• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Discussion Intel current and future Lakes & Rapids thread

Page 755 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
I really cannot see Intel having RPL-R take on Zen 5 & Zen 5 V-Cache if that's true (though DigiTimes AFAIK have been inaccurate at times with MTL partially having TSMC N3), because if they can't get MTL-S out then they have to get ARL-S out ASAP even if it's just a less ambitious core count. They can't afford to mess that up.
 
Alder Lake is 10+++.

Yes, I totally forgot about

Cannon Lake 10nm 3.2GHz
Ice Lake 10+ 4.1GHz
Tiger Lake 10++ 5GHz
Alder Lake 10+++ 5.5GHz
Raptor Lake 10++++ 6GHz
Raptor Lake Refresh 10+++++

Now we're back in the "+" business again!

What frequency improvements could RLR bring?
 
Was
It being mobile focused is nothing to do with it being their first attempt at chiplets.

MTL/RWC just isn't good enough to be worth bringing to desktop. ST perf gains just ain't there against desktop RPL.

EDIT: Also I have no clue what the whole CPU+SoC+IO die thing is all about. That's just nonsense.
I have heard this too. Lion cove in arrow lake is the next desktop platform.
 
Yes, I totally forgot about

Cannon Lake 10nm 3.2GHz
Ice Lake 10+ 4.1GHz
Tiger Lake 10++ 5GHz
Alder Lake 10+++ 5.5GHz
Raptor Lake 10++++ 6GHz
Raptor Lake Refresh 10+++++

Now we're back in the "+" business again!

What frequency improvements could RLR bring?
no more with Arrow Lake desktop, Intel will go to TSMC 3nm according to rumours.
 
Intel Xeon W9-3495X Reviews are going to be released today at 9 a.m PT make sure to tune to STH, PugetSystem, Phoronix and a few others, there arent going to be many
 
Sapphire rapids seems like an utter failure, cant even beat Zen3 in nether ST or MT performance, while sucking down much more power ?

1677079587282.png
1677079489833.png1677079521537.png1677079536364.png1677079548248.png1677079559193.png1677079756234.png

1677079764500.png
 
Last edited:
I really cannot see Intel having RPL-R take on Zen 5 & Zen 5 V-Cache if that's true (though DigiTimes AFAIK have been inaccurate at times with MTL partially having TSMC N3), because if they can't get MTL-S out then they have to get ARL-S out ASAP even if it's just a less ambitious core count. They can't afford to mess that up.

Might make sense for other reasons to stick with Raptor Refresh... DDR5 prices, N3 wafer prices...
 
I am

giphy.gif
 
At first glance, it seems like Intel's attempt at doing "chiplets" is even less efficient than AMD's method from a perf/w perspective and maybe even just performance. Losing to Zen3 on single/lightly threaded work is very surprising to me. I expected that to be a clear win for SPR with a minor win overall in MT. Losing outright is a shock. No wonder Intel is pushing their accelerators so hard in their PR presentations.
 
At first glance, it seems like Intel's attempt at doing "chiplets" is even less efficient than AMD's method from a perf/w perspective and maybe even just performance. Losing to Zen3 on single/lightly threaded work is very surprising to me. I expected that to be a clear win for SPR with a minor win overall in MT. Losing outright is a shock. No wonder Intel is pushing their accelerators so hard in their PR presentations.
I agree.
 
You are. Intel's communications have been confusing, but ARL is using some combination of N3 and 20A, with the latest rumor being N3 for desktop, and 20A for mobile.
Okay I feel stupid, because I thought 18A to start, realised that Lunar Lake is the one that's 18A and that ARL is 20A instead and then wrote 18A anyway

Though that being said I didn't know about the desktop one utilising N3.
 
142W idle power for the high performance setting... Wtf is going on there?
Considering the boards were designed and ready for release many months ago it's genuinely very surprising to see what I can only assume is half baked firmware causing that excessive idle power.

I mean it can't be EMIB causing it, one of the biggest reasons it's even there is because it saves a tonne of power (at a given bandwidth) compared to normal packaging techniques.
 
Back
Top