Question Anandtech.com article on 13900k and 7950x power scaling

[adamge]

A Lighter Touch: Exploring CPU Power Scaling On Core i9-13900K and Ryzen 9 7950X

www.anandtech.com

It's an interesting article, but IMO severely crippled by the reliance on "configured" power limit rather than measuring power usage. It's even more frustrating when one of the datapoints is measured power usage, so the author has access to this data.

In this scenario where configured power limits are basically hand waving guidelines, measured power usage (specifically, average power usage by the CPU during the benchmark run) is the only useful, meaningful, and concrete way to analyze the performance results relative to the power usage.

 

[Hitman928]

Anyways, back to anandtechs data

7950x at 105w = 35975 score @ 145.6w =
247.08pts per watt

13900k at 105w = 29372 scors @ 118.6w =
247.65pts per watt



Clearly saving hundreds of watts x 3 daily. No questions about it.

But in your case, the 7950x would complete its tasks 22.5% faster which means either you can be 22.5% more productive in a day, or the system gets to go to sleep sooner while the Intel system is still computing away so the total energy used will be significantly in favor of the AMD system.

Realistically, equalizing for efficiency is kinda silly because there is no inherit limit or motivation to do so. You can equalize for power use if you are power/cooling limited or, if there is a specific performance level you need to reach, you can equalize for performance and then see which CPU uses the least amount of power to reach that performance. If you equalize for power, then AMD comes out ~15% ahead in efficiency (this is for their current desktop chips in MT situations, I expect laptop and server comparisons to go much more in AMD's favor). If you equalize for performance, AMD comes out significantly more ahead due to the fact that both CPUs are pushed well beyond their efficiency points to the point where AMD at 90W can match Intel at 140W. Most people will want to equalize for power so it makes more sense to use that method. As such, AMD has an efficiency lead, but it's not all that significant for the desktop SKUs.

 

[Hitman928]

Oh come on now, TPUP uses CPU only power draw and it shows a 52% efficiency delta. And I wasn't even the one that brought up the review, abwx did, now that it proves how inefficient amd is at doing simple tasks, you discarded the review and going for the next one...

But funny thing, even with the results you just posted , there is still a 30-40% efficiency delta. Compare the wattage with the score and see for yourself.

You can tilt at windmills for as long as you want, no one is ever going to care.

 

[Mopetar]

Weird that a new poster shows up out of seemingly nowhere and has time to dredge through old threads and posts to be familiar with things that people have supposedly said months ago, but almost a week later can't be bothered to actually link to any of these posts.

 

[Mopetar]

What? That's totally normal and not at all indicative of some kind of viral marketing strategy.

I don't know if I'd go that far. I don't know if the quality of the work is worthy of pay and I don't think Intel would keep a position like this around while cutting technical staff.

I suppose there's always the horrifying possibility that this is being done for free, but I'd at least like to think that even if the reward isn't monetary it's being done for dopamine deposits.

And on along that line of reasoning, I'm still split between whether to tell them to go back into the drawer or back underneath the bridge. Perhaps a drawer tucked away under a bridge.

 

[Hitman928]

Well, although you are not wrong, the question still remains why does that game run better on X cpu while only using 2 cores instead of Y cpu while also only using 2 cores. And the most likely answer is that X and Y cores are not equal, at least not in gaming workloads

Why does that game run better on X cpu while only using 2 cores instead of Y cpu while also only using 2 cores. And the most likely answer is that X and Y cores are not equal, at least not in this particular game but in other games, Y will run better than X.

FTFY.

 

[Thunder 57]

Yeah, you posted a peak power draw graph and then concluded that you are saving 100w a day. Scientific stuff

Your continued use of watts per (unit of time) is just wrong and hard to read? A watt is already a unit of time, a joule/s.

 

[Markfw]

Ah, but you did, you replied to carefax with some screenshots from f@h claiming that zen 2 beats pcores.

I mean it's fine, whatever, my opinion is irrelevant, if you don't think you are incredibly biased then so be it.

Link my post. I will either explain it or say you are right. Otherwise it didn't happen.

 

[Hitman928]

Anyways, since you seem to trust tpup, we do agree that in light workloads the 13900k is VASTLY more efficient than the 7950x,right? Its also plenty faster in games as well. Am i right?

Nope and nope.

 

[Hitman928]

Then why bother posting a review you dont agree with? That seems weird

Single threaded does not equal lightly threaded in 2023. Additionally, if single threaded or very lightly threaded performance and efficiency are key driving factors for someone, they wouldn't be looking at any of these CPUs to begin with.

 

[Hitman928]

So you agree that the 7950x gulps way more power than the 13900k in st workloads, but you dont think its an important metric. Don't want to misinterpret what you said, so please feel free to correct me

Way more is an exaggeration. Yes, there is a power penalty for AMD having a non-monolithic design that mostly shows up at idle or when using only 1-2 cores. This has been known for quite a long time, you are not bringing new information to the table. We are talking like, 10-11 watts of delta between the two. That's an LED light bulb worth of power, it's not a significant amount overall and again, we are talking about CPUs with 32 threads. This is like comparing the fuel efficiency of two 18 wheel trucks while they are idling or completely unloaded. No one cares who wins. The idle/light core usage is far more important in laptops and SFF systems, where AMD has monolithic designs to fight that battle.

 

[Carfax83]

In EVERY scenario I saw, the 13900k was consuming a LOT more power(like as much as 100 watt or more than the 7950x), except this one test. Way to cherry pick.

Wow, every scenario? I don't know why you bother even commenting on it since it's clear you didn't watch it at all. I mean, it's quite easy to disprove what you are saying. The gaming sections are very close in terms of power draw for instance.

Also, browsing the internet is by far the most common form of computing that the vast majority of people do, so I don't consider that cherry picking. It emphasizes exactly what others and I have been saying about RPL having an advantage in light to moderate loads.

 

[Timmah!]

This may surprise you, but because of ray tracing, being CPU limited is much more prevalent now than it has ever been. That HWUB graph has already been heavily critiqued because they never tested any CPU bound scenarios, which is why the graph looks as bunched up as it does.

If you're GPU limited, then the CPU hardly matters at all so why they even bothered with all the testing if it's meaningless in the end? To do proper CPU testing, reviewers need to find the most CPU limited areas in the games and also enable settings like ray tracing which increase CPU dependency. Compare the HWUB graph to something like what PCGH.de does with their testing.

This is a worst case scenario for Zen 4, because The Witcher 3 Next Gen has terrible CPU optimization which makes it hammer 2 cores (due to non native DX12), and when you factor in the BVH calculations from ray tracing and the increased crowd density and draw distances from the next gen enhancements, then it's literally the perfect storm for CPU testing.

Here the 13900K is a good 40% faster than the 7950x at 720p to maximize CPU dependency. Currently, no CPU can maintain 60 FPS in this title at all times and even at 4K you can be CPU bottlenecked.

If you know the game is terribly optimized and uses only 2 cores, this really should not be taken into consideration as a proof some CPU is inadequate. You could take any modern CPU to crawl with Sins of a Solar Empire on big maps with 10 players, only because the game is single threaded. That should hardly make any impact on your decision what to buy nowadays.
If its a modern game like Cyberpunk, that uses 8 or more cores, then fair enough. But Witcher? Meh. I dont know why they even bothered to add RT to that game, i mean it was pretty game back in the day, but nowadays it looks dated, and adding more realistic shadows or GI is not really going to change that.

 

[Hulk]

As I wrote in the comments section for the article I'm grateful for any new content on Anandtech. But with that being said...

1. More power usage data points needed to draw conclusions. 125W and then a jump to 230W. Too large. There is a lot going on in that gap that needs to be investigated.
2. Benchmarks aren't the be all end all application for power usage. Outside of benchmarks not a lot of applications use all of the cores or even use all of the resources in a core. So while 5.5GHz all cores loaded under Cinebench might result in a ridiculous power usage for the 13900K, in a real world application such as an NLE those frequencies are reached an maintained with a package power of 160W in many cases. This can be seen in the gaming benches, which are real world and things even out quite a bit.
3. Bursty loads aren't taken into account either and it should be noted that cranking power up to 250W for 3 seconds every now and then while working so you aren't waiting on the computer might be a worthwhile tradeoff in terms of keeping a creative flow going.

 

[Mopetar]

As I wrote in the comments section for the article I'm grateful for any new content on Anandtech. But with that being said...

1. More power usage data points needed to draw conclusions. 125W and then a jump to 230W. Too large. There is a lot going on in that gap that needs to be investigated.

It's not ideal to have a gap that large, but from the data that's available it's pretty easy to conclude that AMD pretty much hits a wall at 105W and that Intel can at least scale past that point, even if it's not particularly efficient.

At least there are results in the other direction and performance is measured below 65W. Most people aren't likely to tune to those settings, so it's more academic than anything, but I still found it interesting.

 

[nicalandia]

Whoever argues in favor of the 13900K Being more efficient than the 7950X is delusional. No amount of tweaking will make the 13900K be more efficient than a similar tweaked 7950X

 

[lopri]

I find it peculiar what he describes as "scaling better" is the one that does not respond to increased voltage. His test shows 13900K scales well with power while 7950X does not. What that tells me is that AMD should have released 7950X as a 105W part, not 170W (230W actual) part.

And his 13900K running y-cruncher is only at 86C.. hmm I remember my 13700K getting close to 110C. Well, with an air cooler, so there's that. He's using an AIO 360mm.

I do not think the testing methodology is clearly explained in the article. The author says he is reading AIDA64 which in my experience corresponds to HWiNFO64's reading. It seems to me this chart shows how AMD's Eco Mode operates. (e.g. Eco Mode 65W in BIOS -> 90W PPT in Windows) I thought setting 90W PPT in BIOS resulted in 90W in Windows as well according to HWiNFO64. I might be misremembering it.

 

[Hulk]

@Hulk While you may have a point or two, I would have to say that I agree with the article overall. The 7950x is more efficient from almost any angle. I run my 3 at 142 watt, since that is Co -25, and I set temps to 85. It absolutely kills my 5950x's in anything, and I have used the avx-512 at times.

That's a fair comment. I just wanted more power points tested and more real world applications.
Most of the games tested showed the same result so instead of beating a dead horse why not test some other actual applications?

 

[TESKATLIPOKA]

I see a big problem with R9 7950X or rather R9 7945HX's temperature.

In laptops you won't have such a powerful cooling solution.
So you are most likely limited to 65W(88W PPT).
But Intel can be set to even 125W.

Performance will be in favor of Intel in CB R23.

 

[scannall]

I see a big problem with R9 7950X or rather R9 7945HX's temperature.

In laptops you won't have such a powerful cooling solution.
So you are most likely limited to 65W(88W PPT).
But Intel can be set to even 125W.

Performance will be in favor of Intel in CB R23.

Not so sure about that. I have a Ryzen 5800, and a Ryzen 6900HX. The 5800 idles at 18-20 watts. The 6900HX idles at 3ish. And performs better than the 5800. https://browser.geekbench.com/v5/cpu/compare/17339395?baseline=19505769 Both systems are at stock settings. The 6900HX tops out at 60 watts, and the 5800 at 78.

 

[TESKATLIPOKA]

I did not see any laptop benchmarks. I don't know how you can say what will happen. Not only is cooling different, the chips are different. The only article's I saw that had real chips, heavily favored AMD's 7000 series.

I am talking about Dragon Range. That's Raphael or not?
In most laptops is an AIR cooling solution and It even has to cool the GPU. This simply can't compare to EKWB EK-AIO Elite 360 D-RGB 360mm which only cools the CPU.
These charts are telling us a lot.

 

[TESKATLIPOKA]

Not so sure about that. I have a Ryzen 5800, and a Ryzen 6900HX. The 5800 idles at 18-20 watts. The 6900HX idles at 3ish. And performs better than the 5800. https://browser.geekbench.com/v5/cpu/compare/17339395?baseline=19505769 Both systems are at stock settings. The 6900HX tops out at 60 watts, and the 5800 at 78.

Those are only 8C16T Zen3+ at 6nm process, heat is not that concentrated.
What I am talking about is Zen4 5nm 16C32T.
You have that chart about temperatures and that CPU should be the same as R9 7945HX with lower TDP.

 

[scannall]

Those are only 8C16T Zen3+ at 6nm process, heat is not that concentrated.
What I am talking about is Zen4 5nm 16C32T.
You have that chart about temperatures and that CPU should be the same as R9 7945HX with lower TDP.

What I am saying is you can't infer much from that. Until the mobile parts are in the wild and actually benchmarked then it's speculation.

There is a lot of cheer leading for peoples favorite team happening around here lately. How about be happy that there are great things happening. Both Intel and AMD have compelling offerings, so buy according to your needs, and it will be fine.

 

[lopri]

Trying to infer laptop thermals and power consumption from this is unwise in my view.

Going back to the y-cruncher peak power, I hope there is a clarification with regard to testing methodology. That chart can invalidate the rest of the charts.

Edit: It is peak power, so I guess it really doesn't matter either way.

 

[moinmoin]

I am talking about Dragon Range. That's Raphael or not?

That's the same dies, but very likely not the same behavior set in the firmware. E.g. STAPM (Skin Temperature Aware Power Management) is very likely a major limiting factor in Dragon Range that wasn't in Raphael.