Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads

Tigerick · Aug 22, 2022

As Hot Chips 34 starting this week, Intel will unveil technical information of upcoming Meteor Lake (MTL) and Arrow Lake (ARL), new generation platform after Raptor Lake. Both MTL and ARL represent new direction which Intel will move to multiple chiplets and combine as one SoC platform.

MTL also represents new compute tile that based on Intel 4 process which is based on EUV lithography, a first from Intel. Intel expects to ship MTL mobile SoC in 2023.

ARL will come after MTL so Intel should be shipping it in 2024, that is what Intel roadmap is telling us. ARL compute tile will be manufactured by Intel 20A process, a first from Intel to use GAA transistors called RibbonFET.

Comparison of upcoming Intel's U-series CPU: Core Ultra 100U, Lunar Lake and Panther Lake

Model	Code-Name	Date	TDP	Node	Tiles	Main Tile	CPU	LP E-Core	LLC	GPU	Xe-cores
Core Ultra 100U	Meteor Lake	Q4 2023	15 - 57 W	Intel 4 + N5 + N6	4	tCPU	2P + 8E	2	12 MB	Intel Graphics	4
?	Lunar Lake	Q4 2024	17 - 30 W	N3B + N6	2	CPU + GPU & IMC	4P + 4E	0	12 MB	Arc	8
?	Panther Lake	Q1 2026 ?	?	Intel 18A + N3E	3	CPU + MC	4P + 8E	4	?	Arc	12

Comparison of die size of Each Tile of Meteor Lake, Arrow Lake, Lunar Lake and Panther Lake

	Meteor Lake	Arrow Lake (20A)	Arrow Lake (N3B)	Lunar Lake	Panther Lake
Platform	Mobile H/U Only	Desktop Only	Desktop & Mobile H&HX	Mobile U Only	Mobile H
Process Node	Intel 4	Intel 20A	TSMC N3B	TSMC N3B	Intel 18A
Date	Q4 2023	Q1 2025 ?	Desktop-Q4-2024 H&HX-Q1-2025	Q4 2024	Q1 2026 ?
Full Die	6P + 8P	6P + 8E ?	8P + 16E	4P + 4E	4P + 8E
LLC	24 MB	24 MB ?	36 MB ?	12 MB	?
tCPU	66.48
tGPU	44.45
SoC	96.77
IOE	44.45
Total	252.15

Intel Core Ultra 100 - Meteor Lake

As mentioned by Tomshardware, TSMC will manufacture the I/O, SoC, and GPU tiles. That means Intel will manufacture only the CPU and Foveros tiles. (Notably, Intel calls the I/O tile an 'I/O Expander,' hence the IOE moniker.)

Wolverine2349 · Thursday at 7:16 PM

MS_AT said:
I understand enthusiasm but at this pace, Skymont will turn out to be better than yet to be announced Apple M5. So if we start to pay attention to what intel claimed

under following conditions

So first 2% IPC parity has +/- 10% error.

Next Intel used for comparison GCC12.1 at O2 optimization level. No further flags were given. Is therefore Intel trying to hide something? Nope I believe they are as honest as one can get, since Intel used GCC12.1 without specifying -march/mtune options the code will be tuned for baseline x64 target that iirc doesn't use anything more fancy than old SSE 2 or 4 at best. Read 128b per execution unit, so exceeding what Raptor Cove can do at that bit width, what places Raptor Cove at slight disadvantage as there are SPEC subtests that can be autovectorized by the compiler and if allowed to would make use of 256b SIMD units of Raptor Cove. But this is more relevant for non gaming workloads.

Last they are given the same fabric so the same amount of L3 cache. It's up for debate if they were therefore measuring 1 Skymont core from the cluster, so it had 2MB of L2 cache or not. I guess they did as these cores go in clusters so if it was meant to be representative it should be this. In reality especially for MT workloads each Skymont core on average will have 1/4 of this, so 0.5MB. This is more relevant for gaming workloads.

[source for intel quotes https://edc.intel.com/content/www/us/en/products/performance/benchmarks/computex-2024/]

So as you can see the IPC parity has very strict conditions.

We of course need to wait for real tests, but overhyping Skymont might end up in the same way as overhyping Zen5. Approach with caution and pat them on the back if they really deliver.

I see the points. Though 4 Skymont cores have a cluster of 4MB L2 cache. So assume 1MB per core as 4 cores in a cluster. Half that of Raptor Cove and 80% that of Golden Cove for 1MB per core where Golden Cove had 1.25MB L2 per core I think.

So maybe it could still be very good and Golden Cove IPC and latency which once again still very strong and viable platform.

I would buy a 10-12 P core 5GHz Golden Cove/Alder Lake CPU on a single ring bus in a heartbeat if it existed.

H433x0n · Thursday at 8:21 PM

ondma said:
I would think so. 7800x3D is a few percent ahead of flat out 14900k, Going to the tile architecture and with a <10% 1T performance increase, I cant see ARL being more than a few percent at best faster than 14900K, and thus at best equal to 7800x3D. In fact, worst case, with the new architecture and minimal ST performance gain, I could see ARL showing no gain or even a slight regression vs RL.

I don’t think it’s going to be as bad as expected. There was a leaked Aida64 latency benchmark and the memory latency is ~5-10ns worse than RPL but ARL has better L3$ performance.

I don’t think it’ll perform much better but I no longer think it’ll be a regression.

Wolverine2349 · Thursday at 8:25 PM

H433x0n said:
I don’t think it’s going to be as bad as expected. There was a leaked Aida64 latency benchmark and the memory latency is ~5-10ns worse than RPL but ARL has better L3$ performance.

I don’t think it’ll perform much better but I no longer think it’ll be a regression.

Arrow Lake compared to what the 14900K in gaming or 7800X3D or both?

Do you think it ties or beats 7800X3D in games that do not scale past 6-8 threads? Or loses to it?

Or close enough that in heavily threaded games it beats 7800X3D due to 8 Lion Cove then 12-16 Skymont e-cores??

Or hard to say?

H433x0n · Thursday at 8:36 PM

Wolverine2349 said:
Arrow Lake compared to what the 14900K in gaming or 7800X3D or both?

Do you think it ties or beats 7800X3D in games that do not scale past 6-8 threads? Or loses to it?

Or close enough that in heavily threaded games it beats 7800X3D due to 8 Lion Cove then 12-16 Skymont e-cores??

Or hard to say?

It's impossible to project. Gaming performance is hard to nail down, even for currently released CPUs. It varies a decent bit from reviewer to reviewer. I personally don't think the 7800X3D outperforms the 14900K, they both have their strengths and weaknesses and I view them as roughly equal. In my opinion, most reviewers benchmark games in an odd way. You'll almost never see games tested with max settings + RT (if available), which is how those with high end gaming rigs will run their games.

Anyway, my point is that it's hard to say - I would expect them to all be within 10% of each other.

Wolverine2349 · Thursday at 8:48 PM

H433x0n said:
It's impossible to project. Gaming performance is hard to nail down, even for currently released CPUs. It varies a decent bit from reviewer to reviewer. I personally don't think the 7800X3D outperforms the 14900K, they both have their strengths and weaknesses and I view them as roughly equal. In my opinion, most reviewers benchmark games in an odd way. You'll almost never see games tested with max settings + RT (if available), which is how those with high end gaming rigs will run their games.

Anyway, my point is that it's hard to say - I would expect them to all be within 10% of each other.

I think you are correct.

The 14900K kind of ties 7800X3D.

Having those more cores can be helpful. I was once onto them, but the stability and degradation and random other unknown issues (weird crashes and insane heat output) of 13th and 14th gen turned me off and had me go back to 7800X3D.

Arrow Lake should have much lower power draw and be stable so its something to consider for me unlike Intel 13th and 14th Gen.

alcoholbob · Friday at 1:26 AM

H433x0n said:
It's impossible to project. Gaming performance is hard to nail down, even for currently released CPUs. It varies a decent bit from reviewer to reviewer. I personally don't think the 7800X3D outperforms the 14900K, they both have their strengths and weaknesses and I view them as roughly equal. In my opinion, most reviewers benchmark games in an odd way. You'll almost never see games tested with max settings + RT (if available), which is how those with high end gaming rigs will run their games.

Anyway, my point is that it's hard to say - I would expect them to all be within 10% of each other.

The 7800X3D clearly outperforms in tech reviewer’s standard runs, which is typically run at similar memory speeds (DDR5-5200 to DDR5-6000).

The thing is 13th/14th gen *will* beat the 7800X3D by a pretty substantial margin, maybe up to 10+%, if you run really fast RAM (i.e. 8000 MHz or higher). It’s just reviewers usually don’t bother to benchmark at these settings. But it’s something Intel users can do to juice their performance since they don’t have to deal with the infinity fabric. So realistically, Arrow Lake will most likely beat the 9800X3D too if you run some high memory speeds. It’s just going to look much worse on a typical online or tech tube review because memory speeds are usually equalized.

511 · Friday at 1:31 AM

9200MT/s is gonna be insane

H433x0n · Friday at 1:48 AM

alcoholbob said:
The 7800X3D clearly outperforms in tech reviewer’s standard runs, which is typically run at similar memory speeds (DDR5-5200 to DDR5-6000).

The thing is 13th/14th gen *will* beat the 7800X3D by a pretty substantial margin, maybe up to 10+%, if you run really fast RAM (i.e. 8000 MHz or higher). It’s just reviewers usually don’t bother to benchmark at these settings. But it’s something Intel users can do to juice their performance since they don’t have to deal with the infinity fabric. So realistically, Arrow Lake will most likely beat the 9800X3D too if you run some high memory speeds. It’s just going to look much worse on a typical online or tech tube review because memory speeds are usually equalized.

I wasn't referencing memory speeds but rather the quality settings used in benchmarks. If you run 1080P with maxed settings + RT, the 14900K will often outperform the 7800X3D whereas in reviewer benchmarks it'll show the opposite result - a good example would be Jedi Survivor & Plague Tale Requiem.

Det0x · Friday at 4:06 AM

H433x0n said:
It's impossible to project. Gaming performance is hard to nail down, even for currently released CPUs. It varies a decent bit from reviewer to reviewer. I personally don't think the 7800X3D outperforms the 14900K, they both have their strengths and weaknesses and I view them as roughly equal. In my opinion, most reviewers benchmark games in an odd way. You'll almost never see games tested with max settings + RT (if available), which is how those with high end gaming rigs will run their games.

Anyway, my point is that it's hard to say - I would expect them to all be within 10% of each other.

Wolverine2349 said:
I think you are correct.

The 14900K kind of ties 7800X3D.

Having those more cores can be helpful. I was once onto them, but the stability and degradation and random other unknown issues (weird crashes and insane heat output) of 13th and 14th gen turned me off and had me go back to 7800X3D.

Arrow Lake should have much lower power draw and be stable so its something to consider for me unlike Intel 13th and 14th Gen.

If you dont mind me asking, do you guys think the latest KB5041587 patch would change anything for this comparison ? (7800X3D vs the 14900K)

Do you guys also think this patch should be enabled by default by all reviewers going forward ? --> As in when Arrow Lake releases it should be compared aginst the AMD 7000 and 9000 series with KB5041587 enabled ?

cebri1 · Friday at 5:32 AM

285K、265K、265KF、245K 與 245KF 等 5 款 Intel Core Ultra 2 系列桌上型處理器規格定案 - BenchLife.info

Intel 新一代的 Core Ultra 2 桌上型處理器，應該值得期待一下。代號 Arrow […]

benchlife.info

Mmm... 285K all p-cores 5.6Ghz according to the latest leak.

	Core Ultra 9 285K	Core Ultra 7 275K	Core Ultra 7 275KF	Core Ultra 5 245K
Cores（P + E）	8 + 16	8 + 12	8 + 12	6 + 8
Threads	24	20	20	14
Intel Smart Cache（L3）	36MB	30MB	30MB	24MB
Total L2 Cache	40MB	36MB	36MB	26MB
Intel TVB	Up to 5.7GHz	Up to 5.5GHz	Up to 5.5GHz	Up to 5.2GHz
Intel TBMT 3.0	Up to 5.6GHz	Up to 5.5GHz	Up to 5.5GHz	-
P-Cores Max Turbo	Up to 5.6GHz	Up to 5.4GHz	Up to 5.4GHz	Up to 5.2GHz
E-Cores Max Turbo	Up to 4.6GHz	Up to 4.6GHz	Up to 4.6GHz	Up to 4.6GHz
P-Cores Base	3.7GHz	3.9GHz	3.9GHz	4.2GHz
E-Cores Base	3.2GHz	3.3GHz	3.3GHz	3.6GHz
Processor Base Power	125W	125W	125W	125W
Maximum Turbo Power	250W	250W	250W	159W

mmaenpaa · Friday at 5:45 AM

Det0x said:
If you dont mind me asking, do you guys think the latest KB5041587 patch would change anything for this comparison ? (7800X3D vs the 14900K)

Do you guys also think this patch should be enabled by default by all reviewers going forward ? --> As in when Arrow Lake releases it should be compared aginst the AMD 7000 and 9000 series with KB5041587 enabled ?

KB5041587 is already installed on W11 systems if one allows automatic updating.

I see no valid reason why reviewers would not test with latest updates. Also I really would like to see 14X00K/Arrowlake tests with new Intel default bios settings also.

511 · Friday at 5:55 AM

mmaenpaa said:
KB5041587 is already installed on W11 systems if one allows automatic updating.

I see no valid reason why reviewers would not test with latest updates. Also I really would like to see 14X00K/Arrowlake tests with new Intel default bios settings also.

Good news default BIOS will be the default BIOS

Henry swagger · Friday at 7:07 AM

cebri1 said:
285K、265K、265KF、245K 與 245KF 等 5 款 Intel Core Ultra 2 系列桌上型處理器規格定案 - BenchLife.info

Intel 新一代的 Core Ultra 2 桌上型處理器，應該值得期待一下。代號 Arrow […]

benchlife.info

Mmm... 285K all p-cores 5.6Ghz according to the latest leak.

Core Ultra 9 285K Core Ultra 7 275K Core Ultra 7 275KF Core Ultra 5 245K
Cores（P + E） 8 + 16 8 + 12 8 + 12 6 + 8
Threads 24 20 20 14
Intel Smart Cache（L3） 36MB 30MB 30MB 24MB
Total L2 Cache 40MB 36MB 36MB 26MB
Intel TVB Up to 5.7GHz Up to 5.5GHz Up to 5.5GHz Up to 5.2GHz
Intel TBMT 3.0 Up to 5.6GHz Up to 5.5GHz Up to 5.5GHz -
P-Cores Max Turbo Up to 5.6GHz Up to 5.4GHz Up to 5.4GHz Up to 5.2GHz
E-Cores Max Turbo Up to 4.6GHz Up to 4.6GHz Up to 4.6GHz Up to 4.6GHz
P-Cores Base 3.7GHz 3.9GHz 3.9GHz 4.2GHz
E-Cores Base 3.2GHz 3.3GHz 3.3GHz 3.6GHz
Processor Base Power 125W 125W 125W 125W
Maximum Turbo Power 250W 250W 250W 159W

What is the all core boost for the 14900k ?

cebri1 · Friday at 7:34 AM

Henry swagger said:
What is the all core boost for the 14900k ?

5.7Ghz for the 8 pcores. Maybe the lack of HT is helping with frequencies.

Josh128 · Friday at 9:32 AM

cebri1 said:
285K、265K、265KF、245K 與 245KF 等 5 款 Intel Core Ultra 2 系列桌上型處理器規格定案 - BenchLife.info

Intel 新一代的 Core Ultra 2 桌上型處理器，應該值得期待一下。代號 Arrow […]

benchlife.info

Mmm... 285K all p-cores 5.6Ghz according to the latest leak.

Core Ultra 9 285K Core Ultra 7 275K Core Ultra 7 275KF Core Ultra 5 245K
Cores（P + E） 8 + 16 8 + 12 8 + 12 6 + 8
Threads 24 20 20 14
Intel Smart Cache（L3） 36MB 30MB 30MB 24MB
Total L2 Cache 40MB 36MB 36MB 26MB
Intel TVB Up to 5.7GHz Up to 5.5GHz Up to 5.5GHz Up to 5.2GHz
Intel TBMT 3.0 Up to 5.6GHz Up to 5.5GHz Up to 5.5GHz -
P-Cores Max Turbo Up to 5.6GHz Up to 5.4GHz Up to 5.4GHz Up to 5.2GHz
E-Cores Max Turbo Up to 4.6GHz Up to 4.6GHz Up to 4.6GHz Up to 4.6GHz
P-Cores Base 3.7GHz 3.9GHz 3.9GHz 4.2GHz
E-Cores Base 3.2GHz 3.3GHz 3.3GHz 3.6GHz
Processor Base Power 125W 125W 125W 125W
Maximum Turbo Power 250W 250W 250W 159W

This does not indicate "all P-core boost clocks" in MT workloads. It indicates max boost of 5.7 of low / single threaded apps with TVB, and max boost of 5.6 of low / single threaded apps when TVB requirements are not met or exhausted.

Hulk · Friday at 10:09 AM

cebri1 said:
285K、265K、265KF、245K 與 245KF 等 5 款 Intel Core Ultra 2 系列桌上型處理器規格定案 - BenchLife.info

Intel 新一代的 Core Ultra 2 桌上型處理器，應該值得期待一下。代號 Arrow […]

benchlife.info

Mmm... 285K all p-cores 5.6Ghz according to the latest leak.

Core Ultra 9 285K Core Ultra 7 275K Core Ultra 7 275KF Core Ultra 5 245K
Cores（P + E） 8 + 16 8 + 12 8 + 12 6 + 8
Threads 24 20 20 14
Intel Smart Cache（L3） 36MB 30MB 30MB 24MB
Total L2 Cache 40MB 36MB 36MB 26MB
Intel TVB Up to 5.7GHz Up to 5.5GHz Up to 5.5GHz Up to 5.2GHz
Intel TBMT 3.0 Up to 5.6GHz Up to 5.5GHz Up to 5.5GHz -
P-Cores Max Turbo Up to 5.6GHz Up to 5.4GHz Up to 5.4GHz Up to 5.2GHz
E-Cores Max Turbo Up to 4.6GHz Up to 4.6GHz Up to 4.6GHz Up to 4.6GHz
P-Cores Base 3.7GHz 3.9GHz 3.9GHz 4.2GHz
E-Cores Base 3.2GHz 3.3GHz 3.3GHz 3.6GHz
Processor Base Power 125W 125W 125W 125W
Maximum Turbo Power 250W 250W 250W 159W

If this information is correct it is quite interesting.

First, the base P/E frequencies have increased from 3.2/2.4 for the 14900K to 3.7/3.2 at 125W. We don't know what load Intel is applying but assuming it is the same between the two these numbers would imply a significant efficiency boost for ARL compared to Raptor Lake Refresh.

Furthermore, the base frequencies for the 13900K were 3/2.2, which are lower than the 14900K, indicating there was some process improvement from Raptor to Raptor Refresh. Finally, my 13900K on Auto BIOS settings when running CB R23 MT at 125W were 4.4/3.2. These are much higher than Intel's numbers indicating their test load is "heavier" than CB R23.

3.7/3=1.23, meaning ARL P cores look to be 23% more efficient with the Intel 125W test load. 3.2/2.2=1.45, so 45% better efficiency for the E cores compared to the 13900K.

Scaling up the ARL base numbers ...
3.7x 1.23=4.5, 3.2x1.45=4.6 This implies that ARL running CB R23 at 125W would be clocking 4.5/4.6. Seems high but promising.

Also it seems as though Intel as learned something about extremely high single/dual core boost frequencies. While is it exciting for the advertising department to be able to write "6.0GHz!!!" it is also useless from a performance point-of-view and requires cpu degrading frequencies to achieve.

Notice that Intel has specified the top frequency for the silicon as 5.7GHz. Period. Across the stack. In reality 5.7GHz will be hard to achieve because Intel has most likely set a pretty low temperature required for that boost frequency, so 5.6 is more likely the top actual frequency, which is very close to the 5.5GHz I find realistic for my 14900K, which coincidentally I'm running with HT off.

The point is that it looks like Intel has specified a max voltage they find acceptable for ARL and at that voltage the silicon can probably run at 5.6GHz up to 100C or whatever max they determine. But, if you can keep the chip cool enough, at that same voltage it CAN clock up to 5.7GHz. So it appears they're not just throwing volts at the silicon. I think there is some nuance here in Intel's decisions if the numbers provided are in fact legit.

In conclusion it looks like ARL may be measurably more efficient than Raptor Refresh and, since Intel has removed the ridiculously high 1/2 core boost frequencies (and the high voltages that go along with those frequencies) these chips may likely not have the degradation issue.

dullard · Friday at 10:33 AM

cebri1 said:
Core Ultra 9 285K Core Ultra 7 275K Core Ultra 7 275KF Core Ultra 5 245K
Cores（P + E） 8 + 16 8 + 12 8 + 12 6 + 8
Threads 24 20 20 14
Processor Base Power 125W 125W 125W 125W
Maximum Turbo Power 250W 250W 250W 159W

Wait, wait, wait. We have our own poster that claimed in many posts that it would be 297 W for PL2. We all know that Abwx is always correct.

Abwx said:
Guess that currently Intel is devising about the exact power they ll use since at 250W they will have a win only in CB R20/R23 for MT, and a marginal one.

They may well set 297W as PL2, or use an extended 297W PL4 since that s their apps performance profile for ARL, otherwise they ll lose in too much benches to be convincing, so we ll surely be good for another round of exagerated TDPs.

And likely that AMD limiting their CPU at 200W has somewhat disrupted their plans as they surely expected 230W, wich would had rendered their own 250W/297W TDP acceptable.

Abwx · Friday at 10:53 AM

dullard said:
Wait, wait, wait. We have our own poster that claimed in many posts that it would be 297 W for PL2. We all know that Abwx is always correct.

I was right that it would be at least 250W when the crowd was still hoping for a fairy taled 177W, and the 297W PL4 is stil not demonstrated as irrelevant.

Hail The Brain Slug · Friday at 10:53 AM

Hulk said:
If this information is correct it is quite interesting.

First, the base P/E frequencies have increased from 3.2/2.4 for the 14900K to 3.7/3.2 at 125W. We don't know what load Intel is applying but assuming it is the same between the two these numbers would imply a significant efficiency boost for ARL compared to Raptor Lake Refresh.

Furthermore, the base frequencies for the 13900K were 3/2.2, which are lower than the 14900K, indicating there was some process improvement from Raptor to Raptor Refresh. Finally, my 13900K on Auto BIOS settings when running CB R23 MT at 125W were 4.4/3.2. These are much higher than Intel's numbers indicating their test load is "heavier" than CB R23.

3.7/3=1.23, meaning ARL P cores look to be 23% more efficient with the Intel 125W test load. 3.2/2.2=1.45, so 45% better efficiency for the E cores compared to the 13900K.

Scaling up the ARL base numbers ...
3.7x 1.23=4.5, 3.2x1.45=4.6 This implies that ARL running CB R23 at 125W would be clocking 4.5/4.6. Seems high but promising.

Also it seems as though Intel as learned something about extremely high single/dual core boost frequencies. While is it exciting for the advertising department to be able to write "6.0GHz!!!" it is also useless from a performance point-of-view and requires cpu degrading frequencies to achieve.

Notice that Intel has specified the top frequency for the silicon as 5.7GHz. Period. Across the stack. In reality 5.7GHz will be hard to achieve because Intel has most likely set a pretty low temperature required for that boost frequency, so 5.6 is more likely the top actual frequency, which is very close to the 5.5GHz I find realistic for my 14900K, which coincidentally I'm running with HT off.

The point is that it looks like Intel has specified a max voltage they find acceptable for ARL and at that voltage the silicon can probably run at 5.6GHz up to 100C or whatever max they determine. But, if you can keep the chip cool enough, at that same voltage it CAN clock up to 5.7GHz. So it appears they're not just throwing volts at the silicon. I think there is some nuance here in Intel's decisions if the numbers provided are in fact legit.

In conclusion it looks like ARL may be measurably more efficient than Raptor Refresh and, since Intel has removed the ridiculously high 1/2 core boost frequencies (and the high voltages that go along with those frequencies) these chips may likely not have the degradation issue.

I don't know about the E cores, but no HT on the P cores would explain most or all of that increased baseclock.

Hitman928 · Friday at 10:57 AM

alcoholbob said:
eat the 7800X3D by a pretty substantial margin, maybe up to 10+%, if you run really fast RAM (i.e. 8000 MHz or higher). It’s just reviewers usually don’t bother to benchmark at these settings. But it’s something Intel users can do to juice their p

Source? Both TPU and HWUB showed very small gains going from DDR5-6000 to DDR5-7200 in gaming. I haven't seen data that shows RPL overtaking Zen4X3D with any memory speed in a fair comparison. But if you have a reviewer showing different, please share.

H433x0n said:
I wasn't referencing memory speeds but rather the quality settings used in benchmarks. If you run 1080P with maxed settings + RT, the 14900K will often outperform the 7800X3D whereas in reviewer benchmarks it'll show the opposite result - a good example would be Jedi Survivor & Plague Tale Requiem.

Link? HWUB was the only outlet I know that investigated CPU performance with RT on and off. Their CPU comparison ranking was virtually unchanged whether RT was on or off.

dullard · Friday at 11:02 AM

Hail The Brain Slug said:
I don't know about the E cores, but no HT on the P cores would explain most or all of that increased baseclock.

Thanks. People still don't seem to realize that HT has drawbacks. One of which is the extra transistor flipping and data movement uses more power and creates more heat. Meaning HT limits clock rates (not a lot, but some). And the efficiency improvements come in-part by no longer having to do the logic checks to keep threads apart on the same core for security reasons (you don't want one thread to be able to access even the cache from another thread).

csbin · Friday at 11:05 AM

cebri1 said:
285K、265K、265KF、245K 與 245KF 等 5 款 Intel Core Ultra 2 系列桌上型處理器規格定案 - BenchLife.info

Intel 新一代的 Core Ultra 2 桌上型處理器，應該值得期待一下。代號 Arrow […]

benchlife.info

Mmm... 285K all p-cores 5.6Ghz according to the latest leak.

Core Ultra 9 285K Core Ultra 7 275K Core Ultra 7 275KF Core Ultra 5 245K
Cores（P + E） 8 + 16 8 + 12 8 + 12 6 + 8
Threads 24 20 20 14
Intel Smart Cache（L3） 36MB 30MB 30MB 24MB
Total L2 Cache 40MB 36MB 36MB 26MB
Intel TVB Up to 5.7GHz Up to 5.5GHz Up to 5.5GHz Up to 5.2GHz
Intel TBMT 3.0 Up to 5.6GHz Up to 5.5GHz Up to 5.5GHz -
P-Cores Max Turbo Up to 5.6GHz Up to 5.4GHz Up to 5.4GHz Up to 5.2GHz
E-Cores Max Turbo Up to 4.6GHz Up to 4.6GHz Up to 4.6GHz Up to 4.6GHz
P-Cores Base 3.7GHz 3.9GHz 3.9GHz 4.2GHz
E-Cores Base 3.2GHz 3.3GHz 3.3GHz 3.6GHz
Processor Base Power 125W 125W 125W 125W
Maximum Turbo Power 250W 250W 250W 159W

P All Core 5.4

511 · Friday at 11:21 AM

Abwx said:
I was right that it would be at least 250W when the crowd was still hoping for a fairy taled 177W, and the 297W PL4 is stil not demonstrated as irrelevant.

You know PL4 is the max limit that is never exceeded Unlike Pl 1/2 also intel allowing it doesn't mean it gains anything cause it is not a intel node famous for their well High Power

maddie · Friday at 12:02 PM

dullard said:
Thanks. People still don't seem to realize that HT has drawbacks. One of which is the extra transistor flipping and data movement uses more power and creates more heat. Meaning HT limits clock rates (not a lot, but some). And the efficiency improvements come in-part by no longer having to do the logic checks to keep threads apart on the same core for security reasons (you don't want one thread to be able to access even the cache from another thread).

What are you on about?

On average, HT leads to increases in perf/W. Show me where it loses efficiency when averaged over a wide and varied code base.

dullard · Friday at 12:04 PM

maddie said:
What are you on about?

On average, HT leads to increases in perf/W. Show me where it loses efficiency when averaged over a wide and varied code base.

I didn't mention anything about efficiency on average of HT vs no-HT. Of course there are benefits to HT. There are cases where it increases performance per watt. I was mentioning the drawbacks that are real and often ignored--especially when one only talks about things "on average". Fact: HT increases the power used at a given frequency. There is no way to do any extra calculations without changing transistor states and moving data--thus it uses more power. Whether or not that increases performance/W is application dependent.

In other words: at a given power, turning off HT can let you increase frequency. At a given frequency, turning off HT can increase IPC if you no longer have so much security microcode and have more possible optimizations available. But, turning off HT might decrease performance.

Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads

Senior member

Attachments

Senior member

Golden Member

Senior member

Golden Member

Senior member

Diamond Member

Senior member

Golden Member

Golden Member

Senior member

Member

Senior member

Senior member

Senior member

Senior member

Diamond Member

Elite Member

Lifer

Diamond Member

Diamond Member

Elite Member

Senior member

Senior member

Diamond Member

Elite Member