Question AMD Phoenix/Zen 4 APU Speculation and Discussion

Page 42 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

inf64

Diamond Member
Mar 11, 2011
3,884
4,692
136
I think the price cuts on Raphael made Phoenix on desktop redundant. Just suck it up and buy a dGPU if you need gaming performance.
Yeah but the iGPU with that kind of performance (Phoenix) would be a great value for many AM5 customers on a budget.
 
  • Like
Reactions: Tlh97 and Kaluan

Heartbreaker

Diamond Member
Apr 3, 2006
5,160
6,778
136
Yeah but the iGPU with that kind of performance (Phoenix) would be a great value for many AM5 customers on a budget.


I wouldn't want to be waiting on Phoenix for a build. They still haven't delivered the previous generation Rembrandt APU for AM4 socket users, as they all went to laptop and OEM pcs. I expect the same will happen for Phoenix.
 

ryanjagtap

Member
Sep 25, 2021
146
214
126
So essentially AMD could do a launch of A620 chipset motherboards alongside ryzen 5 and ryzen 7 Rembrandt CPUs? It could be a good budget offering to get customers on the AM5 platform. (But will they do it?)
 

Heartbreaker

Diamond Member
Apr 3, 2006
5,160
6,778
136
Afaik Rembrandt doesn't support DDR4 so can't come to AM4.

Ok, fair enough, but there have been AM5 MB since Sept last year, so they could have released it on AM5.

The point is AMD is slow to provide APUs to end users, and instead wants to mainly sell to OEMs for laptops.
 

BorisTheBlade82

Senior member
May 1, 2020
710
1,132
136
View attachment 77567
Notebookcheck.net
AMD clearly wins. I would love to see how would Intel perform on better process.
The thing is, that these are not even the Mobile First SKUs. Although they have a clear core count advantage, it remains to be seen if the monolithic Phoenix Point parts aren't even a lot better in terms of efficiency (for ST sure as hell, for MT it might get close).

Do we have any leaks regarding desktop Zen 4 APUs?
We do not even have hard currency for the real mobile ones - see above.
 
  • Like
Reactions: Tlh97 and moinmoin

Shivansps

Diamond Member
Sep 11, 2013
3,918
1,570
136
I think the price cuts on Raphael made Phoenix on desktop redundant. Just suck it up and buy a dGPU if you need gaming performance.

Not really, with Windows 10/11 you can easily change gpus per game, having a good igp is a pro because you can run on it old or light games on it whiout using your big dgpu.

And in my case sometimes i dont have power and i run my pc on a inverter, in those cases i could run the games only on the igp if i had one.

Definately going for an APU on AM5, just no now. maybe in one/two years.
 

soresu

Diamond Member
Dec 19, 2014
4,112
3,569
136
Definately going for an APU on AM5, just no now. maybe in one/two years.
Same.

Strix Point or its inevitable Zen5/RDNA4 successor should have pretty decent performance for SFF PCs, biting at the heels of a RX 6600 XT for raster and better for RT.
 

Kaluan

Senior member
Jan 4, 2022
515
1,092
106

It seems we have some new Phoenix score

As support for LPDDR5X-7500 is also present, it would be interesting to see some score with that.
CPU score seems very low (5700G/6900HS typical score is 9000-9500 w/ adequate wattage).

Might this be ran on a very modest TDP?

Either way, glad the trend with the leaks is upward.


Edit: Dug around a bit and based on the first Phoenix TS leak, here's 7940HS with much older drivers at 54W on the left and 25W on the right:
AMD-RADEON-780M-3DMARK.jpg

The 25W result's CPU score lines up perfectly with this new result (at a unknown TDP).

Conclusion: Take it with a pinch of salt but this 3000 TS graphics score may be from a 25W configuration. For context, that's 50% higher than 6900HS/680M @25W.

Edit2: Here's a 7840HS (CPU) result from MLID @25W as well, just slightly under the 7940HS:
RYZEN-7840HS-IGPU-TEST.jpg

No doubt with higher TDP, finalized drivers and LPDDR5X-7500 these could do mid 3000s in TS graphics.
I expect the AM5 version to hit around RX 570 performance under the right circumstances.
 
Last edited:

careymariah

Junior Member
Mar 10, 2023
1
2
41
As of my knowledge cutoff date of 2021, there were no plans for an iGPU (integrated graphics processing unit) infinity cache that is distinct from L3 (level 3 cache). However, it's always possible that new developments could arise in the future.
 
  • Like
Reactions: Tlh97 and maddie

TESKATLIPOKA

Platinum Member
May 1, 2020
2,696
3,260
136
I wouldn't mind If Minisforum made a thin form factor for Phoenix(Raphael) + GPU(something with more than 8GB Vram) similar to NUCXI7.
Link
NUCXi7_5.jpg

Not really upgradable expect 2x SSD and 2x RAM, but It looks great and is thin, while cooling is pretty good and not being noisy.

P.S. Honestly, It could have been smaller, inside is still a lot of unused space in the front part.
 

leoneazzurro

Golden Member
Jul 26, 2016
1,114
1,867
136
Last edited:

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
Ryzen 9 7945HX vs Intel i9-13980HX Benchmarks! How to Undervolt Ryzen! Scar 16 vs Zephyrus Duo 16! - YouTube

One of the first reviews available. Note that the Zephirus Duo has a 25W TDP deficit compared to the I9 notebook. Gaming is comparable, Ryzen seems better in application benchmarks.

He s comparing a 7945HX that use 108W while the Intel CPU use about 190W to get comparable MT scores, but of course it s all about idling power deficit...

 

tamz_msc

Diamond Member
Jan 5, 2017
3,865
3,730
136
From the 7945HX review it is clear that AMD's sub-optimal memory controller coupled with the chiplet architecture is hurting gaming performance, yet again.

V-Cache should have been reserved for mobile CPUs. Not only would the extra L3 help with the memory deficit, it would also lead to lower power consumption in gaming, allowing more headroom to the GPU.
 

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
27,247
16,108
136
From the 7945HX review it is clear that AMD's sub-optimal memory controller coupled with the chiplet architecture is hurting gaming performance, yet again.

V-Cache should have been reserved for mobile CPUs. Not only would the extra L3 help with the memory deficit, it would also lead to lower power consumption in gaming, allowing more headroom to the GPU.
The title says that it beats the Intel counterpart, so why is its gaming performance getting hurt ????? Its winning !
 

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136


The GPU in AMD s based laptop is set at a lower TDP than in the Intel counterfparts, wich render all your "explanations" above completely fantasmagoric, here the full review of the laptop :



Asus offers the new ROG Zephyrus Duo 16 with the Nvidia GeForce RTX 4080 Laptop or the GeForce RTX 4090 Laptop like on our review unit right now. In theory, it is the fastest version of the GPU, but the maximum TGP of 175 Watts (150W TGP + 25W Dynamic Boost) is reserved for the manual power profile. The Turbo profile, which was used for all the benchmarks, is limited to 155 Watts (140W TGP + 15W Dynamic Boost).
 

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
27,247
16,108
136
From the conclusion:

Initial Verdict: AMD has the best mobile processor
The benchmark results above obviously do not give us a full picture yet, but the initial results are extremely positive. AMD managed to close the gap in terms of single-core performance with lower power consumption. The new Zen 4 chip is even faster in multi-core benchmarks and still consumes considerably less power at just 120 Watts, which means laptop manufacturer get more headroom for the cooling of the GPU, for example.
 

tamz_msc

Diamond Member
Jan 5, 2017
3,865
3,730
136
The GPU in AMD s based laptop is set at a lower TDP than in the Intel counterfparts, wich render all your "explanations" above completely fantasmagoric, here the full review of the laptop :
It is losing to the RTX 4080. And no, AMD memory controller sucks with JEDEC timings - it is a proven fact.

Edit - Downvotes won't change the reality:

1678897763029.png
 
Last edited: