Open Source Software Community - it's about ethics in Code of Conducts

[Ed Special]

Nobody's ever going to make a nerd box that sits on your desk and replicates the function of modern CPUs, allowing you to bypass Intel and AMD.

Best we can hope for is for VIA to decide they want to do something with x86/x64 again (besides just licensing their license out to Chiniggers)

 

[Lady Ivlia de Nîmes]

But sometimes you can get a good price, for example if you were a university you could get custom chips for basically free because they sometimes have empty space on wafers if things don't align right

Keep in mind that the resolution of lithography procedures most universities operate isn't that high, so you suffer a similar issue as if you were using a FPGA; just significantly less flexible.

 

[Cedric_Eff]

Best we can hope for is for VIA to decide they want to do something with x86/x64 again (besides just licensing their license out to Chiniggers)

That probably never gonna happen considering you still need to license from AMD to make anything AMD64 related.

 

[Creative Username]

That probably never gonna happen considering you still need to license from AMD to make anything AMD64 related.

what if we just threw x86 in the garbage where it belongs?

 

[Coolio55]

i think it's actually somewhat possible to get scarily close to the state of the art without relying on expired patents
just look at av1 for instance
we could probably get a weirdly anachronistic core 2 duo or something that way

It makes a man think. CPU speeds have barely progressed in the last 15 years or so.
The hard part is the GPU part.

 

[snotang]

Nobody's ever going to make a nerd box that sits on your desk and replicates the function of modern CPUs, allowing you to bypass Intel and AMD.

I have used three eras of CPUs directly: ones from 2001, 2007-2008ish and ones from 2012. I am still using the one from 2012, which is a 4-core AMD APU, and it works quite well for my purposes, except that I can't use linux-libre and have reasonably-functioning video output at the same time.

All that to say that if you're chill enough with old hardware you don't need to replicate the function of modern CPUs to bypass Intel and AMD. Although to be fair, I still don't internally think of the 2012 one as anything but a "modern CPU", since I went straight from playing call of duty 4 at 15 fps on a celeron D running at 2GHz to playing at 90fps at max settings.

While we're talking about open hardware, anyone have any updates on lkclware? Will 2030 be the year of the eoma68 desktop? Is the libre-riscv-but-not-really-actually-libre-power9-actually-with-some-custom-extensions coming along well, or has the critical mass of autism fallen into politisperging in these trying times?

 

[prollyanotherlurker]

Best we can hope for is for VIA to decide they want to do something with x86/x64 again (besides just licensing their license out to Chiniggers)

I have hopes for RISC-V. We actually are starting to see them get decent enough to start making their way into proper laptops. It's still early on. But I think it's a good sign. It's definitely one of the best bets we have now.

At least for more open hardware.

That said. I wonder what kind of nightmares go on in things like modern intel wifi chips. I've never looked into it. But knowing intel I wouldn't be surprised if there is some eyebrow raising stuff happening there.

 

[HootersMcBoobies]

Nobody's ever going to make a nerd box that sits on your desk and replicates the function of modern CPUs, allowing you to bypass Intel and AMD.

There are options, but not great ones. Maybe RISC-V (one day). Ampere makes a desktop ARM board

https://www.ipi.wiki/products/ampere-altra-developer-platform?variant=45122307096738

and Raptor still makes POWER9 boards:

https://www.raptorcs.com/content/TL2B02/intro.html

Best we can hope for is for VIA to decide they want to do something with x86/x64 again (besides just licensing their license out to Chiniggers)

That probably never gonna happen considering you still need to license from AMD to make anything AMD64 related.

Yea I was wondering about that. There are other companies that had old x86 licenses (like IBM), but does anyone else actually have both a license for x86 and amd64?

So many companies dropped the ball yeas ago, like Transmeta. Maybe the move away from Windows 11 will move the needle on alternative architectures.

 

[Ferryman]

That said. I wonder what kind of nightmares go on in things like modern intel wifi chips. I've never looked into it. But knowing intel I wouldn't be surprised if there is some eyebrow raising stuff happening there.

You can probably assume anything Intel/AMD/Broadcom = tainted. None of these companies has a good track record of consumer-friendly firmware implementations.

Nobody's ever going to make a nerd box that sits on your desk and replicates the function of modern CPUs, allowing you to bypass Intel and AMD.

The peeps at MNT are certainly trying with their Reform laptops. Just like the POWER9 rigs, they're currently very, very pricy for the power they offer, but its still a good option if you happen to have very deep pockets.

 

[Gatdam Animal Person]

PreserveTube
Linux Frog man pretends to be impartial about the NixOS CoC drama, saying he doesn't pick a side between the moderators and CoC committee, then completely contradicts himself by disavowing somebody saying CoCs are causing such issues like within NixOS. Makes it seems like he's subtly agreeing with the overreach of CoCks.

Do anyone of you knows a YouTube channel about daily linux news without this guy's completely retarded takes AND without looking like an utter retard like Lunduke?

 

[Cedric_Eff]

So many companies dropped the ball yeas ago, like Transmeta. Maybe the move away from Windows 11 will move the needle on alternative architectures

Like RISC-V or like ARM?

 

[Ed Special]

That probably never gonna happen considering you still need to license from AMD to make anything AMD64 related.

Yea I was wondering about that. There are other companies that had old x86 licenses (like IBM), but does anyone else actually have both a license for x86 and amd64?

VIA has been making amd64 SoCs since Bush was still in office. This is why the Chiniggers have been licensing their licenses for their own "home-grown" silicon.

 

[Cherenkovblue]

idk they seem to be some pretty fucking good programs
also i don't think we would have the gpl and the general good state of things without him

1. As someone who has to use them on a regular basis... no, they're really not. They just happen to be free.

2. The GPL was always a horrible license predicated on the false premise that source code was valuable. It wasn't, and it isn't. Ideas, not source code, are what matter, and those aren't protectable by copyright. People are just starting to figure that out as autists and corporate overlords play with LLMs to scrape codebases and use them to generate their way to fame and fortune. LLMs are a very clumsy way to do this, but it's the most accessible way to do it at the moment. As more reliable methods becomes available for natural language processing you'll see what I mean.

did he ever actually say that it was literally impossible, or just not worth chasing at the moment because 90s fpgas weren't good enough to run modern software?

Free software is often available for zero price, since it often costs you nothing to make your own copy. Thus the tendency to confuse ``free'' with ``gratis''. For
hardware, the difference between ``free'' and ``gratis'' is more clear-cut; you can't download hardware through the net, and we don't have automatic copiers for
hardware. (Maybe nanotechnology will provide that capability.) So you must expect that making fresh a copy of some hardware will cost you, even if the hardware
or design is free. The parts will cost money, and only a very good friend is likely to make circuit boards or solder wires and chips for you as a favor.

Because copying hardware is so hard, the question of whether we're allowed to do it is not vitally important. I see no social imperative for free hardware designs
like the imperative for free software. Freedom to copy software is an important right because it is easy now--any computer user can do it. Freedom to copy
hardware is not as important, because copying hardware is hard to do. Present-day chip and board fabrication technology resembles the printing press. Copying
hardware is as difficult as copying books was in the age of the printing press, or more so. So the ethical issue of copying hardware is more like the ethical issue of
copying books 50 years ago, than like the issue of copying software today.

-- Richard Stallman -- On "Free Hardware", 22 Jun 1999, Linux Today

He literally shit on the very idea of "free hardware" because:
1. He didn't realize what was then possible with FPGAs; and
2. He didn't think there was a social imperative for it in the same sense as free software because of that.

i'm not saying that, just that fpgas are a far different beast than dedicated hardware and you can't get 2004 cpus with a 2004 fpga or 2025 cpus with 2025 fpgas

No one said you could, but more to the point, why would you want to? Do you realize just how much cruft is buried in the x86 architecture? There's a reason OEMs of mobile devices switched to ARM.

still, the thought of running everything on fpgas is just somehow not very satisfying... and i don't like the idea of having to go all the way back to 2004-era hardware. that is not exactly "modest needs", it's more "retrocomputing enthusiasm" in my book
it is good for like 80% of the things i would ever use a computer for, but that last 20%...

What are the 20% of things you think you can't do? 3D gaming? Mining bitcoin? Running bloated LLMs?

I'd love to see a system that uses solely open code and expired patents - something that emulates maybe a Pentium II or an AMD Opteron and uses ps/2 and vga. 100% open and with a pinout board so we can experiment with networking protocols and such. It wouldn't replace my existing system, at least not right away

Well, you know what you need to do, right? Grab the ao486 FPGA core and add the appropriate missing opcodes. Not a very huge undertaking since the majority of the work is effectively already done. At that point you'd more or less just be adding MMX support.

I don't know how much work that would be, or if that work has already been done. The expired patents represent technology that has already been developed but can be built upon, and I'm noit sure about how much effort it would be to make a CPU roughly equivalent to a Core 2 Duo from scratch without basically copying the existing design.

Answer: Not very much effort since there's already the ao486 core to start from. Even if you mean "I want it in raw silicon, not an FPGA" this still applies as there exist development pathways to convert an FPGA design into hardwired designs.

some stuff like VGA and PS/2 allows you to use existing technology instead of having to build displays and keyboards from scratch that use a different standard. I think USB doesn't require a license, so yeah you could probably make a relatively modern system that doesn't rely on patents (which of course means that once a design is made Chinese manufactures will churn them out by the billions)

You are very, Very, VERY wrong about USB. If you want to say your stuff is USB compatible you most definitely have to pay a licensing fee for USB-IF's certification marks.

Nobody's ever going to make a nerd box that sits on your desk and replicates the function of modern CPUs, allowing you to bypass Intel and AMD.

"Nobody is going to make a nerd OS that sits on your desk and replicates the function of Unix, allowing you to bypass AT&T."

I already have a box the size of a pack of playing cards that replicates a vintage 2000 PC, and FPGAs continue to advance in capability. It's only a matter of time.

 

[Ed Special]

Do you realize just how much cruft is buried in the x86 architecture?

This is actually a greatly exaggerated issue. Most modern CPUs actually use a lot of clever emulation tricks to fake Good Enough™ approximations of legacy features to avoid losing performance and efficiency to them.

There's a reason OEMs of mobile devices switched to ARM.

Literally the first company Apple approached looking for CPUs for the original iPhone was Intel, and Intel refused because they thought the iPhone was going to be a commercial failure. ARM was their backup plan. Mind you, this was while AMD's Athlons and Phenoms were actively kicking the shit out of Intel at every turn on power efficiency. Stop uncritically eating up the revisionist tech history myths.

 

[Betonhaus]

You are very, Very, VERY wrong about USB. If you want to say your stuff is USB compatible you most definitely have to pay a licensing fee for USB-IF's certification marks.

Can you just call it the unlicensed rectangular port and get away with that?

 

[Creative Username]

- snip -

you are even more completely fucking delusional than i am good job

Can you just call it the unlicensed rectangular port and get away with that?

i have a weird suspicion that this is what many usb devices do in practice

 

[prollyanotherlurker]

I'm not a member so I can't watch it. Idk if this is different from the other video he did. But c'mon lunduke. You know that's bullshit.

 

[volatile puar]

Literally the first company Apple approached looking for CPUs for the original iPhone was Intel, and Intel refused because they thought the iPhone was going to be a commercial failure. ARM was their backup plan. Mind you, this was while AMD's Athlons and Phenoms were actively kicking the shit out of Intel at every turn on power efficiency. Stop uncritically eating up the revisionist tech history myths.

FYI, Steve Jobs approached Intel hoping to secure their XScale ARM mobile chips for the iPhone, it wasn't until later that Intel sold their rights and announced an x86 based mobile chip.

 

[The Noise]

The GPL was always a horrible license predicated on the false premise that source code was valuable.

Reminder that Stallman, the inventor of the GPL, had to invent a much more permissive copyright license for gcc headers because nobody would have used his compiler if it was under the GPL. The GPL infects codebases and inconveniences people by design. I much prefer the MIT license for this reason.

View attachment 7998040

I'm not a member so I can't watch it. Idk if this is different from the other video he did. But c'mon lunduke. You know that's bullshit.

Is BcacheFS made by the guy who keeps trying to cowboy his patches past the release window? Linus telling someone to STFU is honestly pretty tame for him.

 

[Creative Username]

Reminder that Stallman, the inventor of the GPL, had to invent a much more permissive copyright license for gcc headers because nobody would have used his compiler if it was under the GPL.

yes, there were many other c compilers out there, and he wanted a superior c compiler that anybody would be able to use for anything the other c compilers could be used for, which means a fully gpl c compiler would be kind of stupid
also he has advocated licensing certain libraries under very permissive licenses so absolutely everything can e.g. support ogg. if people's proprietary shitware supports more freedom-friendly file formats, the state of user freedom is still improved

The GPL infects codebases and inconveniences people by design.

yes it does! thankfully most of the people it inconveniences are people trying to write proprietary software
this is a very good effect because severe inconvenience is exactly what proprietary software writers deserve

I much prefer the MIT license for this reason.

cool fact: the intel management engine is (was?) based on a permissively licensed kernel called minix made by andrew tanenbaum
they did not even notify him after they profited off his work for free
using a cuck license can be incredibly stupid. remember: when you license under mit, you're doing it for big tech for free