Raptor Lake Leaks + Intel 4 developments

[Armageus]

Then why did you post a review that wasnt even testing for your theory and then pretended it supported your theory??? Go ahead then, post a review where when both tested at the same frequency, the e core is more efficient. There isnt any cause that is not the case

Half a paragraph further down it literally says as much..

That means running Gracemont above 3.2 GHz is pointless if energy efficiency is your primary concern. Running the E-Cores at 3.8 GHz basically makes them worse P-Cores. But that’s exactly what Alder Lake does by default.

Below 3 GHz, Gracemont shines. It’s able to maintain better performance at very low power targets (as we saw in the previous section) while consuming less energy. With an integer workload, it’s especially efficient.

And that is relevant...? Why would they run them at 3.5 ghz?

Because to get better efficiency than E cores that's where they need to be clocked?

 

[Bencher]

Half a paragraph further down it literally says as much..

And it still doesnt say that its more efficient at the same clockspeeds...

Because to get better efficiency than E cores that's where they need to be clocked?

And that is relevant how? Your point is that intel doesnt use 16 p cores because of power consumption, yet they are using e cores that are less efficient. Therefore your theory does not hold any water and power consumption is not the reason they don't have 16 p core parts. Yes?

 

[Armageus]

And it still doesnt say that its more efficient at the same clockspeeds...

Because it isn't, it's more efficient when run at the clock speeds it's designed to run at.

It's like driving a Prius everywhere at 120mph and then wondering why the fuel economy is worse than a Ferrari

 

[humbug]

And that is relevant...? Why would they run them at 3.5 ghz?

If its not relevant why did you post it then? I know its not, you're the one who keeps posting it....

 

[Bencher]

If its not relevant why did you post it then? I know its not, you're the one who keeps posting it....

I posted that they should run the p cores at 3.5 ghz?

 

[Deleted User 456458]

P and E cores are different core architectures designed for different uses cases. Not fully sure what value this 'comparison' will yield.

The only thing they have in common is access to the shared L3 and being on the same die (for now). That's is.

 

[Armageus]

P and E cores are different core architectures designed for different uses cases. Not fully sure what value this 'comparison' will yield.

Because bencher seems to think P cores are more efficient

 

[Grim5]

Aren't there rumours of a 24C Zen4 CPU?

Not happening

 

[Grim5]

SiSoft are Intel.

PS: its also the software Intel used to prove the 15% performance improvements they claimed for every quad core of the decade long refresh, all of them turned out to be 3%.

Link to show it's Intel? Wiki says it's a British company in London

 

[Troezar]

Not happening

Prove it Unless you work for AMD it is just as good a guess as anyone else.

 

[humbug]

I posted that they should run the p cores at 3.5 ghz?

And you don't think that running the P cores at 3.5Ghz would put them behind Zen 3 is at all relevant?

 

[humbug]

Not happening

Why not?

 

[Ikana]

Prove it Unless you work for AMD it is just as good a guess as anyone else.

Why not?

Isn't it what Zen 4C is about? I doubt the original Zen 4 will have any more cores unless you can fit a third die into the CPU somehow.

 

[Bencher]

And you don't think that running the P cores at 3.5Ghz would put them behind Zen 3 is at all relevant?

Running them at 1 mhz would also put them behind zen 3. So...they wouldn't run them at 1 mhz.

 

[Bencher]

Because bencher seems to think P cores are more efficient

They are? You seem to think they are not. I mean sure, maybe (that's a MAYBE), if you drop them down to like 1.5 ghz they end up more efficient, but you are not going to do that on a desktop CPU.

 

[Armageus]

They are? You seem to think they are not. I mean sure, maybe (that's a MAYBE), if you drop them down to like 1.5 ghz they end up more efficient, but you are not going to do that on a desktop CPU.

Eh?

P series are most efficient when run between 3ghz and 4ghz. E series are most efficient when run below 3ghz.

If you drop a P series down to 1.5ghz it wouldn't do enough work per watt to be more efficient than E series cores around their ideal 3ghz speed.


Edit: it's like arguing with a wall, so I'm not going to bother replying any further.

 

[humbug]

Running them at 1 mhz would also put them behind zen 3. So...they wouldn't run them at 1 mhz.

You just keep going round and round in circles......

 

[Deleted User 456458]

Because bencher seems to think P cores are more efficient

And they are, for their use case. They’re more efficient at voltage scaling in relation to frequency. You can easily test this. Let me know E cores scale up with vcore. Have you tuned and played with any adl?

Efficiency is a broad term. I showed above v/f scaling factor for efficiency. You can use a host of other baselines in the eff realm.

Asides from that, I can’t repeat enough how useless it is to try and baseline two diverging architectures for specialized use cases.

 

[humbug]

And they are, for their use case. They’re more efficient at voltage scaling in relation to frequency. You can easily test this. Let me know E cores scale up with vcore. Have you tuned and played with any adl?

Efficiency is a broad term. I showed above v/f scaling factor for efficiency. You can use a host of other baselines in the frequency realm.

Asides from that, I can’t repeat enough how useless it is to try and baseline two diverging architectures for specialized use cases.

More efficient than what?

 

[LePhuronn]

The mental gymnastics here trying to justify Intel's technical failings...