Question Raptor Lake - Official Thread

Page 99 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
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.
 
  • Like
Reactions: vstar

Wolverine2349

Member
Oct 9, 2022
152
56
61
MLID is hardly a reliable source, especially when it comes to AMD related leaks. That said, the purpose of cache is to mitigate the latency and bandwidth penalties associated with accessing main memory. CPUs with slower main memory should theoretically benefit more from increased cache than a CPU with faster memory.

I know that Zen 3 3D didn't benefit from memory overclocking as much as standard Zen 3 did, because it had a much bigger L3 cache.

Similarly, Raptor Lake's ability to utilize much higher DDR5 memory speeds than Zen 4 3D should help to compensate somewhat for the smaller L3 cache.

DDR5 memory kits with 7ghz+ speeds are already going to be available later this month.


So would having less L3 cache say 16MB instead of 32MB be mitigated by faster RAM on Zen 3. Or is that different as Zen 3 is designed for 32MB cache where as adding 64MB on top is niche and specialized app and does not benefit in all cases?? But 32MB is required for good performance of Zen 3 regardless of memory speed with faster memory just improving performance but if L3 cache was cut in half performance would be crippled no matter how fast you could tune the system RAM?? But different with 96MB as it is a niche use case??

I have also heard that the 5800X3D and maybe the 3D versions in general the extra cache provides some type of latency penalty or cuts some bandwidth in half of some sort which can hurt performance in some things even at same clock speeds despite it being good for most games?? Is that true with cache stacking in general and how exactly does that work?? And would it be the same on Zen 4 3D cache versions???
 
Last edited:

Wolverine2349

Member
Oct 9, 2022
152
56
61
I know no one is interested in it for the IGP, but...
DDR5-5200

Stock
3gj8YbL.png

Dv3rjCM.png


2100mhz (the max allowed in the bios)
UqcvpVD.png

P5Z9bTg.png


Its not that bad of a result, definately better than what AMD has on Ryzen 7000X. DDR4-3600 was just 1fps less. It is around 3200G stock performance i belive for Shadow, and W3 is around 5600G perf.
I dont know whats up with W3 1C usage


Is that the 13900K playing Witcher 3 with no discrete GOU and using the integrated graphics. And better than AMD Ryzen 7000? DO you mean Ryzen 7000 with their integrated GPU??

The real test are these CPUs with a 3090 Ti or even better a 4090.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
Is it memory speed or latency that matters when applied to V-cache improvements in performance?

Not sure I understand what you are asking here as the question is phrased strangely....

Faster memory generally do not improve latency, as faster has higher C#, leading to little to none absolute improvements in latency.

Higher frequency memory does improve both latency and bandwidth. CAS latency only affects memory latency when the memory has to activate a new column. If I remember correctly, CAS stands for column access strobe. Also, DDR5 has other tricks up it's sleeve to reduce effective latency. DDR5's evolution will likely see it max out at around the 8.4ghz for JEDEC standard modules, while much higher for overclocked ones. At those speeds, the effective latency is below 10ns even with a CAS latency of 40.

40/8400x2000= 9.5ns
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
So would having less L3 cache say 16MB instead of 32MB be mitigated by faster RAM on Zen 3

Absolutely. But there are limitations due to how far away the system memory is from the CPU.

Or is that different as Zen 3 is designed for 32MB cache where as adding 64MB on top is niche and specialized app and does not benefit in all cases?? But 32MB is required for good performance of Zen 3 regardless of memory speed with faster memory just improving performance but if L3 cache was cut in half performance would be crippled no matter how fast you could tune the system RAM?? But different with 96MB as it is a niche use case??

You're overthinking it. Cache is SRAM memory, which is much faster and less capacity than DRAM.

Also cache is built directly next to the CPU cores themselves which drastically reduces access latency.

Even with hyper fast system memory, cache will always have lower latency because of it's proximity to the CPU.

I have also heard that the 5800X3D and maybe the 3D versions in general the extra cache provides some type of latency penalty or cuts some bandwidth in half of some sort which can hurt performance in some things even at same clock speeds despite it being good for most games?? Is that true with cache stacking in general and how exactly does that work?? And would it be the same on Zen 4 3D cache versions???

Increasing the cache capacity does increase latency as it takes longer to access the data, but it's still far less latency than accessing system memory.

That's precisely why V-cache provides such a significant increase in game performance. Anything that gets data to the CPU faster is going to increase performance.
 
  • Like
Reactions: Kaluan and moinmoin

maddie

Diamond Member
Jul 18, 2010
4,723
4,628
136
Not sure I understand what you are asking here as the question is phrased strangely....



Higher frequency memory does improve both latency and bandwidth. CAS latency only affects memory latency when the memory has to activate a new column. If I remember correctly, CAS stands for column access strobe. Also, DDR5 has other tricks up it's sleeve to reduce effective latency. DDR5's evolution will likely see it max out at around the 8.4ghz for JEDEC standard modules, while much higher for overclocked ones. At those speeds, the effective latency is below 10ns even with a CAS latency of 40.

40/8400x2000= 9.5ns
The ns latency has remained fairly constant as you show and which I clumsily stated. This is what I mean when I said DDR5 does not lower access times by being faster. In programs with a lot of small data package access needed, V-cache will give lots of benefits even though DDR5 has higher bandwidth. That is my claim.
 

Henry swagger

Senior member
Feb 9, 2022
356
235
86

DrMrLordX

Lifer
Apr 27, 2000
21,583
10,785
136
40k at 290 watts.. dvlr at work 🤔

292W stock? Ew.
 
  • Like
Reactions: Kaluan

nicalandia

Diamond Member
Jan 10, 2019
3,330
5,281
136
Some people paid around $630 for the pre-orders. There is a chance that the i9-13900K/KF SKUs may get scalped hard, coz 4090 gamers will want the best gaming CPU and a lot of them are bent towards Intel.
Who is to say that the 13900K will be the best for 4090? Does CPU matters at 4K? For Bragging rights at 480p and 720p reviews on Anandtech?
 
Jul 27, 2020
15,761
9,830
106
Who is to say that the 13900K will be the best for 4090? Does CPU matters at 4K? For Bragging rights at 480p and 720p reviews on Anandtech?
1440p matters to a lot, especially those with gaming monitors with 120 Hz and above.

Zen 4 sales are not good. It's possible that a lot of people looking to upgrade are waiting to see what Intel has to offer.
 
  • Like
Reactions: Henry swagger

nicalandia

Diamond Member
Jan 10, 2019
3,330
5,281
136
1440p matters to a lot, especially those with gaming monitors with 120 Hz and above.
1440p Really? So one has the funds to build a $3000 Gaming CPU so you can get a skimp on a 1440p monitor? Do you really need a 4090 to play games at that resolution? If you have the budget to get a a High End CPU like 13900K/7950X Fast DDR5 and a 4090 GPU, you have the $ to get a Top of the line 4K Gaming Monitor
 
  • Like
Reactions: Kaluan
Jul 27, 2020
15,761
9,830
106
1440p Really? So one has the funds to build a $3000 Gaming CPU so you can get a 1440p High Hz monitor?
Funds can be limited. Not a lot of people have the funds to buy the best of every hardware component. Their build specifications may be very disproportionate and unbalanced, paying particular attention to the CPU and GPU but skimping everywhere else.
 

deasd

Senior member
Dec 31, 2013
513
724
136
Zen 4 sales are not good. It's possible that a lot of people looking to upgrade are waiting to see what Intel has to offer.

Because so much gamers burst out to buy 5800X3D after Zen4 review, due to FPS/price ratio. Nothing from Intel would change that..... only the next Zen4 X3D series would worth waiting.
 

Wolverine2349

Member
Oct 9, 2022
152
56
61
Funds can be limited. Not a lot of people have the funds to buy the best of every hardware component. Their build specifications may be very disproportionate and unbalanced, paying particular attention to the CPU and GPU but skimping everywhere else.


Plus some would rather do 1440P especially if they only have a 27 to 32 inch monitor where 4K makes no difference in visual quality.
 
  • Like
Reactions: igor_kavinski

Wolverine2349

Member
Oct 9, 2022
152
56
61
Some people paid around $630 for the pre-orders. There is a chance that the i9-13900K/KF SKUs may get scalped hard, coz 4090 gamers will want the best gaming CPU and a lot of them are bent towards Intel.


That would really be horrible. I thought the shortage of consumer grade computer parts was over unlike other things.

Well the RTX 4090 did sell out day 1, but they were available at local MIcro Center throughout the day but were gone at end of the day. Far cry from the days were there was nothing and on launch day of RTX 3080 and 3090, they were gone in seconds and none in stock to be found at the end. I also think cards selling out on day 1 is normal just not in seconds, but throughout the day.
 

nicalandia

Diamond Member
Jan 10, 2019
3,330
5,281
136
People on Twitter. Hey look how 4090 was out of stock within minutes of release and then saying why Ryzen 7000 can't do that? Well for one you can put that GPU on all range of Motherboards released in the past 6 years and play with it without any issues. If Ryzen 7000 could be just put on a AM5 you bet it would have been sold out too.
 
  • Like
Reactions: lightmanek

Harry_Wild

Senior member
Dec 14, 2012
830
150
106
Who is to say that the 13900K will be the best for 4090? Does CPU matters at 4K? For Bragging rights at 480p and 720p reviews on Anandtech?
The bandwidth matters aka PCIe. PCIe 5 is like warpspeed compare to previous PCIe versions like 3 and earlier! Of course no PCIe 5 graphics cards yet! Only PCIe 4s!