RTX 5000 allegedly has 512-bit memory bus (32GB VRAM)

[John Elden Ring]

Nvidia Leaker Kopite7kimi mentions that multiple industry sources have confirmed that the next-generation Nvidia flagship (let's call it the RTX 5090) will feature a 512-bit memory interface. If true, this marks the first time since the GTX 280 and GTX 285 days — all the way back in 2008/2009! — that Nvidia has used a 512-bit bus width.

At a maximum throughput of 32 Gbps, a 512-bit bus width would translate into 2 TB/s throughput — double that of the RTX 4090. Considering how even $1,500 cards are nowadays insufficient to run the latest games at full settings and native resolution (looking at you, Remnant II), there may be something to that idea. With 2GB memory chips, a 512-bit interface would also translate to 32GB of VRAM total on such a GPU, or potentially even 64GB and 128GB professional versions with 4GB chips.

 

[Nhomnhom]

$2199.

 

[PlacidusaX]

A worthy release.

 

[Entropi]

Is the 4090 not able to run some games at native resolutions? That doesn't sound right.

Also: $3000 at least

 

[Frozone]

Nvidia will continue to ship new innovative technology every 2yrs. They really are a monster company to be fair. Totally understandable with the high price tag for their hardware.

 

[ethomaz]

512bits bus in GPU is not something new.
But it is a hell of expensive to build a PCB for that.

 

[JAHGamer]

Is the 4090 not able to run some games at native resolutions? That doesn't sound right.

Also: $3000 at least

It could, but not when Nvidia pushes DLSS so heavily and pays devs to overemphasize its value. It enables both developers and Nvidia to do less work for more money.

 

[Systemshock2023]

If I were Jen tsun wan I'd use this song while introducing the next top of the line GPU.

Nvidia is on top of the game, they should find a way to bring SLI back.

 

[ethomaz]

The sad part is AMD barely doing any competition at all with RDNA3 worst efficiency chip ever created with it boasted FLOPS that doesn’t translate in performance

It is even worst than GCN in efficiency.

 

[anonpuffs]

Delicious... Gonna be needing 600watt for that beast. Hopefully AMD comes back swinging or nvidia gonna be charging $3000...

 

[Systemshock2023]

The sad part is AMD barely doing any competition at all with RDNA3 worst efficiency chip ever created

The truly sad part is that AMD and NVIDIA CEOs are family. They collude like crazy. That's why I want Intel to succeed and more companies to get in the GPU game.

 

[Nhomnhom]

The truly sad part is that AMD and NVIDIA CEOs are family. They collude like crazy. That's why I want Intel to succeed and more companies to get in the GPU game.

Dark times when Intel becomes our hope.

 

[Hezekiah]

Skipping the 5000 series, the reality is a 4090 runs everything at high frames even without frame generation. That will be the case until next gen consoles are released.

Last Nvidia rumour I saw was that the 5000 series this won't be out until 2025 so expecting a nice jump.

 

[anonpuffs]

Skipping the 5000 series, the reality is a 4090 runs everything at high frames even without frame generation. That will be the case until next gen consoles are released.

Last Nvidia rumour I saw was that the 5000 series this won't be out until 2025 so expecting a nice jump.

Depending on pricing, vram capacity, memory bus width, and power usage I could see myself upgrading from my 6800xt to a 5080. Not for $1200 though. And not if it only has 16gb vram.

 

[Hezekiah]

Depending on pricing, vram capacity, memory bus width, and power usage I could see myself upgrading from my 6800xt to a 5080. Not for $1200 though. And not if it only has 16gb vram.

HAS to have more than 16gb VRAM. That much in 2025 would be a pisstake for an 80 card.

 

[anonpuffs]

HAS to have more than 16gb VRAM. That much in 2025 would be a pisstake for an 80 card.

It's nvidia lol