AMD whitethorn scrap nan high-end options of its adjacent procreation of Radeon gaming GPUs to divert scarce resources into building GPUs for AI and high-performance computing (HPC) alternatively – a conception that’s undergoing thing of a boom.
When AMD launches its RDNA 4 family of GPUs, perchance adjacent year, location won’t beryllium an AMD Radeon RX 8800 aliases 8900, according to TechSpot. This will springiness its rival Nvidia a clear tally astatine manufacturing nan best GPUs to meet nan high-end gaming market, but could besides service to constrain proviso and spike prices.
The line-up will lucifer nan RDNA 1 family of AMD GPUs, according to sources speaking to nan publication, wherever nan astir powerful introduction was nan RX 5700 XT GPU. Subsequent generations included higher-end units specified arsenic 6800, 6900, and 6950 successful RDNA 2, and 7800 and 7900 successful past year’s RDNA 3 series.
AMD wants successful connected nan AI boom
This rationale is simple. There’s a unreserved for hardware and components to work generative AI workloads – alongside a constricted proviso of resources and manufacturing capacity – and AMD wants to get successful connected nan action.
Indeed, this is simply a conception successful which there’s presently a shortage, pinch chipmaking elephantine TSMC lacking nan capacity to ramp up accumulation from vendors for illustration Nvidia to meet manufacture demand.
Nvidia’s A100 and H100 chips, incidentally, presently lead nan measurement successful an AI servers marketplace that’s reportedly group to surge to $150 cardinal by 2027, pinch AMD hoping to beryllium a portion of it. The main logic is Nvidia is enjoying profit margins of 823%, according to sister tract Tom’s Hardware, connected its H100 GPUs.
Rather than diverting semiconductors into its high-end user GPUs, nan patient will attraction connected section programmable gross arrays (FPGAs) and general-purpose graphics processing units (GPGPUs). This is according to Brits and Chips - Eng.
The circuitry of nan erstwhile is highly suited to instrumentality learning and heavy learning, while nan second are GPUs that besides grip computational workloads usually undertaken by the champion CPUs. They’re some perfect to meet nan rising request for GPUs for AI.
With constrained supply, however, does does mean we whitethorn spot a return to nan shortages and spiking prices for GPUs that we past saw successful 2020. With less options, gamers whitethorn find themselves paying supra nan likelihood erstwhile building PCs, for example.
More from TechRadar Pro
- We've rounded up nan best GPUs you tin bargain today
- Read our guideline connected nan best AMD graphics cards you tin find astatine nan moment
- These are nan best 4K graphics cards for gamers and creatives