Nvidia Removes Some Video Encoding Limitations From Consumer GPUs

Nvidia has quietly eliminated some simultaneous video encoding limitations from client graphics processing items to allow them to now encode as much as 5 streams. The transfer will make video lovers’ lives simpler, however Nvidia’s datacenter-grade {and professional} GPUs will proceed to have an enormous benefit over client merchandise as Nvidia not restricts the variety of simultaneous periods on them.

Based on the corporate’s personal knowledge, Nvidia has elevated the variety of simultaneous NVENC encodings on client GPUs from three to 5. Video Encoding and Decoding GPU Support Matrix (opens in new tab). That is efficient for dozens of merchandise launched within the final eight years, primarily based on Maxwell Gen 2, Pascal, Turing, Ampere, and Ada Lovelace microarchitectures (except for some MX sequence merchandise). Nevertheless, the variety of simultaneous NVDEC decodings on these GPUs continues to be restricted to 1 stream.

The change doesn’t have an effect on the variety of NVENC and NVDEC {hardware} items enabled on Nvidia’s client GPUs. For instance, the newest AD102 graphics processor primarily based on Nvidia’s Ada Lovelace structure has three {hardware}. NVENC encoders (opens in new tab) and three NVDEC decoders. All three are enabled on Nvidia’s RTX 6000 Ada and L40 motherboards for workstations and knowledge facilities, however solely two are enabled on the consumer-grade GeForce RTX 4090.

Traditionally, Nvidia has restricted the variety of simultaneous decoding and encoding periods throughout all graphics playing cards. Client-oriented GeForce playing cards supported as much as three simultaneous NVENC video encoding periods and as much as one NVDEC video decoding stream. In distinction, Nvidia’s workstation and datacenter options operating the identical silicon and concentrating on ProViz, video streaming companies, recreation streaming companies, and digital desktop infrastructure (VDI) can help 11-17 concurrent NVENC periods relying on high quality and {hardware}. It turned out a couple of years in the past that Nvidia’s restrictions may very well be lifted with a comparatively easy hack.

However Nvidia’s stance on NVENC and NVDEC limitations has clearly modified. Now the patron GPU helps as much as 5 simultaneous NVENC encoding periods and as much as 1 NVDEC decoding session. In distinction, there are not any restrictions on workstation-class and datacenter-class boards, and the precise variety of simultaneous periods will depend on the precise {hardware} capabilities and video high quality.