The stuff Hacker News actually recommends.

GPUs and CPUs

AMD Ryzen 5950X vs NVIDIA GeForce RTX

Hacker News owners slightly favor the NVIDIA GeForce RTX.

AMD Ryzen 5950XNVIDIA GeForce RTX
Owners with an opinion56617
Positive verdicts76% of 3776% of 416
Positive/negative since 202413/6214/71
Happy after 3+ years110
Longest happy owner4 yrs6 yrs
Praised forperformance and valueperformance and value
Criticized forreliability and noisenoise
Newest comment

Updated

What owners wrote about both

AMD Ryzen 5950X: Likes itNVIDIA GeForce RTX: Neutral
i'd say it depends (as with everything) i'm still using an old i7 3770k @ 4.8ghz with 16gb (ddr3) ram running linux for random tasks like executing tests. obviously, power consumption is higher. my main machine is a MBP M1 Max which i use for everything. i also have my main linux desktop workstation that has a 5950x with 128gb (ddr4) ram. i'll probably get 2-5 years out of my MBP, and my AMD workstation will probably be good for another 5-10 years. i'm not a gamer, but i have a 3080. i'm sure my 5950x will still be good for gaming in 10 years if paired with a modern GPU.
AMD Ryzen 5950X: NeutralNVIDIA GeForce RTX: Doesn’t like it
I have 70B at q4_0 running at 20 tokens/s inference using 2 Founders Edition RTX3090s with NVLink. It's an older AMD 5950x with a motherboard with 4-slot spacing between the PCI-e (iirc, the motherboard is an ASRock x570 Taichi). For comparison, I tried the same rig with a single RTX 4090, but it generated an abysmal 2.4 tokens/s, which I would consider nigh unusable. The difference between all GPU VRAM is significant.
AMD Ryzen 5950X: NeutralNVIDIA GeForce RTX: Likes it
Just a quick note, it's worth pointing out that for most people (eg, wanting to chat to a model in realtime), I don't think running locally on a CPU is a very viable option unless you're very patient. On my 16c Ryzen 5950X/64GB DDR4-3800 system, llama-2-70b-chat (q4_K_M) running llama.cpp (eb542d3) and testing doing a 100 token test (life's too short to try max context), I got 1.25 tokens/second (~1 word/second) output. Compiled with cuBLAS w/ `-ngl 0` (~400MB of VRAM usage, no layers loaded) makes no perf difference. The max layers I can load on a headless 24GB 4090 is 45/83 layers (running `-ngl 45 --low-vram`) which brings speeds up to 2.5 t/s. A little less painful, but still not super pleasant. For reference, people have reported performance of 12-15 t/s with 2x4090s w/ exllama (GPTQ). People are using a 14,20 split and able to load a full (NTK Rope scaled) 16K context into 48GB of VRAM.
AMD Ryzen 5950X: Likes itNVIDIA GeForce RTX: Likes it
I'm using qwen 3.6 35B unsloth 4 bit with my 5950x (128 gb memory) and a 3060 12 gb gpu with a self made harness. At 10k context I get about 40 tps generation and 500 tps prefill. At 100k context I get about 25 tps generation and 400 tps prefill. It works, but I often use gpt or claude to make a detailed enumerated plan of what I want to do first, then have qwen follow it. I'm not sure if it is economical or not, but I have solar on the roof so the power use is not really an issue and I already have the hardware. The biggest benefit for me is that it's all done locally, and I know the harness is not uploading anything or sending telemetry to someone else.
AMD Ryzen 5950X: Likes itNVIDIA GeForce RTX: Likes it
I got an entire Prebuilt PC when POW ended from a miner. AMD 5950x, 64gb ram, 3090, 2tb SSD, all for $1700. This was 4 years ago when the 3090 was $1500 by itself, and with minor upgrades it's still going strong today. It's wild to think that the system now is worth at least as much as I paid for it then if not much more than that. I saw a similar one going for $2500.
AMD Ryzen 5950X: Likes itNVIDIA GeForce RTX: Likes it
I didn't really understand the performance table until I saw the top ones were 8B models. But 5 seconds / token is quite slow yeah. I guess this is for low ram machines? I'm pretty sure my 5950x with 128 gb ram can run this faster on the CPU with some layers / prefill on the 3060 gpu I have. I also see that they claim the process is compute bound at 2 seconds/token, but that doesn't seem correct with a 3090?
AMD Ryzen 5950X: Doesn’t like itNVIDIA GeForce RTX: Doesn’t like it
The criteria for purchase for anybody trying to use it is "run slowly but acceptably" vs. "run so slow as to be unusable". My memory is wrong, it was the 32b. I'm running the 70b against a similar prompt and the 5950X is probably going to take over an hour for what the M4 managed in about 7 minutes. edit: an hour later and the 5950 isn't even done thinking yet. Token generation is generously around 1 token/s. edit edit: final statistics. M4 Pro managing 4 tokens/s prompt eval, 4.8 tokens/s token generation. 5950X managing 150 tokens/s prompt eval, and 1 token/s generation. Perceptually I can live with the M4's performance. It's a set prompt, do something else, come back sort of thing. The 5950/RTX3080's is too slow to be even remotely usable with the 70b parameter model.
AMD Ryzen 5950X: NeutralNVIDIA GeForce RTX: Neutral
The baseline draw is still much higher than lower spec alternatives, and it means you would need to cater for the high end scenario in your battery estimations if you intend to actually use it during power outages. I switched from a 5950X + 3080, to a 5700G APU, to finally an M1 MBP + Steam Deck last year for this exact reason. Far cheaper to have a 250wh battery that can handle those two for the ±2h outages every day.
AMD Ryzen 5950X: Likes itNVIDIA GeForce RTX: Likes it
+1. it's great that it compiles code faster, but i would also like when i'm writing code and hit enter, a newline is added and all lines under it goes down by 1 without any "lag". That's why i decided to go with a 5950x + 3090 rtx + gamer ram for my current "workstation" instead of a 5995wx + rtx 6000 + ecc ram for my current machine. even tho its not workstation rated, it seemed like the better way to go for quality of life feel. When the next generation of cpu/gpu/ram drops in the next couple eyars, i can re-evaluate the gamer vs professional offering to see if it matters.
AMD Ryzen 5950X: NeutralNVIDIA GeForce RTX: Neutral
if you installed via steam, check if the ultra-wide resolution fix will help - if you're trying to run in 4k. The 2080 ti has 12GB of vram, right? This is a mod that allows 2550x1440 or whatever it is to display properly, and yes, you will get letterboxing (or whatever it's called) - but it may run and then you'll know it is something about 4k on your hardware/software combo. if you're not trying to run 4k, there is a vague possibility that the number of threads is hampering your ability to play. I know, i don't really believe it myself, but lots of people mentioning that trying to disable SMT (or whatever) helped. FTR: RTX 3090, Ryzen 5950x, 128GB, game installed on sata SSD, and never had a graphical issue related to the hardware.
AMD Ryzen 5950X: NeutralNVIDIA GeForce RTX: Neutral
I'm in the process of testing the various self-hosted LLMs. I have an M2 MBA laptop and a 5950X w/ 64GB RAM and an RTX 4090 (24GB VRAM). I've used ChatGPT 3.5 and 4 quite a bit, and have done a bunch of comparisons w/ nat.dev's Playground between a variety of models (claude-instant provides gpt-3.5-turbo level output and is about 3-4X faster; gpt-3.5-turbo, text-davinci-003 to me are about equal and about the cutoff level of where they are generally useful for me - reliability as an end user for summarizations, Q&A, code assistance, etc). I found all the raw LLaMA variants I could run (up to 30B) to not be very coherent or useful. pythia, gpt-j, gpt-neox, chatglm and the other open raw models I found to be much worse than what the various eval scores would suggest (PIQA, HellaSwag, WinoGrande, ARC-e, etc)? I did a fair amount of playing w/ inference hyper-parameters early on to no avail, but did not do much k-shot learning or proper prompts (like the one's Scale AI uses for training). I tried a bunch of other Alpaca/instruction-tuned models and they're better, but IMO still not very good. GPT4All w/ the unfiltered checkpoint was the only one that did OK until I tried Vicuna (13B load-8-bit on GPU; I tried Baiz but wasn't impressed, have yet to try Koala, but don't have high expectations). Vicuna does a better job than GPT4All, but I did notice some of the going off the rails/not stopping - it however strongly leans on "as an AI language model..." responses - IMO, any fine-tu…
AMD Ryzen 5950X: NeutralNVIDIA GeForce RTX: Neutral
Counting my monitors and the small network switch on my desk, I can hit around 750-800 watts sustained with a 5950X and a 3090 running compute intensive workloads as measured at the wall by a power meter. So yeah, there's plenty of head space but I have also hit circuit breaker issues recently on a proper (non consumer) workstation so the limit is fresh in my mind.