V5-6000 & RAMPAGE Team

[osckhar]

Hello,

Yesterday I received the 6000s from feinripp. It is a very nice 3400 A3 with 5'4ns. Card looks very nice sadly it has CR3/4 diodes blown and damaged the PCB. I always recommend to take out at least 1 diode and if there is a wrong voltage PCB will be not damaged.

V5-6000 3400 A3



LINE ERROR FAST MODE OPENGL


LINE ERROR 2xFSAA OPENGL


STATUS:
1.- Using Win98SE and original driver card only shows lines errors in FAST MODE or 2xFSAA but only in OpenGL mode. In D3D card works very well.

2.- This A3 comes with different reworks I never saw before in others A3.

3.- Later Isolated different VSA100 error comes from first and master VSA100 but in single mode card does not shows errors.

4.- VSA100 works 3V instead 2.8V and card has 'boost @3V' turned on in R737 after booting up.

28/02/2017
At the moment not luck. I checked components where error could come without luck. IC & SMD are working in perfect condition. Seem to find the error will be more difficult than I thought.

I still need to check others components where line problem can be.

23/03/2017
Later a lot of hours testing component by component seems I found the problem.

As you can see in the pictures there are 2 pads in short on VSA100. GND+MA_A6 and it can cause the lines on memory. Such pad make reference to Memory Address Bus:

"This multiplexed bus supplies the address to the SGRAM/
SDRAMs providing MD[63:0].
The exact connections of the MA, MBA, and MCS lines to the SGRAM/SDRAM inputs
depends on the memory configuration. See Chapter 5 for details."



Is it time for BGA rework???

24/03/2017
I started again to work on 3400 A3 and reemplaced MASTER VSA100 CHIP.

TAKING OUT MASTER VSA100 CHIP


CLEANING BGA PADS



STARTING PROCESS BGA REWORK


PERFECT ENDED BGA REWORK


BOOTING UP 3400 A3


QUAKE 3 FAST MODE WITH NOT LINES ERRORS.


Very nice feeling when later a BGA REWORK you see again the card is booting up without problems - each BGA rework is a risk - since I dont own any kind of stuff for checking if process is correct or not and if there are balls in short or balls are soldered correctly - I would need a X-Ray machine - and when I run a Quake 3 in fast mode or 2xFSAA and lines vanish- not words here. Amazing feeling of 'peace'.

This time this card was driving me crazy since it has been very difficult to fix it since I was looking for the problem in others components.

Anway, Not bad, another 3Dfx V5-6000 fixed.

Regards,
Oscar.
_________________
¿Leyenda? ¿Mito? ¿Ficción? No... ¡Realidad!
3Dfx Collection

[feinripp]

Pretty interesting observations so far! Oscar knows this article of course, but for the sake of completeness here is what I found out: https://robertkrau.se/blog/3dfx-voodoo-5-6000-stripe-artifacts/

Note that the error occurs with all APIs for me, not only OpenGL.

[osckhar]

UPDATED FIRST POST.

Regards,
Oscar.
_________________
¿Leyenda? ¿Mito? ¿Ficción? No... ¡Realidad!
3Dfx Collection

[Obi-Wan Kenobi]

Hey good to see the card is doing something
But ow a cure if there is one

Also I PMe'd Oscar about it after I read your thread at VA Feinripp
Heck I even tried to make contact with Hank Semenec as well via Facebook, since this was to me a very critical thing, but no reply back from him, I will keep in touch if I do get any response back though.

Anyways good to see that this problem is getting looked in to, if I recall correctly the Rev.A3 3400's were all quite different from each other when it came to reworks. Ram wise 5.4ns was the most common SDRAM used on 3400's 5.5ns was rarer, as for Rev.A2 2600's these mostly had 5.5ns Hynix or 5.4ns Toshiba, similar to what Rev.A1 1500's has as the 5.4ns Toshiba went.

It seems that Feinripps card does have a RAMDAC issue, as I had thoughts of this issue earlier, but then again it was just a hypothesis from my end, since I could not scientifically prove it for it to be a theory.

If the problem can be determined might be an idea to explain what is causing this and what would fix this in the following, just for good resource archiving of such situations.

Although 3dfx used anything they had at the time, as ram types & ram spd's go, hence those minor differences.

Anyhow I hope you can heal this card osckhar, I have a good feeling you will succeed though, May The Force Be With You old friend
_________________
Blue-Leader: 2CPU Opteron 2435, 32GB DDR2-800 Kingston, Supermicro H8DAE-2
Firefox: 1CPU Opteron 6140, 32GB DDR3-1600 @ 1333 HP, Supermicro H8SGL-B/F
Valkyrie: 2CPU Opteron 6180 SE, 48GB DDR3-1600 @ 1333 HP, Supermicro H8DG6-F

[osckhar]

23/03/2017

FIRST POST UPDATED.
_________________
¿Leyenda? ¿Mito? ¿Ficción? No... ¡Realidad!
3Dfx Collection

[osckhar]

24/03/2017

Post updated and card fixed.

Another 6000s fixed. Yeahhhhhhhhhhhhh!!!!

Regards,
Oscar.[/b]
_________________
¿Leyenda? ¿Mito? ¿Ficción? No... ¡Realidad!
3Dfx Collection

[Obi-Wan Kenobi]

Yaaaaaaaaaaaaaaaaaaas Grats dooh <3 Huuuuuuuuuuugz <3 thanks for helping our Feinripp with this one Osckhar xD

It may have been complicated, but see it as very challenging, educate and worth doing

Glad to see you found your way through this nasty issue
And now you know what caused don't you get that feeling of oh was this really it xD
_________________
Blue-Leader: 2CPU Opteron 2435, 32GB DDR2-800 Kingston, Supermicro H8DAE-2
Firefox: 1CPU Opteron 6140, 32GB DDR3-1600 @ 1333 HP, Supermicro H8SGL-B/F
Valkyrie: 2CPU Opteron 6180 SE, 48GB DDR3-1600 @ 1333 HP, Supermicro H8DG6-F