Discussion Intel current and future Lakes & Rapids thread

Page 516 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

mikk

Diamond Member
May 15, 2012
4,168
2,204
136
I have doubts of any real IPC gain. It's going to be big enough as it is with the extra small cores. I'm beginning to think the "Improved Performance" is the small cores and that's it. Maybe enabling AVX-512. I don't know how much L3 they can add, that can't be much but you could say it's better.


Much better CPU cache for gamers doesn't sound trivial to me.
 

Hulk

Diamond Member
Oct 9, 1999
4,351
2,215
136

Observe this post in particular:




Golden Cove is already massive area-wise, and it's been on the drawing board for a long time. You would think they would have already worked it over multiple times to squeeze out more performance.

True. But at some point you have to "lock in" a design to move it along to actual being produced. When that happens, as new ideas come along they have to "save them" for the next revision. This goes for any item heading from drawing board to production.
 

dullard

Elite Member
May 21, 2001
25,165
3,586
126
Compare that to the 5950x and get back to me. Perf/watt matters. That is what I mean by Intel being behind.
You said "Intel 14nm CPUs are severely gimped if constrained to 125W". I responded that actually that Intel processors are not gimped when limited in power. And now your response is that AMD is better? How does that even remotely have anything to do with your comment or my reply? The discussion is about PL2 power levels of processors.

Yes, the 5950x is a great chip and yes, it is far better than the equivalent that Intel sells right now in many use cases. But you still are flat out incorrect about "Intel 14nm CPUs are severely gimped if constrained to 125W". If you want to correct your statement, then delete the "if constrained to 125W" part. You could just leave it at "Intel 14nm CPUs are severely gimped" when you need more cores.
 

Abwx

Lifer
Apr 2, 2011
11,143
3,840
136
Last edited:

Thunder 57

Platinum Member
Aug 19, 2007
2,791
4,066
136
If the top end SKU is priced like a 5950X then it means that it doesnt perform better, otherwise you could had expected a 20% premium for 5% better perf..

I don't know what performance might be. Those could be "scalper" prices from a store. They are just higher than I expected.
 

Hitman928

Diamond Member
Apr 15, 2012
5,526
8,593
136
You said "Intel 14nm CPUs are severely gimped if constrained to 125W". I responded that actually that Intel processors are not gimped when limited in power. And now your response is that AMD is better? How does that even remotely have anything to do with your comment or my reply? The discussion is about PL2 power levels of processors.

Yes, the 5950x is a great chip and yes, it is far better than the equivalent that Intel sells right now in many use cases. But you still are flat out incorrect about "Intel 14nm CPUs are severely gimped if constrained to 125W". If you want to correct your statement, then delete the "if constrained to 125W" part. You could just leave it at "Intel 14nm CPUs are severely gimped" when you need more cores.

Something is wrong with Tom's "power limit enforced" 11700k result. For some reason their power limit enforced shows higher power usage than their power limit unlocked result, both of which are breaking the Intel power 'spec' for the CPU.





Gamers Nexus showed the difference when power limits are properly enforced, the CPU without power limits will be ~15% faster (edit: at 70+ W higher). That's just the 11700k. Some motherboards will allow even higher power usage for increased performance.

 
Last edited:

dullard

Elite Member
May 21, 2001
25,165
3,586
126
Ah, the expected doom-posting from news of pre-release prices
Yes. It always happens this way. Some small company, in a foreign currency, with VAT, giving a pre-release price list. They never quite translate into US prices for some reason. I just can't fathom why not.
 

coercitiv

Diamond Member
Jan 24, 2014
6,341
12,598
136
Something is wrong with Tom's "power limit enforced" 11700k result. For some reason their power limit enforced shows higher power usage than their power limit unlocked result, both of which are breaking the Intel power 'spec' for the CPU.
At first I thought the same, until I started reading the review piece associated to those graphs:
motherboard vendors are free to exceed those recommendations, even at stock settings, to differentiate their motherboards. As such, performance has long varied by motherboard based on the respective power settings. Our standard policy is to allow the motherboard to exceed Intel's recommended power limits, provided the chip remains within warrantied operating conditions.
As such, be aware that our stock settings reflect performance with lifted power limits to reflect the experience most enthusiasts will encounter, provided they top the chip with a capable cooler.

In other words, the CPU was operating at motherboard stock, not Intel stock.
 

Hitman928

Diamond Member
Apr 15, 2012
5,526
8,593
136
At first I thought the same, until I started reading the review piece associated to those graphs:



In other words, the CPU was operating at motherboard stock, not Intel stock.

Still doesn't explain why the "motherboard limited" (which doesn't mean anything because motherboards can essentially set whatever limit they want) has higher power consumption than the "unlimited" results. They really need to rethink their setup, or at least their naming convention.
 

Thunder 57

Platinum Member
Aug 19, 2007
2,791
4,066
136
This is a future lakes thread, right? I never said it was accurate, but hey, people speculate. What fun would it be if we couldn't? I just posted a link with speculated prices and commented on them. Don't be a buzzkill.
 

Abwx

Lifer
Apr 2, 2011
11,143
3,840
136
Something is wrong with Tom's "power limit enforced" 11700k result. For some reason their power limit enforced shows higher power usage than their power limit unlocked result, both of which are breaking the Intel power 'spec' for the CPU.

View attachment 49665

View attachment 49666

Gamers Nexus showed the difference when power limits are properly enforced, the CPU without power limits will be ~15% faster (edit: at 70+ W higher). That's just the 11700k. Some motherboards will allow even higher power usage for increased performance.

View attachment 49667

In Cinebench R20 about 180W without AB (adaptative boost) and 230W when set on, that s for a 11900K.
In the AVX loading test package power is 270W and 319W with AB on.


Methink that THG dont know what they are talking about...
 

ModEl4

Member
Oct 14, 2019
71
33
61
Regarding 1K unit pricing, if there is an increase, I don't expect the 12000K series to be more than $25 expensive in relation with the 11000K series (maybe $35 for the 12900K) and the KF series -$25 vs K series like in the previous gen (although with the VGA increase-even in the entry models- there was an opportunity for Intel to increase the difference to $30)
And the retail pricing in Europe at least, should be close if there is availability, Mindfactory for example was selling the 11000K series even below than 1K pricing (except 11900K/KF which had high demand or should i say low availability, although the last days is priced at 519€ including VAT...) so max 1K unit price difference in worst case scenario imo:
12900K vs 11900K $574-564 vs $539
12700K vs 11700K $424 vs $399
12600K vs 11600K $297 vs $262
This mean that the KF series which is the direct competition for the upcoming Zen3+cache CPUs will be at max:
$549-539 12900KF
$399 12700KF
$262 12600KF
Even if the new 12 core Zen3+ is $599 instead of $549, then for me it will be better value than 12900KF because it will be around 10% faster in both single thread and in multithreading, while the 8 core Zen3+ will be slightly slower in multithreading while costing $50 more in relation with 12700KF (although probably will be 10% faster in single thread) and the 6 core Zen3+ clearly slower than the 12600KF in multithreading but also more than 10% faster in single thread) but still at least 14% more expensive...
The easy choice for gamers will be the new 6 core Zen3+ at $299, for everyone else that is a casual gamer and do mostly work, the better value regarding performance/CPU price will be Intel 12000 series (except 12900K/KF maybe) and regarding the possible price increase, look at it this way: Even if AMD does not increase any price, it will offer you 15% more performance at the same price as Zen3, while Intel will offer probably up to +57% / +38% / +40% more multithreading performance in 12900K/12700K/12600K vs last gen. Also with the above assumptions a $249 5600X does not make sense for a new windows 11 system, 12600KF will be just 5% more expensive while offering at least comparable gaming and single thread performance but will be up to +20% faster in multithreading applications like rendering (performance differences assumed with windows 11 testbed) Although if Alder Lake needs DDR5 to show these performance differences and if you take account that the new Intel chipsets will be more expensive than Z590/B560 etc. while AMD won't need new chipsets, then the tables are turning in favour of AMD regarding performance/platform cost...
 
Reactions: Tlh97

coercitiv

Diamond Member
Jan 24, 2014
6,341
12,598
136
No non-K at initial release?
AFAIK they're doing a limited release, the rest is coming in 2022.

Still doesn't explain why the "motherboard limited" (which doesn't mean anything because motherboards can essentially set whatever limit they want) has higher power consumption than the "unlimited" results. They really need to rethink their setup, or at least their naming convention.
I'm not defending Tom's here, that portion of the review is all sorts of weird (at best). My point was we shouldn't care at this point, since the only reason it was brought into discussion turned out to be irrelevant: they did not measure actual stock performance.
 

DrMrLordX

Lifer
Apr 27, 2000
21,768
11,088
136
True. But at some point you have to "lock in" a design to move it along to actual being produced.

Golden Cove - or some variant thereof - has been sitting on Intel's shelves for up to two years waiting for a suitable process. Not having participated in the development of Golden Cove, I can't tell you for sure exactly when it would first have been considered ready for early test silicon. But even if we accept that Cannonlake launching in late 2017 was normal and acceptable (it wasn't), had 10nm Cannonlake not been a disaster, Intel should have had:

Sunny Cove/IceLake: 2018
Willow Cove/TigerLake: 2019
Golden Cove/???: 2020, maybe 2021 (no, I don't think it was initially Intel's plan to launch anything like Alder Lake, at least not so soon)

Had Intel been hitting on all cylinders, Cannonlake would have launched instead of KabyLake, and Golden Cove should have launched (possibly on 7nm?) in 2019 or early 2020. The design teams should have been able to keep up with that cadence had the process tech been available for them to act.

Or I'm completely off my rocker, and everything beyond Sunny Cove/IceLake is a mess they cobbled together after it was obvious that 10nm wasn't working out, in which case Golden Cove may be more of a last-minute hack job than an organized plan that's been waiting for execution since 2019.
 

eek2121

Diamond Member
Aug 2, 2005
3,027
4,215
136
Regarding 1K unit pricing, if there is an increase, I don't expect the 12000K series to be more than $25 expensive in relation with the 11000K series (maybe $35 for the 12900K) and the KF series -$25 vs K series like in the previous gen (although with the VGA increase-even in the entry models- there was an opportunity for Intel to increase the difference to $30)
And the retail pricing in Europe at least, should be close if there is availability, Mindfactory for example was selling the 11000K series even below than 1K pricing (except 11900K/KF which had high demand or should i say low availability, although the last days is priced at 519€ including VAT...) so max 1K unit price difference in worst case scenario imo:
12900K vs 11900K $574-564 vs $539
12700K vs 11700K $424 vs $399
12600K vs 11600K $297 vs $262
This mean that the KF series which is the direct competition for the upcoming Zen3+cache CPUs will be at max:
$549-539 12900KF
$399 12700KF
$262 12600KF
Even if the new 12 core Zen3+ is $599 instead of $549, then for me it will be better value than 12900KF because it will be around 10% faster in both single thread and in multithreading, while the 8 core Zen3+ will be slightly slower in multithreading while costing $50 more in relation with 12700KF (although probably will be 10% faster in single thread) and the 6 core Zen3+ clearly slower than the 12600KF in multithreading but also more than 10% faster in single thread) but still at least 14% more expensive...
The easy choice for gamers will be the new 6 core Zen3+ at $299, for everyone else that is a casual gamer and do mostly work, the better value regarding performance/CPU price will be Intel 12000 series (except 12900K/KF maybe) and regarding the possible price increase, look at it this way: Even if AMD does not increase any price, it will offer you 15% more performance at the same price as Zen3, while Intel will offer probably up to +57% / +38% / +40% more multithreading performance in 12900K/12700K/12600K vs last gen. Also with the above assumptions a $249 5600X does not make sense for a new windows 11 system, 12600KF will be just 5% more expensive while offering at least comparable gaming and single thread performance but will be up to +20% faster in multithreading applications like rendering (performance differences assumed with windows 11 testbed) Although if Alder Lake needs DDR5 to show these performance differences and if you take account that the new Intel chipsets will be more expensive than Z590/B560 etc. while AMD won't need new chipsets, then the tables are turning in favour of AMD regarding performance/platform cost...

My guess is around $588 for the 12900k in the US.
 

ModEl4

Member
Oct 14, 2019
71
33
61
My guess is around $588 for the 12900k in the US.
maybe, who knows.
I guess if the launch is around windows 11 launch, many people working in the IT sector (intel distributors, OEMs working with Intel, etc...) will have knowledge at least 2 weeks before launch, maybe a month before, so they will know a lot earlier than us, also there could be a leak even earlier, i'm sure in the coming weeks will see new informations regarding prices.
 

Exist50

Platinum Member
Aug 18, 2016
2,452
3,101
136
There is a big core IPC increase with Raptor Cove and improvements in performance/watt. Intels says it has a much better CPU cache capability for gamers.

It's a stopgap gen. I think expecting anything more than <5% IPC, 100-300MHz peak, and some more cache is just asking to be disappointed.
 
Reactions: Tlh97 and uzzi38

eek2121

Diamond Member
Aug 2, 2005
3,027
4,215
136
It's a stopgap gen. I think expecting anything more than <5% IPC, 100-300MHz peak, and some more cache is just asking to be disappointed.

I agree. ADL-S will be decent for sure, but Raptor Lake is the architecture to keep an eye on.
 

RanFodar

Junior Member
May 27, 2021
19
17
51
It's a stopgap gen. I think expecting anything more than <5% IPC, 100-300MHz peak, and some more cache is just asking to be disappointed.

Meh. I would ask for a ~10% IPC increase, but that's fine. There's also the doubling of Gracemont cores, so there's that.
 

mikk

Diamond Member
May 15, 2012
4,168
2,204
136
The text says "Improved CPU Cache for Gaming" which if Intel added an additional 6 MB would fit.


I'm not talking about the text of this slide, I'm talking about Intels Raptor Lake comment which is a bit more telling than the text.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |