Hacker News Reader

Comment by neilc | Hacker News Reader

Is ARM's power advantage really that significant for devices larger than the smart phone / tablet form factor? If an Atom-based CPU consumes ~2-3 extra watts but offers marginally better performance and (more importantly) compatibility with an enormous base of existing applications, that doesn't seem like a very compelling argument for switching.

View on HN

Think of a data center situation though where you can get a ~50% power requirement difference for marginal performance decreases. The cost savings would be huge even if you require more servers to make up for the performance difference.

For most home users I think you're right for now but for businesses it could be huge.

Of course, you've got to remember that which chip has the best performance per watt depends on the job. A sufficiently smart data center could determine this automatically and take that into account in its job scheduling.

> Think of a data center situation though where you can get a ~50% power requirement difference for marginal performance decreases.

50% lower processor power isn't the same as 50% less system power.

Is ARM's power advantage really that significant for devices larger than the smart phone / tablet form factor?

I'll bet $5 that the MacBook Air in the future (18-24 months) will switch to ARM once the >2GHz ARM processors start shipping en masse.

The Asus Eee PCs running Android with Snapdragon processors already embed the Cortex-A9 MPCores.

"4GB RAM ought to be enough for anybody."

ARMS's Achilles' heel is that it's only 32-bit, and I don't think anyone wants to go back to a segmented address space any time soon.

That's a great point. Are there any 64 bit ARM chips on the horizon? I don't know much about chips, so maybe it's not even possible given the architecture.

I don't recall hearing any announced, though I wouldn't be surprised at all if it were currently in the works. The A15 mentioned in the article, however, I believe has something in the spirit of x86's PAE that offers an expanded (36 bits or so?) physical address space, for better or for worse (better in that you can address more memory, worse in that it's kind of an ugly hack and not a great long-term solution).

It's possible, but just like the x86-64 transition it won't be pretty.

It might go better for ARM given that most of the more tightly controlled software that is delivered on ARMs (app store or custom embedded).

I'm curious; have you faced any problems in that transition?

(seriously, I'm not trying to troll. I've had exactly one - making sure a project of mine would compile as PIC and that was pretty minor after the required reading)

There haven't been any major issues, it's just been glacially slow. Most software running on capable CPUs is still 32 bit, Visual Studio still doesn't have first class support for 64 bit and my MacBook still runs a 32 bit kernel.

I expect ARM's transition to be quite different though--with the highest ends of their business moving in only a couple product cycles, and the lower ends of their business sticking with 32 bit indefinitely.

There were a lot of stories in the 2000 time frame, but nothing seems to have been actually done.

One of the article's contentions is that the vast majority of users are satisfied with Windows and Microsoft programs as they are right now. Most of those programs are 32-bit.

By extension, if they are satisfied with 32-bit programs and don't feel a need to switch, then they won't miss 64-bit architectures.

Most Windows sold today is 64-bit, and it allows you to run more 32-bit programs without running out of memory.

Most people could care less about more applications...

Virtually all of the computer users that I know limit their use to an office application and a web browser on top of Windows. That's it. The web browser gives them everything they really need. I wonder how many computer owners could even tell you if they have 32-bit or 64-bit or how the two differ. Probably the same percentage that can tell you detailed specs about their car engines. Virtually none.

The only computer specs that matter anymore --exempting programmers-- are weight and battery life. A few hardcore gamers still care, but they're in the minority, even for video gamers. XBox anyone?

You think Apple is going to switch to an architecture that breaks VMWare?

Apple have demonstrated time and again that they are happy to break existing functionality if it suits their aims.

And just as important--Apple's customers have demonstrated time and again that they will follow wherever Apple leads.

What's your best 2 examples?

OSX and Intel processors...

OS X supported OS9 apps until long after people were puking at the mere sight of OS9 apps.

And when they converted to Intel, they also provided Rosetta, a completely ridiculous binary translation system to make PPC apps work.

Those are bad examples.

The most striking for me when Jobs announced move to Intel was the bit where he told that they had OS X builds for Intel since the very first version of OS X. Who knows what are they cooking in their secret kitchens.

NeXTStep was ported to x86 back in 1993, years before it was acquired and rebranded as Mac OS X. I would have been surprised if they had neglected to maintain an asset like that.

Motorola 68k -> Apple/Motorola/IBM PowerPC -> Intel ???

Prediction recorded: http://predictionbook.com/predictions/1861

Hope you're still around in 2 years; it can be interesting to ponder a falsified prediction and wonder, why did I believe that?

Except that the Atoms at same (and even a bit higher) clock don't even offer marginally better performance.

And another thing that I came to think of that weighs heavily in ARM's favor is the multitude of composite, multifunctional chips, SoCs etc. currently available in the ARM flavor. Is there ANYTHING like this for Atom, or any x86 at all?

Yes, AMD's Fusion line (http://en.wikipedia.org/wiki/AMD_Fusion) aims to provide SoC functionality.

Except that the Atoms at same (and even a bit higher) clock don't even offer marginally better performance.

Depends if we are talking Arm Cortex-A8 or Cortex-A9. Atom is much more powerful at same clock speed that Cortex-A8. Cortex-A9 is getting closer competitive, but doesn't have much of a power advantage over it. (Lies, damn lies and benchmarks, but A8 gets ~2000 MIPS, A9 ~2500 MIPS and singlecore 1st gen Atom gets ~3300 MIPS)

Also, multicore-Atom has been in the market for a while. Multicore ARM is just coming online now.

It's worth noting that Intel can speed up Atom whenever they want. There are plenty of easy wins in there (eg, better execution core, faster memory controller) that they can put in if they need to keep their lead.

(Lies, damn lies and benchmarks, but A8 gets ~2000 MIPS, A9 ~2500 MIPS and singlecore 1st gen Atom gets ~3300 MIPS)

Measured how, and at what frequencies? Also ARMs are really starved for bandwidth. Has that improved with the Cortex A9?

Reply on news.ycombinator.com