To those who lost a GF4 on 8RDA+

[SteveI]

To RacerX1,
Something you said... made me recall something...
I vaguely recall reading a review of 4600's when they first came out. Quite a few did not follow the reference design, and the reviewer was criticizing the mfgrs for making the cards too large and clumsy, and had difficulties installing on some mobos...

Not sure if I'm adding anything useful, but if this rings any bells for anyone, I'll try to find a link.

If some 4600's are not following spec or reference design....

[rocky500]

The 2 Leadtek Geforce 4600's that we had go on the 1 board were brought in July 2002.
We have already RA'd them so can't check exactly the No.s on them.
They did not have a revision number as we looked for that before sending them back.

[RacerX1]

Originally posted by SteveI
To RacerX1,
Something you said... made me recall something...
I vaguely recall reading a review of 4600's when they first came out. Quite a few did not follow the reference design, and the reviewer was criticizing the mfgrs for making the cards too large and clumsy, and had difficulties installing on some mobos...

Not sure if I'm adding anything useful, but if this rings any bells for anyone, I'll try to find a link.

If some 4600's are not following spec or reference design....

Yup that very well could be but I just bought my card less than a month ago.

I really feel my card shorted out. Right below the memory controller or regulator there are about 5 or 6 tiny resistors. The white tab presses firmly against those resistors with card installed. You wouldn't think the plastic tab could short out anything, right?

Maybe instead of getting another MSI card in return I will change to a different brand. I will try for Abit since it cost alittle less.

[Corn]

My GF4 was a gainward 4400 Ultra/700XP.
I got it 12 of june.
I died as I fired up the MB for the 1:st time.
And if anything it was properly seated in the AGP, I triple checked that
I had two 256M memory modules running in 128bit mode.
2400XP

[EluSiOn]

corn... have you email that information to your epox tech support nearest you?

[Corn]

nope...
Id rather email it to someone at epox who knows of this problem.

[EluSiOn]

Originally posted by Caper
I don't know if this post will be fine here, might be in AMD Motherboards?

Anyway, I spoked to EPoX support in Netherlands.
They want all that lost their GF4 on 8RDA+ to mail them.

Subject: TI4600 on 8RDA+

support@epox.nl

Describe what happened and what hardware you was using.
It would be nice is you say bump when you done it.

do that..

[Corn]

bump

[MiGGinZ]

I have an idea on what could kill some gf4's.

When you press it in there are a couple of small components on the cards back which almost catch on the ram clips ( I noticed this when installing mine today, Now im not saying anyones stupid or anything but maybe when you have to give it that firm push you could easily clip one of them off..

[Stanley Jobson]

Originally posted by MiGGinZ
I have an idea on what could kill some gf4's.

When you press it in there are a couple of small components on the cards back which almost catch on the ram clips ( I noticed this when installing mine today, Now im not saying anyones stupid or anything but maybe when you have to give it that firm push you could easily clip one of them off..

That can't be the cause!

I press the cards always completely carefully into the AGP slot ... and it's smoked nevertheless ...


@ET:

My smoked Leadtek was a Rev. A (can't find the mentioned code on the card pics - the card itself was replaced by Leadtek) the only code I found was this - ... CR2890 PCB REV. A


@caper: BUMP

[MiGGinZ]

I said SOME, not all.

[drow_elf]

okay so my understanding of the three theories are thus:

1. somehow while installing ram the vid card was slightly unseated/misaligned in the slot and this caused death (i find it unlikely because the only way i see this killing a card is if a high voltage pin touched a low voltage contact, but the "key" in the slot should prevent that)

2. because of the largness of the vid card some componets short, on plasitc ( still unlikely unless plastic not conducts electricity)

3. again because of uncommon size several componets where scraped off (once again unlikely since most surface mount componets are pretty tough)

[rocky500]

Here is another:
Mabey on intial Power up there is a intial Voltage spike to the AGP slot in which certain cards cannot take.
There is only few cards failing & majority have no problems.

I spoke to my friend and his is 99% sure he flashed the bios before putting in the 2nd card.

Both failed on intial power up without anything comming up on the monitor.
He remembers on first boot with each card he got a "7F".
Each try after that to boot it would come up "FF".
2 Kingmax DDR memory sticks were used each time.

[LL_NEO]

Yeah but the "initial boot spike" scenario doesn't answer to the question why we have reports of "smoked gf4" after the change of ram. I think that Caper has reported about 2 users having that sort of a problem. And since you mention it, my 8RDA+ always shows the FF sign, I take that is the "all systems go" sign?

[EluSiOn]

it is very likely to bend the GF4 card when you swap RAM... because of the 6 pci slot design... epox mobo's agp slot is too close to ram slot...

[xgman]

Does rev. 1.1 of the board have the same spacing of the clips to the vid card as rev. 1.0 of the 8rda+?

[kuzma]

Visiontek gf4 4600

180-10083-0000-A02 REV B

[EluSiOn]

****.... It is a rev B card....