11:28 "no date code of any type". In the 1980s, some chip manufacturers left the "8" off the date code to avoid confusion with chip type numbers that also started with 8. That chip says "42300" which probably means that it was made in week 23 of 1984.
I have no idea why I am spending time learning how to detect fake Commodore or TI chips but I can watch Adrian for hours at a time doing almost anything he cares to deliver! Awesome video Adrian, thank you!
I would love counterfeit chips as long as they function like the real thing. So I actually hope they make a really good copy before these old chips are gone for good.
@@cfothough You are never going to get a SID replacement that's exactly the same as the SID you know because 10 real SIDs all sound way more different from each other than an ARMSID does from whatever one it's closest to. But it's good, especially ARM2SID since you get two sids inside the one chip.
I have a feeling that smaller pin spacing is 2.5mm instead of those headers' 0.1" (2.54mm). Across 32 pins, the accumulated error would be 1.28mm, which I'd guess would be uncomfortable but not impossible to shove together, just as described.
I tried to use the TMS9918 in my homebrew computer. It didnt work properly. Spend hours and hours on it, thrying to fix it. Turned out it was a bad IC... Bought another one and it worked perfectly.
The 2382 chip sounds fine, it probably has unusually high, "8580-esque" cutoff values. You may try Terra Cresta by Martin Galway or Lightforce by Rob Hubbard, those are excellent for testing "8580-esque" 6581s.
One of the major reasons that the 6581 is so hard to clone is the fact that they were made from masters that deviated a lot from the design specs, the 8580 masters on the other hand were much closer to the specs, which greatly improved their technical quality but also greatly reduced the stray current that enabled the digisampling with just the master volume control that the 6581 was famous for.
I think some eastern European / soviet countries used the slightly off pin spacing as well. 2.54 was .1 inch and 2.5 was an attempt to make it "metric"
That's what I was thinking too, metric spacing. Russians were well known for using metric DIP chips. It must have been real fun trying to put those socket pins in there. But what TI was thinking, who knows. Clearly a dead genuine chip; they probably had no way to test it. At least it's clearly not a 68000!
Genuine TMS9900's were 64-Pin, .9" wide DIP packages with .1" lead spacing. It is possible that the chip he received was a soviet clone having 2.5mm spacing and repainted as a TMS9900 - obviously a bad chip. At the time of the TMS9900 (1976), the 64-pin DIP, .9" wide was unusual but was chosen to prevent having to multiplex the address and data lines. Motorola would eventually use this package for the 68000. Honestly a repainted 64-pin chip could be anything so it might have been a 68000 instead of a 9900 - there is really no way to know for sure... you risk smoking a vintage machine plugging in these repainted chips - in my case I have built fully buffered "go between" sockets for these types of tests but even these aren't failsafe...
As someone that's programmed for the TMS9900, let me tell you, the pin spacing isn't even CLOSE to the only weird thing about that chip. For starters it doesn't even have a stack pointer.
who needs a stack pointer? even if you want one you can easily use on of the general purpose 16 bit registers for that. With the indirect auto increment mode you will get pretty close the the normal stack pointer behaviour. The TMS9900 architecture seems to be 'inspired' by the pdp11, a computer that brought us c, and unix. Actually the orthogonality of the TMS9900 was rare at the time the processor was launched.
@@available898 I actually learned assembly language on the 9900 as a little kid. When we switched to the C64 (cuz that's what everyone else had) it took a long time for my brain to shrink (as I like to describe it) before I could do 6502 assembly.
I stacked up the work spaces and used the 'registers' as local variables. The Otis 401 family of lift controllers that are all over the world used TMS9900s and its family of peripherals.
The workspace idea I really like, kind of like direct page on the 6809. I just am not a huge fan of the TMS9900 syntax, though maybe that's just the TI-99/4a assembler package. Also it's really annoying that you need both the 32k expansion and the disk drives to even use the assembler package. PEBs are just ridiculously expensive these days. I've used MAME for all the TI-99 assembly I've done, which isn't much. I'd absolutely kill to find a PEB for a decent price... people want like $500-1000 for them!
@Ken Mason The SID chips we like best are the ones we had in our own machines we had when we were young. 😊 It's the same as when there's this one particular mix of a song we prefer even though it may objectively sound worse than others. Which is why I don't really believe in "audiophilism" (unless I get to experience it myself but that's unlikely as my hearing is somewhat ruined at this point), and find it to be a combination of preference, elitism and autosuggestion. But we all have our own weird quirks so whatever. Icecast sounds interesting though, I may want to look into that sometime.
Surprised with regards the SID there isn’t an analyser of some sort, eg Android app that takes the audio out via a line in and checks (think guitar tuner apps)
The filter on the R1 SIDs sounds very different than R4 -- there are some tunes that sound like they have missing basslines on the R1. I had an early model breadbin growing up that played Hubbard tunes (among others) completely differently than my friend's, and always thought there was something wrong with my system. Many years later, I blew out the R4 SID in my SIDstation and replaced it with a spare R1 I had from my parts bin, and manually sweeping the filter made it very clear what was going on.
Thank you so much for the tip with the acetone, Adrian! 💪 I find myself almost addicted to your videos. Kind of in the same way as when we as kids would wait impatiently for the next episode of our favorite cartoon or TV-show. The fact that I don’t care at all for CRTs or that it is pure torture for a fellow sweet-loving diabetic to watch you munching on and (what feels like) endlessly talking about sweets… quite oddly doesn’t seem to have much impact at all. 😁 I think what I love the most is that we get to see when something goes wrong for you as well. It is kind of hard to learn from other people’s mistakes when we don’t get to see them. It also brings less of a teaching or tutorial feeling and more of a “just along for the ride” feel to it, which I really like. I seem to have collected quite a crazy number of EPROMs and EEPROMs over the past few years. Some from eBay, some from AliExpress, and some from random web-shops, which I have forgotten the names of. Doesn't really matter as I don't remember which chip came from where anyway either. Some of them work, some kind of semi-work, and some of them I couldn't get working at all. But like you, I always keep broken chips... hence the crazy collection. For quite a while now I have suspected a portion of them not to be actual fakes but maybe resurfaced, painted, and relabelled. Because many of my 27C160 EPROMs (the ones I have the most of) are marked as "M27C160", "ST", and "Singapore". Yet there is quite a variance in how they look. Some have a darker and almost black finish to the ceramics, others are light grey-brownish, but most lie somewhere in between. The key difference lies in the size and the partitioning of the exposed chip beneath the window. This makes me suspect that some of them may be smaller, like 27C400s or 27C800s only marked as M27C160. They may not all be from Singapore, maybe made by SGS-Thomson or some other brand instead. We often find ourselves having strange ideas and biases. I don't know why I have never dared to use acetone. I have tried cleaning these chips with isopropanol and other cleaning solutions without revealing much. But if acetone is safe for human fingernails, it shouldn't really be too harsh on plastic EEPROMs and not at all on ceramic EPROMs, right? (doh!). So, after watching this video, I instantly went and broke out the acetone.... and it obviously removes any paint with a vengeance. Some of the EEPROMs have clearly been resurfaced and laser-edged. But I didn't expect to find that quite a few of my EPROMs have also gotten the same treatment. Because as you indicate, the presence of the glass window would seem to make that process harder to pull off. But no, the acetone did remove a ton of paint off some of those as well. A few of the semi-working ones had what to me seemed like an excessive amount of paint on top, spreading out onto the sides. There was even a thin layer on the pins, which I just hadn’t been able to see with the naked eye, and obviously why they were behaving oddly. But it sadly didn’t bring out any of the old markings on any of them. I don’t really care if they are a different brand, speed, or even a smaller data-capacity. Because I can use 27C400s or 27C800s just as well. But the mislabelling is annoying when you are attempting to burn something onto them. Because my (up until now) favorite GQ-4x4 USB programmer needs an adaptor for burning these kinds of 16 Bits EPROMs. And this adaptor makes it no longer able to detect the ID of the EPROM, if the chip is seated wrong, or there is a problem with the connection to any of the pins. It just simply fails instead, and then all I can do is put the EPROM back into the eraser for another try 20-30 minutes later. I have just gotten one of the fairly new XGecu T56 programmers, maybe I should try using that one instead. Anyway, thanks again for the tip… at least now I have the means to “get right to it” as you say 😉 Greeting from Southern Jutland, Denmark 🙂
I'm shocked the SID chip hasn't started being manufactured again in small quantities. I love the sound of that chip. I know some DJs and musicians have utilized the sid chip for producing music.
Having been through about 4 SID chips (which were WAY more than $25 each) trying to get my C64 repair/restoration finished... and each of them have had their own issues; I -think- I hate you now... If I were to order a SID for AliExpress it wouldn't even be an IC... it'd be a Dorito with some staples glued to it or something. How the hell do you keep getting working SIDs for such good prices!? It's killing me, man! I did FINALLY get to order a SwinSID Ultimate and it'll get here from Germany, eventually... but I'd've much rather had a fully working, real, SID. Keep up the great videos and keep up the C64 love!
34:50 - the suspence was killing me, I was rooting for the chip to work and when the right screen showed up I was cheering like a 4 year old in a candy store :D It just made me so happy that the chip is finally working :) 36:50 - maybe you could try to use a socket that has the 2 metal thingies (or is it just one bent into a V shape), cut it up and solder the sides in so you'll have more horizontal clearance and play on the chips instead of just a single hole per leg. Just an idea. It might not look as clean but it might be easier to take the CPU in and out. And run the bodge wires inder the socket or something, but that's just me...
Chip doesn't work. "Time to hit it with Deoxit!" :D Loved the analysis! And, yeah, there are some chips that have 2.5mm spacing instead of the standard 2.54 (i.e. 0.1 inches).
I'm across the river from you on the Washington side-- I'm set up for x-ray inspection of chips and boards and the like, so if you ever want to find out (non-destructively) what's inside those packages and compare them to a known-good one, give me a shout!
Could it be, that the "TMS 9900" actually is a 68000? They both have the same package and the M68k actually is a lot cheaper to get, from what little I know.
Wasn't the TMS 9900 the first single-chip 16bit CPU? I seem to recall they kinda set the standard on the 64-pin DIP packaging, which was later used by a bunch of other chips, usually high-spec stuff like the 68000 or high-integration parts like the Z180. That said, the package notching and markings look like TI's TMS 9900. If I had to guess, it's a legit TI part, but was probably just toast. If it's not a dead "refinished" TMS 9900, then it's some other TI part that came in 64bit DIP packaging.
I've not seen a m68k with such large circular marks, in particular the semi circular mark on the left. They seem to be smaller on Motorola DIP64 packages.
One of our New Zealand marine research vessels - RV Tangaroa - has an onboard comms system powered by a TMS9900 from one of my TI99/4as. I got a tour of the ship as a thank you :)
Ah the TI-99 that brings back memories, I used to do in-house repair of those back in the day. Right before they were closing down all of the production lines for those we could buy them for $50.
Re: process to make IC legs look new... They will straighten them, brush them, then dip in molten solder in an effort to make them look new. Most of the dram, eproms and old chips from China seem to arrive here like that anyway. I've had problems with them 'binding' in sockets, or introducing reliability problems, so I try to solder them in directly if I have to use these.
They may not be fake. Most of these chips are recycled from old machine parts, clean and legs re-tinned. It's a gamble because the seller does not has the means to test them out.
It's my understanding that IC chips are full of discrete components. So would it not be possible to essentially unpack a SID chip and make it large scale?? If possible, I'd like to see that done as a project here.
Usually they contain just a single silicon chip. You could potentially reverse engineer it and make something fully discrete, bit it would not be trivial.
@@TomStorey96 and that silicon chip contains all the discrete components. I realise making a full size replica wouldn't be easy. And the process of grinding the surface of the chip down to the wafer and then using an electron microscope to look at the contents is beyond any normal person's capability - but if there are schematics of the dies inside, it would be interesting to see it done and to see what sort of sounds that version would make.
The use of machine sockets will make it harder install the TMS cpu, but the cheaper sockets may in fact be a better choice since they have spacing to better account for the slight pin spacing.
Adrian, nice SID chips. I study these enigmas, trying to build a mega synthesiser and I’m buying as many as I can. I read that MOS grew them organically from silicon or something, and why can’t we do it 37 years later, but nevertheless my construction is already the size of a table, involves 24 SID chips. I wanna blow the windows off anywhere I take this thing. Considering using a Leslie speaker, and being an audio engineer, I’m just collecting 64s and chaining via midi. I need help.
This brought back memories - so many hours of my life swapping around components to try to keep a system running. In the early '90s I ran a computer room at a small school. When I first got there it had about 20 donated IBM PCs, and as each one failed I would replace its bad part with another from a machine that had failed in some different way. I managed to keep the place going on a shoestring, but I don't want to think about how many weekends I spent on the floor of that room trying to get some Frankenstein's monster of a machine to last one more month. By the time we closed the room ~8 years after I arrived, I think we were down to 12~13 working systems (and I broom closet full of scrap parts).
Whenever I socket something with a nonstandard pin spacing, I take a gold plated dip socket with a compatible pin size and use nippers to pick off one pin at a time and after its soldered in, clip off the rest of the plastic,, then if you want to get real fancy, you can use pieces of poster board to build a wall around the pins on top of the board and backfill with potting compound to the height of the pin sockets.. Either way, no funky Bending.
I was really hoping that fake CPU didn't blow the rest of the machine! Less colored circles, more accurate VUs :P Oh and the Donkey Kong splash screen on the C64 in the background would be fitting :) Very nice demo to whoever wrote it! :)
No, it's regular 0.100". The problem is inserting a desolderd 64 pin chip with very slightly bend pins into an precision turn DIP or actually 2 rows of precision turned SIL... (Pin diameter typical ~0.150", hole diameter of a turned socket typical ~0.200") 64 pin DIP is the only case in which I would prefer a regular socket. Even if it they were new from the factory these 64 pin DIP chips have to be hard to insert in a socket like this. There was a reason why they stopped at 64 pins...
Lucky you that you got 2 genuine, working SIDs from China. I recently bought 10 6581R4’s from two different electronics parts vendors out of Hong Kong. The picture they showed me before purchase was of a genuine 6581R4. (I actually already had one with totally identical markings as their photo.). What I received were clearly re-marked chips. I didn’t have to do the acetone test to check, plus they had return policies and didn’t want to “damage” them from the acetone test and risk not being able to return them. In the end, I got my money back. The strange part is that of the 20 chips, many of them did not register as either a 6581 or 8580 in my C64 Reloaded board, but several did. Of those, most had some degree of functionality of a SID, so it seems like what they do is either: 1. Mark a completely different chip with the same package to be the chip you’re ordering (in this case a 6581R4 SID), or 2. Clean up and re-mark an older SID with the newer markings to pass it off as the newer version but don’t check if the chip actually works. I did find a few SIDs that were maybe 60% or 70% functional, and even found one that was probably 99% functional, but had to send it back to get my refund.
I have very fond memories of growing up with a TI 99/4a. I remember we had some math game, Jawbreaker II, and Pole Position for it. We also had a game called Tunnels of Doom that loaded off of a cassette (we pillaged the cassette player from our home organ at one point to play it). My brother still has it all these years later and he can't wait to show my niece how to use it.
Today I fixed 2 Quickshot Turbo II joysticks, both had missing wires, one was just yanked off, one was cut. I had to tinker with the wiring with the new cable (used a 9pin dsub extension cable and cut off one of the ends), cleaned the microswitch contacts and they are just like new, I'm so happy!
What did you do for strain relief where the cable comes through the plastic shell? I'm curious because i would like to find a good method myself. I've tried with 4 zip-ties in a little square on both sides of the shell, but i think there must be a better way.
@@BertGrink One of the joysticks still had the strain relief on the cable, I janked out the old cable and put it on the new one and put some electrical tape behind it so I won't be able to pull it through the strain relief plastic. The other one sadly didn't have the rubber strain relief so I just put a huge winding of electrical tape to keep the wire inside the shell so pulling on the cable won't stress the soldering. You can reuse old strain relief rubbers, they are not permanently attached to the wires and I also wanted to buy some online but it's hard to find the right size. Luckily the my new wire was the same size as the old one.
@@Dukefazon I think you have been quite lucky then; most strain reliefs i have come across have been bonded to the outer isolation of the cables, making it very difficult to separate the cable from the strain relief.
Wow. Thanks for some great info. I'd like recommend also InSIDious which is a totally Faithfull vst recreation with patches programmed by Rob Hubbard, Jeroen Tell, Martin Galway etc...
Needed something to watch while entering data, more listen, but does this ever fit the bill. This kind of video fascinates me. That cart with the song is awesome! Love that sound.
A tip: Chrome has a picture-in-picture addon you can download from the extensions page (made by Google, dunno why it's not just included). Then you can press ALT-P to pop any HTML5 video player into a little resizeable always-on-top window. Probably still not great to use at work, but plays really nicely with games - I use it a lot when I'm playing Runescape.
@@GdotWdot just have my cell phone next to me and glance at it when he’s talking about something I need to see. Thanks for the tip though. Going to check that out for home.
@@xabhax Yes, some of them do that. Others don't ship at all and then claim that your package was delivered, based on having given you someone else's real tracking number. This is the scam I'm referring to.
@@andycraig7734 When I see a packet having been delivered in Russia, I apply for a refund. Got it every time so far. And most of the time a couple weeks later the real packet shows up. So yeah, I'm fine with them hurting themselves...
@@ligametis There are daily freight trains from China to Germany, but normally such goods would be using airmail. These never land in Russia because they take crazy fees from the airlines.
Thank you for the video! I enjoy learning about these counterfeit or fake ICs, and it is good to have more awareness. For the leg cleaning, I have been able to do this with one of those solder pots, melting pot to have the pool of molten solder, like the kind you might use to tin the tips of wire. If you put some flux on the pins and then quickly dip the entire legs into the solder it comes out to look like this for me anyway. I know its not really related to this video.. but I really wanted to ask, how do you get your audio so good? Your voice seems very clear and articulate. and no background / room noises, but also I don't hear artifacts of a noise cancelling too. Is this just from that Rode Go lavalier mic!?
i see that the more expensive ICs on your longboard have passive cooling. So i guess you recommend the use of passive heat sinks. But i wonder, if you would recommend heat sinks on the shortboards with the later variants of those ICs too?
31:50 that socket has a 45° angled shape that designates pin 40 for some reason. And that fake TI CPU... I would have measured ohms on some select legs like last time you checked fake ICs
The 100- AMD-27c400's i got fro0m china were actually Micronix 27c4100's pin compatible but the eprom burner identified it internally. remarked. Chinese chips are like a box o chocolate,you never know what your gonna get.
the age of the CPU chip I think anyone would be happy with pin to pin compatible replacement, with any name on it? if it got there computer going, newer ones would probably be capital of running faster with being newer, as with most CPU's
Many old devices will not function properly due to moisture ingress - very slow over time - when the device is not in service for a long time. As in - no heating. A time-honored solution, particularly with memory parts, is to bake them for 12 - 24 hours at a temperature of 95 to 110 degrees centigrade. This technique has been used in PCBA facilities for many years when the devices have been outside their original foundry shipping sealed packaging beyond the prescribed warehouse-storage-life of the device.
Hi Adrian, I see some 24 pin EPROMs there, WSI 57C49C and TMS2532. I have been looking for some drop in replacement EPROMs for Commodore's MOS2332 and MOS2364, seems like you've got what I've been looking. How did you program these? My TL866II+ doesn't have them in it's list.
I've seen it where the fake chips are just empty carriers marked up... so no active die inside there. IMO if it fails the acetone test, its junk and unless there is some indication of what the part actually is, it would go into the bin rather than risk damaging the circuit its being used in
Adrian in his basement, cup of coffee, sit back, good times. Had a dream last night about a game called Nodrub’s Quest and woke up to some Ti99 video. That’s called a kwinkidink
In regards to the EPROMs; It's possible they weren't trying to fool anyone to accept substandard chips, but rather make it more probable that they find potential buyers (people who want replacement parts who doesn't know which chips WOULD work the same in their computer, so they just search for the one type they KNOW works).
I recently got a batch of fake DRAM (batch of 20 - all exactly matched and looking way too new, plus incorrect US date codes on a Japanese part!), however they survived the acetone test!! Looking closely I could see they were sanded, then maybe sand-blasted to give them a texture and finally laser-etched. Ie. no paint used in the process! Just something more for folks to watch for!!!
that non working TMS9900 CPU...64 pins I wonder if they relabelled a motorola 68000 which also has 64 pins ? Might be worth trying it in an atari ST or amiga
The TMS9900 could of been a 68000 just marked as a TMS9900 , as they both are 64 pin DIP...?? So might be worth trying it in an ATARI ST or An Amiga ??
The 82 sid chip is a Revision 2 i have the exact same datecode sid in my Ultimate64, the later AR has more Bass and the Revision 3 is the Screamer Sid!
Fun video. I especially liked the fan-made 8 Bit Party - it put a big smile on my face. May suggest you take keys from the TI 99/4A that needs repair and put it on the one you fixed last year to complete it?
I guess the re-brand can made them easier to sell, since some younger people know nothing about Hitachi or STM EPROM, I got an re-brand AMD also, TL866II detected the ID is STM. Famous chips can sell higher price.
The fake TMS9900 could very well be a 68000 or 68010, it has the correct number of pins (64) and looks almost identical to 68000 and 68010 parts I have. It'd be interesting to try it and see if it works after the 12v insult it got. You can put a socket in an old compact Macintosh from the original 1984 Mac to the SE, which used the long DIP package 68000. The later classic I think changed to the QFP variant. The limiting factor is I think the pin pitch is not the standard 0.1" / 2.54mm, and is a bit narrower, so getting a socket may be a bit of a challenge.
I have a solution for the pin spacing. Permanently mount the cpu in a ic socket and then the spacing will match putting it in another of the same socket by putting one socket in another!
Hard to notice but are u using a ESD wristband, cos i would with those precious MOS chips for sure. For any handling of PCB with ESD devices i would use one.
Strange, according to the TI datasheet for the TMS9900 processor the pin spacing is 0.1", which is totally standard. Admittedly, this is for the ceramic package, but why would they change this for the plastic package?
FYI the reason that Ti used the TMS9900 because that was what their industrial machine tool controllers were based on. So there were more out there than you knew about, but still not a lot.
i had the same impression regarding the old 6581 version. it sound distorted. not pleasant for my ears at all. this is the advantage of newer re-implementaions, they all do the filters digitally which gives higher precision (almost) without distortion - if correctly done.
I actually have a board that uses the TMS9000 chipset on it, used as a serial terminal interface.... Does have the socket as well, with that special spacing. Likely the Ali one is the OTP version of the TMS system, they did make some with built in ROM and RAM, as it is not likely to be a 68000 series, unless TI was acting as second source for them. Could also be one of the TI DSP families, as they used that package for them early on as well, needing to get all the pins out, and having a parallel data and program bus to interface with the programming.
Had the same problem with the AMD 27C400 ROM's from Aliexpress, It schould be AMD but the first batch was Toshiba (TC574200) with a big erase window and the paint was still wet and the second batch was Intel (D27C400-150V10) which had an accesstime of 150ns instead of 95 ns.
Every time I see a TI99 I find out more about its weirdness. We were considering one back when I was a kid but I got a Vic 20 instead, the right choice I think
11:28 "no date code of any type". In the 1980s, some chip manufacturers left the "8" off the date code to avoid confusion with chip type numbers that also started with 8. That chip says "42300" which probably means that it was made in week 23 of 1984.
I have no idea why I am spending time learning how to detect fake Commodore or TI chips but I can watch Adrian for hours at a time doing almost anything he cares to deliver! Awesome video Adrian, thank you!
I would love counterfeit chips as long as they function like the real thing. So I actually hope they make a really good copy before these old chips are gone for good.
even if there close enough. like most of the time you dont need %0.0002 precision for old hardware.
There's always FPGA
There's nothing wrong the the ARMSID and FPGASID. Both very good and better in some ways than the originals.
@@thomassmith4999 I've heard the ARMSID has some filtering errors
@@cfothough You are never going to get a SID replacement that's exactly the same as the SID you know because 10 real SIDs all sound way more different from each other than an ARMSID does from whatever one it's closest to. But it's good, especially ARM2SID since you get two sids inside the one chip.
"Would you look at that!"... so satisfying to hear everytime! ;D
I have a feeling that smaller pin spacing is 2.5mm instead of those headers' 0.1" (2.54mm). Across 32 pins, the accumulated error would be 1.28mm, which I'd guess would be uncomfortable but not impossible to shove together, just as described.
Adrian makes himself a Cuba Libre. Someone is having a good time there.
The ice is still ice 🤣
I tried to use the TMS9918 in my homebrew computer. It didnt work properly. Spend hours and hours on it, thrying to fix it. Turned out it was a bad IC... Bought another one and it worked perfectly.
19:50 because acetone is literally used in paint remover, so of course it's better.
The cruelty of having to wait for that bass drop. Phew!
The 2382 chip sounds fine, it probably has unusually high, "8580-esque" cutoff values. You may try Terra Cresta by Martin Galway or Lightforce by Rob Hubbard, those are excellent for testing "8580-esque" 6581s.
One of the major reasons that the 6581 is so hard to clone is the fact that they were made from masters that deviated a lot from the design specs, the 8580 masters on the other hand were much closer to the specs, which greatly improved their technical quality but also greatly reduced the stray current that enabled the digisampling with just the master volume control that the 6581 was famous for.
"It fixes the legs so that they easily go into the socket, without any kind of issue." 😂
Yea talk about "Jinxing" ;)...
I actually thought "wouldn't it be typical if he bent a pin after saying this".. And lo and behold...;)
"it places the lotion in the basket"
It puts the socket on the pins or else it goes into the bin.
@@kitsophrenik Exactly the scene I was thinking of.
It sounds like the pin spacing on that CPU is 2.5mm instead of 2.54mm
I think some eastern European / soviet countries used the slightly off pin spacing as well. 2.54 was .1 inch and 2.5 was an attempt to make it "metric"
@@stonent yes, Soviet ICs had usually 2.5 mm spacing
That's what I was thinking too, metric spacing. Russians were well known for using metric DIP chips. It must have been real fun trying to put those socket pins in there. But what TI was thinking, who knows. Clearly a dead genuine chip; they probably had no way to test it. At least it's clearly not a 68000!
Genuine TMS9900's were 64-Pin, .9" wide DIP packages with .1" lead spacing. It is possible that the chip he received was a soviet clone having 2.5mm spacing and repainted as a TMS9900 - obviously a bad chip. At the time of the TMS9900 (1976), the 64-pin DIP, .9" wide was unusual but was chosen to prevent having to multiplex the address and data lines. Motorola would eventually use this package for the 68000. Honestly a repainted 64-pin chip could be anything so it might have been a 68000 instead of a 9900 - there is really no way to know for sure... you risk smoking a vintage machine plugging in these repainted chips - in my case I have built fully buffered "go between" sockets for these types of tests but even these aren't failsafe...
As someone that's programmed for the TMS9900, let me tell you, the pin spacing isn't even CLOSE to the only weird thing about that chip. For starters it doesn't even have a stack pointer.
who needs a stack pointer? even if you want one you can easily use on of the general purpose 16 bit registers for that. With the indirect auto increment mode you will get pretty close the the normal stack pointer behaviour. The TMS9900 architecture seems to be 'inspired' by the pdp11, a computer that brought us c, and unix. Actually the orthogonality of the TMS9900 was rare at the time the processor was launched.
@@available898 I actually learned assembly language on the 9900 as a little kid. When we switched to the C64 (cuz that's what everyone else had) it took a long time for my brain to shrink (as I like to describe it) before I could do 6502 assembly.
I stacked up the work spaces and used the 'registers' as local variables.
The Otis 401 family of lift controllers that are all over the world used TMS9900s and its family of peripherals.
The workspace idea I really like, kind of like direct page on the 6809. I just am not a huge fan of the TMS9900 syntax, though maybe that's just the TI-99/4a assembler package. Also it's really annoying that you need both the 32k expansion and the disk drives to even use the assembler package. PEBs are just ridiculously expensive these days. I've used MAME for all the TI-99 assembly I've done, which isn't much. I'd absolutely kill to find a PEB for a decent price... people want like $500-1000 for them!
The CPU might be a repainted genuine chip to look new, but faulty and they did'nt check it.
It's funny that it was the 70% alcohol that was available - because that was the one that was more effective for the purposes of human malware.
...and now I want oscilloscope traces and whatever FFT analysis on different versions of SID chips :-D
This.
For sure this :)
Your ears have more dynamic range and a better sense of the quality of the sound.
@Ken Mason The SID chips we like best are the ones we had in our own machines we had when we were young. 😊 It's the same as when there's this one particular mix of a song we prefer even though it may objectively sound worse than others. Which is why I don't really believe in "audiophilism" (unless I get to experience it myself but that's unlikely as my hearing is somewhat ruined at this point), and find it to be a combination of preference, elitism and autosuggestion. But we all have our own weird quirks so whatever.
Icecast sounds interesting though, I may want to look into that sometime.
Maybe dual wiper headers would be better than turned pin as that's what the IC legs are designed for.
I loved it with you jamming out to the music lol Great vid. thanks
Surprised with regards the SID there isn’t an analyser of some sort, eg Android app that takes the audio out via a line in and checks (think guitar tuner apps)
Evie Salomon's BackBit I.C. Chip Tester?
@@MattKasdorf That would test if the chip is functional, not if it can generate the correct tones etc?
The filter on the R1 SIDs sounds very different than R4 -- there are some tunes that sound like they have missing basslines on the R1. I had an early model breadbin growing up that played Hubbard tunes (among others) completely differently than my friend's, and always thought there was something wrong with my system. Many years later, I blew out the R4 SID in my SIDstation and replaced it with a spare R1 I had from my parts bin, and manually sweeping the filter made it very clear what was going on.
Looks like their investment in acetone resistant paint has finally paid off
Thank you so much for the tip with the acetone, Adrian!
💪
I find myself almost addicted to your videos. Kind of in the same way as when we as kids would wait impatiently for the next episode of our favorite cartoon or TV-show. The fact that I don’t care at all for CRTs or that it is pure torture for a fellow sweet-loving diabetic to watch you munching on and (what feels like) endlessly talking about sweets… quite oddly doesn’t seem to have much impact at all.
😁
I think what I love the most is that we get to see when something goes wrong for you as well. It is kind of hard to learn from other people’s mistakes when we don’t get to see them. It also brings less of a teaching or tutorial feeling and more of a “just along for the ride” feel to it, which I really like.
I seem to have collected quite a crazy number of EPROMs and EEPROMs over the past few years. Some from eBay, some from AliExpress, and some from random web-shops, which I have forgotten the names of. Doesn't really matter as I don't remember which chip came from where anyway either.
Some of them work, some kind of semi-work, and some of them I couldn't get working at all. But like you, I always keep broken chips... hence the crazy collection.
For quite a while now I have suspected a portion of them not to be actual fakes but maybe resurfaced, painted, and relabelled.
Because many of my 27C160 EPROMs (the ones I have the most of) are marked as "M27C160", "ST", and "Singapore". Yet there is quite a variance in how they look. Some have a darker and almost black finish to the ceramics, others are light grey-brownish, but most lie somewhere in between. The key difference lies in the size and the partitioning of the exposed chip beneath the window. This makes me suspect that some of them may be smaller, like 27C400s or 27C800s only marked as M27C160. They may not all be from Singapore, maybe made by SGS-Thomson or some other brand instead.
We often find ourselves having strange ideas and biases. I don't know why I have never dared to use acetone. I have tried cleaning these chips with isopropanol and other cleaning solutions without revealing much. But if acetone is safe for human fingernails, it shouldn't really be too harsh on plastic EEPROMs and not at all on ceramic EPROMs, right? (doh!).
So, after watching this video, I instantly went and broke out the acetone.... and it obviously removes any paint with a vengeance.
Some of the EEPROMs have clearly been resurfaced and laser-edged. But I didn't expect to find that quite a few of my EPROMs have also gotten the same treatment. Because as you indicate, the presence of the glass window would seem to make that process harder to pull off. But no, the acetone did remove a ton of paint off some of those as well. A few of the semi-working ones had what to me seemed like an excessive amount of paint on top, spreading out onto the sides. There was even a thin layer on the pins, which I just hadn’t been able to see with the naked eye, and obviously why they were behaving oddly. But it sadly didn’t bring out any of the old markings on any of them.
I don’t really care if they are a different brand, speed, or even a smaller data-capacity. Because I can use 27C400s or 27C800s just as well. But the mislabelling is annoying when you are attempting to burn something onto them. Because my (up until now) favorite GQ-4x4 USB programmer needs an adaptor for burning these kinds of 16 Bits EPROMs. And this adaptor makes it no longer able to detect the ID of the EPROM, if the chip is seated wrong, or there is a problem with the connection to any of the pins. It just simply fails instead, and then all I can do is put the EPROM back into the eraser for another try 20-30 minutes later.
I have just gotten one of the fairly new XGecu T56 programmers, maybe I should try using that one instead.
Anyway, thanks again for the tip… at least now I have the means to “get right to it” as you say 😉
Greeting from Southern Jutland, Denmark 🙂
Yeah, that XGecu T56 programmer does seem to make things a lot easier 😉
I'm shocked the SID chip hasn't started being manufactured again in small quantities. I love the sound of that chip. I know some DJs and musicians have utilized the sid chip for producing music.
Twisted electrons made a c64 synthesizer called therapsid mk2 they haven't made any more because they don't have c64 chips
Having been through about 4 SID chips (which were WAY more than $25 each) trying to get my C64 repair/restoration finished... and each of them have had their own issues; I -think- I hate you now... If I were to order a SID for AliExpress it wouldn't even be an IC... it'd be a Dorito with some staples glued to it or something. How the hell do you keep getting working SIDs for such good prices!? It's killing me, man!
I did FINALLY get to order a SwinSID Ultimate and it'll get here from Germany, eventually... but I'd've much rather had a fully working, real, SID.
Keep up the great videos and keep up the C64 love!
34:50 - the suspence was killing me, I was rooting for the chip to work and when the right screen showed up I was cheering like a 4 year old in a candy store :D It just made me so happy that the chip is finally working :)
36:50 - maybe you could try to use a socket that has the 2 metal thingies (or is it just one bent into a V shape), cut it up and solder the sides in so you'll have more horizontal clearance and play on the chips instead of just a single hole per leg. Just an idea. It might not look as clean but it might be easier to take the CPU in and out. And run the bodge wires inder the socket or something, but that's just me...
Chip doesn't work. "Time to hit it with Deoxit!" :D Loved the analysis! And, yeah, there are some chips that have 2.5mm spacing instead of the standard 2.54 (i.e. 0.1 inches).
I'm across the river from you on the Washington side-- I'm set up for x-ray inspection of chips and boards and the like, so if you ever want to find out (non-destructively) what's inside those packages and compare them to a known-good one, give me a shout!
Could it be, that the "TMS 9900" actually is a 68000? They both have the same package and the M68k actually is a lot cheaper to get, from what little I know.
Wasn't the TMS 9900 the first single-chip 16bit CPU? I seem to recall they kinda set the standard on the 64-pin DIP packaging, which was later used by a bunch of other chips, usually high-spec stuff like the 68000 or high-integration parts like the Z180. That said, the package notching and markings look like TI's TMS 9900. If I had to guess, it's a legit TI part, but was probably just toast. If it's not a dead "refinished" TMS 9900, then it's some other TI part that came in 64bit DIP packaging.
I've not seen a m68k with such large circular marks, in particular the semi circular mark on the left. They seem to be smaller on Motorola DIP64 packages.
So we paid the Chinese to take our e-waste. They then sold it back to us. That's one awesome business model!
I used to repair marine radios and the TMS9900 was used in some radios. I was surprised as I thought it was a 99/4a exclusive.
One of our New Zealand marine research vessels - RV Tangaroa - has an onboard comms system powered by a TMS9900 from one of my TI99/4as. I got a tour of the ship as a thank you :)
Ah the TI-99 that brings back memories, I used to do in-house repair of those back in the day. Right before they were closing down all of the production lines for those we could buy them for $50.
The hand dance is hilarious. Never change man, never change. :)
Re: process to make IC legs look new... They will straighten them, brush them, then dip in molten solder in an effort to make them look new. Most of the dram, eproms and old chips from China seem to arrive here like that anyway. I've had problems with them 'binding' in sockets, or introducing reliability problems, so I try to solder them in directly if I have to use these.
They may not be fake. Most of these chips are recycled from old machine parts, clean and legs re-tinned. It's a gamble because the seller does not has the means to test them out.
It's my understanding that IC chips are full of discrete components. So would it not be possible to essentially unpack a SID chip and make it large scale?? If possible, I'd like to see that done as a project here.
Usually they contain just a single silicon chip. You could potentially reverse engineer it and make something fully discrete, bit it would not be trivial.
@@TomStorey96 and that silicon chip contains all the discrete components. I realise making a full size replica wouldn't be easy. And the process of grinding the surface of the chip down to the wafer and then using an electron microscope to look at the contents is beyond any normal person's capability - but if there are schematics of the dies inside, it would be interesting to see it done and to see what sort of sounds that version would make.
I need to buy backup chips for my colecovision and atari. Just in case
The use of machine sockets will make it harder install the TMS cpu, but the cheaper sockets may in fact be a better choice since they have spacing to better account for the slight pin spacing.
I am too silly, if I could code I definitely would put you and Rami together dancing to the Donkey Kong song 😉😂👍
Adrian, nice SID chips. I study these enigmas, trying to build a mega synthesiser and I’m buying as many as I can.
I read that MOS grew them organically from silicon or something, and why can’t we do it 37 years later, but nevertheless my construction is already the size of a table, involves 24 SID chips. I wanna blow the windows off anywhere I take this thing.
Considering using a Leslie speaker, and being an audio engineer, I’m just collecting 64s and chaining via midi. I need help.
My 6581 SID chip was soldered. Only socketed chip on my board was the VIC-II… until I started socketing others. ;)
This brought back memories - so many hours of my life swapping around components to try to keep a system running. In the early '90s I ran a computer room at a small school. When I first got there it had about 20 donated IBM PCs, and as each one failed I would replace its bad part with another from a machine that had failed in some different way. I managed to keep the place going on a shoestring, but I don't want to think about how many weekends I spent on the floor of that room trying to get some Frankenstein's monster of a machine to last one more month. By the time we closed the room ~8 years after I arrived, I think we were down to 12~13 working systems (and I broom closet full of scrap parts).
Can you upload the 8-Bit Dance Party demo to The Archive or CSDB?
And can we also get Adrian‘s tools as an EPROM image? Or maybe even the full cartridge to buy somewhere? :)
Whenever I socket something with a nonstandard pin spacing, I take a gold plated dip socket with a compatible pin size and use nippers to pick off one pin at a time and after its soldered in, clip off the rest of the plastic,, then if you want to get real fancy, you can use pieces of poster board to build a wall around the pins on top of the board and backfill with potting compound to the height of the pin sockets.. Either way, no funky Bending.
I was really hoping that fake CPU didn't blow the rest of the machine!
Less colored circles, more accurate VUs :P
Oh and the Donkey Kong splash screen on the C64 in the background would be fitting :)
Very nice demo to whoever wrote it! :)
Pin spacing in CPU is metric? 2.5mm vs 2.54mm (0,1"). Soviet chips were 2.5 spacing.
No, it's regular 0.100". The problem is inserting a desolderd 64 pin chip with very slightly bend pins into an precision turn DIP or actually 2 rows of precision turned SIL...
(Pin diameter typical ~0.150", hole diameter of a turned socket typical ~0.200") 64 pin DIP is the only case in which I would prefer a regular socket.
Even if it they were new from the factory these 64 pin DIP chips have to be hard to insert in a socket like this. There was a reason why they stopped at 64 pins...
I do how ever love your energy . im glad at least most of the chips worked.
Lucky you that you got 2 genuine, working SIDs from China. I recently bought 10 6581R4’s from two different electronics parts vendors out of Hong Kong. The picture they showed me before purchase was of a genuine 6581R4. (I actually already had one with totally identical markings as their photo.). What I received were clearly re-marked chips. I didn’t have to do the acetone test to check, plus they had return policies and didn’t want to “damage” them from the acetone test and risk not being able to return them. In the end, I got my money back. The strange part is that of the 20 chips, many of them did not register as either a 6581 or 8580 in my C64 Reloaded board, but several did. Of those, most had some degree of functionality of a SID, so it seems like what they do is either: 1. Mark a completely different chip with the same package to be the chip you’re ordering (in this case a 6581R4 SID), or 2. Clean up and re-mark an older SID with the newer markings to pass it off as the newer version but don’t check if the chip actually works. I did find a few SIDs that were maybe 60% or 70% functional, and even found one that was probably 99% functional, but had to send it back to get my refund.
Everyone is buying the 99% isopropyl to disinfect stuff...
the 70% is better for disinfecting. 99% evaporates too quickly.
I have very fond memories of growing up with a TI 99/4a. I remember we had some math game, Jawbreaker II, and Pole Position for it. We also had a game called Tunnels of Doom that loaded off of a cassette (we pillaged the cassette player from our home organ at one point to play it). My brother still has it all these years later and he can't wait to show my niece how to use it.
Today I fixed 2 Quickshot Turbo II joysticks, both had missing wires, one was just yanked off, one was cut. I had to tinker with the wiring with the new cable (used a 9pin dsub extension cable and cut off one of the ends), cleaned the microswitch contacts and they are just like new, I'm so happy!
Good work, it's always nice to repair things. I'm currently working on restoration of a 1981 'fruit machine' from an amusement arcade
What did you do for strain relief where the cable comes through the plastic shell? I'm curious because i would like to find a good method myself. I've tried with 4 zip-ties in a little square on both sides of the shell, but i think there must be a better way.
@@MattyEngland Wow, sounds really nice! I love restoration works, it's just so satisfying to do or watch.
@@BertGrink One of the joysticks still had the strain relief on the cable, I janked out the old cable and put it on the new one and put some electrical tape behind it so I won't be able to pull it through the strain relief plastic. The other one sadly didn't have the rubber strain relief so I just put a huge winding of electrical tape to keep the wire inside the shell so pulling on the cable won't stress the soldering. You can reuse old strain relief rubbers, they are not permanently attached to the wires and I also wanted to buy some online but it's hard to find the right size. Luckily the my new wire was the same size as the old one.
@@Dukefazon I think you have been quite lucky then; most strain reliefs i have come across have been bonded to the outer isolation of the cables, making it very difficult to separate the cable from the strain relief.
I love the 8-bit dance parties. It's been too long. Thank you Adrian!
Wow. Thanks for some great info.
I'd like recommend also InSIDious which is a totally Faithfull vst recreation with patches programmed by Rob Hubbard, Jeroen Tell, Martin Galway etc...
Needed something to watch while entering data, more listen, but does this ever fit the bill. This kind of video fascinates me. That cart with the song is awesome! Love that sound.
A tip: Chrome has a picture-in-picture addon you can download from the extensions page (made by Google, dunno why it's not just included). Then you can press ALT-P to pop any HTML5 video player into a little resizeable always-on-top window. Probably still not great to use at work, but plays really nicely with games - I use it a lot when I'm playing Runescape.
@@GdotWdot just have my cell phone next to me and glance at it when he’s talking about something I need to see. Thanks for the tip though. Going to check that out for home.
AliExpress seems to have more than their share of "wrong tracking number" scammers IME.
@@xabhax Yes, some of them do that.
Others don't ship at all and then claim that your package was delivered, based on having given you someone else's real tracking number. This is the scam I'm referring to.
Yes. Details in tracking will show it going through some weird locations.
@@andycraig7734 When I see a packet having been delivered in Russia, I apply for a refund. Got it every time so far. And most of the time a couple weeks later the real packet shows up. So yeah, I'm fine with them hurting themselves...
@@HenryLoenwind to be fair they often travel through Russia
@@ligametis There are daily freight trains from China to Germany, but normally such goods would be using airmail. These never land in Russia because they take crazy fees from the airlines.
Great video, I have a working C64 from 1982 which my father bought me as a child. Still use it today
Thank you for the video! I enjoy learning about these counterfeit or fake ICs, and it is good to have more awareness.
For the leg cleaning, I have been able to do this with one of those solder pots, melting pot to have the pool of molten solder, like the kind you might use to tin the tips of wire. If you put some flux on the pins and then quickly dip the entire legs into the solder it comes out to look like this for me anyway.
I know its not really related to this video.. but I really wanted to ask, how do you get your audio so good? Your voice seems very clear and articulate. and no background / room noises, but also I don't hear artifacts of a noise cancelling too. Is this just from that Rode Go lavalier mic!?
34:46 The chip works! Keep doubting whether something is broken or not!
"Because of human malware" has to be the best filter-dodge euphemism for covid I've heard for a while
These are quickly becoming some of my favorite videos on the channel! Chip Busters! Is what the wife and I call it!
for a quick reference - TI TMS 9900 data sheet at
datasheets.chipdb.org/TI/9900/TMS9900_DataManual.pdf
For some reason I thought the TI-99 you repaired was silver.
i see that the more expensive ICs on your longboard have passive cooling. So i guess you recommend the use of passive heat sinks. But i wonder, if you would recommend heat sinks on the shortboards with the later variants of those ICs too?
31:50 that socket has a 45° angled shape that designates pin 40 for some reason.
And that fake TI CPU... I would have measured ohms on some select legs like last time you checked fake ICs
The 100- AMD-27c400's i got fro0m china were actually Micronix 27c4100's pin compatible but the eprom burner identified it internally. remarked. Chinese chips are like a box o chocolate,you never know what your gonna get.
That picture of you in "Adrian's Dance Party' is "SQUEEZED" by 3/4!
Square pixels weren't a thing on home computers until VGA.
Wait. Wasn't the TMS9918 chip used in MSX computers? Afaik they were quite widely used at least in Japan?
The original MSX did, yes.
changing the branding is simply cause certain chips get more money
the age of the CPU chip I think anyone would be happy with pin to pin compatible replacement, with any name on it? if it got there computer going, newer ones would probably be capital of running faster with being newer, as with most CPU's
Have you found circuit digrammes for the pal euro version ?, I have not !!!!!!!!!!!!!!!!!
u have not !!!!!!
@@uiopuiop3472 why you kow ?
@@johndavis577 why me kow !!!!!
Many old devices will not function properly due to moisture ingress - very slow over time - when the device is not in service for a long time. As in - no heating.
A time-honored solution, particularly with memory parts, is to bake them for 12 - 24 hours at a temperature of 95 to 110 degrees centigrade.
This technique has been used in PCBA facilities for many years when the devices have been outside their original foundry shipping sealed packaging beyond the prescribed warehouse-storage-life of the device.
Have you done the Stereo SID mod yet? you ought to try and make a stereo sid plug in card!
Its also worth noting body static WILL kill these cpu's.
Adrian can you review an ARMsid?
Hi Adrian, I see some 24 pin EPROMs there, WSI 57C49C and TMS2532. I have been looking for some drop in replacement EPROMs for Commodore's MOS2332 and MOS2364, seems like you've got what I've been looking. How did you program these? My TL866II+ doesn't have them in it's list.
I've seen it where the fake chips are just empty carriers marked up... so no active die inside there. IMO if it fails the acetone test, its junk and unless there is some indication of what the part actually is, it would go into the bin rather than risk damaging the circuit its being used in
Adrian in his basement, cup of coffee, sit back, good times. Had a dream last night about a game called Nodrub’s Quest and woke up to some Ti99 video. That’s called a kwinkidink
[Everyone runs to AliExpress to look for $20 legit SID chips, finds they were all gone a year ago] ;-)
Yea methinks that 8bitguy was complaining just a tad bit too much about gettung 30 year old chips.
In regards to the EPROMs; It's possible they weren't trying to fool anyone to accept substandard chips, but rather make it more probable that they find potential buyers (people who want replacement parts who doesn't know which chips WOULD work the same in their computer, so they just search for the one type they KNOW works).
That #8bitdanceparty program is just epic! :D
I recently got a batch of fake DRAM (batch of 20 - all exactly matched and looking way too new, plus incorrect US date codes on a Japanese part!), however they survived the acetone test!!
Looking closely I could see they were sanded, then maybe sand-blasted to give them a texture and finally laser-etched.
Ie. no paint used in the process!
Just something more for folks to watch for!!!
that non working TMS9900 CPU...64 pins
I wonder if they relabelled a motorola 68000 which also has 64 pins ?
Might be worth trying it in an atari ST or amiga
The pin spacing oddity would not happen with a 68k so not worth burning a mobo to test that.
@@benbaselet2026
what pin spacing oddity ?
both are 2.54mm pin pitch
The TMS9900 could of been a 68000 just marked as a TMS9900 , as they both are 64 pin DIP...?? So might be worth trying it in an ATARI ST or An Amiga ??
The 82 sid chip is a Revision 2 i have the exact same datecode sid in my Ultimate64, the later AR has more Bass and the Revision 3 is the Screamer Sid!
Fun video. I especially liked the fan-made 8 Bit Party - it put a big smile on my face. May suggest you take keys from the TI 99/4A that needs repair and put it on the one you fixed last year to complete it?
I guess the re-brand can made them easier to sell, since some younger people know nothing about Hitachi or STM EPROM, I got an re-brand AMD also, TL866II detected the ID is STM. Famous chips can sell higher price.
The fake TMS9900 could very well be a 68000 or 68010, it has the correct number of pins (64) and looks almost identical to 68000 and 68010 parts I have. It'd be interesting to try it and see if it works after the 12v insult it got. You can put a socket in an old compact Macintosh from the original 1984 Mac to the SE, which used the long DIP package 68000. The later classic I think changed to the QFP variant.
The limiting factor is I think the pin pitch is not the standard 0.1" / 2.54mm, and is a bit narrower, so getting a socket may be a bit of a challenge.
I have a solution for the pin spacing. Permanently mount the cpu in a ic socket and then the spacing will match putting it in another of the same socket by putting one socket in another!
Hard to notice but are u using a ESD wristband, cos i would with those precious MOS chips for sure. For any handling of PCB with ESD devices i would use one.
Strange, according to the TI datasheet for the TMS9900 processor the pin spacing is 0.1", which is totally standard. Admittedly, this is for the ceramic package, but why would they change this for the plastic package?
FYI the reason that Ti used the TMS9900 because that was what their industrial machine tool controllers were based on. So there were more out there than you knew about, but still not a lot.
What is the intro song? It's really good
they HASL the chip pins, hot air solder leveling,like on some pcbs.
i AM the munch MAN
im going to need a full version of the sid version of the dkc1 intro music
For the IC legs, they just re-tin them. You can see they are a little lumpy now.
i had the same impression regarding the old 6581 version. it sound distorted. not pleasant for my ears at all. this is the advantage of newer re-implementaions, they all do the filters digitally which gives higher precision (almost) without distortion - if correctly done.
The TI 99 and TI 99/4a are different systems.
I actually have a board that uses the TMS9000 chipset on it, used as a serial terminal interface....
Does have the socket as well, with that special spacing. Likely the Ali one is the OTP version of the TMS system, they did make some with built in ROM and RAM, as it is not likely to be a 68000 series, unless TI was acting as second source for them. Could also be one of the TI DSP families, as they used that package for them early on as well, needing to get all the pins out, and having a parallel data and program bus to interface with the programming.
Had the same problem with the AMD 27C400 ROM's from Aliexpress, It schould be AMD but the first batch was Toshiba (TC574200) with a big erase window and the paint was still wet and the second batch was Intel (D27C400-150V10) which had an accesstime of 150ns instead of 95 ns.
I buy my acetone from Sally Beauty Supply. It comes in several different sizes up to one gallon and it is very pure.
HAHA I buy acetone from the hardware store when I'm not using it for my nails.
@@megan_alnico lmao
Great investigation of the chips. Nice and detailed /cheers ScrapBongo
Every time I see a TI99 I find out more about its weirdness. We were considering one back when I was a kid but I got a Vic 20 instead, the right choice I think