AMD Ryzen AI 300 Series Dominates Intel Core Ultra 7 Lunar Lake Performance For Linux Developers & Creators

[zachw]

In the future, it'd be cool to include test results from an Apple MacBook running Asahi Linux. That might shed some additional light on if the battery life angle Intel seems to be taking is competitive with ARM CPUs.

[Nempk]

Some of these results look suspicious:

In the audio encoding benchmarks (I guess all single threaded?), how is Zen5 not even able to beat Tiger Lake?

I remember Intel having better encoder performance than AMD for some time, nut sure is about CPU

[Teggs]

I can understand a quasi-quadcore CPU being overloaded in some of these tests. But those efficiency numbers... with onboard RAM. Ouch. Intel was so focused on this product, I didn't think they would drop the ball to have day 1 performance like that.

I expect Intel to go bankrupt at this point, but if they can't even make a decent launch of a decent product, they'll get there in freefall instead of in a handbasket tied to a sinking balloon. (Am I mixing metaphores? Well, some of the people at Intel can go to hell for what they've done to the company).

[Kabbone]

I'm also a bit confused of some results. There are benchmarks where the lunar lake is the fastest, used the least amount of energy but performance per Watt is worse than AMD, so it's a bit hard to compare without taking the energy used into account.

[kylew77]

For a business laptop that is gonna be running Edge or Chrome, MS Teams, MS defender, and Windows 11 I think they hit the nail on the head as evident by the web browser scores. This is not however a good programers laptop as evident by the compiling and multithreaded scores, but it makes a good business laptop and beats out the competition there to me at least.

[pWe00Iri3e7Z9lHOX2Qx]

As much as I bag on Intel for their (at least) half decade of fab incompetence, I feel like some of the Lunar Lake coverage here is missing the point. These SKUs are targeted specifically at efficiency, especially at 15W or less. These are supposed to be road (air these days!) warrior machines with battery life to compete with Qualcomm and Apple. At 15W, performance is quite good, and iGPU performance is class leading for x86-64 (in Windows).

[Eumaios]

Michael writes, "The Core Ultra 7 256V was consuming less power than the other laptop CPUs which is nice if you are doing a lot of code compilation on battery". But I read in a review of Lunar Lake on another website that the processor's performance drops considerably when running on battery.

Michael, do you test these laptops on wall power or battery power?

[Mitch]

I wonder how this might get impacted by AMD's recent PR for hetero CPU architecture.

[blackshard]

54W 12 cores/24 threads dominate 37W 8 cores/8 threads?

You are ignorant:

• Intel choose to remove hyperthreading from their processors. They said it was not worthy the silicon space for their architecture. If you want to blame someone, blame Intel
• The ryzen AI HX have a configurable cTDP from 15 to 54W. On the Asus S16 they are configured as 28W parts. This Intel Core 256V is a non-configurable 17W part
• The Ryzen 365 AI HX is a 10 core part (4 * Zen5 + 6 * Zen5c)

I would think so. It looks like the weight categories are quite different, don't you think?

If you look only at the performanca graphs, it may be. But when you look at the efficiency graphs, the less consuming chips usually gets an advantage and, even in that case, the Intel Core 256V is not able to reach the Ryzen efficiency anyway.

What about battery life though? What about video playback life though?

Battery life will be awesome, but battery life depends upon the notebook and not about the processor only. Anyway the Asus S14 will be "slow" but battery life will be awesome. Are you happy?

What about idle power consumption? AFAK it's close to 4W for Ryzen HX 370 and roughly 0.5W for Intel Core 256V.
These are laptop CPUs, not workstations, Michael. And people rarely use them for heavy MT tasks, unless you don't mind a high-pitched noise close to your ears.

Indeed: you buy a top-edge notebook to watch videos and keep it idle. Personally, I bought a top-edge notebook because I need it for compilation, which is totally an MT task. If you don't need a notebook for heavy MT tasks, you can any buy lower-end Ryzen 3/5 or Intel i3/i5 because they will serve any light task purposes well enough.

[blackshard]

I can understand a quasi-quadcore CPU being overloaded in some of these tests. But those efficiency numbers... with onboard RAM. Ouch. Intel was so focused on this product, I didn't think they would drop the ball to have day 1 performance like that.

I expect Intel to go bankrupt at this point, but if they can't even make a decent launch of a decent product, they'll get there in freefall instead of in a handbasket tied to a sinking balloon. (Am I mixing metaphores? Well, some of the people at Intel can go to hell for what they've done to the company).

You're completely off-target: this processor is not intended to compete as a top performer, otherwise it would not be a 17W part.
This new architecture and its fabrication process are promising, the efficiency graphs show a great leap and the gap against AMD is greatly reduced. While I expect the gap to widen a bit with higher power chips (but it could even not be the case), this is indeed a valid architecture.

Furthermore: Intel will not go bankrupt for this. They still have something like 80% of the market, they sold crap until yesterday for years and did not go bankrupt, they won't go bankrupt today nor tomorrow.