News Intel Panther Lake processors could pack up to 16 cores, maximum of four performance cores according to leak

[YSCCC]

The i7-13700's PL2 is 219W. So, you're being less than generous, when you cut it down to 125W. It's misleading for people who fall for your ruse and believe you're doing quite the opposite.


Quite projecting.

Speaking of the 7700X, that's just your usual M.O. of drawing sweeping conclusions from cherry-picked, rigged benchmarks.

And then the Chief buys 9950 X3D...

 

[YSCCC]

It is how normal people that care about efficiency use their cpus. If they don't care about it sure, I agree with you, but why does it matter how people that don't care about efficiency use their cpus when the discussion is about efficiency? Those people are irrelevant here.

If a friend asks for a very efficient MT cpu at 300$ obviously I'd strongly strongly advice him to buy an Intel chip (and then cap it to whatever power he feels appropriate) than to buy the amd chip (and cap it to whatever power he seems appropriate) cause the intel chip will be literally (literally literally) 2 generations ahead in both perofmenace and efficiency. The Zen 5 ryzen 7 (which is also more expensive) hasn't caught up with a 2022 i7, and probably neither will zen 6. But hey, let's pretend amd is more efficiency, God knows why.

There is no world we're you'll end up buying an amd chip cause of its MT efficiency (excluding the 7950x), that's a laughable argument and you simply don't really care about efficiency if you do that. Which is perfectly fine, my point is stop pretending you care about it when you are buying chips that are insanely inefficient

It's not even about Intel vs amd, there are people who bought a 5950x over a 7950x because "it's more efficient" and I laughed at them too. The 7950x is leagues ahead of the 5950x. Right here you are basically arguing that the 5950x is more efficient than the 5950x which is a joke.

I believe most ppl here almost never saw someone who will just buy a part and limit it for efficiency, for 99%+ ppl, buying the CPU and use it stock is the easiest way to go, and by effieciecy, they want something cool at a lower cost which will do the job they need, if one use it to render, they will want the fastest part with lowest power draw at a higher than average user power usage, coz they need the work to be done quickly for, work. it doesn't need to be extreme low power usage, nor they do only MT performance in a very specific power level.

For those who really want some power efficient stuffs, they usually do not need much performance in the first place, and for the power efficient minded enthuiast, they want TOTL performance at a lower TDP.

And even INTEL don't claim they have the efficiency crown, guess why.

 

[TheHerald]

If you really believe that, I don't know what to say. Even the most realistic die hard Intel people say the performance is there but not the efficiency. And they believe 20A/18A will fix that. We should have Lunar Lake news in a few hours. That will tell us something.

I don't believe anything. Belief isn't data. I go with data, not beliefs. A 13700 at 125w is around 25% faster and around 25% more efficient than a 7700x at the same power. Tests from eg. ComputerBase clearly show that. If you believe anything other than what the data suggests then you are clearly in the wrong here.

 

[TheHerald]

I believe most ppl here almost never saw someone who will just buy a part and limit it for efficiency, for 99%+ ppl, buying the CPU and use it stock is the easiest way to go, and by effieciecy, they want something cool at a lower cost which will do the job they need, if one use it to render, they will want the fastest part with lowest power draw at a higher than average user power usage, coz they need the work to be done quickly for, work. it doesn't need to be extreme low power usage, nor they do only MT performance in a very specific power level.

For those who really want some power efficient stuffs, they usually do not need much performance in the first place, and for the power efficient minded enthuiast, they want TOTL performance at a lower TDP.

And even INTEL don't claim they have the efficiency crown, guess why.

There is a difference between a cpus efficiency and how efficient a cpu is with out of the box settings. I'm clearly talking about the first, you are clearly talking about the latter.

When it comes to cpu efficiency intel is leading by a lot in most segments. You measure that either by iso power or iso performance. Iso power is extremely biased towards AMD and I choose to use that to remove any notion of me being intel biased. Not that that prevented you from doing so, but if I went with iso performance, the 7700x would need around a kwh and ln2 canisters to match a 13700 chilling at 125w.

If you are talking about how efficiently they come preconfigured then obviously K chips are inefficient out of the box, but why are you focusing on k chips in that case and not the non k and t chips?

At the end of the day, intel leads in both fronts, cpu efficiency and out of the box efficiency. Other than personal attacks I haven't seen anyone posting data to refute that. ComputerBase has made plenty of comparisons. Just take a guess which cpu is the most efficient they have ever tested? It's not an amd one....

 

[TheHerald]

I have absolutely no confidence in you being truthful about the results.

I won't bother.

You miss the point of a forum. By a lightyear (ahead).

In other words, you are wrong. Noted

 

[YSCCC]

I don't believe anything. Belief isn't data. I go with data, not beliefs. A 13700 at 125w is around 25% faster and around 25% more efficient than a 7700x at the same power. Tests from eg. ComputerBase clearly show that. If you believe anything other than what the data suggests then you are clearly in the wrong here.

you clearly only pick one setup at the market segment suggested by both camps and then not use them at the vendor's out of factory settings, why don't you compare 7950x with 14900k then? wait, that is a sure losing war

 

[MacZ24]

In other words, you are wrong. Noted

No wonder you have a hard time understanding anything .. you constantly makes wrong inferences.

You are a fanboy, and reality doesn't faze you. If you are not paid by or invested in Intel, it's quite pathetic.

Otherwise, you are just corrupt.

 

[YSCCC]

No wonder you have a hard time understanding anything .. you constantly makes wrong inferences.

There's a chinese saying that "everyone is drunk with only me being awake"......

 

[TheHerald]

you clearly only pick one setup at the market segment suggested by both camps and then not use them at the vendor's out of factory settings, why don't you compare 7950x with 14900k then? wait, that is a sure losing war

I picked it cause it's a very easy example to prove my case. Do you agree that the 13700 is a lot more efficient than the 7700x? If yes sure, I can compare more parts no problem.

I've already acknowledged multiple times that the 7950x is more efficient than the 14900k at iso power. Thats the only segment amd has the lead. In every other segment they do not.

 

[TheHerald]

No wonder you have a hard time understanding anything .. you constantly makes wrong inferences.

You are refusing to share that workload that will do better with only pcores cause you know yourself you are wrong. As ive said a couple of posts ago, this is clearly trolling.

 

[MacZ24]

You are refusing to share that workload that will do better with only pcores cause you know yourself you are wrong. As ive said a couple of posts ago, this is clearly trolling.

You are a fanboy and can't be honest. It's clear by your posts that you have a side.

If I coded such a workload, you will admit to nothing, spin it every way, nitpick and otherwise lie.

Why should I trust a fanboy that is speaking its porfolio ?

The fact you are even proposing this without realizing it's impossibile to arrive at a sastifying conclusion one way or another just tells everybody everything there is to know about your cognitive capacities. And your lack of self awareness.

It's funny.

 

[TheHerald]

You are a fanboy and can't be honest. It's clear by your posts that you have a side.

If I coded such a workload, you will admit to nothing, spin it every way, nitpick and otherwise lie.

Why should I trust a fanboy that is speaking its porfolio ?

The fact you are even proposing this without realizing it's an impossibility to arrive at a sastifying conclusion one way or another just tells everybody everything there is to know about your cognitive capacities.

It's funny.

If you coded such a workload then it means that currently such a workload doesn't exist which is my point.

It is very possible to arrive at a satisfying conclusion, in every single mt workload 6+8 is considerably faster than 8+0. It's only you that disagrees here while not even offerring me a workload to test. That's insane.

 

[MacZ24]

If you coded such a workload then it means that currently such a workload doesn't exist which is my point.

It is very possible to arrive at a satisfying conclusion, in every single mt workload 6+8 is considerably faster than 8+0. It's only you that disagrees here while not even offerring me a workload to test. That's insane.

Your lack of self awareness if staggering. abysmal. titanic.

You cannot be honest. There is no point of even trying to convince you.

It's like asking a serial killer to judge himself.

That's your level of grotesque and inability to understand.

And your inability to understand that you are unable to understand is really funny.

 

[TheHerald]

So your position is that full pcores is faster in an undisclosed workload that basically doesn't even exist but you are coding one right now, but you won't share it with us after you are done cause we are going to falsify the results.

Clearly, im in the wrong here bud.

 

[MacZ24]

So your position is that full pcores is faster in an undisclosed workload that basically doesn't even exist but you are coding one right now, but you won't share it with us after you are done cause we are going to falsify the results.

Clearly, im in the wrong here bud.

The simple fact that there are 8 p-cores on your CPU just indicates that there are workloads that require 8 p-cores.

Not 2 p-core and x e-cores
Not 4 p-cores and x e-cores
Not 6 p-cores and x-e-cores

Otherwise there wouldn't be 8 p-cores.

If you can do the same workload with 6 p-cores and x e-cores, why 8 p-cores ?
If you can do the same workload with 4 p-cores and x e-cores, why 8 p-cores ?
If you can do the same workload with 2 p-cores and x e-cores, why 8 p-cores ?

What is the black magic that decides that the limit of this is 8 p-cores ?
Why can't there be workload that require 10 p-cores rather than 8 p-cores + some e-cores ?
Why not 12, 14 ,16 ?

But this is obviously too much logic at once. You should take a breather.

 

[TheHerald]

The simple fact that there is 8 p-cores on your CPU just indicates that there are workloads that require 8 p-cores.


But this is obviously to much logic at once. You should take a breather.

And this argument completely disproves your whole thesis, since according to your logic the simple fact that there are 8 e-cores on my CPU also indicates blablaba. Right?

 

[YSCCC]

I picked it cause it's a very easy example to prove my case. Do you agree that the 13700 is a lot more efficient than the 7700x? If yes sure, I can compare more parts no problem.

I've already acknowledged multiple times that the 7950x is more efficient than the 14900k at iso power. Thats the only segment amd has the lead. In every other segment they do not.

so, you have one setup and have to manually limit the power hungry raptor lake to win in some MT tests and one setup where raptor lake loses, and then you have gaming type workload where X3D just slauters intel at basically all power levels, and proves your case? great logic.

How about ignore the name and compare the same thread counting CPUs, naming 7900x vs 13700k?

 

[MacZ24]

And this argument completely disproves your whole thesis, since according to your logic the simple fact that there are 8 e-cores on my CPU also indicates blablaba. Right?

Defeated by too much logic...

 

[danny009]

Hey Intel let me help you with naming:

Bra Lake - Female Windows users gets %5 CPU performance boost out of box for free.

 

[bit_user]

And scheduling, yes ... but when your display thread is waiting for your physics thread, it has to wait ... because it doesn't know where the objects are. That its physical thread is whisked away is irrelevant : it still can't draw.

Okay, so I realized I was overlooking the most obvious point about your thread-synchronization argument: utilization.

The way you can tell whether some threads on faster cores are being hamstrung by threads on slower ones is that task manager (or top, on Linux) would show reduced overall CPU utilization. Like I said, when a thread blocks on a synchronization primitive, like a mutex or barrier, it's becomes suspended and relinquishes the CPU core. If there's nothing else ready to run on that core, it becomes idle.

So, take your favorite multithreaded workload/benchmark/etc. and check overall CPU utilization during the run. If it's virtually 100%, then you know you're not losing performance due to a performance due to core speed disparity! Unfortunately, if you do see utilization below 100%, we can't necessarily conclude why. However, what we can conclude is that 100% utilization should be a sign that what you claim is not happening.

I'd also point out that most multithreaded workloads don't work by spawning N threads to do the same work, then wait on barrier before exchanging results and continuing. That's basically the main way that a performance disparity between the cores would be the kind of liability you're talking about and it's a pattern I've rarely seen in practice. For most other cases, you should see at least some benefit from some of the threads running on faster cores.

 

[Thunder64]

so, you have one setup and have to manually limit the power hungry raptor lake to win in some MT tests and one setup where raptor lake loses, and then you have gaming type workload where X3D just slauters intel at basically all power levels, and proves your case? great logic.

How about ignore the name and compare the same thread counting CPUs, naming 7900x vs 13700k?

It's mental gymnastics.

 

[MacZ24]

I'd also point out that most multithreaded workloads don't work by spawning N threads to do the same work, then wait on barrier before exchanging results and continuing. That's basically the main way that a performance disparity between the cores would be the kind of liability you're talking about and it's a pattern I've rarely seen in practice. For most other cases, you should see at least some benefit from some of the threads running on faster cores.

If your multithreaded application spawn threads that do their work in their corner then infrequently synchronize to exchange data (like for example zipping a large file), thread speed disparity is not a problem, granted.

When all those threads synchronize frequently, like every frame in a game, you will encounter difficulties. The principal being that you have a hard time determining the size of the workload to give each thread when execution speed may vary from 50% to 100%.

In my experience, the OS trying to automagically decide stuff without all the relevant data always lead to subpar performances and crappy behavior.

The fact is that this e-core business (and c cores from AMD) is a specialized hardware, and as such its use may be useful but in my mind it has also drawbacks, that some people just don't want to understand.

Just like Apple having memory on-chip, which is great for performances but is not exempt from drawbacks if you suddenly find yourself without enough memory.

Or SMR or QLC, all specialized hardware that have advantages and disadvantages.

For me General > Specialized, even if it means losing some performance in some cases (who does compress files all day long ?). I'll take the option to change my RAM, thank you.

And laptop sales being 60% higher than desktop play a major role in this "optimization" coming to dektop.

 

[TheHerald]

so, you have one setup and have to manually limit the power hungry raptor lake to win in some MT tests and one setup where raptor lake loses, and then you have gaming type workload where X3D just slauters intel at basically all power levels, and proves your case? great logic.

I have to manually limit both power hungry chips to 125w. The 13700 is still 25% faster than the 7700x. How is that possible, what you just called a power hungry chip being much faster while consuming the same power as a non power hungry chip??? Shouldn't it be slower, since it's more power hungry??

How about ignore the name and compare the same thread counting CPUs, naming 7900x vs 13700k?

Why would you ever compare thread counts? Do you compare GPUs based on cuda cores?

 

[YSCCC]

For me General > Specialized, even if it means losing some performance in some cases (who does compress files all day long ?). I'll take the option to change my RAM, thank you.

I agree completely on this, and for built in ram, maybe coz I am old, I am skeptical for it's longevity in mid term, memory IME always partially fail and runs into error before the CPU gets degraded, say I have 1 pair of corsair vengenance and 1 pair of gskill DDR3 died on the Sandy Bridge, not frequent, but each pair died like 5 years on service and some random error in a single chip arise. And ram, usually degrade quicker at high temp. while DDR5 can sustain 85C, modern CPU runs at.... 100C TJ max, and like running gaming workload, both intel and AMD nowadays runs at some 60C-70C and is considered cool, a ram chip right next to the die don't really sounded like a good idea

 

[YSCCC]

I have to manually limit both power hungry chips to 125w. The 13700 is still 25% faster than the 7700x. How is that possible, what you just called a power hungry chip being much faster while consuming the same power as a non power hungry chip??? Shouldn't it be slower, since it's more power hungry??


Why would you ever compare thread counts? Do you compare GPUs based on cuda cores?

How about trying limit both to say, 45W and compare? you run one basically at it's 142W to 125 and 13700k is 253W stock, both likely runs on the more inefficient range of the cores to gain more performance at "acceptable" power draw, so you are only doing mental gymnastics to try to get some weird fake fair comparison

WHy wouldn't one compare thread counts? it is precisely how the cores/threads a architecture can handle work at specific power, AKA efficieny