Hacker News Reader

Comment by Mizza | Hacker News Reader

I have an AI chemistry project that appears to work on my test data, but I had to put it on the backburner because I simply can't (couldn't) find a 3070/80 anywhere! I stopped looking 6 months ago, does anybody know a reliable place where I can snag one?

View on HN

If you don't mind having a gaming PC (stupid RGB lights and fans) then Build Redux [1] is not price gouging. You can basically just think of the GPU markup as a builder fee. They are incredibly slow to ship but the lead time is something like a month which beats your 6 months. At this point, it would take bitcoin dropping below 10k for the market not to be silly.

[1] - https://buildredux.com/

Isn't it Ethereum driving the GPU shortage? I thought Bitcoin was not really profitable without an ASIC rig.

That's correct. Saying BTC was lazy of me but I meant that basically the price of all things crypto need to be quartered due to the correlations.

You know what, this might actually be the guy. Good tip. It's a bummer to pay for a Windows license I don't want, but the fee for that is less than the markup I'd pay to some scalper just to get the card.

If you don't want Windows, you can remove it!

It is probably about the principle.

Given that the price goes down $109 with Windows removed, I'd suspect you're not buying a license in this case.

Have you considered renting cloud GPUs by the minute? You can rent some pretty powerful GPUs from cloud providers (i.e. https://cloud.google.com/compute/docs/gpus)

If you are really on a tight budget, but have time to fiddle, you can try to make something work on top of Google's colab, too.

I wonder why you've not teamed up with some uni-lab, they usually have some hardware and also some free labor if you share credits. A single GPU also means, whatever your idea, it probably would have finished already on a CPU ;).

Good idea with the uni lab!

About CPU vs GPU: things seldom work out on first try. A GPU gives you much lower latency for trying out a series of ideas.

If you're willing to pay scalper prices, there's a relatively consistent availability. If you're looking for closer to MSRP, they're in short supply and your choices are either waiting lists or racing other people for online restocks. I got a 3080 a month ago only after waiting on EVGA's wait list for just over a year.

Why do you need a 3070/3080 specifically? If it's to run something like Tensorflow or CUDA code more generally, could you do it with an older card, or the more available 3060s?

I dont know if it has enough ram for your purposes but I also gave up and just bought an entire computer to get a 3080 and got it in 3 weeks before Christmas https://skytechgaming.com/product/shiva-amd-ryzen-5-5600x-nv...

> because I simply can't (couldn't) find a 3070/80 anywhere

This is false.

You can easily find thousands of brand new Nvidia 3070/3080 GPUs online.

The problem is, you wish to pay MSRP. Supply and demand doesn’t work in your favor here.

This is a little disingenuous, that is like saying scarcity doesn't exist since you can buy almost anything at any price. There is always an implied "in a reasonable budget".

In this situation, yes.

In general: price competition might be effectively outlawed in some circumstances.

Eg you can't really buy a replacement kidney for any amount of money (outside of Iran).

What about moving to the cloud? Something like CoLab?

Yeah, I guess I could go down that route, I've used the AWS Gx instances for projects before, but this dataset would justttt fit in memory for a 3080, which really simplified the rest of the code, and the speed of iteration, and, at the end of the day, quite frankly I just want one. I'll do more weird stuff if I don't have the meter of how much I'm paying to Jeff Bezos running in my head every time I run an experiment.

Colab GPUs have more vram than a 3080 let alone a 3070.

just bought one with a PC from falcon northwest. Not sure about buying it standalone but they sell them with their PCs

vast.ai, nothing else is remotely as cheap

The price is great, the downside is that you do not know whether the server owner reads your data. The "jobs" run in a Docker container on somebody's machine.

Yeah. Use fresh ssh keys, and don't train with sensitive data. If that's an option for you, vast is great.

Reply on news.ycombinator.com