AMD units its sight on adopting an aggressive strategy within the AI business, because it discloses its plans for next-gen MI400 Intuition AI accelerators.
AMD Goals At Enhancing its Software program Platform to Present Enhanced Help For Generative AI Functions, Confirms Subsequent-Gen Intuition MI400 Accelerators
AMD has formally confirmed its intentions to develop the “MI400” Intuition sequence. This affirmation comes as no shock since Lenovo’s VP already confirmed the existence of the next-gen accelerators on AMD’s agenda.
Throughout the firm’s current Q2 earnings name, CEO Lisa Su hinted on the future Intuition MI400 AI Accelerators, with out disclosing any specifics, therefore making a cliffhanger. However similar to the MI300 sequence, the MI400 accelerators may also are available a spread of configurations.
Whenever you look throughout these workloads and the investments that we’re making, not simply at present, however going ahead with our subsequent era MI400 sequence and so forth and so forth, we positively consider that we have now a really aggressive and succesful {hardware} roadmap. I believe the dialogue about AMD, frankly, has at all times been concerning the software program roadmap, and we do see a little bit of a change right here on the software program facet.
Dr. Lisa Su (AMD CEO)
The CEO’s assertion mentions the truth that the Intuition lineup is well-equipped within the {hardware} division, providing top-of-the-line specs. Nevertheless, it’s felt that AMD lacks software program growth, which incorporates in depth assist for generative AI functions. NVIDIA has been on the forefront for now, with the introduction of options corresponding to “NVIDIA ACE” and “DLSDR”. AMD plans at enhancing issues on the software program facet, therefore we are able to see main adjustments being launched by AMD, which can hopefully elevate the Intuition platform.
As beforehand mentioned, aside from the next-gen MI400 Intuition lineup, AMD additionally disclosed plans for the event of “cut-down” MI300 variants for Chinese language markets in an effort to adjust to US commerce insurance policies. Whereas precise specs aren’t sure but, we consider that Crew Purple will undertake the same strategy taken by NVIDIA with their “H800 and A800” GPUs.
NVIDIA has largely been capitalizing from the AI “gold mine”, reaching great ranges of gross sales and demand. Rivals like Intel and AMD have been late, however they may present powerful competitors given the truth that they create higher efficiency and worth onboard.
AMD Radeon Intuition Accelerators
Accelerator Title | AMD Intuition MI400 | AMD Intuition MI300 | AMD Intuition MI250X | AMD Intuition MI250 | AMD Intuition MI210 | AMD Intuition MI100 | AMD Radeon Intuition MI60 | AMD Radeon Intuition MI50 | AMD Radeon Intuition MI25 | AMD Radeon Intuition MI8 | AMD Radeon Intuition MI6 |
---|---|---|---|---|---|---|---|---|---|---|---|
CPU Structure | Zen 5 (Exascale APU) | Zen 4 (Exascale APU) | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A |
GPU Structure | CDNA 4 | Aqua Vanjaram (CDNA 3) | Aldebaran (CDNA 2) | Aldebaran (CDNA 2) | Aldebaran (CDNA 2) | Arcturus (CDNA 1) | Vega 20 | Vega 20 | Vega 10 | Fiji XT | Polaris 10 |
GPU Course of Node | 4nm | 5nm+6nm | 6nm | 6nm | 6nm | 7nm FinFET | 7nm FinFET | 7nm FinFET | 14nm FinFET | 28nm | 14nm FinFET |
GPU Chiplets | TBD | 8 (MCM) | 2 (MCM) 1 (Per Die) |
2 (MCM) 1 (Per Die) |
2 (MCM) 1 (Per Die) |
1 (Monolithic) | 1 (Monolithic) | 1 (Monolithic) | 1 (Monolithic) | 1 (Monolithic) | 1 (Monolithic) |
GPU Cores | TBD | Up To 19,456 | 14,080 | 13,312 | 6656 | 7680 | 4096 | 3840 | 4096 | 4096 | 2304 |
GPU Clock Velocity | TBD | TBA | 1700 MHz | 1700 MHz | 1700 MHz | 1500 MHz | 1800 MHz | 1725 MHz | 1500 MHz | 1000 MHz | 1237 MHz |
FP16 Compute | TBD | TBA | 383 TOPs | 362 TOPs | 181 TOPs | 185 TFLOPs | 29.5 TFLOPs | 26.5 TFLOPs | 24.6 TFLOPs | 8.2 TFLOPs | 5.7 TFLOPs |
FP32 Compute | TBD | TBA | 95.7 TFLOPs | 90.5 TFLOPs | 45.3 TFLOPs | 23.1 TFLOPs | 14.7 TFLOPs | 13.3 TFLOPs | 12.3 TFLOPs | 8.2 TFLOPs | 5.7 TFLOPs |
FP64 Compute | TBD | TBA | 47.9 TFLOPs | 45.3 TFLOPs | 22.6 TFLOPs | 11.5 TFLOPs | 7.4 TFLOPs | 6.6 TFLOPs | 768 GFLOPs | 512 GFLOPs | 384 GFLOPs |
VRAM | TBD | 192 GB HBM3 | 128 GB HBM2e | 128 GB HBM2e | 64 GB HBM2e | 32 GB HBM2 | 32 GB HBM2 | 16 GB HBM2 | 16 GB HBM2 | 4 GB HBM1 | 16 GB GDDR5 |
Reminiscence Clock | TBD | 5.2 Gbps | 3.2 Gbps | 3.2 Gbps | 3.2 Gbps | 1200 MHz | 1000 MHz | 1000 MHz | 945 MHz | 500 MHz | 1750 MHz |
Reminiscence Bus | TBD | 8192-bit | 8192-bit | 8192-bit | 4096-bit | 4096-bit bus | 4096-bit bus | 4096-bit bus | 2048-bit bus | 4096-bit bus | 256-bit bus |
Reminiscence Bandwidth | TBD | 5.2 TB/s | 3.2 TB/s | 3.2 TB/s | 1.6 TB/s | 1.23 TB/s | 1 TB/s | 1 TB/s | 484 GB/s | 512 GB/s | 224 GB/s |
Type Issue | TBD | OAM | OAM | OAM | Twin Slot Card | Twin Slot, Full Size | Twin Slot, Full Size | Twin Slot, Full Size | Twin Slot, Full Size | Twin Slot, Half Size | Single Slot, Full Size |
Cooling | TBD | Passive Cooling | Passive Cooling | Passive Cooling | Passive Cooling | Passive Cooling | Passive Cooling | Passive Cooling | Passive Cooling | Passive Cooling | Passive Cooling |
TDP (Max) | TBD | 750W | 560W | 500W | 300W | 300W | 300W | 300W | 300W | 175W | 150W |
Information Sources: Videocardz, SemiAnalysis
Originally posted 2023-08-05 17:31:03.