Linus Torvalds says ARM just doesn't look like beating Intel

[Karel]

Linus Torvalds believes ARM has little chance of overhauling x86, because the latter has built an open hardware ecosystem that the former just doesn't look like replicating.

http://www.theregister.co.uk/2016/10/10/linus_torvalds_says_arm_just_doesnt_look_like_beating_intel/

[ataradov]

He is not blaming ARM as architecture, but rather specific implementations. ARM has performance issues as well, but that's not relevant here.

And what open hardware we are talking about with all that proprietary stuff (UEIFI) that has been added recently?

X86 is more uniform, since there is only one vendor and a number of on-chip peripherals is limited and more or less standard.

This can be easily rectified by ARM if they see it as an actual problem. They did it once for MCUs by introducing CMSIS.

[G7PSK]

Intel have all but given up on mobile platforms.

[ataradov]

Intel have all but given up on mobile platforms.

Yep. And also, what Linus is really saying is "my life would be so much easier if there was only one ARM device or all of them were exactly the same" Understandable, but not going to happen.

[rob77]

Intel have all but given up on mobile platforms.

unfortunately...

R.I.P. Moorefield  R.I.P.  SoFIA     

[timb]

Intel have all but given up on mobile platforms.

Yep. And also, what Linus is really saying is "my life would be so much easier if there was only one ARM device or all of them were exactly the same" Understandable, but not going to happen.

Yeah, he got tired of the influx of kernel patches and drivers that he had to slog through, each time a new ARM board appeared. This is why we have the absolutely horrible "Device Tree Overlay" system now.

He basically forced a few developers to quickly hack together something. DTO is very much not ideal.

[dannyf]

“It does not have the same unified models around the instruction set as we do in the PC space, but it is getting better.”

He's failing to understand the business aspect of ARM vs. Intel. ARM's core competency is in the microcontroller mkt by a long margin. It may move to the processor (desktop + server) markets but that progression will take time / resource, and they have to consider the potential of Intel using a microcontroller offensive as a defensive move.

What's why Linus is a techie and not a business visionary.

[Macbeth]

What's why Linus is a techie and not a business visionary.

But he makes a killing with his Tech Tips on youtube! (I kid, I kid )

[rob77]

ARM will never ever replace intel x86 arch... it simply won't happen. what is already happening is that certain activities like reading mails and browsing the web are moved from desktops to mobile devices where ARM is prevalent (and since Intel ditched their mobile platforms ARM is the one and only)... but ARM will never take over the desktops. it simply won't happen because of the software. if there is no windows and office for ARM, then it means there will be no ARM for mainstream desktop. for server market it's the very same... some services might/will move to specialized ARM platform (e.g. WEB hosting could be moved to a massive multi-core ARM platform). but in general... as long as there is no windows servers, ms sql, oracle , SAP and many others  for ARM... there will be no significant market share for ARM.

[ataradov]

as long as there is no windows servers, ms sql, oracle , SAP and many others  for ARM... there will be no significant market share for ARM.

But that's actually changing with AArch64 and ARMv8.

[rob77]

as long as there is no windows servers, ms sql, oracle , SAP and many others  for ARM... there will be no significant market share for ARM.

But that's actually changing with AArch64 and ARMv8.

server market is about business, business is about profit... porting existing applications to a new platform is a unnecessary investment which recrudesces profit => won't happen.
sorry , today's world is all messed up... tech companies are no tech anymore... they are led by businessman without the slightest clue about tech stuff... nowadays it's all about more and more profit, buzz words.. and stupid decisions talking about the giants in the industry... and the small companies with new fresh blood will be sooner or later swallowed by the giants because the giants can't innovate so they're buying the small ones in hope if buying innovation.. but right after buying them they kill them by stupid decisions... and that's the reason why there will be no existing software ported to any new platform any soon.

[dannyf]

but ARM will never take over the desktops.

that's a highly static view of the world today - which is unimportant for this discussion.

ARM has demo'd server chips and there are ARM-based server boards out there. Having lived with Xeon for the last 10 or so years, I am more and more convinced that the future of (personal) computing is in those multi-core CPUs where individual cores are reasonably but not terribly fast - so they provide adequate performance on older applications. Such CPUs can take on dynamic load situations - sudden onslaught of multiple tasks running at the same time. This approach provides robustness of performance, unlike single core CPUs or CPUs of low core counts.

So if you project 10 - 15 or 20 years down the road, it wouldn't surprise me if ARM or ARM-like architectures become a mainstay on the desktop / server market.

[rrinker]

Intel have all but given up on mobile platforms.

Yep. And also, what Linus is really saying is "my life would be so much easier if there was only one ARM device or all of them were exactly the same" Understandable, but not going to happen.

 Right, and my choice of OS would be much easier if there was only 1 Linux, too.....

[EEVblog]

Bugger the ARM/Intel thing:

Linux Lord also feels Internet of Things hardware is mostly doomed, like his old Sinclair QL

[nctnico]

Just some random thoughts.
- For starters I'm not convinced Linus is right about anything because he says so and sometimes I get the feeling he should leave because he is holding progress back.

- What is open about the PC platform? For starters try and get open source video drivers for decent graphic cards.

- I also don't see why ARM should move to the desktop and why the power hungry x86 architecture should move into embedded devices.

- What is the problem anyway? Neither ARM or x86 are going to dissapear in the next 20 years because both are bound by the same limitations which come with making chips.

[ataradov]

- What is the problem anyway?

He was on this subject for a very long time, that's why we have Device Tree. He said that again, so "media" picked it up again, since they have nothing useful do to.

[linux-works]

Intel have all but given up on mobile platforms.

quite the opposite.  they are laying off their pc people and desktop chip people in an attempt to 'follow' the IoT idiots.

intel will fail at it, as they seriously do not understand what it takes to play well in the maker and/or iot market.  all their attempts are overpriced, overdone and not at all iot or maker friendly.

but BK (the ceo, as he's known inside) still thinks that there's a chance for intel in the small systems, mobile and iot market.

[linux-works]

we are seeing ARM take over the browser market since tablets and phones are so common and new purchases are mostly for phones, not pc's anymore.

the x86 arch is going to be put into servers and mostly left there.  MS is seeing this and that's why their whole push for win10 is 'software as a service' where the end system does not matter (its a thin client; what's old is new again!) and the software all runs in data centers (the 'cloud').  if that's the case, ARM can render html and that crappy junk called ecmascript (lol) just fine.  no need for x86 at home anymore once MS has killed off local windows installs.

[raptor1956]

Intel spent too much of there efforts up through about 5 years ago trying to kill off AMD.  They pretty much succeeded at that but in the process they did not pay enough attention to the mobile market.  But, and it's a huge but, Intel would never be able to get anything close to the margins they make in the PC space unless they were able to do unto ARM as they did to AMD.  They are late to that game as they were too focused in fighting the last war.


Brian

[Dago]

I wouldn't be surprised if the answer to the "what will happen to the desktop market" -question will be "nothing" for the foreseeable future... Development has almost completely stalled (I am running a 2009 or something Xeon I paid 30 € for, for something like 250 € I could get MAYBE twice as fast new CPU, that is nothing in terms of development in seven years) and desktop and laptop computer sales have dropped a LOT. I don't really see why anyone would be interested in entering or even changing this market. Intel is pretty much the only one who has a stake and an interest in this market but I agree with previous posters that they are not really trying, just repackaging their existing x86 Atom stuff and trying to pass that off, but people are not really buying it because it is a poor compromise in terms of speed/price/power use/technology.

[Red Squirrel]

I would love to see an open computing platform though, Intel is closed, and the latest revelation about them putting spy tech built into the chip is quite scary.  An open source hardware - at the semiconductor level, would be quite something.  That would be a HUGE feat though.  Especially if all the parties involved could collaborate and stick to a certain set of standards.  (good luck with that though - look at how fragmented Linux is, and that's just software, I'm talking about hardware too!)

[ataradov]

An open source hardware - at the semiconductor level, would be quite something.

That's what RISC V is. And there is actual silicon being made. But obviously they can't move as fast as Intel does.

Here are some recent news on the MCU side of it - http://hackaday.com/2016/10/10/the-journey-toward-a-completely-open-microcontroller/

There are also MPU implementations out there, but nothing accessible to regular consumers.

[StuUK]

as long as there is no windows servers, ms sql, oracle , SAP and many others  for ARM... there will be no significant market share for ARM.

But that's actually changing with AArch64 and ARMv8.

server market is about business, business is about profit... porting existing applications to a new platform is a unnecessary investment which recrudesces profit => won't happen.
sorry , today's world is all messed up... tech companies are no tech anymore... they are led by businessman without the slightest clue about tech stuff... nowadays it's all about more and more profit, buzz words.. and stupid decisions talking about the giants in the industry... and the small companies with new fresh blood will be sooner or later swallowed by the giants because the giants can't innovate so they're buying the small ones in hope if buying innovation.. but right after buying them they kill them by stupid decisions... and that's the reason why there will be no existing software ported to any new platform any soon.

What makes you think they are stupid decisions? Many companies have been bought to kill the disruption to their core market rather than foster them, that's business for ya....

[DimitriP]

I am more and more convinced that the future of (personal) computing is in those multi-core CPUs where individual cores are reasonably but not terribly fast - so they provide adequate performance on older applications.

I'l take 4 cores at 4+ GHz and let everyone else with 16 cores @ 2.6 GHz gloat about their benchmark results

[rob77]

as long as there is no windows servers, ms sql, oracle , SAP and many others  for ARM... there will be no significant market share for ARM.

But that's actually changing with AArch64 and ARMv8.

server market is about business, business is about profit... porting existing applications to a new platform is a unnecessary investment which recrudesces profit => won't happen.
sorry , today's world is all messed up... tech companies are no tech anymore... they are led by businessman without the slightest clue about tech stuff... nowadays it's all about more and more profit, buzz words.. and stupid decisions talking about the giants in the industry... and the small companies with new fresh blood will be sooner or later swallowed by the giants because the giants can't innovate so they're buying the small ones in hope if buying innovation.. but right after buying them they kill them by stupid decisions... and that's the reason why there will be no existing software ported to any new platform any soon.

What makes you think they are stupid decisions? Many companies have been bought to kill the disruption to their core market rather than foster them, that's business for ya....

decision to "streamline" the workforce in the newly bought company and integrate the new-bought into the rotten structures of the giant is quite stupid and equals to killing the bought innovation and that happens most of the time, i would say always