Nvidia’s Tiny RTX 4000 SFF 20GB Provides RTX 3070 Efficiency at 70W

Posted on

There are a great deal of compact trendy workstations that pack fairly succesful CPUs, however on the identical time they lack the house to accommodate a regular high-performance workstation-grade graphics card. That usually limits them to entry-level GPUs with mediocre efficiency. For those who desire a compact SFF workstation with extra graphics oomph, Nvidia has launched a brand new ProViz-oriented RTX 4000 SFF Ada Technology (opens in new tab) graphics card. It is one of many extra attention-grabbing choices within the current years, packing a high-end GPU right into a low-profile form-factor, with an influence consumption of simply 70W.

The Nvidia RTX 4000 SFF Ada board makes use of the corporate’s AD104 graphics processing unit with 6144 CUDA cores enabled (out of 7680 in complete). That is the identical GPU because the RTX 4070 Ti however with fewer energetic cores, and the enhance frequency will get capped at round 1560 MHz to decrease complete board energy. Alternatively, the graphics card comes with 20GB of GDDR6 reminiscence with ECC that connects to the GPU utilizing a 160-bit interface, so numerous reminiscence for workstation use.

The GPU comes with two NVENC encoders and two NVDEC decoders activated, although Nvidia has not touched upon actual capabilities of those models. They need to be much like the NVENC and NVDEC utilized in different Ada playing cards, and you may see the video encoding efficiency and high quality in our current roundup of GPUs.

The GA104 chip on this configuration delivers peak single precision efficiency of 19.2 TFLOPS, making it theoretically similar to a GeForce RTX 3070. It has peak RT efficiency of 44.3 TFLOPS, and peak FP8/INT8 tensor efficiency of 306.8 TFLOPS/TOPS.

(Picture credit score: Nvidia)

Nearly 20 FP32 TFLOPS could also be dwarfed by the overwhelming efficiency of Nvidia’s RTX 6000 Ada Technology or GeForce RTX 4090, however the RTX 4000 SFF is a low-profile dual-slot graphics card that may match into virtually any desktop pc, even one that doesn’t have a spare auxiliary PCIe energy connector. Apparently, RTX 4000 Ada’s 153/306.8 INT8 TFLOPS (with out and with sparsity, respectively) efficiency could be very near that of Nvidia’s GeForce RTX 3090 Ti that’s each dearer and way more energy hungry.

Swipe to scroll horizontally
Nvidia RTX 40-Sequence Specs
Row 0 – Cell 0 GPU FP32 CUDA Cores FP32 TFLOPS INT8 TFLOPS Reminiscence Configuration TBP MSRP
GeForce RTX 4070 Ti AD104 7680 40 TFLOPS 160/320 TFLOPS 12GB 192-bit 21 GT/s GDDR6X 285W $799
GeForce RTX 4070 AD104 5888 (?) ? ? 12GB 192-bit 21 GT/s GDDR6X 250W (?) ?
RTX 4000 Ada Technology AD104 6144 19.2 TFLOPS 153/307 TFLOPS 20GB 160-bit 16 GT/s GDDR6 ECC 70W $1,250
GeForce RTX 3090 Ti GA102 10,752 40 TFLOPS 160/320 TFLOPS 24GB 384-bit 20 GT/s GDDR6X 450W $1,999
GeForce RTX 3070 GA104 5888 20.31 TFLOPS 81/160 TFLOPS 8GB 256-bit 14 GT/s GDDR6 220W $499

Supply hyperlink

Leave a Reply

Your email address will not be published. Required fields are marked *