NVIDIA RTX 4080 Rumored To Feature 420 W TDP

[ZoneDymo]

That's an assumption if ever there was one...
Where I live, all domestic electricity is hydro generated.

Im sorry but this requires a reality check.

1. We are talking about GLOBAL warming, if the 400+ watt parts were only sold in your area then sure* but they are sold world wide.
2. I would love to know where you even live because I kinda doubt it is all hydro generated and if that limit is not already reached, not resisting against this ever growing power demand will mean that at some point it will be overwhelmed.

3. And this is perhaps the more painful part, The Grand Tour, Season 4, episode 1.
They went to Cambodia to go on a river ride trip, only to find the rivers bare and the houses on stilts looking mighty silly with no water running underneath them
Where did the water go? has it evaporated as a result of global warming? well no.
China is where that same river goes through first and they made....you guessed it, a nice big fat hydrodam, so they can generated all the power for themselves and absolutely destroy all life following after.
The people in Cambodia are completely dependent on that river and James May mentioned that damming china is doing, the results of it, are pretty much on par with Nuclear War....

So yeah, aint all that easy or free from consequences.

Somebody from gov should regulate this nonsense. This is getting ridiculous. A GPU that consumes more then a standard PSU from a couple of years ago..

I fully agree with this, I think we could see lovely developments coming from restraints.
Force the gpu's to not consume more then 300 watts (or perhaps evne less), save for absolute (low quantity) halo products that are meant for OC records etc perhaps.

 

[Deleted member 24505]

Hands up who will buy this..........................Liars you know you will

 

[ARF]

Hands up who will buy this..........................Liars you know you will

I am looking forward to the Radeon RX 7000 series

 

[BSim500]

Unfortunately, there is no way around it. Dennard scaling hit a wall while Moore is still going. Transistors/die can still increase significantly but power/transistor isn't going down at the same rate. If you want more performance, you have to give it more power.

Whilst true, I'm far more interested in seeing better optimised games than blindly trying to brute-force it with unwanted space-heaters. The other half the of "My GPU isn't powerful enough for this game" equation is "this game dev can't optimise for sh*t"...

Edit: Looking back on the past 35 years of PC gaming, I remember being blown away in the late 90's to mid 2000's when games like Thief, Half Life and FEAR came out. Not just the graphics, but the whole package, eg, The Dark Engine's audio propagation physics on an Aureal 3D sound card, enemies communicating with each other in FEAR in an actually believable way, etc, and thinking damn what are games going to be like in 25 years time. 25 years on, and advances in these areas are flat as hell. Everything is solely about graphics and even that's starting to experience the same depreciating excitement issue as Hollywood CGI, ie, once you've seen the 300th similar looking photo-realistic clone, it's no longer that 'interesting' anymore and you start craving stuff that actually looks and feels different, eg, Dishonored's water-colour or Bioshock's underwater art-deco styles. I'd take more of these type of games that run exceptionally well on an average GPU anyday than the 'need' for a 1500w RTX 6090Ti Super Titan for Crysis 6's 7hr long 'content'...

 

[bug]

Whilst true, I'm far more interested in seeing better optimised games than blindly trying to brute-force it with unwanted space-heaters. The other half the of "My GPU isn't powerful enough for this game" equation is "this game dev can't optimise for sh*t"...

Current cards run two engines: raster and ray-tracing. And it will be another decade easily before we can move completely past rastering.

 

[BSim500]

Current cards run two engines: raster and ray-tracing. And it will be another decade easily before we can move completely past rastering.

I don't think we ever will "completely move past it". For every AAA FPS that can be ray-traced, there are like 200x Indie's often of a style or genre where real-time ray-tracing hardly adds anything at all to the game.

 

[chrcoluk]

Somebody from gov should regulate this nonsense. This is getting ridiculous. A GPU that consumes more then a standard PSU from a couple of years ago..

Could happen in Europe over next few years if energy crisis doesnt ease up, is already talks of energy rationing in UK. People buying generators in anticipation of power cuts.

 

[Guwapo77]

Somebody from gov should regulate this nonsense. This is getting ridiculous. A GPU that consumes more then a standard PSU from a couple of years ago..

There is no need for government intervention when the consumer can refuse to purchase it. There is/are other option(s)...AMD/Intel.

 

[chrcoluk]

I don't think we ever will "completely move past it". For every AAA FPS that can be ray-traced, there are like 200x Indie's often of a style or genre where real-time ray-tracing hardly adds anything at all to the game.

Pretty much this, sometimes people forget, the especially picked AAA games typically used by reviewers for graphics tech showcasing are something like 0.01% of the market.

 

[ARF]

And not only this, but also process nodes and transistor count. You have to throw many, many transistors, while in the next decade there will be two, maximum three process node shrinks. So, no, it won't happen, if ever...

 

[Guwapo77]

Err

RTX Titan - 900 W
RTX 4090 Ti - 700 W
RTX 4090 - 600 W
RTX 4080 - 420 W
RTX 4070 - 300 W
RTX 4060 - 250 W
RTX 4050 - 170 W
GeForce RTX 4000 “Lovelace“ power draw might exceed 800 W - HWCooling.net
Nvidia GeForce RTX 4090 GPU may demand 600W from your PSU | PCGamesN

Not nice

I thought those wattages were debunked. But if they are true, Nvidia certainly taking the wrong path moving forward. We might end up with Nvidia Certified PSUs again...

 

[looniam]

yep nvidia - where every fermi meme is a challenge and smoldering gpus will be a feature, adding real life ambient occlusion.

 

[bug]

I don't think we ever will "completely move past it". For every AAA FPS that can be ray-traced, there are like 200x Indie's often of a style or genre where real-time ray-tracing hardly adds anything at all to the game.

There will be a point where ray tracing will be easy enough to implement, it won't matter. E.g. a lot of games don't need Pixel Shader 3.0, but since it's there and easy to use, why not?

 

[randomName]

I do also agree that this power consumption needs to be regulated, because it is getting out of hands really. Game makers can easily add gazillion of extra polygons, so GPUs have a hard time processing and so GPU makers will make even bigger chips and even more power hungry ones, then games will be made even more demanding and this will never end. Also add crypto currency to that.

High power CPU and GPU should be licensed and used by those who really need it, not by simple end-users who want to play games.

 

[ZoneDymo]

Hands up who will buy this..........................Liars you know you will

Im not supporting Nvidia, if only because too many are already doing so.

 

[MarsM4N]

NVIDIA Ada Lovelace GPUs To Have A Node Advantage Over AMD RDNA 3, Rumored To Utilize TSMC 4N Process

NVIDIA Ada Lovelace GPUs for next-gen GeForce RTX 40 Gaming graphics cards will have a node advantage over AMD RDNA 3 in the form of TSMC 4N.

wccftech.com

​

"The reasons why NVIDIA may have selected TSMC's 4N as the candidate for its next-gen gaming GPU lineup are kind of obvious. The upcoming cards will be real power-hungry and NVIDIA & the company is going to optimize them as much as they can by utilizing the 4N process node. AMD on the other hand will be utilizing a mix of TSMC 5nm and 6nm process nodes for its upcoming MCM and monolithic GPUs based on the RDNA 3 graphics architecture and while they don't bring the optimizations that 4N does, they will feature an MCM approach that is expected to be highly efficient. So at the end of the day, NVIDIA gets the better node while AMD delivers a better design approach."

Well, let's wait & see who has the more efficient package this round. Power efficency will be a big selling point for many.

That's an assumption if ever there was one...
Where I live, all domestic electricity is hydro generated.

Can you even see your monitor with the bush fires raging down there?

 

[BSim500]

There will be a point where ray tracing will be easy enough to implement, it won't matter. E.g. a lot of games don't need Pixel Shader 3.0, but since it's there and easy to use, why not?

It won't be "easy enough" anywhere near that justifies a drastic loss in frame-rate for games that won't benefit if we hit an obvious wall for lack of efficiency increase (and we have, GTX 960 -> GTX 1060 = +70-90% FPS increase in one year at same wattage and only +20% price. GTX 1660Ti -> RTX 3050 = barely +5-10% FPS increase after 3 years for actually same price-tier, even using 20% lower than reality MSRP for the latter). Everything else only got faster by pumping up the wattage (GTX1660Ti -> RTX3060 = 50% faster purely because 180w vs 120w). Perf-per-watt improvements have already ground to a halt on the most popular price-tier of GPU's, and yet most of us still don't want space-heaters. Nor is creating games the most inefficient way possible going to be used by default for the bulk of games that won't benefit and games devs are not that stupid they'd voluntarily kill off 70% of their sales for the bulk of the sub $500 GPU market.

Here's a healthy reality check of what most normal people are actually playing, ie, ray-traced Cyberpunk 2077 is barely more popular than "Russian Fishing 4" and on the brink of falling out of the chart. In the middle of the chart, Skyrim, Age of Empires 2, Don't Starve, Stardew Valley, etc have been there for years. No one cares about ray-tracing for those. Now look at the top of the chart. All the competitive CS:GO, Fortnite, etc, players are going to do is turn RT off for better FPS. It really is a literal truth to say less than 5% of the gaming market is interested in RT (and not all of those are interested in it "at any cost"). Perceptions of "everyone will want this" future popularity of the newest gimmick has often been heavily skewed on hardware enthusiast tech forums. Ray-tracing is like VR, 3D glasses, etc, some will want it and it'll be a premium feature for those who do, but many won't and just like VR it certainly isn't going to become a new game development "minimum requirement baseline" though for the bulk of the market and not just due to GPU cost.

 

[ARF]

NVIDIA Ada Lovelace GPUs To Have A Node Advantage Over AMD RDNA 3, Rumored To Utilize TSMC 4N Process

NVIDIA Ada Lovelace GPUs for next-gen GeForce RTX 40 Gaming graphics cards will have a node advantage over AMD RDNA 3 in the form of TSMC 4N.

wccftech.com

​

"The reasons why NVIDIA may have selected TSMC's 4N as the candidate for its next-gen gaming GPU lineup are kind of obvious. The upcoming cards will be real power-hungry and NVIDIA & the company is going to optimize them as much as they can by utilizing the 4N process node. AMD on the other hand will be utilizing a mix of TSMC 5nm and 6nm process nodes for its upcoming MCM and monolithic GPUs based on the RDNA 3 graphics architecture and while they don't bring the optimizations that 4N does, they will feature an MCM approach that is expected to be highly efficient. So at the end of the day, NVIDIA gets the better node while AMD delivers a better design approach."

Well, let's wait & see who has the more efficient package this round. Power efficency will be a big selling point for many.

TSMC N4 is a rebranded N5 process, in the best case +5%. It is N5+.
If the nvidia micro-architecture is much worse, then those 5% won't matter at all.

 

[Bomby569]

i heard a rumour you should not pay attention to rumours, even more as this is just another rumour, and this one contradicts the previous rumour (that the numbers were much higher)

idiots after their 15 minutes of fame (and don't came after me, because at least one of them is, and maybe both)

 

[bug]

It won't be "easy enough" anywhere near that justifies a drastic loss in frame-rate for games that won't benefit if we hit an obvious wall for lack of efficiency increase (and we have, GTX 960 -> GTX 1060 = +70-90% FPS increase in one year at same wattage and only +20% price. GTX 1660Ti -> RTX 3050 = barely +5-10% FPS increase after 3 years for actually same price-tier, even using 20% lower than reality MSRP for the latter). Everything else only got faster by pumping up the wattage (GTX1660Ti -> RTX3060 = 50% faster purely because 180w vs 120w). Perf-per-watt improvements have already ground to a halt on the most popular price-tier of GPU's, and yet most of us still don't want space-heaters. Nor is creating games the most inefficient way possible going to be used by default for the bulk of games that won't benefit and games devs are not that stupid they'd voluntarily kill off 70% of their sales for the bulk of the sub $500 GPU market.

Here's a healthy reality check of what most normal people are actually playing, ie, ray-traced Cyberpunk 2077 is barely more popular than "Russian Fishing 4" and on the brink of falling out of the chart. In the middle of the chart, Skyrim, Age of Empires 2, Don't Starve, Stardew Valley, etc have been there for years. No one cares about ray-tracing for those. Now look at the top of the chart. All the competitive CS:GO, Fortnite, etc, players are going to do is turn RT off for better FPS. It really is a literal truth to say less than 5% of the gaming market is interested in RT (and not all of those are interested in it "at any cost"). Perceptions of "everyone will want this" future popularity of the newest gimmick has often been heavily skewed on hardware enthusiast tech forums. Ray-tracing is like VR, 3D glasses, etc, some will want it and it'll be a premium feature for those who do, but many won't and just like VR it certainly isn't going to become a new game development "minimum requirement baseline" though for the bulk of the market and not just due to GPU cost.

You're assuming AMD and Nvidia will want to drag two engines with them forever. They won't. They'll find a breaking point and go full RT. But again, 10 years is a (very) optimistic estimation for when that will happen.

 

[big_glasses]

Here's a healthy reality check of what most normal people are actually playing, ie, ray-traced Cyberpunk 2077 is barely more popular than "Russian Fishing 4" and on the brink of falling out of the chart. In the middle of the chart, Skyrim, Age of Empires 2, Don't Starve, Stardew Valley, etc have been there for years. No one cares about ray-tracing for those. Now look at the top of the chart. All the competitive CS:GO, Fortnite, etc, players are going to do is turn RT off for better FPS. It really is a literal truth to say less than 5% of the gaming market is interested in RT (and not all of those are interested in it "at any cost"). Perceptions of "everyone will want this" future popularity of the newest gimmick has often been heavily skewed on hardware enthusiast tech forums. Ray-tracing is like VR, 3D glasses, etc, some will want it and it'll be a premium feature for those who do, but many won't and just like VR it certainly isn't going to become a new game development "minimum requirement baseline" though for the bulk of the market and not just due to GPU cost.

While I agree with you on RT, you are pulling a SP game with a (imo) limited replayability vs MP games (and Skyrim with it's massive mod base), at peak it had 800k+ players.
And it's also for the times that people play the SP games they want the gorgeous graphics, and I'd argue that most people don't want to turn down too much of details (unless they are the top percenter)

and a couple of those still requires a bit of power to output the high FPS

 

[ARF]

i heard a rumour you should not pay attention to rumours, even more as this is just another rumour, and this one contradicts the previous rumour (that the numbers were much higher)

idiots after their 15 minutes of fame (and don't came after me, because at least one of them is, and maybe both)

A single 16-pin power connector is also present. It is too early to speculate on RTX 40 series power draw, but it is quite clear we will be seeing a 600W card in this generation such as RTX 4090 Ti or a new TITAN, it may simply not launch in the coming months but later. Such TDP is still within the capability of a single 16-pin connector, which as far as we know, should now debut on all RTX 40 cards.

Alleged GeForce RTX 4090 series PCB leak shows NVLink and 16-pin power connectors but no USB Type-C display output - VideoCardz.com

 

[Bomby569]

Alleged GeForce RTX 4090 series PCB leak shows NVLink and 16-pin power connectors but no USB Type-C display output - VideoCardz.com

"It is too early to speculate on RTX 40 series power draw, but it is quite clear we will be seeing a 600W card in this generation"

it's to early but let me speculate anyway

 

[ARF]

"It is too early to speculate on RTX 40 series power draw, but it is quite clear we will be seeing a 600W card in this generation"

it's to early but let me speculate anyway

I think what they mean is where exactly it will fall between 600 and 1000 watts TGP.

 

[InVasMani]

NVIDIA Ada Lovelace GPUs To Have A Node Advantage Over AMD RDNA 3, Rumored To Utilize TSMC 4N Process

NVIDIA Ada Lovelace GPUs for next-gen GeForce RTX 40 Gaming graphics cards will have a node advantage over AMD RDNA 3 in the form of TSMC 4N.

wccftech.com

​

"The reasons why NVIDIA may have selected TSMC's 4N as the candidate for its next-gen gaming GPU lineup are kind of obvious. The upcoming cards will be real power-hungry and NVIDIA & the company is going to optimize them as much as they can by utilizing the 4N process node. AMD on the other hand will be utilizing a mix of TSMC 5nm and 6nm process nodes for its upcoming MCM and monolithic GPUs based on the RDNA 3 graphics architecture and while they don't bring the optimizations that 4N does, they will feature an MCM approach that is expected to be highly efficient. So at the end of the day, NVIDIA gets the better node while AMD delivers a better design approach."

Well, let's wait & see who has the more efficient package this round. Power efficency will be a big selling point for many.



Can you even see your monitor with the bush fires raging down there?

IF RDNA 3 ends up being more efficient it's a huge success for AMD and huge failure for Nvidia given the context of the situation with the node difference. We'll have to see how big a impact MCM provide AMD, but we've seen with CPU's it's worked really nicely.