[ad_1]
The latest Ampere leak from VideoCardz could have spoiled Nvidia’s announcement. The post has gotten new information that supposedly gives credence to the rumored specs for the GeForce RTX 3090 and GeForce RTX 3080 that will soon compete for a spot on our Best GPUs list.
Starting from the top, the Ampere silicon will apparently be made at the 7nm process node. The leaked spec sheets don’t specify the casting, but TSMC’s 7nm FinFET fabrication process is the first thing that comes to mind. Amp-based graphics such as GeForce RTX 3090 and RTX 3080 will take advantage of the PCIe 4.0 interface. Both models are rumored to use the same GA102 die. Nvidia SLI only supports GeForce RTX 3090 via the chipmaker’s NVLink interface.
Physically, Gainward’s custom models match what we’ve seen in the Founders Edition and Zotac Gaming GeForce RTX 3090 Trinity HoLo. The GeForce RTX 3090 and GeForce RTX 3080 will be gigantic graphics cards. Gainward models adhere to a 2.7 slot design and are based on a robust triple fan cooling solution.
Nvidia GeForce RTX 3090, RTX 3080 Specifications
| Graphic card | CUDA colors | Memory clock | Memory capacity | Memory bus | memory bandwidth | TGP |
|---|---|---|---|---|---|---|
| GeForce RTX 3090 * | 5,248 | 19 Gbps | 24GB GDDR6X | 384 bits | 935.8 GBps | 350W |
| Titan RTX | 4,608 | 14 Gbps | 24GB GDDR6 | 384 bits | 672 GBps | 280W |
| GeForce RTX 3080 * | 4,352 | 19 Gbps | 10GB / 20GB GDDR6X | 320 bits | 760 GBps | 320W |
| GeForce RTX 2080 Ti | 4,352 | 14 Gbps | 11GB GDDR6 | 352 bits | 616 GBps | 250W |
* Specifications are not confirmed.
The GeForce RTX 3090 seems to be equipped with 5,248 CUDA cores and 24GB of GDDR6X memory after all. In terms of clock speeds, the Gainward GeForce RTX 3090 Phoenix Golden Sample has a boost clock of 1,725 MHz.
The GDDR6X memory in the GeForce RTX 3090 runs at 19.5 Gbps on a 384-bit memory interface. If you do the math, the maximum theoretical memory bandwidth adds up to 936 GBps, a little less than 1 TBps. The TGP (Total Graphics Power) of the GeForce RTX 3090 is rated at 350W. Gainward’s model confirms early rumors about using twin 8-pin PCIe power connectors. When it comes to video outputs, the GeForce RTX 3090 Phoenix Golden Sample in particular provides three DisplayPort 1.4a outputs and one HDMI 2.1 port.
As for the GeForce RTX 3080, Gainward’s spec chart has the graphics card with 4,352 CUDA cores and 10GB of GDDR6X memory. The company’s custom model has a boost clock of 1,740 MHz. Memory is clocked at 19 Gbps with a 320-bit memory bus, which equates to a memory bandwidth of 760 GBps.
The TGP of the GeForce RTX 3080 appears to be 320 W, only 30 W lower than that of the GeForce RTX 3090. This should merit the presence of two 8-pin PCIe power connectors. The video outputs of the GeForce RTX 3080 Phoenix Golden Sample are identical to those of its bigger brother.
The GeForce RTX 3090’s 24GB VRAM screams that it’s the replacement for the Titan RTX. This is puzzling because it’s not like Nvidia to release a Titan-style graphics card so early in a product launch. Titan models usually arrive a few months later. It’s not like Nvidia is shooting itself in the foot, either. The GA102 silicon presumably houses up to 7,552 CUDA cores, and only 5,248 CUDA cores are enabled in the GeForce RTX 3090. If necessary, Nvidia has enough headroom to install a Titan-powered Ampere graphics card later or even a GeForce RTX 3080 Ti.
The GeForce RTX 3080, on the other hand, is a curious case. The graphics card reportedly has the same number of CUDA cores as the GeForce RTX 2080 Ti, but it loses 1GB of VRAM. In defense of Nvidia, the GeForce RTX 3080 received a memory upgrade to increase memory bandwidth by 23.4%. However, the graphics card has a new architecture, so it remains to be seen how the situation plays out.
Nvidia’s big announcement is this Tuesday. It will be interesting to see how the chipmaker sells Ampere to the public.
