A bit of misinformation in the article- the leaker said/hinted nothing about it being a 6000G (desktop) part. And it is not likely that it is. Laptops with H-series, based on Rembrandt, are about to be announced- and this is the most likely source of the score leak.

Your complaint has been forwarded to the trash where it belongs.
[/quote]

Okay boomer.

Quote from: deksman2 on December 07, 2021, 21:56:17
Depending on which uArch they go with for 6000 series APU, and if they implement MCM in the iGP design (which is currently being used in the MI200 accelerator that's been released - and consumer version of MCM chips aren't slated until 2022 anyway - 1 year later) then sure, we could easily see a 2x increase in performance.

There is no need for MCM to double Cezanne's performance, and we already know everything that's in Rembrandt. It's 12 RDNA 2 CUs.

+20% performance per CU, +50% CUs vs. Vega 8, +10% clock speed easily gets you to doubled performance. And it has all the bandwidth it needs from DDR5.

Quote from: Rob Stan on December 07, 2021, 22:21:23
Is this notebookcheck or desktopcheck, Rembrandt is a laptop APU first and foremost and it's obvious the leak is from a mobile part (like a Ryzen 6800H Engineering Sample).

Not sure why you guys bring up stuff like 5700G, a dekstop part. Shouldn't you guys mention stuff like performance of 5800H's iGPU instead? Which, judging by your own data, scores between 1200 and 1400 in Time Spy (Overall score, as the one leaked for Rembrandt).

Desktop Rembrandt isn't even confirmed yet and you guys are talking as if that's launching at CES on January 4th. It's not, the mobile parts are... which you guys are all about supposedly...

Your complaint has been forwarded to the trash where it belongs.

Is this notebookcheck or desktopcheck, Rembrandt is a laptop APU first and foremost and it's obvious the leak is from a mobile part (like a Ryzen 6800H Engineering Sample).

Not sure why you guys bring up stuff like 5700G, a dekstop part. Shouldn't you guys mention stuff like performance of 5800H's iGPU instead? Which, judging by your own data, scores between 1200 and 1400 in Time Spy (Overall score, as the one leaked for Rembrandt).

Desktop Rembrandt isn't even confirmed yet and you guys are talking as if that's launching at CES on January 4th. It's not, the mobile parts are... which you guys are all about supposedly...

I wouldn't be surprised if this turns out to be the case.
AMD mentioned that RDNA 3 for example will have another 50% performance per watt increase... and RDNA 2 already has that over NAVI.

Depending on which uArch they go with for 6000 series APU, and if they implement MCM in the iGP design (which is currently being used in the MI200 accelerator that's been released - and consumer version of MCM chips aren't slated until 2022 anyway - 1 year later) then sure, we could easily see a 2x increase in performance.

But we'll see what happens. From what we know, it was mentioned (somewhere) that consumer versions of MCM will be seen only in top-end GPU's... so that would (seemingly) exclude the iGP's from getting it (at least until 2023).