help explain TDP and TcaseMax

All about them.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Hifriday
Patron of SPCR
Posts: 237
Joined: Thu Aug 05, 2004 3:32 pm

help explain TDP and TcaseMax

Post by Hifriday » Fri Apr 14, 2006 1:52 pm

I have to admit that I haven't paid much attention to TDP in the past, but now that AMD is programming the actual TDP of each individual chip makes it especially interesting for selecting the perfect CPU for a silent PC. Thanks to SPCR's and The Coolest's TcaseMax for bringing our attention to this. Unfortunatley my knowledge on the matter was quite limited, however I did find some useful information in AMD Opteron Processor Power and Thermal Data Sheet (pdf) and AMD Athlon 64 Processor Power and Thermal Data Sheet (pdf) and have compiled my interpretation below. I hope those more knowledgeable in this topic would help correct, clarify, and add to this.


Starting with it’s Revision E cores, AMD has begun programming individual chip’s TDP rating into their processors. In the past AMD used a conservative maximum TDP rating that was often the highest theoretical TDP across several models and all production batches. TDP was often used to compare how power hungry certain brands/lines of processors were against others, but now with this chip specific TDP, there is a way to quantify the sample-to-sample production variance and determine how a particular chip will perform in terms of power consumption. This becomes much more interesting to overclocking and silent PC enthusisasts. A low TDP chip will consume less power leading to less heat that needs to be dissipated (from both the CPU as well as the PSU) which means it can be cooled more easily or quietly. Conversely a higher TDP or a chip's ability to consume more power, is also speculated to indicate a better overclocking chip.


TDP – WHAT IS IT?
TDP stands for Thermal Design Power, or also called Thermal Dissipation Power.

In AMD support forum we find this definition:
This is the maximum theoretical amount of power (in Watts) that a processor may consume and therefore dissipate as heat. TDP-values are crucial when it comes to designing cooling solutions as they specify the maximum amount of power (=heat) that a cooling solution must be able to dissipate.Note that AMD's TDP-values are absolute maximum, i.e. 'worst case' ratings.During normal operation, an AMD Processor will typically not reach its specified TDP. Other manufacturers may have a different definition for their TDP-values, e.g. they might give typical instead of absolute maximum ratings, i.e. they might specify the amount of power dissipated during normal operation (='typical' conditions).
In AMD’s Processor Power and Thermal Data Sheets (APPTDS) we find in the notes:

For the maximum theoretical ratings:
Thermal Design Power (TDP) is measured under the conditions of Tcase Max, IDD Max, and VDD=VID_VDD, and include all power dissipated on-die from VDD, VDDIO, VLDT, VTT, and VDDA.
And for the chip specific ratings:
Thermal Design Power (TDP) and IDD max for Rev E and later processors are the limits at the highest Tcase max in the specified range for the corresponding OPN. Products will conform to the TDP and IDD Max limits at all valid voltages.

TDP – HOW TO FIND IT?
Unfortunately the individual chip’s TDP is not indicated on the exterior or packaging of the CPU. It is however programmed into the chip as a TcaseMax value. From APPTDS
Tcase max is the maximum case temperature specification which is a physical value in degrees Celsius. This value is programmed into Rev D and later processors..
Tcase max is programmed during device manufacturing with part-specific values for Rev E and later processors with 'Variable' indicated by the Case Temperature OPN character, and can be any valid Tcase max value in the range specified for the corresponding OPN
How is TcaseMax related to TDP? In APPTDS we find a series of thermal profile tables that translates Tcase Max to TDP based on the specific thermal profile of the processor.
The thermal profile is used to define the relationship between Tcase max and devicespecific Thermal Design Power for Rev. E and later processors with “Variable” indicated by the Case Temperature OPN character.
From the above definitions, it seems that during the manufacturing process, AMD will load each chip at the rated voltage/speed and measure the maximum case temperature and/or power consumption to obtain TcaseMax/TDP?

So how to read the TcaseMax from a Revision E or later CPU? A nice little program called AMD64 TcaseMax will read this value off your chip and automatically translate the value based on APPTDS tables. This handy utility can be downloaded here.


TDP – CLARIFICATIONS
Will the TDP rating be affected by the actual clock or voltage the CPU is running at? No, TDP is measured at the rated voltage, max P-state (or rated/stock speed), and assume under maximum load so regardless of your clock speed/voltage, the TcaseMax utility should give the same TDP. The actual power consumption of the chip however will vary with changes in clock speed and voltage.

So does this mean that a dual-core Opteron 180 with a chip rating of 1.35v, 49C TcaseMax, and 35.0W TDP consumes only 35.0W when running at stock speed of 2.4Ghz under CPU load? Less than half the power of a slower dual-core Athlon X2 3800+ with a smaller L2 cache rated at 1.35v, 71C TcaseMax, and 89.0W TDP running at 2.0Ghz?! Even nearly half the power of a single-core Athlon 3500+ with a TcaseMax of 65C, 67.0W TDP running at 2.2Ghz? Well if our interpretation of TDP is correct, then yes that is exactly what it means.

Borrowing numbers from SPCR’s Desktop CPU Power Survey we see various A64 CPU’s with their TDP read by TcaseMax and their measured CPU power consumption. Although the measured CPU power draw does not exactly match up to the rated TDP, it is within -2.3 to 13.2W. However if we rank the CPUs by measured power consumption under load (at stock voltage/speed) the order is nearly the same as the ranking by TDP. The only exception is the 3500+ Venice which has a slightly higher TDP rating (by 1.4W) than the X2 3800+ but measured in at 4.2W less. Considering that the motherboard used may not have been giving the exact rated voltage, the temperature/cooling differences of the CPU, power loss through the motherboard voltage regulation, as well as power measurement error, overall it seems that rated TDP does give a relatively good idea how the chip will perform in terms of power consumption.

Image

In addition, even with all CPUs idling at the same clock and voltage (1Ghz 1.1v) there is a difference in power consumption, and we see the ranking also very close to the TDP ranking. Again the only exception being the 3500+ Venice, which this time is drawing slightly more power (1.0W) than the higher TDP X2 4800+, but again this could easily be due to the factors mentioned above.

How does TDP affect the ability of a particular chip to be overclocked or undervolted? Well that is a question that unfortunately APPTDS does not seem to address. Any input that can help answer this question is welcome.


TCASE MAX IN DETAIL
Studying the APPTDS we find several figures that seem to determine the processor’s TDP. Each model will have a Thermal Profile that gives a specific Thermal Resistance (case to ambient) in C/W as well as a Tcase Max range. In addition there is a Local Ambient Temperature or Tambient. Plotting the various the Processor Thermal Profile tables, it seems that the relationship between Tcase Max and TDP is linear and can be roughly calculated using the following formula:

TDP = ( TcaseMax – Tambient ) / Thermal Resistance

Image

Each series in the graph represents a different specific Thermal Resistance (C/W). We can see AMD slightly rounded down the highest TDP for some profiles. Most processor profiles are based on a Tambient of 42C, except the FX series which has a Tambient of 40C.

We can see that different models have different Thermal Profiles and in general lower Thermal Resistance will give a higher TDP for the equivalent Tcase Max (and vice-versa). Will this give us any clues which models/lines of processors will have lower or higher TDP? Let’s look at a summary of APPTDS Thermal/Power Specifications tables. I’ve only included the Rev E chips, and this is per the last update of November 2005 for the Opterons, and March 2006 for all other models. In addition to giving the Thermal profile/Thermal resistance, the voltage range (VID_VDD) and TcaseMax range are also given for each CPU model/revision.

Image
** Data points corrected as these seemed to be typos on the APPTDS.

Before we start salivating at the 20.6W TDP (for silent PC enthusisasts) or the 110.0W TDP (for overclockers), please keep in mind this is just a general range given in AMD’s specs and does not mean that there are chips that fall on these extremes. However we do see that in general the dual-core chips have a lower Thermal Resistance of 0.20, notably the higher speed X2s and Opterons. This would suggest a higher TDP, but from user’s posting on the TcaseMax utility forum and on SPCR, it seems most dual-core Opterons have an exceptionally low 35.0W TDPs (which happens to correspond to the lowest possible TDP according to the specs) although one Opteron 165 also had a rating that went up to 105.0W (close to the maximum TDP spec of 110W). It seems that it would be too soon to draw any generalizations with the limited data on actual chip TDPs.

However it is interesting to note we do find an Opteron Thermal Profile with a very high Thermal Resistance of 0.95C/W and an extremely low 7.4W-30.0W TDP range, but alas none of the listed models have this profile. The lowest TDP range can be found for the s940 dual-core 260/860 series OSAxxxFAA6CB/CC with 0.51C/W Thermal Resistance, 1.15/1.20v rating, and 13.7-55.0W TDP range; as well as the s940 single-core 240/840 series OSAxxxFAA5BL/BM with 0.53C/W Thermal Resistance, 1.35/1.40v rating, and 13.2-54.7W TDP range.


AND SOCKET 754 PROCESSORS
None of the s754 processors seem to have the individual chip TDP ratings despite having also moved to Rev E. However we can look at their maximum TDP rating just for comparison purposes.

The Rev E Athlon 64 3000 (ADA3000AIK4BX) as well as the Sempron 64s both have a rating of 1.4v, 42C Tambient, and 0.45C/W Thermal Resistance. The TcaseMax is 65C giving a 51W TDP for the Athlon, and TcaseMax of 69-70C giving a 59/62W TDP for the Semprons. Again this is the overall maximum TDP and not the TDP of the individual chip which will likely to be lower. The high Thermal Resistance, however does suggest that these chips in general should have lower TDPs.

The Turion (MT only, no ML) are also included in the APPTDS. The Thermal Resistance of these chips are amazingly high at 2.00-2.08 C/W. However instead of TcaseMax we find a very high TdieMax of 95C, and also the Tambinet is a lower 36C. Possibly this is because Turions come without the heat spreader found on other AMD64 processors. The maximum TDP is listed at 24-25W. Using the formula from above we get slightly higher figures of (28.4-29.5W).

EDIT: Added RevE specific TDP notes from APPTDS; added mb vr loss.
Last edited by Hifriday on Sun Apr 16, 2006 8:35 am, edited 1 time in total.

ST
Posts: 67
Joined: Tue Apr 11, 2006 11:18 pm
Location: Silicon Valley

Post by ST » Fri Apr 14, 2006 3:33 pm

There is a lot of speculation on TCASEMAX and TDP in regards to temperature and overclockability but everything has been inconclusive to date.

Take for example, i had a X2 3800 stepping LDBHE 0601XPDW . My tcasemax = 71c, TDP: 89.0Watts. It was one of the coolest running dual cores i ever had (i have had about 6 now), showing only 40C at 2900MHz (http://i59.photobucket.com/albums/g319/ ... 0cnq-a.jpg).

One of the other benefits i see with high tcasemax and tdp chips is the ability to undervolt nicely. On my present Opty 165 with a high tcase max and tdp of 61C / and 95W respectively, i have been able to undervolt the chip down to .99V while still using a respectable 1938Mhz on it (http://i59.photobucket.com/albums/g319/ ... 4-1938.jpg).

So, in theory silent enthusiast should look for real low tdp/tcasemaxe numbers, there are still other underlying real world factors that cannot be accounted for readily.

Devonavar
SPCR Reviewer
Posts: 1850
Joined: Sun Sep 21, 2003 11:23 am
Location: Vancouver, BC, Canada

Post by Devonavar » Sat Apr 15, 2006 12:22 pm

Just one small comment. The CPU data that you took from SPCR's testing includes power losses in the VRMs on the motherboard, so the actual power consumption of the CPUs should be 10~20% lower than our data. As the amount of power drops, the proportionate amount of power lost to the VRMs increases, so as much as 50% of the power consumed when CnQ is enabled could be lost.

eternizer
Posts: 55
Joined: Fri Mar 03, 2006 12:36 pm

Post by eternizer » Sun Apr 16, 2006 3:52 pm

Hifriday wrote:Will the TDP rating be affected by the actual clock or voltage the CPU is running at? No, TDP is measured at the rated voltage, max P-state (or rated/stock speed), and assume under maximum load so regardless of your clock speed/voltage, the TcaseMax utility should give the same TDP.
I'm confused. In the AMD A64 TDP Poll, I saw different TDPs with seemingly the same CPUs (look at the different TDPs reported for venice 3000+ E6 CPUs).

Hifriday
Patron of SPCR
Posts: 237
Joined: Thu Aug 05, 2004 3:32 pm

Post by Hifriday » Mon Apr 17, 2006 6:22 am

eternizer wrote:I'm confused. In the AMD A64 TDP Poll, I saw different TDPs with seemingly the same CPUs (look at the different TDPs reported for venice 3000+ E6 CPUs).
Even if two chips have the exact same part number (ADA3000DAA4BP-Rev E3 Venice for example), their individual TcaseMax/TDP could be different. This is what is meant by the sample-to-sample variance or the "variable" rating given by AMD. This is also why it's interesting to see the exact same model (for example Opteron 165) being reported with TDP as low as 35.0W for one sample, and high as 105.0W for another.

However what I meant that TDP doesn't change with clock/voltage is in reference to each individual chip. For example if you own an Opteron 165 and it has an individual chip TcaseMax of 49C programmed in, then regardless if you are running with CnQ at 1.0ghz /1.0v or overclocked to 2.8ghz/1.6v, the software utility TcaseMax will still give you the same 35.0W TDP rating (as it is only converting the TcaseMax value programmed into the chip).
ST wrote:Take for example, i had a X2 3800 stepping LDBHE 0601XPDW . My tcasemax = 71c, TDP: 89.0Watts. It was one of the coolest running dual cores i ever had (i have had about 6 now), showing only 40C at 2900MHz
Temperature/cooling can be affected by many factors, but if my understanding is correct then in the exact same setup a 35.0W chip should certainly run cooler than a 89.0W chip (even if they aren't pulling their exact rated TDP, the 89W chip should be consuming more power). Even undervolted to the same voltage, I would expect the 35.0W chip to still draw less power (from the idle figures we see in SPCR's article). Did you have the other chips in a similar setup, do you recall their TDP/operating temps?

However which chip would overclock or undervolt better is much more speculative and I agree there seems to be no conclusive evidence or good theories yet?

SlaveToSilence
Posts: 218
Joined: Wed Aug 01, 2007 4:50 pm
Location: United Kingdom

Post by SlaveToSilence » Sun Jan 04, 2009 9:49 am

http://www.thecoolest.zerobrains.com/fo ... c.php?t=83

link doesnt work, is there somewhere else we can get it?

AZBrandon
Friend of SPCR
Posts: 867
Joined: Sun Mar 21, 2004 5:47 pm
Location: Phoenix, AZ

Post by AZBrandon » Sun Jan 04, 2009 2:38 pm

SlaveToSilence wrote:http://www.thecoolest.zerobrains.com/fo ... c.php?t=83

link doesnt work, is there somewhere else we can get it?
I never understood the purpose of this tool, but I think I do have a copy of it, although the copy I have is only for AMD. It is version 1.18 from February 2006.

SlaveToSilence
Posts: 218
Joined: Wed Aug 01, 2007 4:50 pm
Location: United Kingdom

Post by SlaveToSilence » Sun Jan 04, 2009 2:44 pm

would that be good to find out which 6000+ i have? i think its the 90nm but am not sure

AZBrandon
Friend of SPCR
Posts: 867
Joined: Sun Mar 21, 2004 5:47 pm
Location: Phoenix, AZ

Post by AZBrandon » Mon Jan 05, 2009 8:19 am

SlaveToSilence wrote:would that be good to find out which 6000+ i have? i think its the 90nm but am not sure
Maybe so. PM me an email address and I'll email it to you to try out.

gb115b
*Lifetime Patron*
Posts: 289
Joined: Tue Jun 13, 2006 12:47 am
Location: London

Post by gb115b » Thu Sep 10, 2009 6:57 am

sorry this is a bit of a n00b question

should all CPUs of equivalent TDP be roughly as easy to cool

basically i'm looking to upgrade an old 2.13 Ghz Core2Duo - which is rated at 65W TDP

and can see that i can go to a 3. something Ghz one which is 65W TDP

or even a Quad Core (S Series) which are also 65W TDP

should these remain equally as cool? or is that a gross approximation?

i assume their idle power draws would vary considerably (but maybe faster processors would spend longer in idle rather than at load? - would it even out?)

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Post by CA_Steve » Thu Sep 10, 2009 8:15 am

gb115b wrote:
should all CPUs of equivalent TDP be roughly as easy to cool
Roughly, yes. Note that AMD and Intel group different speed grades in the same TDP rating. So, the higher clocked CPUs will consume more power than the slower clocked CPUs.

gb115b
*Lifetime Patron*
Posts: 289
Joined: Tue Jun 13, 2006 12:47 am
Location: London

Post by gb115b » Fri Sep 11, 2009 12:38 am

so we're still saying the faster it is the harder it'd be to cool...

TDP isn't an absolute measure of how many watts it's drawing?

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Post by CA_Steve » Fri Sep 11, 2009 6:41 am

I think the very first post in this thread covers this in good detail...TDP is an absolute max based on max design conditions, but is not representative of the actual power draw of YOUR CPU in YOUR mobo as you won't actually be running at max conditions. So, everything being equal for 2 different 45W TDP CPUs, their load power will vary from each other dependant on mfg process variation as well as clock speed.

whispercat
Posts: 376
Joined: Tue Jul 22, 2008 6:05 pm
Location: Calgary, AB, Canada

Re: help explain TDP and TcaseMax

Post by whispercat » Fri Sep 26, 2014 3:29 pm

If one processor is rated at 140w and another at 65w, will they each draw the same power up to their maximum load when performing identical tasks?

Would they each idle at the same wattage?

In other words, is the only difference the headroom at max load?

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Re: help explain TDP and TcaseMax

Post by CA_Steve » Fri Sep 26, 2014 7:29 pm

whispercat wrote:If one processor is rated at 140w and another at 65w, will they each draw the same power up to their maximum load when performing identical tasks?

Would they each idle at the same wattage?

In other words, is the only difference the headroom at max load?
no, no, and no.

It's not just TDP rating, it's the processor architecture, it's frequency and voltage of operation, the process it's designed in, the motherboard design and components used, and more. If you said you had two Haswell i5 processors (same process, same architecture, same mobo), one rated at 65W and another at 84W, then we could drill down and realize one is just a lower speed grade part. At idle, they perform the same. Up to the speed of the S part, they will perform the same for both power use and speed of the application.

whispercat
Posts: 376
Joined: Tue Jul 22, 2008 6:05 pm
Location: Calgary, AB, Canada

Re: help explain TDP and TcaseMax

Post by whispercat » Sat Sep 27, 2014 11:55 am

CA_Steve wrote:
whispercat wrote:If one processor is rated at 140w and another at 65w, will they each draw the same power up to their maximum load when performing identical tasks?

Would they each idle at the same wattage?

In other words, is the only difference the headroom at max load?
no, no, and no.

It's not just TDP rating, it's the processor architecture, it's frequency and voltage of operation, the process it's designed in, the motherboard design and components used, and more. If you said you had two Haswell i5 processors (same process, same architecture, same mobo), one rated at 65W and another at 84W, then we could drill down and realize one is just a lower speed grade part. At idle, they perform the same. Up to the speed of the S part, they will perform the same for both power use and speed of the application.
Thanks. Sorry if I didn't preface my question with more qualifications. Yes, what I was after was: whether say, a Xeon processor rated at 140W would run any hotter than a i7 processor rated at 65W, given the same clock speed, same motherboard, same voltage (not overclocked), same components, same operating system, and performing the same tasks (at least the same tasks up to the point where the processor is taxed beyond drawing 65W)?

edh
Posts: 1621
Joined: Mon Mar 17, 2008 1:49 pm
Location: UK

Re: help explain TDP and TcaseMax

Post by edh » Sat Sep 27, 2014 11:45 pm

whispercat wrote:a Xeon processor rated at 140W would run any hotter than a i7 processor rated at 65W, given the same clock speed, same motherboard, same voltage (not overclocked), same components, same operating system, and performing the same tasks (at least the same tasks up to the point where the processor is taxed beyond drawing 65W)?
That's still not comparable. If there is any difference between the processors in physical architecture then you can not compre them like that.

Also what is 'hotter'? More power usage (heat) or higher temperature? Temperature is a measure of heat density and the two are not the same. Just to confuse things more if the Xeon has a larger die area then for the same power output you could expect it to run at a lower temperature as there is a larger conductive area to take the same amount of heat to the same heatsink.

What are you actually hoping to achieve? Maybe then we advise you better.

whispercat
Posts: 376
Joined: Tue Jul 22, 2008 6:05 pm
Location: Calgary, AB, Canada

Re: help explain TDP and TcaseMax

Post by whispercat » Tue Sep 30, 2014 12:29 pm

By hotter, I mean heat output into the case.

I'm thinking of getting a Xeon E5 1630 v3 for my next build. But only have experience with i7s as a thermal reference. I was just curious whether the Xeon would run hotter at idle or for doing the same work.

CA_Steve
Moderator
Posts: 7650
Joined: Thu Oct 06, 2005 4:36 am
Location: St. Louis, MO

Re: help explain TDP and TcaseMax

Post by CA_Steve » Tue Sep 30, 2014 1:32 pm

I think Intel got lazy and didn't bother to differentiate the 4/8 core/thread CPUs vs the higher count ones. No way will the E5 1630 V3 approach 140W. It's more likely to be the same as the socket 1150 i7's at 84-88W.Comparing the Devil's Canyon i7, Haswell-E 6 core and Xeon 4 core, leads me to say:
- The Xeon will consume about the same power as the socket 1150 Haswell in all cases (idle, light load, heavy load). What can completely alter the power use is difference in motherboard builds.
- The Xeon may run cooler than the socket 1150 i7 due to the difference in sockets and because there isn't an iGP.

edh
Posts: 1621
Joined: Mon Mar 17, 2008 1:49 pm
Location: UK

Re: help explain TDP and TcaseMax

Post by edh » Wed Oct 01, 2014 1:57 am

whispercat wrote:I'm thinking of getting a Xeon E5 1630 v3 for my next build. But only have experience with i7s as a thermal reference. I was just curious whether the Xeon would run hotter at idle or for doing the same work.
What are you intending to use the system for?

Post Reply