Innovation Cooling's Diamond 7 TIM test results

[don_xvi]

A late result !
Finally replaced my old mount (about 6 weeks old) which was either Arctic Silver 5 or Antec Formula 5 (I'd originally thought this was AS5 until I got a tube of real AS5 and noted the consistency difference).
Processor is Q6600 at 3.4GHz, Vcore loaded about 1.28. Load temp reduction across all cores of 4-5 deg C. Neat stuff !
The old paste may have been a bit thick upon disassembly inspection, but that's a for sure performance improvement !

[DDTUNG]

It was a crude attempt on my part to steer the discussion slightly away from an obviously very technical direction towards something a little more practical.

DDTUNG

[tastymannatees]

It was a crude attempt on my part to steer the discussion slightly away from an obviously very technical direction towards something a little more practical.

DDTUNG

As 123 bob pointed this exercise is part beta testing and part marketing. We are dancing some angels on the heads of pins in some of the discussions but I like to think there is some method to the madness. We will find out if pressure or contact is more significant which is practical. I may learn something practical about Temp sensors etc.

I know what the paste does inside and out, backwards and upside down. I am chasing user experience here, Lab results cannot duplicate that and short of having a forum party in my lab this is the next best thing. This is real world testing. I am picking a lot of info on a variety mounting hardware, pressure contact info on a on a cross section of systems. Through forum feedback we have refined paste application and am acquiring info on trouble shooting and the marketing message.

I could not get there from here without the effort of the forum members here which is greatly appreciated. You people have practical a knowledge, background experience I do not have and would be hard pressed to come by. Thanks again for your efforts and patience.

[64dragon]

just got back from my senior project competition (in Tennessee) and still don't have a package of pressure film tasty, did you send me any? i've got some reading to catch up on around here

[tastymannatees]

just got back from my senior project competition (in Tennessee) and still don't have a package of pressure film tasty, did you send me any? i've got some reading to catch up on around here

You were one of the two I ran short on in the first round- However I have new film and loaded up the envelopes this weekend and in the mail Monday. PA should be there in a couple of days

[123bob]

It was a crude attempt on my part to steer the discussion slightly away from an obviously very technical direction towards something a little more practical.

DDTUNG

DDT,

Your question is a valid question to us here. In the time frame you were asking, it was a bit early. Now is the time, and now we have some results. If I can clock the farm up from this experience, so much the better.....I'm all for that.

EDIT: Farm-07 has spent almost 24 hours at the new clock, no errors, and valid scores are coming in on the
WUs. If you had not asked the question, I may not have re-clocked it for a while. So, thanks for steering me back to it. Now I'm going to see what happens on the other machines that I'm done TIM testing on. Of course, I'll report on it...

To all,

To the other owners of "QX" series processors, I would like to ask a couple of you to download Real Temp and run the Sensor test on it. Please report your results. I'd like to know if the QX series of CPUs is the only ones that are fully calibrated. Please report your results and stepping. Even if you are not doing this paste test. Thx.

EDIT: It should only take a couple of minutes to run. You put the machine at idle, stabilize the idle temp, then hit the button. I'm wondering if you get the same number on each core like I do on my main rig. (11,11,11,11 in my case)

Regards,
Bob

[64dragon]

You were one of the two I ran short on in the first round- However I have new film and loaded up the envelopes this weekend and in the mail Monday. PA should be there in a couple of days

thats fine, i've not had free time until now anyways. i cant wait to see my laped ihs and heatsink do

[ShootStraight]

9,8,8,7 QX6700 B1 @ Stock.

-SS

[123bob]

9,8,8,7 QX6700 B1 @ Stock.

-SS

Thx SS. It may just be a fluke that my only QX shows the same number across the cores. I was hoping that at least the QX series might be calibrated. For the premium we pay for those, I would have expected it......

Regards,
Bob

[ShootStraight]

It could also be that I didnt stabilize the temps too much prior to testing - maybe 15 secs or so, and this B1 could double as a cigarette lighter.

-SS

[SiGfever]

@Bob and SS,

Do you find Real Temp to be more accurate than Core Temp? My two Q6600's read cooler in Real Temp but my e6600 reads the same with both programs.

[123bob]

@Bob and SS,

Do you find Real Temp to be more accurate than Core Temp? My two Q6600's read cooler in Real Temp but my e6600 reads the same with both programs.

I don't really know. They use different TjMax values which changes the offset some. RealTemp reads lower than Coretemp. This was based on some testing the author did. Info is in the link I provided above. It could all be cleared up very quickly if Intel would just release the figures. They have not done so yet. I've been running the two programs side by side. They appear to give similar numbers when you consider the TjMax difference. I have come to understand that coretemp may be misreading VID. This is in the thread on our forum. CPUz agrees with Realtemp on this.
(There's a config change you can do in CPUz to make it read VID and not the actual. It's in the thread.)

I had hoped at least the QX series were fully calibrated. I'd like to see more results on that. If they are accurate, we may be able to accomplish some real work with the QXs and perhaps provide some better data. I own exactly one of them. It is the only chip I have that reads the sensor test equally among the cores. (My test group included Xeons, Q6600s, and a couple special chips....I would have expected at least those to be consistent, but they are not.)

Regards,
Bob

[David_L6]

this B1 could double as a cigarette lighter.

-SS

I have a QX6700 B1 like that. It runs HOT!!!


A Q6700 (with a TRUE) and two Q6600's (one TRUE, one Zalman 9700) are averaging just about 3°C cooler with the Diamond 7 TIM than with AS5. The Diamond 7 has impressed me.

[ShootStraight]

@Bob and SS,

Do you find Real Temp to be more accurate than Core Temp? My two Q6600's read cooler in Real Temp but my e6600 reads the same with both programs.

While I'm no expert (I'm not even a novice) on temps, I'm not sure its giving any more accurate of a temp. My reading suggests that the cooler reading is more of an manipulation (interpolation) rather than actual difference in temp, and it only matters under the "idle" parameter.

http://www.tomshardware.com/forum/24...town-real-temp

If I'm wrong here, or been mislead I'm sure some sharp person in the know here will set this straight.

-SS

[[XC] riptide]

My 1st Quad, QX6700 B1 Extra Special, runs at a cool ~70C ........ under decent Water. @1.45vcore 3.2Ghz.

[[XC] riptide]

To be fair to the ID7 I can see how in some situations those 2-3 degrees would help achieve stability at a certain clock and voltage.

DDTUNG

This is true. 2-3C is the difference between random reboots every couple of hours/days to no reboots.

Victor, I believe how you can get a higher overclock, simply because you can supply a little extra voltage if needed. However, for us crunchers, I don't think it'll make much of a difference, since we're never on the bleeding edge of stability.

What? You should be on the bleeding edge of stability, given the thermal ups and downs of your house/room. 2-3C can break it.

[tastymannatees]

""Now calibration.......I agree with your points above and must give it some more thought too. I may post up the questions in the Xtreme thread on the topic. (When I have intelligent questions to ask...) Maybe there is some mis-understanding of what is meant by ambient. My thought on it is that it's the room temp outside the box. Why I would calibrate to something that may or may not correlate to what's inside the box makes no sense to me either. I would think I would want correlation to the inside of the box.""

Why not try it?

[D_A]

Ya know, I thought what we were after was a relative decrease in temperature from one setup to the next, but I do tend to over simplify things a tad.

[123bob]

Ya know, I thought what we were after was a relative decrease in temperature from one setup to the next, but I do tend to over simplify things a tad.

That is the desired result. But with everything temp related in computers, it never seems to stay that simple. There are so many variables at play that it makes it hard to get accuracy. I have found this thread to be more useful to me than most others I've seen.
The nice thing is that for those who do not wish to dive deep into this, they don't have to. For those who do, well come on in, the water's fine. (Even if we can't figure out how warm it is yet...) If we find some "eureka" stuff, we can then share it with everyone, and it will be based on solid facts.
Do I really need to spend the amount of time on this that I'm doing? No, but I'm finding it interesting, I managed to increase the farm clocks, and it's keeping me playing with hardware while I wait for Nehalem to come out... This is cheaper than visiting Dr SiG every couple of days.....

Regards,
Bob

[tastymannatees]

New Temps and clocks.
3.318 gig - CPUz volts @ load 1.368
Ambient 26.4
Coretemp idle 47,47,44,45
Coretemp load 68,68,65,67

I was playing around with 123bob's numbers. using the Zalman sink. his delta number 19C and 150 watts I come up with a 12.7 C/W.

For the overclocked 6600 approx running 200W. so the temp rise over the ambient would be around 25.3 C.

For calibration approx. 22 C ambient + 6C idle adjustment = 28C CPU temp should be 53.3 C. Average core temp is 67C So you have a 14C delta on the on the absolute measurement. I would assume some adjustment higher on the ambient side closed case but it is a ball park calc. for your absolute temp number

[tastymannatees]

emailed some pressure tests results today

[123bob]

I was playing around with 123bob's numbers. using the Zalman sink. his delta number 19C and 150 watts I come up with a 12.7 C/W.

For the overclocked 6600 approx running 200W. so the temp rise over the ambient would be around 25.3 C.

For calibration approx. 22 C ambient + 6C idle adjustment = 28C CPU temp should be 53.3 C. Average core temp is 67C So you have a 14C delta on the on the absolute measurement. I would assume some adjustment higher on the ambient side closed case but it is a ball park calc. for your absolute temp number

I'll try some closed case temp tests shortly. One of the interesting things to try. I'm quite sure my case temps are higher. Of note is that I'm measuring ambient on this machine with a Radio Shack device taped to my front panel on the rack. This is not exactly what the inlet of farm-07 sees. It is however consistent to itself anyway. On Farm-09 all those ambient are reported using a small weather temp gauge at the inlet of the fan I have blowing across the board. This rig is on an open shelf so it's ambient is close to what the CPU and TRUE are really seeing.

I know we are looking more for delta ambients here, so I'm not sure if it matters much. This is to compare our coretemp/realtemp numbers to so we can adjust them.

[SiGfever]

Here are my test results. "Sample 2A" and "Sample 2B"

[tastymannatees]

What happend to the first one? USP runvover it with a truck? I can send you more on that one looks like it was damaged in shipping?

SiGfever and others a Heads up please include the thermal before and after and also systemem, specs - cpu & heat sink as the analysis is hard to do without going through multiple posts and it would be easier to take it all in at a glance and compare to other results.

Also the full spectrum of prints would be nice to see where the pressure is happening on the IHS, whether it is biased to one side or the other more in middle or edges etc

[SiGfever]

What happend to the first one? USP runvover it with a truck? I can send you more on that one looks like it was damaged in shipping?

SiGfever and others a Heads up please include the thermal before and after and also systemem, specs - cpu & heat sink as the analysis is hard to do without going through multiple posts and it would be easier to take it all in at a glance and compare to other results.

Also the full spectrum of prints would be nice to see where the pressure is happening on the IHS, whether it is biased to one side or the other more in middle or edges etc

Sounds like a plan. I was apprehensive about shipping it without some sort of cover to protect them, now I know why. Sorry for not including the specs, I assumed you recorded them from out first posts.