Intel Announces 13th Gen "Raptor Lake" - Linux Benchmarks To Come

[coder]

You lack some knowdlege.

We'll see. Really, don't we all?

On intel most of chipset is not literally chipset, it is inside CPU. Majority of PCI-E lanes, SATA, USB ports are wired directly to CPU, not chipset.

Well, if that's what you mean, then don't call it a "chipset". More to the point, it does fall in the definition of what Intel means by "package power". If they didn't mean power to the entire CPU package, they wouldn't call it "package power" - they'd call it "cores power" or some such thing!

What you're describing used to be known as "North Bridge", before the CPUs absorbed the memory controller and PCIe lanes. Lower-bandwidth peripherals would be connected via the "Southbridge", which (in desktop platforms) is still a separate physical package on the motherboard. In mobile (BGA) CPUs, you'll see it as a separate die inside the CPU package.

So, I thought you meant the Southbridge, when you said "chipset".

PL1/PL2 applies only to processor unit itself, I/O and iGPU is outside of it.

I think your quote doesn't support that claim, nor did you cite a source.

One piece of evidence to the contrary are the KS-series CPUs. Their iGPU is disabled, which allows the CPU to have a tiny bit more power budget. Hence, they tend to be pretty popular with overclockers.

people forget how most motherboards of AMD X470 had literal fan on top of chipset and that chipset was consuming often up to 15W of power while Amd CPUs still do a lot of I/O by itself.

I remember that being true of the X570, but maybe you're right about the X470, also. One reason they used a lot of power is that they were made on a 14 nm GF process node and had to support PCIe 4.0.

Intel maximum spiking 271W instead of 241W (and most of time being 250-255W for very big stress test) seems reasonable for how much stuff is on Intel CPU.

If the stress test wasn't hitting any of those peripherals or the iGPU, then they're irrelevant. Idling them uses just a few Watts, at most.

[piotrj3]

Then Intel's product literature for their desktop CPUs needs to specify the range, like they do for their mobile parts. Anything else would be misleading, especially if common practice is to set them higher.

Gen 13 will be another chance for Intel. Let's hope they don't embarrass themselves, but I wouldn't count on it.


That's not true. Package power refers to the CPU package, which is what they specify and regulate. I think you're confusing it with "system power"?

You lack some knowdlege.

On intel most of chipset is not literally chipset, it is inside CPU. Majority of PCI-E lanes, SATA, USB ports are wired directly to CPU, not chipset. Not to mention Intel CPUs commonly also have iGPU which is also part of CPU package. PL1/PL2 applies only to processor unit itself, I/O and iGPU is outside of it.

Note: The System on Chip processor integrates multiple compute cores and I/O on a
single package. Platform support for specific usage experiences may require additional
concurrency power to be considered when designing the power delivery and thermal
sustained system capability.

This is issue with measurements from those tests. For example people forget how most motherboards of AMD X470 had literal fan on top of chipset and that chipset was consuming often up to 15W of power while Amd CPUs still do a lot of I/O by itself.

But again that is OK. Intel maximum spiking 271W instead of 241W (and most of time being 250-255W for very big stress test) seems reasonable for how much stuff is on Intel CPU. AMD 5950X going 120W with declared TDP of 105W is also ok. Going 80W over on average long term without OC and without messed up setup from motherboard is NOT.

And Intel's product literature exactly specify that. There are 2 public volumes specifically about information like that and GamersNexus was pointing it out in the past that Intel's has own recommendations that motherboards do not have to comply with and very often don't. A lot of motherboards loved:

- pushing BCLK 1-5% higher (and 5% in some edge cases could make stability issues),
- enable multi-core-enhancment aka remove Intel's turbo limits for multi-core workload (Intel's recommendations (and default) were that all-core was boosting less then single core and that had big efficiency diffrence, but motherboards often allowed all cores at once to boost to max speed what eaten efficiency in multi-core workloads alive),
- remove turbo limit times by default - doesn't matter anymore with 12th gen Alder lake K CPUs, but did matter a ton in 11th gen and older,

[coder]

You know PL2 limits (and stuff like "Multi core enhancment") are set by motherboard manufacturers?

Then Intel's product literature for their desktop CPUs needs to specify the range, like they do for their mobile parts. Anything else would be misleading, especially if common practice is to set them higher.

Gen 13 will be another chance for Intel. Let's hope they don't embarrass themselves, but I wouldn't count on it.

And again Package power is power of CPU, GPU, chipset related stuff in chip and I/O. Intel power limits are only towards CPU, package power is all of those together.

That's not true. Package power refers to the CPU package, which is what they specify and regulate. I think you're confusing it with "system power"?

[piotrj3]

Looks to me like > 150 seconds. That's more than a blip. A lot more.


I've read that it was motivated by compatibility with AM4 coolers. If true, definitely not a worthwhile tradeoff!


But they don't have to. Nothing says you need to hit 250 W. AMD didn't guarantee any performance that requires you to dissipate that much power. It's merely a bonus that you can unlock more performance with extreme cooling. That's how I see it. It's a much better situation than AMD building in some completely artificial threshold, IMO.


Direct-die cooling commonly yields cooler temps by more than 10 C. And the hotter the baseline temps, the greater the gains.

You know PL2 limits (and stuff like "Multi core enhancment") are set by motherboard manufacturers? This topic was visited by Gamersnexus on previous Intel CPUs (11th and earlier) and they proven when motherboard is set inline with Intel guidelines you won't see 270W. The only reproducable significant outlier on Intel's recommended settings was AVX512 workloads but they are gone on 12th gen.

And again Package power is power of CPU, GPU, chipset related stuff in chip and I/O. Intel power limits are only towards CPU, package power is all of those together.

[Anux]

No, Intel advertises 125W as "Base frequency" power.

Ah, they must have renamed that at some point, most shops still advertise 125 W TDP.

Thermal Considerations
The Processor Base Power (a.k.a TDP) is the assured sustained power that should be
used for the design of the processor thermal solution, Design to a higher thermal
capability will get more Turbo residency. Processor Base Power is the time-averaged
power dissipation that the processor is validated to not exceed during manufacturing
while executing an Intel-specified high complexity workload at Base Frequency and at
the maximum junction temperature as specified in the Datasheet for the SKU segment
and configuration.
Note: The System on Chip processor integrates multiple compute cores and I/O on a
single package. Platform support for specific usage experiences may require additional
concurrency power to be considered when designing the power delivery and thermal
sustained system capability.

But that again says Processor Base Power = TDP = 125 W. That makes the actual > 200 W much worse.

Don't get me wrong, I don't care what Intel or AMD claim and only look at actual tests for purchasing decisions.

But they don't have to. Nothing says you need to hit 250 W. AMD didn't guarantee any performance that requires you to dissipate that much power. It's merely a bonus that you can unlock more performance with extreme cooling. That's how I see it. It's a much better situation than AMD building in some completely artificial threshold, IMO.

We already had "the cooling solution is the limit" in notebooks much longer. I wouldn't say that I find it good but they give you the possibility to limit it down to 65 W TDP or 88 "real" watts. Possibly with a low end cooler for maybe 35 W it would automatically scale even lower.

[coder]

Intel actually almost always abides those limits. In Blender/Cinebench/7zip etc. you won't see 12900k consuming for extended periods of time more on average then 241W (+- few watts). Only extreme stress testers like prime95 kind of pushes limit more but i wouldn't call it representative.

Looks to me like > 150 seconds. That's more than a blip. A lot more.

AMD for some outrageous reason decided to make heat spreader extra tall,

I've read that it was motivated by compatibility with AM4 coolers. If true, definitely not a worthwhile tradeoff!

most people can't even make it to 250W without some atrocius setup.

But they don't have to. Nothing says you need to hit 250 W. AMD didn't guarantee any performance that requires you to dissipate that much power. It's merely a bonus that you can unlock more performance with extreme cooling. That's how I see it. It's a much better situation than AMD building in some completely artificial threshold, IMO.

der8auer who made direct die cooling (remove heat spreader and put cooler directly on top of chip) decreased tempretures by 22C on 7950X.

Direct-die cooling commonly yields cooler temps by more than 10 C. And the hotter the baseline temps, the greater the gains.

[coder]

I'm not sure I understand your point. Well, at least in theory it's possible for TDP be lower than PPT given only a fraction of PPT is required to sustain base frequency. For example, TDP is 125W, PPT is 170, you put on 125W solution, your CPU starts to draw more than 125W until it reaches a thermal limit during which algorithm will drop frequency and voltage until power draw is reduced at say ~130W which would be sufficient for base frequency.

Okay, if the base frequencies are computed according to TDP, but the max sustained power draw is PPT, then I can at least understand that.

[piotrj3]

But 12900k advertises 125 W TDP.

https://ark.intel.com/content/www/us...-5-20-ghz.html
No, Intel advertises 125W as "Base frequency" power. It also mentions maximum power draw (241W) for boost loads over periods longer then 1 second (quoting)

Maximum Turbo Power
The maximum sustained (>1s) power dissipation of the processor as limited by current and/or temperature controls. Instantaneous power may exceed Maximum Turbo Power for short durations (<=10ms). Note: Maximum Turbo Power is configurable by system vendor and can be system specific.

Also Intel made it public in 12900k annoucment that for K chips PL2 is no longer time limited, and PL1/PL2 time limits are existing in BIOSes as well as official intel extreme tuning utility (and if you visit data sheet link that is more documented above, you can find them how they work too)

From volume 1

Thermal Considerations
The Processor Base Power (a.k.a TDP) is the assured sustained power that should be
used for the design of the processor thermal solution, Design to a higher thermal
capability will get more Turbo residency. Processor Base Power is the time-averaged
power dissipation that the processor is validated to not exceed during manufacturing
while executing an Intel-specified high complexity workload at Base Frequency and at
the maximum junction temperature as specified in the Datasheet for the SKU segment
and configuration.
Note: The System on Chip processor integrates multiple compute cores and I/O on a
single package. Platform support for specific usage experiences may require additional
concurrency power to be considered when designing the power delivery and thermal
sustained system capability.

So TDP according to Intel is power draw at BASE frequency, without including SoC additional power draw (considering majority of chipset is inside CPU and it is responsible for a lot of IO it makes sense to exclude it). So few watts above power limit (if it happens) comes from IO/SoC related stuff. This happened both for Intel and AMD (5950x by comparison has commonly 105W TDP but likes to load up to 120W, but it is acceptable considering you can't quite count in those watts spent on IO/motherboard etc. related tasks).

But because TDP is useless metric in boosting era of processors intel gave you PL1 and PL2 (and PL3/4 but they are disabled by default) to better represent those loads and in chapter just after one i quoted they are very well described how in general they work. Somewhere there is also information about 12th gen K chips being not limited to PL2.

[Anux]

Intel actually almost always abides those limits. In Blender/Cinebench/7zip etc. you won't see 12900k consuming for extended periods of time more on average then 241W (+- few watts).

But 12900k advertises 125 W TDP.

[piotrj3]

I'm not sure I understand your point. Well, at least in theory it's possible for TDP be lower than PPT given only a fraction of PPT is required to sustain base frequency. For example, TDP is 125W, PPT is 170, you put on 125W solution, your CPU starts to draw more than 125W until it reaches a thermal limit during which algorithm will drop frequency and voltage until power draw is reduced at say ~130W which would be sufficient for base frequency. I'm not saying this is the case for ZEN4, i'm saying this is possible in theory. Though it feels to me that 105W for 7700X and 170W for 7900X/7950X should be really enough to sustain base, power scaling shows that 170W gives around 80-85% (if i recall correctly nuilzoid video) performance for 7950X so it seems you can actually even get some boost there. The price is of course CPU is going to be at very high temp.

TDP is supposed to be thermal design for extended workload, not transient. Of course i don't mind if CPU has temponary higher transient load because thermal mass of radiatiors/coolers will be enough to handle short power spike.

However if extended workload processor consumes more power by 30-40% then TDP indicates, it means AMD's claim is entirly fictional.

Intel actually almost always abides those limits. In Blender/Cinebench/7zip etc. you won't see 12900k consuming for extended periods of time more on average then 241W (+- few watts). Only extreme stress testers like prime95 kind of pushes limit more but i wouldn't call it representative. Meanwhile in Blender AMD made 170W chip push 250W on average in extended period of time. What is more, it is way harder to cool 7950X (comparing to Intel chips) as AMD for some outrageous reason decided to make heat spreader extra tall, and smaller meaning most people can't even make it to 250W without some atrocius setup. der8auer who made direct die cooling (remove heat spreader and put cooler directly on top of chip) decreased tempretures by 22C on 7950X.


arQon Power draw from EPS rail (on average) is kind of equalivent to TDP as essentially all electricity you use is thermal in the end. Of course definition varies but TDP was mostly information made for sake of cooler producers in the past. Since then agressive boosting was introduced Intel dropped TDP and introduced PL1 and PL2 and simply set PL2 either unlimited, or limited in time. And since then with small exception of strong AVX loads/stress testers Intel actually abides it. And AMD with Zen 3 and before was also abiding it up to few watts diffrence.