• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Raptor Lake Leaks + Intel 4 developments

Armageus said:
You know exactly what I mean by real - "P" Cores. It's nice that you still didn't answer the question as to what is stopping them from making "real" 16/24/32/48/64 core parts?
We have 38 Core Xeon's as they are 300W parts, that cannot be sensibly cooled in desktop PCs.
Click to expand...
I did answer. There is no point in making a 16 P core cpu since it would be SLOWER in Mt performance and equal in ST performance compared to an 8 + 32E configuration.
 
Bencher said:
No, e cores are LESS efficient than P cores. They are not used for efficiency.

A cpu can either be 16P cores or 8P cores + 32E cores, right? Okay, both of these CPUs have the same single threaded performance, but the 2nd one has way better multithreaded performance. So why would choose the 1st one over the 2nd one?
Click to expand...
Ok I'm confused now, they are less efficient but better at multithreading, because there are more of them? What if you don't need all of them? Say you have software that can use 10 P cores but only a few E cores. Wouldn't 16P cores then be better?
 
Armageus said:
We have 38 Core Xeon's as they are 300W parts, that cannot be sensibly cooled in desktop PCs
Click to expand...

So we can have 38 cores @ 300w but we can't have 16 core at 240w? You need to work on your math.

Again, are you suggesting that the 16 P core part would not stay within the TDP limit set by Intel? I really don't understand what the heck you are saying.
 
Troezar said:
Ok I'm confused now, they are less efficient but better at multithreading, because there are more of them? What if you don't need all of them? Say you have software that can use 10 P cores but only a few E cores. Wouldn't 16P cores then be better?
Click to expand...
Say you have an application that can run on up to 10 cores, then yes, a 10p core part would be faster than an 8 + 16 part. But the norm is - either applications run on a handful of cores (autodesk / photoshop / premiere and the likes), or they go nuts and use ever resource you throw at them.
 
Armageus said:
Eh? So what is stopping them from making "real" 16/24/32/48/64 core parts?

Also if E cores are "just as good", then why aren't there any E-core only parts? You can get 4 E cores in the same space as a P Core, so surely we could have had a 0P 40E core part by now?
Instead of 38 or 40 Core Xeon parts, we could have had 152 or 160 E Core Xeon's which the server market would be lapping up?
Click to expand...

All e-core server parts are coming down the road.

E-cores have two main goals.

- Mobile space scaling
- Node scaling virtualization in the server space

I much rather have a 10p + 8e configuration but that won't come with RPL. At least with L3 being shared you can disable ecore clusters which free's up the L3 for the p-cores to use.
 
The E cores are not a bad idea for scaling MT.

Even for AMD IMO its not a bad idea, 4 Zen 2 cores on each CCD along side the 8 Zen 4 cores wouldn't take up much die space or power on 5nm, effectively you would have a 7950X + a 3700X.
 
Bencher said:
E cores are die space efficient, not power efficient. A P core IS more efficient when it comes to power consumption at same clockspeeds. If you have one, you can test it
Click to expand...
So unlike all the original ARM big.LITTLE concept then?
Not sure about the being the best form of flattery if they couldn't even get that part right then.

I guess we will have to wait until the E cores have their own voltages, because there is something very wrong with Intel's current approach.
 
It certainly doesn't make them "power efficient CPU's" Tim just reviewed the Asus Zen book 6800U vs a Dell with a 12600P, the 6800U has 18% better battery life in mixed idle - workloads for the same capacity battery, the 6800U is a Zen 3+ APU on 6nm.
 
Bencher said:
So we can have 38 cores @ 300w but we can't have 16 core at 240w? You need to work on your math.

Again, are you suggesting that the 16 P core part would not stay within the TDP limit set by Intel? I really don't understand what the heck you are saying.
Click to expand...

The 38 cores aren't at 5Ghz+ however.
A 16P Core part at 5Ghz isn't possible as would probably be 480Watts+

Bencher said:
E cores are die space efficient, not power efficient. A P core IS more efficient when it comes to power consumption at same clockspeeds. If you have one, you can test it
Click to expand...

I don't have one so can't test unfortunately, but other people have.
https://techteamgb.co.uk/2021/11/10...i9-12900k-i5-12600k-p-core-e-core-benchmarks/

P cores use 4x more power than E cores, but only deliver 2.5x more performance - E cores are therefore more efficient in terms of perf/watt, but would have been much better if they hadn't been clocked outside of their actual efficiency curve (which peaks at around 3Ghz)

E cores were supposed to be power efficient, but in Alder Lake they've been driven outside of their efficiency curve.
chipsandcheese.com

Alder Lake’s Power Efficiency – A Complicated Picture

Reviews across the internet show Alder Lake getting very competitive performance with very high power consumption.
chipsandcheese.com chipsandcheese.com


KompuKare said:
So unlike all the original ARM big.LITTLE concept then?
Not sure about the being the best form of flattery if they couldn't even get that part right then.
Click to expand...
They've driven them outside of the power consumption sweet spot, as they've had to in order to beat AMD on performance.
 
Last edited:
humbug said:
It certainly doesn't make them "power efficient CPU's" Tim just reviewed the Asus Zen book 6800U vs a Dell with a 12600P, the 6800U has 18% better battery life in mixed idle - workloads for the same capacity battery, the 6800U is a Zen 3+ APU on 6nm.
Click to expand...
If you are talking about the hwunboxed review, they didn't have the same capacity battery....
 
Armageus said:
The 38 cores aren't at 5Ghz+ however.
A 16P Core part at 5Ghz isn't possible as would probably be 480Watts+
Click to expand...
WHY would it need to run at 5ghz? Wtf are you even talking about, of course it wouldn't be running at 5 ghz, just like the AMD part don't run at 5ghz...I don't even understand what your objection is, yes, higher core parts run at lower frequencies.


Armageus said:
I don't have one so can't test unfortunately, but other people have.
https://techteamgb.co.uk/2021/11/10...i9-12900k-i5-12600k-p-core-e-core-benchmarks/

P cores use 4x more power than E cores, but only deliver 2.5x more performance - E cores are therefore more efficient in terms of perf/watt, but would have been much better if they hadn't been clocked outside of their actual efficiency curve (which peaks at around 3Ghz)
Click to expand...
From the freaking reviews YOU provided, it proves my point. At same clockspeeds the P cores ARE more efficient, lol. This is from your link

Between 3 and 4 GHz, these P-Cores can give the E-Cores a run for their money. In an integer workload, Golden Cove consumes about the same amount of total energy while completing the task faster.
 
Bencher said:
From the freaking reviews YOU provided, it proves my point. At same clockspeeds the P cores ARE more efficient, lol. This is from your link

Between 3 and 4 GHz, these P-Cores can give the E-Cores a run for their money. In an integer workload, Golden Cove consumes about the same amount of total energy while completing the task faster.
Click to expand...

/facepalm
Yes because between 3Ghz and 4Ghz (a Speed that they don't run at by default), is the efficiency sweet spot for these (P Cores), which is then being compared to between 3.2Ghz and 3.8Ghz for the E cores, which is outside the sweet spot for those.

Running both in their sweet spot for performance/watt efficiency then E cores are still more power efficient.
 
It seems pretty clear Intel are behind in the technology but AMD need to keep pushing or they will get caught. The overall package is still in favour of AMD in my opinion but Intel are doing well given where they were. They need to be more competitive on price too.
 
Armageus said:
/facepalm
Yes because between 3Ghz and 4Ghz (a Speed that they don't run at by default), is the efficiency sweet spot for these (P Cores), which is then being compared to between 3.2Ghz and 3.8Ghz for the E cores, which is outside the sweet spot for those.

Running both in their sweet spot for performance/watt efficiency then E cores are still more power efficient.
Click to expand...
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
 
You must log in or register to reply here.
Back
Top Bottom