Comment by izacus
11 hours ago
On the other hand - this is now an opportunity for Linux community to show that they are actually able to fund development of software for their platform, right?
Many HNers promised to pay if developers bring their software to Linux - will that actually happen?
What you say is ridiculous.
The only reason why the "Linux community" cannot create adequate FPGA design tools is that the vendors like AMD refuse to document the necessary details of their products.
A few old AMD FPGAs have been reversed engineered, e.g. some ARTIX-7, so for them there is no need for the rather bad AMD tools, but for most AMD formerly Xilinx FPGAs it is impossible to create better tools for lack of documentation.
As long as AMD refuses to provide the technical documentation required to use their products, it should have been a legal obligation to at least provide basic tools that allows the buyer of such products to actually use "FPGAs", i.e. to "field-program" them, as the name of the sold product claims.
Like many other FPGA developers, I could write myself better FPGA development tools than what AMD provides, if I had access to the complete FPGA technical documentation to which only a few big companies have access, a restriction whose only possible purpose is to prevent competition in the FPGA market.
If AMD had documented the exact format of the bit stream required to program each model of their FPGAs and the complete timing consequences of each synthesis choice, nobody would need any FPGA simulation or synthesis tool provided by AMD in Vivado.
[flagged]
>AMD refuse to document the necessary details of their products.
Because people haven't offered enough money to have a copy privately shared. This is on the Linux community for not ponying up enough money to fund this properly to have a reasonable release date.
The only reason why the "Linux community" cannot create adequate FPGA design tools is that the "Linux community" is completely inadequate in comparison to what's needed.
Reverse engineering tools are pretty good these days. I have no doubt that a dedicated hacker could sit down with Ghidra and the free Windows version of Vivado for two years and come out with something that compiles FPGAs well enough. But there's a shortage of the kinds of people who would do that, they're all busy doing other things, so it doesn't get done.
More easily, someone could get the free Windows version to run on Linux. If it doesn't already work in Wine, they could figure out and implement the needed Wine patches. If Vivado has a DRM scheme they could break it (potentially very difficult), if not then it should be straightforward. Nobody seems to be doing that, either.
The same applies to things like the Nvidia drivers.
In the past, freedom RE projects were handicapped by needing to maintain a Chinese wall. Now it's become obvious you don't need a Chinese wall, you can just straight up decompile someone else's software and use that as a reference as long as you don't copy it directly and you don't make it too obvious to the copyright owners what you're doing. Keeping your anonymity for this sort of project is easier than ever before too. Yet we see less freedom RE projects, not more. Why is that?
Because most don't want to work on something big, with a high risk of being sued into oblivion once released?
2 replies →
Vivado already supports Linux, the development is supported by very large customers that put FPGAs in cars, [REDACTED], and other kinds of objects that crash into other objects.
This is just hurting students and hobbyists.
Schools can join the AMD University Program and get back to where they were, and more. https://www.amd.com/en/corporate/university-program.html
As for hobbyists, in the world of $0.03 microcontrollers, strong competition on the low end from Chinese manufacturers, and where few people learn HDLs, is NRE money in the hobby market really money well spent?
I'm a HW designer and even I use microcontrollers now for most things, but not everything, because it's usually cheaper and faster.
With semiconductor prices coming down as far as they have I think the world has probably fundamentally changed for FPGAs and the niche they occupied is shrinking fast.
And, since there's a pipeline from students and hobbyists to professional use, it's risking the future.
And hurting AMD.
Nah. Why do Windows users get it for free while I have to pay because I'm an "advanced" user?
I'm not rewarding that. I'll reward companies like Valve instead.
This tier of the tool is free on Windows.
It might be a fair criticism that Linux users don't pay for software, but being a dick about it isn't going to get you anywhere.
(It's weird to see people on HN shilling for AMD against Linux, though. Very astroturf flavored)
[flagged]
The notion of being expected to pay for software that was formerly free - when Windows users aren't expected to bear those same costs - does indeed piss me off.
If I were actually using Xilinx FPGAs I'd be more pissed off. Luckily the projects that interest me currently are based around Intel, Lattice and Gowin devices.
More like the notion of seeing different treatment between OSes. No one likes being punished for a choice that shouldn't be any of the selling party's business. That's especially true in the Linux community, which was the target of Microsoft's anticompetitive policies for decades.
That's just like when macOS users got mad when they learned they were targeted by marketing schemes to sell them more expensive stuff [1].
[1]: https://www.npr.org/2012/06/26/155792590/orbitz-targets-mac-...
Maintaining software on Linux just takes more time and effort - due to less stability and number of distros.
1 reply →
I get what you're saying but I can understand the frustration here. Vivado licenses start at $1200/year or $5000 for a perpetual license.. Just to use software to work with hardware that you already paid for. And it's not like they are dropping support for Linux altogether, it would cost them nothing to continue supporting Linux in the free tier.
It just seems like a weird decision on AMD's part.