Question Speculation: RDNA3 + CDNA2 Architectures Thread

[uzzi38]

Man I have been dying to make this one for a while now.

First rumours for RDNA3 are here so new thread time!

Just going to start off with this one for now: kopite7kimi on Twitter: "@VideoCardz Ah, I mean a simple mcm design with 10240 cores is not enough. Because the lift from RDNA2 to RDNA3 is much bigger than from RDNA1 to RDNA2. We should expect many big improvements in GFX11. 🤔" / Twitter

 

[TESKATLIPOKA]

A 150mm2 die makes no sense when they knew they had a 200mm2 Navi 33 coming that blows it away in performance.

What are you talking about?
RX 6500Xt was released a year ago, N33 is not released yet.
That 150mm2 chip should have been released at that time instead of 107mm2 Navi 24.

A stripped down cheap Navi 24 die that can be produced in massive quantities is perfect budget laptop fodder and will be around for years.

How did this perfect budget laptop fodder help AMD in laptops?
How many models are available or at least exist?
It's supposedly produced in massive quantities, but It's even more scarce than a 6800U APU laptop.

Their only mistake was releasing it on desktop. 8GB of GDDR6 costs 2-3X more then the silicon, never-mind the PCB/cooler, so that’s not really a major cost factor here.

The mistake was even releasing such a small chip.
At least I can buy It a desktop version, even If It's not particularly cheap, but there is no better option.

N24 chip is really cheap to make, you just confirmed It.
That's why I said making a bigger one would make more sense.

What would you rather buy:
107mm2 chip with 4GB/8GB Vram for $199/219
1024SP:64TMU:32ROP:16MB ; game Clock: 2610 MHz ; TDP: 107W
or
150mm2 chip with 6GB/12GB Vram for $239/259
1536SP:96TMU:48ROP:24MB; game Clock: 2491 MHz ; TDP: 120W
The second option would provide better perf/W, perf/$, more Vram and would cost only $40 more.
I think the answer is clear.

edit: 12GB version in clamshell is not the best idea.

 

[TESKATLIPOKA]

If a 24 CU APU exists it’s probably not Strix Point. It’s likely a semi-custom design for some sort of handheld console or a dedicated design to service that market and perhaps thin and light gaming laptops. It wouldn’t make sense from a cost/performance perspective anywhere else. Hell it might just be an Xbox Series S refresh that also goes into Surface Books.

If It's a semi-custom design, then that's a problem.
I don't believe It to be for a handheld console. It would consume too much power and handhelds need a good battery life.
There is no OEM who would pay AMD to produce a semi-custom design, maybe Microsoft for Surface, but they don't do gaming laptops.

I think Xbox Series S refresh would be the most likely one If It's a semi-custom design.
Series S has only 20 CUs RDNA2 @ 1.565 GHz, 4.01 TFLOPS
Refresh: 24CU RDNA3 @ 2.6-3 GHz, 8-9.2 TFLOPS (didn't include dual issue ALU)
This actually looks pretty good.

 

[Kepler_L2]

If a 24 CU APU exists it’s probably not Strix Point. It’s likely a semi-custom design for some sort of handheld console or a dedicated design to service that market and perhaps thin and light gaming laptops. It wouldn’t make sense from a cost/performance perspective anywhere else. Hell it might just be an Xbox Series S refresh that also goes into Surface Books.

Or a giant sea monster

 

[Kronos1996]

What are you talking about?
RX 6500Xt was released a year ago, N33 is not released yet.
That 150mm2 chip should have been released at that time instead of 107mm2 Navi 24.

How did this perfect budget laptop fodder help AMD in laptops?
How many models are available or at least exist?
It's supposedly produced in massive quantities, but It's even more scarce than a 6800U APU laptop.

The mistake was even releasing such a small chip.
At least I can buy It a desktop version, even If It's not particularly cheap, but there is no better option.

N24 chip is really cheap to make, you just confirmed It.
That's why I said making a bigger one would make more sense.

What would you rather buy:
107mm2 chip with 4GB/8GB Vram for $199/219
1024SP:64TMU:32ROP:16MB ; game Clock: 2610 MHz ; TDP: 107W
or
150mm2 chip with 6GB/12GB Vram for $239/259
1536SP:96TMU:48ROP:24MB; game Clock: 2491 MHz ; TDP: 120W
The second option would provide better perf/W, perf/$, more Vram and would cost only $40 more.
I think the answer is clear.

Navi 24 was designed for the same market Nvidia’s MX GPU’s serve. As such, they won’t update it often and designed it to be as cheap as possible so it can be made for the next 3-5 years. Anything bigger would be a waste of silicon for it’s primary purpose as a budget laptop GPU. I don’t think they planned on releasing it to desktop at all initially until the shortages happened and they took advantage.

So why waste all that time and silicon on a 150mm2 die that’ll be in market for one year until Navi 33 arrives? You don’t and that’s why they never made one. Navi 33 will probably be between 6700-6800 XT performance on a 200mm2 6nm die with similar VRAM and component cost as this theoretical 150mm2 chip. They can sell it in $300-400 products which will help offset the big inflation which likely makes sub-$300 GPU’s difficult to profit on. So what makes the most sense for AMD? Exactly what they did.

 

[leoneazzurro]

Now which monster it can be
Leviathan?
Kraken?
Tiamat?
Charybdis?

 

[Kaluan]

Everyone just assumes RDNA3+ (assuming same node) WGPs are the exact same size as Phoenix's RDNA3 WGPs. We don't know that, hell we don't even know how big or small the 6 WGPs or whole IGP is relative to the Phoenix die. Let's wait for proper die shots and measurements first. In any case, it should be way smaller than Rembrandt's.
Maybe it's so small AMD lost the CPU:IGP 1:1 equilibrium they reached with Rembrandt? 😅

 

[Glo.]

It's only a rumour, we know how wrong such rumours were about RDNA3 just a month or two before release, Strix Point is still far away.
The rumours about N4 process and being monolithic would make this unrealistic paired with bigger Zen5 cores. Size would be clearly >200mm2.

TDP is also a problem. N4 Phoenix with 12CU boosting to 3GHz has 45W TDP. Game frequency is unknown.
2x bigger IGP at N4 would be impossible unless you significantly reduce clocks.
Even N3(E) with 30-34% reduction in power compared to N5 won't be enough to fit It in 45W If the power draw ratio during gaming is 1:2(15W:30W).
If you can't keep the frequency, then It's kinda pointless to put so much CU in It, 20CU would be a better option.

I still question If 32MB with 55% hitrate is enough in combination with LPDDR5(DDR5). There is no problem with BW of IC.
I would rather have a bigger dynamic SLC(LLC) for both CPU and GPU.

It appears Strix Point is on N3 process.

And you may have your wish about SLC fulfilled, after all.

 

[TESKATLIPOKA]

I don't mean that there would be both N24 and this chip. I meant this one should have been designed and released as N24.

Navi 24 was designed for the same market Nvidia’s MX GPU’s serve. As such, they won’t update it often and designed it to be as cheap as possible so it can be made for the next 3-5 years. Anything bigger would be a waste of silicon for it’s primary purpose as a budget laptop GPU. I don’t think they planned on releasing it to desktop at all initially until the shortages happened and they took advantage.

N24 was aimed against GA107.
MX models are much weaker, with a single exception being MX570.
GeForce MX570 was announced a month earlier than N24 and has comparable performance to a cutdown N24 is based on GA107.
6500M(Full N24) is comparable to RTX 3050(GA107).
Phoenix should already provide the same level of performance as this cutdown N24 making It rather pointless.
There is absolutely no good reason for N24 to be produced for an additional 3-5 years when even now barely anyone wants them, which is evident by the amount of laptops with It.

So why waste all that time and silicon on a 150mm2 die that’ll be in market for one year until Navi 33 arrives? You don’t and that’s why they never made one. Navi 33 will probably be between 6700-6800 XT performance on a 200mm2 6nm die with similar VRAM and component cost as this theoretical 150mm2 chip. They can sell it in $300-400 products which will help offset the big inflation which likely makes sub-$300 GPU’s difficult to profit on. So what makes the most sense for AMD? Exactly what they did.

Because AMD needs something for <=$249.
That 150mm2 chip wouldn't have worse profits than N24, and It also wouldn't cost much more to make.
It's not like N24 is much cheaper to make than N33, when you compare versions with 8GB Vram, yet price will be very different.
Making a 107mm2 GPU which after a year is made pointless by an IGP from the same company doesn't make much sense to me. At least my beefed up version of N24 would still be >50% faster than Phoenix and could be sold at least until Strix is out.

P.S. I think that 12GB version of mine is too costly to make because of clamshell, I would keep only the 6GB version for $239.

 

[insertcarehere]

It appears Strix Point is on N3 process.

And you may have your wish about SLC fulfilled, after all.

Doesn't solve the problem of cost given that N3 is unlikely to be cheaper per transistor than N4.

 

[Glo.]

Doesn't solve the problem of cost given that N3 is unlikely to be cheaper per transistor than N4.

Its still going to be hell of a lot cheaper to design ONE 3 nm product instead of two separate ones.

 

[TESKATLIPOKA]

Its still going to be hell of a lot cheaper to design ONE 3 nm product instead of two separate ones.

Isn't the problem actually the amount of APUs they produce?
If they made a lot more, then a second design would be paid for more easily by selling a bigger amount of chips.
I really wonder how many Rembrandt APUs they actually made, because It doesn't look like It was more than a few million chips.

 

[Glo.]

Isn't the problem actually the amount of APUs they produce?
If they made a lot more, then a second design would be paid for more easily by selling a bigger amount of chips.
I really wonder how many Rembrandt APUs they actually made, because It doesn't look like It was more than a few million chips.

If I were AMD, when Strix Point releases I would just maintain production of ONLY Phoenix, and Strix Point APUs. I don't know how their wafer supply agreements play out with the TSMC, and if they are the reason why AMD has to maintain production of older hardware.

 

[jpiniero]

If I were AMD, when Strix Point releases I would just maintain production of ONLY Phoenix, and Strix Point APUs. I don't know how their wafer supply agreements play out with the TSMC, and if they are the reason why AMD has to maintain production of older hardware.

And not get any sales? It's obvious that AMD is forced to charge a lot more for even Rembrandt (if they want to make similar margins...) compared to Cezanne/Barcelo because of the wafer prices let alone Phoenix. Strix Point is going to be even worse if it uses N3.

Cezanne/Barcelo is going to be around for awhile I think.

 

[Glo.]

And not get any sales? It's obvious that AMD is forced to charge a lot more for even Rembrandt (if they want to make similar margins...) compared to Cezanne/Barcelo because of the wafer prices let alone Phoenix. Strix Point is going to be even worse if it uses N3.

Cezanne/Barcelo is going to be around for awhile I think.

That is the unfortunate downside(the high wafer prices).

What I proposed would simplify the lineup.

 

[IntelUser2000]

Any iGPU in an APU increases the cost, so It's never really free, but It's still pretty useful. Especially when your dGPU dies.

Nah I mean chipset iGPUs added just $5-7. The CPU iGPUs add $30 plus the cost of the faster chip. So $30 if you were going to buy the CPU anyway, but the real cost might be $50-100. On mobile it might be even higher since you often need to buy an i7 to get the top of the line iGPU.

When Iris devices with eDRAM were being made, it was a $200-300 adder on top of the maxed out i7 config!

And $30 is a HUGE amount in terms of production cost. If it adds $30 to the MSRP of the CPU sure, but $30 is likely the cost of the entire chip.

You don't want a backup display GPU that expensive.

@Kronos1996 I second that it's more reasonable to be for a custom APU such as consoles.

I am not going to rule it out entirely but I am saying it doesn't make sense.

Everyone just assumes RDNA3+ (assuming same node) WGPs are the exact same size as Phoenix's RDNA3 WGPs.

And why would it be significantly smaller? + suggests they are likely adding features also it means the changes aren't that big.

Things like RDNA did to double FP32 instruction rate required substantial circuitry changes and they had to cut down on other areas to get there. You don't get things for free. They are not going to have that kind of changes with RDNA3+.

 

[MrTeal]

What would be the point of a 24CU IGP on an APU? That's a pretty massive GPU, just slightly smaller than a RX 7600M/S, but that's a 90W/75W TDP solution. It also has 32MB of IC and 256GB/s of bandwidth out to GDDR6. With dual channel LPDDR5x the whole APU would have 120GB/s bandwidth, with DDR5-5600 it would be less and shared with the CPU. Unless they completely change the memory system, that poor GPU would be completely memory staved.

 

[Stuka87]

And besides RGT is the same "source" that claimed 2.5x performance of RDNA3 over the predecessor! In reality, sometimes it's half that at 1.25x.

In what game is a 7900 XTX only 25% faster than a 6900 XT? (excluding clearly CPU bound resolutions and games obviously)

 

[IntelUser2000]

In what game is a 7900 XTX only 25% faster than a 6900 XT? (excluding clearly CPU bound resolutions and games obviously)

Battlefield V 1080P: 21%
CiV VI: 20.6%, 24.4% at 1440p

Considering the average is 30%, you are almost nitpicking.

 

[Stuka87]

Battlefield V 1080P: 21%

That's a CPU bound resolution though. The 7900 XTX is getting 361 fps. Which essentially makes the benchmark worthless.

At 4K, the 7900 XTX is getting 190fps. Which is 40% faster than the 6900 XT.

 

[IntelUser2000]

That's a CPU bound resolution though. The 7900 XTX is getting 361 fps. Which essentially makes the benchmark worthless.

At 4K, the 7900 XTX is getting 190fps. Which is 40% faster than the 6900 XT.

That's why I also included Civ IV. We're really nitpicking at this point, when the leakers were so far off that in reality we didn't get any increase in perf/$ this generation at all.

 

[GodisanAtheist]

Ok, just cause my memory is getting a little hazy here: Why is AMD even pursuing APUs anymore? What is the benefit of making a whole separate line of much larger monolithic dies with beefier IGPs when they can theoretically just slap something more powerful onto the Zen4 IO die or even package a CPU with a GCD (obviously not a proper RDNA3 GCD but something purpose built) that is much smaller and leverages their existing interconnect/substrate strategy?

APUs felt like they were the solution back before chiplets could be a thing, but now that we have chiplet CPUs and chiplet GPUs... is the clock ticking here on the APU?

 

[Stuka87]

Ok, just cause my memory is getting a little hazy here: Why is AMD even pursuing APUs anymore? What is the benefit of making a whole separate line of much larger monolithic dies with beefier IGPs when they can theoretically just slap something more powerful onto the Zen4 IO die or even package a CPU with a GCD (obviously not a proper RDNA3 GCD but something purpose built) that is much smaller and leverages their existing interconnect/substrate strategy?

APUs felt like they were the solution back before chiplets could be a thing, but now that we have chiplet CPUs and chiplet GPUs... is the clock ticking here on the APU?

Because Intel is. AMD lost out on quite a bit of OEM business because their CPU's required a dedicated GPU. If then want to get in on that business, they have to have a CPU/GPU that can compete with Intel.

 

[insertcarehere]

Because Intel is. AMD lost out on quite a bit of OEM business because their CPU's required a dedicated GPU. If then want to get in on that business, they have to have a CPU/GPU that can compete with Intel.

Except that issue did get remedied by Zen 4 putting a barebones IGP on their IOD. Pretty big difference between that and making the IGP beefy enough to run games well, especially when the extra die area can be used for higher margin products elsewhere for AMD.

 

[Timorous]

That's why I also included Civ IV. We're really nitpicking at this point, when the leakers were so far off that in reality we didn't get any increase in perf/$ this generation at all.

I don't get why TPU measure Civ 6 frame rate. 60fps is fine because it is turn based and more does not make any real difference. It is one of the oddest benchmarks in their suite.

 

[TESKATLIPOKA]

What would be the point of a 24CU IGP on an APU? That's a pretty massive GPU, just slightly smaller than a RX 7600M/S, but that's a 90W/75W TDP solution. It also has 32MB of IC and 256GB/s of bandwidth out to GDDR6. With dual channel LPDDR5x the whole APU would have 120GB/s bandwidth, with DDR5-5600 it would be less and shared with the CPU. Unless they completely change the memory system, that poor GPU would be completely memory staved.

It's not that big, only 2x bigger than a Phoenix without IGP, If my calculation is correct.
I calculated here 227mm2 for a Phoenix with 24CU + 32MB IC.
RX7600S goes as low as 50W including GDDR6 memory and let's not forget It's on 6nm process.
It's not like this 4nm APU must be limited to 35-45W. For this one add 30W for a total 65-75W and you are ready to go. If It can clock at ~2650MHz on average, then It will have 90% of RX 7700s 100W TFLOPs.
32MB IC would certainly help, but If that amount would be enough is pretty questionable.
At worst, I would use a shared 64MB LLC instead of separate 16MB L3 and 32MB IC. This would increase size by ~12-13mm2 to 240mm2.
I think with this amount, even a slower DDR5 shouldn't be a big problem.
Price could be set at $399 for 8C+24CU+64MB LLC, where It could offer better perf/$ ratio than a similarly performing CPU+dGPU combination.

What's the real problem in my opinion is OEM manufactures. As we saw with Rembrandt, they love to put AMD in premium laptops, so this in a laptop would cost more than a combination of CPU+dGPU so It would be pointless.

If they released It for DIY, then It could have a very good price/performance ratio and be a great option in a cheaper gaming machine or small form factor. In this case you can even set TDP a lot higher than 65-75W so IGP could be close to 3GHz on average.
This is my personal opinion.

P.S. The only reason I wouldn't buy this even If It existed is that I am limited to a laptop and a CPU+dGPU option would end up cheaper. Still, buying a new laptop for >€1500 with only 8GB dGPU(N33,Ad106,Ad107) doesn't look so great.