux method that wasn't compatible with the card boards, and instead of the flux going away...naaah, it just wound up contaminating the breadboard stock. On the first 500 or so. But the point is this: when this happened to one of the most anticipated synthesizers of the day, one which made a lot of noise about that CPU being THE way to correct previous polysynth reliability issues (like tuning...this was the exact same period when Yamaha dropped the CS80, only for users to discover that it would go out of tune from merely moving it from one room to another!), welllllll...it took some time to figure out that this contamination was the real cause. And during that period, with frustrated customers, terrified Fender/CBS brass, confused board fabbers out on the prairie, SOMEbody tossed out the unfounded AF idea that... "THE COMPUTER IS RESPONSIBLE!!!" And once that got tossed into the info void at that time, the idea that this brand new feature was to blame went up like a mountain of flash paper wads! Sure, it wasn't TRUE...but nobody had any other idea (yet), so yeah! COMPUTRR NOGO INN SINTH!!! So, remembering the tire fire that resulted, yep, this is just an echo of that. It's like an archetypical memory...like Xenu and the Space Jetliners...but it does seem as if every recurring hiccup of this has that same pearl-clutching going on. C'mon, y'all...just trust your synth designers, OK? They know what they're doing...yes, even Uli, I'll begrudgingly admit...so step back and let 'em breathe. Don't theorize your way out of snagging the cool stuff, either. The tech in this is flying past us at fascinating but disturbing paces, we just need to be more open-minded about what that tech-barrage can do for US.
Had one and sold it after dealing with and constantly hear similar complaints from other buyers. When it worked it was a cinematic synthesizer. Big sweeping effects and rich tonal sounds. But you can literally buy those things as a VST.
Yeah, the thing is that DSP emulations of analog filters have become very, very good -- so there's not much point to the giant analog board (I realize many disagree with me). I have a video comparing the Arturia Prophet VS emulation with the original Sequential hardware (my rack unit, which I've since sold), and asked people to tell which is which, and so far no one's guessed much better than random chance (and the guesses are all over the place from person to person). So although I teach a class on analog circuits for music synthesis because I think analog circuits are fun to tinker with -- as far as practical tools goes, whether something is "real analog" doesn't much matter to me one way or another.
I get the feeling Sequential rushed the reliability testing. Maybe it's a heat issue? Maybe they didn't put in enough stuff to deal with mechanical shocks? Who knows...
Living with and having to maintain a CMI 30A and a Hartmann Neuron I would have say that I'm not so sure that I agree completely... On the other hand keeping the other computers in the studio synths and FX units like the AMS's, PPG 2.3 and Waveterm and all of the vintage synths like the Prophet's, Oberheims, Synergy and Synthex are "relatively" easy (except when they aren't). I guess time will tell!
That being said I do like the idea of using the smaller systems like Raspberry Pi in synths as in the future they will be easier to replace with later iterations of micro. Full PC motherboards can be painful to replace a decade down the track and if these machines can't last that long then they are a bit of a fail given the longevity of the Z80 and 68xxx board so far.
@@Tharsis879 This got me thinking -- it would help if manufacturers would embrace more open standards. Not that they necessarily have to give away all their source code -- but they could publish interface specs that would let someone swap in a newer motherboard that would stay compatible with the rest of the hardware, and enough detail on how the software works that someone could cobble together drivers for a newer OS. (I'm just spitballing here, I realize there's problems with this idea).
Had a Memorymoog Plus that struggled with using what was basically an Z80 micro Single Board Computer and the supporting boards, digital, analog and voice cards all wired up looking like Chef-Boyardee and Bob Moog had a baby synthesizer. It was semi-reliable, but nobody cared because it would melt your speakers. We will cope with knowing that there's way more digital than analog under the hood in today's synths, but we love that knobs and sliders are back and we don't have to be victimized by Microsoft directly, but I do love Cherry Audio VA synths.
I can understand some of your arguments, but comparing a compute module or some other industrial grade motherboard with solid caps and no moving parts to this dirt-cheap ASRock one doesn't work. ASRock is known for putting bad components in, caps can go bad pretty quickly and if your 4K synth fails because of this, this is not ok at all.
this isn't an industrial application. there aren't lives at stake here. it's not getting exposed to temperature or shock extremes. this machine is making bleeps & bloops. a consumer-grade solution is plenty. the reliability and ease of replacement of these components far outweighs the cost that a custom/industrial solution would incur.
Asrock aren't really an outlier when it comes to consumer motherboard quality, or consumer electronics in general. I've had synths with completely custom boards go bad due to cap leaking, it's not like the fail rate will be substantially worse than literally any other synth built in the last 30-40 years.
Great points, talking about cost vs value. One point you can potentially add is that the Linux KERNEL is not an Operating System on its own. It just creates a standard interface for applications to use to access hardware. It likely doesn't have a full "userland" and probably takes just seconds to do Inital Program Load.
Nothing wrong with a good Pi. That's the compute behind the Zoxnoxious synth, much similar in concept to the Prophet with an analog audio path and digital control. I really like your videos, good depth of discussion. Your filter videos were really useful in a recent project!
Edit: oh nevermind, should teach me to watch the whole video before commenting.... The legendary Korg Kronos synth had a pc motherboard too. And nobody ever complained about it....
Yeah, I mention that along with the Hartmann Neuron (although plenty of people complained about that... but I think the Neuron was released before it was ready).
@@raul0ca Indeed, that's a big advantage of a solution built around a microcontroller with more bare-metal programming. Running a full Linux kernel comes with a lot of overhead.
As did its much more expensive predecessor, the Oasys. You could've paid up to 10k for a Pentium 4, but you did also get a beast of a workstation with an absolutely incredible interface.
Great video - I've always been frustrated by this "ugh it's just a cheap PC?" line of reasoning. I 100% put more value well thought out architecture and UX. Yes a lot of what makes up a Hydrasynth could be a VST, but the UX and UI are absolutely worth paying for.
Well... there's other things, like the physical hardware probably isn't going to make you log in to some license server that will go down in five years...
He quit his day job to go full time as a UA-camr, and I think he's struggled to maintain the engagement he used to have. The clickbait video titles and provocative content are probably an attempt to boost his views. A shame, as I loved his older content and even donated a few bits of 80s music gear to him.
I’d love for you to tackle Espen’s “VST in a box” claim about synths like the Prophet 5 rev 4 and Ob-X8 because of their use of digital envelope and voice control with the vintage knob.
Some 80s vintage synths with software envelopes have reputations for having "slow" envelopes (or odd curves) but I don't think people realize that's likely due to the fact that they were being done on a single 1-4mhz 8-bit microcontroller that was responsible for 5 other things on the synth.
@@ddqd123 But that doesn't change the fact that the envelopes were (sometimes) slow. I clearly felt this when comparing a Prophet 5 to a Prophet T8. Without a doubt Sequential could also have used hardware envelopes (CEMs) on the T8 (like they did on the P5) - but they didn't. And to be really picky about this: The T8 used a 6MHz 16-bit processor just for envelopes, LFOs and CVs - and a Z-80 for all the other stuff (scanning controls, keyboard, MIDI and so on). I believe they simply needed to gather experience in writing this stuff - or they simply accepted this (inferior) version as sufficient.
The funniest part about that video was him ranting about cheap digital implementation on analog poly synth while a six track and 106 we’re behind him. Both of which are heavily reliant on digital control. I’m sure he will be mad when he opens up a vintage p5 and an obx to find that they both used off the shelf z80s.
You're damned if you do, you're damned if you don't. Apple is despised for the exact opposite reason. They spare no expense in giving the end user a exceptional experience. They would rather incur the significant expense of developing their own software and hardware than license them from a vendor. They've even gone as far as designing their own microprocessor, which is unheard of, especially for for consumer market. Yet, many complain that Apple products are overpriced, proprietary, and difficult to repair.
Finally, a view on this I actually agree with. I've been a little annoyed at Espen's recent videos, especially because I know what it takes to make the devices he keeps calling "scams". I was particularly annoyed by the implication that Behringer devices are the only "fair" devices, when in my opinion they're as far from it as possible.
If you replace where Espen says "scam" with something meaning "a value proposition that many may not find appealing," then he has a point. But the term "scam" implies deliberate deception, which is unfair. The thing about Behringer is that no one can really compete with their economies of scale. A company like Groove Synthesis buys X number of resistors for a couple versions of basically the same product, whereas Behringer can buy many many many more of that spread over hundreds of products.
I disagree. The computer chips are a way of short cutting work and making your synths cheaper. This is absolutely INSANE when you think about the money they cost. The parts they use now massively reduce repairability, they end up simpler in design (if they had to use actual discrete circuitry to achieve what they do with emulation the synth would be even more complex, not less), and half the time you just end up with problem. Synths are getting buggier and buggier and it’s because of software issues.
I am less concerned about whats inside my boxes as I am on how well they work and are built. Fortunately, I have not ever had to open one. Knock, knock on wood.🤔🎶🎹🎶🎸🎶Play On
I didnt get Espens post either, most of the all the modern - synths ive seen use PC boards? I do think manufacturers are not supporting their ( expensive) older instruments is worth getting pissed about ...but thats a different issue! As an aside, When it come to value for money, Espen promotes the retroactiv midi controllers- they are great piece of kit but their price ( for their limited specialised use) i think is a total gouge😮, so i really think hes damaged his credibility on that front......
The Retroactiv prices make sense from the standpoint that they're not making many of them -- they have a very limited market, so they cannot exploit economies of scale.
@Lantertronics yes I appreciate that, but they also require almost no development cost ( unlike synth software) & have minimal cost components ( no experience is etc)- so personally I see that as overpriced. But like everything (& something I think Espen chooses to ignore with his prophet vid) it subjective & regardless of components etc people will pay more for a sequential synth than a korg etc because they can see the value in it..😉
I'm sad they stopped putting synth chips in my videogame consoles. Ooooh, fancy samples and cd quality! Please. Give me some of that tasty Yamaha or SID instead.
So long as it's replaceable. A PC mobo should be replaceable with any mobo from that CPU generation, so an AM4 board if there's an AMD chip, or the appropriate Intel board, etc. The reason analog hardware is great is that it's not proprietary, anyone who understands analog can repair it, indefinitely, with common parts. That's not true with boards. If they've done proprietary fuckery with this board, then when it fails, either you pay them something silly for the replacement, or worse, they just don't make one, and you're hosed, your $4000 synth is junk now. This UA-cam random did not do the sort of long term reliability testing to verify that, and doesn't get to tell us what's okay. It's not your money. We're not your children.
Alas, it looks like the ease of repairability of analog gear is only true of older gear. All the PCB photos of newer analog gear I'm seeing is all surface mount that's awfully hard to repair unless you have a lot of experience and the right gear.
You seem to have misunderstood my point (admitted, the title alone does not make this entirely clear): I was addressing the question of whether Sequential (or any other manufacturer) is inherently "pulling a fast one" by using an OEM motherboard, which is what a lot of people were claiming. Whether Sequential messed up in the details of *this particular* instance is another issue. As I note at the 6:06 mark, Sequential clearly messed up by doing insufficient reliability testing, and for *that* they should wear the Cone of Shame. Maybe they didn't do a sufficiently good job in mounting the motherboard so that it could withstand the strain of being moved (it would be interesting to know if people who gigged with their Prophet X had more trouble than people who just used it in the studio). Or, maybe there's a heat dissipation problem. Or maybe *that particular* motherboard is just junk and they should have chose a different one.
@@Lantertronics And let's not forget the issue of obsolete/exceptionally hard to find parts. If certain SSM chips fail some gear can quickly turn into a doorstop...
This is a result of social media. People fire off these useless remarks like they are the law when they have 0 clue at all. Same people probably wonder why others feed their dogs when you can just leave them outside to hunt for themselves.
Whilst they literally record, perform and market music to an audience entirely comprised of fuckin computers with a consumer grade motherboard 🤣💀🤣💀🤷🏻♂️
I’d rather just spend 2K to buy a high end PC and use the leftover to grab a nice midi controller. At least that way I won’t have to deal with firmware issues, component breakdown, and I can play video games too. That’s more than Sequentials new computer (Prophet X) can do.
One of the main selling point of the Prophet X is the analog filters, which is important to some people, and you won't get from just your own PC. If that's not important to you, then yeah the Prophet X doesn't have a good value proposition.
Finally, some say what had to be said. I have a PX and I'm glad it has a PC motherboard and a Intel SSD and it sounds great! Kudos to you @lantertronics
I recall the first generation of Sequential synths, when they did have their own microprocessor systems and firmware designed and coded by Dave Smith - so it is a little sad, and just the current practice, to see this outsourced.
Why would you design your own MCU when there is an off-the-self part that does what you want? It's not commercially viable. Maybe back in the day D.Smith did not have a part that would do what he wanted and then design its own MCU.
It's also cool to see the hand-laid PCB designs in the older Prophets, with tinned traces instead of today's ubiquitous green solder mask. No one uses curved traces anymore. 🙂
The firmware was designed and coded this way because the processors were slow and ROM and RAM was expensive. Still the Prophet 5 used a stock Z80. Dave Smith probably could have put a development board into the Prophet but I believe those were rather expensive at the time (and also had additional functionality that wasn't needed).
This makes no sense at all from an engineering viewpoint or a marketing viewpoint. Early synths had didn't have bespoke CPU boards with Z80s and ROMs and RAM because the designers wanted to show off their engineering prowess; they had them because it was the best way to hit a price point at the time. If they could have sold the synth cheaper by using an off-the-shelf SBC (like Lexicon did in their 224), they would have.
I bet most people would not make the difference in a blind A/B test. The people who made the vongon replay first designed a fully analog synth. Then they tried making a digital counterpart with the same "analog" style sound and ended up liking it better so went with that. I find this synth sounds lovely; but they're getting crap for it (and yes it's not a cheap synth). I like analog and it does sound better in certain cases but it really should not matter in the end. The sound and experience of using the thing is what matters.
@@valdir7426 I think thats possible, a similar test has been done with violins costing millions of pounds vs a modern built ones, and they cannot tell. But ! there is not doubt some psychological element that would favour the million pound antique violins. I preferred digital synths because of their accurate reliability, but have always had a sense something is lesser somewhere, and i guess its the virtually imperceptible, but present artifices of the interrupt cue.
I have another video where I compare the original Sequential Prophet VS with the Arturia softsynth version; I randomize which I played first in each pair, and asked people to tell me which I thought. So far no one has guessed significantly more than what you'd get by random chance.
@@Lantertronics I wonder if its possible to sample at a very high rate and thus actually see the slight imperfections of the interrupt queue vs a c continuous analogue signal. At some level it must be demonstrable. This is my only argument really, that anything running on a personal computer is subject to wildly different renderings, at any given moment. its way faster than a 16 or 32 bit sample, practically inaudible, but its there.
@@JackNicklauson Well, considering everyone is ultimately listening to an audio file through a DAC being played by a computer, I'm not sure it matters. ;)
Espen has gone full troll / controversial rage bait with his content. I do love my prophet x and appreciate this video, but I think it's best if we all just ignore him at this point.
The dark side of UA-cam Algorithm is trying hard to convince me to make more content like this (almost 6K views in 3 days, which is a lot for my channel -- I mostly make math-heavy lecture videos about analog electronics, which admittedly I don't expect to be that attractive to a general audience).
Why doesn't the profit X use a quantum computer or at least maybe a giant parallel NUMA cluster. I am triggered and offended that the prophet X does not use at least 200,000 Watts.
Seriously, go look at the comments on the original video. It's non-stop people complaining that Sequential is pulling some kind of "fast one" by using an OEM motherboard.
With the competition from the behemoths it's a life or death proposition for many of them, if they can't make enough spare money they can't recoup their dev costs and will either have to just make the same product forever or close their doors. The disparity between Behringer and their competitors is hard to overestimate, even Roland's outsourced production can't compete in terms of worker salaries and bespoke factories. If you believe "natural selection should run its course" and most companies deserve to close, then fine, but I personally appreciate a more varied market
What exactly are you disagreeing with? I was specifically making a point about about the use of OEM parts in a synth and whether that made sense or not from an engineering standpoint. Whether synth companies are coming out with shite stuff more broadly is a different discussion.
You got it wrong. This isnt about making the synth more expensive due to them RNDing a computer, is about them A) Beign upfront right in the box "Powered by Raspberry Pi" or "Powered by Intel", some people not care but others would be interested, many do electronics, computers and stuff. Why hide it and/or not disclose it? and B) If they are running off shelf parts, why not pass THE SAVINGS to the customer? Why you are so pro business? It IS not only a cheap computer and a mostly barren SMT board, running off a Prophex looking MIDI controller, HARDLY $4k, it is also very lazy design throwing in there an ITX board and calling it day. Put it together DIY? My brother in Christ, plenty people already make standalone MIDI controllers with VSTs inside running off ITX boards, raspberry Pis, FPGAs... you name it. DJs have been doing "laptop less" setups by hiding the computer in the case long before standalones were a thing. RnD costs? Sure you can design the UI but the backend will be done in India for a few rupees and hour, of course, based on some copy/pasted code from Github.
but why does it matter? do you expect manufacturers to list every single chip and components they use? they should make such lists available for people who want to repair their stuff but just because people are familiar with intel and raspberry doesn't make it relevant to advertise. they could as well put "powered by a texas instrument converter" but no one would care. That's an interesting trivia but no it's not something people SHOULD absolutely know. Also: not using "of the shelf" part (sorta for the compute module) would increase the cost of r&d and manufacturing, not the other way around. And finally you can look up the specs of the prophet X and it's not even a digital synth; it's a hybrid synth with fully analog filters; so your point about it being "a vst in a box with a controller" doesn't even make any sense. the motherboard is only a small part of what makes it a synth. you really don't have a clue what this thing is.
"pass THE SAVINGS to the customer" the savings were already passed to the customer. you obviously don't understand how marketing and product pricing works.
There's a tremendous amount of incorrect information and misunderstanding here. How in the world can you look at that massive analog board with all those filter chips and call it a "mostly barren SMT board?" Did you watch the full video? Did you miss the section at 2:52? Sure, try DIYing a Prophet X and see how it goes. Go write the Verilog for the FPGA that handles motherboard-to-analog-board communication. Maybe the value proposition of the Prophet X doesn't appeal to you, and that's fine.
The big difference between a consumer PC motherboard and an SBC like the Pi Compute Module is that latter is supported and manufactured for a decade while consumer products have a much shorter lifecycle. Yes, theoretically you could replace it with the same motherboard if it’s available on the second hand market, but you can’t just drop in any ITX motherboard and expect it to work as the chipsets need to match as well. This is a point that should have been mentioned and the primary reason why their component choice is problematic.
Good point -- I'm going to pin this comment right now.
Forgive me for my ignorance here, but what even is the purpose of the itx board.
It seems to be connected via usb, so you should be able to just pop in a new computer system. with proper software
@@ToniTechBoi It is for file handling / storage and feeding the samples into the analog board. It you look closely you will see, that there is a PCB that connects the motherboards PCIE connector with the analog board. There might be some preprocessing done when you load the samples.
@@cfbdk completely missed that one, those darn small x1 connectors.
Seems to be running QNX, would be interesting to find out what storage the board uses and if there's a full image of it instead of the update packages to try it out on a different board but AFAIK, as long as the chipsets are register compatible and not tied to secure boot it might actually be fairly portable
Some incorrect info floating around: The Prophet X is running QNX. An RTOS very different from Linux.
Oh, that's good to know. I will put a correction in the description.
QNX is also used by NASA
@@SpookyAlien-2001 Interesting!
Yes and also very not free, like very very not free.
@@hellcoreproductions Eeek -- the website is all like "contact our sales team." That's never a good sign...
People get a bug up their butts because they're anti-DAW snobs and it rubs them the wrong way to have a "computer" inside their Real Musical Instrument.
The audiophile people would have the PC in a separate box. The philosophy there is anything that causes EMI should not be anywhere near anything that does low level audio signals. This is why you have external DACs rather than Soundblaster cards. They run off separate power supplies.
A good retort. Espen admitted in a comment to me on his second video about this subject that he's doing this for the clicks and thus money. BTW, that video is the one with the noose in the thumbnail in which he claimed that Sequential was pissed at him when really all they did was try to tell him that the PSU he ignorantly suggested would not work and indeed damage the Prophet X due to a different pinout.
But yes, even worse than selling any integrity he may have had for limited clicks from synth nerds are the myriad misinformed commenters who evidently have never peered into an electronic instrument before. Since the dawn of digital control of analog synths, ie the original Prophet 5 (designed by the genius, Dave Smith, who also designed the Prophet X obviously), manufacturers have used off the shelf parts to save time and money. The Prophet 5 uses a Z80 and all the standard outboard support ICs it needs to function. It's basically an 8bit computer with it's own ROM, RAM, and I/O. There's nothing custom about it except the software.
Actually it goes back way before that. Guess what op-amp the original, highly regarded Oberheim SEM uses all over the place? The lowly 741. And yet it sounds great.
Contrast that with the Alesis Andromeda, an instrument that uses custom designed ASICS. It's a great synth but if something goes wrong with one of those custom chips, you're screwed. Alesis hasn't made them in 20 years.
The other irony is the only reason he was able to fix his Prophet X is because it IS a cheap, standard PC motherboard in there with a coin-slot battery. Ever change a battery in a DX7? In an Oberheim OB-Xa? Korg PolySix? I'm pretty sure Mr. Kraft wouldn't be able to do it.
Many moons ago, I made a deliberate decision to never monetize the channel. This was for a variety of reasons, but one was that I never wanted to be tempted to do things that would degrade the integrity of my work on this channel for financial reasons. (I do confess that my self esteem has largely become tied to watching those view counts and the subscriber count go up...)
@@Lantertronics Disseminating knowledge can be a reward on its own. Major props to you.
I got curious and looked up some tidbits about the Prophet-X, and it's kind of crazy how, I wouldn't say _over_-engineered, but I guess, how complex it is for what is essentially a very expensive sampler. Like yeah sure, _part_ of it is a relatively cheap PC, but you've got what's likely completely bespoke audio software running on top of a proprietary-licensed OS kernel - they quite possibly had to write their own low-latency audio driver as QNX typically isn't used for embedded audio AFAIK, as well as code to interface with the controller and filter boards. And that's just the 'Freakin Cheap PC'!
$4k is a lot of money, and honestly in spite of the engineering prowess, I would not consider the Prophet X a worthwhile purchase, but Espen's implication that it's a sloppily designed piece of kit, or that Sequential gouged their customers is dumb as rocks.
I feel very seen by that comment about VNAs! We may be using "standard" PC motherboards in the sense that they could be used to run anything (see CRD's little guys series). The main difference is supply chain. When we design around an SBC, we need assurances that the SBC will be available for as long as we will conceivably make the product, and when they do discontinue it they'll give us an opportunity to put in one last big order before their last manufacturing run.
A customer as small as DSI can't get that from the likes of ASRock! When this specific motherboard goes off the market, they'll be on the hook to find something compatible or add support for something new.
Thanks for projecting thoughtful nuance and pragmatism into this space.
The world could definitely use more thoughtful nuance and pragmatism right now!
There’s nothing wrong with using a pre-made computer platform in a synth. However some are better choices than others. Using a standard pc motherboard in a end product that doesn’t include a pc screen and keyboard is not a good idea because the BIOS need those peripherals to communicate errors to the user, and give the users a mean resolve those problems. In the case of Espen’s problem, the BIOS was indicating a problem with the CMOS battery and asked the user to press F1 to bypass this error and boot anyway. Since the prophet X didn’t have a standard computer screen and keyboard it looked like the synth was frozen because the boot process was interrupted and the Prophet X firmware wasn’t running. Had Sequential used a proper embedded computer platform like a Raspberry Pi or similar, this problem could have been properly handled because those platforms are designed to be embedded in applications that don’t include a standard pc screen and keyboard. This is a poor hardware choice coupled with a poor firmware that gives a sub standard user experience IMO.
But like wearing a suit at a business meeting, you should show your customers some respect. Don't let them see a low cost ASUS branded board in your product. HP once sold a low cost minitower PC at Walmart where it had a laptop mobo with an external power adapter. It was essentially a repurposed older laptop model sold as a desktop. It was shocking to see such a janky presentation come from an otherwise highly respectable company. Apple would never do that!
Someone make a thin client version of AV Linux and throw them on little buddy boxes immediately
The public'll love it, you'll make a million
@@neilpatrickhairless Every high tech company wants the Burger King Rodeo Burger. Repurpose cheap hardware and freely available software, slap a label on it and call it a day.
I don’t think that’s a fair assessment of the Prophet X, though. That’s a very complicated analog board.
All great points. I think you nailed it when you said 'you're paying for the interface'. These synths could just as easily be sold as a pc peripheral with some bundled software to install on your own PC... and charge roughly the same for them. I think these companies/products reach a point where the decision is made 'it's already costing thousands of dollars, we may as well integrate the pc into it so we have a controlled environment for our software'.
That option more or less exists with the NI Komplete Kontrol S series, or Arturia controllers, too
There's a company called SoundForce that sells custom small desktop MIDI controllers built for specific VST's like TAL U-NO-LX. Incidentally, the controllers cost only slightly less than actual small synths like the Nymphes, Pro-800 and DeepMind 6. When you buy a hardware synthesizer you're essentially buying a MIDI controller which just happens to come with a synth built in.
That's a very good take on the issue; the 224 immediately came to mind for me as well. Other examples would be the original PPG Waveterm A which was AFAIK also built around an existing SBC (not sure about the Waveterm B). When it comes to operating systems for synths or samplers it gets even more messy- the Fairlight III being built on top of OS9 (not MacOS 9!) or some Mackie stuff on top of BeOS (the d8b? or some harddisk recorder?).
I believe at least some of the synths of the late 70's and early 80's were conceived with the help of microprocessor development boards (similar to the SBC in the 224); and when development was finished the hardware was redesigned using only those parts that were actually needed for the tasks at hand. This was more cost efficient and also made the devices smaller (again: the SBC in the 224 is quite large). Seeing how much the size has shrunken since then it isn't exactly a surprise for me that standard motherboards are used nowadays.
Oh I didn't know about the BeOS thing -- that's interesting!
Yeah, using a development board as a "reference design" and then customizing bits here and there would have been quite common.
@@Lantertronics IMO, BeOS is one of the best operating systems no one knows about. It's unnatural how smooth and responsive the UI remains on ancient hardware loaded to 100% CPU utilization. The folks that designed and implemented their interrupt handling and scheduler were truly gifted.
@@stevepreskitt283 Have you tried Haiku OS?
@@stevepreskitt283 I like to imagine an alternate reality in which Apple based its new OS on BeOS instead of NeXTSTEP. (Of course that would never have happened with Jobs at the helm, and it's hard to imagine Apple surviving the 90s without Jobs at the helm).
First, you hit the main points very well. The reason you use a "standard" of anything, is because it reduces (long list), plus if enough of a certain item is made, then someone will make it for a long time as a repair part (see car parts that many manufacturers will make, not OEM, but then, some are with a different name). Whenever building something, using an "off of the shelf" (aka standard), just makes sense and is best practice. Even at the OS level. If people really knew how hard it is to make an OS. Furthermore, doing low level code is pretty F'n hard. Better to use proven APIs, if boot time is a problem, they should plan ahead. Good video, good points. Custom means, not better, just means, special (and hope it doesn't break).
The antidote to the Kraft Singles rage bait....thank you Professor :)
You are welcome! :)
This is the problem with the internet. The uninformed (which, let’s face it…that’s most people) rise above all and some stupid, baseless narrative can catch on like wildfire out of nowhere. The internet gave a voice to the previously voiceless and we had no idea how annoying that would ultimately become…
Dunning-Kruger reigns supreme.
A great response. Also non-zero chance this causes someone to fall out of love with their Lexicon 224 and list it on Reverb.
hah!
I just would like to mention that Blofeld is based on DSP IC destined for home theaters
Yup -- if it does the job, it does the job. :)
Years ago I had a Prophet 600 (one of the very first MIDI synths available) -- it ran on a bone stock Z80 CPU and several MUX/DEMUX chips sending data through digital-to-analog chips to create the control voltages to all those wonderful Curtis CEMxxxx VCOs, VCAs, and VCFs. It worked, and it worked well (until something inside went wacky and it wouldn't tune right - and I traded it in on a Casio CZ1 - another early MIDI synth, which I still have, and use).
So what's the big deal?
Some people seem to be really upset that a manufacturer might choose to use an OEM PCB with a microprocessor on it designed by another manufacturer in their product vs. in-house designing their own PCB with that same microprocessor, which is seen as more honest and pure, or something... ;)
@@Lantertronics OK, but apart from maybe Yamaha (which ironically enough, made synthesizer chips to put in computers), how many companies have chip fabs to create their own chips. And how many of the old synths had those Curtis chips in them? quite a few, I would imagine...
This guitar i bought is made of wood. I have trees in my backyard...I can just cut them down and make my own guitar...what a ripoff
lol exactly. really tired of these people.
The days of purpose-built hardware seem to be gone, which makes sense, since computing power is so abundant that you don't need to build specialized chips for everything. Even early digital synths were already using some off-the-shelf chips. I mean even my Ensoniq Mirage is really just a fancy Commodore 64 with a nice soundcard and a keyboard.
A relative of mine bought a broken Lexicon 224. He had to make a couple minor repairs to two of the internal PCBs, and then he designed his own remote for it in a new style case, and it's flawless now. Quite a cool project. I hate it when people who don't do any design work or building, or anything else at all (sometimes) go an complain about the design choices of designers. Or the price complaints. Like you, I tell people, buy it or don't, enjoy it or don't, but things cost what they cost, and design choices were made for very good reasons (most of the time). Excellent video!
Thanks!
The first level headed take I have seen on this so far. Thank you for that :)
Thank you! :)
_"Urgh! They used off-the-shelf capacitors instead of rolling their own!"_
kornos was also an atom based syth,every sound is produced with the x86 board.still good for every artist
Great video, Dr. L. I really don't see a problem with using OTS components if it allows a quicker/cheaper development process and allows the manufacturer to easily source parts that they'd have to have custom-made otherwise. We're not back in the 80's when Yamaha had to design the OPS/EGS chipset to be able to offer a rudimentary but usable FM/PM synthesis capability. Any one of the $30 Raspberry Pis sitting on my desk has more computational horsepower than a full $250K Synclavier or Fairlight rack system had back then, and SO MUCH of what used to have to be done with discrete hardware can now be done in software, in real time, for very little cost. Is $4000 a lot for an instrument? Sure it is, but consider that the original Prophet-5 cost the same amount of money in 1978, which is over $20K after adjusted for inflation, and the Prophet-X is a much more capable instrument overall, "vintage analog" arguments aside. I will say that I would MUCH rather my Yamaha SY-77 had a simple SBC than the stacks and stacks of bespoke boards that are in it - just changing the display in that beast is a major operation because of everything that has to be removed to get to it. 😀
Except when your synth is completely a VST with a physical interface all you are really paying for is the looks/feel/ergonomics of the synth. Like why not just get a midi controller and spend that money on a nice PC to play as many VSTs as you’d like. I’d rather have an actual working computer if I’m just going to be purchasing a computer (ala the Prophet X). The Hydrasynth is a better value proposition than the X in this regard.
@@rustyshackleford634 But that's NOT what the Prophet X is - it's a true digital/analog hybrid. The X is controlled by a regular PC board, but there's quite a bit of custom circuitry in there (primarily the analog filters and UI hardware) that the synth simply doesn't work without. I totally agree that VSTs are a better value proposition for many (myself among them), but just calling the Prophet X a VST disregards the very real amount of additional stuff that's in the audio path.
@@rustyshackleford634 omg I can't with this argument. some people want to have an actual instrument; who cares if it runs on tubes or an atx motherboard. Yes you're paying for the looks; feel; ergonomics; software dev; engineering; sound of the synth; basically yes you're paying for a synth. and again; the prophet X is a hybrid synth; with analog filters; so the furthest thing from just being a "vst in a box". and even a fully analog synth has a digital part; so basically a little computer; so in fact aren't all analog synth vst's in a box? why are you triggered by the fact this synth uses a part that you can recognizer? Do you know they're not manufacturing every single analog chip they use in their design either?
The Prophet X isn't like the Hydrasynth. It has boatloads of analog filter chips. If that's not important to you, the yes, the Prophet X doesn't have a good value proposition.
Saying "It's just a cheap ITX motherboard" makes no sense. Most of the electronics that goes into synths is cheap. You can buy the SSI filter and oscillator chips that go into the $5k OB-X8 for about $5 each (probably much less if you buy in bulk). Capacitors, op-amps etc. are also dirt cheap.
You aren't paying for the individual components, you're paying so the company can recoup the R&D cost and the cost of all the tooling for setting up for mass production. Of course the more units they sell, the more people there are to share this cost, which is how Behringer are able to keep their prices so low.
I don’t think people buying that synth are worried about the price….
All you need these days is a PC, a midi keyboard, some nice speakers connected to a decent soundcard and some virtual synth software. You could possibly build a system that sounds like an expensive synth for around a thousand quid these days
the people complaining about this give me a real "doesnt understand a thing about hardware, software or general product development" vibe
Roland samplers in the 1980 were also based on a standard MSX computer.
I didn't know that! That's very interesting.
@ I just recently learned that but now it all makes sense that you could connect a monitor, mouse and keyboard to those. Which actually made editing easier.
The PPG digital synths also were all based off of some standard industrial 6809 platform btw.
@@mudi2000a The Buchla 400 is an S-100 bus system (like the Altair or IMSAI 8080).
No, the Roland S series samplers (and W30 workstation) had nothing to do with the MSX architecture. The MSX used an 8 bit Zilog Z80 processor, whereas the S series used processors from the 16 bit Intel 809x family. Earlier Roland products did use the Z80, but nothing about the implementation reflects any influence from the MSX standard.
@ thank you, recently read it and from the looks of the screen it would have been possible. On a vintage synth page I found the info that Roland used the MSX mouse. Maybe that’s where this comes from.
Edit: I checked the service manual and the CPU and it was a 16 bit architecture. That makes more sense than an 8 bit Z-80 architecture.
the idea that off-the-shelf parts are insufficient for an application like this is laughable. these components were sold into a highly-competitive market that's extremely price-sensitive and margins are very tight. every RMA represents a significant loss for the manufacturer - these parts are designed to be fumbled together by sweaty kids in their basements, and just work - so it's in their best interest to maintain a high level of reliability. if Sequential had replicated all the R&D to build their own custom board with the same level of performance and reliability, the price would have been more than double.
Its like industrial machines. They will all have an off the shelf PLC like an Omron Sysmac 120 if your oldschool.
The fundamental philosophy behind technology choices is directly related to purpose. A pc motherboard is a PC motherboard because it facilitates and communicates best for a computing environment with a monitor, keyboard, and mouse. NOT a synth. I have owned synths since 1985, and one pf the bedrock principles of a synth is it needs to work reliably when you turn it on and need to play/perform. For that you need bedrock reliability and something that can withstand transport (banging around). Designing for that purpose delivers a board that facilitates those needs.
In this case Sequential failed to properly perform reliability testing on this particular configuration. Perhaps they didn't mount the motherboard well in a way that reduced it being subject to vibration; perhaps there's a problem with thermals; or a combination of those and other things.
I don't recall Korg customers complaining about motherboards failing in the Kronos, so the use of a motherboard itself isn't inherently the problem.
One annoying thing about hardware running a full OS is the boot time. There's signal generators at work that take like a minute to boot up.
This was exactly my thought when that video first came out. Like do people not realize that everything has a computer in it nowadays lol? And the DSP isnt usually even done in that part of the architecture. Espen kraft is on this kick of shaming hardware companies and users and raking in the views. It comes off to me as major projection for his own vintage synth GAS, from which he’s now repented and been born again. Its just corny as shit
Sequential isn't involved in price gouging. BUT the prophet rev2 was released at €1443 in 2018 and now costs €2485. That's a 72% price increase. If you consider inflation, the price should be around €1800.
Inflation isn't the same across the board though - some things are more affected than others. So the price increase might reflect their actual increase in operating costs - or it might be even worse price gouging. We don't really know.
I guess some people are going to be really pissed that the Korg Kronos used an off-the-shelf Intel atom motherboard running Linux, the newer Akai MPCs used an ARM 32-bit processor on a SOM module running Linux, and Korg's smaller synths such as the Wavestate, Modwave, Opsix, and now the Multi/Poly synths use Raspberry PI compute modules also running a real-time fork of Linux. Edit: Sorry for the repeated info here, I commented before hearing the rest of the video.
I didn't know about the Akai MPCs -- but it doesn't surprise me at all. It makes sense to use OEM modules.
apropos to LMNC's recent video: Peter Zinovieff built his MUSYS system around two 'off-the-shelf' DEC PDP-8's.
Just saw a "Look Mum, No Computer" video where he got to play on Zinovieff's EMS VCS4. He mentioned in that video the rest of the EMS studio was lost in a flood while in storage, presumably including the PDP8s.
I'll need to check out that video! (Love LMNC)
A lot of the later PDP-8s were sold as "OEM" parts to use in bigger systems.
I've got a kit called a "PiDP8" that I need to get around to building. Appropriately enough considering your video's subject, it's a working replica of a PDP8 control panel with a Raspberry Pi inside that runs an emulation of the DEC mini computer itself.
@@chriswareham I've seen it! I'm obsessed with the PDP-8 so I've been tempted to get it, but I have to be realistic and realize I'd probably never get around to building the kit...
1:35 well put together except for the part where he recommends a power supply with the wrong pin out.
Eeeeeek!
Yeah, that's not good...
Not sure how this is different from the 1980s Synclavier that came with a Mac II, apart from that the Mac II wasn't stripped down to the pcb and incorporated in the Synclavier rack.
The Mac was basically just the terminal for the Synclavier which still was running its own OS on its own processor. So the Synclavier was basically "outsourcing" a part of the user interface to the Mac (and admittedly some other functions - but sound generation and sequencing was still completely inside the Synclavier itself).
@@maro_from_germany
Sure but my point is they could have done that in-house and in hardware like Fairlight.
Using a Mac instead seems effectively the same as using a mass-market PC motherboard inside the keyboard. It's just that computers have become so compact that it now fits inside rather than being a separate unit.
@@jonathanhendry9759 The Synclavier used external terminals from the start (first a DEC, then another one that I don't remember right now); and back in the day NED even had a mouse that connected directly to the mainframe. Using a Mac was basically the continuation of this; in large parts it still was that "dumb" terminal that was not much more than a glorious console connecting to a serial port. When using it beyond that you needed to add a bespoke Nubus card from NED.
Wow, isn't this silly!
The "computer inside" thing isn't anything new. It's not even something from this century. In fact, the first synth using a CPU is something which many synthesists would gleefully agree to unpleasant bodily alterations in order to have one. I bring you...
The ARP Chroma. Notice that I put the ARP part in there, because they developed it. But what they developed it around was a logic board powered by an Intel 80186. Yep, that's a "1" there; this chip was intended for use in industrial and early embedded use. Yes, it was used to control the parameters, deal with user preferences, and I would figure it provided the VCO tuning and the like.
I did get to play one of the 50-ish ARP builds, though...and the comparisons are alarming. And I mean "alarming" as in a bizarre drift problem that the CPU was supposed to be preventing...but which wouldn't be fixed even if you hooked it up to Deep Blue!
No, it was the VOICE cards...and NOT the processor...that were at fault. Or more correctly, it was a bunch of people who thought they knew better at Gulbranson's old Hoopeston, Illinois factory, where the manufacturing moved post-ARP. Gulbranson used a flux method that wasn't compatible with the card boards, and instead of the flux going away...naaah, it just wound up contaminating the breadboard stock.
On the first 500 or so.
But the point is this: when this happened to one of the most anticipated synthesizers of the day, one which made a lot of noise about that CPU being THE way to correct previous polysynth reliability issues (like tuning...this was the exact same period when Yamaha dropped the CS80, only for users to discover that it would go out of tune from merely moving it from one room to another!), welllllll...it took some time to figure out that this contamination was the real cause.
And during that period, with frustrated customers, terrified Fender/CBS brass, confused board fabbers out on the prairie, SOMEbody tossed out the unfounded AF idea that...
"THE COMPUTER IS RESPONSIBLE!!!"
And once that got tossed into the info void at that time, the idea that this brand new feature was to blame went up like a mountain of flash paper wads! Sure, it wasn't TRUE...but nobody had any other idea (yet), so yeah! COMPUTRR NOGO INN SINTH!!!
So, remembering the tire fire that resulted, yep, this is just an echo of that. It's like an archetypical memory...like Xenu and the Space Jetliners...but it does seem as if every recurring hiccup of this has that same pearl-clutching going on. C'mon, y'all...just trust your synth designers, OK? They know what they're doing...yes, even Uli, I'll begrudgingly admit...so step back and let 'em breathe. Don't theorize your way out of snagging the cool stuff, either. The tech in this is flying past us at fascinating but disturbing paces, we just need to be more open-minded about what that tech-barrage can do for US.
Had one and sold it after dealing with and constantly hear similar complaints from other buyers.
When it worked it was a cinematic synthesizer. Big sweeping effects and rich tonal sounds. But you can literally buy those things as a VST.
Yeah, the thing is that DSP emulations of analog filters have become very, very good -- so there's not much point to the giant analog board (I realize many disagree with me).
I have a video comparing the Arturia Prophet VS emulation with the original Sequential hardware (my rack unit, which I've since sold), and asked people to tell which is which, and so far no one's guessed much better than random chance (and the guesses are all over the place from person to person).
So although I teach a class on analog circuits for music synthesis because I think analog circuits are fun to tinker with -- as far as practical tools goes, whether something is "real analog" doesn't much matter to me one way or another.
I get the feeling Sequential rushed the reliability testing. Maybe it's a heat issue? Maybe they didn't put in enough stuff to deal with mechanical shocks? Who knows...
Living with and having to maintain a CMI 30A and a Hartmann Neuron I would have say that I'm not so sure that I agree completely... On the other hand keeping the other computers in the studio synths and FX units like the AMS's, PPG 2.3 and Waveterm and all of the vintage synths like the Prophet's, Oberheims, Synergy and Synthex are "relatively" easy (except when they aren't). I guess time will tell!
I got the impression that the Hartmann Neuron came out before it was really ready.
Oooooh... tell me more about the 30A!
My only complain about the Synergy is the way the daughterboards hook to the motherboard by wire wrap pins jammed into IC sockets.
ZOMG you have a Waveterm!?!?!? Whoa!
That being said I do like the idea of using the smaller systems like Raspberry Pi in synths as in the future they will be easier to replace with later iterations of micro. Full PC motherboards can be painful to replace a decade down the track and if these machines can't last that long then they are a bit of a fail given the longevity of the Z80 and 68xxx board so far.
@@Tharsis879 This got me thinking -- it would help if manufacturers would embrace more open standards. Not that they necessarily have to give away all their source code -- but they could publish interface specs that would let someone swap in a newer motherboard that would stay compatible with the rest of the hardware, and enough detail on how the software works that someone could cobble together drivers for a newer OS. (I'm just spitballing here, I realize there's problems with this idea).
I suppose Mr. Kraft was disappointed it doesn’t run on crystals and unicorn horns, or at least a small troll running on a hamster wheel.
HAH!
Had a Memorymoog Plus that struggled with using what was basically an Z80 micro Single Board Computer and the supporting boards, digital, analog and voice cards all wired up looking like Chef-Boyardee and Bob Moog had a baby synthesizer. It was semi-reliable, but nobody cared because it would melt your speakers.
We will cope with knowing that there's way more digital than analog under the hood in today's synths, but we love that knobs and sliders are back and we don't have to be victimized by Microsoft directly, but I do love Cherry Audio VA synths.
I can understand some of your arguments, but comparing a compute module or some other industrial grade motherboard with solid caps and no moving parts to this dirt-cheap ASRock one doesn't work. ASRock is known for putting bad components in, caps can go bad pretty quickly and if your 4K synth fails because of this, this is not ok at all.
ok but does it actually translate in reality? do people have more issue with the prophet X than other synths; or is it just speculation?
this isn't an industrial application. there aren't lives at stake here. it's not getting exposed to temperature or shock extremes. this machine is making bleeps & bloops. a consumer-grade solution is plenty. the reliability and ease of replacement of these components far outweighs the cost that a custom/industrial solution would incur.
Asrock aren't really an outlier when it comes to consumer motherboard quality, or consumer electronics in general. I've had synths with completely custom boards go bad due to cap leaking, it's not like the fail rate will be substantially worse than literally any other synth built in the last 30-40 years.
@@Theinvalidmusic yeah, 90's caps were 90's caps. it doesn't matter what board they're in consumer or custom. if they leak, they leak.
The complaining about a Sequential synth being a computer in a keyboard case is hilarious given the history.
Hah! I hadn't thought about that. But yeah -- the same "Sequential Circuits" came from somewhere. ;)
Great points, talking about cost vs value. One point you can potentially add is that the Linux KERNEL is not an Operating System on its own. It just creates a standard interface for applications to use to access hardware. It likely doesn't have a full "userland" and probably takes just seconds to do Inital Program Load.
Nothing wrong with a good Pi. That's the compute behind the Zoxnoxious synth, much similar in concept to the Prophet with an analog audio path and digital control.
I really like your videos, good depth of discussion. Your filter videos were really useful in a recent project!
Edit: oh nevermind, should teach me to watch the whole video before commenting.... The legendary Korg Kronos synth had a pc motherboard too. And nobody ever complained about it....
Yeah, I mention that along with the Hartmann Neuron (although plenty of people complained about that... but I think the Neuron was released before it was ready).
I think you can still complain about the 3 minute boot times and the fact that m.2 SATA will be hard to find in the future
heh ;)
@@raul0ca Indeed, that's a big advantage of a solution built around a microcontroller with more bare-metal programming. Running a full Linux kernel comes with a lot of overhead.
As did its much more expensive predecessor, the Oasys. You could've paid up to 10k for a Pentium 4, but you did also get a beast of a workstation with an absolutely incredible interface.
Great video - I've always been frustrated by this "ugh it's just a cheap PC?" line of reasoning. I 100% put more value well thought out architecture and UX. Yes a lot of what makes up a Hydrasynth could be a VST, but the UX and UI are absolutely worth paying for.
I've been saying for ages that the only real point of physical hardware is the user interface experience.
Well... there's other things, like the physical hardware probably isn't going to make you log in to some license server that will go down in five years...
Thank you thank you thank you.
You're welcome you're welcome you're welcome!
Chill. If it was a Roland it wouldn't even have an analog board in there. The PC motherboard would just run a VST ;-)))
Yeah, I don't see why Sequential got so much flack for this when so many other manufacturers have done the same thing.
And here I am wondering why thae eggs that were about a dollar 4 years ago are now almost 5 dollars. I am clearly thinking about the wrong stuff.
Oy vey, Espen's ragebait turn led me unsubscribe. I know it's just business for him, but there's enough wrongheaded misery in this world already.
He quit his day job to go full time as a UA-camr, and I think he's struggled to maintain the engagement he used to have. The clickbait video titles and provocative content are probably an attempt to boost his views. A shame, as I loved his older content and even donated a few bits of 80s music gear to him.
I agree that it's okay, that it isn't an inherent scam. But it is also plain funny
You would be suprised how well a phone oscillates a waveform for basic synthesis, ......................oscilab......works on android 4-13
I’d love for you to tackle Espen’s “VST in a box” claim about synths like the Prophet 5 rev 4 and Ob-X8 because of their use of digital envelope and voice control with the vintage knob.
Oh goodness -- yeah... I thought about wading into that a while back but decided not to at the time. But... yeah.
By Espen’s measure, the Oberheim Xpander is a VST in a box.
Some 80s vintage synths with software envelopes have reputations for having "slow" envelopes (or odd curves) but I don't think people realize that's likely due to the fact that they were being done on a single 1-4mhz 8-bit microcontroller that was responsible for 5 other things on the synth.
@@ddqd123 But that doesn't change the fact that the envelopes were (sometimes) slow. I clearly felt this when comparing a Prophet 5 to a Prophet T8.
Without a doubt Sequential could also have used hardware envelopes (CEMs) on the T8 (like they did on the P5) - but they didn't.
And to be really picky about this: The T8 used a 6MHz 16-bit processor just for envelopes, LFOs and CVs - and a Z-80 for all the other stuff (scanning controls, keyboard, MIDI and so on). I believe they simply needed to gather experience in writing this stuff - or they simply accepted this (inferior) version as sufficient.
The funniest part about that video was him ranting about cheap digital implementation on analog poly synth while a six track and 106 we’re behind him. Both of which are heavily reliant on digital control. I’m sure he will be mad when he opens up a vintage p5 and an obx to find that they both used off the shelf z80s.
You're damned if you do, you're damned if you don't. Apple is despised for the exact opposite reason. They spare no expense in giving the end user a exceptional experience. They would rather incur the significant expense of developing their own software and hardware than license them from a vendor. They've even gone as far as designing their own microprocessor, which is unheard of, especially for for consumer market. Yet, many complain that Apple products are overpriced, proprietary, and difficult to repair.
Thank you.
You're welcome!
Good points and perspective on this topic, Aaron. Keep up the great content as always! 👍
Thank you for your kind words!
I heard they are thinking of putting computers in DAW's next.
is this really a controversy?
(ok reading the comments apparently it is)
Yeah -- and check out the comments on Espen Kraft's recent videos...
Finally, a view on this I actually agree with. I've been a little annoyed at Espen's recent videos, especially because I know what it takes to make the devices he keeps calling "scams". I was particularly annoyed by the implication that Behringer devices are the only "fair" devices, when in my opinion they're as far from it as possible.
If you replace where Espen says "scam" with something meaning "a value proposition that many may not find appealing," then he has a point. But the term "scam" implies deliberate deception, which is unfair.
The thing about Behringer is that no one can really compete with their economies of scale. A company like Groove Synthesis buys X number of resistors for a couple versions of basically the same product, whereas Behringer can buy many many many more of that spread over hundreds of products.
Agreed, I just unfollowed him as I feel that there is a degree of clickbait and sensationalism there sometimes that is unhelpful.
@@Tharsis879 Which is sad, because his content has historically been pretty interesting. It's just gone a bit over the top lately.
@@Lantertronics "over the top" is a good description, I got tired of it.
I disagree. The computer chips are a way of short cutting work and making your synths cheaper. This is absolutely INSANE when you think about the money they cost. The parts they use now massively reduce repairability, they end up simpler in design (if they had to use actual discrete circuitry to achieve what they do with emulation the synth would be even more complex, not less), and half the time you just end up with problem. Synths are getting buggier and buggier and it’s because of software issues.
I am less concerned about whats inside my boxes as I am on how well they work and are built. Fortunately, I have not ever had to open one. Knock, knock on wood.🤔🎶🎹🎶🎸🎶Play On
its turtles all the way down
the tc system 6000 ran on windows NT, iirc. Guess that was just e-waste, too? ;)
I didnt get Espens post either, most of the all the modern - synths ive seen use PC boards? I do think manufacturers are not supporting their ( expensive) older instruments is worth getting pissed about ...but thats a different issue! As an aside, When it come to value for money, Espen promotes the retroactiv midi controllers- they are great piece of kit but their price ( for their limited specialised use) i think is a total gouge😮, so i really think hes damaged his credibility on that front......
The Retroactiv prices make sense from the standpoint that they're not making many of them -- they have a very limited market, so they cannot exploit economies of scale.
@Lantertronics yes I appreciate that, but they also require almost no development cost ( unlike synth software) & have minimal cost components ( no experience is etc)- so personally I see that as overpriced. But like everything (& something I think Espen chooses to ignore with his prophet vid) it subjective & regardless of components etc people will pay more for a sequential synth than a korg etc because they can see the value in it..😉
I'm sad they stopped putting synth chips in my videogame consoles. Ooooh, fancy samples and cd quality! Please.
Give me some of that tasty Yamaha or SID instead.
;)
a controversy? really?
Heh, check out the comments section on Espen Kraft's latest videos.
Good video. Interesting how the 3 people who disagreed with you in the comments sound underage or very stupid
I'd prefer to just think of them as... profoundly misinformed. ;)
I'm extremely bothered and triggered that prophet X did not invent their own real-time operating system and used Linux instead.
Agreed, lazy devs, bare minimum effort imo.
A custom toolchain, OS, hardware platform and system to manage I/O isn't so much to ask.
They didn't use Linux - it runs under QNX, which predates Linux by about a decade.
They didn't use Linux. They used QNX.
I remember reading about QNX -- it was in Byte magazine, I think, back when Byte was interesting.
@@Lantertronics
It's an interesting OS, and a very popular one for embedded systems, but it sure isn't end-user friendly.
So long as it's replaceable. A PC mobo should be replaceable with any mobo from that CPU generation, so an AM4 board if there's an AMD chip, or the appropriate Intel board, etc. The reason analog hardware is great is that it's not proprietary, anyone who understands analog can repair it, indefinitely, with common parts. That's not true with boards. If they've done proprietary fuckery with this board, then when it fails, either you pay them something silly for the replacement, or worse, they just don't make one, and you're hosed, your $4000 synth is junk now. This UA-cam random did not do the sort of long term reliability testing to verify that, and doesn't get to tell us what's okay. It's not your money. We're not your children.
Alas, it looks like the ease of repairability of analog gear is only true of older gear. All the PCB photos of newer analog gear I'm seeing is all surface mount that's awfully hard to repair unless you have a lot of experience and the right gear.
You seem to have misunderstood my point (admitted, the title alone does not make this entirely clear): I was addressing the question of whether Sequential (or any other manufacturer) is inherently "pulling a fast one" by using an OEM motherboard, which is what a lot of people were claiming. Whether Sequential messed up in the details of *this particular* instance is another issue.
As I note at the 6:06 mark, Sequential clearly messed up by doing insufficient reliability testing, and for *that* they should wear the Cone of Shame. Maybe they didn't do a sufficiently good job in mounting the motherboard so that it could withstand the strain of being moved (it would be interesting to know if people who gigged with their Prophet X had more trouble than people who just used it in the studio). Or, maybe there's a heat dissipation problem. Or maybe *that particular* motherboard is just junk and they should have chose a different one.
@@Lantertronics And let's not forget the issue of obsolete/exceptionally hard to find parts. If certain SSM chips fail some gear can quickly turn into a doorstop...
@@maro_from_germany Fortunately a lot of those chips have been reissued. But, indeed, not all!
This is a result of social media. People fire off these useless remarks like they are the law when they have 0 clue at all. Same people probably wonder why others feed their dogs when you can just leave them outside to hunt for themselves.
Running-Kruger reigns supreme!
@@Lantertronics HAHAHAHAHA Wow that's funny.
Whilst they literally record, perform and market music to an audience entirely comprised of fuckin computers with a consumer grade motherboard 🤣💀🤣💀🤷🏻♂️
I'm extraordinarily bothered and triggered that the profit ex did not invent and develop its own op-amps and resistors.
I’d rather just spend 2K to buy a high end PC and use the leftover to grab a nice midi controller. At least that way I won’t have to deal with firmware issues, component breakdown, and I can play video games too. That’s more than Sequentials new computer (Prophet X) can do.
One of the main selling point of the Prophet X is the analog filters, which is important to some people, and you won't get from just your own PC. If that's not important to you, then yeah the Prophet X doesn't have a good value proposition.
If you want a synth that can play videogames you should buy a minilogue, that one has breakout
@@greentoaster Hah, I didn't know that! That's funny!
You can make Linux boot quicker by stripping it down
True. (Although as an aside it turns out the Prophet X uses QNX, another Unix-type OS focused on real-time performance).
@Lantertronics interesting. I know next to nothing about QNX
@@tolkienfan1972 I know next to next to nothing about QNX. I just remember reading about it in Byte magazine when I was a teenager (I think).
Finally, some say what had to be said.
I have a PX and I'm glad it has a PC motherboard and a Intel SSD and it sounds great!
Kudos to you @lantertronics
I recall the first generation of Sequential synths, when they did have their own microprocessor systems and firmware designed and coded by Dave Smith - so it is a little sad, and just the current practice, to see this outsourced.
Why would you design your own MCU when there is an off-the-self part that does what you want? It's not commercially viable. Maybe back in the day D.Smith did not have a part that would do what he wanted and then design its own MCU.
It's also cool to see the hand-laid PCB designs in the older Prophets, with tinned traces instead of today's ubiquitous green solder mask. No one uses curved traces anymore. 🙂
The firmware was designed and coded this way because the processors were slow and ROM and RAM was expensive. Still the Prophet 5 used a stock Z80. Dave Smith probably could have put a development board into the Prophet but I believe those were rather expensive at the time (and also had additional functionality that wasn't needed).
This makes no sense at all from an engineering viewpoint or a marketing viewpoint. Early synths had didn't have bespoke CPU boards with Z80s and ROMs and RAM because the designers wanted to show off their engineering prowess; they had them because it was the best way to hit a price point at the time. If they could have sold the synth cheaper by using an off-the-shelf SBC (like Lexicon did in their 224), they would have.
The original Prophet 5 cost about what you can get one of these for.
That was 1978, $3395.00. Now it would be $16431.80. Do the math.
Everything is software for last 25 years.
I think it's a matter of proportion. Everything has *some* software in it.
VST's have come full circle.
My 224 is still rocking 🤙🤙
There is always something less appealing about sound being created via an interrupt cue than done in analogue electronics.
I bet most people would not make the difference in a blind A/B test. The people who made the vongon replay first designed a fully analog synth. Then they tried making a digital counterpart with the same "analog" style sound and ended up liking it better so went with that. I find this synth sounds lovely; but they're getting crap for it (and yes it's not a cheap synth). I like analog and it does sound better in certain cases but it really should not matter in the end. The sound and experience of using the thing is what matters.
@@valdir7426 I think thats possible, a similar test has been done with violins costing millions of pounds vs a modern built ones, and they cannot tell. But ! there is not doubt some psychological element that would favour the million pound antique violins. I preferred digital synths because of their accurate reliability, but have always had a sense something is lesser somewhere, and i guess its the virtually imperceptible, but present artifices of the interrupt cue.
I have another video where I compare the original Sequential Prophet VS with the Arturia softsynth version; I randomize which I played first in each pair, and asked people to tell me which I thought. So far no one has guessed significantly more than what you'd get by random chance.
@@Lantertronics I wonder if its possible to sample at a very high rate and thus actually see the slight imperfections of the interrupt queue vs a c continuous analogue signal. At some level it must be demonstrable. This is my only argument really, that anything running on a personal computer is subject to wildly different renderings, at any given moment. its way faster than a 16 or 32 bit sample, practically inaudible, but its there.
@@JackNicklauson Well, considering everyone is ultimately listening to an audio file through a DAC being played by a computer, I'm not sure it matters. ;)
Well, that’s nothing. Yamaha’s flagship keyboards run on off the shelf arm processors.
It took a computer to grant us analog poly. It takes a pc to give such large data rates in an affordable package. Ergo, shutup and play!
Espen has gone full troll / controversial rage bait with his content. I do love my prophet x and appreciate this video, but I think it's best if we all just ignore him at this point.
Oh hai! I love your channel.
The dark side of UA-cam Algorithm is trying hard to convince me to make more content like this (almost 6K views in 3 days, which is a lot for my channel -- I mostly make math-heavy lecture videos about analog electronics, which admittedly I don't expect to be that attractive to a general audience).
Why doesn't the profit X use a quantum computer or at least maybe a giant parallel NUMA cluster. I am triggered and offended that the prophet X does not use at least 200,000 Watts.
Are you for real, this blatantly straw-manning?
What straw-man argument do you believe I have engaged in?
Those are quotes from Espen's comment section.
Seriously, go look at the comments on the original video. It's non-stop people complaining that Sequential is pulling some kind of "fast one" by using an OEM motherboard.
Or for that matter, look at some of the comments on this very video.
i'll happily wit for the behringer clone of the PX. suckers
I doubt this would be on Behringer's list of things to clone.
Strongly disagree. He’s completely right about the shite stuff synth companies are coming out with.
With the competition from the behemoths it's a life or death proposition for many of them, if they can't make enough spare money they can't recoup their dev costs and will either have to just make the same product forever or close their doors.
The disparity between Behringer and their competitors is hard to overestimate, even Roland's outsourced production can't compete in terms of worker salaries and bespoke factories.
If you believe "natural selection should run its course" and most companies deserve to close, then fine, but I personally appreciate a more varied market
What exactly are you disagreeing with? I was specifically making a point about about the use of OEM parts in a synth and whether that made sense or not from an engineering standpoint. Whether synth companies are coming out with shite stuff more broadly is a different discussion.
You got it wrong. This isnt about making the synth more expensive due to them RNDing a computer, is about them A) Beign upfront right in the box "Powered by Raspberry Pi" or "Powered by Intel", some people not care but others would be interested, many do electronics, computers and stuff. Why hide it and/or not disclose it? and B) If they are running off shelf parts, why not pass THE SAVINGS to the customer? Why you are so pro business? It IS not only a cheap computer and a mostly barren SMT board, running off a Prophex looking MIDI controller, HARDLY $4k, it is also very lazy design throwing in there an ITX board and calling it day. Put it together DIY? My brother in Christ, plenty people already make standalone MIDI controllers with VSTs inside running off ITX boards, raspberry Pis, FPGAs... you name it. DJs have been doing "laptop less" setups by hiding the computer in the case long before standalones were a thing. RnD costs? Sure you can design the UI but the backend will be done in India for a few rupees and hour, of course, based on some copy/pasted code from Github.
but why does it matter? do you expect manufacturers to list every single chip and components they use? they should make such lists available for people who want to repair their stuff but just because people are familiar with intel and raspberry doesn't make it relevant to advertise. they could as well put "powered by a texas instrument converter" but no one would care. That's an interesting trivia but no it's not something people SHOULD absolutely know. Also: not using "of the shelf" part (sorta for the compute module) would increase the cost of r&d and manufacturing, not the other way around. And finally you can look up the specs of the prophet X and it's not even a digital synth; it's a hybrid synth with fully analog filters; so your point about it being "a vst in a box with a controller" doesn't even make any sense. the motherboard is only a small part of what makes it a synth. you really don't have a clue what this thing is.
"pass THE SAVINGS to the customer" the savings were already passed to the customer. you obviously don't understand how marketing and product pricing works.
I don't quite remember an "Powered by Intel" sticker on the Polysix (and many other synths) either.
There's a tremendous amount of incorrect information and misunderstanding here. How in the world can you look at that massive analog board with all those filter chips and call it a "mostly barren SMT board?" Did you watch the full video? Did you miss the section at 2:52? Sure, try DIYing a Prophet X and see how it goes. Go write the Verilog for the FPGA that handles motherboard-to-analog-board communication.
Maybe the value proposition of the Prophet X doesn't appeal to you, and that's fine.