Twitter user named @hongxing2020 We recently shared a photo of the mysterious MSI GeForce RTX 3080 Ventus 3X 20G OC. The user claims that up to 100 units are available for between $432 and $576 and these are the rest of the cards in the mining operation. It is hinting that it may be. The core specs are unknown, but the graphics card could be the long-rumored 20GB variant of the vanilla RTX 3080 modified to support Micron memory chips with double the capacity.
This isn’t the first time Nvidia has reportedly attempted to double the VRAM capacity of its high-end GPUs. With nearly every generation of his Nvidia graphics cards, there have been rumors and allegations pointing to his 2x higher capacity variant even after the GPU’s first release.
In the past, Nvidia has sold different memory capacity variants of the same GPU model, with one variant featuring double the memory capacity of the other. That strategy was common in the Kepler generation and older models where cards like the GTX 780 3GB and GTX 770 2GB were supplemented with 6GB and 4GB models respectively.
Creating a new GPU with twice the memory capacity is not difficult. One of the easiest ways to add more VRAM capacity to your GPU. All Nvidia has to do is replace the memory ICs with ones with twice his capacity in previous variants.
If Nvidia wanted to add various levels of VRAM less than double or more, creating such a configuration would be much more complicated. This includes changing memory bus widths and memory organization. But that’s mostly because memory manufacturers choose to make memory ICs based on his 512MB, 1GB, or 2GB memory configuration, with no capacity in between.
We often see memory configurations on some Nvidia GPUs, such as the RTX 3090 and RTX 3090 Ti with 24GB of VRAM and the RTX 3060 with a very strange 12GB of memory. In many cases, if you need more VRAM, it’s easier to double the amount of VRAM. You don’t want to create new dies with more or fewer memory controllers just for a specific capacity, or worse, disable memory controllers to reach less than double the capacity. Nvidia does this. Previous.
However, if Nvidia had considered the RTX 3080 20GB, it would have caused more problems than benefits. Of course, the most important factor is his RTX 3080 12GB. It will be a direct competitor. Especially now that it has been revived this year.
The RTX 3080 12GB is more than just a 2GB capacity upgrade over the RTX 3080. As mentioned earlier, Nvidia had to upgrade his RTX 3080 12GB to add a memory controller and CUDA cores and add his 2GB to the card. Additionally, his CUDA core capacity of the RTX 3080 12GB has been increased by another 256 cores and, best of all, the bus width of the card has been brought to the same specs as the 3080 Ti and RTX 3090.
The RTX 3080 20GB is more of a downgrade for gamers than anything else. 20GB is a perfectly even double upgrade of 10GB, so Nvidia should run the exact same core specs and memory bus width as his 10GB model. This means that the RTX 3080 12GB is noticeably faster than the RTX 3080 20GB in titles that don’t use more than 12GB of memory.
Another issue with the 20GB model is memory usage in games. Only some titles running at very high resolutions and maximum texture packs can destroy his 12GB of VRAM on Nvidia GPUs, thanks to Nvidia’s excellent memory compression technology. As a result, the RTX 3080’s 20GB extra his 8GB of his VRAM is of little use to most gamers.
Technically, the 20GB SKU is a better upgrade than the 10GB model, but with 12GB variants already out there, a 20GB model for gamers doesn’t make sense.The only place where the RTX 3080 20GB makes sense is , the prosumer side of the market where large memory capacities are beneficial for non-gaming tasks such as 3D rendering and high-definition video timelines with various visual effects. But Nvidia already has his RTX 3090 and RTX 3090 Ti that fill this gap, making the RTX 3080 20GB even more pointless.
