Stripping a 260 gtx

Intexity

Senior member
Jan 10, 2009
299
0
0
Just curious. I have a EVGA 260 gtx and it peaks at 55 c under load and was wondering if I took the plastic casing off the card would it help or hinder. It looks to me like the fan is mounted to the board and figured with two fans on the side panel one positioned for the gpu if it would help to cool it. Unfortunately before I go taking it apart (the cover decal is conveniently placed just off center of the screw holes to show a voided warranty) was curious if anyone had tried this.
 

alkalinetaupehat

Senior member
Mar 3, 2008
839
0
0
What Acolyte said + I took mine apart and the plastic shrouding keeps the airflow pretty smooth. You *might* see some benefits from swapping the thermal paste (I saw ~5C), but if you're getting 55C LOAD, you're about 15C below me, and ~45C below the thermal threshold. You're golden dude.

I can upload my pics of when I stripped the card, the sticker went back on so well you can't tell I swapped the thermal paste...;)
 

PCTC2

Diamond Member
Feb 18, 2007
3,892
33
91
The plastic casing directs the airflow from the blower at parts, so it would be wise to keep it on. However, the GTX 260 is a HOT card and 55C is cooler than my 8800GTS. I say those are great temps.
 

Intexity

Senior member
Jan 10, 2009
299
0
0
ok ok so im bored. I can't finish my case mod (cosmos rc1000) and i hate this computer case. Too loud!! What??? anyways.... was trying to find a way to bring it down i was seeing some issues when i started playing stranglehold with a temp at 59+ load temp but after cleaning the msconfig and taking off the side panel managed to wrangle it back under control. I am still seeing issues SOMETIMES very rarely it will freeze but im not so sure if its not the game itself. yeah i have a dead 8800gts im still waiting for an rma on. damn thing was getting way too hot and ended up frying itself or so i think. hence the 260 gtx. anyways thanks for the informative responses. don't think i will mess with it cause i payed way too much money on the card to void the warranty in any way. video cards seem to fry at the most inoppurtune times.
 

alkalinetaupehat

Senior member
Mar 3, 2008
839
0
0
Originally posted by: Intexity
ok ok so im bored. I can't finish my case mod (cosmos rc1000) and i hate this computer case. Too loud!! What??? anyways.... was trying to find a way to bring it down i was seeing some issues when i started playing stranglehold with a temp at 59+ load temp but after cleaning the msconfig and taking off the side panel managed to wrangle it back under control. I am still seeing issues SOMETIMES very rarely it will freeze but im not so sure if its not the game itself. yeah i have a dead 8800gts im still waiting for an rma on. damn thing was getting way too hot and ended up frying itself or so i think. hence the 260 gtx. anyways thanks for the informative responses. don't think i will mess with it cause i payed way too much money on the card to void the warranty in any way. video cards seem to fry at the most inoppurtune times.

Ya know there are times like every two months for about a one week period where my video card will lock up the system and require a reboot to fix. Weird you mention your card is doing the same thing.

If you want to quiet down your system you could go for Scythe S-Flex's or Slipstreams, or some more sound deadening. You could see what 1/2 speed fans do for your temps/noise, general temps in my case go up about 5C when all the fans are turned down. The size room you're in makes a big difference too; my PC is much more noticeable in my 10x12 bedroom than in the 15x15 living room (which opens into the dining room and kitchen, so yeah). At an old house I was in an 8x10 room with my old PC, I could tell you where every little noise it made came from after a month.

Different card manufacturers have different rules on what you can do with your card; EVGA allows overclocking on their cards, BFG does too, and XFX allows both modding and OC'ing. The trick with XFX though is if you RMA your card, you have to send it back with the stock cooler on (Isn't hard to do). I ripped my card apart ~2 weeks after I got it, lol. Same one as is in the sig.
 

OCNewbie

Diamond Member
Jul 18, 2000
7,596
24
81
I'm on my 2nd EVGA GTX 260 now and both hit 90+C after 10-15 minutes or so of Furmark. I guess I should RMA this 2nd card too... argh. They both idle around 45C but absolutely cook when fully loaded. My CPU (C2Q Q9550) stays around 28C at idle and low 50's full load with a Thermaltake Big Typhoon, so I don't think it's my case temps that's causing this.

Edit: So 55C load (not sure what you're considering load) is super low for that chip.
 

OCGuy

Lifer
Jul 12, 2000
27,224
37
91
Originally posted by: OCNewbie
I'm on my 2nd EVGA GTX 260 now and both hit 90+C after 10-15 minutes or so of Furmark. I guess I should RMA this 2nd card too... argh. They both idle around 45C but absolutely cook when fully loaded. My CPU (C2Q Q9550) stays around 28C at idle and low 50's full load with a Thermaltake Big Typhoon, so I don't think it's my case temps that's causing this.

Edit: So 55C load (not sure what you're considering load) is super low for that chip.

If it hits 90C under heavy load, but it is stable, what does it matter? Those chips have auto-throttling at 100-105C, so they can handle 90.

What fan speed percentage is that? Sometimes the "auto" feature doesnt set it as high as it should.
 

chizow

Diamond Member
Jun 26, 2001
9,537
2
0
Originally posted by: OCNewbie
I'm on my 2nd EVGA GTX 260 now and both hit 90+C after 10-15 minutes or so of Furmark. I guess I should RMA this 2nd card too... argh. They both idle around 45C but absolutely cook when fully loaded. My CPU (C2Q Q9550) stays around 28C at idle and low 50's full load with a Thermaltake Big Typhoon, so I don't think it's my case temps that's causing this.

Edit: So 55C load (not sure what you're considering load) is super low for that chip.
Are you experiencing problems in actual games? Those temps are quite normal for Fur Mark, which is certainly not a good indication of full load temps as no game will stress the GPU as heavily as Fur Mark. Fur Mark is good for testing stability, but even then, I'd set manually set the fan speed to 80% or higher while testing.
 

mindless1

Diamond Member
Aug 11, 2001
8,722
1,735
126
Originally posted by: Ocguy31
Originally posted by: OCNewbie
I'm on my 2nd EVGA GTX 260 now and both hit 90+C after 10-15 minutes or so of Furmark. I guess I should RMA this 2nd card too... argh. They both idle around 45C but absolutely cook when fully loaded. My CPU (C2Q Q9550) stays around 28C at idle and low 50's full load with a Thermaltake Big Typhoon, so I don't think it's my case temps that's causing this.

Edit: So 55C load (not sure what you're considering load) is super low for that chip.

If it hits 90C under heavy load, but it is stable, what does it matter? Those chips have auto-throttling at 100-105C, so they can handle 90.

What fan speed percentage is that? Sometimes the "auto" feature doesnt set it as high as it should.

It matters because there's momentary stability then there's llifespan. We often say chips will last longer than we'd want to use them, but that's if they aren't running so hot. My crystal ball is broken so I can't predict the day it will die from running at 90C, but I wouldn't expect 2 years.
 

OCGuy

Lifer
Jul 12, 2000
27,224
37
91
Originally posted by: mindless1
Originally posted by: Ocguy31
Originally posted by: OCNewbie
I'm on my 2nd EVGA GTX 260 now and both hit 90+C after 10-15 minutes or so of Furmark. I guess I should RMA this 2nd card too... argh. They both idle around 45C but absolutely cook when fully loaded. My CPU (C2Q Q9550) stays around 28C at idle and low 50's full load with a Thermaltake Big Typhoon, so I don't think it's my case temps that's causing this.

Edit: So 55C load (not sure what you're considering load) is super low for that chip.

If it hits 90C under heavy load, but it is stable, what does it matter? Those chips have auto-throttling at 100-105C, so they can handle 90.

What fan speed percentage is that? Sometimes the "auto" feature doesnt set it as high as it should.

It matters because there's momentary stability then there's llifespan. We often say chips will last longer than we'd want to use them, but that's if they aren't running so hot. My crystal ball is broken so I can't predict the day it will die from running at 90C, but I wouldn't expect 2 years.

"Running hotter than you think it should" is not a valid RMA reason.

Again, is that with 90-100% fan speed?

EVGA will replace the card anyway if it randomly dies.

 

mindless1

Diamond Member
Aug 11, 2001
8,722
1,735
126
I never suggested running hotter than you think it should is a valid RMA reason, but it is a reason why it's undesirable in general and should be avoided if possible. Additionally, what if the card dies after the warranty is up? What is EVGA's warranty these days, 1 year plus 1 addt'l if you register the card?
 

OCGuy

Lifer
Jul 12, 2000
27,224
37
91
Originally posted by: mindless1
I never suggested running hotter than you think it should is a valid RMA reason, but it is a reason why it's undesirable in general and should be avoided if possible. Additionally, what if the card dies after the warranty is up? What is EVGA's warranty these days, 1 year plus 1 addt'l if you register the card?

Lifetime on the higher end.
 

OCNewbie

Diamond Member
Jul 18, 2000
7,596
24
81
Oops, didn't realize I had had responses.

This is getting to 90C within 10 minutes. This is manually setting the fan to 100% via EVGA's Precision utility, then letting the card idle/settle for a few minutes down to around 44C, then while the fan is still running at 100% running Furmark. Like I said, within 10 minutes it's at 90C (up from 44C before starting Furmark) and then is very slowly continually rising (about 93C after 20 mins). I believe I've read others report max temps in the 70's or low 80's on Furmark with their fans on Auto and saying their fans only maxed around ~70% anyway.

Just seems like with the fan at 100% I would expect lower than 90+C. The Fan at 100% is way too loud.

Oh, and isn't TJMax for Intel CPU's like 95 or 100C? But nobody seems to think it's fine to let their CPU's get anywhere close to maximum safe temps, so why is it ok to do so with a GPU?
 

OCNewbie

Diamond Member
Jul 18, 2000
7,596
24
81
Oh, just to add... I believe I was getting just shy of 90C in Oblivion the other day. This is at 1024x768, 4xAA 16AF, I believe the fan was either at manual 55% or auto. Just seems too hot to me. This was on my replacement card, not the first one they sent me.