Slash Boxes

SoylentNews is people

posted by martyb on Wednesday August 25 2021, @04:35AM   Printer-friendly [Skip to comment(s)]

AMD Unveils New Ryzen V-Cache Details at HotChips 33:

AMD gave us more information about its upcoming V-Cache at Hot Chips this year, the annual conference where semiconductor engineers from all over the industry come together to crow over disclose details regarding their technical achievements in the past 12 months.

Earlier this year, AMD announced that it would not advance directly from Zen 3 to Zen 4. Instead, it would iterate on the Zen 3 core by stacking a full 64MB of 7nm L3 cache vertically on the core. AMD claims this can improve performance by up to 15 percent based on 1080p gaming results. The improvement in other applications is unknown.

Original Submission

Related Stories

AMD Announces Milan-X Epyc With 3D V-Cache, Bergamo, and First MCM GPU: Instinct MI200 16 comments

AMD has announced its "Milan-X" Epyc CPUs, which reuse the same Zen 3 chiplets found in "Milan" Epyc CPUs with up to 64 cores, but with triple the L3 cache using stacked "3D V-Cache" technology designed in partnership with TSMC. This means that some Epyc CPUs will go from having 256 MiB of L3 cache to a whopping 768 MiB (804 MiB of cache when including L1 and L2 cache). 2-socket servers using Milan-X can have over 1.5 gigabytes of L3 cache. The huge amount of additional cache results in average performance gains in "targeted workloads" of around 50% according to AMD. Microsoft found an 80% improvement in some workloads (e.g. computational fluid dynamics) due to the increase in effective memory bandwidth.

AMD's next-generation of Instinct high-performance computing GPUs will use a multi-chip module (MCM) design, essentially chiplets for GPUs. The Instinct MI250X includes two "CDNA 2" dies for a total of 220 compute units, compared to 120 compute units for the previous MI100 monolithic GPU. Performance is roughly doubled (FP32 Vector/Matrix, FP16 Matrix, INT8 Matrix), quadrupled (FP64 Vector), or octupled (FP64 Matrix). VRAM has been quadrupled to 128 GB of High Bandwidth Memory. Power consumption of the world's first MCM GPU will be high, as it has a 560 Watt TDP.

The Frontier exascale supercomputer will use both Epyc CPUs and Instinct MI200 GPUs.

AMD officially confirmed that upcoming Zen 4 "Genoa" Epyc CPUs made on a TSMC "5nm" node will have up to 96 cores. AMD also announced "Bergamo", a 128-core "Zen 4c" Epyc variant, with the 'c' indicating "cloud-optimized". This is a denser, more power-efficient version of Zen 4 with a smaller cache. According to a recent leak, Zen 4c chiplets will have 16 cores instead of 8, will retain hyperthreading, and will be used in future Zen 5 Ryzen desktop CPUs as AMD's answer to Intel's Alder Lake heterogeneous ("big.LITTLE") x86 microarchitecture.

Also at Tom's Hardware (Milan-X).

Previously: AMD Reveals 'Instinct' for Machine Intelligence
AMD Launches "Milan" Epyc Server CPUs, with Zen 3 and up to 64 Cores
AMD at Computex 2021: 5000G APUs, 6000M Mobile GPUs, FidelityFX Super Resolution, and 3D Chiplets
AMD Unveils New Ryzen V-Cache Details at HotChips 33
AMD Aims to Increase Energy Efficiency of Epyc CPUs and Instinct AI Accelerators 30x by 2025

Original Submission

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 0) by Anonymous Coward on Wednesday August 25 2021, @07:46AM (1 child)

    by Anonymous Coward on Wednesday August 25 2021, @07:46AM (#1170709)

    SoylentNews is going to the birds? Thought we were just rid of one feathered raptor.

  • (Score: 3, Interesting) by Opportunist on Wednesday August 25 2021, @10:04AM (2 children)

    by Opportunist (5545) on Wednesday August 25 2021, @10:04AM (#1170741)

    Face it, it doesn't matter at all how good, fast or powerful your chip is if it's essentially unobtanium. Try to solve the supply crisis first before announcing some wonder chips, nobody gives a fuck about it unless we can actually buy it without paying a fortune to scalpers.

  • (Score: 0) by Anonymous Coward on Wednesday August 25 2021, @09:37PM (1 child)

    by Anonymous Coward on Wednesday August 25 2021, @09:37PM (#1170988)

    OTOH it's a gpu not a cpu so, if many here are like me and are text heavy and not miners, I can understand why few comments. Enjoy your 10% year over year performance bump based on iterative design, this impacts me only insofar as the expected value and impact of GPU-gated activities (folding, mining, actual big data, gaming) shift in the world. This is good for GME - minor HW bump means market dollars going into PC HW and more into GME's offerings. This only barely nudges expected foundry throughput expectations. Etc.

    None of these are big news, but it's nice to know as a datapoint.

    If nobody had complained I'd have simply had the above thoughts and not replied either, btw, since all are minor. Again, can see why few replies != few reads. Some things are worth knowing but have little to talk on.

    • (Score: 2) by takyon on Wednesday August 25 2021, @09:56PM

      by takyon (881) <reversethis-{gro ... s} {ta} {noykat}> on Wednesday August 25 2021, @09:56PM (#1170990) Journal

      OTOH it's a gpu not a cpu so

      Well, you got it wrong right out of the gate. Maybe follow your own advice next time and don't comment.

      Having said that, rumor has it that RDNA 3's "Infinity Cache" will also be stacked, or if not, it will be massively increased from the 128 MB of RX 6800 (XT)/RX 6900 to up to 512 MB [].

      [SIG] 10/28/2017: Soylent Upgrade v14 []