Replaced Xigmatek Dark Knight with TRUE Cu...20C drop in temps

[LennyRhys]

Bask in the greatness of...
The TRUE Cu

(image below can be downloaded for use as a wallpaper, or just a cool pic of a sweeeet cooler - click for 1920x1200 size)



As the thread title says, I just swapped out my Dark Knight (S1283) with a TRUE Copper. Here are some pics of the coolers together:



Gave both heatsinks a good rub with Brasso - SHINY!!



Side-by-side the TRUE Cu is fractionally smaller, but 3x the weight





Was going to lap the base of the TRUE, but really, with the performance it delivers, why bother...LOL



Just for kicks (and because this is Xtreme Systems) I thought I'd take a stab at some priming at 4GHz, 1.33v, HT on, with passive air cooling

Temps stayed at about 65C for about five minutes, and then they went to 85C and I stopped the test. After this little endeavour, the TRUE Cu was RED HOT




And now the goods - I ran Prime at 4.2GHz (200x21) with 1.4 vcore, HT on, with the same fan on both CPU coolers: Silverstone FM122 @ 2400 rpm.

Xigmatek Dark Knight: 90C and rising (Test aborted < 3 mins)
TRUE Cu: 68C max after one hour

Yes, I did reset Real Temp, but look at the 62-65C load here



So there you have it...the TRUE Cu stomps all over the "Heat-pipe Direct Touch" nonsense. I expected maybe 5 degrees, 10 at the most...not once did I expect the TRUE to outperform the Dark Knight by 20C+ at a vcore of 1.4.

I will post more pics tomorrow to show the fan setup.

[BeepBeep2]

Wow, thats really interesting.
Going from a Xigmatek SD964 to water cooling (see sig) gained me 10c on a 965BE @ 4 Ghz 1.6v. (~61 to 51)

You might want to check your CPU IHS for flatness...if it wasn't flat chances are the TRUE's base fit it better than the HDT cooler. (Everyone knows TRUE's and how bad the base flatness can be)...

But a Xigmatek S1283 should perform better than what it did...

also, switch that fan out. It's one of the worst high-RPM fans on the market. I own an FM92 btw, it's trash.

[Metroid]

I have the TRUE CU. It is really neat. I used it for few months, temps used to go above 80c because the case + room temperature were +33c.

[zalbard]

You have really nice temps. My i7 920 @ 1.4V vCore used to have some crazy temps on air, lol.

[LennyRhys]

But a Xigmatek S1283 should perform better than what it did...

Nope - it just isn't capable of producing the goods when vcore is as high as I had it. With the FM122, the Xigmatek did remarkably well up to about 1.35v, where load temps peaked at 74C (by comparison, the TRUE peaked at 61C ). Beyond that vcore, it simply couldn't produce a proportionate rate of thermal transfer, unlike the TRUE.

Re the Silverstone fan being trash: I've had it for over a year now and it's not a bad fan; it might not be as good as other high RPM fans, granted, but it's a darned sight better than any 120 x 25mm fan I have, and it's better on a TRUE than two silent fans push/pull (I tested this last year)

[poparamiro]

Xigmatek for the victory...ahahahahahahaaaaa

[BeepBeep2]

Re the Silverstone fan being trash: I've had it for over a year now and it's not a bad fan; it might not be as good as other high RPM fans, granted, but it's a darned sight better than any 120 x 25mm fan I have, and it's better on a TRUE than two silent fans push/pull (I tested this last year)

I mean noise to airflow ratio It makes way more noise than say, a Yate Loon High Speed or Delta.

[prtuc2]

This thread want to make me use my TRUE Cu again, then the weight is probably too much for vertical position.

[LennyRhys]

I mean noise to airflow ratio It makes way more noise than say, a Yate Loon High Speed or Delta.

Maybe a Yate Loon, but not a Delta...never lol; guaranteed it's a noisey 2400rpm fan, but compared to a 3200rpm Delta it's not noisey. Point taken though - the noise-to-airflow ratio is pretty poor compared to some of the 38mm fans out there. I'm looking into getting something with better static pressure and CFM to use as a heatsink fan; just got six 120mm Xilence Red Wings (again) to use as case fans (they are brilliant, and best of all they are red and black ), and I'm thinking about getting a Gentle Typhoon to use on the TRUE.

This thread want to make me use my TRUE Cu again, then the weight is probably too much for vertical position.

Actually it's not. You don't even need to support the cooler in the vertical position (Thermalright tested it themselves apparently for a week and saw no ill effects to the motherboard), but I'm doing it anyway.

After being mounted, the thing is absolutely SOLID; doesn't twist or sway unless the motherboard twists or sways with it.

To support it, I have just a small piece of thread tied to the top of the case - even just a slight bit of tension in the thread takes most of the strain off the motherboard/backplate, and in my opinion that's all it needs to be as safe as any other cooler.

Contrary to popular opinion, it's not the weight of the cooler that matters but how the weight is distributed, i.e. its centre of gravity. The TRUE Cu is pretty heavy at the base so the bulk of the weight is near the motherboard rather than being far away from it.