Monday, January 6th 2025
First NVIDIA GeForce RTX 5090 GPU with 32 GB GDDR7 Memory Leaks Ahead of CES Keynote
NVIDIA's unannounced GeForce RTX 5090 graphics card has leaked, confirming key specifications of the next-generation GPU. Thanks to exclusive information from VideoCardz, we can see the packaging of Inno3D's RTX 5090 iChill X3 model, which confirms that the graphics card will feature 32 GB of GDDR7 memory. The leaked materials show that Inno3D's variant will use a 3.5-slot cooling system, suggesting significant cooling requirements for the flagship card. According to earlier leaks, the RTX 5090 will be based on the GB202 GPU and include 21,760 CUDA cores. The card's memory system is a significant upgrade, with its 32 GB of GDDR7 memory running on a 512-bit memory bus at 28 Gbps, capable of delivering nearly 1.8 TB/s of bandwidth. This represents twice the memory capacity of the upcoming RTX 5080, which is expected to ship with 16 GB capacity but 30 Gbps GDDR7 modules.
Power consumption has increased significantly, with the RTX 5090's TDP rated at 575 W and TGP of 600 W, marking a 125-watt increase over the previous RTX 4090 in raw TDP. NVIDIA is scheduled to hold its CES keynote today at 06:30 pm PT time, where the company is expected to announce several new graphics cards officially. The lineup should include the RTX 5090, RTX 5080, RTX 5070 Ti, RTX 5070, and an RTX 5090D model specifically for the Chinese market. Early indications are that the RTX 5080 will be the first card to reach consumers, with a planned release date of January 21st. Release dates for other models, including the flagship RTX 5090, have not yet been confirmed. The RTX 5090 is currently the only card in the RTX 50 series planned to use the GB202 GPU. Pricing information and additional specifications are expected to be revealed during the upcoming announcement.
Source:
VideoCardz
Power consumption has increased significantly, with the RTX 5090's TDP rated at 575 W and TGP of 600 W, marking a 125-watt increase over the previous RTX 4090 in raw TDP. NVIDIA is scheduled to hold its CES keynote today at 06:30 pm PT time, where the company is expected to announce several new graphics cards officially. The lineup should include the RTX 5090, RTX 5080, RTX 5070 Ti, RTX 5070, and an RTX 5090D model specifically for the Chinese market. Early indications are that the RTX 5080 will be the first card to reach consumers, with a planned release date of January 21st. Release dates for other models, including the flagship RTX 5090, have not yet been confirmed. The RTX 5090 is currently the only card in the RTX 50 series planned to use the GB202 GPU. Pricing information and additional specifications are expected to be revealed during the upcoming announcement.
81 Comments on First NVIDIA GeForce RTX 5090 GPU with 32 GB GDDR7 Memory Leaks Ahead of CES Keynote
The greater VRAM is for AI related stuff.
Wish that I could afford this :love:
that's said it's stupid and too expensive
What I whish is to have 5080 cards with only 2,5 slot thickness or less, not even dreaming about a "lean" 5090.
If it's not more than 30% faster generally over the outgoing 4090, technically essentially it's a bit of a waste of time.
High-end workstations/AI should have been for the Quadro lineup
This 5090 with 32GB is not really for gaming per se, this is again for AI, no game requires 32GB at 4k and the price will suffer for people who wanted this just for gaming
if DLSS4 required more vRam buffers than the 5080 surely would have had more than 16GB..
so yes once again,..this is an AI company now, given how much billions it generates I understand it but I’m in the market for a high-end GAMING gpu
They have gaming as a afterthought at this point, if they could they would stop selling just gaming gpus probably. They shitify the low end with low vram and insane prices and try to move us all to spend more and more, and if the AI crowd could snatch all the high end even better so they can ask more and more money with each generation.
We need EU to step up and stop this nonsense, since no other Country outside EU seems not to care.
Is it because of power, the people that travel in private jets will stop other people from using whatever gpu they want?
In a few years, electric consumption will not be a problem for the environment. However, CO2 it's and will be. The same way, water consumption in datacenters it's a problem. But not 600W in a home GPU.
People (used to) worry about CPU cooler height but the GPU makers basically went "hold my beer" and just went nuts.
It goes in the wrong direction about power consumption for gpu's. I hope performance are there after.
Your 'provider' company may generate from renewables but what gets put into the grid and taken out by everyone connected will be a mix, e.g.:
A bit shy of 50%...
www.iea.org/countries/spain/electricity
Combine this with the push by some for BTF motherboards necessitating a thicker area behind the motherboard and cases are gonna become more cube shaped. The chart is just the overall mix - not renewable or fossil specific. Guy claimed it was 50% - it isn't. Nuclear isn't renewable but also doesn't count towards CO2 emissions (technically).
Maybe there's demand still to put it in LLM, but still seems like there's few revenue generating things that can be done with those. Plus you can rent GPU time for LLM from datacenters anyway and prices been falling like a rock for that...
Quadro is not what you think it is. Quadro is specifically for programs (CAD, CAM mostly) that require the Quadro drivers and certification. It's a seperate market from CUDA tasks. The super high end CUDA market are the accelerators way above all this that few companies can even afford.
Nvidia stopped being a gaming company when they released the 8800 GTX and CUDA. They have stated this. Repeatedly. You keep not getting that.
The high end GAMING gpu is the RTX 5080. Period. Anything higher than that you are buying a PROFESSIONAL gpu and not really using it and setting money on fire to game on it. That's only for products that require the Quadro drivers such as CAD, CAM. Titan and X090 are prosumer for CUDA based stuff and nvidia does have Creator drivers for the Geforce line as well. Nvidia has been very clear about it. They are not a graphics company since the 8800 GTX and CUDA. These super high end cards are CUDA products and not gaming products. But if you have the money you can game on them.
That gamers don't want to believe what has been true for over a decade now and has been repeatedly stated by nvidia is just gamer syndrome at it's finest. Serve it up with a cup of crying about women in video games not being hentai'd up and you have the perfect cocktail.
The dilution of features between the different product ranges is great for tinkerers and people developing on a budget, but has been detrimental to some product segments. I'm surprised that Nvidia haven't rectified this, not from some sort of attempt to maintain supply or stimulate sales at lower prices, but purely to maintain profit on selling higher end / higher feature parts for higher prices / profit margins.
There is a reason why so many spyware infected Chinese companies are banned in US and EU.
Power regulation is the same. Why do you think there are regulations for housing appliances and how much power they can use?