Live data from Hacker News

Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

bloomberg.com

221–230 of 239 posts

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#221

Has it been long enough for people to forget NT on Alpha, MIPS and x86 (and i860, though not released)? And to forget both PReP ( https://en.wikipedia.org/wiki/PowerPC_Reference_Platform ) and CHRP ( https://en.wikipedia.org/wiki/Common_Hardware_Reference_Plat... ) which were going to break the Intel stranglehold once and for all?

Imagine having a time machine, going back to Microsoft in the 90's and telling them to forget SGI, IBM, and HP, port NT to ARM...

They'd laugh at you since at that time ARM was something that ran in either small-format low power low-clock devices (like a Newton) or in fringe machines like the Acorn Archimedes.

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#222

Earlier quoted context omitted.

linux as a primary gaming platform works just fine for me, thanks to valve. Confirmed by a happy occasional CS:GO player here.

I wish it worked on Debian, it used to but the last time I tried I lost a day and still didn't get it to run...

Its works fine in Ubuntu.

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#223

Earlier quoted context omitted.

Time and time? We do see them in Linux. The ARM server ecosystem has been in the works for a few years. TechCon'16 we saw the announcement that the software is fully there: and this is not hype, all major distributions support all the major players out of the box. Download ISO, install, and use it. HW-wise we are now in the 3rd gen of chips (QC Centriq, XGene3, ThunderX2) all of which compare with E5, not Atoms. Firs…

So if my code is parallel enough to take advantage of all the cores, like a web server serving 96 requests simultaneously, one per core, would I get better performance than Intel? Better performance per watt?

You would run 96 processes one for each core. Performance more depends on IO that CPU in reality.

What ARM will bring is competitive market and no Intel Tax. Average 1U in ARM will be massively cheaper. For servers OpenPOWER is more interesting[1]

[1]https://www.theregister.co.uk/2016/04/07/open_power_summit_p...

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#224

Earlier quoted context omitted.

Google have been doing something with POWER CPUs for a few years now, at least on an experimental basis. 2016, Power 9: http://www.computerworld.com/article/3052951/cloud-computing... 2014, Power 8: http://www.pcworld.com/article/2149080/google-shows-homegrow...

Each time Google looks into POWER, it seems to be when it's time to renegotiate with Intel. This is a tried and true tactic for better Intel pricing at this point. I was even talking about it (and got somewhat of a corroboration) a few days ago[1]. 1: https://news.ycombinator.com/item?id=13786170

That's rather sad. POWER has seemed to be lingering in this half-life for many years now, where it's valuable enough for development to continue, but it's not competitive against Intel for broad use.

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#225

Earlier quoted context omitted.

linux as a primary gaming platform works just fine for me, thanks to valve. Confirmed by a happy occasional CS:GO player here.

I wish it worked on Debian, it used to but the last time I tried I lost a day and still didn't get it to run...

I've been running it on Debian testing here for nearly two years now, no major problems. Since jessie it's just an apt-get away.

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#226

Earlier quoted context omitted.

To cut a long story short: Yes, most applications can just be ported. The number of applications that rely on the actual clock rate being steady is extremely close to zero. Bear in mind nearly everything runs fine on laptop machines, which generally have a much wider range of clock states, and have no problems. Similarly, each core in modern machines is likely to be running at a different speed, and again, no problem…

I'm struck by how hard it would be to track down a bug that depends on the consistency of the clockrate...

Bugs like the ones you describe come down to timing.

But it's not like "the old days" where for a specific target, there was only 1 chip/clockspeed so devs simply relied on it for timing since an RTC was or not available, or too slow to access.

These days, it would be very hard to actually write code that relies on a specific clock-rate and work reliably. It's a lot easier and reliable to use the clock for time-sensitive stuff.

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#227

Earlier quoted context omitted.

Perhaps spurred by Google's recent announcement that they have Skylake chips in their cloud: https://cloudplatform.googleblog.com/2017/02/Google-Cloud-Pl...

Sure. And also, may be the Google-Inte Skylake comes from past year, when Google was testing POWER9 chips... [1] [1] https://www.theregister.co.uk/2016/04/07/open_power_summit_p...

There's more recent news on that front: https://blog.rackspace.com/the-latest-zaius-barreleye-g2-ope...

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#228
post #117
post #91

Earlier quoted context omitted.

a ton of major issues have been recently fixed. It doesn't seem to be on the back burner at all, it's just that building/updating an entire OS is a bit different from building/updating just steam. Also, a lot of fixes are edge cases that most people don't view as fixes, because it was never an issue for them. Most of the issue revolves around GPU drivers, which in theory valve is not responsible for, but they have hi…

> Even then, most devs don't optimize for OpenGL. That's again on the dev, and not valve, Sony, Nintendo, Microsoft, specially Microsoft, go the extra mile to make sure games perform well on their platforms, even special casing APIs for specific games.

they also push those platforms as their only game platform. and try to discourage devs from using other platforms.

Valve isnt trying to discourage devs from using windows. Windows is still their main income source. Linux only devs aren't going to make valve money during the transition to windows. The goal is to build the game library on linux so it is a valid competitor.

Also, they can't really make a ton of special OS level APIs, since they are trying to keep things generic. they dont want to fork Linux, they want to maintain compatibility. I mean, open is the entire premise of leaving windows. And if you mean higher level APIs, there's things like: https://valvesoftware.github.io/steam-audio/ one of the top things in porting a game, is that directX covers more than just graphics. opengl + SDL2 covers most of it. steam audio makes it so you can write code once and have it work everywhere.

So yeah, they are working on making APIs that make it easier for devs to release linux/SteamOS games without needing the player base already there, because linux "just works" with the middleware.

The short of it is: they are actively doing a lot of stuff to advance linux gaming still, just a lot of it is behind the scenes now and aimed at devs or not directly steam related (you dont see the fact that they hire mesa devs through steam updates).

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#229

Earlier quoted context omitted.

Not true. When multiarch mattered (NT 3 and 4), Microsoft was the only vendor that delivered the exact same OS, device support and development environment across x86, Alpha, MIPS and PowerPC, and pushed the industry towards standardization. Microsoft has always taken arch and platform independence seriously, and this was evident even in 2010 when I worked on NT. You could rebuild the whole system for any supported ar…

Great points. It is an odd fact that Microsoft generally avoided writing operating systems on or for Intel x86 processors [1], and started the development of both NT and CE on Risc processors. Meanwhile, Linus had a PC with an Intel 386 processor, so that's what he started Linux on and for... [1] MS DOS was based on code that Microsoft bought in, not having time to develop it from scratch. However, Microsoft did have…

Not that odd. The NT group specifically wanted a portable design, and they made the right call to initially target i860, MIPS and only then i386 machines. Contrast that to OS/2 development - the (never shipped) PowerPC OS/2 port was based on top of of an IBM fork of Mach, because the kernel was too x86 specific (derived from the pre-virtual memory 16-bit protected mode code).

Re: Microsoft Pledges to Use ARM Server Chips, Threatening Intel's Dominance

#230
post #96

Earlier quoted context omitted.

Dell did this for ages with AMD chips. Same time every year you'd see an article go across all the magazines about Dell planning to introduce AMD chips to their servers, and just as quickly vanish once Dell had finished negotiating with Intel.

They did more than plan, I have a rack full of older Dell servers with AMD CPUs.

When I say "did this for ages", I'm reaching back to the turn of the millennium. It started happening about as soon as AMD brought their first competitive chip to the market.

Dell did eventually start shipping servers with AMD processors in them, but it was long after they first started saying "We may start selling AMD based servers."

Post reply on HN