|03-24-2009 07:54 PM|
I know what you mean. And,if it passes that, it is 110% stable so you can usually rule it out when you have an issue.
|03-24-2009 01:28 PM|
ikjadoon: It's all good.
The reason Intel and AMD doesn't release those kinds of stress-testers is because there's no point. They are so over-the-top, that in absolutely no possible situation could you ever achieve the same. IntelBurnTest and anything else that uses Linpack is pretty much the same. I've never seen any real process burn up the CPU so bad, so in a way, it's probably overkill. But, if it deems the CPU unstable, I still consider it unstable. I like to reside in the fact of knowing that even when pushed to the brink, my CPU won't fail.
Unregistered: Thanks for the info! I never heard of that before, but I just downloaded it and plan to test later.
|03-23-2009 11:54 PM|
|Unregistered||There is a linpack for i7, its called linx. Burntest doesn't properly support i7 ht yet, but linx stresses all 8 threads.|
|03-22-2009 11:21 PM|
Really? Huh, well, I guess that makes sense, different architecture and all, but you'd think a CPU was a CPU. Ah....we get the striped down version. Makes sense. I heard Thermalright issuing a statement that people shouldn't use FurMark to test their GPU stability because it makes the cards too hot. Maybe Intel is worried of the same!
|03-22-2009 10:33 PM|
Linpack was and still is an Intel creation, although the free version they allow people to use in their own applications (IntelBurnTest, OCCT) isn't as robust. It doesn't currently work properly on i7, for example. I am really, really hoping that an updated version comes out sooner than later. I'm forced to use the old-school Prime95 test (inside of OCCT) for i7 in the interim.
As for that quote... wrong website ;-)
Nate's a good friend though, so kudos regardless!
|03-22-2009 10:23 PM|
Hehe, good to be around.
Really?! That sounds ridiculous. Now that would be a torture test, lol....Wasn't LINPACK originally an Intel creation or something?
Off-topic, but kudos, you guys got a little snippet in that recent Anandtech article about SSDs:
LegitReviews was one of the first to spot the SSD slowdown phenomenon, good work Nate.
|03-22-2009 08:49 PM|
Ibrahim, good to see you around!
|03-21-2009 06:05 PM|
I agree there. In fact, there is very little that will push the same load as Prime95.
To me, when I want to declare my overclock stable, I want to stress it beyond any level it will ever see in real use. If it isn't Linpack stable then in my book it just isn't stable. I always use Prime95 to start my testing and once I have a good solid 12hours there I move on the Linpack testing to finalize my OC. There have been times that I never made any changes and it went on to pass, but more often than not I need to change just a few small details.
My whole point is exactly what you brought up. With all these coolers within a few degrees C, I like the extra load Linpack offers to separate them. I have not come across a single heatsink, including the almighty TRUE 120, that could tame my QX9650 @ 4GHz. Only water was able to do that.
|03-21-2009 05:28 PM|
At the same time, there are no programs in existence (correct me if I'm wrong) that will ever stress the CPU to levels of LINPACK and, in the same vein, Prime95. I doubt I'll ever use a program that will bring a quad-core @ 4Ghz to 67 degrees on load with this cooler. It is nice to know, however, that it does it with ease. Just my two cents.
|03-21-2009 04:31 PM|
|b1lk1||Agreed on the Linpack testing. Considering the majority of reviewers are still running Core 2 Duo processors they have no reason to stick to Prime95 which is just not going to give us the full picture of how hot the CPU can run.|
|03-21-2009 04:14 PM|
Sweet review. I only wish it wasn't so expensive, but you get what you pay for, definitely.
EDIT: Never mind, missed that paragraph after the picture.
P.S. Glad to see LINPACK used. That test is a mofo monster, for sure.
|03-20-2009 03:45 PM|
Also that's Bill for the review, i've been waiting to see that monster reviewed since I first saw it.
|03-20-2009 01:49 PM|
|b1lk1||I can state with certainty that the V10 has very high build quality and I don't see any reason for any early failure. The TEC is not a high powered unit and it does have it's own 4 heatpipes to keep it cool. Also, the V10 does have the built in variable controller to only push the TEC when it needs to otherwise it cuts back on the power delivered to it.|
|03-20-2009 01:44 AM|
Good GAWD... this thing is a beast!
I'd be very interested to know the reliability of the TEC. A friend of mine had a similar unit from another company and it died a horrible death.... twice I believe (his original and his replacement).
I trust CM's products more than I trust the unit he was using, so I am hopeful that it has a more quality build feel to it, and the proper gauge wiring was used.
|03-19-2009 04:23 PM|
We have a single i7 platform right now, and it's here with me, not Bill, so sadly, this testing won't happen, at least soon. I'd still expect that the cooler would perform well though, since this did prove to be the best one to hit our QX9650. I don't think we could guarantee a 4GHz clock out of the 920, but who knows. The problem with the i7 and overclocking is heat, and since the V10 is just as efficient as a $200 water-cooling setup, I'd expect it to do well.
Sorry we can't get you the results you are looking for. We'll have to work on getting an i7 into Bill sooner than later.
Welcome to the forums :-)
|This thread has more than 15 replies. Click here to review the whole thread.|