News AMD Allegedly Testing Hybrid Processor with Zen 4 and 4c Cores

[bit_user]

In your previous post you mentioned actual draw versus artificial power limits but now you use the artificial power limits as an argument...

I used them as a way of trying to compare products rated & marketed similarly. I did also mention the actual power data provided in this article:

https://www-computerbase-de.translate.goog/2023-02/intel-core-i5-13500-test/?_x_tr_sl=de&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=wapp

If you look at that, it shows the 7700X uses only 95% as much power as the i5-13500 (both stock), in the "Full Load (PC)" benchmark. In the AVX load, the 7700X drops to 89% of the i5. At Full Load (package power), it's 87%. It's only at Idle and Partial Load where the i5 pulls ahead (using 88% and 91%, respectively). As I've said before, I think the chiplet architecture is adding overhead which hurts these processors' efficiency on lightly-threaded workloads. That should be a non-issue for the APUs, like those mentioned in the article.

So, there's actual power data for you. And, again, their multi-threaded performance rating shows the 7700 at 1% faster and 5% faster at single-threaded performance.

They didn't have the 13500 but the 13400 has an PL2 of 148 just 6W below the 13500 and it barely goes above 70W
(And no, the 7700 doesn't hit its max power either)
https://www.techpowerup.com/review/intel-core-i5-13400f/21.html

So, why is WebXPRT at like 1 W? That's implausible for a web benchmark. And if it's a DNF, then do we know they excluded it from the mean?

Also, you refer to 7700, but where's the data on that? I didn't find a review of it, on TechPowerUp. Otherwise, it'd be great to compare like-for-like.

That's your opinion...

The point is irrelevant, because we're talking about laptops here. In particular, the CPU in this article sounds like a very low-end laptop that would struggle to hit 95 C without even a heatsink!
: D

And even if it is due to the packaging it still is what it is.

We can agree the thermal efficiency of their AM5 packaging solution sucks.

You know what's important for laptops?!
Having a super efficient iGPU that can do hardware acceleration on your browsing/ media consumption and face time or can stream video to the net, one that also has basic AI features to sort pictures and apply filters on them without having to use the cpu cores or the power hungry dGPU.
Where are the benches on all that?

I am as eager as you to see benchies on the 780M and the XDNA engine in Phoenix. We'll just have to wait. And I will wait for professional reviews - I'm not going to waste time on this or that "leaked" benchmark.

 

[jeremyj_83]

Don't be drawn into a pitched battle between K vs. X models. It's a classic misdirection tactic. He's lured you onto the terrain he feels is most defensible.

This is an article about laptop CPUs and the relevant point is how well these cores scale down, not up. Never forget the context.

Even in the X vs K discussion, the Zen chip is more efficient. Before the non X models were released Anandtech did a scaling test and limited the CPUs in TDP. The Zen chip with a 65W TDP was usually faster than the Intel with 105W or 125W TDP, except in Cinebench.

 

[jeremyj_83]

Do you do them for more than 10% of your time on the PC??
Because I can bet you anything you want that the average person will be using them less than that, if at all.
The only people that will be using them for more time would be semi-pros or outright server/workstations.

My use case is irrelevant. Same as typical use case. This is nothing more than a deflection tactic because the data says one thing so you pivot and try to move the goal posts to fit your narrative.

 

[TerryLaze]

So, why is WebXPRT at like 1 W? That's implausible for a web benchmark. And if it's a DNF, then do we know they excluded it from the mean?

Also, you refer to 7700, but where's the data on that? I didn't find a review of it, on TechPowerUp. Otherwise, it'd be great to compare like-for-like.

Because that's how webXPRT runs...¯\(ツ)/¯
https://www.techpowerup.com/review/amd-ryzen-7-7700-non-x/22.html

 

[TerryLaze]

My use case is irrelevant. Same as typical use case. This is nothing more than a deflection tactic because the data says one thing so you pivot and try to move the goal posts to fit your narrative.

It's very relevant for selling laptops.

 

[jeremyj_83]

It's very relevant for selling laptops.

Again we have seen time and time again that with Zen 2, Zen 3, and Zen 3+ that laptops with those CPUs have BETTER battery life and performance than the Intel ones (note this would be similar specs from same manufacturer) at the same TDP. This stems from the efficiency of Zen compared to Core. This isn't saying that Core isn't a good uArch. Just that Core isn't as efficient as Zen.

 

[rtoaht]

...And with that certain folks who always bashed big-little concept will suddenly start acting like it is the best thing ever.

 

[bit_user]

...And with that certain folks who always bashed big-little concept will suddenly start acting like it is the best thing ever.

I do expect the reception to be better, but that will be mainly because we'll have had 2 years of experience + software tuning to round off most of the sharp edges and get used to the idea. Also, AMD won't be dropping AVX-512, which was another source of criticism leveled at Intel's approach.

Most criticism I've seen of Intel's Big + Little approach is aimed at desktop usage, not laptops. And if AMD limits it to laptop CPUs, at least initially, that would make it much less controversial than what Intel did.

Personally, I credit Intel for taking the lead on Big + Little. It was a brave move, but it probably had to happen sooner or later.