Nvidia's RTX 5080 is Actually an RTX 5070 - Wait, What?

I was able to get the 5090 FE model.
I never would have gone for the 5080 because I wanted, ultimately, to get the maximum amount of VRAM available. I always expect the "90" version to simply be the best of the best until a "90 Ti" version comes due.

But lets face it: the games mostly are lackluster and don't demand this level of performance - performance you can only really trace using benchmark programs. If the performance is noticeably poor with my eyes, that's when I know I need to upgrade hardware. At this point we are just upgrading to upgrade. I saw a huge difference playing Cyberpunk when I upgraded from a 2080Ti to a 3090. No benchmark...I just "saw" it. I'm sure the same amount of fun I'm having with my 5090, I'd be able to have with a 5080 or my old 4090 or my old 3090. The games themselves just don't demand this kind of investment.

As for Deepseek... that's just a kneejerk reaction to "there's something better". Most investors really don't understand what it is Nvidia represents and they're just shuffling money around. Deepseek is programming. What they don't have is hardware. They don't cater to a captive audience of computer gamers willing to drop thousands of dollars of video game parts.

You make very strange choices. If this is your take on games (and I agree) you had no reason to upgrade from the 3090, and even less reason to go from 4090 to 5090. That's just mindboggling.
 
I completely understand the reasoning of the Author and do not disagree. But as we have seen the cards sell out instantly. So I completely understand why Nvidia is doing what its doing. In fact, based on the sales numbers I believe Nvidia will learn that they should price higher. Maybe next year we will get a 60 class card for an 80 class price. That will be because there is enough demand and money in the market to warrant it.

And that will suck yes but these are complete luxury goods, we have absolutely no entitlement to a reasonably priced brand-new graphics card. It's like complaining that Porsches cost too much.

In short, its futile to complain about the pricing of these things. It's not going to change in the direction you want. It's going to continue to get more and more expensive to buy the latest GPUs. Just as its more and more expensive to buy the fastest sports cars every year. You should just accept it and enjoy the product for what it is or get a different hobby. I'm so fed up with hearing people whining about pricing in this industry.
 
People waited days and some hours just to buy one 5080 at $1500 when the 4080 super was available at $949 on black Friday. Some drove hours just to come out empty handed too.

🤡
That is stupidity at it's finest hour. How the puck should you pay 50% more for 10% more performance? And that waiting in line at the MC store reminds me of the days that Communism was still a thing in my country.
 
The 5080 comes undertuned from the factory as well. Jays2cents tested overclocking and it easily clocked 350mhz on the gpu and 500mhz on the memory with 0 instability. This gave him an almost 10% increase in performance. This shows us that there absolutely will be a model named 5080 super or whatever and Nvidia doesn't have to do anything - they can just up the speed on a factory level.
Then they can release a "ti" model with just using 24gigabyte and 384bit bus . They have plans for sure
 
The 5080 comes undertuned from the factory as well. Jays2cents tested overclocking and it easily clocked 350mhz on the gpu and 500mhz on the memory with 0 instability. This gave him an almost 10% increase in performance. This shows us that there absolutely will be a model named 5080 super or whatever and Nvidia doesn't have to do anything - they can just up the speed on a factory level.
Then they can release a "ti" model with just using 24gigabyte and 384bit bus . They have plans for sure
ASUS GeForce RTX 5080 Astral OC Review - Relative Performance | TechPowerUp https://search.app/peqquUGPaJzxSoDR6

3.3 ghz on core and 375 plus on vram a
at 400 watts max power just to come 5% of stock 4090. FYI
 

Attachments

  • relative-performance-3840-2160 (2).png
    relative-performance-3840-2160 (2).png
    98.7 KB · Views: 5
  • overclocked-performance (1).png
    overclocked-performance (1).png
    35.5 KB · Views: 5
Thx, great article ! Interesting and well written, every line was worth reading.
I feel TechSpot is raising the bar with these kind of articles (like the DLSS4 MFG in-depth review).

There was a time when computer stuff became better/faster while getting cheaper at the same time ... late '90s, I remember buying an internal HP CD-writer (1x or 2x speed writing) for what would be 250€ at that time ... a couple of years later, you had combo CD/DVD writers that were a lot better and faster and they were costing 50€ :D

Granted, not the same type of hardware as graphics cards ... still, a high-end card that is almost similar (price and performance) to cards from a couple of years ago is indeed underwhelming.
When looking at a performance graph, it's certainly not stagnating (just look how steep the graphs drop for the first couple of cards, like 5090 to 4090 to ...), it's that this performance seems to (artificially ?) stay in the very high-end instead of moving down to mid tier cards ...
 
The GTX285 was faster and launched at only $359, by then a perfectly reasonable proposition compared to the $650 GTX280 from the very same year. Only AMD competition prevented Nvidia gouging, and this was fifteen years ago. Nothing can be done today unless there is a good part from a rival. Let's hope 9070XT is that part.
Yes, that would be great :) ... imagine AMD 'not pursuing the high-end' because they consider the 9070XT to be mid range this generation: meaning same performance as last gen high-end, but much lower price ?
 
This is exactly why I wouldn't pay for a 4080, but happily went for the 4070 Ti Super. It's flatly a superior card for the money and the lower power requirements will give it a long life. Sure they can keep fighting physics and making bigger cards that burn through more power, but that's barely progress. That balance of efficiency is everything, and they're losing it.
 
People keep buying Nvidia cards so there is no incentive for Nvidia to change its behavior. That's really all there is to it.
True but this will likely change when the ai datacenters get saturated with supply and demand starts to plautau. Guess who they'll beg for their business and the performance will magically reaper again? This week we learned that Nvidia stock is very vulnerable and volatile.
 
I applaud Tim on this RTX 50 Series breakdown.
It's nice to see nGreedia getting it's dues.


Quick note and not sure why not mentioned right off the bat with nVidia's newest Blackwell architecture, is that the $1999 RTX 5090:

^Does not use the full GB202-300-A1 GPU core. (it kinda implied, but not mentioned or even graphed)

The RTX 5090 will have only have 170 SMs enabled out of the total of 192 SMs and will feature only 21,760 cores instead of the total of 24,576 cores. 11.4% redux, a tad more than the -11.1% redux of 4090 versus its full Ade Lovelace AD102 die.
 
This is exactly why I wouldn't pay for a 4080, but happily went for the 4070 Ti Super. It's flatly a superior card for the money and the lower power requirements will give it a long life. Sure they can keep fighting physics and making bigger cards that burn through more power, but that's barely progress. That balance of efficiency is everything, and they're losing it.
I had a NumberNine video card..!

Imagine if.. you would have bought a XTX instead of the 4070ti Super though...
 
All they did was to create the worlds most powerful GPUs, how dare they not do more! Lol just buy from the also-ran micro devices company that's good enough for you.
 
RTX5080=1500$ minimum entry level: pny, gainward
Now 2025 February

so need correctly
 
Last edited:
Ultimately what most of us want is actually an RTX 4060 12/16 GB VRAM at existing 4060 prices.
that would be far more exciting ...
 
So, the evolution of the cores over 6 years of the 80 range is on average 72%.
The RTX 5080 only evolves by 49% and should find its place in the 70 range.
Despite the manufacturing cost that has been multiplied by 2.5 over the past 10 years, added to inflation, Nvidia is considerably inflating its margins to ensure the fruit of its future research by selling its 5080 about $220 more than inflation.
In the end, this is double the normal price for the consumer... because it has no competitor on its direct products (RT, DLSS4 ....).
For the price of the 5080 on sale, Nvidia should have developed a better configuration of its card (more bandwidth, more vram and more cores), it is actually a 5070.
I love Techspot's conclusion: "Does NVIDIA want to sell an RTX 5080 at a price more in line with the 70-class hardware? Absolutely not. Could they do it if they really wanted to? yes, and probably quite comfortably. »

That's all said, there's enough to turn to AMD graphics cards now...
But I may wait to have a 5080 Super Ti at a more than reasonable price in ... 2 years, just to limit the scam.

 
I was 20 years PC gamer. At current price tags on gaming pc hardware and Windows invasiveness combined with nVidia latest and greatest 50 series I finally voted with my wallet. PlayStation.

I don’t have the luxury to upgrade my hardware every three years. I don’t have the time for Microsoft gimmicks, bugs, updates. I don’t want any AI telling me what I didn’t ask for and don’t care about.

PlayStation generational upgrade cycle is around 10 years. I turn it on. I play the game.

That was all I needed from PC
 
NVIDIA has always priced its GPUs as high as the market would allow. Even when AMD offered competitive alternatives at lower prices, consumers still overwhelmingly chose NVIDIA. The brand dominance, superior software ecosystem, and stronger marketing made NVIDIA the default choice for most buyers.

AMD, historically, has struggled to truly surpass NVIDIA or Intel. While they’ve made strides with their Ryzen CPUs, the generational uplift from the 7000 series to the 9000 series isn’t groundbreaking. The performance improvements are incremental at best, and history suggests that AMD’s current lead in CPUs won’t last long.

In the GPU space, AMD has failed to provide meaningful competition and has basically given up. Without true competition, NVIDIA has no reason to lower prices. And coupled with "I want it" or "Just buy it" mentality these are the real issue, competition does drive prices down, but right now, there simply isn’t any, and as long as everyone keeps opening their wallets to Nvidia, this will not change.

But I digress, it has been shown that Nvidia doesn't really care about the consumer GPU market anymore. They give crumbs and and out come the credit cards.

 
And finally we have a 70 series class at $999, good job Nvidia you have outdone yourself this time. Not even if it had the ratio of the 3070 vs 2080 Ti I wouldn't consider it. This means equal or better than the 4090.
But when looking at the current prices and the new tarrifs coming in soon, this can be the $2500-3000 card real fast.
PC gaming is dead and beside stupid games last years Nvidia is putting the cherry on top.
For sure. I want next gen performance around 1k, if they cant deliver im out. They CANT this generation. 5080 cant even pass the last gen card. Ive never seen that. Usually cards are even or a bit faster. This is just crazy. I might agree to pay 1k for a 70 card... but this is even worse than a 70 card!!! So RIP PC gaming and these stupid prices for meh cards. It really seems like the end.
 
Nvidia will keep doing this as long as people keep buying them. Unfortunately, AMD hasn't been able to offer much competition at the high end to provide consumers with another option. Maybe Intel will actually stick around in the market long enough where their performance increases can become a factor.
 
I disagree with the base premise. The 5080 is the exact same size as 4080 Super, on the same process node.

Just because NV had to push the reticle limit to produce a flagship that was distinct from it's previous chip doesn't mean the performance class has shifted.

I'd look at it from the viewpoint that if the 5090 was constrained to the same size as the 4090, it would also probably be showing only a 5-10% uplift. But because it's 20% larger physically, that's the space for the extra performance bump, beyond minor architecture updates.

They can't well launch a new line of GPUs that has almost no uplift, top to bottom.
 
AMD, historically, has struggled to truly surpass NVIDIA or Intel. While they’ve made strides with their Ryzen CPUs, the generational uplift from the 7000 series to the 9000 series isn’t groundbreaking. The performance improvements are incremental at best, and history suggests that AMD’s current lead in CPUs won’t last long.
That's total BS. AMD finally resolved issues with AVX512 power consumption. Something Intel solved by, well, disabling AVX512. Intel can start catching AMD by enabling AVX512 again.

AVX512 has always offered huge performance improvements, however previous implementations had serious limitations and/or power consumption was serious issue. AMD solved both problems so saying around 40% difference between 7000 and 9000 series isn't groundbreaking is just BS. Oh yeah, enabling AVX512 take few seconds from developers. So that's not an issue either.
 
Back