-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
NVIDIA 5000 SERIES
- Thread starter scubes
- Start date
HACO. The guy that would always predict much higher prices every new generation as I recall.
You made your dreams a reality and now you live them
Got 2000 and 4000 pretty much spot on, and 3000 very wrong
and my prediction for this year is £2000 and £1200.Got 2000 and 4000 pretty much spot on, and 3000 very wrong
Yeah. That sounds likely unfortunately.
Associate
I honestly find it sad that in 2024 and with the launches in 2025 that my 1080ti still has more Memory than GPUs that are being released in the present day.
Yeah, it's not good. What depresses me is that if you want to buy a mid-range card for sensible-ish money (~£300) your only option in what will soon be 2025 is to go for an RTX 4060
Which basically means buying a card that's little more spec and performance-wise than a rebadged RTX 3060 from over 4 years ago!Truly grim ....
Last edited:
I think the B580 is now the go to option in that price range assuming you can find one in stock.Yeah, it's not good. What depresses me is that if you want to buy a mid-range card for sensible-ish money (~£300) your only option in what will soon be 2025 is to go for an RTX 4060
Truly grim ....
Associate
- Joined
- 15 Aug 2012
- Posts
- 265
I find the idea of an 8GB 5060 and 12GB 5070 pretty appalling for what they’re likely to be charging for these cards.
Associate
Mrk trying to justify whatever price he pays for his inevitable 5090 purchase
Last edited:
Associate
i love the idea of midrange buyers subsidizing enthusiast grade GPUs
Jensen?People acting as if GDDR7 comes cheap, and 32GB of it too, and new build process, and the massive die size etc etc
Surely when you typed this up, you realized literally any counter argument or even... piece of hardware from the last 30 years would dispense massive dents into this nonsense. Like there's never been a node change before? Is GDDR7 the first ram etched into diamond or something?
As for the rest of you guys, the 5080 will probably be about the 4090, and they aren't going to give 4090 performance away for 4080 prices. Every launch we've foolishly tried applying 2017 logic to this stuff, and yet it's very clear that nvidia is just running with it now. With their GPU mindshare, they can do whatever they want and they are going to continue doing that. You want to believe that a generational leap will account for a price reduction. But nope, that's not how the GPU market works now. They just add the generational improvements to the last gens price, instead of replacing it. 4080 is double 3080 performance, so you pay double. 5080 will add 20% to 4080, so you can just add 20%. There's no improvements. You can't even just buy 20% improvement. They really got us. Have to buy the whole new die for 2 grand that cost nvidia like what, 100 bucks lol?
My buddy at work who knows a lot more about PC's than I do, in the sense of coding and lanes and yada yada, completely buys the userbenchmark slant. It's the first link that comes up when people go to compare. So he believes their little blurbs, that AMD is horrifically overpriced and has an army of propaganda bots and all that jazz. That's what normal people think of the GPU landscape. Nevermind the XTX is half the price and 80% the performance. It doesn't matter what AMD puts out now. They can't break gen z's/young millennials mindlock, so Nvidia is free to do whatever they want.
Associate
the facts.. unfortunately right now jensen can sell the same die for $15k a pop
my concern is that somewhere around $8000 the market for enthusiast gpus will implode meaning nvidia will just not see enough demand to keep this segment alive.
it all boils down to how intel/samsung can keep up with tsmc, GPUs need to be produced in a different fab cant have them competing for capacity with AI chips
my concern is that somewhere around $8000 the market for enthusiast gpus will implode meaning nvidia will just not see enough demand to keep this segment alive.
it all boils down to how intel/samsung can keep up with tsmc, GPUs need to be produced in a different fab cant have them competing for capacity with AI chips
Soldato
I think it's worth remembering a few points;
If you *need* a gpu it dosnt have to be the latest and greatest to have a good time.
Because of the lack of progress (without a price increase) GPUs bought now will last a long time and hold most of their value.
While margins are incredibly high, the costs to make these things has gone through the roof. 32gb of DDR7 is $400 dollars alone. Wages, transportation, inflation all have a culmative effect.
If you *need* a gpu it dosnt have to be the latest and greatest to have a good time.
Because of the lack of progress (without a price increase) GPUs bought now will last a long time and hold most of their value.
While margins are incredibly high, the costs to make these things has gone through the roof. 32gb of DDR7 is $400 dollars alone. Wages, transportation, inflation all have a culmative effect.
Associate
blasphemer
i cant see it dropping buy that much mate .At launch maybe. But give it a few months and you would jump in as it eats at you day by day
Man of Honour
As before, it won't get 4090 performance, it doesn't have enough cores or memory bus or VRAM - It might with Neural rendering whatever that is, but as also mentioned, you will then have a new category of naysayers that sitr alongside frame gen haters and upscaling haters. That's not quite the same thing and it's not going to negate the need for VRAM at 4K output anyway as enough games now use over 11GB VRAM fairly easily, with more VRAM needed if you enable frame gen and maybe even this neural rendering assuming it's a global feature instead of games needing to be updated to support it like current upscaling and frame gen tech.
Otherwise please explain how you get more from less, this isn't the big bang
Associate
5080 definitely won't have the performance in all situs as the 4090 does, AI or neural rendering aside (by "neural", they likely mean better algos that evaluate a frame to identify images to replicate -- it's not magic). But it will be in the realm, most likely at lower res/IQ, and nV (and their army of "reviewers") will push this point to the utmost. 5080 is designed with a few things in mind, foremost being to allow nV to state they are providing "4090 tier" performance for ~£1200.As before, it won't get 4090 performance, it doesn't have enough cores or memory bus or VRAM - It might with Neural rendering whatever that is, but as also mentioned, you will then have a new category of naysayers that sitr alongside frame gen haters and upscaling haters. That's not quite the same thing and it's not going to negate the need for VRAM at 4K output anyway as enough games now use over 11GB VRAM fairly easily, with more VRAM needed if you enable frame gen and maybe even this neural rendering assuming it's a global feature instead of games needing to be updated to support it like current upscaling and frame gen tech.
Otherwise please explain how you get more from less, this isn't the big bang
Soldato
"We're giving you /almost/ the performance now for £1249 that you could have had for over 2 years at £1600"
I mean.... It's 'progress' for sure, but it's less than is generationally expected and certainly not something nVidia should/would be wanting to shout about.
Soldato
Given the rate of tech advancement to me it seems pretty good (minus less vram).