So basically all along you agreed with what I was saying then. Too funnyHmm I think I'd still bite..![]()

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.
So basically all along you agreed with what I was saying then. Too funnyHmm I think I'd still bite..![]()
What if it has double the RT of 3090, 10% extra rasta, but only 10gb vram?![]()
That’s your specialityHammer it
Some decent efficiency gains should be expected when moving from the inefficient Samsung 8nm which is barely better than TSMCs 12nm down to TSMC 5nm if that's what nvidia goes with.Ah yeah just checked, "stock" 3080 tops out at 318w and a "stock" 3090 at 346w or there abouts anyway.
Still an impressive showing in efficiency improvement if a "stock" 4060 matches/beats a "stock" 3080 whilst pulling 100w less imo.
Won't be long until it's just TSMC and Intel versus the ultra cheapos stuck several gens behind.Samsung just fired its head of semi conductors, rumoured to be because the 4nm node is performing awfully
Cough cough m2 chip from apple..Won't be long until it's just TSMC and Intel versus the ultra cheapos stuck several gens behind.
That'll definitely bring price downCough cough m2 chip from apple..
Cough cough m2 manufactured by TSMC.Cough cough m2 chip from apple..
Be interesting to see how the new gen will undervolt and manage the heat output/power draw?Some decent efficiency gains should be expected when moving from the inefficient Samsung 8nm which is barely better than TSMCs 12nm down to TSMC 5nm if that's what nvidia goes with.
I guess we'll find out but the 4000s series could be a bad/lazy gen, pump em out asap while pricing is still going nuts? They can capitalise on the new cooler design and just ramp up the power. Whist saying stuff like 'we made every effort to make sure there is plenty of stock this time'![]()
Why do you think this? Performance is what matters. If they can deliver a 35% performance improvement between 3090 and 4090, it will sell out just the same as every generation. I picked the 35% number as this is how much faster the 3090 was compared to the 2080ti at 4K, relative performance across all games (Techpowerup review source).
From what we've seen so far, the performance gap will be more than 35%, though this comes with the downside of even higher TDP.
One thing to note, when people upgraded to a 3080/Ti/90 from another gen before, they likely had a PSU capable of handling the requirement. Some still upgraded PSU's as they were on the limit or had old units.
However, if your going top end assuming 40 series does require more headroom than Ampere did, this will force more PSU upgrades than before. We dont know yet what power draw these cards will take. I assume the lower sku's wont have to worry and folk with 750w units should be fine.
I have a Seasonic Ti 750W but I'd be doing my research if I was thinking of an XX80 class card.One thing to note, when people upgraded to a 3080/Ti/90 from another gen before, they likely had a PSU capable of handling the requirement. Some still upgraded PSU's as they were on the limit or had old units.
However, if your going top end assuming 40 series does require more headroom than Ampere did, this will force more PSU upgrades than before. We dont know yet what power draw these cards will take. I assume the lower sku's wont have to worry and folk with 750w units should be fine.
Performance and cost is what matters. They may want to release the product soon while the market is still nuts and people will pay whatever it costs. Remember we should be getting a 4070 for £360 MSRP - graphics card are supposed to be same price and more performance each gen right? Do you think that'll be the price, or maybe for the first 1000 cards then sold out, then the price won't be MSRP anymore.Why do you think this? Performance is what matters. If they can deliver a 35% performance improvement between 3090 and 4090, it will sell out just the same as every generation. I picked the 35% number as this is how much faster the 3090 was compared to the 2080ti at 4K, relative performance across all games (Techpowerup review source).
From what we've seen so far, the performance gap will be more than 35%, though this comes with the downside of even higher TDP.
Performance and cost is what matters. They may want to release the product soon while the market is still nuts and people will pay whatever it costs. Remember we should be getting a 4070 for £360 MSRP - graphics card are supposed to be same price and more performance each gen right? Do you think that'll be the price, or maybe for the first 1000 cards then sold out, then the price won't be MSRP anymore.
Anyone can make a GPU faster by giving it more power, that's a lazy approach. What would be impressive if they gave more performance and less power usage.
Do you have such knowledge in electrical engineering where you are confident making the claim that newer performance GPU architecture coming out with higher power draw is just "lazy design" as opposed to the limitations and requirements of the current level of technology?Anyone can make a GPU faster by giving it more power, that's a lazy approach. What would be impressive if they gave more performance and less power usage.
One thing to note, when people upgraded to a 3080/Ti/90 from another gen before, they likely had a PSU capable of handling the requirement.
Even electrical engineers would be lost on this question considering it has sweet eff all to do with GPU architectureDo you have such knowledge in electrical engineering where you are confident making the claim that newer performance GPU architecture coming out with higher power draw is just "lazy design" as opposed to the limitations and requirements of the current level of technology?