Hacker News Reader

Comment by Trisell | Hacker News Reader

I think the bigger thing for Open Source Chips is going to be the expiration of a ton of the x86 patents next year. That is going to free up a whole host of people to take a look at trying to build A better open x86 processor. Maybe then we can finally get to having a truly open source server from chip through software.

View on HN

What’s the point of open source for consumers, expect for maybe education, when you can’t produce something from the source yourself nor be sure the product you buy is made from the source?

It’s very cost prohibitive to run a batch through a semiconductor fab. You are not going to request one offs unless you have a few millions to spare.

It’s not like running make after cloning some sources.

Well the barriers are pretty low today. Even fairly small companies like Axis (they make security cameras) or ubiquiti (routers and access points) can work with CPU vendors to license their tech and get a custom CPU. With an opensource IP the barrier will be even lower. Maybe even within a well funded kickstarter range.

Maybe for a killer home router with a 100% open hardware+software platform that can route 8 ports of 5Gbit/port at line speed that would more trusted than random commodity hardware.

Or something to handle say 8 security cameras and use machine learning to handle all 8 streams for not just motion detection, but also identifying what person/object is in each stream.

I'd certainly pay a premium for smart devices in my home that I knew that I could trust, wasn't spying on me, doesn't require cloud connections, used open APIs/standards, and wouldn't die with the next time a company dies, gets bored, gets greedy, get purchased, etc.

This is not accurate, Axis is NOT a small company by any standards, they have done ASICs since the 1980's and a tape-out there now is in the $10M range or more I'm sure, and then we're talking about the same amount in salaries to do all design and testing.

It's fun to have access to HDL code but it's simply not possible to actually turn it into chips as an OSS effort.

FPGA-stuff is the way to go I think. The HW cost will be much higher but you can muck around with it as a lone engineer in your home with a turnaround time of minutes.

I’m curious about the ubiquiti custom CPU, can you share a link to more info ? Thanks

"Custom CPU" seems like a bit of a stretch. They have ASIC people on staff (eg https://rocketreach.co/ya-chau-yang-email_61973123), but based on the fact that you can flash basically all their devices with OpenWRT (including those disc-shaped APs), I think the "ASIC" is just a conventional ARM or MIPS core with some peripherals geared toward hardware acceleration of common network tasks— switch routing, vlan tagging, etc.

This page talks about the specific features available for hardware offloading:

https://help.ubnt.com/hc/en-us/articles/115006567467-EdgeRou...

>What’s the point of open source for consumers

faster iteration and development and more competition means more innovation and cheaper prices and more choice for consumers.

The primary benefactor of open source has never literally been the consumer because only an exceedingly small niche of people makes use of the source directly. The biggest benefactor is the ecosystem as a whole.

For consumers the great thing about open source has never been that they can run make in a terminal. It's been that people can run make in a terminal who can then produce products for consumers.

There are efforts such as https://libresilicon.com/ to design open source processes to manufacture silicon chips at the small scale. I am hopeful that these efforts will bring down one-off cost to the 10k USD level or less.

> What’s the point of open source for consumers

See Venezuela and Adobe. See the trade war with China and Huawei's position...

I don't understand why this is. As the high end stuff gets down to 10nm-ish, shouldn't it have gotten way cheaper to fabricate micron-precision chips?

AIUI, you can just use FPGA on a 10nm-ish process and get roughly the same performance as hardwired logic on a micron-precision chip. Still useful for analog though, and I imagine that going even a bit lower but still very cheap (in the hundred-nm range, perhaps) would still yield better performance than a state-of-the-art FPGA, given the one- or two-orders-of-magnitude overhead that's inherent in those.

Is there's open-source FPGA? (I mean not reverse-engineered which relies on mercy of the manufacturer who can discontinue it anytime)

Just for that reason there's no point really, you can just switch in the next generation FPGA (more or less) as the high-level HDL stays the same.

Then there is no point in open source C compiler, either? The language stays the same...

Open source doesn't have to mean without resources. Imagine the big tech companies banding together to build better processors in the open.

For example Google's OpenTitan.

It opens the source so many more companies can work on products enjoyed by consumers.

Wouldn't it be x86-64 rather than x86 patents which are due to expire next year? It's weird that we've seen no interest so far in an open source reimplementation of x86 16-bit or 32-bit. You would think that it ought to garner quite a bit of interest among retrocomputing enthusiasts, if nothing else.

I'm sure there's tons of patents, but the p6 core came out in 1995 and most of the 32 bit stuff predates that and has all been patent free for quite a while. I guess if you're building your own processor, why make an x86?

Which specific patents ? Isn't the ISA protected by copyright rather than patents ? And implementing x86 is anyway not that attractive due to the complexity of the frontend. If you had to do something from scratch, why would you do x86 today ?

> Isn't the ISA protected by copyright rather than patents ?

Interfaces are uncopyrightable, according to everyone but the Court of Appeals for the Federal Circuit. And the case where the latter said it was copyrightable has been appealed to the Supreme Court specifically to overturn that ruling, and the Supreme Court will probably do so.

I'm aware of the google case, but "ownership" of ISAs seems to be a settled thing. That's partly what ARM's business is (i.e., to license the ISA to Apple for example). Do you have a clarifying source on copyright v/s patents regarding ISAs? RISC_V would also not be so significant if it were just a matter of running out patent clocks. MIPS is quite old too and would have expired soonish by the same logic (MIPS's own open-source announcement notwithstanding). Intel has also been passive-aggressively threatening MS and others legally so that they don't emulate x86 on ARM hardware. I remember when MS showcased some ARM laptop, Intel Legal sent out some press briefing claiming how they'll protect their IP. Qemu gets a pass though since it's non-commercial and there is nobody to sue. It would also be bad PR.

Edit: You may be right that it's only patents afterall and not copyright. This was Intel's briefing:

"However, there have been reports that some companies may try to emulate Intel’s proprietary x86 ISA without Intel’s authorization. Emulation is not a new technology, and Transmeta was notably the last company to claim to have produced a compatible x86 processor using emulation (“code morphing”) techniques. Intel enforced patents relating to SIMD instruction set enhancements against Transmeta’s x86 implementation even though it used emulation. In any event, Transmeta was not commercially successful, and it exited the microprocessor business 10 years ago."

which patents? does it include 64 bit?

surely most of the x86 patents have expired, yet i don't see any interest in open source implementations.

The AMD64 patents. The first Athlon 64 was released in 2003, so the architecture is still protected. Until those patents expire, the best you can do is a 32-bit P6/K7 equivalent.

> I think the bigger thing for Open Source Chips is going to be the expiration of a ton of the x86 patents next year.

AMD has been building x86 chips and the patents haven't stopped it. ARM has been building non-x86 chips and the patents haven't stopped it. I'm failing to see how the expiration of x86 patents will have any effect.

That’s because AMD licenses the x86 patents from intel and Intel licenses the amd64 patents from AMD. There are a few other niche x86 licensees (VIA, DM&P, etc.) that don’t make anything in the high end market.

Reply on news.ycombinator.com