PC@LIVE wrote on 2024-06-09, 21:56:

Thermalwrong wrote on 2024-06-09, 20:08:

PC@LIVE wrote on 2024-06-09, 14:11:

I had an ASUS CUV4X in my hands, which needed to be tested, with a P3 370 733MHz, after the first attempts didn't go well, I understood that there was a RAM problem, I tried different SDRAMs, and the third module I tried, solved it the problem, the PC started, it's alive!!!!!!!!!!
While waiting to gather more hw, I tried booting from DOS, and launched a couple of programs, diagnostics and bench.

That heatsink seems like an odd one for a pentium III, is that maybe from some Socket 7 system?

Yes, it's not from a socket 370, it should be from an S7, perhaps coming from a Compaq PC, for temporary use anyway, that's fine, I'll put another one in when I've finished, and found the right mix between CPU RAM and cards, not I have decided nothing, but I think of increasing the RAM, putting a suitable AGP VGA, and if I wanted I could change the P3 with a more overclockable one, for example a 700, or I can simply change it with an 800-1000.

Oh, I thougt it was a big heatsink from being in a system with semi-passive cooling 😁

zuldan wrote on 2024-06-10, 05:59:

If you don’t mind me asking, what is the name of the DOS program that is showing the CPU and Motherboard information?

Astra32 v7.0 for DOS from Floppy image:
http://www.sysinfolab.com/download.htm

It's part of system tools on Ultimate Boot CD: https://www.ultimatebootcd.com/

Just got awe64 from eBay today
Can anyone tell me if it real or fake

stef80 wrote on 2024-06-10, 07:31:

It's part of system tools on Ultimate Boot CD: https://www.ultimatebootcd.com/

That reminds me. I really need to get my PXE server set up to serve things too big to be practical for PXELINUX's memdisk chainload module. (eg. Adding an NFS server so I can efficiently offer up Linux boot options.)

Currently, my menu is just memtest86+, a bunch of 1.44MiB and 2.88MiB floppy images for things like PartitionMagic and various boot disks, and a Damn Small Linux "everything in the initrd" setup that isn't useful for anything with less than 128MiB of RAM. (And is more an X11-based tmux than a GUI, as far as which apps will start goes, when netbooting with only 128MiB of RAM.)

EDIT: Seriously, if you haven't used PXELINUX's menu module to set up a "homebrew Apple Internet Restore for retro PCs", do so. You'll thank yourself... the secret sauce to it not being painful is the dhcpdump tool. I was just reminded how annoying it is to not have when I bought an Igel 3/2 thin client and discovered that, unlike my newer HP thin clients, it ignores the DHCP server's instruction to look to a different IP address for the TFTP server. (Luckily I run OPNsense for my router, so that's fixable, if annoying, by moving the TFTP server to it.)

...which reminds me of another thing to do: Open up my AST Adventure! 210 and check what kind of SIMMs I need to buy to upgrade it from the 80MiB of RAM I happened to have lying around to the 128MiB it can take.

DataDragons wrote on 2024-06-10, 08:31:

Just got awe64 from eBay today
Can anyone tell me if it real or fake

Why would it be fake? It looks like any other CT4520 I have seen and I have never heard of a fake Sound Blaster of any type. There are cards that are functional sound blaster "clones", but they are clearly different, always had their own unique name and I don't believe any clone of the AWE32\64 line was ever made.

https://www.google.com/search?client=firefox- … ih=944&dpr=1.33

Nothing much. Shopping for some Startech adapters as I have a few OGXbox bricks to throw SATA drives at 🤣

I designed a simple support that lets me keep the mt-32 on top of the sc-55.

I spent the better part of my day getting things somewhat organized while setting up a new computer space in my family's house. I also attempted to see if I could find an MT-32 hat for my Pi, but it seems that the MISTer has ruined any chances of me looking for non-MISTer versions.

BetaC wrote on 2024-06-10, 22:37:

I spent the better part of my day getting things somewhat organized while setting up a new computer space in my family's house. I also attempted to see if I could find an MT-32 hat for my Pi, but it seems that the MISTer has ruined any chances of me looking for non-MISTer versions.

Would this work?

https://www.tindie.com/products/retrofletch/b … ro-pc-and-more/

Tried flashing EEPROM chips for the Trangg Bow AMD286 with Headland HT12A motherboard, which is from a CNC machine, as it turns out. So far no luck, just BIOS checksum errors. As the flashing happens without errors, I can only assume its the bios itself. Silly question - The original CNC BIOS had 'low' and 'hi' chips, and a lot of bios have 'odd' and 'even' . Which corresponds to the other? I thought low would be 'odd' and high would be 'even'.

BitWrangler wrote on 2024-06-10, 00:16:

Thermalwrong wrote on 2024-06-09, 20:08:

That heatsink seems like an odd one for a pentium III, is that maybe from some Socket 7 system? […]
Show full quote

PC@LIVE wrote on 2024-06-09, 14:11:

I had an ASUS CUV4X in my hands, which needed to be tested, with a P3 370 733MHz, after the first attempts didn't go well, I understood that there was a RAM problem, I tried different SDRAMs, and the third module I tried, solved it the problem, the PC started, it's alive!!!!!!!!!!
While waiting to gather more hw, I tried booting from DOS, and launched a couple of programs, diagnostics and bench.

That heatsink seems like an odd one for a pentium III, is that maybe from some Socket 7 system?

I got hold of a Sony PRD-250 CD-ROM Discman player, which is a SCSI CD-ROM drive but is missing its cable and I've decided I like SCSI drives now. But it didn't read discs at all, just hunting around when trying to play pressed audio CDs.
Using a junky webcam, I confirmed that the laser was working and the focus system was working well, the carriage could move freely so it was probably a laser problem, like it's worn out.

I was getting nowhere and was going to give up then realised I have nothing to lose since they're not common drives and I had found the PRD-650 service manual which gave me some understanding of the TEST mode it could be put into by soldering together a solder jumper. I also learned that the laser assembly (DAX-02S) is probably irreplaceable.
I've never fixed a CD player before but I have two oscilloscopes, one is the Owon HDS2102S portable scope and the other is the Rigol DS1054Z - the latter gets used less because it is big and cumbersome in my opinion, but my view of it improved while trying to fix this drive. The control system is so much nicer than on a portable scope, and the screen is bigger.

So anyways, I took the player apart and started by taking pictures of everything - all the trimpots original positions and I tried marking them, this probably saved me.
IMG_3622 (Custom).JPG
Because I hooked up the test points RFO (RF output?) to see the diamond pattern, VC as the ground point, TE for the E-F balance. Initially I tried using the Owon scope but I didn't know about its persistence settings so could not get anything resembling an 'eye' pattern that the service manual says to look for. Then settings were adjusted wildly for a while.
IMG_3624 (Custom).JPG
I then switched to the Rigol scope, watched some videos of other people looking at this signal and started to understand what I was looking at. Soon after I could see something resembling the diamond type pattern but the voltage levels (vpp) were too low and it was noisy, so the laser did need boosting I think...?

Took off some capacitors to test whether those could be causing issues and no, they're not leaking and the ESR values were in spec for the 2 I tested.

Through the course of the night I adjusted everything, all the trimpots though only 3 mattered -
- The trimpot on the laser diode itself which of course the service manual didn't mention because that's set in the factory. It started off at something like 1.7v and I put it up to 1.71v or something - that was measuring on some test points by the laser diode so I'm pretty sure that's the voltage it's being fed?
- Then the Focus Bias adjustment which is the one that really made the difference, very finnicky adjustment
- The E-F Balance which seemed to set the high/low of the voltage and also made a difference to the eye pattern, in that the eye pattern only looked good once the E-F was set in the middle so the upper and lower waveforms matched and weren't clipping
There's also the Focus / Tracking gain adjustment pots but adjusting them did nothing positive so I pretty much have them back in their original positions which worked best.

I probably had the player functional last night since it was playing audio sometimes but broken up - that's going to happen in test mode because the carriage can be moved in/out using the |<< and >>| buttons, it's not playing an audio track, just a repeating bitstream.
Eventually I gave up thinking there was no hope and went to bed.

Got back to it in the morning and somehow I had the player reliably making sound in test mode, then playing audio and giving a nice 'diamond' / 'eye' pattern in something like 20 minutes of fiddling with it again.
prd-250-repair-eye-pattern.JPG
From the above picture I realised the amplitude was now too high so put the laser voltage back down and the RFO test point now shows 1.2v vpp which is in spec I think. It's now adjusted just a little bit higher than its original setting.
prd-250-repair-eye-pattern-2.JPG

Hmmm my 1Mhz scope is suddenly sounding more useful.

Yeah, I just don't have the room for an analogue scope, I barely have enough space for the DS1054Z right now... Has anyone else adjusted the laser on a CD drive / player before? I don't know whether I did the right thing by boosting the laser power or if I could've fixed it just with careful recalibration, I've read and seen some conflicting ideas on the topic.

I've had a testing backlog that I've been building up for a while and the other day decided to get through it.

My Leadtek FX5700LE that would start a benchmark but crash in like a second of running - I replaced its capacitors and checked its voltages and those weren't the fault. In the end it was fixed by re-fitting the heatsink with screws instead of the spring push-pin type things. Now it seems to actually work which is great because it wasn't working and I fixed it. But my fix is a stupid hack because the real fault is the 'bumps' on the flip-chip aren't good anymore.

Gonna have to try this on the Radeon 9800 card I have as well. I thought it worked 100% but the card seems to give artifacts sometimes and bending the card a bit makes it better/worse.

Then was testing some Pentium II / III slot 1 cpus. Some were ones that I'd tested as not working a few years ago, but most of these were mystery CPUs that came from a scrap / gold lot so the cases were gone, components bashed off of some. And my new Coppermine 650MHz with no heatsink - I was scared of crunching it so I put some 0.8mm PCB scraps on there, these also push on a copper shim which fills the gap left when I removed the chewing gum thermal interface that the CoolerMaster SECC2 heatsink had on it.

Some of the CPUs had little crunch marks on the CPU core but it seems that early intel flip chips were fairly forgiving. All the mystery CPUs were PII-350 / PII-400 so nothing special 😒
One CPU started working after I refitted a broken off inductor and some little bypass caps. The other ones that didn't want to work all worked after cleaning the edge contacts with isopropyl alcohol. Every single CPU worked in the end, the coppermine seems much faster than the katmai & deschutes processors

On the fence over whether or not I drill holes in my Agilent Arcticooler to make it a SECC2 cooler, which I don't have enough of.

Thermalwrong wrote on 2024-06-11, 02:00:

Has anyone else adjusted the laser on a CD drive / player before? I don't know whether I did the right thing by boosting the laser power or if I could've fixed it just with careful recalibration, I've read and seen some conflicting ideas on the topic.

I thought, I know, I'll get you an authoritative view on the subject, my repair tome, "Compact Disc Player Maintenance and Repair" McComb/Cook 1987 ... and the only thing about adjusting the laser in there is below.... yeahhhh now I know why this book stays on the shelf..

Flicking through it again, it pads out a load of really "Noddy" stuff, like making sure you've got your audio plugs in right holes and mixer set to right input, no shit Einstein.

DataDragons wrote on 2024-06-10, 08:31:

Just got awe64 from eBay today
Can anyone tell me if it real or fake

That's a real AWE64 sound card.

Ozzuneoj wrote on 2024-06-10, 19:52:

Why would it be fake? It looks like any other CT4520 I have seen and I have never heard of a fake Sound Blaster of any type. There are cards that are functional sound blaster "clones", but they are clearly different, always had their own unique name and I don't believe any clone of the AWE32\64 line was ever made.

There are castrated versions of the CT4520, without the EMU8011 chip and RAM upgrade headers. Those gets recognized as a SB16 instead of AWE64.

Gmlb256 wrote on 2024-06-11, 04:09:

That's a real AWE64 sound card. […]
Show full quote

DataDragons wrote on 2024-06-10, 08:31:

Just got awe64 from eBay today
Can anyone tell me if it real or fake

That's a real AWE64 sound card.

Ozzuneoj wrote on 2024-06-10, 19:52:

Why would it be fake? It looks like any other CT4520 I have seen and I have never heard of a fake Sound Blaster of any type. There are cards that are functional sound blaster "clones", but they are clearly different, always had their own unique name and I don't believe any clone of the AWE32\64 line was ever made.

There are castrated versions of the CT4520, without the EMU8011 chip and RAM upgrade headers. Those gets recognized as a SB16 instead of AWE64.

Right, but they aren't fake. It's just a sloppy attempt by Creative to reuse old AWE64 PCBs for a later budget-oriented product (as late as 1999 from what I've seen) without bothering to change the model number on the board. They don't have AWE64 written on them anywhere, and are clearly missing the parts that would make them an AWE64. I'm surprised anyone even knows about those fairly rare cards without also knowing that the missing chips are what makes them undesirable.

Anyway, yeah it's real. Enjoy. 😀

Ensign Nemo wrote on 2024-06-10, 23:08:

BetaC wrote on 2024-06-10, 22:37:

I spent the better part of my day getting things somewhat organized while setting up a new computer space in my family's house. I also attempted to see if I could find an MT-32 hat for my Pi, but it seems that the MISTer has ruined any chances of me looking for non-MISTer versions.

Would this work?

https://www.tindie.com/products/retrofletch/b … ro-pc-and-more/

Thanks, I'll give it a look.

Currently working on my newly acquired Mac Classic. Removed corrosion from the logic board and it boots up better, but the speaker doesn't output anything.

Ordered cap kits and it'll be fully repaired once I get said caps.

Thermalwrong wrote on 2024-06-11, 02:00:

Then was testing some Pentium II / III slot 1 cpus. Some were ones that I'd tested as not working a few years ago, but most of t […]
Show full quote

Then was testing some Pentium II / III slot 1 cpus. Some were ones that I'd tested as not working a few years ago, but most of these were mystery CPUs that came from a scrap / gold lot so the cases were gone, components bashed off of some. And my new Coppermine 650MHz with no heatsink - I was scared of crunching it so I put some 0.8mm PCB scraps on there, these also push on a copper shim which fills the gap left when I removed the chewing gum thermal interface that the CoolerMaster SECC2 heatsink had on it.
IMG_3612 (Custom).JPG
Some of the CPUs had little crunch marks on the CPU core but it seems that early intel flip chips were fairly forgiving. All the mystery CPUs were PII-350 / PII-400 so nothing special 😒
One CPU started working after I refitted a broken off inductor and some little bypass caps. The other ones that didn't want to work all worked after cleaning the edge contacts with isopropyl alcohol. Every single CPU worked in the end, the coppermine seems much faster than the katmai & deschutes processors

maybe the examples weren't special as such but rescuing them all is - especially from scrap, maybe these will find homes once more, in due course

iraito wrote on 2024-06-10, 20:29:

I designed a simple support that lets me keep the mt-32 on top of the sc-55.

Have you got a picture of it printed?

zuldan wrote on 2024-06-11, 07:47:

iraito wrote on 2024-06-10, 20:29:

I designed a simple support that lets me keep the mt-32 on top of the sc-55.

Have you got a picture of it printed?

I'm the the process of priming and sanding, after that I'm gonna finish with a dark gray.

BetaC wrote on 2024-06-10, 22:37:

I spent the better part of my day getting things somewhat organized while setting up a new computer space in my family's house. I also attempted to see if I could find an MT-32 hat for my Pi, but it seems that the MISTer has ruined any chances of me looking for non-MISTer versions.

I built this one for myself and it works perfectly: https://github.com/chris-jh/mt32-pi-midi-hat

Serdaco also sells a very nifty one for 3A but works on 3B: https://www.serdashop.com/MP32L