This site may earn affiliate commissions from the links on this page. Terms of use.

If you're a fan of AMD's GPUs, the by few years have been a fleck rough. Polaris has offered competitive performance in the midrange consumer market, but AMD'due south Vega vicious short of overall expectations. In the wake of Vega's launch, AMD appear that it would build a 7nm GPU based on Vega and aimed at the auto intelligence market. AMD hasn't formally shipped MI60 all the same, but it appear the part today and confirmed that it would ship before the cease of the year.

The new MI60 will use PCIe 4.0 for additional bandwidth and offers up to 1.25x more performance within the same ability envelope or fifty percent reduced power at 14nm clock speeds. That 25 percent gain is in-line with what we predicted for potential 7nm performance on Vega, and the power reduction at same-clock is also as expected — GPUs tend to depict much more ability when pushed out of their clocking 'sweet spot.' It's the start GPU to utilise Infinity Cloth over the PCIe bus and packs 13.2B transistors in a 331mmii package, compared with 12.5B transistors in 486mmii for the original Vega.

The MI60 volition pack 32GB of HBM2 with a massive 1TB/southward of memory bandwidth. That's significantly higher than any GPU currently on the marketplace; Nvidia'southward V100 GPUs elevation out at 900GB/southward. Those of you lot who were holding out some hope that 7nm Vega would debut in whatever kind of personal SKU or consumer production, nosotros've got bad news — AMD explicitly confirmed on stage that this card is solely intended for enterprise markets.

Performance in DGEMM and SGEMM benchmarks is said to exist directly comparable to Nvidia'south Tesla V100 and substantially faster than the MI25 (shown to a higher place). But of course, these are vendor-provided tests and should exist taken with a grain of common salt until we know more well-nigh the conditions under which the tests were run.

Overall, the MI60 seems to be an aggressive loftier-end part for this market, simply it'south non clear what kind of adoption it will encounter. Nvidia has a meaning lead over AMD in this space, and while offering a high-end 7nm solution for AI/ML workloads is absolutely required for AMD to compete, it will take fourth dimension for enterprises and companies to evaluate and adopt the solution. AMD aggressively talked up existence the outset company to send a 7nm GPU and the first to support PCIe four.0, only software compatibility and ease-of-porting will probably be more critical to overall uptake.

AMD confirmed that it will ship the bill of fare before the end of the twelvemonth and that information technology'south already working on MI-Adjacent, merely gave no details on any upcoming consumer cards based on Navi.

Now Read: AMD launches Radeon Instinct GPUs to tackle deep learning, artificial intelligence, Epyc Achievement: AMD Now Bachelor for Oracle Cloud Compute Instances, and AMD Moves All 7nm CPU, GPU Production to TSMC