Hacker News Reader

Comment by cameldrv | Hacker News Reader

If there were ever a time where Intel had its customers by the balls the least in the past twenty years, it's now. AMD is extremely competitive in the server and desktop space, ARM/Apple is becoming competitive in the laptop space, Intel has no product in the phone/tablet space, and they are noncompetitive in the HPC/DL accelerator space.

If there's anything that has typified Intel in the past decade it's a generalized failure to execute. There's been very little innovation in the core product since Haswell in 2013.

View on HN

Yeah for anyone who was a computer enthusiast in the 90s and 00s the rise of ARM has been an astonishing development comparable to the ubiquity of Linux and decline of Windows.

> ubiquity of Linux and decline of Windows

Should I read this as sarcasm?

>> ubiquity of Linux and decline of Windows

> Should I read this as sarcasm?

It could only be sarcastic if he's referring to "Linux on the desktop." I don't have any numbers, but I wouldn't be surprised if he was correct about servers.

Correct. I am referring to everything except desktop computing. I can see how this would be confusing as I was just referencing desktop prossesor market patterns of the 90s/00s.

But I do think Windows's position in the desktop space is somewhat unsustainable at this point. I personally prefer Windows but I just don't see how a heterogenous user interface experience (with users on iOS and Android for mobile and Windows for desktop) will abide for long.

I think it's only a matter of time before mobile OSes make some serious inroads to desktop. Though I suppose Google has been salivating at that thought for neaely a decade.

If you look at the overall "computing devices with a reasonably high performance CPU" market, Linux has won the server market, and through Android, won the phone/tablet market. If you squint a bit, you can lump MacOS and iOS in this in that they have a Unix core and use a lot of the same software stack as Linux.

50 years ago, IBM owned the computing market with mainframes. They still dominate the mainframe market, that's just become a smaller and smaller portion of the overall computing market as time has gone on.

In both the server and especially mobile space Windows has lost its mojo and Linux has been a beneficiary.

Linux is ubiquitous. It runs on embedded devices, cell phones, and tablets which surround us. Linux runs most of the computers which underpin websites and services we interact with daily, and it is the only serious choice for HPC. A few people even run it on their laptops.

Windows has declined. It is still used extensively, but Microsoft no longer dictates the browser market. Microsoft Office is commonplace, but has genuine marketshare competition in alternatives like Google Docs. Developers increasingly build "desktop" applications as web-based or multiplatform-first instead of Windows-exclusive, with the notable exception of AAA games. The Linux subsystem for Windows is a direct admission by Microsoft that developers want to work with Linux-based software for many tasks.

I think he interpreted the statement as meaning linux displacing Windows on desktop, which isn't happening. Phones are displacing Windows on desktop.

> Intel has no product in the phone/tablet space

To be pedantic, Intel does have phone and tablet processors. This doesn't invalidate your point however. I can't easily find a breakdown of mobile cpu market share but I am guessing Intel isn't doing very well.

https://www.intel.in/content/www/in/en/products/devices-syst...

https://www.intel.com/content/www/us/en/products/devices-sys...

They canceled the atom for phones product. Right around the time that Microsoft announced the Continuum feature of Windows Phone 10 that gave a desktop(ish) experience when docked, and would have been perfect with an x86 phone. :(

Then its only a matter of time before they go bankrupt.

They have the cash to get out of this if they make the right moves.

Exactly. A single well-time and well-executed acquisition could change the whole game. Of course we all know that executing acquisitions well is easier said than done, and I don't have a specific target in mind when I say this. It's more of an "in principle" speculation than a concrete suggestion at the moment.

Softbank are looking to sell some or all of their ARM holdings.

Very unlikely. Just because they are no longer "without alternatives" in many use-cases (for business customers) doesn't mean they go bankrupt. They will just lose money and maybe shrink a bit.

Also thy are in many other areas then just CPU's including FPGA and hard drives.

Reply on news.ycombinator.com