Nvidia is working on its next iteration of the graphics card to be called RTX 30xx lineup. We are already aware that the company is calling the new architecture – Ampere. With AMD offering tough competition to the brand, the company is working hard to launch the product as soon as possible. In fact, the new lineup of the graphics card might see its official unveiling as soon as March. For those who’re unaware, Nvidia will host its GTC conference in the month of March which is nearing. Well, we now have the very first information on the Nvidia RTX 3080 and RTX 3070 GPU.
Nvidia GeForce RTX 3080, RTX 3070: Specs, Features
As per the recent emerging report, two new GPUs have appeared online with core codes GA103 and GA104. Both the GPUs are part of the high-end offering of the Ampere family. In an unrelated note, this is the first time that an Nvidia core code is ending with a numerical 3. Coming to the specifications, the GA103 will come with 10/20GB GDDR6 VRAM capacity. IT will also feature SM (Streaming Multiprocessors) count of 60, 3480 stream processors and 320-bit video memory.
Gizchina News of the week
The GA104, on the other hand, will feature 8/16GB GDDR6 VRAM memory capacity along with 48 sets of SM arrays. There will be 3072 stream processors and 256-bit video memory. Looking at the specs, the GA103 points towards the Nvidia RTX 3080 GPU while the latter indicates RTX 3070. Both the new GPUs will be manufactured using 7nm architecture. It still remains unknown who will manufacture the graphics card. We might see Samsung doing that considering good relations of the South Korean band and Nvidia. Furthermore, TSMC is already pretty packed with its current 7nm production capacity.
When compared with the previous generations, the all-new RTX seems to offer significant improvements if the leak is to be believed. The SM count and memory capacity has been greatly improved. Even the current high-end RTX 2080 Super comes with 8GB of video memory capacity. It would be interesting to see how well the new Ampere GPUs will perform.
“the GA103 will come with 10/20GB GDDR6 VRAM capacity. IT will also feature SM (Streaming Multiprocessors) count of 60, 3480 stream processors and 320-bit video memory.”
then
“The GA104, on the other hand, will feature 8/16GB GDDR6 VRAM memory capacity along with 48 sets of SM arrays. There will be 3072 stream processors and 256-bit video”
So the G104 according to your own reporting is the lesser of the two. Making it align with the 3070 in this context.
Good grief.
Explain, i don’t understand.
The “rumored” spec leaks would put this thing 4-10% slower then a 2080Ti. (Not taking into account architectural upgrades, and faster ram.)
This probably means GA 102 probably has 72SM , with next gen Titan rtx/rtx 3080ti having 72/68SM probably and 32/16 GB vram.
So when AMD are looking at a significant boost to current 2080ti speeds, Nvidia are going for the same old 30% max over the previous generation.
Where they out done AMD with the 1080ti (and themselves, evident from the low 2080ti sales for average 5-10% advantage in games over a 1080ti) Yhey look set to lose out in this generation to the new Navi range…