Hacker News Reader

Comment by kaelinl | Hacker News Reader

Since it's in the title, note that the TDP on AMD cores (Intel and the rest too, but especially AMD) is misleading. They engineer the number to fit their marketing; there's no expectation that 65 watts of power is put into nor dissipated out of this chip.

It's probably reasonable as a relative comparison point to other recent AMD parts, but that's about it.

See e.g.: https://www.gamersnexus.net/guides/3525-amd-ryzen-tdp-explai...

View on HN

- "Finally, it's also worth reminding everyone that, as has been the case for several processor generations now, TDP is not the maximum power consumption value for either AMD or Intel – nor is the definition of TDP even identical between the two. For AMD, the metric to watch is Package Power Tracking (PPT), which is the power level that the CPU socket allows the CPU to draw in terms of power. On the AM5 platform, AMD defines PPT as being 1.35x the TDP – so in the case of today's 65 W Ryzen processors, that means an 88 W socket power limit."

(This is on Anandtech's first page).

In other words, they know that the "65 Watts" in their title is meaningless but included it anyway. Saying "88 Watts" would at least mean that the number referred to an amount of power. But even then, it's the actual draw as measured on the traces that really matters, rather than the marketing material.

Not making any judgements re: whether this chip or others are power-efficient, just making it clear that the power number in the title and body was chosen by a marketing team and has no other meaning.

> especially AMD

https://www.anandtech.com/show/17601/intel-core-i9-13900k-an...

Intel Core i9 13900K has a TDP of 125/253 (base/turbo) but draws 334W (peak). 81W or 32% over max turbo TDP, 209W or 167% over base.

AMD Ryzen 9 5950X has a TDP of 105W but draws 142W (peak). 37W or 35% over.

AMD Ryzen 9 7950X has a TDP of 170W but draws 221W (peak). 51W or 30% over.

AMD Ryzen 9 7900 has a TDP of 65W but draws 90W (peak). 25W or 38% over.

So slightly higher % wise on some chips, but not outside the ballpark by any means. I don't think Intel is doing us any favors!

The "K" in 13900K stands for "draws as much current as physically possible." The ones without the K are a little more mainstream, and were just recently released.

They were released, but they weren't publicized, nor were they seeded to reviewers. In other words, unlike the non-X Zen 4's, the non-K Intel Raptor Lakes likely perform a lot worse than the Intel K's.

I'm not sure that's really going to be the case. E.g. the 13900 a score in CineBench R23 Multi close to the best AMD CPU in this article, just a few % off the 13900K.

In highly multithreaded workloads the 13900 probably does fine (24 cores are 24 cores). It's the lightly threaded and gaming workloads where it's expected to suffer the most. Especially since the L2 cache is also cut on some of the non-K parts, which might be a first. The 13600 drops from 20mb on the k to 11.5mb on the non-K, that's a big drop.

It's the heavily threaded workloads that will suffer the most. In a single threaded workload that single core can get as much power as it wants in both K and non-K. In a multi-threaded workload the cores are power limited, and K has a lot more power than non-K so the effect of power throttling is felt more on non-K.

Heavily threaded will suffer the most between the K & non-K, but not necessarily enough to drop below the "only" 16-core competition which was the comparison point here, not the K version. Especially since it's primarily the 8 P-cores that are really going to be impacted in such a scenario.

Hence why I specified "lightly threaded" not "single threaded" workloads. There's a whole huge market of workloads that sit at around 4-8 threads. They're called "video games" and it's what Intel has been bragging about being the best at for a while. The new crop of non-K's look prime to suffer the most in this workload relative to AMD's non-X competition.

On the contrary, the other article at the top of Anandtech right now shows that the K-series CPUs don't benefit from higher power levels in games. As far as games are concerned they might as well be 95W CPUs.

Anandtech sadly only tested 2 games in a fairly narrow set of parameters. If you look at 13900K reviews, such as this one by techpowerup: https://www.techpowerup.com/review/intel-core-i9-13900k/22.h... you'll see gaming loads where the 13900K uses well past 120w. Civ 6, for example, sits at nearly 190w.

But also again remember the 13600 non-K has much less L2 than the 13600k. It's not just a power/turbo restriction.

But we'll see for sure when the 13th gen non-K reviews happen. That Intel didn't sample these to reviewers isn't a great sign, though.

"we were pretty impressed with the overall performance on offer, despite being limited to a 65 TDP limit. Although power consumption on the CPU package was around 90 W at full load, this is still considerably lower than other processors the trio of 65 W Ryzen 7000 chips was pitted against."

This is from the Conclusion page. Still pretty impressive.

> note that the TDP on AMD cores (Intel and the rest too, but especially AMD) is misleading

Reviewers should put this in BIG ALL-CAPITALS BOLD ITALICS at the top of their reviews, all the time. The TDP rating of any given CPU in most cases is completely unrelated to the actual performance of said CPU. This is most egregious in notebooks, where it's actually the PL1/PL2 limits that matter. In my notebook's case: it contains a Xeon W-11955M (TDP supposedly 45 W), but while gaming or rendering, the CPU regularly draws nearly twice that. In fact, in notebooks, the thermal setup frequently affects the performance of the CPU and GPU by more than ±20-30% compared to the baseline.

CPUs (and now GPUs, too) also include a variety of clock-scaling mechanisms like P-states, C-states, SpeedStep, Speed Shift (hardware P-states), etc that allow the CPU to sip a mere 0.2–0.5 W while doing nothing, and ramp up to 150+ W when demanded.

For accurate comparisons, reviewers should lock CPUs to a maximum power draw, and test CPUs at that fixed power draw.

> For accurate comparisons, reviewers should lock CPUs to a maximum power draw, and test CPUs at that fixed power draw.

This is done by some reviewers, but it also only accurately tests efficiency rather than performance (or potential for performance given suitable conditions).

This is also why you mostly see laptop performance reviews instead of laptop CPU reviews.

Linus Tech Tips published the actual power consumption for both X and non-X in their video here: https://youtu.be/CTiRNnSg0jA

Power consumption depends a lot on the specific workload so picking a single number is mostly an exercise in marketing. It seems close enough in practice.

What is the purpose of even providing a figure if it's not accurate?

(Maybe I'm wrong, I thought the TDP figure is the maximum sustained power that has to be dissipated by the cooling setup, the CPU can output more heat briefly but over a long period of time it'll average 65 W assuming it's fully utilized?)

It's made up but does provide rough tiers within a product generation.

Also, modern CPUs and GPUs pretty effectively throttle to the capacity of their cooling solution. I would not necessarily recommend it, but you could probably get by without any issue slapping a solution that reliably dissipates 65W of heat on any Ryzen, and have a stable computer. It would just happen to run up to its thermal limit regularly and clock down.

If you actually want a lower power/heat setup, you can undervolt and underclock the CPU and not worry about bouncing off the thermal limits.

Sure, I should have included *without causing throttling etc. Things have come a long way in the 18 years(!) since this video was created...

https://www.youtube.com/watch?v=Xf0VuRG7MN4

That's backwards really. The control loop works in the other direction by monitoring the temperature inside the CPU. The CPU will draw as much or as little power as the cooling system permits, as long as the demand exists.

Marketing. It's the same reason AMD decided to tier the Ryzen lineup into a 3, 5, 7, and 9.

Intel had products with TDP of 65w and 125w, so AMD "matched" that by making up a TDP formula to get the results they wanted. At the time, Intel's TDP was actually mostly accurate. It would turbo above that, but only for a few seconds before dropping to the TDP limit.

Then when AMD started winning, Intel had to get match them in return. If AMD's "125w TDP" parts could turbo forever, then why shouldn't Intel's? And so that's what they did. And just like that over a few generations TDP became entirely stupid & without a shred of meaning.

Reply on news.ycombinator.com