Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

theverge.com

Post reply on HN
Print view View original

401–410 of 776 posts

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#401

Post by memhole »

terhechte wrote: post #179
blixt wrote: post #156

Pretty interesting watching their tech explainers on YouTube about the changes in their AI solutions. Apparently they switched from CNNs to transformers for upscaling (with ray tracing support) if I understood correctly though for frame generation makes even more sense to me. 32 GB VRAM on the highest end GPU seems almost small after running LLMs with 128 GB RAM on the M3 Max, but the speed will most likely more than…

If you have 128gb ram, try running MoE models, they're a far better fit for Apple's hardware because they trade memory for inference performance. using something like Wizard2 8x22b requires a huge amount of memory to host the 176b model, but only one 22b slice has to be active at a time so you get the token speed of a 22b model.

I haven’t had great luck with the wizard as a counter point. The token generation is unbearably slow. I might have been using too large of a context window, though. It’s an interesting model for sure. I remember the output being decent. I think it’s already surpassed by other models like Qwen.

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#402

Post by williamDafoe »

bowsamic wrote: post #162
wruza wrote: post #151

Earlier quoted context omitted.

Why do you need one of those as a gamer? 1080ti was 120+ fps in heavy realistic looking games. 20xx RT slashed that back to 15 fps, but is RT really necessary to play games? Who cares about real-world reflections? And reviews showed that RT+DLSS introduced so many artefacts sometimes that the realism argument seemed absurd. Any modern card under $1000 is more than enough for graphics in virtually all games. The gamin…

> is RT really necessary to play games? Who cares about real-world reflections? I barely play video games but I definitely do

1. Because you shoot at puddles? 2. Because you play at night after a rainstorm?

Really, these are the only 2 situations where ray tracing makes much of a difference. We already have simulated shadowing in many games and it works pretty well, actually.

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#403

Post by williamDafoe »

PaulKeeble wrote: post #11

Looks like most of the improvement is only going to come when DLSS4 is in use and its generating most of the frame for Ray Tracing and then also generating 3 predicted frames. When you use all that AI hardware then its maybe 2x, but I do wonder how much fundamental rasterisation + shaders performance gain there is in this generation in practice on the majority of actual games.

Given that Jensen completely omitted ANY MENTION of rasterization performance, I think we can safely assume it's probably WORSE in the 5000 series than the 4000 series, given the large price cuts applied to every card below then 5090 (NVidia was never happy charging $1000 for the 4080 super - AMD forced them to do it with the 7900xtx).

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#404

Post by Reason077 »

dabinat wrote: post #380
jbarrow wrote: post #288

The increasing TDP trend is going crazy for the top-tier consumer cards: 3090 - 350W 3090 Ti - 450W 4090 - 450W 5090 - 575W 3x3090 (1050W) is less than 2x5090 (1150W), plus you get 72GB of VRAM instead of 64GB, if you can find a motherboard that supports 3 massive cards or good enough risers (apparently near impossible?).

This is the #1 reason why I haven’t upgraded my 2080 Ti. Using my laser printer while my computer is on (even if it’s idle) already makes my UPS freak out. But NVIDIA is claiming that the 5070 is equivalent to the 4090, so maybe they’re expecting you to wait a generation and get the lower card if you care about TDP? Although I suspect that equivalence only applies to gaming; probably for ML you’d still need the highe…

Does a laser printer need to be connected to a UPS?

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#405

Post by mbreese »

MisoRamen wrote: post #326
themaninthedark wrote: post #236

Earlier quoted context omitted.

It seems that they have been tightening what they allow their partners to do, which caused EVGA to break away as they were not allowed to deviate too much from the reference design.

It is an ever uphill battle to compete with Nvidia as a AIB partner. Nvidia has internal access to the new card way ahead of time, has aerodynamic and thermodynamic simulators, custom engineered boards full of sensors, plus a team of very talented and well paid engineers for months in order to optimize cooler design. Meanwhile AIB partners is pretty much kept in the blind until a few months in advance. It is basicall…

Which is why EVGA stopped working with Nvidia a few years ago... (probably mentioned elsewhere too).

https://www.electronicdesign.com/technologies/embedded/artic...

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#406

Post by memhole »

logankeenan wrote: post #264
terhechte wrote: post #179

Earlier quoted context omitted.

If you have 128gb ram, try running MoE models, they're a far better fit for Apple's hardware because they trade memory for inference performance. using something like Wizard2 8x22b requires a huge amount of memory to host the 176b model, but only one 22b slice has to be active at a time so you get the token speed of a 22b model.

Do you have any recommendations on models to try?

I planted garlic this year. Thanks for documenting! I can’t wait to see what I get harvest time.

I like the Llama models personally. Meta aside. Qwen is fairly popular too. There’s a number of flavors you can try out. Ollama is a good starting point to try things quickly. You’re def going to have to tolerate things crashing or not working imo before you understand what your hardware can handle.

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#407

Post by Yizahi »

intellix wrote: post #351
chmod775 wrote: post #281

> will be two times faster [...] thanks to DLSS 4 Translation: No significant actual upgrade. Sounds like we're continuing the trend of newer generations being beaten on fps/$ by the previous generations while hardly pushing the envelope at the top end. A 3090 is $1000 right now.

Why is that a problem though? Newer and more GPU intensive games get to benefit from DLSS 4 and older games already run fine. What games without DLSS support could have done with a boost? I've heard this twice today so curious why it's being mentioned so often.

I also like DLSS, but the OP is correct that it is a problem. Specifically it's a problem with understanding what are these cards capable of. Theoretically we would like to see separately performance with no upscaling at all, then separately with different levels of upscaling. Then we would be able to see easier what is the real performance boost of the hardware, and of the upscaler separately.

It's like BMW comparing new M5 model to the previous gen M5 model, while previous gen is on the regular 95 octane, and new gen is on some nitromethane boosted custom fuel. With no information how fast the new car is on a regular fuel.

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#408

Post by iwontberude »

Reason077 wrote: post #404
dabinat wrote: post #380

Earlier quoted context omitted.

This is the #1 reason why I haven’t upgraded my 2080 Ti. Using my laser printer while my computer is on (even if it’s idle) already makes my UPS freak out. But NVIDIA is claiming that the 5070 is equivalent to the 4090, so maybe they’re expecting you to wait a generation and get the lower card if you care about TDP? Although I suspect that equivalence only applies to gaming; probably for ML you’d still need the highe…

Does a laser printer need to be connected to a UPS?

It's not connected to the UPS directly, it's causing voltage dip on the circuit tripping the UPS.

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#409

Post by mikepurvis »

epolanski wrote: post #353
marricks wrote: post #347

Earlier quoted context omitted.

I got into desktop gaming at the 970 and the common wisdom (to me at least, maybe I was silly) was I could get away with a lower wattage power supply and use it in future generations cause everything would keep getting more efficient. Hah...

Yeah, do like me, I lower settings from "ultra hardcore" to "high" and keep living fine on a 3060 at 1440p for another few gens. I'm not buying GPUs that expensive nor energy consuming, no chance. In any case I think Maxwell/Pascal efficiency won't be seen anymore, with those RT cores you get more energy draw, can't get around that.

I feel similarly; I just picked up a second hand 6600 XT (similar performance to 3060) and I feel like it would be a while before I'd be tempted to upgrade, and certainly not for $500+, much less thousands.

Re: Nvidia announces next-gen RTX 5090 and RTX 5080 GPUs

#410

Post by ryao »

jbarrow wrote: post #288

The increasing TDP trend is going crazy for the top-tier consumer cards: 3090 - 350W 3090 Ti - 450W 4090 - 450W 5090 - 575W 3x3090 (1050W) is less than 2x5090 (1150W), plus you get 72GB of VRAM instead of 64GB, if you can find a motherboard that supports 3 massive cards or good enough risers (apparently near impossible?).

I wonder how many generations it will take until Nvidia launches a graphics card that needs 1kW.
Post reply on HN