LowSpec486 wrote on 2023-11-19, 15:16:

DerBaum wrote on 2023-11-19, 13:36:

Slowly we should realise that we just buy stuff for the dump from china.

Make no mistake. ALL hardware you buy, is from China.

China makes Apple kit as well as counterfeit kit.

Sure.
The quantity of unregulated poor quality products flooding the market for decades is also ruining the rest of the industry there.
The fake stuff is a huge part of the economy. If that goes away they basically have nothing desirable left for private buyers.

I can imagine if the market for pre made trash goes away that the goverment has to raise taxes on good quality products sold.
Wich in turn would make china less desirable as a production place for proper tech.
This is why i think the trash is hurting not just us but also china itself...

DerBaum wrote on 2023-11-19, 15:25:

LowSpec486 wrote on 2023-11-19, 15:16:

DerBaum wrote on 2023-11-19, 13:36:

Slowly we should realise that we just buy stuff for the dump from china.

Make no mistake. ALL hardware you buy, is from China.

China makes Apple kit as well as counterfeit kit.

The quantity of unregulated poor quality products flooding the market for decades is also ruining the rest of the industry there.

I don't know were are you getting your facts from, but the industrial sector in China is not precisely in ruins.

China has a tiered industrial sector. It can produce whatever you want up to any level of quality you may demand - provided you are ready to pay a premium for high quality produced kit. It so happens that the world is a big place and there is a world wide high demand for cheap low quality kit - because most of the world happens to be poor, you know.

Kahenraz wrote on 2023-11-19, 11:55:

GigAHerZ wrote on 2023-11-19, 10:21:

vutt wrote on 2023-11-18, 14:30:

I upgraded my NAS op system from ubuntu 18.04 to 20.04 LTS. Result: My Win98SE box could not connect to my read only transfer sh […]
Show full quote

I upgraded my NAS op system from ubuntu 18.04 to 20.04 LTS. Result: My Win98SE box could not connect to my read only transfer share for v1SMB retro boxes.
It looks like distros are killing v1 protocol from default configurations.

So adding this section did the trick for me:

Copy code to clipboard

1    client min protocol = NT1
2    server min protocol = NT1
3    ntlm auth = yes
4    lanman auth = yes

I very strongly recommend you not do that on your NAS! SMBv1 is very unsecure protocol.

Create a simple VM with some minimal linux instal (e.g. debian netinstall) and configure it to mount your nas shares and then re-share them with SMBv1 support. Whenever you need your retro machines to access nas, you just run your premade vm. And when you are done, you shut the vm down. (NB! You can also use that VM to mount iso images and the drive can also be shared for retro usage - there's your "CD drive emulator"!)

Behind a NAT firewall on a private home network, there is no reasonable surface area to be attacked. It's not a big deal.

These days you have oh-so-many devices in your private network, doing who-knows-what and sometimes even drilling upnp port-holes in. Peoples' homes are littered with "IoT" and "Smart" devices, any can be an entry point. So there is quite a bit of surface. 😉
And oh boy, when someone was reading such "tutorial" and is going to follow them with their every-day laptop. 😁

But what had to be said, has been said. Some people still insist on not putting fire alarm in their ceilings, too...

Come benchmark with us!! Netburst: Aiming for the Stars

LowSpec486 wrote on 2023-11-19, 18:29:

DerBaum wrote on 2023-11-19, 15:25:

LowSpec486 wrote on 2023-11-19, 15:16:

Make no mistake. ALL hardware you buy, is from China.

China makes Apple kit as well as counterfeit kit.

The quantity of unregulated poor quality products flooding the market for decades is also ruining the rest of the industry there.

I don't know were are you getting your facts from, but the industrial sector in China is not precisely in ruins.

China has a tiered industrial sector. It can produce whatever you want up to any level of quality you may demand - provided you are ready to pay a premium for high quality produced kit. It so happens that the world is a big place and there is a world wide high demand for cheap low quality kit - because most of the world happens to be poor, you know.

Like i said these are no FACTS! just thoughts.
I got so much trash over the years that i will not order anything from aliexpress, wish or temu even you can find neat things there.
The risk of getting scammed there is extremely high.

And the only way to show that i am not happy with that "straight to the bin quality" is to simply not use these platforms.

Benchmarked a couple ISA cards in my 386 DX-40 build.

Diamond Speedstar 24 (Tseng ET4000AX)

3D Bench: 15.8
PC Player Benchmark (320x200): 3.8
Doom (high detail, full screen, 320x200): 7.0

Diamond Speedstar 24X (Western Digital WD90C31-LR)

3D Bench: 15.1
PC Player Benchmark (320x200): 3.7
Doom (high detail, full screen, 320x200): 6.7

Conclusion: The Tseng ET4000AX is the faster of the two, at least in my 386. This contrasts with other benchmarks I've seen putting the two cards neck-and-neck.

Did even more video card testing.

Starting a Pentium Pro 200 build and trying out different PCI video cards and seeing how they perform.

Unfortunately the Diamond Stealth 3D 3000 has a vertical line artifact in the middle of the screen in mode 13h. And for some reason it won't run Quake in 640x480.

Matrox cards seemed to do okay though.

I put Windows 2000 on a 486 DX2-66 for the lols this weekend. It's on a 4GB CF card accessible through XTIDE. Has 32MB of RAM and has about 4-5MB free when it's idle. 3Com 3C509 brings it online for a bit.

I initially installed SP4 without smartdrv which took ages. Then I installed a stripped down nLite'd copy of RTM with smartdrv and it installed and ran noticeably faster, though it's still really slow.

Not sure why CPU-Z shows none of the 256K L2 cache though, cachechk reports everything is working.

Replaced the broken retention system on my Chaintech CT-9BJA0 (i845 SDR) and replaced the Celeron 2.2GHz Northwood with a P4 2.5GHz Northwood I had around.

All it needs now is a cooler. I would have participated in the "Aiming for the Stars" as well, but I am absolutely clueless in overclocking.

Cosmic wrote on 2023-11-20, 04:12:

I put Windows 2000 on a 486 DX2-66 for the lols this weekend. It's on a 4GB CF card accessible through XTIDE. Has 32MB of RAM and has about 4-5MB free when it's idle. 3Com 3C509 brings it online for a bit.

I initially installed SP4 without smartdrv which took ages. Then I installed a stripped down nLite'd copy of RTM with smartdrv and it installed and ran noticeably faster, though it's still really slow.

Not sure why CPU-Z shows none of the 256K L2 cache though, cachechk reports everything is working.

Nice to see the old logos in CPU-Z.
I do miss the original P4 and P4 HT logos...

ubiq wrote on 2023-11-17, 22:44:

darry wrote on 2023-11-17, 09:17:

Not sure it qualifies as purely retro, but I putzed around Windows 98 and its included DOS 7.1 in a KVM powered QEMU VM. […]
Show full quote

Not sure it qualifies as purely retro, but I putzed around Windows 98 and its included DOS 7.1 in a KVM powered QEMU VM.

I actually got PCI passthrough to work, to a point, in pure DOS on a Voodoo 3, but there are still lots of unresolved issues.

The main trick (for me at least) to actually get video output in DOS and Windows 98 GUI, is to force a VGA re-initialization right after the virtual POST . I adapted this [1] into a onliner that I added to autoexec.bat and I was greeted by working video output. Presumably Seabios does not init a passed through PCI VGA card for some reason (I am setting X-VGA to true for the Voodoo 3's hostdev and had tries fruitlessly specifying the path to a VGA ROM dump in the BAR definition). A Promise Ultra133's normall

Unfortunately it is very slow in Doom (maybe 10FPS or less) and I lose Keyboard and mouse once Doom launches .

The slowness might be due to PCIE/PCI issues.

The keyboard and mouse issues might be something
fixable through various possie means

[1]
Re: Single board computer 386-sx40 skipping VGA BIOS?

This sounds pretty neat, but I don't know enough about what you're doing to understand what you're talking about. Are you using a VM to control retro HW?

Essentially, yes.

I'll post details in a new thread soon .

Since my last post, I got 3D acceleration (Glide) in a Windows 98 VM on an an actualy physical (via PCI passthrough) Vooodoo 3 . Specifically, I tried the Need For Speed 2 SE (Glide) demo and a few old Glide demoscene demos .

The host machine has a Ryzen 5 4600G CPU and a B450 chipset .

EDIT : I was finally able to succesfully (mostly) PCI passthrough a real PCI Voodoo 3 2000 to a Windows 98 SE QEMU KVM VM

Thermalwrong wrote on 2023-11-11, 02:22:

Wah, I've been buying broken 3DFX cards for a while to see if I can fix them. Today I was looking over a Voodoo 1 that I'd bough […]
Show full quote

Wah, I've been buying broken 3DFX cards for a while to see if I can fix them. Today I was looking over a Voodoo 1 that I'd bought last year and I could never get it to do anything. It would only report 0xDEAD for the TMU status with 57005 TMUs. I'd thought it was down to a broken pin since there were a few on this card - but now I know how to actually use mojo.exe to generate the SST.LOG file which gives actual information:

This 0xDEAD result in Mojo

Info for Voodoo board # 0:
=====================================================
Virtual Base Address: 0xe6000000
Physical Base Address: 0xe6000008
PCI Device Number: 0x9
Vendor ID: 0x121a
Device ID: 0x1
FBI Revision: 2
FBI Memory: 0 MB
FBI PowerOn Sense: 0x0
TMU PowerOn Sense: 0xdead
FBI DAC Output Color Format: 24BPP
Scan-Line Interleaved? No
TMU Revision: 57005
Number TMUs: 57005
WARNING: Board 0: Bogus number of TMUs (57005)!

Was caused by

sst1Init Routines: InitCode $Revision: 8 $
sst1InitMapBoard(): BoardsInSystem = 1
sst1InitMapBoard(): vAddr:0xc8dc0000 pAddr:0xdd000008 Dev:0x9 Board:0
sst1InitRegisters(): Setting TREX-to-FBI FIFO THRESHOLD to 0x8...
sst1InitRegisters(): Setting PRELIM FT-CLK delay to 0x8...
sst1InitRegisters(): Could not detect DAC...

This reference design Voodoo 1 card did have a very bad hit to the empty PCB area and I'd already had to rebuild the VGA pass-through circuitry, but it looks like the DAC took a hit too:
3dfx voodoo1 TVP3409 DAC traces damaged.JPG
Those are all damaged traces where you see the thin metal legs, three of them were the red green and blue signals. I found that after I got the DAC to communicate with the FBI chip by resoldering legs at the upper part, which are the digital connections, the MOJO result changed into a good one:

Mojo and SST log of working Voodoo 1

SST:
sst1Init Routines: InitCode $Revision: 8 $
sst1InitMapBoard(): BoardsInSystem = 1
sst1InitMapBoard(): vAddr:0xc8dc0000 pAddr:0xdd000008 Dev:0x9 Board:0
sst1InitRegisters(): Setting TREX-to-FBI FIFO THRESHOLD to 0x8...
sst1InitRegisters(): Setting PRELIM FT-CLK delay to 0x8...
sst1InitRegisters(): Storing TREX0INIT0=0x5441
sst1InitRegisters(): Storing TREX0INIT1=0xf420
sst1InitRegisters(): Storing TREX1INIT0=0x5441
sst1InitRegisters(): Storing TREX1INIT1=0xf420
sst1InitRegisters(): Storing TREX2INIT0=0x5441
sst1InitRegisters(): Storing TREX2INIT1=0xf420
sst1InitRegisters(): Setting up FAST DRAM Configuration
sst1DeviceInfo: Board ID: Obsidian GE
sst1DeviceInfo: FbiConfig:0x2, TmuConfig:0x11
sst1DeviceInfo: FBI Revision:2, TMU Revison:1, Num TMUs:1
sst1DeviceInfo: FBI Memory:2, TMU[0] Memory:2
sst1DeviceInfo: Dac Type: TI TVP3409
sst1DeviceInfo: SliDetect:0
sst1InitRegisters(): LFB Writes go through memory FIFO...
sst1InitRegisters(): TEXTURE Writes go through memory FIFO...
sst1InitRegisters(): exiting with status 1...
sst1InitShutdown(): Shutting down SST-1 #0...
sst1InitShutdown(): Returning with status 1...
Mojo result:
Info for Voodoo board # 0:
=====================================================
Virtual Base Address: 0xe6000000
Physical Base Address: 0xe6000008
PCI Device Number: 0x9
Vendor ID: 0x121a
Device ID: 0x1
FBI Revision: 2
FBI Memory: 2 MB
FBI PowerOn Sense: 0x2
TMU PowerOn Sense: 0x11
FBI DAC Output Color Format: 24BPP
Scan-Line Interleaved? No
TMU Revision: 1
Number TMUs: 1
TMU 0 RAM: 2 MB

It even has full RAM showing for both the texture and frame buffer - so I started a game and only saw a totally black screen, but the computer was still running. I was probing at the DAC while it was running to see if the VCCA (DAC analog side power) and ground were connected when my probe on ground slipped and maybe completed the circuit since the ground trace there was completely broken away. But when my multimeter probe slipped something good happened instead of something bad - I could see the unreal flyby in blue!
Pressing the DAC area more I think I managed to get red as well and turned it off to work on the DAC. I thought the DAC might be damaged initially but if it's working well enough to generate 2 colours of 3 and make an image it should be fine. Instead of using hot air I used the low melt solder stuff and that got the DAC off pretty cleanly except as you can see in the picture a bunch of traces were broken probably before the chip was removed.

Took an hour or two to get all the traces in place and then I marked the DAC legs where there were wire legs instead of copper traces, so I could squash some of the "J" shaped legs into "L" legs - allowing the remaining good traces to still make contact with the DAC's PLCC legs, while giving clearance for the wire leg 'traces'.
That went great though, I put the card back together again and could actually test it - it went straight into tomb raider demo with the 3DFX splash screen showing all colours 😁

This card has been with me for something like 18 months and I've now and then taken another go at getting it working with no result each time because I didn't know what the diagnostics could show, didn't know that no DAC would upset the whole card. I've taken chips off, replaced RAM, broken traces in the process of replacing RAM, resoldered the QFP legs probably multiple times... It's had a rough time and I've never been able to test it.
I found that although it can now run games with proper display, it's glitching sometimes - I resoldered the RAM again and added some more capacitors to no avail, but there were some 0-ohm resistors on the back that had somehow had the markings rubbed off and replacing those seemed to stabilise the display. To test it I ran the unreal flyby for a while and it looked fine.
I was celebrating a job well done when...
IMG_2328 (Large).JPG
Party mode activated... I may still have a bit more work to do...

Huh, I should probably buy a lottery ticket today.
Just pulled out the soldering microscope to play about with an alternative screen for it and I pulled up this broken 3Dfx Voodoo 1 card. The first place I looked was at this resistor network and started measuring on the card in my previous post which was making rainbow colours:

I had somewhat given up and was about to start swapping RAM again, I had resoldered every pin on both the FBI & TMU just 'cause and to no avail. Nope, it was this resistor network that connects up the CAS lines to each memory chip - the one with a box around it was reading 400 ohms going from point to point. Resoldered it and it's down to 0 ohms again.
The one off to the side was 0.6 ohms and broken a bit so that got a little jumper wire since it's a 0 ohm resistor.

I didn't expect it to work but now the unreal flyby looks perfect and tomb raider is working properly again. Before it was giving rainbow colours in unreal but most notably black blocks on textures in the 3dfx splash logo and in tomb raider, in this case it all was because one of the texture memory chips wasn't being written to or read because of the high resistance on the CAS line.

------------------
Just look at this poor card:

• Memory swapped out all over the place - all mismatched. I think only the TMU memory is original
• Broken traces under the memory where I pulled the chips a bit too quick
• Broken legs on the FBI and TMU - one ground wire and one important wire on the FBI and one important wire on the TMU
• Display switching circuitry replaced and bodged since so many traces were damaged
• RAMDAC circuitry rebuilt
• Broken VGA connector replaced
• 3D printed bracket because its bracket went to another card while it was in the parts bin
• You can't see the back but there were several broken inductors and capacitors which have been replaced. Some of the inductors were just replaced with wire links since I ran out
• Bodged resistor network for the CAS lines

And it works 😁

Law212 wrote on 2023-11-18, 20:08:

Got my Pentium 4 Set up and running in the new place. PLayed some Quake 3 on it, ran 3D mark 2001. Good to go to play some games […]
Show full quote

Got my Pentium 4 Set up and running in the new place. PLayed some Quake 3 on it, ran 3D mark 2001. Good to go to play some games on it now.

GTA 3 wouldnt run, but I knew it ran on this computer before. I downgraded the video driver to one that came out closer to when servicepack 3 came out. So this new driver is from 2008 where the other one was 2011.

Not only did GTA 3 work, but I got a better 3D mark score as well. I need to try some other games as well. I know Turok didnt work last time I checked .

Lol now I'm getting lighting glitches and out of place polygons on this Voodoo 1, guess now the framebuffer memory is having some trouble... I did clean the flux and in doing so probably dislodged yet another chip leg.

DerBaum wrote on 2023-11-19, 13:36:

There are chinese fakes out there that use "Sony branding". https://i.ebayimg.com/00/s/MTAwMFgxMzMz/z/t4cAAOSwpaVkhvBc/$_12.JPG […]
Show full quote

PcBytes wrote on 2023-11-19, 12:01:

brostenen wrote on 2023-11-19, 10:12:

Yup.... There are Sony branded wireless Ps2 controller out there in the wild. Found a few on eBay and Amazon.

Not sure what you found but they're not genuine.

Sony never released a wireless controller until the PS3. The only licensed wireless PS2 controller is the Logitech Cordless Action Controller.

There are chinese fakes out there that use "Sony branding".

Easy to think its real.

I am really not sure how long china can produce crap like this and people are still buying this and think they get a good product.
Slowly we should realise that we just buy stuff for the dump from china.

Aye these are the two that showed up on my search. I was laughing a solid minute at how blatantly fake they are 🤣

I thought only the PS3 was plagued by fake controllers - litte did I know this BS was from back in the PS2/PS1 days 🤣

Thermalwrong wrote on 2023-11-20, 18:30:

Lol now I'm getting lighting glitches and out of place polygons on this Voodoo 1, guess now the framebuffer memory is having some trouble... I did clean the flux and in doing so probably dislodged yet another chip leg.

I wish I had the patience and skill to even attempt to do repairs like that.

Thermalwrong wrote on 2023-11-20, 18:05:

Huh, I should probably buy a lottery ticket today. Just pulled out the soldering microscope to play about with an alternative s […]
Show full quote

Thermalwrong wrote on 2023-11-11, 02:22:

Wah, I've been buying broken 3DFX cards for a while to see if I can fix them. Today I was looking over a Voodoo 1 that I'd bough […]
Show full quote

Wah, I've been buying broken 3DFX cards for a while to see if I can fix them. Today I was looking over a Voodoo 1 that I'd bought last year and I could never get it to do anything. It would only report 0xDEAD for the TMU status with 57005 TMUs. I'd thought it was down to a broken pin since there were a few on this card - but now I know how to actually use mojo.exe to generate the SST.LOG file which gives actual information:

This 0xDEAD result in Mojo

Info for Voodoo board # 0:
=====================================================
Virtual Base Address: 0xe6000000
Physical Base Address: 0xe6000008
PCI Device Number: 0x9
Vendor ID: 0x121a
Device ID: 0x1
FBI Revision: 2
FBI Memory: 0 MB
FBI PowerOn Sense: 0x0
TMU PowerOn Sense: 0xdead
FBI DAC Output Color Format: 24BPP
Scan-Line Interleaved? No
TMU Revision: 57005
Number TMUs: 57005
WARNING: Board 0: Bogus number of TMUs (57005)!

Was caused by

sst1Init Routines: InitCode $Revision: 8 $
sst1InitMapBoard(): BoardsInSystem = 1
sst1InitMapBoard(): vAddr:0xc8dc0000 pAddr:0xdd000008 Dev:0x9 Board:0
sst1InitRegisters(): Setting TREX-to-FBI FIFO THRESHOLD to 0x8...
sst1InitRegisters(): Setting PRELIM FT-CLK delay to 0x8...
sst1InitRegisters(): Could not detect DAC...

This reference design Voodoo 1 card did have a very bad hit to the empty PCB area and I'd already had to rebuild the VGA pass-through circuitry, but it looks like the DAC took a hit too:
3dfx voodoo1 TVP3409 DAC traces damaged.JPG
Those are all damaged traces where you see the thin metal legs, three of them were the red green and blue signals. I found that after I got the DAC to communicate with the FBI chip by resoldering legs at the upper part, which are the digital connections, the MOJO result changed into a good one:

Mojo and SST log of working Voodoo 1

SST:
sst1Init Routines: InitCode $Revision: 8 $
sst1InitMapBoard(): BoardsInSystem = 1
sst1InitMapBoard(): vAddr:0xc8dc0000 pAddr:0xdd000008 Dev:0x9 Board:0
sst1InitRegisters(): Setting TREX-to-FBI FIFO THRESHOLD to 0x8...
sst1InitRegisters(): Setting PRELIM FT-CLK delay to 0x8...
sst1InitRegisters(): Storing TREX0INIT0=0x5441
sst1InitRegisters(): Storing TREX0INIT1=0xf420
sst1InitRegisters(): Storing TREX1INIT0=0x5441
sst1InitRegisters(): Storing TREX1INIT1=0xf420
sst1InitRegisters(): Storing TREX2INIT0=0x5441
sst1InitRegisters(): Storing TREX2INIT1=0xf420
sst1InitRegisters(): Setting up FAST DRAM Configuration
sst1DeviceInfo: Board ID: Obsidian GE
sst1DeviceInfo: FbiConfig:0x2, TmuConfig:0x11
sst1DeviceInfo: FBI Revision:2, TMU Revison:1, Num TMUs:1
sst1DeviceInfo: FBI Memory:2, TMU[0] Memory:2
sst1DeviceInfo: Dac Type: TI TVP3409
sst1DeviceInfo: SliDetect:0
sst1InitRegisters(): LFB Writes go through memory FIFO...
sst1InitRegisters(): TEXTURE Writes go through memory FIFO...
sst1InitRegisters(): exiting with status 1...
sst1InitShutdown(): Shutting down SST-1 #0...
sst1InitShutdown(): Returning with status 1...
Mojo result:
Info for Voodoo board # 0:
=====================================================
Virtual Base Address: 0xe6000000
Physical Base Address: 0xe6000008
PCI Device Number: 0x9
Vendor ID: 0x121a
Device ID: 0x1
FBI Revision: 2
FBI Memory: 2 MB
FBI PowerOn Sense: 0x2
TMU PowerOn Sense: 0x11
FBI DAC Output Color Format: 24BPP
Scan-Line Interleaved? No
TMU Revision: 1
Number TMUs: 1
TMU 0 RAM: 2 MB

It even has full RAM showing for both the texture and frame buffer - so I started a game and only saw a totally black screen, but the computer was still running. I was probing at the DAC while it was running to see if the VCCA (DAC analog side power) and ground were connected when my probe on ground slipped and maybe completed the circuit since the ground trace there was completely broken away. But when my multimeter probe slipped something good happened instead of something bad - I could see the unreal flyby in blue!
Pressing the DAC area more I think I managed to get red as well and turned it off to work on the DAC. I thought the DAC might be damaged initially but if it's working well enough to generate 2 colours of 3 and make an image it should be fine. Instead of using hot air I used the low melt solder stuff and that got the DAC off pretty cleanly except as you can see in the picture a bunch of traces were broken probably before the chip was removed.

Took an hour or two to get all the traces in place and then I marked the DAC legs where there were wire legs instead of copper traces, so I could squash some of the "J" shaped legs into "L" legs - allowing the remaining good traces to still make contact with the DAC's PLCC legs, while giving clearance for the wire leg 'traces'.
That went great though, I put the card back together again and could actually test it - it went straight into tomb raider demo with the 3DFX splash screen showing all colours 😁

This card has been with me for something like 18 months and I've now and then taken another go at getting it working with no result each time because I didn't know what the diagnostics could show, didn't know that no DAC would upset the whole card. I've taken chips off, replaced RAM, broken traces in the process of replacing RAM, resoldered the QFP legs probably multiple times... It's had a rough time and I've never been able to test it.
I found that although it can now run games with proper display, it's glitching sometimes - I resoldered the RAM again and added some more capacitors to no avail, but there were some 0-ohm resistors on the back that had somehow had the markings rubbed off and replacing those seemed to stabilise the display. To test it I ran the unreal flyby for a while and it looked fine.
I was celebrating a job well done when...
IMG_2328 (Large).JPG
Party mode activated... I may still have a bit more work to do...

Huh, I should probably buy a lottery ticket today.
Just pulled out the soldering microscope to play about with an alternative screen for it and I pulled up this broken 3Dfx Voodoo 1 card. The first place I looked was at this resistor network and started measuring on the card in my previous post which was making rainbow colours:
2016_0211_232253_001.jpg

I had somewhat given up and was about to start swapping RAM again, I had resoldered every pin on both the FBI & TMU just 'cause and to no avail. Nope, it was this resistor network that connects up the CAS lines to each memory chip - the one with a box around it was reading 400 ohms going from point to point. Resoldered it and it's down to 0 ohms again.
The one off to the side was 0.6 ohms and broken a bit so that got a little jumper wire since it's a 0 ohm resistor.

I didn't expect it to work but now the unreal flyby looks perfect and tomb raider is working properly again. Before it was giving rainbow colours in unreal but most notably black blocks on textures in the 3dfx splash logo and in tomb raider, in this case it all was because one of the texture memory chips wasn't being written to or read because of the high resistance on the CAS line.

------------------
Just look at this poor card:
BEST-DATA-3D128.jpg

• Memory swapped out all over the place - all mismatched. I think only the TMU memory is original
• Broken traces under the memory where I pulled the chips a bit too quick
• Broken legs on the FBI and TMU - one ground wire and one important wire on the FBI and one important wire on the TMU
• Display switching circuitry replaced and bodged since so many traces were damaged
• RAMDAC circuitry rebuilt
• Broken VGA connector replaced
• 3D printed bracket because its bracket went to another card while it was in the parts bin
• You can't see the back but there were several broken inductors and capacitors which have been replaced. Some of the inductors were just replaced with wire links since I ran out
• Bodged resistor network for the CAS lines

And it works 😁

Congrats on the save!

I don't remember what it was, but I got something happen like that a decade ago, condemned piece of hardware, had not worked in years, was warming up the iron to pull a cap off it I needed elsewhere, when I noticed a bad solder joint on the board for the first time, touched it, wired it all up and tadaa, perfect function, well shit 🤣

Playing around with a Northwood 2.8C (800 MHz FSB HyperThereaded) Just trying to play a little more along here Netburst: Aiming for the Stars

Overclocked to 3.333 @ 238 FSB; 1:1 memory divider for PC4000 DDR RAM on an Asus P4P800 SE

Transplanted my 486 DX-33 build into a desktop case. It now takes the place of my previous 386 DX-40 build which was admittedly a bit redundant given the relative performance and specs of both setups.

This build even scales performance-wise better than the 386 did:

Turbo ON / L1 Cache enabled = 486 DX-33
Turbo OFF / L1 Cache enabled = 386 DX-33
Turbo ON / L1 Cache disabled = 386 SX-25
Turbo OFF / L1 Cache disabled = 286-8

I love the look of that system. It turned out very nice indeed. Now all you need to do is decide what to put on it for a case badge.