I don't know why, but I really like the repair videos. And I feel like the 64 is in that perfect space of being complicated enough that the machine can do interesting things (sound, graphics), but simple enough that you can actually diagnose it by hand.
I think that’s a great summation, I had a C64 in 1982-3, it was tons of fun, imagine typing in basic game listings from magazines, character by character and then debugging, and then you could save it to floppy if you liked the game! I would love to fix these myself! It was definitely one of the first really good home machines.
I think it is incredible how you manage to pump out two videos (both channels combined) per week, despite working full time. And your videos are not some random garbage just to satisfy the UA-cam algorithm but very well made and always worth watching, both for their content and also your very enjoyable style of persentation. (I think there isn't a single video listed on either of your channels that I haven't watched..😅) Big thanks from Haribo country!
Great video Adrian, love the split screen format. For those wondering, the part number for the Hakko drill cleaning tool is B1303. I've had it for a while, and it really has saved me from tossing out expensive tips that I thought were ruined. The spring tool that comes with the gun just wasn't working for them.
Adrian: When in doubt about making a C=64 repair video, make a C=64 repair video. :) We love seeing the particular symptoms of each machine and thinking along about the possible causes. It's like trying to solve a mystery before Holmes or Poirot.
Not filler, it's why i watch your videos. Fixing things is it's own reward. Taking a broken thing and returning it to usable condition is just awesome.
I enjoyed the video. You could look at it as not filler, but a training film. Trouble shooting, diagnosis, repair and testing the repair. It is a great educational video.
You can never do too many C64 repair videos! Lots of people have them and they are prone to so many failure modes. And, you're also doing a very good job of imparting general troubleshooting knowledge every time.
Even if I’m not understanding what you’re talking about because it’s far over my head, I love going through the troubleshooting process with you. Will not get sick of c64 repairs!
Yep, that was definitely an (audible) issue with the low pass filter on that SID. But I would leave it in there. As long as the envelope & tone generators, amplitude & ring modulator and the bandpass filter work it should be ok to listen to.. Those pops give it some personality as you get with a vinyl record ;-) I wouldn't be surprised if that 'fault' has been in there from the factory as not all SID were created equal to begin with and the way the low pass filter behaves is the most variable component in my experience ;-)
In fact my original C64 sounded just like that when I got it back in 1984. After a couple years I wrecked my SID when I tried to hook the audio output up to a phone line, and I got a replacement chip that sounds more like what we're used to hearing. I never liked it as much as my original "crackly underwater filter" SID (probably because I had become so used to it). I think that sound is specific to a certain batch of 6581 SIDs, and it's not broken at all. I suggest Adrian listen grab the program "Synth Sample" or "Swinth" - a program with a handful of early SID demos. Listen to "Stationary Ark" with that SID. It is glorious. I'd love to find a SID that sounds like that.
Adrian, definitely appreciate these videos - no matter how many you make, I'll watch them. Your channel is one of the ones I've watched over the last couple of years that helped spark my interest in repairing retro electronics. Your various troubleshooting and repair videos have helped me immensely with the information and techniques (soldering, troubleshooting, etc...) that you demonstrate. With what I've learned, I've built a 'new' C64 using a Sixty-Clone board and was able to use an oscilloscope to diagnose an issue I had in the timing circuit for the VIC-II. I've also managed to assemble various project boards for the C64 such as a switch-less kernal switcher and a modern VIC-II timer circuit replacement board. Thanks for sharing your knowledge and expertise so willingly.
I'm not really into the C64 HOWEVER I will happily watch you repair them till the cows come home! Just love hearing your thought process about how you go about fixing issues, it's just great communication and a total pleasure. I like to think your little tidbits it'll come in handy when I start fixing some of my broken machines that I have piling up (Electron, SMS2 and a 48k ZX), so keep me coming!
Thanks for fixing all those machines all the time. I love watching each single one of the repair videos. Even if it is the same error again or another retrobright. Just as some people enjoy watching the same sport again and again, i love to see this again and again
I think I'll always find repair videos satisfying to watch, no matter how easy the diagnosis and fix are. The process of zeroing in on the issue using such common tools and a bit of logic is an interesting and useful skill, and one I'd like to learn myself. I've composite modded my father's old TI-99 and I'd like to get it working again someday (it's currently just displaying a solid, light blue screen)
Back in 1986, I designed an IC tester that would test all logic ICs of the time that was programmed into it. In circuit, while powered. All the user had to do, was to remove the crystal of the CPU or any clocking source, and type in the part number after connecting the ribbon dip connector to the chip under test. The tester would pulse the inputs and latch in the outputs and compare it with the truth table in memory. It was built around a Z80A. While testing, I saw a new similar product in Popular electronics from B&K Precision. I became so upset, that I never worked on it again. B&K pulled it off the market. It was like they came out with it, just to shit on my dreams and then left the room!!
I still remember playing Commodore Crunch and Chopper Command as a kid on my VIC20. I remember having to load some games off the tape drive and instruct the computer to load and run the software while others were on cartridges. Good times.
The repair vids are some of the best content, they are always appreciated. I could not have fixed my own machine if I hadn't watched you fix all-together too many machines on here first !! I am never going to get "tired" of yet another C64 on the bench .. heck, you could just talk about a C64 in bits on the bench and strip it for parts and I would still watch and enjoy it.
13:40 - The chip tester pro is one of those things that seems like an amazing thing, but it's far less useful in general than people realize. It won't tell you definitively if a chip is good, but it will only tell you if it's definitively bad. Adrian (and many others) have run into this many times. A lot of these older machines rely on the chip's electrical characteristics and each circuit and the retro chip tester just isn't going to exercise the chip the same way the target hardware will. So a chip may test good just fine in it, but bad in context of the circuit. It's far cheaper to just have a cache of common 244/245s, etc... on hand than it is for the retro chip test's price. If you've gone to the trouble of swapping/removing the chip in the first place, just swap in a known good one to see if it works. It takes less time. The only place I've found the retro chip tester to be appropriate is if you have a pile of chips and want to weed out the bad ones, or perhaps see if a ROM image is good (provided the retro chip tester knows what it is).
A note on the ram tests of the dead test and 'official' diagnostics... The dead test only tests a small part of the system ram, but in such a way that it doesn't depend on the zeropage and stack etc working. The diagnostics tests all ram, but requires you first fixed any issues reported by the dead test cart, because diagnostics heavily relies on the zeropage and stack and some other bits of ram working properly to do the complete ram test (and do a more thorough test). For that matter, the diagnostics copies its ram test to ram, and runs it from ram. Consequently, if there are errors in the ram the diagnostics relies on to work, the output of the diagnostics will often be wrong, or it will just crash. So the proper way to do this is to always first run the dead test cart, and fix any ram, pla etc errors it reports, and only after it no longer reports errors, run the full diagnostics to test the remainder of the machine. Starting out with just running diagnostics may give the correct result, or not, there is no way of knowing without first having ensured the things it depends on are functioning properly.
Thank you! I am glad I am not the only one to comment on his "deadtest" comments. I have found since the late eighties that deadtest is the first one to go to if you suspect ram problems. The first test tests each 64k chip and then reports the bad ram by flashing the chip number. It doesn't use any ram (including zero page) until rams tests ok. It requires the vic chip to work (vic pin 17 generates the ø0 for the 6510), PLA mostly to work (otherwise no cartridge would work). The bad part is it stops at the first bad ram and is dependent on ram support chips to work correctly. Just so you know, the first 7 bits are tested and then falls through to show one flash for bit 7. So if the first 7 bits are ok and even if the 8th bit is ok but something else is wrong you get one flash over and over.
@@johnsaller2481 I independently came up with a highly similar bit of code to do a 'ram test' without using ram. Only being able to report the first bit error is a simple consequence of not being able to use ram, and the relatively limited bitwise operations available on a 6510 cpu. My code is intended for the C128, but ended up using an almost identical approach.
The thing about C64 repairs is that every single one is unique in some way. We want to watch that journey of diagnosis, planning, action, success or failure, and followup. Everything is content. Record it all! (Honestly, just give me any reason at all to listen to that intro music - so good!)
Great video. We love that you care for the computers. That's why we keep coming back. 👍 Side note: that RAM chip was saying "It's not me man. It's the NEC chips next to me. It's not me." 😂
Repair videos never get old. Everyone you repair another machine you always tend to learn something new. I enjoyed every repair video over the past 10 years.
Thanks for the pointer about that drill from Hakko - I have the same desoldering gun (and it's totally great) but recently destroyed a tip when I couldn't unblock it. Ended up breaking off part of the plating by trying to force the blunt wire cleaner it comes with through the hole using way too much force 🤦♂️
Did you notice both times the diagnostic test froze that there was random " ( " instead of " " . If you looked up the Petscii character codes (converted to binary) for both open bracket and space (ie. $20hex ( for " " ) = 0010 0000 binary, and $28hex (for " ( " ) = 0010 1000), you would see the only difference is $8hex (ie. 1000 binary) which is Data bit # 3. From this information you can either check the C64 circuit diagram to determine which RAM is used for Data bit 3 or from the C64 Dead Test manual looking at the RAM error flash code chart again , you would see that this also points to U10 being faulty. Note: it's not uncommon when you have a memory fault(s) for random characters to appear on screen, and using the Petscii character codes you can determine the offending data bit or bits (if you see more than one random character displayed) and then lookup the offending RAM IC(s). There's always an exception to this rule (eg, bad MUX IC ) but will still help point you in the right direction.
@@matthewtempest7756 on the SX-64 the 74LS373 definitely can affect the displayed characters , but personally haven't seen it produce what you are describing ... what I have seen is UB2 (that's the SX-64 circuit reference for U26 as on a C64 board) corrupting several / all of the displayed characters. Have you already gone down the Petscii Character code path? Which would suggest if you see an " & " character instead of a space on screen that 2 x Data bits are affected (ie. Data bits # D1 and D2) , which point to RAM IC's UA7 & UB6. On a C64 if the RAM was not socketed I might suggest trying to piggyback the suspect RAM IC(s) , but on the SX-64 access is poor and if you don't have additional diagnostic tools to assist it isolating / confirming the cause of the fault, then just replace the (cheap , readily available) Suspect IC's and test again hopefully confirming the diagnosis.
I have been using the TL866A/minipro for testing some dozen logic chips and other ICs recently, and while the supported range of IC of course is not as huge as with dedicated testers, the testing results where 100% correct, yet! No false positives or negatives.
Thanks AB (and Patreon's :) for another C64 repair video! I love these as that's what I used to do back in the shop I worked on. Although back in the 80's we usually just replaced the board and sent the defective one back to C=ommodore...specially during the VIC chip "twinkle" era.
Just amazed that you put out 2 videos per week! I so miss the behind the scenes content but I had to downgrade my Patreon tier because money is tight at the moment with being ill again.
I used to repair C=64, Vic 20 and all Commodore products professionally while they were still under warranty. And I have the burned finger tips to prove it. I didn't have any decent temperature sensing devices until years later, and so it was the finger that was used to find hot RAM chips. Went through a lot of sockets repairing these units. I even repaired Commodore PCs when they came out....at component level. That stopped after the '286. Also serviced Atari products.
In 1983 MOS Technology used a defective process in chip production: too much boron in the passivation layer, ironically the outer coating intended to protect the chip. Thus, the horrible long-term reliability of many chips in 1983 and 1984 machines.
Wait... are you saying "MT" ram chips are MOS Technology, not Micron? (never knew MOS made DRAM?) All this time I've been trying to figure out why Adrian is always sh**ing on MT ram chips, when Micron was always the most reliable ram brand (perhaps 2nd to Samsung) the 20+ years I had a computer store.
@@cliffshockley4406 No, MOS made ROMs, but not RAMs, IIRC. This is just referring to the issues with the 7400 series logic chips. MT is definitely referring to Micron Technology.
@@cliffshockley4406 MT = Micron, who had their own problems. However... Commodore bought Frontier Semiconductor. Everything made in the Frontier Semiconductor fab in 1983 and 1984 suffers from the boron problem. Well known chips that fail because of the boron problem are the PLA, 8501 CPU and the MOS 74xx clones. Chips made in the Norristown fab during those years are reliable.
@@danielmantione Just so surprising to me, as in later years, I can't recall ever having defective Micron MT ram chips, though I do suppose they later had their Spektec subsidiary which they used to sell their "substandard" (my interpretation, don't sue me Spektec) ram, so maybe they just got better at testing for ram that was off-spec.
Defective to you and me. A literal gold mine for Commodore and MOS when the things failed after a few years and people had to buy new computers, or get new parts. :)
Love the C64 repair videos. If I learn any tiny thing Im happy. Also think of the library of repair information you building up here on the channel. Love it.
I always love your repair videos Adrian, you always explain in such a clear, fun and educative way. Everytime you post a repair video I get a nostalgic trip down memory lane :) I am looking forward to more of them.
I really enjoy all Commodore repair videos! I hope that's not boring for YOU, though. Please enjoying repairing and recording what YOU like; we're here to watch. Aloha!
Excellent repair as always Adrian! Why are RAM tests so bad on the C64? I've also had all sorts of inaccurate results with them. Is it really not possible to detect them reliably? Maybe I'm going to have to write a C64 diagnostics ROM like I did for the Amstrad CPC 😃
In this case it failed during the stack test on a diagnostic that actually uses it. I suspect they probably are more reliable if the error occurs in general purpose ram
The ram test is just too simple to be that useful .. it likely only quickly writes a couple patters to some pages and calls it a day. That’s not thorough enough but it’s fast.
Noal that would be awesome to be able to test the ram fully .Like you ive loads of different results . As i believe the tests we have now only write to small random places ..
The first thing I do every time is to push all the chips well into their sockets because they can pop out very easily, especially if someone has used cheap sockets. Many times something worked straight after pushing the chips in well.
Might be interesting to build a program that analyzes the output of the SID over a microphone port. Could do a waveform comparison between a known good 8-bit dance party and a test one.
By far my favorite repair channel! Do you think one day you could do a video on how you set up your Easy Flash Cartridge? I bought one and a bit confused on it lol
SIDs are a bit clicky anyway, that's how they originally did sample playback. Filters always vary, pro-synths with analogue filters have calibration but the SID never did.
I have seen the com port test fail where a trace had corroded between the DB9 port and the PIA chip ended up running a bodge wire to fix. Quick way to test is see if a joystick has lost a function :)
Take a look at acetylene torch cleaning bit sets. They're cheap and have quite a few different size bits on them for cleaning torch tip holes. They also work pretty well for the older Hakko models that I've used.
Ya know Adrian, I don't have nor ever used a C64's, or any game mechanics as I don't play games but I watch your videos anyway. Anyone that survived the IRQ and Memory wars of the 90's computers and games an still like to play games, I take my hat off to ya. Me, it away all the fun an ruined my game playing as a computer tech years ago sad to say... :-( The reason I do watch, is I do fix alot of my own electronics as a hobby, if it is still broken when I'm done, well it was broken to start with... LOL Also mostly because I just plain enjoy your videos and seeing what it takes and how you go about fixing them. Believe it or not I have fixed several thing just from watching your videos, with that I say a BIG Thank YOU! Take care! LLAP 🖖
Need an interface from the ZIF socket on the "retro chip tester" that you could place on a RAM IC while it's in situ. Akin to how stacked chips came about but this one with a ribbon cable leading to the ZIF socket :D
Hi Adrian, did you notice the message in the diagnostics test that says "4066 @ U28 BAD"? This might have something to do with the SID popping like that. Haven't got the schematics at hand but I am pretty sure there's a 4066 involved somewhere around the SID circuitry. Might be worth taking a look at that.
Well, this one is going to be an interesting set of videos, it would be nice if you could also make some videos about deeper diagnostics and repairs, maybe even upgrades?
Arent the SID filters dependant on some of the capacitors next to the SID? I have a vague recollection of recapping that area solving some filter issues.
Most of my MT ram was bad. I sent the computer over to Ray Carlsen and he replaced it with a "modern equivalent". Now it works great and I do not foresee any memory issues again on this machine.
SID chip sounds fine… it’s a tune made for 8580 chips, it will sound clicky on a 6581. Better to test 6581 chips with some other music, I’d recommend Evolver by Jeff…
Except that Adrian has listened to that tune on a hundred 6581 SIDs, and by extension so have his viewers! I agree with Adrian that the filter sounds unusual for a 6581, but I don't think it's broken - I think it is a dark, warbly filter that is characteristic of a specific batch of 6581 SIDs.
@@infindebula there's hundreds of different 6581's that sound different even those that are made on same week plus there are variants like S and AR (R4). Especially R3's. That SID sounds fine. I'm a SID musician, those things are dodgy especially played with the R5 in mind.
@@-CrippledNinja- By "specific batch" I could be referring to even a subset of the date code that's marked on the IC itself. My 1984 C64 had that exact filter sound, back in 1984. And I've heard it on a select few other machines. But I never had the opportunity to see if they even fall within the same revision / date code. Despite what I said, I acknowledge that what we're hearing could be a specific way SIDs fail, but to my ears it sounds really sweet. I know it's an unusual filter sound. (PS: while I wouldn't call myself a "SID musician", I'm definitely a SID lover. Years ago I built a MIDIbox "MB-6582" with 8x 6582 SIDs, and I use it a lot, along with my other synths)
I don't know why, but I really like the repair videos. And I feel like the 64 is in that perfect space of being complicated enough that the machine can do interesting things (sound, graphics), but simple enough that you can actually diagnose it by hand.
Adrian is *the* C64 guy of all UA-camrs!
I liken it to binge watching "This Old House" only for 8 bit computers.
I like them because I'm learning
@@dannytipple1232 In '84 was fine for 14 year olds after school, the 64 is perfectly comprehensible😀
I think that’s a great summation, I had a C64 in 1982-3, it was tons of fun, imagine typing in basic game listings from magazines, character by character and then debugging, and then you could save it to floppy if you liked the game! I would love to fix these myself! It was definitely one of the first really good home machines.
I'll always watch C64 repair videos. It's more about the troubleshooting process for me.
+1 I really enjoy every one of those videos, very interesting and informative.
Absolutely agree with You :)
It gives you all the sort of shortcuts to repair knowing which chips are weaker than others.
I know nothing about soldering or repairing electronics but I always watch and enjoy the repair videos.
I am a simple man. If Adrian repairs a C64, I want to watch him do it.
I think it is incredible how you manage to pump out two videos (both channels combined) per week, despite working full time. And your videos are not some random garbage just to satisfy the UA-cam algorithm but very well made and always worth watching, both for their content and also your very enjoyable style of persentation. (I think there isn't a single video listed on either of your channels that I haven't watched..😅) Big thanks from Haribo country!
I came here to say pretty much this so yeah, I appreciate all the work Adrian. You're awesome!
I would pay for this content.
@@BartKliebert Become Patreon if you aren't already.
Adrian is a cyborg. He comes from the past.
@@stevenmitchellproductions8572 an 8-bit cyborg? 😂 👍
Great video Adrian, love the split screen format. For those wondering, the part number for the Hakko drill cleaning tool is B1303. I've had it for a while, and it really has saved me from tossing out expensive tips that I thought were ruined. The spring tool that comes with the gun just wasn't working for them.
Exactly! Thanks for putting the part number
@@adriansdigitalbasement No problem, thank you for all you do!
12 in a row is fine! Bring them on! These repair videos are may favorite part of your channel.
Adrian: When in doubt about making a C=64 repair video, make a C=64 repair video. :) We love seeing the particular symptoms of each machine and thinking along about the possible causes. It's like trying to solve a mystery before Holmes or Poirot.
Not filler, it's why i watch your videos. Fixing things is it's own reward. Taking a broken thing and returning it to usable condition is just awesome.
I enjoyed the video. You could look at it as not filler, but a training film. Trouble shooting, diagnosis, repair and testing the repair. It is a great educational video.
You can never do too many C64 repair videos! Lots of people have them and they are prone to so many failure modes. And, you're also doing a very good job of imparting general troubleshooting knowledge every time.
The C64 repair videos are always fantastic to watch Adrian - please don't stop making them !
Even if I’m not understanding what you’re talking about because it’s far over my head, I love going through the troubleshooting process with you. Will not get sick of c64 repairs!
Yep, that was definitely an (audible) issue with the low pass filter on that SID. But I would leave it in there. As long as the envelope & tone generators, amplitude & ring modulator and the bandpass filter work it should be ok to listen to.. Those pops give it some personality as you get with a vinyl record ;-) I wouldn't be surprised if that 'fault' has been in there from the factory as not all SID were created equal to begin with and the way the low pass filter behaves is the most variable component in my experience ;-)
In fact my original C64 sounded just like that when I got it back in 1984. After a couple years I wrecked my SID when I tried to hook the audio output up to a phone line, and I got a replacement chip that sounds more like what we're used to hearing. I never liked it as much as my original "crackly underwater filter" SID (probably because I had become so used to it). I think that sound is specific to a certain batch of 6581 SIDs, and it's not broken at all.
I suggest Adrian listen grab the program "Synth Sample" or "Swinth" - a program with a handful of early SID demos. Listen to "Stationary Ark" with that SID. It is glorious. I'd love to find a SID that sounds like that.
Adrian, definitely appreciate these videos - no matter how many you make, I'll watch them. Your channel is one of the ones I've watched over the last couple of years that helped spark my interest in repairing retro electronics. Your various troubleshooting and repair videos have helped me immensely with the information and techniques (soldering, troubleshooting, etc...) that you demonstrate. With what I've learned, I've built a 'new' C64 using a Sixty-Clone board and was able to use an oscilloscope to diagnose an issue I had in the timing circuit for the VIC-II. I've also managed to assemble various project boards for the C64 such as a switch-less kernal switcher and a modern VIC-II timer circuit replacement board.
Thanks for sharing your knowledge and expertise so willingly.
I'm not really into the C64 HOWEVER I will happily watch you repair them till the cows come home! Just love hearing your thought process about how you go about fixing issues, it's just great communication and a total pleasure. I like to think your little tidbits it'll come in handy when I start fixing some of my broken machines that I have piling up (Electron, SMS2 and a 48k ZX), so keep me coming!
I love these. So satisfying.
Love c64 repairs it never gets old for me.
Thanks for fixing all those machines all the time. I love watching each single one of the repair videos. Even if it is the same error again or another retrobright.
Just as some people enjoy watching the same sport again and again, i love to see this again and again
Love the split screen view!
I think I'll always find repair videos satisfying to watch, no matter how easy the diagnosis and fix are. The process of zeroing in on the issue using such common tools and a bit of logic is an interesting and useful skill, and one I'd like to learn myself. I've composite modded my father's old TI-99 and I'd like to get it working again someday (it's currently just displaying a solid, light blue screen)
c64 repair video's are never boring, keep them coming. don't skip anything
Back in 1986, I designed an IC tester that would test all logic ICs of the time that was programmed into it. In circuit, while powered.
All the user had to do, was to remove the crystal of the CPU or any clocking source, and type in the part number after connecting the ribbon dip connector to the chip under test.
The tester would pulse the inputs and latch in the outputs and compare it with the truth table in memory. It was built around a Z80A.
While testing, I saw a new similar product in Popular electronics from B&K Precision. I became so upset, that I never worked on it again.
B&K pulled it off the market. It was like they came out with it, just to shit on my dreams and then left the room!!
Maybe it's time to bring this back as a open kit of some kind? An in-circuit tester for these old machines would save people some effort.
I still remember playing Commodore Crunch and Chopper Command as a kid on my VIC20. I remember having to load some games off the tape drive and instruct the computer to load and run the software while others were on cartridges. Good times.
The repair vids are some of the best content, they are always appreciated. I could not have fixed my own machine if I hadn't watched you fix all-together too many machines on here first !! I am never going to get "tired" of yet another C64 on the bench .. heck, you could just talk about a C64 in bits on the bench and strip it for parts and I would still watch and enjoy it.
The C64 repair videos are always a pleasure to watch.
They're even better when the 8-bit dance parties are included!
I'm glad you got such a positive response to more C64 repair videos. They're among my favorites on your channel.
For people, like myself who loved them back when they were "the boss" and still do, we love to see C=64's get deserved TLC!
ABSOLUTELY keep the C64 Repair Videos coming...! They rank at the top of my personal choice in Videos to watch (along with Amiga Repairs. -Mark.
13:40 - The chip tester pro is one of those things that seems like an amazing thing, but it's far less useful in general than people realize. It won't tell you definitively if a chip is good, but it will only tell you if it's definitively bad. Adrian (and many others) have run into this many times. A lot of these older machines rely on the chip's electrical characteristics and each circuit and the retro chip tester just isn't going to exercise the chip the same way the target hardware will. So a chip may test good just fine in it, but bad in context of the circuit.
It's far cheaper to just have a cache of common 244/245s, etc... on hand than it is for the retro chip test's price. If you've gone to the trouble of swapping/removing the chip in the first place, just swap in a known good one to see if it works. It takes less time. The only place I've found the retro chip tester to be appropriate is if you have a pile of chips and want to weed out the bad ones, or perhaps see if a ROM image is good (provided the retro chip tester knows what it is).
Exactly!😁
Also, it's often just not possible to completely test a chip, especially if it, like the SID, has analog output.
A note on the ram tests of the dead test and 'official' diagnostics...
The dead test only tests a small part of the system ram, but in such a way that it doesn't depend on the zeropage and stack etc working. The diagnostics tests all ram, but requires you first fixed any issues reported by the dead test cart, because diagnostics heavily relies on the zeropage and stack and some other bits of ram working properly to do the complete ram test (and do a more thorough test). For that matter, the diagnostics copies its ram test to ram, and runs it from ram.
Consequently, if there are errors in the ram the diagnostics relies on to work, the output of the diagnostics will often be wrong, or it will just crash.
So the proper way to do this is to always first run the dead test cart, and fix any ram, pla etc errors it reports, and only after it no longer reports errors, run the full diagnostics to test the remainder of the machine. Starting out with just running diagnostics may give the correct result, or not, there is no way of knowing without first having ensured the things it depends on are functioning properly.
Thank you! I am glad I am not the only one to comment on his "deadtest" comments. I have found since the late eighties that deadtest is the first one to go to if you suspect ram problems. The first test tests each 64k chip and then reports the bad ram by flashing the chip number. It doesn't use any ram (including zero page) until rams tests ok. It requires the vic chip to work (vic pin 17 generates the ø0 for the 6510), PLA mostly to work (otherwise no cartridge would work). The bad part is it stops at the first bad ram and is dependent on ram support chips to work correctly. Just so you know, the first 7 bits are tested and then falls through to show one flash for bit 7. So if the first 7 bits are ok and even if the 8th bit is ok but something else is wrong you get one flash over and over.
@@johnsaller2481 I independently came up with a highly similar bit of code to do a 'ram test' without using ram. Only being able to report the first bit error is a simple consequence of not being able to use ram, and the relatively limited bitwise operations available on a 6510 cpu. My code is intended for the C128, but ended up using an almost identical approach.
@@c128stuff I did the same thing, and I had it for both. Suspended it until I can get back to it because of real world events!
The thing about C64 repairs is that every single one is unique in some way. We want to watch that journey of diagnosis, planning, action, success or failure, and followup. Everything is content. Record it all! (Honestly, just give me any reason at all to listen to that intro music - so good!)
I'm just glad that there are a lot of C64 machines that still exist. Awesome videos Adrian, awesome.
I have not gotten into c64 yet but want to. This has given me insight into what to look for. Thanks Adrian.
Love the repairs, Adrian. It’s not filler at all. Great content.
Great video. We love that you care for the computers. That's why we keep coming back. 👍
Side note: that RAM chip was saying "It's not me man. It's the NEC chips next to me. It's not me." 😂
Repair videos never get old. Everyone you repair another machine you always tend to learn something new. I enjoyed every repair video over the past 10 years.
I dig the new camera view for showing your face as you narrate!
Agreed! New camera angle is good!
Thanks for the pointer about that drill from Hakko - I have the same desoldering gun (and it's totally great) but recently destroyed a tip when I couldn't unblock it. Ended up breaking off part of the plating by trying to force the blunt wire cleaner it comes with through the hole using way too much force 🤦♂️
Did you notice both times the diagnostic test froze that there was random " ( " instead of " " . If you looked up the Petscii character codes (converted to binary) for both open bracket and space (ie. $20hex ( for " " ) = 0010 0000 binary, and $28hex (for " ( " ) = 0010 1000), you would see the only difference is $8hex (ie. 1000 binary) which is Data bit # 3. From this information you can either check the C64 circuit diagram to determine which RAM is used for Data bit 3 or from the C64 Dead Test manual looking at the RAM error flash code chart again , you would see that this also points to U10 being faulty.
Note: it's not uncommon when you have a memory fault(s) for random characters to appear on screen, and using the Petscii character codes you can determine the offending data bit or bits (if you see more than one random character displayed) and then lookup the offending RAM IC(s). There's always an exception to this rule (eg, bad MUX IC ) but will still help point you in the right direction.
i have a sx64 when warm it shows "&" on the screen but the software is still running fine i think its the U26 74LS373
@@matthewtempest7756 on the SX-64 the 74LS373 definitely can affect the displayed characters , but personally haven't seen it produce what you are describing ... what I have seen is UB2 (that's the SX-64 circuit reference for U26 as on a C64 board) corrupting several / all of the displayed characters. Have you already gone down the Petscii Character code path? Which would suggest if you see an " & " character instead of a space on screen that 2 x Data bits are affected (ie. Data bits # D1 and D2) , which point to RAM IC's UA7 & UB6. On a C64 if the RAM was not socketed I might suggest trying to piggyback the suspect RAM IC(s) , but on the SX-64 access is poor and if you don't have additional diagnostic tools to assist it isolating / confirming the cause of the fault, then just replace the (cheap , readily available) Suspect IC's and test again hopefully confirming the diagnosis.
C64 repairs were the main reason why I found your channel. Could watch them all day long
C64 repair video’s are my absolute favorite
I have been using the TL866A/minipro for testing some dozen logic chips and other ICs recently, and while the supported range of IC of course is not as huge as with dedicated testers, the testing results where 100% correct, yet! No false positives or negatives.
Thanks AB (and Patreon's :) for another C64 repair video! I love these as that's what I used to do back in the shop I worked on. Although back in the 80's we usually just replaced the board and sent the defective one back to C=ommodore...specially during the VIC chip "twinkle" era.
Just amazed that you put out 2 videos per week! I so miss the behind the scenes content but I had to downgrade my Patreon tier because money is tight at the moment with being ill again.
I used to repair C=64, Vic 20 and all Commodore products professionally while they were still under warranty. And I have the burned finger tips to prove it. I didn't have any decent temperature sensing devices until years later, and so it was the finger that was used to find hot RAM chips. Went through a lot of sockets repairing these units.
I even repaired Commodore PCs when they came out....at component level. That stopped after the '286.
Also serviced Atari products.
In 1983 MOS Technology used a defective process in chip production: too much boron in the passivation layer, ironically the outer coating intended to protect the chip. Thus, the horrible long-term reliability of many chips in 1983 and 1984 machines.
Wait... are you saying "MT" ram chips are MOS Technology, not Micron? (never knew MOS made DRAM?) All this time I've been trying to figure out why Adrian is always sh**ing on MT ram chips, when Micron was always the most reliable ram brand (perhaps 2nd to Samsung) the 20+ years I had a computer store.
@@cliffshockley4406 No, MOS made ROMs, but not RAMs, IIRC. This is just referring to the issues with the 7400 series logic chips. MT is definitely referring to Micron Technology.
@@cliffshockley4406 MT = Micron, who had their own problems. However... Commodore bought Frontier Semiconductor. Everything made in the Frontier Semiconductor fab in 1983 and 1984 suffers from the boron problem. Well known chips that fail because of the boron problem are the PLA, 8501 CPU and the MOS 74xx clones. Chips made in the Norristown fab during those years are reliable.
@@danielmantione Just so surprising to me, as in later years, I can't recall ever having defective Micron MT ram chips, though I do suppose they later had their Spektec subsidiary which they used to sell their "substandard" (my interpretation, don't sue me Spektec) ram, so maybe they just got better at testing for ram that was off-spec.
Defective to you and me. A literal gold mine for Commodore and MOS when the things failed after a few years and people had to buy new computers, or get new parts.
:)
I appreciate seeing more c64s repaired!
Love the C64 repair videos. If I learn any tiny thing Im happy. Also think of the library of repair information you building up here on the channel. Love it.
I always love your repair videos Adrian, you always explain in such a clear, fun and educative way. Everytime you post a repair video I get a nostalgic trip down memory lane :) I am looking forward to more of them.
Not too much. Looking forward to every single one of the c64s getting repaired.
Please keep up the good job.
C64's and retro repair just go hand in hand. The C64 was so modular, I wish computers were still like that now.
I really enjoy all Commodore repair videos! I hope that's not boring for YOU, though. Please enjoying repairing and recording what YOU like; we're here to watch. Aloha!
Always great to see Commodore 64 diagnosed and fixed :)
We will watch Commodore 64 repair videos like a squirrel stares at a bag of walnuts. Carry on good sir.
Excellent repair as always Adrian! Why are RAM tests so bad on the C64? I've also had all sorts of inaccurate results with them. Is it really not possible to detect them reliably? Maybe I'm going to have to write a C64 diagnostics ROM like I did for the Amstrad CPC 😃
In this case it failed during the stack test on a diagnostic that actually uses it. I suspect they probably are more reliable if the error occurs in general purpose ram
The ram test is just too simple to be that useful .. it likely only quickly writes a couple patters to some pages and calls it a day. That’s not thorough enough but it’s fast.
@@adriansdigitalbasementthis shit is fkn lame
It sounds like the challenge is to do a really good test on the zero page using nothing more than ROM and the registers in the CPU.
Noal that would be awesome to be able to test the ram fully .Like you ive loads of different results . As i believe the tests we have now only write to small random places ..
I noticed the address of the failure was 3xx ish. >256
Thanks for the video! I was wondering about that 4066/U28 that the diagnostic cart reported bad, when you ran the diagnostic with the harness.
It's just a false reading. The joystick ports work perfectly -- I think one of the ICs has gone bad on my test harness.
The first thing I do every time is to push all the chips well into their sockets because they can pop out very easily, especially if someone has used cheap sockets. Many times something worked straight after pushing the chips in well.
Might be interesting to build a program that analyzes the output of the SID over a microphone port. Could do a waveform comparison between a known good 8-bit dance party and a test one.
By far my favorite repair channel! Do you think one day you could do a video on how you set up your Easy Flash Cartridge? I bought one and a bit confused on it lol
Totally.. its the best part doing the fixing well watching you doing the fixing and chasing the gremlins..
Adrian, I used to check for bad chips spraying compressed air. The first one that evaporates means has more temperature than the others.
Really awesome. I especially like the hand acting. Reminds me of Mr Bill and Mr Hands!
C64 repairs are some of my personal favorites.
SIDs are a bit clicky anyway, that's how they originally did sample playback. Filters always vary, pro-synths with analogue filters have calibration but the SID never did.
Dude. Love the intro! Been watching alot of your videos lately and the intro gives me 90s TV vibes. It's nostalgic!
Hey @Adrians_Digital_Basement Look at this scammer! You should go through your videos and delete these losers. Some people might fall for it
Thanks to your Patreons for saying yes, repairathons are great!
I have seen the com port test fail where a trace had corroded between the DB9 port and the PIA chip ended up running a bodge wire to fix.
Quick way to test is see if a joystick has lost a function :)
Take a look at acetylene torch cleaning bit sets. They're cheap and have quite a few different size bits on them for cleaning torch tip holes. They also work pretty well for the older Hakko models that I've used.
Neat! Good “tip” :-)
Ya know Adrian, I don't have nor ever used a C64's, or any game mechanics as I don't play games but I watch your videos anyway.
Anyone that survived the IRQ and Memory wars of the 90's computers and games an still like to play games, I take my hat off to ya. Me, it away all the fun an ruined my game playing as a computer tech years ago sad to say... :-(
The reason I do watch, is I do fix alot of my own electronics as a hobby, if it is still broken when I'm done, well it was broken to start with... LOL
Also mostly because I just plain enjoy your videos and seeing what it takes and how you go about fixing them. Believe it or not I have fixed several thing just from watching your videos, with that I say a BIG Thank YOU!
Take care!
LLAP 🖖
Keep ‘em coming! I enjoy this fix videos
Need an interface from the ZIF socket on the "retro chip tester" that you could place on a RAM IC while it's in situ. Akin to how stacked chips came about but this one with a ribbon cable leading to the ZIF socket :D
The intro song playing in stereo is just awesome.
Hi Adrian, did you notice the message in the diagnostics test that says "4066 @ U28 BAD"?
This might have something to do with the SID popping like that.
Haven't got the schematics at hand but I am pretty sure there's a 4066 involved somewhere around the SID circuitry.
Might be worth taking a look at that.
Repair videos are always interesting. Definitely not what I would consider filler.
Yea!! The RCT!
Great video. Very enjoyable! You are getting so good at explaining troubleshooting! Well done!!
I use it , too. Great device, indeed.
would be cool to make a 64 or amiga with ssds etc just inside the case..? maybe some coolable dual core..? noteboox??
Thanks for all your great videos. have a great weekend
Very good explanation of diagnostics and repairs! Thank you as always for your videos!
I love your videos. It just struck me, tho, how much I love your intro music!
Definitely like these repair videos, could watch all day.
Well, this one is going to be an interesting set of videos, it would be nice if you could also make some videos about deeper diagnostics and repairs, maybe even upgrades?
LOVE C64 videos! Thanks for showing off the cool tester again; even if it’s not 100% accurate.
Thank you Adrian another great episode.
I would gladly watch all C-64 repair vids!
Arent the SID filters dependant on some of the capacitors next to the SID? I have a vague recollection of recapping that area solving some filter issues.
I would watch endless C64 repairs for the rest of my damn life.
Great video came here from recommendation on Techmoan comment. Subbed.
Adrian: without further ado let’s…
Me: He’s gonna say it!
Adrian: start looking at these machines!
Me: 😳
Haha. I like to mix it up to keep everyone on their toes!
Most of my MT ram was bad. I sent the computer over to Ray Carlsen and he replaced it with a "modern equivalent". Now it works great and I do not foresee any memory issues again on this machine.
That was quick! Good work as usual.
SID chip sounds fine… it’s a tune made for 8580 chips, it will sound clicky on a 6581. Better to test 6581 chips with some other music, I’d recommend Evolver by Jeff…
Yep, propably filter change bug. It is not audible on 8580 but is on 6581 (because of the filterswitching bug which they later fixed in 8580).
Except that Adrian has listened to that tune on a hundred 6581 SIDs, and by extension so have his viewers!
I agree with Adrian that the filter sounds unusual for a 6581, but I don't think it's broken - I think it is a dark, warbly filter that is characteristic of a specific batch of 6581 SIDs.
@@infindebula there's hundreds of different 6581's that sound different even those that are made on same week plus there are variants like S and AR (R4). Especially R3's. That SID sounds fine. I'm a SID musician, those things are dodgy especially played with the R5 in mind.
@@-CrippledNinja- By "specific batch" I could be referring to even a subset of the date code that's marked on the IC itself.
My 1984 C64 had that exact filter sound, back in 1984. And I've heard it on a select few other machines. But I never had the opportunity to see if they even fall within the same revision / date code.
Despite what I said, I acknowledge that what we're hearing could be a specific way SIDs fail, but to my ears it sounds really sweet. I know it's an unusual filter sound.
(PS: while I wouldn't call myself a "SID musician", I'm definitely a SID lover. Years ago I built a MIDIbox "MB-6582" with 8x 6582 SIDs, and I use it a lot, along with my other synths)
@@infindebula trust me, it's just the filter bug "click" and you can hear it with filter off /on on 6581 easily.
Retro computer repair video's are always awesome!
The 8-bit-dance-party is excellent excercise for a stiff shoulder. ;) Get well soon.
Good video!!! Love any and all Commodore content! TY
Always love the 64 videos!!
C64 repair videos are the best for me.
Great video Adrian as always .. love watching your videos :)
Can´t get enough of the C64 repairs :)
You almost always learn something on a repair. Well worth this vid.