With 7.9, shaping (read: bandwidth rate limiting) no longer tops out at 4 Gbps. PF could always process/transfer beyond 4 Gbps, presuming you had fast enough hardware to handle such bandwidth. The discussed limit was specific only to queues, when using such to shape traffic.
If you're asking about OpenBSD/PF's general network performance, it is finally performing acceptably since a couple of years back. You can easily saturate a 2.5 GbE interface with low-end hardware.
thanks a lot. yeah these models are way better than our previous launch. our 15M model now is better than our previous 80M model and we expect to continue seeing this rate of improvement.
> it can generate a 50 steps 512x512 image around 1 minute and 50 seconds.
I have the 4650G APU, and the best way to describe it is: lacking of support. This was even more true 3 yo than now. rocm (is) was absolutely dogshit then, I know this because I tried to do the same when that post was made. You have to compile everything from scratch, get the relevant patches, and even then, xformers which is a library that accelerate diffusion model inferencing was not supported for renoir or rocm back then. Yes, you can generate an image, but it was much slower, and rigged with bugs. You couldn'rt update rocm because it broke compatibility, and it was partly the reason I got into nixos. That being said, those APUs are a power house. Nowadays I can run decent agentic workflows on them (I have 64gb of ddr4 ram, ie APU can suck as much as it needs with the latest linux kernels).
Just note, diffusion models are still second class citizens on AMD apus even GPUs. But then again, nothing close right now on the market except for what apple offers.
The Ryzen AI CPU/GPUs (Ryzan AI 395+ etc) seem to have increasing support - https://lemonade-server.ai/ now has support for the NPU as well as the combined CPU/GPU (which I guess is a APU but is different to the G series of APUs I think?)
But I'm always interested in first hand experiences of how good is it really - I'm pretty cynical about the idea that AMD actually knows what it takes to build good software end-to-end.
I also have one, and indeed support is very much frictionless now compared to a year ago. But again, not thanks to AMD, as initially it was purely community driven. Strix halo was not even supported by ROCm (officially), and we had to deal with therock images, then donato made the toolbox, and then lemonade came through. I am really surprised how AMD approached this. They made big promises, they threw the hardware out, it really is amazing piece of hardware given what you can do with it, but it was left hanging without support for AI stack for months even though it had it in its name. Contrast that with the DGX spark (yes it had and still had bugs in its kernels, but cuda worked on day 1) and you can see the difference. Nvidia is selling an ecosystem, AMD is selling hardware. I really hope AMD focus on the software layer more.
Note that Lemonade Server uses NPU low-level code that's proprietary, not available as open source. It would be nice to work on a fully open alternative, perhaps by exposing the NPU itself as a Vulkan Compute-capable device, that shaders can be auto-compiled to.
I run DNSSEC (to facilitate DANE) and with regards to DNSSEC I haven't had to manually edit my zones in years either. Unlike yourself I don't consider DNSSEC deployment or ZSK rotation / KSK roll-over scary or complex, and seeing an adept technician dole out warnings on the level of "don't run with scissors" is pretty peculiar.
HTTPS also has expiring keys that also need to be rotated. Most people outsource this to a service provider for them - as is the case with DNS. It's weird how people gripe about standard cryptography/PKI when it comes to DNSSEC but not HTTPS.
And stuffing your PC case with various disco lights doesn't affect its performance either. But an evidently large portion of gamers are, well, special people.
I like it. I don't actually do it myself, the only lights in my case are the ones on my 4090 FE (just white ones, the FE are very understated designs). But I don't really mind it either. I like cyberpunk aesthetics.
I do often dress myself up with RGB lights however :)
It does, however, open up your computer to fun security bugs that were totally unnecessary, but you know, you gotta make your computer look like modern art.
Aesthetics are more important for some people than The 0.01% percent chance of an obscure, 8 year old security annomaly. I think those computers look horrible, but they are harmless.
Because you have motivated reasoning to dislike these companies, even though Blackrock and Blackstone are bog standard financial services companies and a random naming scheme is easy to grab onto.
All the worst companies seem to all be LOTR themed.
Technically the Palantiri were a force for good in the hands of Elves and Men, and could still be used for good, like Aragorn using it to challenge Sauron and forcing Sauron’s hand. So that’s a defense to the self-awareness argument. In fact that ambiguity is likely intentional.
Btw I always wondered why I was seeing droves of Palantir swag on Stanford campus back in early 2010s. I wouldn’t wear something that has a 50%+ chance of being interpreted as evil.
The Palantir themselves aren’t evil, they were made by the elves long before the events of LOTR. Essentially they are just a tool.
However I heard that Thiels favourite book is the rewrite of LOTR from the perspective of Sauron, where Gandalf and the elves seek to destroy humanity and technology (at least that’s how I understood the gist, haven’t read it)
If you're asking about OpenBSD/PF's general network performance, it is finally performing acceptably since a couple of years back. You can easily saturate a 2.5 GbE interface with low-end hardware.
reply