Question Raptor Lake - Official Thread

[Hulk]

Since we already have the first Raptor Lake leak I'm thinking it should have it's own thread.
What do we know so far?
From Anandtech's Intel Process Roadmap articles from July:

Built on Intel 7 with upgraded FinFET
10-15% PPW (performance-per-watt)
Last non-tiled consumer CPU as Meteor Lake will be tiled

I'm guessing this will be a minor update to ADL with just a few microarchitecture changes to the cores. The larger change will be the new process refinement allowing 8+16 at the top of the stack.

Will it work with current z690 motherboards? If yes then that could be a major selling point for people to move to ADL rather than wait.

 

[biostud]

Yeah, cause the performance numbers are way off. He has the 12900k @ 125 getting 18k points in CBR23, which is just....not right

Ok, what should he be getting?

 

[Just Benching]

Ok, what should he be getting?

Ask someone with a 12900k to power limit it to 125w and run CBR23. 10$ says he will score 24 to 24.5k in high priority.

TPUs numbers are more flawed the lower you go in power consumption. His testing at 50 and 75w is wayyy off, I score almost double of what he does :O

 

[nicalandia]

Agree, the 200mhz will be useless. But then again, the zen 4 this time around has DDR5, so we don't need for sure how much improvement the bigger cache will bring. Zen 3 was memory starved so the 3d actually made a huge difference.

It's been leaked that the X3D Version of the 7000 Series will bring a 30% Gaming performance boost. Now I believe that a 15% will be enough to keep the gaming lead until Meteor Lake.

 

[Just Benching]

It's been leaked that the X3D Version of the 7000 Series will bring a 30% Gaming performance boost. Now I believe that a 15% will be enough to keep the gaming lead until Meteor Lake.

View attachment 67427

If these are not just cherrypicked games then sure, the gains look impressive.

 

[Abwx]

Yeah, cause the performance numbers are way off. He has the 12900k @ 125 getting 18k points in CBR23, which is just....not right

Computerbase use Cinebench for their power test, whole plateform use 199W for the 5950X and 342W for the 12900K wich manage 4% better perf in those conditions, this should close the debate about efficency.

FTR the 12700K is at 255W, those are abyssimal efficencies and RPL could had made quite a difference without beating the 5000 series, but his opponents will use a smaller node, so there s no hope in this front.

 

[biostud]

Ask someone with a 12900k to power limit it to 125w and run CBR23. 10$ says he will score 24 to 24.5k in high priority.

TPUs numbers are more flawed the lower you go in power consumption. His testing at 50 and 75w is wayyy off, I score almost double of what he does :O

I don't know anyone with a 12900k

 

[Just Benching]

Computerbase use Cinebench for their power test, whole plateform use 199W for the 5950X and 342W for the 12900K wich manage 4% better perf in those conditions, this should close the debate about efficency.

FTR the 12700K is at 255W, those are abyssimal efficencies and RPL could had made quite a difference without beating the 5000 series, but his opponents will use a smaller node so there s no hope in this front.

Those numbers are with stock power limits. Who cares, if you care about efficiency you can change them

Efficiency should be measured at same wattage. Everything else is just flawed. If you don't think that everything else is flawed, then prepare to eat your words, the 13900T will be the most efficient CPU on planet Earth. And since it's obvious how objective you are, you will obviously start calling zen 4 abysmally inefficient compared to the 13900t

 

[Henry swagger]

The sooner they release 13900KS the sooner AMD will release zen4 with 3D cache, so full steam ahead intel

V cache will have lower clocks and low volume

 

[exquisitechar]

Why? Are those extra 200Mhz made out of gold or something? Do you really believe that 200Mhz will change the playing field? Really?

Mind you a 4.5 Ghz 5800X3D overcame a 18% speed deficit, A RAM defiit(DDR4 vs DDR5) and matched a 12900K on average.

The 7800X3D Will have MUCH less to worry about this time.

It’s not the KS’s 200MHz that matters, I just think that tuned Raptor Lake in general will be very fast. Anyway, we don’t even know how Raphael without V-Cache fares against Raptor Lake in terms of gaming performance yet, so I wouldn’t make definitive statements about the X3D parts. All we have are AMD’s benchmarks for Raphael and shoddy leaks for Raptor Lake.

 

[uzzi38]

Those are numbers are with stock power limits. Who cares, if you care about efficiency you can change them

Efficiency should be measure at same wattage. Everything else is just flawed. If you don't think that everything is flawed, then prepare to eat your words, the 13900T will be the most efficient CPU on planet Earth. And since it's obvious how objective you are, you will obviously start calling zen 4 abysmally inefficient compared to the 13900t

As someone that actually knows how efficient Zen 4 is and has seen actual numbers comparing it to Zen 3, you're selling it pretty short.

But you'll see the actual numbers soon enough.

 

[Just Benching]

As someone that actually knows how efficient Zen 4 is and has seen actual numbers comparing it to Zen 3, you're selling it pretty short.

But you'll see the actual numbers soon enough.

You mean zen 4? I don't know, it might be 2nd coming, im talking about zen 3 right now

 

[Abwx]

Those are numbers are with stock power limits. Who cares, if you care about efficiency you can change them

Efficiency should be measure at same wattage. Everything else is just flawed. If you don't think that everything is flawed, then prepare to eat your words, the 13900T will be the most efficient CPU on planet Earth. And since it's obvious how objective you are, you will obviously start calling zen 4 abysmally inefficient compared to the 13900t

As said if you limit the 12900K to 125W it will be short of 1% vs a 5900X.

As for efficency it increase as you reduce frequency, those 13900T will be highly innefficient compared to a similarly power reduced 7950X, to give you an idea at 55-60W it will roughly match a stock 5950X and at 35W it will be around a stock 5900X.

Of course we re talking of actual TDP, not of Intel s benching the 13900T at 100W PL2 and then branding it as a 35W TDP SKU as if it were to perform the same at this power...

 

[uzzi38]

You mean zen 4? I don't know, it might be 2nd coming, im talking about zen 3 right now

You said Zen 4 will look abysmally inefficient compared to the 13900T. I'm just saying that at similarly restricted power limits Zen 4 will look a lot better than you think. Most I'll say for now is mid-4GHz is the new mid-3GHz.

 

[nicalandia]

It’s not the KS’s 200MHz that matters, I just think that tuned Raptor Lake in general will be very fast.

Very Fast indeed, but so will the Ryzen 7000 series.

 

[nicalandia]

OMG The 13700T is faster than the 5800X

Intel's 35W Core i7-13700T Raptor Lake CPU Is Faster Than The 105W Ryzen 7 5800X & 125W Core i5-12600K In Leaked Benchmark

New benchmarks of the Intel Core i7-13700T, a 35W Raptor Lake CPU, have leaked out which show impressive performance figures.

wccftech.com

7600X enters the Chat...

 

[Det0x]

Why are we discussing this again ?

If you handicap Zen3 to compete core for core with CG at a given power level, the Alder lake part can and will win in Cinebench performance to power ratio, with the expense of much larger cores/die.
(But there is a reason why Intel is only producing consumer dies with maximum 8 CG cores)

If you let go of the false narrative that you only should compare core for core, the Ryzen parts smokes Alder Lake cpus in performance/power as ive shown again and again..
The 12900k cant touch the 5950x when both are turned for maximum performance to power:

Spoiler: power testing data proving 5950x is much more power efficient

Snipped from other thread here:
Few words before we dive into data:

• There have been no telemetry trickery with these numbers
• SOC powerusage is ~2w together with 13w usage for rest of CPU. Anything over this constant ~15w draw is going to the cores/caches on the CCD.
• For all the new runs a very wimpy LLC together with set 0.75vcore was used to reach close to 0.7v under load on the lowest power runs

Below we have results for a simulated 8core Zen3 5800x:

8 Zen3 cores @ 33 watt = 11149 points in Cinebench r23 (~lowest powerlevel i could run with my current setup)
3200mhz @ 0.719 vcore under load -> ~2.25watt per core under load
11149points/33watt = 337 points per watt


8 Zen3 cores @ 41 watt = 12603 points in Cinebench r23
3625mhz @ 0.8 vcore under load -> ~3.25watt per core under load
12603points/41watt = 307 points per watt


8 Zen3 cores @ 50 watt = 13698 points in Cinebench r23
3950mhz @ 0.88 vcore under load -> ~4.375watt per core under load
13698points/50watt = ~274 points per watt


8 Zen3 cores @ 64 watt = 15099 points in Cinebench r23
4350mhz @ 0.98 vcore under load -> ~6.125watt per core under load
15099points/64watt = ~235 points per watt


Results for a real 16core Zen3 5950x:

16 Zen3 cores @ 49 watt = 20441 points in Cinebench r23 (~lowest powerlevel i could run with my current setup)
3100mhz / 2950mhz @ 0.7 vcore under load -> ~2.125watt per core under load
20441points/49watt = 417 points per watt


16 Zen3 cores @ 64 watt = 23701 points in Cinebench r23
3575mhz / 3425mhz @ 0.781 vcore under load -> ~3.063watt per core under load
23701points/64watt = 370 points per watt


16 Zen3 cores @ 88 watt = 27037 points in Cinebench r23
4075mhz / 3925mhz @ 0.887 vcore under load -> ~4.5625watt per core under load
27037points/88watt = 307 points per watt

So what have we learned by this comparison ?
A underclocked 12900k can be more efficient then a underclocked 5800x when you handicap Zen3 with its size advantage and only compare core for core with GC, but at the same time it cant touch a underclocked 5950x in energy efficiency as the numbers show.

Like i said earlier, in the end it all boils down to GC physical size, they are so big that intel could only put 8(10) of those on a consumer cpu(die) and keep the price in check at the same time. (10P cores would score lower than 8p+8E in full multithreaded benchmarks)
But that is no reason to handicap desktop Zen3 with a artificial limit for 8 cores maximum in this efficiency comparison when we both have to 5900x and 5950x as normal desktop consumer cpus for sale today

If you dont think this is correct, please provide your own data (with benchmate screenshot) showing comparable "points per watt" for any Alder Lake CPU you wish..
And please dont start speaking of a hypothetical 16 core CG this time, as a 32core TR would eat it for breakfast the same way

CG cores are simply too big too compete in performance/power with the nimble Zen3 cores, in any actual reallife products/cpus you can buy.

 

[Just Benching]

You said Zen 4 will look abysmally inefficient compared to the 13900T. I'm just saying that at similarly restricted power limits Zen 4 will look a lot better than you think. Most I'll say for now is mid-4GHz is the new mid-3GHz.

Oh, you are talking about same wattage? Sure, don't disagree with that, was talking out of the box though

 

[Just Benching]

As said if you limit the 12900K to 125W it will be short of 1% vs a 5900X.

As for efficency it increase as you reduce frequency, those 13900T will be highly innefficient compared to a similarly power reduced 7950X, to give you an idea at 55-60W it will roughly match a stock 5950X and at 35W it will be around a stock 5900X.

Of course we re talking of actual TDP, not of Intel s benching the 13900T at 100W PL2 and then branding it as a 35W TDP SKU as if it were to perform the same at this power...

Ah, so now that zen 4 will lose in efficiency out of the box, you want to run both at same wattage? I see you are being very consistent

 

[Just Benching]

Why are we discussing this again ?

If you handicap Zen3 to compete core for core with CG at a given power level, the Alder lake part can and will win in Cinebench performance to power ratio, with the expense of much larger cores/die.
(But there is a reason why Intel is only producing consumer dies with maximum 8 CG cores)

If you let go of the false narrative that you only should compare core for core, the Ryzen parts smokes Alder Lake cpus in performance/power as ive shown again and again..
The 12900k cant tough the 5950x when both are turned for maximum performance to power:

Spoiler: power testing data proving 5950x is much more power efficient

Snipped from other thread here:
Few words before we dive into data:

• There have been no telemetry trickery with these numbers
• SOC powerusage is ~2w together with 13w usage for rest of CPU. Anything over this constant ~15w draw is going to the cores/caches on the CCD.
• For all the new runs a very wimpy LLC together with set 0.75vcore was used to reach close to 0.7v under load on the lowest power runs

Below we have results for a simulated 8core Zen3 5800x:

8 Zen3 cores @ 33 watt = 11149 points in Cinebench r23 (~lowest powerlevel i could run with my current setup)
3200mhz @ 0.719 vcore under load -> ~2.25watt per core under load
11149points/33watt = 337 points per watt


8 Zen3 cores @ 41 watt = 12603 points in Cinebench r23
3625mhz @ 0.8 vcore under load -> ~3.25watt per core under load
12603points/41watt = 307 points per watt


8 Zen3 cores @ 50 watt = 13698 points in Cinebench r23
3950mhz @ 0.88 vcore under load -> ~4.375watt per core under load
13698points/50watt = ~274 points per watt


8 Zen3 cores @ 64 watt = 15099 points in Cinebench r23
4350mhz @ 0.98 vcore under load -> ~6.125watt per core under load
15099points/64watt = ~235 points per watt


Results for a real 16core Zen3 5950x:

16 Zen3 cores @ 49 watt = 20441 points in Cinebench r23 (~lowest powerlevel i could run with my current setup)
3100mhz / 2950mhz @ 0.7 vcore under load -> ~2.125watt per core under load
20441points/49watt = 417 points per watt


16 Zen3 cores @ 64 watt = 23701 points in Cinebench r23
3575mhz / 3425mhz @ 0.781 vcore under load -> ~3.063watt per core under load
23701points/64watt = 370 points per watt


16 Zen3 cores @ 88 watt = 27037 points in Cinebench r23
4075mhz / 3925mhz @ 0.887 vcore under load -> ~4.5625watt per core under load
27037points/88watt = 307 points per watt

If you dont think this is correct, please provide your own data (with benchmate screenshot) showing comparable "points per watt" for any Alder Lake CPU you wish..
And please dont start speaking of a hypothetical 16 core CG this time, as a 32core TR would eat it for breakfast the same way

CG cores are simply too big too compete in performance/power with the nimble Zen3 cores, in any actual reallife products/cpus you can buy.

I did post my numbers with the full 12900k on and it wiped the 5950x with 434 points per watt. And that's with default motherboard settings, no tuning

 

[Zucker2k]

It's been leaked that the X3D Version of the 7000 Series will bring a 30% Gaming performance boost. Now I believe that a 15% will be enough to keep the gaming lead until Meteor Lake.

View attachment 67427

(2) All samples were limited to the same tdp? Is this a power consumption test or performance test? I'm beginning to smell some desperation from AMD with this "leak." 😅

 

[nicalandia]

(2) All samples were limited to the same tdp? Is this a power consumption test or performance test? I'm beginning to smell some desperation from AMD with this "leak." 😅

Are you Blind or something? It's relative performance benchmarks at same TDP.

 

[Abwx]

Ah, so now that zen 4 will lose in efficiency out of the box, you want to run both at same wattage? I see you are being very consistent

There will be non X SKUs as well, i expect the 7700X/7900X, and why not the 7950X, to be down powered to 45-55W for use in gaming laptops, they dont need to go as low as 35W for top level efficency.

A full node advantage in perf/watt is what it is and cant be circumvented other than by 2 x the core advantage, 8 + 16 would make sense on a throughput and perf/Watt basis if the contender was a 7 nm based Zen 4.

 

[Zucker2k]

Are you Blind or something? It's relative performance benchmarks at same TDP.

A gaming benchmark at same tdp? Where has something like this ever been done?

 

[Just Benching]

There will be non X SKUs as well, i expect the 7700X/7900X, and why not the 7950X, to be down powered to 45-55W for use in gaming laptops, they dont need to go as low as 35W for top level efficency.

A full node advantage in perf/watt is what it is and cant be circumvented other than by 2 x the core advantage, 8 + 16 would make sense on a throughput and perf/Watt basis if the contender was a 7 nm based Zen 4.

Sure, until those non x 45w tdp cpus show up, the 13900T will be the most efficient cpu out of the box, right?

 

[coercitiv]

A gaming benchmark at same tdp? Where has something like this ever been done?

At some point in the future I hope we'll start distinguishing between leaks and MLID leaks. The second category is "special"