New GeForce RTX 4080 Super, RTX 4070 Ti Super, RTX 4070 Super

Post by anonym29 » Mon, Jan 08, 2024, 5:12 PM UTC

jasonjmcghee wrote: post #11

laweijfmvo wrote: post #3

> All three GeForce RTX 40 SUPER Series GPUs are faster than their predecessors It's weird that they're only comparing the new cards to the RTX 30's and 20's, and not the "v1" 40's. I assume the 4080 SUPER is faster than the 4080 (based on name?) but it seems cheaper and there's absolutely no comparison data

This was what I came to comment. Very strange. This article https://beebom.com/nvidia-rtx-4080-4070-ti-super-gpu-specs-r... says "the RTX 4070 Ti Super is up to 10% faster than the non-Super 4070 Ti on average" So, on average, up to 10% faster, yeah seems pretty incremental.

Post by kllrnohj » Mon, Jan 08, 2024, 5:19 PM UTC

alyandon wrote: post #6

lp0_on_fire wrote: post #2

The important bits... > The GeForce RTX 4080 SUPER arrives January 31st, starting at $999 > The GeForce RTX 4070 Ti SUPER launches January 24th, starting at $799 > The GeForce RTX 4070 SUPER launches January 17th, starting at $599

I wonder if nVidia is going to commit to providing enough stock to keep prices at or under those MSRP values.

Post by jasonjmcghee » Mon, Jan 08, 2024, 5:21 PM UTC

anonym29 wrote: post #12

jasonjmcghee wrote: post #11

Earlier quoted context omitted.

This was what I came to comment. Very strange. This article https://beebom.com/nvidia-rtx-4080-4070-ti-super-gpu-specs-r... says "the RTX 4070 Ti Super is up to 10% faster than the non-Super 4070 Ti on average" So, on average, up to 10% faster, yeah seems pretty incremental.

The big difference with the 4070 Ti Super is that it's using the AD103 chip (with a full 256-bit memory bus and 16GB of VRAM) found in the 4080, which is a huge leap over the AD104 chip found in the 4070 Ti (non-Super), which only touts a 192-bit memory bus and 12GB of VRAM. While the TFLOPS of the Super variant does only see a ~10% increase as you note, memory bandwidth jumps by 42% and the memory capacity jumps by…

Post by baobabKoodaa » Mon, Jan 08, 2024, 5:22 PM UTC

loloquwowndueo wrote: post #15

eurekin wrote: post #13

Earlier quoted context omitted.

Thanks! That rules them out completely EDIT: For training

Why?

Post by ShamelessC » Mon, Jan 08, 2024, 5:25 PM UTC

eurekin wrote: post #13

woodson wrote: post #7

If you’re interested in ML training, to save you some time, RTX 4080 Super and RTX 4070 Ti Super have 16GB GDDR6X, the RTX 4070 Super has 12GB.

Thanks! That rules them out completely EDIT: For training