Monday, April 28th 2025

NVIDIA Reportedly Working on GeForce RTX 5080 SUPER 24 GB & RTX 5070 SUPER 18 GB Designs

Mere months after the launch of GeForce RTX 5080 16 GB and RTX 5070 12 GB graphics cards, industry rumors are already swirling about a potential "SUPER" semi-successor. Sections of NVIDIA's previous-gen "Ada Lovelace" GeForce RTX 40-series family received mid-generation refresh treatment around early 2024. It is not clear whether it will take a year+ for the speculated arrival of upgraded "Blackwell" desktop gaming solutions, but members of the Chiphell forum have been openly discussing alleged "in-progress" GeForce RTX 5080 SUPER 24 GB and RTX 5070 SUPER 18 GB card designs. Past weekend theorizations were highlighted by ITHome and VideoCardz—in particular, one Chiphell participant posited the two follow-ups will not be deployed with noticeably "faster" performance.

Larger pools of onboard GDDR7 VRAM could pave the way for improvements in AI productivity, although greater capacities could lead to beneficial conditions in gaming scenarios—e.g. extra provisions for large textures and complex assets. The speculated GeForce RTX 5080 SUPER card could borrow aspects from Team Green's GeForce RTX 5090 Mobile 24 GB model; namely the latter's usage of 3 GB GDDR7 memory modules. Press interpretations—of fresh Chiphell leaks—point to company engineers enabling existing PCB designs with 8×3 GB (24 GB) and 6×3 GB (18 GB) parts; thus preventing a major overhaul of board layouts. Global PC gaming hardware communities have often expressed a dislike of NVIDIA's repeated deployment of 8 GB and 12 GB capacity products. A theoretical GeForce RTX 5070 SUPER 18 GB option could receive a warmer welcome. Very early speculation proposes an introduction—of SUPER "Blackwell" models—just before or during CES 2026.
Sources: Chiphell, VideoCardz, Wccftech, TechRadar, Hot Hardware, Tom's Hardware, ITHome
Add your own comment

76 Comments on NVIDIA Reportedly Working on GeForce RTX 5080 SUPER 24 GB & RTX 5070 SUPER 18 GB Designs

#26
Prima.Vera
This is a complete mess.
The 5070 is 12GB-192bit card
The 5070 Ti is 16GB-256bit card
The 5070 Super is 18GB-192bit card??
Posted on Reply
#27
Onasi
Sol_Badguy@Onasi

I was under the impression that it does work. If not then my bad. I see others have thought about this.
But the good news then is that nVidia is forced to make more hefty VRAM bumps.
So if I was wrong then I'm glad I was. :)
VRAM fuckery CAN be done, but the usual modern way is to make sure it doesn’t result in an asymmetrical config. The 3060 8Gb mentioned in the thread you linked is running a 128-bit bus as opposed to 192-bit on the OG 3060 12Gb. And it, from what I have seen, only has chips for 8Gb. Other examples are a product of VBIOS hacking and are usually a result of a card already damaged/faulty with said hacks employed as a temporary solution. A 15Gb GPU just cannot be done without either going asymmetrical with memory/bus or resorting to firmware shenanigans. Even then, it would be extremely inefficient in terms of production costs.
Posted on Reply
#28
Sol_Badguy
@Onasi

Going back to the GTX 550 Ti and GTX 660 Ti for which nVidia used mismatched capacity modules, I somehow thought that if a memory controller can access a chip of a different capacity then the same controller can access a chip of no capacity i.e. no chip and achieve the same result but in reverse.
Well I am glad I was wrong, it means nVidia has less methods that can be used to gimp cards.
Posted on Reply
#29
mama
More memory on the higher skews where it is arguable as to how useful that is and still an 8GB card for the masses. So out of touch.
Posted on Reply
#30
phanbuey
So basically a 4090 2 years later. Still $1699
Posted on Reply
#31
igormp
Sol_Badguy@Onasi

I was under the impression that it does work. If not then my bad. I see others have thought about this.
But the good news then is that nVidia is forced to make more hefty VRAM bumps.
So if I was wrong then I'm glad I was. :)
You'd be effectively reducing the bus-size, which would reduce the memory bandwidth, and thus overall performance.

So I'd say you can discard the idea of refreshes that have a smaller bus width than their regular counterparts.
Posted on Reply
#32
Ninehell
Got myself 5070ti not even a week ago. I guess I can go for 5080 super for vram to replace my 4060. It probably gonna cost $1800+ here
Posted on Reply
#33
Dr_b_
everyone rightfully complaining about 8GB 5060, but 5080 @ actual retail for $1400+ AIB cards, is DOA with only 16GB
Posted on Reply
#34
tsunami2311
5070 with 18gb could be nice but know nvidia they will tack on $100 or more. on an already over priced sku, and xx80 even worse

But none this matter if there is not stock let alone stock at msrp
Posted on Reply
#35
Hyderz
5080 24gb super will be a waste of extra vram if the performance jump is like the 4080 to 4080 super and 4080 super to 5080....
Posted on Reply
#36
A Computer Guy
Event HorizonSomeone here called it, and they've done this before. Launch with lower than expected VRAM and then refresh with the "right amount".
If we could only get the right amount of 12VHPWR connectors that would be a relief.
Posted on Reply
#37
RejZoR
AusWolfMedia coverage.
They got enough of that because of broken drivers...
Posted on Reply
#38
cinemaware
It's amazing how some folks take this as gospel, even thought all that happened is that somebody on a forum somewhere said something.

(btw, there's only one primary "source", far as I know, all the other outlets just posted about it ).
Posted on Reply
#39
Chomiq
5080 should have 24 gigs of vram from the start.
Posted on Reply
#40
Vayra86
Event HorizonSomeone here called it, and they've done this before. Launch with lower than expected VRAM and then refresh with the "right amount".
Yup.
Sol_BadguyVery juicy subject. I like it, I like it a lot.

A full-die type Super is only possible with the GB205 (currently used on the 5070) and the GB202 (currently used on the 5090).
Regarding the GB202 I am willing to bet money that nVidia will not give such a unicorn to the consumer market.

But the GB205 is very likely. With perhaps a slight speed bump it will probably be a bit faster (in raster) than the 9070.
Landing at the same $550 MSRP as the 5070 it will replace means it could potentially be disruptive against the 9070 but not so much for the 9070 XT which will still be faster.
However regarding VRAM capacity, going from 12 to 18 would be most welcome, and going with 18 would definitely help the "5070 Super" be disruptive against the 9070 XT, although giving so much more at the same MSRP has a big question mark next to it.
nVidia sort of did it before with the 4070 Ti Super which represented a substantial upgrade over the 4070 Ti, compare this with the absolute joke of going from 4080 to 4080 Super, that was mostly a PR move for using the "MSRP reduction card".

Back to the 18GB dilemma, this move while very effective against 9070 series cards forces nVidia to make substantial VRAM upgrades further up the stack. An 18GB 5070 Super would make sense next to the 5060 Ti 16GB but would make the current 5070 Ti with 16GB look gimped, meaning the 5070 Ti needs at least 18GB or more to make sense in the lineup. For the 5070 Ti 18GB is possible with 3GB chips by enabling only 6 out of 8. Or perhaps 21GB by enabling 7 out of 8.
And so for the 5080 the possibility is with 21GB or 24GB. To give users 50% more VRAM at the same MSRP for the 5080 I imagine nVidia is foaming at the mouth.
And all of this thanks to "good guy" AMD.

The 5060 Ti is in a weird position, 3GB chips would make a 12GB version possible which has the advantage of having chips on only one side of the PCB and also 50% bump over the 8GB version but it would represent a buffer downgrade compared to the current 16GB version.
If this 12GB version gets the $380 MSRP then it would be an okay deal, but still coexisting with the 16GB version would be very weird.

There is another version, one in which nVidia is more stingy with the VRAM. So the 5070 Super gets only 15GB by having 5 out of 6 chips enabled. This means the 5070 Super won't be as disruptive against the 9070 series cards but this version is almost guaranteed to land at the same $550 MSRP as the 5070. Compared to the 5060 Ti 16GB (if that is still kept) it would look slightly gimped. But this means that the 5070 Ti could remain unchanged, or if nVidia wants to use 3GB chips across the stack then 18GB is the most likely. Then for the 5080 18GB is still possible but 21GB is more likely. There would be no reason that nVidia will go for 24GB.
In this version the lineup could look like this:
5060 Ti Semi-Super 12GB - $380 MSRP
5070 Super 15GB - $550 MSRP
5070 Ti Super 18GB - $750 MSRP
5080 Super 21GB - $1000 MSRP
3GB increments, only the 5060 Ti gets all chips enabled. 5060 Ti 16GB is discontinued and everything else higher up is replaced.
The lower-tier cards can still be countered by AMD's offer, maybe with a $50 price cut.

The rationale for the 50 series refresh is simply a testing ground for the 3GB GDDR7 chips, to see how they perform in dGPU consumer products in preparation for the 60 series. And of course to maintain the appearance of progress, of giving more for same or less money and all of that.
This is the sole reason in my opinion, there is no problem with how attractive the current cards are, the market is already adjusting all GPUs according to the meaningful variables: supply and demand, MSRP and actual performance. We see that the 9070 series were pushed upwards towards the 5070 Ti and not the other way around. We see the 5070 getting lower street price compared to the 9070.
They ain't gonna reduce bus widths I reckon just to arrive at memory configs, but other than that, interesting look at things
Posted on Reply
#41
Wirko
OnasiThat would still end up being an asymmetrical configuration that cannot work on a 192-bit bus. You can’t just “disable” a memory chip and keep the same bus. I suppose theoretically it can be done in firmware to make only 15 gigs accessible while being physically and electrically present, just not technically addressable by applications, but that’s silly since you still end up paying for all 6 chips in manufacturing.
To make more money, you need to think out of the box.

15 GB : 6 chips = 2.5 GB per chip

No bandwidth lost.

Where do all defective 3GB RAM chips go, some of them with a single bad bit? Yeah, those dies get sorted and put on storage. Later, when stock is large enough, they will get labelled as 2.5 GB and sold for 25% less. Fewer bad chips dusted down - everybody wins. JHH wins more.

History lesson: The Sinclair ZX Spectrum 48 KB 8-bit computer had eight 4532 DRAM chips (32-kilobit each, 200 ns cycle time) for the upper 32 KB. They were actually 4164 chips (64-kilobit each) with one half working. The damn Z80 processor even had logic for DRAM refresh buried somewhere within its 8500 transistors.
Posted on Reply
#42
Dimitriman
Only Jensen's hubris could sell more Ai software as "generational upgrades", but also decline to increase VRAM capacity, kneecapping the card's ability to perform such Ai tasks.
Posted on Reply
#43
Sol_Badguy
@igormp @Vayra86

I never implied bus width reductions in my first post.
I (wrongly) assumed that disabling chips is still possible while keeping the same bus width. As I commented in a subsequent post:
Sol_BadguyGoing back to the GTX 550 Ti and GTX 660 Ti for which nVidia used mismatched capacity modules, I somehow thought that if a memory controller can access a chip of a different capacity then the same controller can access a chip of no capacity i.e. no chip and achieve the same result but in reverse.
Well I am glad I was wrong, it means nVidia has less methods that can be used to gimp cards.
Again I was wrong.

However there are still potential product segmentation issues with the future lineup.
As I said, for the people that are planning to get a 5060 Ti 16GB, that version is perfectly fine (for them, that's why they want it), sure it wouldn't hurt if it was cheaper.
But for the 8GB version to be replaced by the 12GB one and having two 5060 Tis coexisting I find questionable.
The 12GB version replacing both current versions means it will upset a lot of people who are satisfied with the performance level of the 5060 Ti but want extra VRAM for other workloads. They will get a downgrade if they wait until then and possibly miss out on the 16GB version.

Also the 5070 Ti would be in a peculiar position being sandwiched between the 5070 Super with 18GB and the 5080 Super with 24GB. Receiving the same 24GB as the 5080 Super I find implausible, that would basically be a gift for the history books made by nVidia. At the same MSRP I mean.
So keeping the 5070 Ti untouched means that relative to the rest of the lineup it will have worse value than in today's context.
But maybe that's the plan, to upsell the 5080 Super.
Being discontinued is also possible (but improbable in my opinion), although there will be a very big gap between the 5070 Super and the 5080 Super, a gap too big to be effective for upselling.

Anyway the 5080 Super with 24GB (for the same MSRP as 5080) would be very good value (for non-gaming especially) in my opinion, the previous card from nVidia with that kind of buffer was the 4090 at $1600. Yes it is more powerful but value-wise it would get smoked by this 5080 Super.
And compared to the 7900 XTX there is no contest, that one had the same MSRP but productivity-wise it doesn't have the same potential.

So if a 50-series refresh does happen, I feel that it would make people happier rather than unhappier assuming it will be done with a similar approach to the 40-series refresh.
But more importantly available stock is what's needed now and then, not a lineup refresh with paper launches.
Only if the stocks are plentiful and cards are sitting on shelves will the retailers be forced to lower prices.
Posted on Reply
#44
AusWolf
RejZoRThey got enough of that because of broken drivers...
Yep. And they're trying to fix that more cards that no one asked for.
Event HorizonSomeone here called it, and they've done this before. Launch with lower than expected VRAM and then refresh with the "right amount".
Might have been me. :D

It's pretty basic maths, though. They've been playing the "look, there's more" card since Turing. It's getting boring if you ask me.
Posted on Reply
#45
mgk
Can't stop laughing at these comments honestly. How are some actually believing that the 5070 super will have 18gb?I mean yeah it can happen but then the msrp will be way higher that 5070ti is. It also doesn't make any sense at all to put more vram on 5070s over the 5070ti and especially the 5080,they just sabotage themselves. People need to realize Nvidia is not making gpus to last for a decade,they almost never put the right amount of vram just so you upgrade more frequently. 4070 super and ti had same vram as the base 4070,same goes with the 4080 super. It's unlikely for them to add this amount on the super models.Maybe the 5080 super will have like 20gb?As the difference is too big from it to the 32gb rtx 5090,they did this small upgrade back with the 3080 10gb up to the 3080ti 12gb as the difference was too big against the 24gb 3090.I believe they are most likely to release these "much bigger vram models" on the 60 series.As they did in the past,3070 had the same 8gb vram as 2070 & 2070s and not even the 3070ti got upgraded,while the 3060 had 12gb,basically same as this gen with the 5070 and 5060ti.
Posted on Reply
#46
OneSorcerer
mgkCan't stop laughing at these comments honestly. How are some actually believing that the 5070 super will have 18gb?I mean yeah it can happen but then the msrp will be way higher that 5070ti is. It also doesn't make any sense at all to put more vram on 5070s over the 5070ti and especially the 5080,they just sabotage themselves. People need to realize Nvidia is not making gpus to last for a decade,they almost never put the right amount of vram just so you upgrade more frequently. 4070 super and ti had same vram as the base 4070,same goes with the 4080 super. It's unlikely for them to add this amount on the super models.Maybe the 5080 super will have like 20gb?As the difference is too big from it to the 32gb rtx 5090,they did this small upgrade back with the 3080 10gb up to the 3080ti 12gb as the difference was too big against the 24gb 3090.I believe they are most likely to release these "much bigger vram models" on the 60 series.As they did in the past,3070 had the same 8gb vram as 2070 & 2070s and not even the 3070ti got upgraded,while the 3060 had 12gb,basically same as this gen with the 5070 and 5060ti.
I find more plausible we get a 5060 ti refresh with 12GB with a 10% perf. improvement, matching a 4070 in raster performance.
Posted on Reply
#47
DarkChib
Worst is that I want a 5080 but the reference design is so good now that I will only buy it, but you cant get any of the stock for love nor money and their preordering is only in the US right now. 5080 Super with a reorder option and I will buy one!
Posted on Reply
#48
csendesmark
Hope they will produce a 32GB VRAM version too! :toast:
Posted on Reply
#49
Sol_Badguy
mgkHow are some actually believing that the 5070 super will have 18gb?I mean yeah it can happen but then the msrp will be way higher that 5070ti is.
Then how do you explain the 4070 Ti Super replacing the 4070 Ti at the same MSRP?
mgkPeople need to realize Nvidia is not making gpus to last for a decade,they almost never put the right amount of vram just so you upgrade more frequently.
The 4060 Ti 16GB has more VRAM than GPU muscle.
The 5060 Ti 16GB has more VRAM than GPU muscle.
The 4090 has more VRAM than GPU muscle.
The 5090 has more VRAM than GPU muscle.
Therefore nVidia must be retarded, right?
And yes make no mistake a game that would go over 20GB would absolutely murder the 4090, a 5090 would also succumb before 30GB.
These cards (and surely some other ones) have more VRAM than they can take advantage of for gaming.
And when you will need to upgrade from them it won't be because of insufficient VRAM but because of insufficient GPU power.

Look here at the difference (in 4K) between the 5080 and 4080 Super, 11.5% more for the 5080. Achieved on a similar sized die and the same process node. Now look at the difference between the 4090 and the 5080, 16% more for the 4090.
Imagine the RTX 6080, similar sized die to the 4080 Super and 5080, but on a smaller node. More processing units in the same die space. It's reasonable to expect a performance level over the 4090, a card with 24GB. What reaction do you think the public will have if nVidia keeps the same 16GB buffer on the 6080?
mgkI believe they are most likely to release these "much bigger vram models" on the 60 series.
nVidia doesn't want to risk other GPU makers launching consumer cards with 3GB chips first, they want to be the first at everything, marketing/PR is one of nVidia's biggest strengths.
Plus it's a good opportunity to test a larger buffer before the next-gen, they also give the AIBs and chip makers more work, which means more money, everybody happy.
Posted on Reply
#50
matar
( SUPER stands for )
Super hard to find Anywhere
Super hard to find at MSRP
Super money for Scalpers
Posted on Reply
Add your own comment
Nov 19th, 2026 17:02 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts