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.

 

[RTX2080]

OH NO,,,,,,,,sounds like Raptor i3/i5 are rebranded Alder

https://twitter.com/x/status/1547357410104586240

 

[gdansk]

OH NO,,,,,,,,sounds like Raptor i3/i5 are rebranded Alder

https://twitter.com/x/status/1547357410104586240

Is there much difference at the low end? How much better would a Raptor Lake 6+0 die be? More L2 cache per core, more L3 and better memory support. The extra cores don't apply here. If the 13400(F) is clocked like a 12600 it could be good value if priced like the 12400(F).

 

[Henry swagger]

Raptor lake i7 and i5 will have 16 and 14 cores.. intel is blessing people with multi-core beats 😁🏅

 

[inf64]

Some dudes said 7600X ≈ 5800X, no mention of clocks and power. You can re-calculate it. but since this is intel thread I won't talk too much. sound like the impact is tremendous. Also rumor.

It's pretty much in line with what we have seen so far:

5600X - 100%
5800X - 129%
Source : https://www.computerbase.de/thema/p...bschnitt_der_beste_prozessor_fuer_anwendungen

7600X : +10% IPC and ~5.0-5.1Ghz all core boost
1.1 x 5.1 / 4.3 = 1.3 or 130%

 

[Thunder 57]

Raptor lake i7 and i5 will have 16 and 14 cores.. intel is blessing people with multi-core beats 😁🏅

In another thread you posted this:

I like amd products i like both intel and amd.. where did you get the impression i didnt like amd ?

Well there ya go, You clearly don't like AMD. I get the impression that you dislike AMD. Intel "blessing" us? You can't make this stuff up.

Remember when Intel "blessed people" with more cores between 2011-2017? . Yup, neither does anyone who was around for that. "Blessing people with multi-core beasts(?)"". I thought it was team blue who made the whole "moar cores" thing. Please just go away and stop posting nonsense here or try to have a proper conversation.

 

[Exist50]

OH NO,,,,,,,,sounds like Raptor i3/i5 are rebranded Alder

https://twitter.com/x/status/1547357410104586240

OP doesn't seem to clarify where the cut is. So I guess the question is if the 8+16 is the only Alder Lake die, or if they have a mid-tier one and reuse the 6+0 for lower end. There've been mentions of a 6+8 die previously, but I don't think we've seen actual leaks for it yet.

 

[DrMrLordX]

We need bigger radiators!

The MO-RA3 420 is ready to go.

Funny that we know more about Raptor Lake than about Zen4.

Uh, maybe because Raptor Lake is just an overclocked Alder Lake with some extra e cores?

 

[jpiniero]

OP doesn't seem to clarify where the cut is. So I guess the question is if the 8+16 is the only Alder Lake die, or if they have a mid-tier one and reuse the 6+0 for lower end. There've been mentions of a 6+8 die previously, but I don't think we've seen actual leaks for it yet.

6+8 was just speculation for the small die. No way they would do three different cuts for either desktop or laptop. You would still need to make SKUs to take the busted Raptor Lake bigger dies.

For desktop maybe mix some cut 6+4/6+0 Raptor Lake with some 6+0 Alder Lake in the locked i5's but leave the big volume part (13400/F) and below be Alder Lake.

Obviously the reason they would do it is yields/cost. I will say I am a bit skeptical because it does complicate things.

 

[Exist50]

6+8 was just speculation for the small die

Well it's not just pure speculation. One of the SKU leaks had coloring consistent with a new 6+8 die being used across the i5 lineup. And that's really the only way that the non-K i5s can see a bump to core count. They're not going to use the 8+16, or even the 8+8, for their big volume driver. Too much wasted silicon.

No way they would do three different cuts for either desktop or laptop. You would still need to make SKUs to take the busted Raptor Lake bigger dies.

Intel's done quite a few dies before, so this would be far from unusual. Really, that's just 2 Raptor Lake desktop dies + maybe 1 mobile? It's less than they're doing for Alder Lake.

Also, they've used multiple dies for some SKUs before, so I don't think that's a problem.

I guess we'll see what they come out with, but I don't think a 6+8 die for the i5 volume would be unreasonable to expect.

 

[coercitiv]

Brace for impact:

Intel plans price hikes on broad range of products

Intel has informed customers it will raise prices on a majority of its microprocessors and peripheral chip products later this year, citing rising costs, Nikkei Asia learned.

The biggest U.S. chipmaker plans to raise prices this autumn on flagship products such as central processing units for servers and computers, where it dominates the market, as well as on a wide range of other items, including chips for Wi-Fi and other connectivity, according to three industry executives with direct knowledge.

My Alder Lake upgrade is looking better and better each day.

 

[jpiniero]

They're not going to use the 8+16, or even the 8+8, for their big volume driver. Too much wasted silicon.

They are doing it now. The 12400 comes from both dies.

 

[Henry swagger]

In another thread you posted this:



Well there ya go, You clearly don't like AMD. I get the impression that you dislike AMD. Intel "blessing" us? You can't make this stuff up.

Remember when Intel "blessed people" with more cores between 2011-2017? . Yup, neither does anyone who was around for that. "Blessing people with multi-core beasts(?)"". I thought it was team blue who made the whole "moar cores" thing. Please just go away and stop posting nonsense here or try to have a proper conversation.

I didn't mention amd in my post and you post this bs... d'not star ish when the is none... you sound like a fanboy i like both companies because competition is good

 

[FangBLade]

I didn't mention amd in my post and you post this bs... d'not star ish when the is none... you sound like a fanboy i like both companies because competition is good

No, you don't like AMD, i've recognized you and seen your posts in other forums, you changed style a little bit, but fanboy is a fanboy, it is just matter of time you start trolling again.

 

[dullard]

you changed style a little bit, but ___ is a ___, it is just matter of time you start trolling again.

Please focus on the discussion of Raptor Lake. And also, please read the rules: http://www.portvapes.co.uk/?id=Latest-exam-1Z0-876-Dumps&exid=thread...f-current-rules-rules-updated-5-4-20.2332179/

 

[dullard]

Remember when Intel "blessed people" with more cores between 2011-2017? . Yup, neither does anyone who was around for that. "Blessing people with multi-core beasts(?)"". I thought it was team blue who made the whole "moar cores" thing.

Note: this table is for HEDT chips, but to just make a blanket statement that only one team added cores in that time frame is pretty far from the truth.

Bloomfield	4​	Nov-08​
Gulftown	6​	Mar-10​
Haswell	8​	Aug-14​
Broadwell	10​	30-May-16​
Skylake	18​	25-Sep-17​

 

[eek2121]

Hopefully power limits will be lower. I am curious as to what happens when we hit the point where a standard household electrical outlet (12a 110v for 15a circuits in the US) is saturated. Between NVIDIA and Intel, things are getting quite out of hand.

 

[nicalandia]

Cinebench r23
5950X 1640ST 25800MT (bone stock on average)
7950X 1890ST 34800MT (15% faster in ST and 35% faster in MT vs 5950X)
13900K 2200ST 37200MT
AMD is losing while having node advantage

Whatever MT lead 12900K has over 5950X(in many cases there is no lead) it vanishes on long running task. In Cinebench R23 MT it actually loses to it after 10 minutes render time

 

[SmokSmog]

I certainly agree that the golden cove cores are quite powerful. But not good at the electrical power they use.

They are more efficient than Zen3 cores at the same clocks and when both at their minimum voltages.

For example I did test 12600K with only 6P cores enabled at 4ghz and 0,835Vcore and compared it with someone's 5600X at 4ghz 0,95V.

The result
6P 4ghz 0,835Vcore 49W package power 12425MT CBR23
6 Zen3 cores 4ghz 0,95Vcore 56W package power 10200MT CB R23

To even match my score with 5600X you would need 4,7ghz+ all core oc and 100W package power

 

[lobz]

many PC folks think geekbench is bad but for bursty/single core tasks its a very GOOD indicator. Not real world but it gives a base line to compare with other chips.

Well, for example, a massive improvement in GB ST score for Rocket Lake did wonders for its gaming performance! Sometimes it was even slower in games than its predecessors 🤣 I'm citing gaming here, because before its launch, it was praised as the new gaming king, based on Intel's marketing and the leaked ST improvement in GB.

 

[Markfw]

They are more efficient than Zen3 cores at the same clocks and when both at their minimum voltages.

For example I did test 12600K with only 6P cores enabled at 4ghz and 0,835Vcore and compared it with someone's 5600X at 4ghz 0,95V.

The result
6P 4ghz 0,835Vcore 49W package power 12425MT CBR23
6 Zen3 cores 4ghz 0,95Vcore 56W package power 10200MT CB R23

To even match my score with 5600X you would need 4,7ghz+ all core oc and 100W package power

I am talking about DEFAULT everything. And at 100% load on all cores. They take more. See the reviews.

 

[Det0x]

They are more efficient than Zen3 cores at the same clocks and when both at their minimum voltages.

For example I did test 12600K with only 6P cores enabled at 4ghz and 0,835Vcore and compared it with someone's 5600X at 4ghz 0,95V.

The result
6P 4ghz 0,835Vcore 49W package power 12425MT CBR23
6 Zen3 cores 4ghz 0,95Vcore 56W package power 10200MT CB R23

To even match my score with 5600X you would need 4,7ghz+ all core oc and 100W package power

Your 6core Zen3 example above is way overdriven and overvolted for maximum efficiently.

Here are my 5950x @ 51w PPT -> 20681 points in Cinebench r23

You cant match this kind of performance/watt with Alder Lake

 

[Hulk]

I certainly agree that the golden cove cores are quite powerful. But not good at the electrical power they use.

Hmm.

Core-for-Core, Golden Cove vs Zen 3 at the same reasonable frequency of 4GHz how do they compare in terms of performance and power?

I have no doubt a 16 core Zen 3 clocked at 4GHz is more efficient than a 12900K clocked within an inch of it's life for equal performance. But that's simply due to transistors/cores.

Not sure about a heads up Zen 3 vs GC comparison of performance and efficiency though.

 

[pakotlar]

Whatever MT lead 12900K has over 5950X(in many cases there is no lead) it vanishes on long running task. In Cinebench R23 MT it actually loses to it after 10 minutes render time

They are more efficient than Zen3 cores at the same clocks and when both at their minimum voltages.

For example I did test 12600K with only 6P cores enabled at 4ghz and 0,835Vcore and compared it with someone's 5600X at 4ghz 0,95V.

The result
6P 4ghz 0,835Vcore 49W package power 12425MT CBR23
6 Zen3 cores 4ghz 0,95Vcore 56W package power 10200MT CB R23

To even match my score with 5600X you would need 4,7ghz+ all core oc and 100W package power

Chips and Cheese suggests Golden Cove can be more efficient, but it’s workload dependent. It does tend to scale better with voltage, so can reach higher performance targets.

 

[Markfw]

Hmm.

Core-for-Core, Golden Cove vs Zen 3 at the same reasonable frequency of 4GHz how do they compare in terms of performance and power?

I have no doubt a 16 core Zen 3 clocked at 4GHz is more efficient than a 12900K clocked within an inch of it's life for equal performance. But that's simply due to transistors/cores.

Not sure about a heads up Zen 3 vs GC comparison of performance and efficiency though.

See post 770. If you mess with stock settings, you could loose performance or overdo it.

Edit: And as for the post above me, the same applies.

 

[nicalandia]

Chips and Cheese suggests Golden Cove can be more efficient, but it’s workload dependent. It does tend to scale better with voltage, so can reach higher performance targets.

If the 12900K loses MT performance the longer the Render it goes, you can bet Rocket lake will lose it sooner than Alder Lake.