Friday, March 25th 2022
NVIDIA GeForce RTX 4090/4080 to Feature up to 24 GB of GDDR6X Memory and 600 Watt Board Power
After the data center-oriented Hopper architecture launch, NVIDIA is slowly preparing to transition the consumer section to new, gaming-focused designs codenamed Ada Lovelace. For starters, the source claims that NVIDIA is using the upcoming GeForce RTX 3090 Ti GPU as a test run for the next-generation Ada Lovelace AD102 GPU. Thanks to the authorities over at Igor's Lab, we have some additional information about the upcoming lineup. We have a sneak peek of a few features regarding the top-end GeForce RTX 4080 and RTX 4090 GPU SKUs. According to Igor's claims, NVIDIA is testing the PCIe Gen5 power connector and wants to see how it fares with the biggest GA102 SKU - GeForce RTX 3090 Ti.
Additionally, we find that the AD102 GPU is supposed to be pin-compatible with GA102. This means that the number of pins located on GA102 is the same as what we are going to see on AD102. There are 12 places for memory modules on the AD102 reference design board, resulting in up to 24 GB of GDDR6X memory. As much as 24 voltage converters surround the GPU, NVIDIA will likely implement uP9512 SKU. It can drive eight phases, resulting in three voltage converters per phase, ensuring proper power delivery. The total board power (TBP) is likely rated at up to 600 Watts, meaning that the GPU, memory, and power delivery combined output 600 Watts of heat. Igor notes that board partners will bundle 12+4 (12VHPWR) to four 8-pin (PCIe old) converters to enable PSU compatibility.
Source:
Igor's Lab
Additionally, we find that the AD102 GPU is supposed to be pin-compatible with GA102. This means that the number of pins located on GA102 is the same as what we are going to see on AD102. There are 12 places for memory modules on the AD102 reference design board, resulting in up to 24 GB of GDDR6X memory. As much as 24 voltage converters surround the GPU, NVIDIA will likely implement uP9512 SKU. It can drive eight phases, resulting in three voltage converters per phase, ensuring proper power delivery. The total board power (TBP) is likely rated at up to 600 Watts, meaning that the GPU, memory, and power delivery combined output 600 Watts of heat. Igor notes that board partners will bundle 12+4 (12VHPWR) to four 8-pin (PCIe old) converters to enable PSU compatibility.
107 Comments on NVIDIA GeForce RTX 4090/4080 to Feature up to 24 GB of GDDR6X Memory and 600 Watt Board Power
Is there any laws or regulations world wide limiting this? Semiconductor makers and designers should be subject to the same regulations that apply for general household appliances and car manufacturers, where efficiency is prioritized, we need more FLOPs per Watt.
Dunno about the 4080 ...
I'd be surprised if this thing tops 450w stock! Those extra watts have always been aimed at insane overclocker dreamers (because, 50% higher power consumption for 15% faster performance == GUD)!
Man, and people said the R9 390 and Vega 64 were power hungry. Damn.
edit: Ah, it was the last year's one, missed that.
but there will be people that will want that and i don't see the big deal if Nvidia is making it. Nobody will be forced to use those cards. I suspect that they will be very expensive as the card will need to have a really beefy power section to be able to manage and deliver all this power cleanly to a GPU that will boost up and down.
I think the problem would be if those cards would be mainstream. but the titans just showed both Nvidia and AMD that some people will pay any amount just to have the best of the best. Those cards are for those people and i am not sure if that is a significant portion of the gaming market.
If Nvidia deliver a high end (not top end) cards that is way better than what we currently have and run at 300w, i don't see that as a problem.
The other thing is how much power AMD will use to achieve the same performance. It remain to be seen but rumors is RDNA3 will have an efficiency advantage and that is the main reason Nvidia is pushing so hard those GPU. If AMD cards use much less power to deliver same performance, the card should also be cheaper to produce since it will require less cooling, less power stages, etc..
Nvidia had those series of GPU that were not that much competitive and got pushed really hard. but they always fought back quickly. Time will tell.
It kinda sucks that they just put out more and more power hungry cards instead of increasing the efficiency. You guys remember how efficient Maxwell and Pascal were? Damn.
24 phases for voltage is hardly anything to write home about on a high-end SKU, considering the 3090 FE has 19 and 3090 overclocking SKUs can have even more (e.g. Galax RTX 3090 Hall of Fame packs in 25).There is no such thing as a "PCIe Gen5 power connector". There is only the 12VHPWR connector which is part of Intel's ATX 3.0 specification.It's "fares".Igor does not "note" anything. He does, however, make a large number of very bold claims.
If I ran the GPU in the UK without undervolt and no FPS cap, I think I would pay more in electric than buying the card in the first year. :)