Boosting Computational Fluid Dynamics Performance with AMD MI300X

Re:Boosting Computational Fluid Dynamics Performance with AM... by gymbeaux on Monday January 20, 2025 @02:53AM (#42766061)

I recognize that selling enterprise hardware one or two units at a time to people like me is not cost effective and is why AMD isn’t doing it, but I don’t think there’s anything stopping them from relying on distribution partners like Sapphire, Gigabyte and XFX to handle everything but the GPU die. Demand would be low relative to consumer stuff, and after cutting these partners in they’re probably selling at cost or on thin margin, but again, if they want to carve out market share it’s going to be VERY slow-going if they continue with this “charge as much money as possible and only sell to datacenters” approach. Nvidia can do that because they’ve cornered the market.

Meanwhile I can’t even find an MI300X on eBay. I can at least poach enterprise Nvidia GPUs like the A100 on eBay. This tells me AMD’s shipping far fewer units and therefore enterprise GPUs aren’t doing much for their balance sheet (though I’d have to look at their quarterly and annual reports to know for certain). To me this strengthens the case for selling to individuals/startups, and at prices that offset the risk of picking AMD over Nvidia and potentially running into software shortcomings.

I’m set with two RTX 3090s at the moment, but it’s very neat that you’ve been able to bootstrap essentially a cloud service provider in the age of AWS, Azure and GCP (and DO and Vultr and Linode et al).

[ Parent | on HN ]

Re:Boosting Computational Fluid Dynamics Performance with AM... by latchkey on Monday January 20, 2025 @01:15PM (#42771425)

> I don’t think there’s anything stopping

There absolutely is. The current form factor is not standard PCIe. It is a OAM/UBB board that is custom designed by AMD to support Infinity Link. It only comes in an 8x configuration. Now, you're asking for a totally different design and that requires a huge investment that would take away from their existing focus on enterprise.

> Meanwhile I can’t even find an MI300X on eBay.

https://www.ebay.com/itm/305850340813

[ Parent | on HN ]

Re:Boosting Computational Fluid Dynamics Performance with AM... by gymbeaux on Tuesday January 21, 2025 @06:53PM (#42786806)

I just want the card, not an entire server with a cluster of them. You can’t get an MI300X on eBay. But yes I didn’t realize they don’t come in the standard PCIe form factor. Their previous Instict “GPUs” did.

They should have considered the RX 480 approach for the Instinct accelerators.

[ Parent | on HN ]

Re:Boosting Computational Fluid Dynamics Performance with AM... by latchkey on Tuesday January 21, 2025 @07:06PM (#42786929)

What you're asking for is HPC level of compute, but as a consumer product. They moved away from that market because they realized that it wasn't going to compete with the AI farms that Nvidia is building with companies like CoreWeave. The 355x is going to move even further away cause it'll require even more high end datacenter deployments.

So, on one hand, people want them to compete with Nvidia, but on the other hand we want them to ignore the market that is actually going to make them money. As much as it would be nice to have, we can't have it both ways.

The middle ground is to rent a single one from us billed by the minute (or another neocloud). We handle all the detailed problems for you (don't forget the massive upfront capex spend), and you get to build your products/companies on that. Once you grow to the point of being able to buy your own equipment, we can even help you deploy it.

[ Parent | on HN ]