Ratdude's Main rig V3.5

[ratdude747]

Re: Ratdude's Main rig V3.5

Those novabench results were not on the opteron system, they are the results of my current office machine, which is a dual quadcore xeon. I never did novabench the opti box before I dismantled it....I guess I should have.

Umm...

Topcat did a Novabench run on his current setup awhile back:



Too bad it wasn't the previous system, as it used the same board and all.

I know dude, I know.

[Topcat]

Re: Ratdude's Main rig V3.5

^
kk, lotsa luck, wiseass.

[ratdude747]

Re: Ratdude's Main rig V3.5

^
kk, lotsa luck, wiseass.

I was just using it as a comparison point. My whole "wish it was the previous system" comment was about having a better comparison point...

I hate miscommunictions

[shovenose]

Re: Ratdude's Main rig V3.5

Granted I'm watching Netflix in HD at the same time and also have about a billion programs open but...

[Topcat]

Re: Ratdude's Main rig V3.5

I was just using it as a comparison point. My whole "wish it was the previous system" comment was about having a better comparison point...

I hate miscommunictions

I didn't catch the last line actually.....its all good man.

[ratdude747]

Re: Ratdude's Main rig V3.5

decided to furmark it... and when the GPU's got to 100C, I killed furmark.

Uh oh...

When I redid the heatsinks, I used this for the GPU die:

http://www.radioshack.com/product/in...oductId=154389

I take it that's not good enough... What is suggested to fix this?

[370forlife]

Re: Ratdude's Main rig V3.5

Furmark pushes a GPU to it's absolute limits. What cooler is on the card? I use to have an 8600GT with the dinky stock cooler. Heated up like crazy. Put on a big ol' cooler and it never got over 40C after that.

[Pentium4]

Re: Ratdude's Main rig V3.5

Yeah Furmark pushes them pretty hard. Were the fans actually at 100% or were they lower? I've seen some stupid video cards that even when getting up to 90C+ only went to 80-90% fan speed

[Sparkey55]

Re: Ratdude's Main rig V3.5

decided to furmark it... and when the GPU's got to 100C, I killed furmark.

Uh oh...

When I redid the heatsinks, I used this for the GPU die:

http://www.radioshack.com/product/in...oductId=154389

I take it that's not good enough... What is suggested to fix this?

If you are talking about TIM, I use ArcticSilver Ceramique. Never had a problem.

[ratdude747]

Re: Ratdude's Main rig V3.5

Furmark pushes a GPU to it's absolute limits. What cooler is on the card? I use to have an 8600GT with the dinky stock cooler. Heated up like crazy. Put on a big ol' cooler and it never got over 40C after that.

Stock cooler for an 8600GTS (it's a bit bigger than a GT IIRC). My only worry about swapping is that if you look, there is only one slot of space between GPUs. Kinda tight.

Yeah Furmark pushes them pretty hard. Were the fans actually at 100% or were they lower? I've seen some stupid video cards that even when getting up to 90C+ only went to 80-90% fan speed

100% @ 90C or so.

If you are talking about TIM, I use ArcticSilver Ceramique. Never had a problem.

I heard ceramique was junk and AS5 was a better choice... but I'm open to ideas.

[shovenose]

Re: Ratdude's Main rig V3.5

AS5 is a bit old at this point but still excellent and it seems to have come down in price a tiny bit.

The differences between all of the really good pastes including AS5 are like 1-3C so just find one you like an use it AS5 would be my recommendation that said.

[ratdude747]

Re: Ratdude's Main rig V3.5

Done some looking, not as sold on AS5... there's some 200 hour burn-in period on that stuff, and that's almost a deal breaker, as the tests I've read show it's not bad (but far from the best) when fresh.

This stuff any good?

http://www.newegg.com/Product/Produc...82E16835426020

[mariushm]

Re: Ratdude's Main rig V3.5

I don't know, looks like generic paste to me. If it was any good, it would list the specs on it.
The gelid site lists it as 8.5W/mK which isn't anything special.

Look for pastes that have better thermal conductivity, for example the Prolimatech pastes are advertised at 10.2w/mK
http://www.newegg.com/Product/Produc...82E16835242024

But then again, Tuniq TX-4 is recommended by a lot of sites as a good general paste and it's only rated as 6.3w/mK.

It makes a lot of difference how thin the paste is applied and so on...

[Sparkey55]

Re: Ratdude's Main rig V3.5

Done some looking, not as sold on AS5... there's some 200 hour burn-in period on that stuff, and that's almost a deal breaker, as the tests I've read show it's not bad (but far from the best) when fresh.

This stuff any good?

http://www.newegg.com/Product/Produc...82E16835426020

All TIMs have some burn-in period, whether from 5 minutes to 5 days. The microscopic gaps need time to get filled in at normal conditions. You should read some NASA or Mil-Standard Qualification data sheets for further understanding.

[c_hegge]

Re: Ratdude's Main rig V3.5

I always use AS5, and it seems to do great.

As far as the furmark temps, I once did it to a passive 8500GT with a somewhat small heatsink (Thanks ASUS!). It maxed out at 105*C. You could just about cook eggs on it at that temp.

[ratdude747]

Re: Ratdude's Main rig V3.5

All TIMs have some burn-in period, whether from 5 minutes to 5 days. The microscopic gaps need time to get filled in at normal conditions. You should read some NASA or Mil-Standard Qualification data sheets for further understanding.

I see.

[lti]

Re: Ratdude's Main rig V3.5

I remember one company that had a warning on every model of paste they made saying that it must be reapplied every 12 months. I can't remember what brand that was now, and I don't see any that has that warning now.

I never had good luck with Ceramique. It worked fine for a year, and then the temperatures started increasing. I'd rather not take the heatsink off that often.

[shovenose]

Re: Ratdude's Main rig V3.5

Use whatever the OEMs (HP, Dell, eMachines/Acer/Gateway, Lenovo, Toshiba, etc) use because most people run their ten year old PCs and have never reapplied the TIM nor even know what the heck it is LOL...

Alright, maybe I'm kidding, a little bit, maybe not?

[Sparkey55]

Re: Ratdude's Main rig V3.5

I remember one company that had a warning on every model of paste they made saying that it must be reapplied every 12 months. I can't remember what brand that was now, and I don't see any that has that warning now.

I never had good luck with Ceramique. It worked fine for a year, and then the temperatures started increasing. I'd rather not take the heatsink off that often.

I am using a computer right now that has Ceramique in use with a Athlon XP 2600+ used every day almost 24/7 for over a year now and no temp problems at all in that time. Do you clean out the dust bunnies often?

[RJARRRPCGP]

Re: Ratdude's Main rig V3.5

I remember one company that had a warning on every model of paste they made saying that it must be reapplied every 12 months. I can't remember what brand that was now, and I don't see any that has that warning now.

I'm not surprised if it's the run-of-the-mill white stuff!
Unfortunately, the RadioShack stuff likes to fade away after some months of use.

Took socket 462 heatsinks off more than once, IIRC, to see it disappeared from the Athlon!

No such bull with Arctic Silver 5.