Hacker News Reader

Comment by SantiagoElf | Hacker News Reader

Yeah, not what I expected.

Here is my result:

https://browser.geekbench.com/v5/cpu/16820778

Given that I didn't stop anything and was playing Dota 2 while running it ...

Still, I believe at some point once newer iterations of X86-64 drop support for legacy instructions (they still have to support code running on 8080/8086 through 486) - we will have smaller, more efficient X86-64 chips.

Actually, it is darn impressive that the latest Intel/AMD chips keep up with the brand new M1/M2 chips that have zero legacy instructions and the Intel/AMD chips carry 50 years of legacy :)

View on HN

Noticed that I ran the previously GeekBench on balanced power plan.

Here are the results on Performance Power Plan - https://browser.geekbench.com/v5/cpu/16821908

Summarized results - https://i.imgur.com/X5vKtFn.png

Overall, with the new results - AMD Ryzen 9 5900HX is 18% slower in Single Core, 4% faster in Multi Core.

Not that bad ...

Not that bad? Those are absolutely terrible.

4% is probably close to noise level, so essentially you are a couple of percentage points faster on multicore with TWICE the total power envelope. Think about that.

And almost 1/5th slower in single core.

Also, you went from "It will OBLITERATE M2" to moving the goal post "not that bad" pretty quickly. I love it.

True.

But still, pointless to me - I don't care about power efficiency, battery life or any synthetic benchmarks.

I have two concerns with a laptop:

1. Can it run the latest Visual Studio at blazing fast speed?

2. Can it run Dota 2 on Full HD @ >144 fps?

So far ... M2 doesn't fulfill fully neither point 1 nor point 2 :)

That's not such a good result when you're talking about a chip with a 54w TDP (even more peak) vs a ~20w peak TDP for the M2.

4% faster for 2.5x more power isn't a win. The fact that 95% of users care more about that 18% single-threaded performance makes this even worse.

The TDP of the 5900HX is 45W.

Don't forget we are comparing last-gen AMD vs current gen M2.

M2 on Mac makes 1884/8717, but on Paralles -> 1681/7260.

So, the penalty is 11% on single, 17% on multi because of running on Windows via Paralles.

APPLE M1 average Single Core is 1706, average multi is 7421 on MAC.

If we apply the above penalty becomes -> 1706/7421 -> 1522/6180.

AMD 5900HX average Single Core is 1413, average multi is 7656.

So, AMD - 1413/7656 | Apple M1 - 1522/6180 -> Amd is 8% slower in single, 19% faster in multi.

> The TDP of the 5900HX is 45W.

The TDP according to AMD's website is "45+W" which they clarify to be a cTDP up to 54w TDP.

https://www.amd.com/en/products/apu/amd-ryzen-9-5900hx

> once newer iterations of X86-64 drop support for legacy instructions (...) - we will have smaller, more efficient X86-64 chips.

Don't get too excited about it. The old compatibility is going to be tiny in the die space compared to even basic Intel extensions. And it's not just the old code - new code may well contain a "mov al..." so you can't just drop it. All of those instructions will stay with us for decades.

Reply on news.ycombinator.com