Nvidia RTX 5080 may get faster video RAM than we expected, but doubts remain surrounding this GPU
Nvidia’s RTX 5080 graphics card will likely be loaded with faster video memory than expected, in a new twist in the story of rumors surrounding Team Green’s next-generation GPU.
Wccftech has heard from sources (enable skeptical mode now) who claim that the RTX 5080 will be equipped with 16 GB of GDDR7 VRAM (as previously rumored), but that this RAM will run at 32 Gbps, a faster speed than the 28 Gbps memory that speculations that have been suggested in the past. (Samsung initially has 28Gbps and 32Gbps GDDR7 modules available, as you may recall).
With a 256-bit memory bus and 32Gbps VRAM, as Wccftech points out, the total bandwidth in this case will (just) exceed the magical limit of 1TB/s (1024GB/s to be precise). To put that into perspective, the RTX 4080 offers a total memory bandwidth of 736 GB/s.
Another interesting wrinkle here is the claim that the RTX 5090 will use 28 Gbps of memory, but it will have a much wider memory bus at 512-bit, meaning its total bandwidth will still be much greater than the RTX 5080 , as you’d expect (it’ll actually double close – plus the 5090 will presumably offer 32GB of VRAM).
Ultimately, there’s a good chance the RTX 5090 will also use 32Gbps video RAM, but initially this will be the domain of the RTX 5080 – assuming this information from the grapevine is correct (and all other past rumors ). issue – add spices as always).
Analysis: Memory Matters – Any Compensation from Nvidia?
Okay, so assuming this costs money, why would Nvidia make this VRAM speed choice for the RTX 5080? Well, in case you haven’t noticed yet: there has already been quite a reaction to the (alleged) memory expansion with the RTX 5080. Firstly, with the choice to use that 256-bit bus (remaining the same as on the RTX 4080, and not upgrade there), and more recently the 16 GB VRAM configuration has left many PC gamers dissatisfied.
Nvidia has likely anticipated this, and so this may be Team Green’s way of compensating for these shortcomings to some extent, and ensuring that the overall memory bandwidth is sufficiently snappy. At 1TB/s, or a little more, you can’t really complain – this represents a generational improvement of almost 40% in video memory speed.
Another positive bit of talk is that, while the 16GB RAM expansion is widely seen as Nvidia cutting back on video memory – there could – again – be a 24 GB version of the RTX 5080 on the market.
However, concerns still remain about the RTX 5080 in general, with the rumored core count representing a big drop compared to the RTX 5090. The former will have less than half the CUDA cores of the next-gen flagship, probably a wider gap than ever before. in the RTX 4080 versus the RTX 4090 (where the difference is more like 40% than 50%).
Are we getting a nerfed xx80 series graphics card for the Blackwell generation? That’s what some are suggesting, but honestly such speculation is kind of pointless until we know what Nvidia’s pricing plans are for the RTX 5080. Is the idea to make the GPU a bit more affordable than the RTX 4080?
The problem, though, is that it’s hard to imagine Team Green putting a more reasonable price tag on a 5080 in the given form – it just feels like this is some serious wishful thinking. Still, we can only wait and see. In theory, the RTX 5090 and 5080 could launch together, with an initial reveal at CES 2025 (we could even see Blackwell laptop GPUs).
