Apogee from Swiftech...

[BillA]

Originally Posted by nikhsub1

Yeah this would be a good test. I would bet the storm should really shine over the apogee with no IHS... let us know orkan. HTFU already.

am I correct in understanding that this is the last leg of your and Stew's claim of misrepresentation against Swiftech ?

I see non-Swiftech reports of similar performance for the 2 wbs with IHS, so I'm thinking the 2 of you may have dropped that line of attack ?
w/o ever admitting that Swiftech's representations were 'not unreasonable' of course, this is the 'net where one can dump and run

[nikhsub1]

Originally Posted by BillA

am I correct in understanding that this is the last leg of your and Stew's claim of misrepresentation against Swiftech ?

I see non-Swiftech reports of similar performance for the 2 wbs with IHS, so I'm thinking the 2 of you may have dropped that line of attack ?
w/o ever admitting that Swiftech's representations were 'not unreasonable' of course, this is the 'net where one can dump and run

Bill, I never said there was intentional misrepresntation. I do believe that solely using the TTV is not a smart idea.

[flatline]

http://www.investormarketplace.com/p...MP&ed=20051101

[Budwise]

The good news, there was no cracked top, and i have ZERO leaks on my first watercooling system.

The finish on the block was excellent. No nicks or anything. Extremely smoothe.


Here is during some leak testing


Here is the final look



And the bad news...

As you can see, this is what i saw as i opened it up. There is one shaving just sitting there in plain view. The is obviously poor QC... I truly thought that they had tightened down on QC since the last batch. Guess not...


These are the smaller two shavings that i found. There was one big one about 50% larger than these, but it was small enough to get lost on my desk lol...

[Cathar]

Originally Posted by BillA

am I correct in understanding that this is the last leg of your and Stew's claim of misrepresentation against Swiftech ?

Bill. Cut the crap will you? At no stage did I ever claim that Swiftech intentionally misrepresented anything.

I see non-Swiftech reports of similar performance for the 2 wbs with IHS, so I'm thinking the 2 of you may have dropped that line of attack ?

All that was ever said was that there were suspicions that TTV results were inconsistent, and that IHS's are inconsistent.

w/o ever admitting that Swiftech's representations were 'not unreasonable' of course, this is the 'net where one can dump and run

Troll-bait.


Posted at [H]ardOCP, but ignored:

Originally Posted by Cathar

Thing I don't understand is this. The documents linked to by Marci refer to the Tjunction, which by itself would completely satisfy me, and most any overclocker, about the temperature of the CPU/heat die. Tjunction being the temperature positionally located in the middle of the die, located between the die itself, and the CPU packaging substrate. It'd be the next best thing against the most desirable option, which would be having access to TCC readings ( TCC being the on-die thermal probe located within the ALU on P4's).

So with that, it seems that the present Intel supplied testbeds given out for validation purposes do not allow for the measurement of Tjunction. If they did, then in my mind, that would totally absolve any concerns that overclockers have, being that rather than merely knowing the IHS surface temp, we now know what's going on at the die level as well. Sure, we can argue for days about whether or not Tjunction is truly representative of the die temperature, but I can say one thing, it is better than nothing at all.

Having Tjunction, along with Tcase, would allow for at least some quantification of TIM1 in all scenarios. Intel themselves appear to clearly be using Tjunction to perform internal assessments of TIM1 as pressures vary, and since much of the call for TIM1 quantifications seem to be evidenced in geometrical and pressure variations of cooling devices, this would seem to me to be what is needed to solve the issue, and allow testers to quantify TIM1, and allow testers to quantify the die temperature in a more precise sense than merely measuring Tcase, which is itself subject to variations due to wb flatness.

Using Incoherent's flux block method to quantify TIM2, and then we have pretty much all the answers we're after.

So a call for consistency is now to be interpreted as accusing a company of "intentionally misrepresenting"?

To think that some have the gall to claim that others have some hidden agenda.

[Budwise]

im at 2700 1.45Vcore with a 170. 26C Idle 38C load. Is it just me or does this seem high?

[BillA]

agreed Stew, we should bag this whole thing
if you have all you are after, why the need to take issue with CPU testing by others for others ?

"intentionally misrepresenting"
your quotes
identify the source please

yes Stew, I do think you misquoted me intentionally (liar apply ?)
I now view that 'misunderstanding'/wandering off track as a gambit you use repeatedly
I am stopping this game as I do not like the product, endless debate
no more objections from here to your comments Stew (and Scott too),
my insistence on an objective appraisal has produced pointless acrimony - and no testing improvements, my failure

[Cathar]

The misunderstanding of others is not my problem, it's theirs. It only becomes an "issue" when they attempt to turn it into a problem that never existed. Throughout all of it, I have only seen one person disagree, accuse of hidden agendas, and accuse of lying, that being you. I only questioned your agenda after persistent baiting and snide comments, and even then, it was expressed as a rhetorical question, rather than an outright accusation as you so commonly lay at other's feet.

I want to see consistency. I represent the enthusiast overclocker first and foremost, and throughout our discussions in the past we have commonly understood that you and I have differing points of view about what is most important. I want qualification of die-temps, whether exact or as close as in reasonably possible, as these are of most importance to the enthusiast market, such as has been the case for the last 5 years.

When I see testing methodologies recommended that ignore die temps, and make assumptions about die temps, then I have an issue, and I have raised that same point consistently without diversion for the last four years. This whole Apogee vs Storm thing is just crap. It has nothing at all to do with the real issue, and the real point, but it seems that some are hell bent on attempting to dismiss very valid concerns under the weight of slanderous accusations that this is all about some stupid and irrelevant hidden agenda.

I have worn such accusations from yourself, and in private from Gabe, and I am sick to death of it. I represent the voice of the overclocking community that is being trampled on for purposes of CPU manufacturer testing convenience that quite easily and provably does nothing to serve the direct needs of the overclocking community.

Some, or even most, seem to understand that clear as day. A few, it seems, want to silence it, or call it irrelevant, for whatever reasons I do not know and still do not understand, suffice to say that those reasons appear to be due to a lack of understanding of the community's wants.

[gone_fishin]

http://www.xbitlabs.com/articles/cpu/print/p4-temp.html
That is a good article explaining the ALU temp diode and its function on the P4.

Quote
"Intel claims that the integrated thermal diodes are calibrated specifically for each given CPU on the manufacturing stage" End quote


So as we see, each diode is calibrated differently for each cpu to match against an equally unique "critical value" for each cpu. I do not see them being good comparative values from cpu to cpu for test benching.

[Cathar]

gf, I believe that they meant that the diodes are specifically calibrated for shutdown at 135C during the manufacturing stage. All diodes require calibration, and they just can't leave it to chance that some random diode on some CPU they make won't correctly shut the CPU down when it hits the critical 135C. I believe that this is what is meant by that statement. Otherwise the statement makes no sense. Why would anyone calibrate a diode to read differently to other CPU's? The act of calibration essentially means to bring the diode within acceptable error margins, which would make the ALU temp probe specifically ideal and consistent from CPU to CPU.

[gone_fishin]

Originally Posted by Cathar

gf, I believe that they meant that the diodes are specifically calibrated for shutdown at 135C during the manufacturing stage. All diodes require calibration, and they just can't leave it to chance that some random diode on some CPU they make won't correctly shut the CPU down when it hits the critical 135C. I believe that this is what is meant by that statement. Otherwise the statement makes no sense. Why would anyone calibrate a diode to read differently to other CPU's? The act of calibration essentially means to bring the diode within acceptable error margins, which would make the ALU temp probe specifically ideal and consistent from CPU to CPU.

Each cpu has its own critical heat point. Some can function fine at higher heat while some have a lower threshold.

The ALU diode reading does not shut down the cpu (and readings are spaced in milliseconds). If you read the article you will see that it only triggers the send delay, to cool the alu (which are running at twice the cpu clock btw). The other on die diode is the shut down one, the one that the motherboard can read.

Besides the ALU is rapid functioning and you would only be able to get an average reading because the measurements are taken in milliseconds. So when instructions are being done there will be high heat. If send delay is triggered you will get artificially low averages so theorietically a poor heatsink could give the same reading as a good one.

In conclusion it is a folly to think that if only Intel would let us read the alu diode all would be fine. It is there for a specific function and would not give an accurate die temp for determining heatsink performance.

[Cathar]

GF, you are referring to the Thermal Monitor and the Thermal Control Circuit (Tcc) diode that I occasionally talk about (but I am not referring to Tcc in this post). I think that you will find that Tcc which performs the CPU shut-down is NOT the same diode as the motherboard readable diode .

As all the diodes and circuitry of Thermal Monitor and Thermal Control Circuit are integrated into the CPU core and are independent of the mainboard

For your information: the temperature when THERMTRIP# signal is sent equals about 135oC, according to Intel.

The rest of the article goes on to demonstrate how the motherboard readable diode reads nothing like 135C, when THERMTRIP gets triggered and the CPU shuts down.

Unfortunately, we couldn't continue the test any further as MIR-253 incubator can only provide +50oC in its chamber. So all we could do, was to turn off the CPU cooler. After that, the CPU temperature rose to 94oC in less than one minute and the system shut down.

I also believe that you are nit-picking to the highest extreme. Yes, ALU temperatures will fluctuate, but over the period of a few milliseconds the thermal capacity of the silicon itself will not allow for huge variations in the matter of that short period.

The diode is calibrated against the a standard temperature scale, I fail to see how that cannot be taken to be representative of being valid across all CPU's of that type.

In any even, when I referred to Tjunction above, I was not talking about the Tcc probe that is the subject of the XBitLabs article.

[gone_fishin]

The thermal control circuit is what makes the pauses in the cpu cycles, it is not a temp measuring device or diode in any way.


Also,
Quote above article,
"In order to save the CPU from damages in case of such breakdowns, the second thermal diode built into the CPU keeps track of another CPU temperature value. When this value is notched, it doesn't "slow down" the CPU anymore, but sends THERMTRIP# signal for the system to shut down"

Clearly this second diode is what shuts the computer down, it is NOT the one near the ALU.

[Cathar]

Originally Posted by gone_fishin

The thermal control circuit is what makes the pauses in the cpu cycles, it is not a temp measuring device or diode in any way.

Tcc, referring to the Thermal diode that feeds the Control Circuit.

[Cathar]

Originally Posted by gone_fishin

Clearly this second diode is what shuts the computer down, it is NOT the one near the ALU.

How can you come to that conclusion? So where is it then?

I am instead drawing my information from published Intel documents, rather than a third party web-site:

http://www.employees.org/~slf/it04021.pdf

http://www.intel.com/technology/itj/...ol09_art03.pdf

[gone_fishin]

Originally Posted by Cathar

How can you come to that conclusion? So where is it then?

I am instead drawing my information from published Intel documents, rather than a third party web-site:

http://www.employees.org/~slf/it04021.pdf

http://www.intel.com/technology/itj/...ol09_art03.pdf

Ok we'll use your link, the first one.

Here come some lines of text from it.

"In addition to the thermal monitor, all Intel® processors include an on-die thermal diode."

"This thermal diode is separate from the thermal
monitor’s thermal sensor and cannot be used to predict the behavior of the thermal monitor."


Clearly 2 diodes.
the thermal monitor, and the on die thermal diode.

"Because of the speed of
temperature changes in “hot spots” and thermal gradients across the die, it is possible that the TCC may be active while
the thermal diode is measuring a temperature within normal ranges."


Same as I was saying.

[Cathar]

Nowhere does it state that the user visible diode is the one that control THERMTRIP. If it was, then the XBitLabs THERMTRIP test would've shutdown at 135C, rather than the indicated 94C.

Which is the same as I always have been saying about the user visible diode: "Numb and Dumb" - it doesn't reflect the actual peak CPU temperature.

A third diode perhaps? Or perhaps, and more likely, a second thermal trip sensor (ie. THERMTRIP) calibrated to 135C and driven by the same thermal diode that drives the PROCHOT trip sensor, which in turn drives the Control Circuit.

[gone_fishin]

"PROCHOT# and THERMTRIP#
Processors assert PROCHOT# when activating a thermal control
circuit that can automatically throttle processor clock speed.
When the junction temperature exceeds a critical threshold, the
processor asserts THERMTRIP#. Upon a THERMTRIP# trigger,
Intel processors attempt to stop internal clocks and halt program
execution to reduce internal temperature and avoid processor
damage"

I get that THERMTRIP is based on a "junction" temperature and PROCHOT is clearly activated by the sensor reading near the ALU from previous readings.

http://www.compactpci-systems.com/co...fs/11.2002.pdf

[Cathar]

Ok, so a 3rd thermal diode then? Or perhaps not.

Reading this document again, the diagram at the top of Page 7, seems to imply that Tjunction is the true name of the thermal diode that the thermal control circuit actuation (PROCHOT) is coming from. As per your document, Tjunction is also driving THERMTRIP, which brings us back to the original statement:

Or perhaps, and more likely, a second thermal trip sensor (ie. THERMTRIP) calibrated to 135C and driven by the same thermal diode that drives the PROCHOT trip sensor, which in turn drives the Control Circuit.

What I referred to as Tcc (that drives the monitor and control circuity) & Tjunction are the one and the same sensor.

Tjunction is what I called for to be user visible.

What's the problem again? Are you arguing that we should not be interested in the peak die temperature? Sorry, just don't know where you're going with all of this, other than arguing semantics over the naming of various sensors.

[gone_fishin]

Originally Posted by Cathar

Ok, so a 3rd thermal diode then? Or perhaps not.

Reading this document again, the diagram at the top of Page 7, seems to imply that Tjunction is the true name of the thermal diode that the thermal control circuit actuation (PROCHOT) is coming from. As per your document, Tjunction is also driving THERMTRIP, which brings us back to the original statement:



What I referred to as Tcc (that drives the monitor and control circuity) & Tjunction are the one and the same sensor.

Tjunction is what I called for to be user visible.

What's the problem again? Are you arguing that we should not be interested in the peak die temperature? Sorry, just don't know where you're going with all of this, other than arguing semantics over the naming of various sensors.

Lol, sorry to laugh Cathar but where are you going with this? First you say the THERMTRIP is wildly inaccurate then you want Intel to make it accessible.

PROCHOT is totally independant of THERMTRIP. From my understanding of your earlier argument you wanted Intel to make the reading at the hotspot available. This would be the PROCHOT sensor reading near the ALU. My argument is that it is independently callibrated per cpu and takes readings in milliseconds near a wildly fluctuating heatsource (ALU) making it unsuitable for our purposes.

[Incoherent]

Originally Posted by gone_fishin

...it is independently callibrated per cpu and takes readings in milliseconds near a wildly fluctuating heatsource (ALU) making it unsuitable for our purposes.

I would argue that that is very suitable for our purposes, as long as you knew what you were looking at, integrated sensibly and established a transfer function to reality.

...

[GlassMan]

Back handed way to use the TCC for measurement http://forums.procooling.com/vbb/showthread.php?t=12516

http://download.intel.com/design/Pen...s/30255304.pdf Section 4 discusses both systems.

[pHaestus]

goddamn you guys will argue about EVERYTHING. ALL temperatures taken from diodes have to be calibrated lol. What that article is saying doesn't mean what you think it means. It's just a part of the equation for transistor ->diode -> temp comversion.

and since it is just a transistor then yea sampling speed and accuracy are inversely related because the reader has a finite number of bits to work with.


I do the same type of calibration on every diode I solder onto to eliminate any possible issues with solder joints, reader problems, and intrinsic diode nonlinearity factors etc.

[Marci]

goddamn you guys will argue about EVERYTHING.

but... but... it's the one thing we're ALL good at!

[BillA]

ah procooling
a larynx cooler, beer of course

(2 points Marci)