Wednesday, April 14th 2021

Rumor: AMD Ryzen 7000 (Raphael) to Introduce Integrated GPU in Full Processor Lineup

The rumor mill keeps crushing away; in this case, regarding AMD's plans for their next-generation Zen designs. Various users have shared pieces of the same AMD roadmap, which apparently places AMD in an APU-focused landscape come their Ryzen 7000 series. we are currently on AMD's Ryzen 5000-series; Ryzen 6000 is supposed to materialize via a Zen 3+ design, with improved performance per watt obtained from improvements to its current Zen 3 family. However, Ryzen 7000-series is expected to debut on AMD's next-gen platform (let's call it AM5), which is also expected to introduce DDR5 support for AMD's mainstream computing platform. And now, the leaked, alleged roadmaps paint a Zen 4 + Navi 2 APU series in the works for AMD's Zen 4 debut with Raphael - roadmapped for manufacturing at the 5 nm process.

The inclusion of an iGPU chip with AMD's mainstream processors may signal a move by AMD to produce chiplets for all of its products, and then integrating them in the final product. You just have to think about it in the sense that AMD could "easily" pair one of the eight-core chiplets from the current Ryzen 5800X, for example, with an I/O die (which would likely still be fabricated with Global Foundries) an an additional Navi 2 GPU chiplet. It makes sense for AMD to start fabricating GPUs as chiplets as well - AMD's research on MCM (Multi-Chip Module) GPUs is pretty well-known at this point, and is a given for future development. It means that AMD needed only to develop one CPU chiplet and one GPU chiplet which they can then scale on-package by adding in more of the svelte pieces of silicon - something that Intel still doesn't do, and which results in the company's monolithic dies.
That this APU integration occurs throughout the whole of AMD's lineup makes sense. For one, the integrated GPU can be used to further accelerate certain tasks on your PC. Add to that the increased reputation that AMD has garnered via its generations of Zen, and you'll see how manufacturers (OEMs especially) might now be looking amenably at integrating AMD solutions into their mainstream, high-volume manufacturing products. The full APU road also makes sense in the AM5 platform - access to DDR5 might improve performance of the integrated GPU chiplets, and it makes sense that the new chiplet pairing requires a new platform and pin layout to make work.
Source: Videocardz
Add your own comment

42 Comments on Rumor: AMD Ryzen 7000 (Raphael) to Introduce Integrated GPU in Full Processor Lineup

#26
watzupken
Minus Infinity
The server market is demanding an iGPU and this is probably why they are doing it. Epyc will come with an iGPU as sure as the sun rises in the east and sets in the west.

I think it's great to have a highly efficient yet very good iGPU to compliment the discrete GPU. And let's face integrated RDNA2 should be pretty darn good and will probably handle 1080p gaming with decent settings so the laptop market will love this.
Not sure about AMD's plans ahead, but laptop segment will have their own APUs. The move to integrate an iGPU in the server and desktop chips will be beneficial because Intel sells a lot of processor with iGPU (even though the iGPU is very dated and slow) every year. If AMD wants to gain market share in the enterprise/ companies segment, they will need to follow suit.
Posted on Reply
#27
TheLostSwede
Wirko
I/O pins are already there - or else, APUs on AM4 wouldn't be able to output video. All B550 boards have a HDMI output, many have DP too, even many X570 boards have one or both.
1080p decoding or 720p encoding for videoconferencing would put some load on the CPU - I have no idea how much, I'm just guessing that one core dedicated to video should be enough. That used to be a problem in 2007, it shouldn't be a problem now.

Managing customers' expectations would be the hardest part. Yes, you can have a nice Ryzen-based office PC with two displays without a dGPU. But no, you can't do exactly everything without a dGPU.
You're aware I/O pins/interfaces can be muxed, right? So yes, the pins might be for display output on the APUs, but they might be PCIe on the CPUs.
As we don't have AMD's pin-outs, it's impossible to make any kind of assumptions.

As for video playback, try H.265 or AV1 and you'll see that software decoding is madness.
Posted on Reply
#28
TumbleGeorge
TheLostSwede
You're aware I/O pins/interfaces can be muxed, right? So yes, the pins might be for display output on the APUs, but they might be PCIe on the CPUs.
As we don't have AMD's pin-outs, it's impossible to make any kind of assumptions.


Is this^?
Posted on Reply
#29
spnidel
Wirko
fpu? What do you mean by that? Electric brain substitute? (Excuse me for trying to do away with common buzzwords, E.B.S. is just an AI!DL!NPU! unit.) That would make sense as it would execute its own program code - like a GPU does, not a part of CPU's code.
fpu is anything but a buzzword lol, it's been a thing for, I dunno, 20+ years?
Posted on Reply
#30
TheLostSwede
TumbleGeorge


Is this^?
I dunno, where did you find that?
spnidel
fpu is anything but a buzzword lol, it's been a thing for, I dunno, 20+ years?
Nah, it's more like 30+ years.
In fact, Intel used to sell a separate FPU for their 386SX CPUs that fitted in a special FPU socket.
en.wikipedia.org/wiki/X87#80387
Looking at that, it would seem they'd made them a lot longer than that, so we're talking late 70's.
Apparently AMD made the first FPUs for Intel as early as 1977...
Posted on Reply
#31
TumbleGeorge
TheLostSwede
I dunno, where did you find that?
Wikichip
Posted on Reply
#32
Chrispy_
EzioAs
Poll has no option for backup GPU when you don't have a dedicated video card or just want to troubleshoot something.
This.
Troubleshoot/Reflash a bricked dGPU BIOS/Temporary solution because dGPUs can't be bought at the moment.
Posted on Reply
#33
BArms
watzupken
Not sure about AMD's plans ahead, but laptop segment will have their own APUs. The move to integrate an iGPU in the server and desktop chips will be beneficial because Intel sells a lot of processor with iGPU (even though the iGPU is very dated and slow) every year. If AMD wants to gain market share in the enterprise/ companies segment, they will need to follow suit.
In case you didn't hear, AMD is selling every 5900X and 5950X they can possibly make, all without iGPUs. Intel has been pushing their iGPU BS for a long time and just because they do doesn't make it "beneficial" because virtually nobody wants them on a high end CPU since they're almost always paired with a high end GPU. Intel's quicksync is nice and so is having a backup GPU but that's no reason for the vast majority to be forced to buy them on high end desktop SKUs.

I get your argument for enterprise though, I"m just worried that AMD will follow intel's lead and release only a single SKU for both sectors so enterprise gets their iGPUs for spreadsheets while enthusiasts and gamers get a useless section of silicon they don't care for or need.
Posted on Reply
#34
bencrutz
Minus Infinity
The server market is demanding an iGPU and this is probably why they are doing it. Epyc will come with an iGPU as sure as the sun rises in the east and sets in the west.

I think it's great to have a highly efficient yet very good iGPU to compliment the discrete GPU. And let's face integrated RDNA2 should be pretty darn good and will probably handle 1080p gaming with decent settings so the laptop market will love this.
what? servers need 2D chip with embedded IPMI functionality, like aspeed ast 2500 series :wtf:
Posted on Reply
#35
Punkenjoy
The main problem is Windows require 3d feature now days to do compositing and stuff. A 2d chips barely work fine. But they will do it at some point. Not sure if it would make sense to add one to the current I/O die that is already quite large. Maybe once it's made with something denser than the current Global Foundry process.


One of the curse of current chips is the vast amount of silicon that are barely or never use. Adding a backup GPU that is just barely good enough won't help. Better add something that people could actually use or don't have it at all. With Intel, quicksync still useful. On AMD, there is no such thing yet.

We will see. Even if AMD is leading currently on the hardware front or on par, they still have so much catch up to do on the software side.
Posted on Reply
#36
r9
Expecting lawsuit from Lego. lol
Posted on Reply
#37
Speedyblupi
TheLostSwede
Looks pretty dashed to me...


Oh, I thought you were talking about the striped patten on some of the colours rather than the dashed line around the box. My mistake.

In that case, you are likely to be correct. It might be indicating that AMD hasn't yet confirmed their decision of whether they want to include Navi2 in Raphael or not.
Posted on Reply
#38
Vanny
An iGPU would be the one thing to push me to upgrade from my 3900X (which needs no upgrading). If my GPU dies, I want to at least be able to use my PC. No backup GPUs available. And a Navi2 iGPU sounds great to me.
Posted on Reply
#39
mechtech
hmmmm interesting

I would assume it's going to be some cut down thing that will be equivalent to what Intel has. I can't see them slapping on an integrated gpu with 10+ CU.

It would be nice though if they do have a little gpu built in as back up for times like these with the ridiculous dedicated gpu market.

However, I would also hope for a really good APU line, meaning I would take 4 fast cores with at least 16CU over an 8 core with 4CUs, or an 8 core with 20CU
Posted on Reply
#40
olymind1
I fear if they include an iGPU on their next (faster) lineup, the cheapest 6 core cpus will cost 350-400€ easily. I hope they will start producing great value cpus again, not just pricey (fast) ones.
Posted on Reply
#41
Wirko
TheLostSwede
Nah, it's more like 30+ years.
In fact, Intel used to sell a separate FPU for their 386SX CPUs that fitted in a special FPU socket.
en.wikipedia.org/wiki/X87#80387
Looking at that, it would seem they'd made them a lot longer than that, so we're talking late 70's.
Apparently AMD made the first FPUs for Intel as early as 1977...
Yeah, I'm aware of that, at least since I didn't buy a 80287 (an 80286 at 12 MHz was my first PC). Thanks for the link anyway, I learned about Intel licencing stuff from AMD back in the 70s, and the dirty trick called the i487.
What I really meant is: the suggestion by @mtcn77 to put the "fpu" on a separate chiplet makes sense if fpu stands for ML accelerator, which executes its own code, independent from CPU's code. FP instructions (scalar, vector, AVX), which are found in the same execution stream as CPU's code (integer, conditions, loops) have no chance of successful acceleration by a separate chiplet. Data transfer and synchronisation would eat away any advantages of fast calculation - which is exactly what made the original 8087 very inefficient.
Posted on Reply
#42
Blueberries
This would allow AMD to not only sell their more powerful and more expensive chips to prebuilt manufacturers but also shrink their design process even further.

It's a long needed change and a continued step in the right direction. Simplifying your product stack is the only way to stay ahead in such a saturated market.
Posted on Reply
Add your own comment