Hacker News new | past | comments | ask | show | jobs | submit login

[flagged]



The 4090s perf/$ is much better than the 4080, so if you are using the card for anything except gaming it makes a lot of sense. For gaming, yeah, you'd need triple 4k screens on a latest AAA game to actually need than kind of performance.


If you need CUDA don't care about AMD.


>If you need CUDA

The 4090 is a gaming card, and thus optimized for games, not compute.

Buying a card that costs that much and draws that much power to do CUDA is not a choice somebody would make within the realm of sanity.

>don't care about AMD.

AMD has HiP, which is supposedly an open CUDA.

I hear that it's enough to replace calls to "CUDA" to calls to the same function names except "HIP" and then you're set, and has swappable backends, working with both AMD and NVIDIA hardware.

Note that I can't vouch for it, as I haven't tried it, but with the swappable backends I'd switch immediately even if I kept using NVIDIA hardware, just to have assurance of not being under vendor lock-in.


HiP (unfortunately) has very limited support on AMD cards, unlike CUDA which runs on ALL NVidia hardware.


That sucks, but the situation is improving. RDNA2 support was added early this year.

Now that AMD isn't constrained by financial dire straits anymore, they seem to be pushing the HiP ecosystem very hard, adding HiP support into relevant software and frameworks.

They were focused on CDNA cards first (aimed at datacenter/computer), but expanding it to cover RDNA2 (the gaming cards) is a key step forward; that's what potential developers already have for gaming, and thus important for HiP adoption.

Otherwise, the widely deployed Vega have worked with HiP for a long time. I agree the RDNA1 hiccup was ugly and hurt adoption.


While it's nice that AMD is becoming more competitive in this space, many small teams still go for NVIDIA to maximize compatibility.

The 4090 is fine as a budget option for many compute tasks, including deep learning, as long as 24GB per card is enough for your use cases.

Single cards in a PC/Workstation can be used to develop the code for a network, even if you run the main training on servers.

But as the 4090 is multiple times cheaper than an A100, a setup with 4 4090's will be preferable to A100s for some workloads.

See for instance Bizon Z5000: https://bizon-tech.com/bizon-z5000.html


This.

Even assuming AMD had support for something as mature and battle-tested as CUDA, a CUDA engineer may consider the price of these cards a no-brainer.


If you're using software that relies and CUDA, and really need the performance upgrade from the 3090, the 4090 may be a bargain.


doesn't matter what AMD is going to do if you value ray tracing, dlss, rtx voice, shadowplay, good drivers


>if you value ray tracing

AMD has Ray Tracing, too. In RDNA3, it might even perform really well. Nov 3rd is close, and we'll know by then.

>dlss

Current FSR (open, works on AMD, NVIDIA and Intel) is considered equivalent by reviewers. Differences exist, but neither is better than the other. These differences are now up to individual taste.

>rtx voice

AMD Noise Suppression.

>shadowplay

AMD ReLive. But you should be using OBS: It's better than either, and open source. That's what most streamers use, pro and else.

>good drivers

AMD are better in my experience. YMMV.


I've never seen a single reviewer say FSR is equal/better than DLSS especially now with 3.0. I would like to see a source for that. I do like that it's open.

AMD's ray tracing isn't good yet and is years behind NVIDIA


Even if you use OBS nvenc is way better than AMD's offering. You'd just use nvenc through OBS, unless you want to dedicate half of your CPU to recording. Even if you do that you'll still take a performance impact in anything else you're doing.


AMD's hardware encoder can of course be used with OBS, and it is actually good as of OBS 28.


The problem isn't with OBS. The problem is with AMD's encoder. Nvenc is much better.


AMD's drivers are so consistently awful that I no longer bother looking at their cards. And if you keep waiting for the next announcement in the pipeline then you'll never buy anything.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: