Printable View
Me too ...2 blue/green caps at the edge of the card...damn.Quote:
Originally Posted by MarioMaster
Anyway, this sucker has 3 years warrenty in only 2 1/2 months old
...so if it dies ...hopefully it will be replaced by the revised version
...or if they stop producing it ...by an 8900GT/9900GT or by X900GT:D (i.e., 10900GT)
Quote:
Originally Posted by MarioMaster
what kind of problems are you getting? 3dmark giving you artifacts/blinking screen or worst BSOD?
I might do the 1.50v mod later and i was thinking of doing it like this
http://xs203.xs.to/xs203/06295/line.jpg
the greenline shows the trace path, would that be safe/correct or its better to add another "L" trace line like the yellow line ?
My card is dead. First time thru 3D marks 01, no probs. Later when i was installning some software the hole screen was filled up with artifacts. Know i can barly boot it up until i have the screen filled with artifacts.
what we need to do is coalate all the caps values. this way we can determin if theyve uped or lowered some of the values. i got in contact with a 7900gt seller on ebay who kindly told me the values on his two bottom caps, as part of my repair problem. interestingly it seems not only do some boards have different values, his were 100uf 16v instead of 100uf 25v, they also switch between surface mount and wire leads from the bottom of the capacitors. my gut feeling on this problem is that someone at nvidia underated the voltage regulators. i mean we shouldnt have to go about fitting heat sinks just to get these cards stable at default clock speeds. when is nvidia going to step forward and anounce they made a big mistake in the design?
MarioMaster
My old evga card looked exactly like yours cept mine had the SS heat sink which covers the ram too instead of the KO heat sink. My card would artifact during 3dmark2001 point sprite test. Even with a heatsink over the IC's i could only run my card at 450mhz core speed.
Don't be TOO worried though because although many of the cards have this problem there are still many that also don't have this problem. Also evga is trying really hard to RMA all the affected cards.
those flashing dots on the horse? thats normal on all videocards ive seenQuote:
Originally Posted by ineedaname
Edit:
is 66~68c on full load at 1.50v ( 650Mhz core ) safe for a 7900GT?
yes, defintely.Quote:
Originally Posted by Demo
I did 3 runs of 99 loops at using deep freeze test.. i checked the screen once in a while for artifacts.. i also notced that these IC ( on the pic ) are extremely hot even at stock volts. added a heatsink on them
http://xs304.xs.to/xs304/06301/hotic.jpg
Demo
No not the flashing dots on the horse. My vid card would just totally crap out after about 1 spin of the horse with dots all over the screen. Then the screen will flash every 2 seconds and go extremely slow.
Also can you get a bigger pic of the IC's ur circling.
It's right behind the PCI-E power connector (see the 3 half-circles on the left edge of the picture).
Quote:
Originally Posted by Bladesinger7x
The edge circle is part of the Vmod the only 2 ones highlighted are the hot spots
problem is i can't test it to see if it artifacts or not, because my am2 rig is the only pci-e computer i have and the cpu is on it's way back to AMDQuote:
Originally Posted by likewhoa
Don't u know any friends with a PCI-E slot to test it on?
Leadtek 7900GT 450/660 - it will be a new revize, was say me, all purple :up:
so I hope that will not be problem
waiting for my pen
first OC with this, through nv drivers optimum 554/784 but set 570/870 def 1,2
atitool no artefact, 3D 2005 runs 2x -- ok
Do you think put on back side card some passive on regulator?
http://img195.imageshack.us/img195/6...1779ez1.th.jpg
http://img134.imageshack.us/img134/1...1782jj4.th.jpg
http://img195.imageshack.us/img195/1...1793ws4.th.jpg
http://img195.imageshack.us/img195/9...1807qv1.th.jpg
well, may as well join the shenanigans.
i have a Gainward 7900GT, 520/660
it runs Deep Freeze and Point Sprites fine, no problems at all.
the concerned part of the card doesnt feel TOO hot to worry about i dont think. however, i have the mint caps...
also, i have OCed the core up to 520 from 450, but have not yet touched the mem, as it is sink-less. i have mailed Gainward to see if they will let me sink the mem without voiding the warranty.
what are people generally managing to get out of the memory with stock volts, and no sinks? just be interested to know how hot people can get the mem on average before problems appear.
thanks.
EDIT: i dont THINK i get the noise, i cant really hear because of the fans in my case. i am fairly sure i dont though...
My mem maxes out at 810 im running them at 800 no vmods or sinks in the mem.
I just noticed during 3d my card makes a fan like noise.. is that normal ?
i have a CF900Cu always running at 100% fan speed
it could be that noise that people were talkin about.
Ugh.
My XFX 7900GT just started to make this coil whine. Does the glue trick really works?
yea, but i seem to remember that it is only a temporary solution to the noise before the card dies on you?Quote:
Originally Posted by ROcHE
Did they ever get to fix this sudden death problem of the card?
my 7900gt died on my 2 weeks ago.. actually massive artifacts.. its now under RMA and i should get my new replacement in 2-3 weeks