Intel 9th Gen Paid Benchmarks Take Advantage of NDA Periods

[leadeater]

4 minutes ago, Carclis said:

But hyperthreading also fairly significantly affects TDP, doesn't it? Each core would spend more time actually doing work.

It shouldn't, probably more about cache and memory bandwidth package power increase if it does. Be interesting to see that actually, I'm sure at least one reviewer has looked at that. Might try it out myself maybe. HT does nothing if the core is fully utilized which good benchmarking tools get fairly close to, CB is like 15%-20% (Edit: HT increase) from memory? That won't actually be 15%-20% extra power draw though, the core is at full power state already.

[Carclis]

5 minutes ago, leadeater said:

It shouldn't, probably more about cache and memory bandwidth package power increase if it does. Be interesting to see that actually, I'm sure at least one reviewer has looked at that. Might try it out myself maybe. HT does nothing if the core is fully utilized which good benchmarking tools get fairly close to, CB is like 15%-20% (Edit: HT increase) from memory? That won't actually be 15%-20% extra power draw though, the core is at full power state already.

If you find a comparison let me know because the 9700k vs 9900k TDP seems suspect even if you could somewhat rationalise the 9600k.

[Carclis]

13 minutes ago, leadeater said:

It shouldn't, probably more about cache and memory bandwidth package power increase if it does. Be interesting to see that actually, I'm sure at least one reviewer has looked at that. Might try it out myself maybe. HT does nothing if the core is fully utilized which good benchmarking tools get fairly close to, CB is like 15%-20% (Edit: HT increase) from memory? That won't actually be 15%-20% extra power draw though, the core is at full power state already.

Well I found this. It seems to suggest that power draw goes up when HT is enabled which would also suggest that more heat is produced.

[JoostinOnline]

On 10/9/2018 at 6:32 PM, Fooshi said:

Oh wow, Intel being scummy shits again. Who'd have thought?

 

Just stop buying Intel. It's literally immoral.

If we stopped buying from companies whenever they did something immoral, nobody would be able to access this forum.

 

Most people seem to either think that this is new behavior for Intel, or that other companies aren't just as bad. It's getting annoying.

 

[mr moose]

1 hour ago, Stefan Payne said:

The facts are that Intel Changed the definition and that the industry agreed on the older INTEL definition that they used for decades.

Look at this definition of TDP:

https://www.intel.com/content/dam/doc/white-paper/resources-xeon-measuring-processor-power-paper.pdf

 

Back in the day they took a shit on AMD's ACP definition, while using the same definition right now!

And it should be allowed to criticize what Manufacturers do, shouldn't it?
Or are you saying one should be banned for criticizing Intel for the shit they're doing?

 

And I am using Intels own definition of TDP as shown in that document!

 

 

Its a manufacturer's definition, not a "tech spec", that should be criticized and talked about.

 

Because right now, when using the Intel Definition, I could claim that my Ryzen 7/1700x has a TDP of 50W. AMD includes the Turbo mode of the Ryzen Desktop processors mostly in their TDP, look up reviews.

nVidia also uses the TDP or TBP in a similar way as the one that was standard for decades in the industry...

 

No, its not.

Its a manufacturers definition, that they themselves changed a while back and that should and must be talked about and criticized. And that is what a Forum should be for.

 

Then why did they redefine their "TDP" from average power with useful software to average power at base frequency and do not mention the maximum power of the Turbo mode??

Especially since it isn't really disclosed that great, is it?

 

Its not like this redefinition of "TDP" has caused problems in the past, has it not?
Like in Notebooks, the ones that are throtteling.

 

And read the Intel Document I've linked. They say so themselves!

And there should be an information about the maximum power consumption of the CPU for normal operation inside the manufacturer's defined operation.

 

And everything a manufacturer says and defines should be up for scrutiny!

That is what Forums are for, to disagree about that!

 

 

But since you said that Manufacturer's Specification aren't up for debate, then an AMD FX8350 is an 8 Core Processor. Because that's how AMD calles it.

And right now I'm using a 4 Core A10-7850K because the manufacturer specified it to have 4 Cores.

 

There was also some criticism towars Intel about the TDP in the past.

I'm totally fine if they would have called it "SDP":
https://www.anandtech.com/show/6655/intel-brings-core-down-to-7w-introduces-a-new-power-rating-to-get-there-yseries-skus-demystified

 

Well and it looks like their TDP Definition makes their processors look more power efficient than they really are...

Criticize manufacturers when they actually do something wrong, not because you don't understand TDP.   If someone kept claiming AMD were rebadging their 8350 as Ryzens in every thread and derailed them ad nuaseum with misplaced quotes and misunderstood papers you'd want them banned too. 

[mr moose]

4 minutes ago, JoostinOnline said:

If we stopped buying from companies whenever they did something immoral, nobody would be able to access this forum.

 

Most people seem to either think that this is new behavior for Intel, or that other companies aren't just as bad. It's getting annoying.

 

It's as annoying as the behaviour itself.

[Carclis]

2 minutes ago, JoostinOnline said:

Most people seem to either think that this is new behavior for Intel, or that other companies aren't just as bad. It's getting annoying.

Not new, but far more egregious. If you make it common knowledge how poor some of these instances are it might start to make the companies responsible look like idiots. A great example was Kyle's latest video.

 

[mr moose]

1 hour ago, Carclis said:

Well you can argue that Intel's motivation was not to handicap and deceive all you like but more recent history would suggest otherwise. Even so what they did was double down on the claims when they were challenged; something that was proven to be wrong later when PT redid their tests. If their motivations were to be transparent and give the public the truth then that would have been reflected in their public statement. Something like this would have sufficed:

"We are actively working with PT to verify and address your concerns."

 

I also believe his appraisal of the situation was affected by a poor understanding of many of the earliest points introduced from what was also a condensed list.

I guess you can't have what you want.

[leadeater]

2 hours ago, Carclis said:

Well I found this. It seems to suggest that power draw goes up when HT is enabled which would also suggest that more heat is produced.

Went up about as much as I'd expect, test needs to be run again though with all boosting disabled. That 3-7 watt difference could be a lot less at base clocks, with no MCE or other stuff that changes defaults which is actually standard on gaming motherboards.

[Carclis]

20 minutes ago, leadeater said:

Went up about as much as I'd expect, test needs to be run again though with all boosting disabled. That 3-7 watt difference could be a lot less at base clocks, with no MCE or other stuff that changes defaults which is actually standard on gaming motherboards.

It probably depends on the workload. If you're running Cinebench you'll see a large performance gain so I'd expect a larger difference in power as well. But who knows...

[mr moose]

20 minutes ago, leadeater said:

Went up about as much as I'd expect, test needs to be run again though with all boosting disabled. That 3-7 watt difference could be a lot less at base clocks, with no MCE or other stuff that changes defaults which is actually standard on gaming motherboards.

3-7 watts after 140 sounds like it's inside the confidence interval anyway.

[leadeater]

Just now, mr moose said:

3-7 watts after 140 sounds like it's inside the confidence interval anyway.

I had no idea sounds was such a good CPU benchmark 

[mr moose]

1 minute ago, leadeater said:

I had no idea sounds was such a good CPU benchmark 

sounds is one of the best, right next to "feels" about right and "looks" like the business.   And like all enthusiasts, near enough is good enough once you chuck em in a graph with no lines and very few markers.

[Stefan Payne]

3 hours ago, mr moose said:

Criticize manufacturers when they actually do something wrong, not because you don't understand TDP. 

AMD should change their TDP to 57W for Ryzen 2/2700x and 1800X so those are ~50% too high to be comparable with Intel.

And 50W for the Ryzen 7/1700 or 2400G.

 

Why?
Because:

https://www.tomshardware.com/reviews/amd-ryzen-7-2700x-review,5571-12.html

Lets use THAT to define the TDP of a CPU!

Use the Gaming loop. Because it looks like that is in line with the Intel TDP...

 

Package Power AMD is somewhat close (+/-20% tolerance), Intel is at 2,4 times TDP for the 8700 and 1,7 times the TDP for the 8700K.

AND THAT should have been mentioned somewhere!

In that review, the i7-8700 is only around 20W under a Threadripper 1950X...

And a Ryzen 7/2700X is comparable to an i5/7500, i3-8350K and i7-7700k in the Torture test...

[Droidbot]

4 minutes ago, Stefan Payne said:

AMD should change their TDP to 57W for Ryzen 2/2700x and 1800X so those are ~50% too high to be comparable with Intel.

And 50W for the Ryzen 7/1700 or 2400G.

 

Why?
Because:

https://www.tomshardware.com/reviews/amd-ryzen-7-2700x-review,5571-12.html

 

Lets use THAT to define the TDP of a CPU!

 

P95 stress: 105w.  

 

That's about in line with the TDP and the absolute maximum wattage the CPU will use. 

 

Also manufacturers define TDP differently. 

Since there's no regulatory body actually outlining a standard for CPU TDP measurements and ratings manufacturers are free to define TDP however they so choose. 

[Stefan Payne]

3 minutes ago, Droidbot said:

Also manufacturers define TDP differently. 

Since there's no regulatory body actually outlining a standard for CPU TDP measurements and ratings manufacturers are free to define TDP however they so choose. 

...and thus it is totally fine to critizize them when they are sugarcoating their TDP definition, wich Intel seems to do, wich leads to problems with notebooks and other stuff...

[mr moose]

9 minutes ago, Stefan Payne said:

Lets use THAT to define the TDP of a CPU!

 

No,  you cannot use anything other than the manufactures definition.  We've been over this countless times, you cannot use wikipedias definition, you certainly  cannot mix n' match AMD and INtel TDP's.

 

The laws of physics do not change, as Leadeater has pointed out, they all perform within their TDP spec, Honestly you must be the last person  still trying to argue that it is wrong even in the face of all the evidence.

[mr moose]

3 minutes ago, Stefan Payne said:

...and thus it is totally fine to critizize them when they are sugarcoating their TDP definition, wich Intel seems to do, wich leads to problems with notebooks and other stuff...

They aren't though.  Why do you keep saying they do? Seriously why?  It's o.k not to understand something, but it's not o.k to keep saying the same thing in hopes that it becomes true. It's not going to become true just because you keep saying it.   The problems with notebooks is not the TDP but the inefficient cooling solution chosen by the manufacturer. 

[leadeater]

10 minutes ago, mr moose said:

They aren't though.  Why do you keep saying they do? Seriously why?  It's o.k not to understand something, but it's not o.k to keep saying the same thing in hopes that it becomes true. It's not going to become true just because you keep saying it.   The problems with notebooks is not the TDP but the inefficient cooling solution chosen by the manufacturer. 

All Intel CPUs have configurable TDP anyway, mobile ones often have an extra configuration mode that the desktop processors don't, low power mode. If the laptop maker wants a thin and light they can reduce the short term and long term TDP limits of the processor using the bios.

 

On gaming motherboards when you change the power options that also configures the TDP of the CPU, every review ever that I have seen use both MCE and default power options that increase the boost TDPs above Intel default, we want that btw. Just have to use a better cooler.

 

Here is what my Asus Rampage IV Black Edition motherboard has configured the CPU TDP for boost states.

Taken with Intel XTU, I did not set them myself using this tool.

 

1000W sure as hell is not Intel default.

[mr moose]

2 minutes ago, leadeater said:

 Just have to use a better cooler.

 

And that is the crux of it, if you want to push a CPU past the intel TDP spec you have to use a better cooler.  

 

Some people think that using an inferior cooler or pushing the CPU harder than the cooler can keep up makes the TDP spec wrong.  Which is my point.

[Canada EH]

Just now, mr moose said:

inferior

Are you saying Rosewill is inferior

Bottom of the barrel in price cant be inferior LOL

[mr moose]

2 minutes ago, Canada EH said:

Are you saying Rosewill is inferior

Bottom of the barrel in price cant be inferior LOL

I think if you use $5 CPU cooler from Ebay and you don't get full boost on all cores running prime95 then it's because intel's TDP is wrong.

 

EDIT: even if you are using an AMD CPU.

[MMKing]

@Stefan Payne

TDP refers to Thermal Design Power, and refers to the minimum amount of heat the end user will have to dissipate from the CPU package in order to run the CPU at stock speeds.

 

https://ark.intel.com/products/126686/Intel-Core-i7-8700-Processor-12M-Cache-up-to-4-60-GHz-

''Thermal Design Power (TDP) represents the average power, in watts, the processor dissipates when operating at Base Frequency with all cores active under an Intel-defined, high-complexity workload. Refer to Datasheet for thermal solution requirements.''

 

https://www.intel.com/content/dam/www/public/us/en/documents/datasheets/8th-gen-core-family-datasheet-vol-1.pdf

 

I agree, it's a bit bullshit that Intel writes 65W TDP on the box, as well as writing max turbo 4.6GHZ. When in reality, 65W TDP refers to minimum heat dissipation to reliably run the CPU at base 3.2GHZ frequency. Max turbo refers to single core turbo, provided the end user dissipates enough heat away from the CPU and that enough power is available.

 

It's tucked away, and even to experiences users it's convoluted to find exactly what you're looking for. But you can't really stretch it further than misleading. You can easily configure your CPU to not boost above it's stated base clock, and it should hover around 65W power usage. But keep in mind that TDP and power consumption is not the same, they are related, hence the confusion, but they are not the same.

[mr moose]

42 minutes ago, MMKing said:

@Stefan Payne

TDP refers to Thermal Design Power, and refers to the minimum amount of heat the end user will have to dissipate from the CPU package in order to run the CPU at stock speeds.

 

https://ark.intel.com/products/126686/Intel-Core-i7-8700-Processor-12M-Cache-up-to-4-60-GHz-

''Thermal Design Power (TDP) represents the average power, in watts, the processor dissipates when operating at Base Frequency with all cores active under an Intel-defined, high-complexity workload. Refer to Datasheet for thermal solution requirements.''

 

https://www.intel.com/content/dam/www/public/us/en/documents/datasheets/8th-gen-core-family-datasheet-vol-1.pdf

 

I agree, it's a bit bullshit that Intel writes 65W TDP on the box, as well as writing max turbo 4.6GHZ. When in reality, 65W TDP refers to minimum heat dissipation to reliably run the CPU at base 3.2GHZ frequency. Max turbo refers to single core turbo, provided the end user dissipates enough heat away from the CPU and that enough power is available.

 

It's tucked away, and even to experiences users it's convoluted to find exactly what you're looking for. But you can't really stretch it further than misleading. You can easily configure your CPU to not boost above it's stated base clock, and it should hover around 65W power usage. But keep in mind that TDP and power consumption is not the same, they are related, hence the confusion, but they are not the same.

You know I've looked on the box,  I can't find an image of one that does.  In fact the spec sheet (in arc or bellow the product) is the only place I have seen the TDP.

[Stefan Payne]

20 minutes ago, MMKing said:

@Stefan Payne

TDP refers to Thermal Design Power, and refers to the minimum amount of heat the end user will have to dissipate from the CPU package in order to run the CPU at stock speeds.

...wich makes it totally useless and sugarcoated, wich is what I'm talking about the whole time.

And Intel should be criticised for such a useless definition without much real world usage...

 

20 minutes ago, MMKing said:

https://ark.intel.com/products/126686/Intel-Core-i7-8700-Processor-12M-Cache-up-to-4-60-GHz-

''Thermal Design Power (TDP) represents the average power, in watts, the processor dissipates when operating at Base Frequency with all cores active under an Intel-defined, high-complexity workload. Refer to Datasheet for thermal solution requirements.''

That is the new definition that Intel changed - and nobody noticed.

I've linked an old document citing the old Intel TDP specification, from the "Measuring Processor Power Document", though its a bit older (April 2011):

"Intel defines TDP as follows:  The upper point of the thermal profile consists of the Thermal Design Power (TDP) and the associated Tcase value.  Thermal Design Power (TDP) should be used for processor thermal solution design targets. TDP is not the maximum power that the processor can dissipate. TDP is measured at maximum TCASE.1.  The thermal profile must be adhered to to ensure Intel’s reliability requirements are met.  Note:  Different processors SKU’s have different TDP’s.  At the time of this writing, Intel® Xeon® processors for 2 socket servers (5600 series) are available with a TDP specification from 40W up to 130W depending on the particular SKU1. "

 

Somewhere between then and now they changed the TDP definition from this upper limit to something towards the Baseclock - without the introduction of another power measurement for maximum boost frequency/turbo...

 

20 minutes ago, MMKing said:

I agree, it's a bit bullshit that Intel writes 65W TDP on the box, as well as writing max turbo 4.6GHZ.

...wich is the point I'm trying to make...

At least one that agrees with me here.

And no, I'm not misunderstanding it, Intel is sugarcoating it and its a mostly useless value with the changes they made sometime in the last couple of years. Because most of the Time, the Intel TDP was the upper end of power to be dissipated, not the lower end like today.

 

They did that because they had to print 125 or maybe even 150W TDP on the box without the change to the specification, if they would have stuck with the old definition they used for more than 10 years (since the Original Pentium! or even before that).

For example:

https://ark.intel.com/products/49949/Intel-Pentium-Pro-Processor-166-MHz-512K-Cache-66-MHz-FSB

https://ark.intel.com/products/49935/Intel-Pentium-II-Processor-233-MHz-512K-Cache-66-MHz-FSB

 

20 minutes ago, MMKing said:

When in reality, 65W TDP refers to minimum heat dissipation to reliably run the CPU at base 3.2GHZ frequency. Max turbo refers to single core turbo, provided the end user dissipates enough heat away from the CPU and that enough power is available.

...wich nobody really does and is also the lower limit of the CPU under load...

So they never should have called this number "TDP" but something else (SDP for example, mentioned before)...

 

20 minutes ago, MMKing said:

It's tucked away, and even to experiences users it's convoluted to find exactly what you're looking for. But you can't really stretch it further than misleading. You can easily configure your CPU to not boost above it's stated base clock, and it should hover around 65W power usage. But keep in mind that TDP and power consumption is not the same, they are related, hence the confusion, but they are not the same.

Thing is that the Turbo boost is part of the default configuration and should be included into the TDP and not excluded like they did here...