Hot Chips 2020 Live Blog: Microsoft Xbox Series X System Architecture (6:00pm PT)by Dr. Ian Cutress on August 17, 2020 9:00 PM EST
- Posted in
- Live Blog
- Xbox Series X
- Hot Chips 32
09:04PM EDT - Final talk of the day is Xbox Series X System Architecture!
09:05PM EDT - Azure Silicon Architecture Team
09:06PM EDT - 3.8 GHz Zen2 Server cores
09:06PM EDT - DXR, VRS, Machine LEarning Acceleration
09:07PM EDT - 14 Gbps GDDR6, 320-bit = 560 GB/s
09:07PM EDT - Hardware accelerators in blue
09:07PM EDT - 120 Hz support, VRR, Xbox Velocity Architecture for MSP Crypto/Decomp on NVMe SSD
09:07PM EDT - Acoustic acceleration
09:07PM EDT - HSP/Pluton RoT - security
09:08PM EDT - 360.4mm2 TSMC N7 enhanced
09:08PM EDT - 15.3B transistors
09:08PM EDT - 2 four core CPU clusters
09:08PM EDT - 10 GDDR6 controllers
09:09PM EDT - GPU 12 FLOPs
09:10PM EDT - AVX256 gives 972 GFLOP over CPU
09:10PM EDT - 16 GB of GDDR6 total
09:11PM EDT - >Says Zen2 server class, but L3 cache is mobile class?
09:11PM EDT - Display processing is kept off the shader engines
09:11PM EDT - IO hub supports PCIe 4.0 x8
09:11PM EDT - Operates on linear light values, not gamma light values
09:12PM EDT - ALLM - Auto Low Latency Mode
09:13PM EDT - Increased die cost on this APU over previous generation
09:13PM EDT - Significantly more expensive!
09:13PM EDT - Trade off
09:13PM EDT - MS created Audio engines - 3 engines, CFPU2, MOVAD, LOGAN
09:13PM EDT - CFPU2 for audio convolution, FFT, reverb
09:14PM EDT - such as Project Acoustics to model 3D audio sources
09:14PM EDT - MOVAD - hyper real-time hardware audio decoder
09:14PM EDT - >300x channels decode at once
09:14PM EDT - best trade off codec, so made in hardware
09:15PM EDT - >100dB signal noise ratio
09:15PM EDT - HW realtime real-time matched to decode based on sampling
09:15PM EDT - Logan is offering also better offload in traditional modes
09:15PM EDT - HSP/Pluton: Root of trust, crypto, SHACK (crypto keys)
09:15PM EDT - MSP supports 5 GB/s high-bw crypto on the SSD
09:16PM EDT - DRAM to SSD balance needed for refill
09:16PM EDT - Load times are always increasing unless SW-to-DRAM bw increases, hence NVMe SSDs
09:17PM EDT - Sampler Feedback System
09:17PM EDT - New metadata for texture portions to pre-load texture caches
09:17PM EDT - Direct Storage
09:17PM EDT - Manages data locations ahead of developer
09:18PM EDT - Distinct savings for most detail texture maps savings
09:18PM EDT - Lossless MS XVA 2:1 compression
09:19PM EDT - Need big GPU - get the tech out of the way
09:19PM EDT - Need raw ops/second increase within PPA and cost
09:19PM EDT - 12.2 supported in HW
09:19PM EDT - 26 active dual CUs (52 CUs)
09:20PM EDT - single geometry supports primatives
09:20PM EDT - DIrectly snoop CPU caches
09:20PM EDT - Dual stream multi-core command processor
09:20PM EDT - Double rate 16-bit math
09:20PM EDT - single cycle issue rate to reduce stalls
09:21PM EDT - CUs have 25% better perf/clock compared to last gen
09:21PM EDT - GPU Evolution: FLOPS have outpaced mem space and BW
09:21PM EDT - Screen pixels has increased in th emiddle
09:22PM EDT - How to fill pixels better without blowing power budget
09:22PM EDT - VRS
09:22PM EDT - supports up to 2x2
09:22PM EDT - 10-30% perf gain for tiny area cost
09:23PM EDT - Full edge detail
09:23PM EDT - SFS
09:24PM EDT - Previously very slow to enable
09:24PM EDT - Two new HW structures for tile-by-tile management for in-DRAM textures
09:25PM EDT - clamps LOD
09:26PM EDT - Tilemaps should stay on die for best latency
09:27PM EDT - SFS: 60% IO/Mem savings for small die area cost
09:27PM EDT - DX Ray Tracing Accel
09:27PM EDT - Not a complete replacement - RT can be applied selectively based on traditional models
09:28PM EDT - Custom ray-triangle units
09:28PM EDT - ML inference
09:29PM EDT - Two virtualized command streams - two VMs
09:29PM EDT - Main title OS vs system OS
09:29PM EDT - 32b HDR rendering, blending display
09:29PM EDT - Optimized games. Unable to show at the event
09:30PM EDT - Q&A Time
09:31PM EDT - Q: TDP? A: Not commenting. There's so many things that are involved in the TDP, and tradeoffs. We're not really able to descibe it without describing it in a technical environemtn
09:32PM EDT - Q: Can you stream into the GPU cache? A: Lots of programmable cache modes. Streaming modes, bypass modes, coherence modes.
09:33PM EDT - Q: Coherency CPU and GPU? A: GPU can snoop CPU, reverse requires software
09:35PM EDT - Q: Are you happy as DX12 as a low hardware API? A: DX12 is very versatile - we have some Xbox specific enhancements that power developers can use. But we try to have consistency between Xbox and PC. Divergence isn't that good. But we work with developers when designing these chips so that their needs are met. Not heard many complains so far (as a silicon person!). We have a SMASH driver model. The games on the binaries implement the hardware layed out data that the GPU eats directly - it's not a HAL layer abstraction. MS also re-writes the driver and smashes it together, we replace that and the firmware in the GPU. It's significantly more efficient than the PC.
09:35PM EDT - Q: Is link between CPU and GPU clocks? A: Hardware is independent.
09:36PM EDT - Q: Is the CPU 3.8 GHz clock a continual or turbo? A: Continual.
09:36PM EDT - Continual to minimize variance
09:37PM EDT - Q: TSMC 7nm enhanced, is it N7P, N7+, or something else? A: It's not base 7nm, it's progressed over time. Lots of work between AMD and TSMC to hit our targets and what we needed
09:38PM EDT - Q: Says Zen 2 is server class, but you use L3 mobile class? A: Yeah our caches are different, but I won't say any more, that's more AMD.
09:39PM EDT - Q: With 20 channels GDDR6, is that really cheaper than 2 stacks HBM? A: We're not religious about which DRAM tech to use. We needed the GPU to have a ton of bandwidth. Lots of channels allows for low latency requests to be serviced. HBM did have an MLC model thought about, but people voted with their feet and JEDEC decided not to go with it.
09:40PM EDT - Q: GDDR6 on sides, not bottom? A: bottom is power, how board interfaces with the chip. GPU has high EDC and currents, and you need clean copper to deliver that. With that much current you need to leave that space unless you use super expensive packaging. We did it the cost efficient way
09:41PM EDT - Q: Why do you need so much math for audio processing? A: 3D positional audio and spatial audio and real world spaces if you 300-400 audio sounds positional in 3D and want to start doing other effects on all samples, it gets very heavy compute. Imagine 20 people fighting in a cave and reflections with all sorts of noises
09:43PM EDT - That's a wrap and we're done for today. Come back tomorrow at 8:30am PT to talk about FPGAs. It's 2:44am here in the UK, time to go to bed.
Post Your CommentPlease log in or sign up to comment.
View All Comments
eastcoast_pete - Monday, August 17, 2020 - linkQuestion: how does the audio processing compare to, let's say, that of a decent audio codec like a 1220?
dersteffeneilers - Thursday, August 20, 2020 - linkI think Discord uses Opus, so like that probably
Crazyeyeskillah - Monday, August 17, 2020 - linkInteresting dialog about the audio pipelines and how much power is really needed. It adds up REAL FAST if you're going to actually source everything in a scene. Image standing in a bustling city square, dozens of npc's footsteps, conversations, sound sources, all being fed in a 3d positional system. 300-400 almost seems like quite a small number when you factor in the thousands that should be present and accounted for in a real environment. At least we're in the right direction!
Cheesecake16 - Monday, August 17, 2020 - linkSorry Ian, I tried to get an answer about the Xbox Series X's GPU TDP because if the TDP is the same as the Xbox One's GPU (120W at worse) then AMD has a monster of a GPU on their hands.
Ian Cutress - Monday, August 17, 2020 - linkThe problem when asking for GPU TDP is that it's monolithic. I guess if you'd said 'SoC' level 'average' power you might have got more of an answer. I think his answer about 'TDP is different based on workload' was a massive cop-out and someone disingenuous given what the definition of TDP actually is, and that engineers in the room understand nuance around TDP numbers. Moreover, it doesn't have to be a number, it could be a range.
Perhaps the question should have been - what thermal profile in watts will the cooling for the chip have to manage? or How big will the power brick be? Sometimes asking questions is an art, so the presenter doesn't block themselves into a corner with their own interpretation of what the question is getting at.
bill54 - Monday, August 17, 2020 - linkwtf I can smell the hostility here that you hold towards MS/Xbox. But now I recall you tried to do some similar "gotcha" BS with the Xbox One back then though I dont remember about what.
If he said "we're not talking about TDP" what's the big deal? at this stage they're not talking about a lot of things, like ya know, the price.
I dont know why you tried to frame it as a hostile "gotcha" Ian, wanna bet the less powerful PS5 will likely have a higher TDP? Considering they've already been show to have a much larger volume case (PS5 is utterly massive), use some crazy liquid metal cooling system, and all that because they to overclock to over 2.2 ghz to try to catch up to Xbox even a little on the spec sheet and get over double digit TF. Now that's bad engineering.
kulareddy - Tuesday, August 18, 2020 - linkno way, Sony (and also Microsoft) will not use Liquid metal TIM.
Eliadbu - Tuesday, August 18, 2020 - linkto be honest this is just speculations. having more capable cooling solution can go into both ways either dissipating more heat or lowering noise levels. I think after PS4 pro fan noise issues Sony decided to have more capable cooling solution to avoid such scenario. we don't know which console consume more power and we don't know if PS5 has the same GPU architecture as Xbox series x, so keep the criticism after we know all the details.
and the criticism on Ian was not justified in my opinion, because he has a point the explanation why he won't comment is disingenuous. TDP is well defined so you don't need to take specific environment or other elements. if he would have just commented that he can't give more information right now, no one would have complained.
d0x360 - Wednesday, August 19, 2020 - linkThe base ps4 and the pro are louder than my PC, xbox one and xbox one x combined lol. I don't think the ps5 will be much better. They can't use a very large slower spinning fan in that case. I just don't see how it could fit (besides sideways) and I don't see how that would be beneficial (sideways) because the system has to work vertical and horizontal.
Eliadbu - Wednesday, August 19, 2020 - linkThere are few possible orientation and sizes. probably the main benefit of having larger volume is having place for larger heat sink which by itself helps enormously with dissipation see for example the 3 slots GPUs they do not usually have better or bigger fans but they do have much larger heat sink which helps with cooling enabling the fans to run significantly lower RPM for the same performance (compared to 2 slots designes) lowering the overall noise produced.