Editorial Wintel Alliance Slowly Crumbling, ARM To Eventually Rule The Desktop?

[jihadjoe]

I don't really buy into the 'wintel' alliance. Microsoft never locked Windows into Intel only platforms.

Windows NT was available for Alpha and Sparc, Windows XP added x64 extensions even when it was primarily AMD's domain and Intel didn't have a 64bit x86 proc.

With ARM increasingly overlapping with netbooks and tablet PCs, it only makes sense for the next Windows version to support ARM.

 

[qubit]

I don't really buy into the 'wintel' alliance. Microsoft never locked Windows into Intel only platforms.

Windows NT was available for Alpha and Sparc, Windows XP added x64 extensions even when it was primarily AMD's domain and Intel didn't have a 64bit x86 proc.

With ARM increasingly overlapping with netbooks and tablet PCs, it only makes sense for the next Windows version to support ARM.

And yes, all those processor architectures have been discontinued, leaving x86/x64 as the only one. ARM's efficiency and performance could well be disruptive here and replace x86. Note what I said in the article, that if the advanced design and process technologies currently applied to x86 were to be applied to an ARM, it would blow x86 away. Take a look at that video, if you haven't seen it yet.

Finally, I'm talking in the time scale of a decade or so, as the whole world runs on x86.

 

[ensabrenoir]

just my 2 cents

Hardware development has far outpaced software development lately. Our cpus & gpus are way more powerful than what the average user will ever need. Were in the midst of a sidegrade so to say where the most important thing now is small form factor and low energy use. While i will still build and use POWER machines...the masses are headed toward potato powered dvd case sized desktops. There will always be business & engineering hardware but I believe there will be a radical change ahead. Imagine a dual 590, Sb-e powered system with the power usage/size of an ipad. Mobile when u need it, plug in a keyboard and 3 30inch monitors at home and game away. end rant

 

[Steevo]

For $16 I could do the same test but with a graphics accelerator since my Galaxy S 2 has one built in.


My phone is OK to browse the web, but not even in the same ballpark as any modern CPU.

And yes, all those processor architectures have been discontinued, leaving x86/x64 as the only one. ARM's efficiency and performance could well be disruptive here and replace x86. Note what I said in the article, that if the advanced design and process technologies currently applied to x86 were to be applied to an ARM, it would blow x86 away. Take a look at that video, if you haven't seen it yet.

Finally, I'm talking in the time scale of a decade or so, as the whole world runs on x86.

A decent amount of the world is moving to X64, even our new Insite and other highly proprietary software is supported exclusively on X64 now. Almost half lf my work machines are X64 or are running hardware that is ready for it now that our software is capable.

 

[qubit]

A decent amount of the world is moving to X64, even our new Insite and other highly proprietary software is supported exclusively on X64 now. Almost half lf my work machines are X64 or are running hardware that is ready for it now that our software is capable.

Yes of course. It was just cumbersome writing x86/x64 all the time. Those two are welded together, so it's just shorthand.

 

[dicobalt]

People here have made some good points. Intel can certainly compete on power efficiency so that old myth is busted. ARM has a huge uphill battle in desktops when it comes to interconnects though. The ARM platform itself is notoriously not standardized. Does ARM even have a PCIe chipset widely available? How do you get current manufacturers to jump ship from PC to ARM? What about drivers and software? Will they have to be custom coded for each ARM platform? That becomes unwieldy and you end up with PCs that are just like poorly updated cell phones thanks to unmanageable software.

 

[Easy Rhino]

since everything is going mobile, and intel is way out ahead of ARM in regards to mobile technology, i see this whole windows vs intel thing as a dog and pony show. windows will continue to buy chips from intel, now at a lower price.

 

[qubit]

People here have made some good points. Intel can certainly compete on power efficiency so that old myth is busted. ARM has a huge uphill battle in desktops when it comes to interconnects though. The ARM platform itself is notoriously not standardized. Does ARM even have a PCIe chipset widely available? How do you get current manufacturers to jump ship from PC to ARM? What about drivers and software? Will they have to be custom coded for each ARM platform? That becomes unwieldy and you end up with PCs that are just like poorly updated cell phones thanks to unmanageable software.

All very good points and something that the big boys will have to address properly if we're to see proper head-to-head competition. However, I don't think it's beyond them.

 

[dicobalt]

Note what I said in the article, that if the advanced design and process technologies currently applied to x86 were to be applied to an ARM, it would blow x86 away.

ARM will acquire that technology at some point but will it matter? Power usage can only go so low before it's not an advantage anymore? Think about it, which of the below would you choose?

1) ARM laptop/tablet that can run 24 hours under constant usage but not have x86 compatibility

2) x86 laptop/tablet that can run 8 hours under constant usage but can run all that long standing and mature desktop software and not have to deal with waiting for an OEM to update your laptop/tablet OS just like cell phone and tablet manufacturers do today.

Of course I also happen to think that Intel will be pushing out faster CPUs and GPUs than ARM systems can provide. ARM will not be able to keep up with the relentless consistent tick tock pace that Intel always delivers on.

 

[qubit]

ARM will acquire that technology at some point but will it matter? Power usage can only go so low before it's not an advantage anymore? Think about it, which of the below would you choose?

1) ARM laptop/tablet that can run 24 hours under constant usage but not have x86 compatibility

2) x86 laptop/tablet that can run 8 hours under constant usage but can run all that long standing and mature desktop software and not have to deal with waiting for an OEM to update your laptop/tablet OS just like cell phone and tablet manufacturers do today.

Of course I also happen to think that Intel will be pushing out faster CPUs and GPUs than ARM systems can provide. ARM will not be able to keep up with the relentless consistent tick tock pace that Intel always delivers on.

Yes, point 2, obviously. That's why I said the likes of IBM would have to get behind it. Bridging the gap between legacy x86 apps and ARM is going to be vital too, as I've said. I couldn't agree more that they have a mountain to climb to oust the entrenched x86 architecture, but these are not small players and together, they might just be able to climb that mountain. Perhaps they could somehow even get Intel on-side (unlikely) but you never know what deals can come about with these things.

The more I think about it, the more convinced I am that the only reason we're stuck with x86 today is because of that sodding exclusivity clause Intel made AMD sign 30 years ago. It's not clear to me why there's a couple of other companies with x86 licences too, such as VIA. I guess a quick google/wikipedia will answer that one, lol.

 

[dicobalt]

I couldn't agree more that they have a mountain to climb to oust the entrenched x86 architecture

Why would those companies bother to climb that mountain? Investing all that time, money, and manpower into solving a problem that has already been solved for them with x86 systems. In the boardroom I can only see that as being a bad investment decision in an economy where they need to squeeze every last penny of profit out of their income. As the open source people say "It's a duplication of effort."

 

[qubit]

Why would those companies bother to climb that mountain? Investing all that time, money, and manpower into solving a problem that has already been solved for them with x86 systems. In the boardroom I can only see that as being a bad investment decision in an economy where they need to squeeze every last penny of profit out of their income. As the open source people say "It's a duplication of effort."

The evidence that they'd bother is there with the fact that Microsoft has written Windows 8 to run natively on ARM and a dual/multi CPU architecture market is unstable, since they have to maintain two development teams at all times to effectively service one product - there's your duplication of effort.

 

[newtekie1]

Sorry but it's simple as that: they can't. ARM is superior on very base level + dedicated processing units spread across SoC make the case even worse for x86.

"The Cortex-A9 has a TDP of 1.9w @2Ghz, the Atom Z550@2GHz is only at 2.5w."
This means that atom would be trashed by ARM performace wise, GPU is already way faster, not to talk about rest of SoC, oh, yeah, SoC, no slow buses etc.

The Cortex-A9 doesn't have a GPU, pay attention here.

But again, right now Sandybridge is absolutely destroying ARM in performance/watt. So taking that same architecture, and crippling it to the point that it performs like an old 486 to match the ARM performance should give power consumption the same or better than ARM.

Like I said, Sandybridge is already down to 17w, that is with two full cores and Hyperthreading, an integrated Northbridge, an integrated GPU, and integrated memory controller, integrated PCI-E lanes, and a huge L3 cache. A lot of stuff that isn't needed in a ARM competitor.

 

[Fourstaff]

The Cortex-A9 doesn't have a GPU, pay attention here.

But again, right now Sandybridge is absolutely destroying ARM in performance/watt. So taking that same architecture, and crippling it to the point that it performs like an old 486 to match the ARM performance should give power consumption the same or better than ARM.

Cortex-A9 doesnt have a GPU, but most of the people who use the Cortex-A9 also add a GPU component on die/on package, so its generally counted "as one".

S-B (and most of the x86 processors out there) destroys ARM only in FLOPs, and a few other benchmarks. Otherwise (watching videos, surfing the net etc) ARM is not bad, and can sometimes tilt the advantage over. I wouldn't call one better than the other, its too task dependent.

 

[Wile E]

All of you are getting your panties in a twist over tech specs.

Guess what - it doesn't matter.

I can Skype, Facebook, send emails and play Angry Birds with a small fraction of the power already available today.

The only relevant 'tech specs' are that Intel is currently incapable of competing with ARM on the all-important idle power consumption. If they are able to address that then they can enter the smartphone race (albeit very late to the party). Intel is currently working very hard at this.

And to Wile E, people use Wintel because that's what the market determined was popular from the 1980s. They didn't choose it because of any tech related specification or power consumption.

Dorks bought a 386sx33mhz w/ 2mb of ram, a 80mb HDD and a Trident video card sitting on an ISA bus. People bought IBM-compatible PCs.

That has no bearing on why it is being used today. Today it is simply the better chip for desktop usage.

Sorry, x86 is here to stay for a long, long time to come.

 

[dicobalt]

The evidence that they'd bother is there with the fact that Microsoft has written Windows 8 to run natively on ARM and a dual/multi CPU architecture market is unstable, since they have to maintain two development teams at all times to effectively service one product - there's your duplication of effort.

Yes indeed Microsoft is willing to make the effort but this will require all major OEMs and chipset manufacturers to also step in and do a whole lot of work on the hardware and drivers too. Microsoft can't do it alone!

 

[newtekie1]

Cortex-A9 doesnt have a GPU, but most of the people who use the Cortex-A9 also add a GPU component on die/on package, so its generally counted "as one".

S-B (and most of the x86 processors out there) destroys ARM only in FLOPs, and a few other benchmarks. Otherwise (watching videos, surfing the net etc) ARM is not bad, and can sometimes tilt the advantage over. I wouldn't call one better than the other, its too task dependent.

S-B generally outperforms ARM in every aspect. ARM is capable of doing low end stuff, surfing the net and watching videos, but SB still is better at those things. To effectively move into the desktop market, ARM is going to have to be alot better than just "good enough to watch videos and surf the net", especially with it being noticeably slower at even managing to surf the web than an Atom, and Atoms are slow pieces of shit in the desktop world. Most business won't even bother to look at them for machines in their office.

 

[Fourstaff]

S-B generally outperforms ARM in every aspect. ARM is capable of doing low end stuff, surfing the net and watching videos, but SB still is better at those things. To effectively move into the desktop market, ARM is going to have to be alot better than just "good enough to watch videos and surf the net", especially with it being noticeably slower at even managing to surf the web than an Atom, and Atoms are slow pieces of shit in the desktop world. Most business won't even bother to look at them for machines in their office.

I don't know about that, but at least in practice I hear the laymen starting to acquire more iDevices and Tablets over Atom netbooks for general day to day internet and entertainment consumption. That is, to me at least, a clear indicator that Intel should be stepping up their efforts if they want to compete effectively against ARM. It wasn't the case last year, and with me living in a clearly more geeky environment (there are people still use Serial ports, Fortran and CUDA rather than newer alternatives), I think that is a sign that times are changing.

 

[bostonbuddy]

Intel must be planning to have 22nm Ivybridge based smartphone/tablet x86 cpu's. Only makes sense. They can't wait 2 years for the next die shrink to start competing in that market.

 

[MadMan007]

This should be an editorial, there is a ton of opinion in this piece.

 

[Damn_Smooth]

This should be an editorial, there is a ton of opinion in this piece.

I can't believe we made it this far before somebody bitched.

 

[Frick]

I can't believe we made it this far before somebody bitched.

Yeah, i'm surprised i didn't.

 

[Wile E]

I don't visit the front page. I just assumed it was an editorial. It's blatantly obvious that this isn't a news story.

 

[qubit]

I don't visit the front page. I just assumed it was an editorial. It's blatantly obvious that this isn't a news story.

You know, this time I agree with you. It started out as a news story (see sources) but as I wrote it up, it developed into an editorial.

Note that editorials shouldn't be seen as any less than a straight news post though, as seems to often be implied in these sorts of discussions. Think about it, with a news post, you're basically just repeating something you've seen somewhere, perhaps gathering info from several sources. With an editorial, the news is analyzed and conclusions drawn from it, as I've done here. Also, to make life more interesting, the line between the two is often blurred.

I've clicked that button.

 

[Wile E]

Never said editorials were bad. Only editorials that are intentionally passed off as news are viewed in a negative light by me. That's obviously not the case here.