Wednesday, March 18th 2020

Sony Reveals PS5 Hardware: RDNA2 Raytracing, 16 GB GDDR6, 6 GB/s SSD, 2304 GPU Cores

Sony in a YouTube stream keynote by PlayStation 5 lead system architect Mark Cerny, detailed the upcoming entertainment system's hardware. There are three key areas where the company has invested heavily in driving forward the platform by "balancing revolutionary and evolutionary" technologies. A key design focus with PlayStation 5 is storage. Cerny elaborated on how past generations of the PlayStation guided game developers' art direction as the low bandwidths and latencies of optical discs and HDDs posed crippling latencies arising out of mechanical seeks, resulting in infinitesimally lower data transfer rates than what the media is capable of in best case scenario (seeking a block of data from its outermost sectors). SSD was the #1 most requested hardware feature by game developers during the development of PS5, and Sony responded with something special.

Each PlayStation 5 ships with a PCI-Express 4.0 x4 SSD with a flash controller that has been designed in-house by Sony. The controller features 12 flash channels, and is capable of at least 5.5 GB/s transfer speeds. When you factor in the exponential gains in access time, Sony expects the SSD to provide a 100x boost in effective storage sub-system performance, resulting in practically no load times.
The secret sauce here is that Sony is using its own protocol instead of NVMe, in supporting 6 data priority tiers versus 2 on NVMe. Each PlayStation 5 ships with an 825 GB SSD, which is expandable using external HDDs over USB, or a selection of third-party M.2 NVMe SSDs certified by Sony. PlayStation 4 games can run directly off your external HDD, but PlayStation 5 games have to be transferred from your HDD to the console's main SSD. Past generations of PlayStation implemented ZLib data compression on Blu-ray and HDD media. PlayStation 5 is implementing Kraken, with hardware-accelerated de-compression via fixed-function hardware built directly into the main SoC.

SoC is where Cerny sounded restrained in what he wanted to disclose. The SoC is a semi-custom chip designed by Sony and AMD, possibly on a 7 nm-class silicon fabrication process. Sony won't specify if it is a monolithic silicon or an MCM, but there are three building-blocks to it: CPU, GPU, and I/O complex. The CPU is based on AMD "Zen 2" x86-64 microarchitecture, and the GPU is based on the company's upcoming RDNA2 graphics architecture.

There are eight "Zen 2" CPU cores, although the company didn't mention if SMT is featured. The maximum CPU clock speed is 3.50 GHz. The GPU is a whole different story from the one on the Xbox Series X Velocity Engine semi-custom chip. Sony decided to go with 36 RDNA2 compute units ticking at up to 2.23 GHz engine clock, compared to 52 compute units running at up to 1.825 GHz on the upcoming Xbox. Sony's GPU ends up with up to 10.3 TFLOPs max compute throughput, compared to Microsoft's 12 TFLOPs.

Sony also shed some "light" on the hardware-accelerated real-time ray-tracing approach AMD is taking with RDNA2. Apparently, each compute unit features a hardware component called "Intersection Engine," with roughly the same function as an RT core on NVIDIA "Turing," which is to calculate the intersection of rays with geometry (such as triangles or polygons) in a scene. This combines with a fairly standardized bounding volume hierarchy (BVH) model to achieve a hybrid of ray-traced elements in an otherwise conventional rasterized 3D scene (pretty much where NVIDIA is right now with RTX). On PlayStation 5, RDNA2's ray-tracing hardware is leveraged for positional audio, global illumination, shadows, reflections, and full ray-tracing.

The third key component of the SoC is the I/O complex. This handles all of the chip's I/O, not just with peripherals and video output, but also storage and memory. There are dedicated I/O co-processors on-silicon designed to reduce the various I/O's processing stack on the CPU cores, and reduce latencies at various stages. There's also a certain amount of SRAM that caches transfers between the various components on the I/O complex. The custom chip leverages AMD SmartShift in power-management.

PlayStation 5 uses 16 GB of GDDR6 memory. Sony did not mention the memory clock, bandwidth, or even the memory bus width. It did drop some hints about memory management. It appears like PlayStation 5 does not partition memory the way Xbox Series X does, and possibly sticks to the hUMA model of the PlayStation 4 (using a common pool of physical memory for system- and video memory).

Lastly, a large chunk of Sony's presentation focused on the next frontier for hardware innovation: positional audio. Sony is investing heavily on positional audio that takes into account the gamer's physical HRTF (head-related transfer function). The company is leveraging the vast amounts of CPU power gained from the upgrade to "Zen 2," to achieve this.
We still don't know what a PlayStation 5 console will look like. Source: Sony Computer Entertainment (YouTube)
Add your own comment

169 Comments on Sony Reveals PS5 Hardware: RDNA2 Raytracing, 16 GB GDDR6, 6 GB/s SSD, 2304 GPU Cores

#1
MxPhenom 216
ASIC Engineer
Based on other reports. The memory is tied to a 256-bit bus.
Posted on Reply
#2
ShurikN
Well, now we definitely know which console is going to be cheaper.
Posted on Reply
#3
R0H1T
btarunr
Sony decided to go with 36 RDNA2 compute units ticking at up to 2.23 GHz engine clock, compared to 52 compute units running at up to 1.825 GHz on the upcoming Xbox. Sony's GPU ends up with up to 10.3 TFLOPs max compute throughput, compared to Microsoft's 12 TFLOPs.
Well that should settle the debate about the upcoming "big Navi" cause it's gonna be a beast w/better IPC, clocks, more CU, hardware RT & who knows 3d audio?
Posted on Reply
#4
ppn
Like a playstation4 pro on steroids, same core count. updated uArch. 3x faster.
Posted on Reply
#5
Rahnak
MxPhenom 216
Based on other reports. The memory is tied to a 256-bit bus.
And memory bandwidth is 448GB/s.

R0H1T
Well that should settle the debate about the upcoming "big Navi" cause it's gonna be a beast w/better IPC, clocks, more CU, hardware RT & who knows 3d audio?
I think the 3D audio is custom. There's a lot of custom hardware in the PS5, just like in the older ones.
Posted on Reply
#6
P4-630
btarunr
Each PlayStation 5 ships with an 825 GB SSD
Which reaching speeds of 5,5GB/s without compression. Compressed up to 8 to 9GB/s.
To expand storage you can use regular m2-nvme-ssd's.

Also there's "3d Audio" support with the "Tempest Engine" which also works with a Stereo setup.
(Which is similar to the Synergistic Processor Unit used in the Sony in de PlayStation 3)

https://tweakers.net/nieuws/164788/playstation-5-krijgt-gpu-met-36-cus-op-2-komma-3ghz-en-825gb-ssd-die-5-komma-5gb-s-haalt.html
Posted on Reply
#7
Ferrum Master
R0H1T
Well that should settle the debate about the upcoming "big Navi" cause it's gonna be a beast w/better IPC, clocks, more CU, hardware RT & who knows 3d audio?
You can scratch 3D audio. On windows it is a pain in the arse. I predicted tad HW audio will come back because of VR and low latency needs, but current state of core windows kernel and PC architecture does not allow that. It really needs a custom design and tailored code not jack of all trades that runs of anything. Tough topic tbh.

Looks like PS5 has more tailored things, especially the GPU cache coherency managers. And it will end up much cooler too imho.
Posted on Reply
#8
R0H1T
Rahnak
I think the 3D audio is custom. There's a lot of custom hardware in the PS5, just like in the older ones.
I know I'm just hoping they bring something like TrueAudio back.

Ferrum Master
It really needs a custom design and tailored code not jack of all trades that runs of anything. Tough topic tbh.
Yes a dedicated DSP would be nice, just like their previous solution.
Posted on Reply
#9
TechLurker
All in all, both the PS5 and Series X seem to be within ~10% of each other.

Xbox has more brute force power, PS5 has faster base speeds and throughput.

Xbox will eke out a few more FPS for the games that demand pure FPS and possibly at higher resolutions. Sony instead offers seamless transition and more dynamic movement; going by how much they've hyped their storage setup and their earlier Spiderman demo (and previous discussion on how character movements were limited by the read speeds of SSDs/HDDs).

The caveat though is that the Xbox seems redundant now that they're sharing releases with PC and thus is going for more of an all-in-one media center with their new Xbox (still a gaming powerhouse, but nevertheless overshadowed by parallel/near-parallel releases with PC), whereas Sony still gets some exclusives and still presents a gaming-first mindset. The ideal combo is looking to be PC + PS5. PC for all the new games releasing alongside Series X on top of an existing Steam/GoG/etc library, and PS5 for exclusives + media center functionality.

Now if only Sony would just release a PSP/Vita successor using AMD's APUs or embedded architecture and offer that seamless transition between on-the-go and at-home-gaming (start a game on one, pick up where you left off on the other, etc).
Posted on Reply
#10
kings
I highly doubt that these 2.23Ghz maximum boost on the GPU will last for a long time, with the inherent cooling limitations that consoles have.

It gives the impression that it was stretched at the last minute, just to reach the double digit 10TF and not feel much weaker than the Xbox Series X.
Posted on Reply
#11
Rahnak
kings
I highly doubt that these 2.23Ghz maximum boost on the GPU will last for a long time, with the inherent cooling limitations that consoles have.

It gives the impression that it was stretched at the last minute, just to reach the double digit 10TF and not feel much weaker than the Xbox Series X.
Cerny said the variable frequency is based on workload, not thermal headroom. Cooling will be unveiled when they do the teardown.

Also, this presentation was recorded before XSX announcement, Eurogamer, for example, saw it a couple of days ago, so a last minute bump in frequency is very unlikely. He said "we" would be happy with their cooling solution though. Whatever that's worth.
Posted on Reply
#13
TheLostSwede
Same, yet so different. It's strange that Sony has put so much importance on storage and the SSD, while having what appears to be a much weaker GPU, compared to Microsoft. I guess it's possible that Sony's API (Vulcan?) is more efficient than DirectX, so they don't need as much raw GPU power?
Not a console gamer, but the hardware is intriguing none the less. We've gone from fairly fixed usage devices to what can only be described as one high-end-ish PC running Windows and another running Linux.
Posted on Reply
#14
sepheronx
I am very curious how much this will be priced at.

With how much power is being proclaimed by Microsoft and Sony regarding their GPU's and CPU's, its hard to imagine they can release this at even a reasonable price. But if RDNA2 is that good, then I am glad I am hanging onto my GTX 1070 till those GPU's come out.
Posted on Reply
#15
R0H1T
TheLostSwede
I guess it's possible that Sony's API (Vulcan?) is more efficient than DirectX, so they don't need as much raw GPU power?
Linux itself is much better with multiple cores than Windows, if MS hasn't heavily tweaked the XB series X scheduler then they'll be wasting a lot of zen2 power.
Posted on Reply
#16
Turmania
Good times for console industry. I never bought a PS but seriously considering my stand on it for PS5.
Posted on Reply
#17
Rahnak
TheLostSwede
Same, yet so different. It's strange that Sony has put so much importance on storage and the SSD, while having what appears to be a much weaker GPU, compared to Microsoft. I guess it's possible that Sony's API (Vulcan?) is more efficient than DirectX, so they don't need as much raw GPU power?
Not a console gamer, but the hardware is intriguing none the less. We've gone from fairly fixed usage devices to what can only be described as one high-end-ish PC running Windows and another running Linux.
Cerny gave a simple explanation on why they went with fewer CUs at higher frequency as opposed to more CUs at a lower frequency. Quoting Eurogamer, quoting Cerny:
Not wishing to draw comparisons with any existing hardware past, present or future, Cerny presents an intriguing hypothetical scenario - a 36 CU graphics core running at 1GHz up against a notional 48 CU part running at 750MHz. Both deliver 4.6TF of compute performance, but Cerny says that the gaming experience would not be the same.

"Performance is noticeably different, because 'teraflops' is defined as the computational capability of the vector ALU. That's just one part of the GPU, there are a lot of other units - and those other units all run faster when the GPU frequency is higher. At 33 per cent higher frequency, rasterisation goes 33 per cent faster, processing the command buffer goes that much faster, the L1 and L2 caches have that much higher bandwidth, and so on," Cerny explains in his presentation.

"About the only downside is that system memory is 33 per cent further away in terms of cycles, but the large number of benefits more than counterbalance that. As a friend of mine says, a rising tide lifts all boats," explains Cerny. "Also, it's easier to fully use 36 CUs in parallel than it is to fully use 48 CUs - when triangles are small, it's much harder to fill all those CUs with useful work."

Sony's pitch is essentially this: a smaller GPU can be a more nimble, more agile GPU, the inference being that PS5's graphics core should be able to deliver performance higher than you may expect from a TFLOPs number that doesn't accurately encompass the capabilities of all parts of the GPU. Developers work to the power limits of the SoC, their workloads affecting frequencies on the fly - but it's those factors that impact the clock speeds, not ambient temperatures.
Price might've been a factor as well, dunno.
Posted on Reply
#18
ppn
Im waiting for the Pro version of it. having this main IO chip on 12nm, and separate GPU CPU, means that whenever 5nm shrink of is available sony will integrate it very quickly. XSX can't do that, it is stuck on 7nm.
Posted on Reply
#19
Rahnak
ppn
Im waiting for the Pro version of it. having this main IO chip on 12nm, and separate GPU CPU, means that whenever 5nm shrink of is available sony will integrate it very quickly. XSX can't do that, it is stuck on 7nm.
I sincerely hope there are no other versions this time around. The Pro versions sole reason for existing was the mass adoption of 4K TVs. I don't see 8K becoming mainstream in the next 5-7 years. Hopefully.

Not counting the usual "slim" versions, of course.
Posted on Reply
#20
kings
Sony may be playing the price card as it did with PS4. If they wanted something more powerful, they would also have it, they have access to the same AMD capabilities that Microsoft has.

Anyway, it is far from being a weak console, 10TF GPU, together with a CPU that makes Jaguar look like from last century, plus super fast SSD, will give a fantastic experience.

With these specs, I start to think that they are probably aiming for $399. And at this price, I have no doubt that it will be a success. I don't see the Xbox Series X being sold for less than $500~$550.
Posted on Reply
#21
Rahnak
399 again would be sweet. But there was that rumor a while ago saying they were having trouble keeping costs under $450, I think it was. Given the custom hardware they're putting on it, I wouldn't be surprised if that were true. Then again, Sony has sold consoles at a loss before.
Posted on Reply
#22
Prince Valiant
R0H1T
I know I'm just hoping they bring something like TrueAudio back.

Yes a dedicated DSP would be nice, just like their previous solution.
I'd like to see that too. The general attitude toward audio in games would need to change. When TA was announced there were a number of articles along the lines of "do we need this".
Posted on Reply
#23
matar
Xbox series X is the clear Winner... 10.3 TFLOPs max compute throughput, compared to Microsoft's 12 TFLOPs , Also Xbox software tends to use the headwear better then sony we have seen in in old 360 where sony had better hardware but yet 360 ran games smother and now we have both on the NEW Xbox better hardware and combine it with better software = super smooth game play.
Posted on Reply
#24
oxrufiioxo
TheLostSwede
Same, yet so different. It's strange that Sony has put so much importance on storage and the SSD, while having what appears to be a much weaker GPU, compared to Microsoft. I guess it's possible that Sony's API (Vulcan?) is more efficient than DirectX, so they don't need as much raw GPU power?
Not a console gamer, but the hardware is intriguing none the less. We've gone from fairly fixed usage devices to what can only be described as one high-end-ish PC running Windows and another running Linux.
Slightly disappointed after the X reveal but considering Sony is more likely to target this hardware as a base I still think Sony first party games may be more impressive.... It will be interesting to see what general consumers do now that basically all multiplayer games are crossplay if the X hits higher framerates or higher resolution it may sway people torwards it.
Posted on Reply
#25
ARF
matar
Xbox series X is the clear Winner... 10.3 TFLOPs max compute throughput, compared to Microsoft's 12 TFLOPs , Also Xbox software tends to use the headwear better then sony we have seen in in old 360 where sony had better hardware but yet 360 ran games smother and now we have both on the NEW Xbox better hardware and combine it with better software = super smooth game play.
You can't know this. Xbox has 10 GB VRAM + 6 GB system RAM, while normal PCs go ****today**** with 16 GB system RAM and 8 GB VRAM.
It will be particularly interesting to see these consoles in 3-4 years when the games will become more demanding for hardware resources.

12 vs 10.3 is not much of a difference, especially when you have a locked FPS at 60 or so.
Posted on Reply
Add your own comment