
World’s Fastest SFF Graphics Card in Retail
Nvidia has began gross sales of the small RTX 4000 SFF Ada Era graphics card that guarantees GeForce RTX 3070-like efficiency at 70W and may slot in nearly any desktop PC. The low-profile, dual-wide motherboard is not low-cost—it is dearer than the RTX 4080, for instance, as a result of it is aimed toward professional customers—however nothing stops you from putting in it on a daily gaming PC.
PNY’s Nvidia RTX 4000 SFF Ada Era graphics card is now $1,444 from ShopBLT, a retailer recognized for getting forward of its opponents with touchdown gear. Subsequently, the board is promoting for a better value than its official MSRP of $1,250. Observe that the motherboard is provided with 4 Mini-DisplayPort connectors, so you will have so as to add the worth of an mDP-DP or mDP-HDMI adapter to the price of this miniature answer.
The Nvidia RTX 4000 SFF Ada Era motherboard options the corporate’s AD104 GPU with 6,144 lively CUDA cores out of a complete of seven,680, together with 20GB of GDDR6 ECC reminiscence that connects to the GPU through a 160-bit interface. The GPU has a restricted enhance frequency of about 1560 MHz to scale back general board energy consumption and is rated for less than 70 W of energy;
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 {dollars} |
GeForce RTX 4070 | AD104 | 5888 | 29 TFLOPS | 116/233 | 12GB 192-bit 21 GT/s GDDR6X | 200W | $599 |
RTX 4000 Island Era | AD104 | 6144 | 19.2 TFLOPS | 153/307 TFLOPs | 20GB 160bit 16 GT/s GDDR6 ECC | 70W | $1,250 |
GeForce RTX 3090 Ti | GA102 | 10,752 | 40 TFLOPS | 160/320 TFLOPS | 24GB 384bit 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 |
Efficiency-wise, Nvidia’s GA104 graphics processor on this configuration chip delivers a peak FP32 efficiency of 19.2 TFLOPS, making it theoretically much like the GeForce RTX 3070. Greater than the RTX 3070 and in addition doubtlessly extra helpful for professionals and AI researchers. The reminiscence configuration in all probability makes use of 2GB GDDR6 chips on either side of the PCB, in any other case the 160-bit interface limits the utmost reminiscence to solely 10GB.
Round 20 FP32 TFLOPS are overshadowed by the superior efficiency of the not too long ago launched GeForce RTX 4070 (29 FP32 TFLOPS). The cardboard additionally options peak RT efficiency of 44.3 TFLOPS and peak FP8/INT8 tensor efficiency of 153/306.8 TFLOPS/TOPS (non-sparse and sparser). FP8/INT8 efficiency has nothing to do with video games in fact, however is an added bonus for the skilled market. In truth, 153/306.8 TFLOPS/TOPS is corresponding to the dearer and power-hungry Nvidia GeForce RTX 3090 Ti (in case you do not thoughts precision, because the 3090 Ti solely has native FP16 assist).
Different advantages of the RTX 4000 SFF Ada embrace assist for Nvidia’s skilled drivers {and professional} software program ISVs. It additionally comes with a 3-pin mini-DIN connector for stereoscopic 3D output (eg Nvidia 3D Imaginative and prescient) and helps Body Lock function for multi-screen functions.
We must always see pricing and availability enhance within the coming weeks as the cardboard turns into extra extensively out there. For now this can be a workstation piece designed for compact programs, aiming for low energy somewhat than most efficiency.
#Worlds #Quickest #SFF #Graphics #Card #Retail