I think Linus wrote a more coherent reply than I could but here are my thoughts anyway.

someone (someone.delete@this.somewhere.com) on September 27, 2013 8:25 am wrote:
> David Hess (davidwhess.delete@this.gmail.com) on September 27, 2013 8:06 am wrote:
> > ARM threatens Intel now having followed the same path Intel took through the low end to
> > threaten established microprocessors that held the workstation, minicomputer, and mainframe markets.
>
> The key differences are 1) Intel was a high volume commodity MPU ODM attacking a
> group of incompatible RISC architectures which had far lower volumes and higher
> fixed costs than Intel even if the MORPers were all combined instead of inefficiently
> duplicating each others efforts in chips, system, OS, and programming toolchain
> design; and 2) Intel's designs were comparable and occasionaly higher in CPU
> performance than even the best of the RISCs while being distinctly cheaper.

The lack of an established software development and hardware infrastructure to support their processors may have been an even larger disadvantage for the RISC players. I took Intel's attempt to sell the i860 as a coprocessor on i486 motherboards as an indication that they had the same problem with their own RISC offerings and were attempting to prime the pump of i860 development by making it more available on standard hardware. There was talk of a common hardware development platform for a simplified version of DEC's 21264 and AMD's Athlon since they shared the same EV6 local bus but I do not remember it coming to anything. I believe it was too late for Alpha by that time anyway. A more modern version is the on again and off again talk from Intel about sharing hardware between Itanium and server side x86.

ARM has the software development and hardware infrastructure problem behind them. They do have incoherence between devices causing segmentation but I expect a shake out with development of a more unified ecosystem which I believe I have seen Linus mention in another forum is happening or at least being worked on.

Previous to the Pentium Pro, Intel did not have higher performance and by the time they had the funds to produce that, it was too late for the RISC manufacturers to stay ahead of them despite their advantage of lower development cost.

> Although ARM based devices sell in far higher quantities than x86 chips do today
> the utter vast majority of those are extremely cheap devices for embedded control.
> Sales of ARM based application processors/SoCs for phones and tablets are still
> far smaller in revenue than x86 chip sales and a good portion of the profits they
> drive go to OEMs who make the phones and tablets, not ARM or the foundries that
> make the chips themselves. ARM products still can't come anywhere close to main
> stream x86 in performance. Low end x86 have been comparable to ARM in both
> power and performance since CT/CT+. With Bay Trail x86 now beats the vast majority
> of ARM chips in power, performance and Intel is increasingly aggressive on pricing.

Are sales of ARM based application processors and SoCs for phones and tablets smaller than revenue for just the x86 chips sales which are also used for these applications? ARM does not *have* to eat the whole elephant at once. If they can successfully compete with Intel at the low end of the processor market which has happened up until now, then they can attempt to incrementally move into the high end like Intel did to the RISC processor market. ARM is not depending on some great leap of faith. They are already established and growing.

Intel can leverage development and fabrication costs from their higher end processors but they still had to develop a separate processor line to compete with ARM and up until now I have not noticed a lot of success there. ARM can do the same in reverse with their lower end processors paying to develop the higher end ones like Intel originally did when it moved x86 into the RISC market.

I am sure Intel has the ability to produce an x86 based ARM killer but can they do that and *not* cannibalize the low end of their x86 processors intended for personal computers? Linus implies in his post that they may be past the politics of market segmentation but I have my doubts. If they deliberately limit the market for Atom and its descendants to prevent cannibalization, then they are effectively fighting ARM with one hand tied behind their back. ARM can then take them on piecemeal from the bottom.

Perhaps even worse, Intel has been somewhat at the mercy of Microsoft when trying to defend the low end of the personal computer space and early Atom netbooks were crippled in the name of market segmentation by both Intel and Microsoft. ARM so far has shown no such limitation that I know of but of course they have not yet made any serious challenges there so by the time they do, Intel and Microsoft may come to their senses. *bursts of laughter*

> The situation today is very much different from that late 80's and '90s'. Simply
> saying ARM will win because it is coming up from below just like x86 did to the
> server RISCs is grossly simplistic as well as almost certainly wrong. The money
> situation is very different and that determines the winner.

Up until this point, the ARM ecosystem *has* gotten enough money to compete successfully with Intel at the low end where Intel has been retreating do to disinterest or not present for fear of cannibalizing existing x86 markets. As ARM as a whole moves up the average selling price tree into markets previously held by x86, it will have more funds to produce better ARM implementations. That is what Intel did to the RISC processor market and at that time, Microsoft was not a loadstone around their neck.