Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Arm is canceling Qualcomm's chip design license

bloomberg.com

Post reply on HN
Print view View original

431–440 of 480 posts

Re: Arm is canceling Qualcomm's chip design license

#431

Post by ksec »

starspangled wrote: post #394
fidotron wrote: post #300

Earlier quoted context omitted.

Brucehoult mentions it further down. > This time last year they were all over the RISC-V mailing lists, trying to convince everyone to drop the "C" extension from RVA23 because (basically confirmed by their employees) it was not easy to retrofit mildly variable length RISC-V instructions (2 bytes and 4 bytes) to the Aarch64 core they acquired from Nuvia. At the same time, Qualcomm proposed a new RISC-V extension that…

I don't really see how this follows. Fixed length instructions are a well known and highly desirable property for instruction decoders (at the cost of some density), which has been known about and done since the first RISCs. And ARMv8 is basically just a warmed-over RISC with very little new or innovative in there. It's just an ISA. ARM Ltd is basically rent-seeking on keeping the ISA proprietary, same as Intel and A…

>If ARM gets upset because someone wanted fixed-length instructions standardized in RISCV that's really the height of hypocrisy.

ARM never said that.

>but not to the value of extracting hundreds of millions of dollars ever year.

ARM makes money on selling ARM design, not from their ISA licensing.

Re: Arm is canceling Qualcomm's chip design license

#432

Post by monlockandkey »

bhouston wrote: post #248
bufferoverflow wrote: post #172

Earlier quoted context omitted.

Samsung and MediaTek make pretty competitive ARM processors. MediaTek 9400 is literally the top-performing SoC on the market. Samsung Exynos 2400, 2400e MediaTek Dimensity 9400, 9300 Plus, 9300 https://nanoreview.net/en/soc-list/rating

The geekbench scores of Apple are highest and then Qualcomm according to this page. MediaTek’s top chip is literally 25% slower according to his listing. The ordering in ranking is a little weird.

People do not seem to understand that these chips are in different price brackets, even between Qualcomm and Mediatek. That is why there is a discrepancy in performance.

Would not be fair to compare a $20 toaster to a $50 toaster and say that the $20 toaster is slower.

Re: Arm is canceling Qualcomm's chip design license

#433

Post by ksec »

rickdeckard wrote: post #240

What is often overlooked on this topic is, that ARM also has a duty to protect its ecosystem. By using its dominant position in Smartphone chipsets, Qualcomm is in progress to establish a custom ARM-architecture as the new standard for several industries, fragmenting the ARM-ecosystem. For decades, ARM is carefully avoiding this to happen, by allowing selected partners to "explore" evolutions of the IP in an industry…

>What is often overlooked on this topic is, that ARM also has a duty to protect its ecosystem.

It is not overlooked. The duty of any company to protect its IP and contract during dispute is largely if not entirely irrelevant on internet inclusive but not limited to HN. They simply want whatever company they like to win and the one they hate to lose.

This has been shown repeatedly on Apple vs Qualcomm in modem and IP licensing. Monopoly trials, or Apple vs US etc.

And just want to say thank you. You are one of the very very few to the point I can count them with fingers, to actually dig into court case and not relying on whatever media decided to expose us to.

Re: Arm is canceling Qualcomm's chip design license

#434

Post by ksec »

adrian_b wrote: post #342
dagmx wrote: post #306

Earlier quoted context omitted.

There’s a LOT of conjecture in your comment. Your very first line is one to begin with. ARM also doesn’t seem to care if QC design their own cores. They just care that they renegotiate the royalty agreement. This is clear if you actually read their statements.

Because Arm wants to increase the royalties for the cores designed by Qualcomm, that is a pretty certain indication that these royalties have been smaller than for the cores licensed from Arm. Therefore Arm cares a lot if Qualcomm designs their own cores, because that would cause a smaller revenue for Arm. If Arm had not cared whether Qualcomm designs their own cores, they would have never sued Qualcomm. The official…

ARM doesn't care Apple design their own cores.

ARM is perfectly happy for Qualcomm to design their own core as long as it was the agreed rate for those sector.

ARM is happy to compete on design, which is what the Cortex X5 is doing. And shown just as competitive against Oryon.

The rest of your comments are like making up stories to back up whatever you think is the truth. And most of them have zero factual basis.

Re: Arm is canceling Qualcomm's chip design license

#435

Post by ksec »

MBCook wrote: post #10

Let’s just assume this happens for a moment. What do Android OEMs do? They can’t use Apple chips, or now Qualcomm chips. Switching to another architecture is a big deal. Would this basically hand the Android market to Samsung and their Exynos chips? Or does another short term viable competitor exist?

1. Qualcomm is still allowed to buy , use and sell ARM's Design IP. i.e Going back to Cortex X series.

2. Mediatek is available, mostly with latest ARM's IP. And extremely competitive. The only thing missing is Qualcomm Modem. It isn't Mediatek's modem are bad, they are at least far better than whatever Apple's Intel Modem had used or planned. The only problem is Qualcomm is go good customers still prefer it for the relatively little price they are paying for.

3. It is not like Android OEM cant make their own SoC. Especially considering the smartphone market can now be largely grouped as Apple, Samsung and Chinese. Together they are 95%+ of Market share.

Re: Arm is canceling Qualcomm's chip design license

#436

Post by snvzz »

x0x0 wrote: post #423
snvzz wrote: post #400

Earlier quoted context omitted.

Note that these are just a person's own opinions, obviously not shared by the architects behind RISC-V. There are multiple approaches here. There's this tendency for each designer to think their own way is the best.

I get that. I just work quite distantly from chips and find it interesting. That said, licensing an instruction set seems strange. With very different internal implementations, you'd expect instructions and instruction patterns in a licensed instruction set to have pretty different performance characteristics on different chips leading to a very difficult environment to program in.

Note that this is not in any way a new development.

If you look at the incumbent ISAs, you'll find that most of the time ISA and microarchitecture were intentionally decoupled decades ago.

Re: Arm is canceling Qualcomm's chip design license

#437

Post by bufferoverflow »

bhouston wrote: post #248
bufferoverflow wrote: post #172

Earlier quoted context omitted.

Samsung and MediaTek make pretty competitive ARM processors. MediaTek 9400 is literally the top-performing SoC on the market. Samsung Exynos 2400, 2400e MediaTek Dimensity 9400, 9300 Plus, 9300 https://nanoreview.net/en/soc-list/rating

The geekbench scores of Apple are highest and then Qualcomm according to this page. MediaTek’s top chip is literally 25% slower according to his listing. The ordering in ranking is a little weird.

You got even that wrong. Single-core GB score is highest, multicore is not. Snapdragon's score is higher.

Re: Arm is canceling Qualcomm's chip design license

#438

Post by MBCook »

6SixTy wrote: post #427
MBCook wrote: post #151

Earlier quoted context omitted.

> If anything, offering RISC-V support as well would further reinforce the idea that Windows is ISA-independent, and not tied to x86 anymore. Anymore? It’s been independent since the 90s. It’s only ISVs that have been an issue. And a rug pull is a fantastic way to scare all the ISVs far far away.

You sure? Microsoft dropped Alpha, MIPS, and PowerPC by the time Windows 2000 rolled around. Beyond that point, only the Xbox 360 and Itanium versions had anything different to the usual X86/64 offering.

While there was only one popular choice, they’ve always kept it flexible. That was a core design decision.

But, as an example, Windows Phone 8 and later were based on the NT kernel. You already mentioned the 360.

Re: Arm is canceling Qualcomm's chip design license

#439

Post by brucehoult »

eqvinox wrote: post #303
brucehoult wrote: post #14

Earlier quoted context omitted.

Qualcomm have a Plan B. This time last year they were all over the RISC-V mailing lists, trying to convince everyone to drop the "C" extension from RVA23 because (basically confirmed by their employees) it was not easy to retrofit mildly variable length RISC-V instructions (2 bytes and 4 bytes) to the Aarch64 core they acquired from Nuvia. At the same time, Qualcomm proposed a new RISC-V extension that was pretty muc…

> Dropping "C" overnight and thus making all existing Linux software incompatible is completely out of the question. For general purpose Linux, I agree. But if someone makes Android devices and maintains that for RISC-V… that's basically a closed, malleable ecosystem where you can just say "f it, set this compiler option everywhere". But also, yes, another commenter pointed out C brings some power savings, which you'…

Qualcomm can do whatever they want with CPUs for Android. I don't care. They only have to convince Google.

But what they wanted to do was strip the "C" extension out of the RVA23 profile, which is (will be) used for Linux too, as a compatible successor to RVA22 and RVA20, both of which include the "C" extension.

If Qualcomm wants to sponsor a different, new, profile series ... RVQ23, say ... for Android then I don't have a problem with that. Or they can just go ahead and do it themselves, without RISC-V International involvement.

Re: Arm is canceling Qualcomm's chip design license

#440

Post by brucehoult »

eqvinox wrote: post #303
brucehoult wrote: post #14

Earlier quoted context omitted.

Qualcomm have a Plan B. This time last year they were all over the RISC-V mailing lists, trying to convince everyone to drop the "C" extension from RVA23 because (basically confirmed by their employees) it was not easy to retrofit mildly variable length RISC-V instructions (2 bytes and 4 bytes) to the Aarch64 core they acquired from Nuvia. At the same time, Qualcomm proposed a new RISC-V extension that was pretty muc…

> Dropping "C" overnight and thus making all existing Linux software incompatible is completely out of the question. For general purpose Linux, I agree. But if someone makes Android devices and maintains that for RISC-V… that's basically a closed, malleable ecosystem where you can just say "f it, set this compiler option everywhere". But also, yes, another commenter pointed out C brings some power savings, which you'…

Qualcomm can do whatever they want with CPUs for Android. I don't care. They only have to convince Google.

But what they wanted to do was strip the "C" extension out of the RVA23 profile, which is (will be) used for Linux too, as a compatible successor to RVA22 and RVA20, both of which include the "C" extension.

If Qualcomm wants to sponsor a different, new, profile series ... RVQ23, say ... for Android then I don't have a problem with that. Or they can just go ahead and do it themselves, without RISC-V International involvement.

Post reply on HN