Monday, June 30th 2025

NVIDIA GeForce RTX 5080 SUPER Could Feature 24 GB Memory, Increased Power Limits
Hot on the heels of rumored specs of the GeForce RTX 5070 SUPER and RTX 5070 Ti SUPER, rumored specs of the RTX 5080 SUPER emerged on VideoCardz. Apparently, NVIDIA will not tap into the larger "GB202" silicon to build the RTX 5080 SUPER despite maxing out the "GB203" silicon to build the current RTX 5080. It will take a slightly different approach, by giving the card additional memory and power limits. 24 Gbit GDDR7 memory chips are a distinct feature of RTX 50-series SUPER graphics cards, and much like the RTX 5070 Ti SUPER, the RTX 5080 SUPER will feature 24 GB of GDDR7 memory across a 256-bit wide memory bus.
NVIDIA could possibly use 30 Gbps memory speeds for this SKU to end up with at least the same kind of bandwidth as the regular RTX 5080, which is 960 GB/s. The other interesting aspect of the RTX 5080 SUPER is expected to be its increased TGP (total graphics power) value of 415 W, a 15% increase over the 360 W TGP of the RTX 5080. This increase in TGP will not just support the higher density memory chips, but also allow NVIDIA to increase GPU clock speeds. This will likely be necessary, given that NVIDIA has no headroom to increase shader counts on "GB203," and will need something to increase performance for games without heavy memory demands.
Source:
VideoCardz
NVIDIA could possibly use 30 Gbps memory speeds for this SKU to end up with at least the same kind of bandwidth as the regular RTX 5080, which is 960 GB/s. The other interesting aspect of the RTX 5080 SUPER is expected to be its increased TGP (total graphics power) value of 415 W, a 15% increase over the 360 W TGP of the RTX 5080. This increase in TGP will not just support the higher density memory chips, but also allow NVIDIA to increase GPU clock speeds. This will likely be necessary, given that NVIDIA has no headroom to increase shader counts on "GB203," and will need something to increase performance for games without heavy memory demands.
115 Comments on NVIDIA GeForce RTX 5080 SUPER Could Feature 24 GB Memory, Increased Power Limits
Always something with Nvidia. You either get this or that. Why can't we have both?
I guess they don't really need to try very hard as AMD keeps shooting themselves in the foot as well.
RTX 5080 already is very weak upgrade over original RTX 4080 only +15% boost @4k and now this shit appears on horizon. :laugh:
Even RTX 4080 Super have some hardware changes more cores and a huge 1.3% boost over RTX 4080 @4k
It's possible they use binned GPU's for this and can also attain a fairly easy 5-10%+ more performance over a standard 5080 if they actually want to, as virtually anyone overclocking a 5080 already can.
Checking the latest RTX 5080 review, it looks like the RTX 5080, at 360W, is not power limited (less than a 1% performance increase after bumping the power limit 40W... and this after overclocking).
That means, the extra 55W power limit will be wasted, outside of extreme edge cases, and aside from the power increase due to denser memory chips.
Perhaps the extra VRAM will let you keep the card in service a little longer, but I would argue that it will run out of grunt in new games at 4k (where 16GB VRAM may start to be a limit in the near-/medium-term) well before the VRAM pool becomes an issue.
SO, while we can all happy with larger VRAM numbers, I really expect this card to perform 100% identically to the vanilla RTX 5080, while drawing more power, and very likely costing more as well.
Best case? Maybe they throw in a 'free' 5% overclock as stock on the new cards, and save everyone 5min of overclocking :shadedshu:
RTX 5080 was supposed to have >16 GB VRAM to begin with.
RTX 4070 Ti S had 16 GB of VRAM ... Just like RTX 5070 was supposed to have 16 GB VRAM.
Don't buy that piece of shit to get a better future performance/cost it's very simple.
6-8GB - Budget basement cards for 1080p and old and retro gaming with obsolete grade consumer AI
12GB - Low end 1440p gaming and very low end consumer grade AI
16GB - Midrange 1440p+ gaming and low end consumer grade AI
24GB - High end 4K gaming, with texture mods, mid range consumer grade AI
32GB - Enthusiast 4K+ gaming with texture mods, high end consumer grade AI
48-64GB - Enthusiast 5K+ gaming with texture mods and very high end consumer grade AI
12GB, 10GB, 8GB will be to short now or little later.
16GB is enough but gpu itself is too weak..... 24gb will change nothing.
5080 super
- Cores 10275
- TMUs 336
- ROPs 112
- Memory Size 24 GB
- Memory Type GDDR6X
- Bus Width 256 bit
Versus (4090)- Cores 16384
- TMUs 512
- ROPs 176
- Memory Size 24 GB
- Memory Type GDDR6X
- Bus Width 384 bit
It needs to match or beat the 4090 at a lower power and lower price, otherwise whats the point.Remember we are talking about this in mid 2025. A year from now and things will be different, and your $1600 card will be struggling. 16GB cards will be obsolete in 2 years time.
And all that while multistreaming AND playing VR, doesn't hit 16GB (only ever maxxed out around 14GB when really pushing it) so if 16GB really doesn't have longevity, something TERRIBLY WRONG has happened in the game industry. Point's to upsell you into buying a refresh or the twice-as-expensive halo model.
Wait, it's getting GD6X ? not GD7 ?
www.techpowerup.com/review/nvidia-geforce-rtx-5080-founders-edition/34.html
The point is it'd be a similar performing card for a heck of a lot less money, now with no memory deficit if you want to use it for AI or whatever.