May 212011
 

Only a month or two after it was published, a detailed report that I wrote was wiped out during a BrightSideOfNews* hard drive crash. That exhaustive report, praised by many throughout the industry as the finest of its kind yet produced, examined the emerging and inevitable ARM versus x86 clash.

It took a little while and cost BSN* a lot of money to recover the data on the hard drive, but that report is now back up and can be read here.

I’m currently working on a followup to that bit of analysis that will include even more hardware than the initial report.  I’m still waiting on a vendor or two, so I can’t promise an ETA yet, but one thing I can state is that the new report will be very interesting.

The computing landscape is changing rapidly and the war between x86 and ARM microprocessors is now underway.  The competitors have dramatically different strengths and weakness, making for a particularly exciting confrontation.

Most importantly, the results of this war will have profound effects well beyond the CPU market, where several companies will possibly see their fortunes upended.  One thing is absolutely certain: computing will never be the same again.

Dec 212010
 

After many months of trying to wring something out of NVIDIA, I have finally obtained a Tegra 2-based device.  It is in the form of the ViewSonic G Tablet, a 10″ Android 2.2 (Froyo) based slate computer.  We bought it from Sears, of all places.  Oddly enough, Sears has one of the largest selections of tablet devices you can find.

Despite its complete lack of refinement, this thing is awesome, but only if you don’t mind wiping out the stock ROM.  The G Tablet’s shipping GUI looks like it was designed for toothless nursing home residents with computer-phobia and a lot more patience than I possess.  Obviously, ViewSonic wanted the device to be embraced by mainstream consumers so they dumbed-down the Android 2.2 interface with a sluggish, buggy and artificially limited mess of an overlay.  To make matters worse, the Android Market is nowhere to be found.

ViewSonic really shot themselves in the foot with the G Tablet.  They were first to the U.S. market with a dual-core Tegra 2-based device.  All they had to do was slap on a standard Froyo installation with a full Android Market and the device would have been a runaway hit for them this Christmas season.  But nooooooooooo!  ViewSonic had to get all greedy with visions of iPad’s success with mainstream buyers.  The resulting, lousy Tap ‘n Tap interface is like pouring a pound of aspartame over a steak dinner.  Sprinkled with bugs, the unsavory kind.

Fortunately, if you are a computer geek then it is not too difficult to flash the G Tablet’s firmware with a proper Android environment.  It’s also a fairly safe process since someone at ViewSonic had the foresight to make the device relatively brick-proof.  I’ve been using TnT Lite 3.0, but there are other options as well.  Yes, there will be headaches along the way, but geeks like me enjoy hacking a new device.

And, frankly, I have not been this excited about a new genre of computing device in many years.  The promise of the iPad was immediately evident to me when we bought one last spring.  However, a properly prepared G Tablet runs circles around the iPad.  Android-based tablets are going to dominate the marketplace by this time next year.

Of course, we bought the device for our business to benchmark and analyze.  Tegra-2 appears to be even faster than I anticipated.  The G Tablet finished under 2.5 seconds on SunSpider using Firefox Mobile 4.0 beta 2.  When I wrote my ARM versus x86 treatise last spring, the 800MHz Cortex-A8 took over 14 seconds on SunSpider while the 1GHz Intel Atom needed over 8 seconds, about as fast a my updated iPad takes today.  Note, however, that Firefox’s JavaScript performance has improved enormously over that time.  On the other hand, remember that JavaScript is still single-threaded, so half of the Tegra-2′s performance is left untapped on SunSpider.

There’s been weeping, moaning and gnashing of teeth over the quality of the G Tablet’s display.  Truth be told, those people are crybabies.  Yes, it’s not as good as the IPS screens on the iPad or the B&N NOOKcolor, but it’s not awful either (its biggest problem is blinding glare, not its relatively limited viewing angles compared to IPS displays).  However, I was expecting more from ViewSonic, a company best known for its outstanding history as a computer monitor vendor.  But given the general unrefinement of the device, I was not too surprised.  I mean, one look at the dingy, off white G Tablet box shows that the challenges of marketing a tablet computer are currently beyond ViewSonic.  I had to take out a ViewSonic monitor box to confirm my suspicion that apparently the monitor and tablet marketing folks at ViewSonic apparently never speak to one another.

Anyhow, it’s not too late for ViewSonic.  They need to ditch tepid Tap ‘n Tap for a real, full Android experience, enable a complete Android Market, push device driver updates to the tablet and recognize the G Tablet for what it is: a Grade A geek toy.  In fact, it appears that ViewSonic decided to take a step in this direction today by promising to push out a new firmware edition before Christmas that will not only improve Tap ‘n Tap, but will also give the user the option to boot into a stock Android interface.

Penetrating the mainstream marketplace will require hardware tweaking like adding an IPS screen, improving the lame webcam, rubberizing the case and bezel, adding mechanical Android buttons and dramatically rethinking the case ink and finish.  If they want a nearly perfect tablet, ViewSonic can add a digital compass, GPS and rear-facing camera.

We’ll be testing the ViewSonic G Tablet and writing benchmarks specifically for this purpose.  Hopefully, we’ll have results to report soon.

Apr 132010
 

BSN* has posted another of my articles. In it, I compare iPad browsing performance against an Intel Atom N450 netbook using the four most popular web browsers. While the Atom-based systems pretty much trounces the iPad except when using Microsoft Internet Explorer 8. IE8 on the Atom makes the iPad look fast.

Additionally, Opera 10.51 beats all comers.

Major kudos are also in order for Opera. Already a big player in the mobile browser market, Opera 10.51 manhandled the competition across all three benchmarks. Safari and Chrome finished a distant second and third respectively. Firefox 3.6.3 kept pace with those two except on the quirky Google V8 benchmark where it stumbled badly.

Apr 082010
 

The popular technology website Bright Side of News* has published an in-depth report I authored comparing an ARM Cortex-A8 microprocessor, used in the Apple iPad’s A4 chip, against a trifecta of x86 CPUs typically found in netbooks, small notebooks and embedded devices.  My report particularly focuses on compute performance.

A major component of that comparison is miniBench, an open source benchmark that I wrote in C++.  I ported miniBench to Linux for both ARM and x86 platforms enabling, for the first time, objective, head-to-head performance comparisons across a wide range of meaningful tests like Dhrystone, Whetstone, FFT, LinPack, MFLOPS, AES, SHA1, SHA256 and many others.

While I worked as head of benchmarking for Centaur Technology, we used miniBench to help isolate performance problems in our microprocessors so that we could optimize out the weakest attributes of our chip designs.

In the BSN* report, I also compare performance across a number of popular JavaScript benchmarks and a few other native tests.

The results surprised me.  It is also worth examinig the relative compute performance between the new Intel Atom N450, the new VIA Nano L3050 and an old AMD Mobile Athlon based upon the Barton core.

You can read my full report here.

I noticed that Theo generously gave me credit for the recall of the 1.13GHz Intel Pentium III.  I want to make it clear that Tom Pabst discovered the speed path defect that manifested when he was trying to compile the Linux kernel.  My part in the recall was representing Tom’s Hardware at Intel.  The representative for the giant chipmaker initially laughed at the issue until I threatened him with an united call to yank the defective part involving a number of major computer hardware ethusiast websites.

Jan 282010
 

Apple introduced its greatly anticipated tablet computer, the iPad, yesterday.  It was impossible for the product to meet the unbelievable hype surrounding the computer’s debut, but even taking this into account, the reaction to the device, which will not ship for months, has been mostly negative.  Critics point out the iPad’s lack of multitasking, the low resolution XGA 4:3 screen, the clunky aesthetics (very unusual for Jobs-ian Apple), the lack of standard ports, no support for Adobe Flash, the high price compared with more capable netbooks and the awkward “iPad” moniker.

But few if any observers have mentioned that the iPad is probably very underpowered compared with much cheaper netbooks.

The Apple iPad uses a custom, 1GHz Apple A4 system-on-chip (SoC).  While details are lacking, the microprocessor is probably either the ARM Cortex-A8 (a superscalar design a little like the Intel Atom, but sans HyperThreading) or Cortex-A9 (an out-of-order superscalar design superficially analogous to the VIA Nano).  If that is the case, this explains why Apple chose to forgo multitasking because ARM chips are very slow.

ARM has been very successful at avoiding direct, objective performance comparisons of their chips with x86 counterparts.  However, I’ve been able to test an 800MHz ARM Cortex-A8 running Ubuntu Linux and compare its results to an 800MHz AMD Mobile Athlon and a VIA Nano L3050 downclocked to 800MHz.  The x86 systems ran the same version of Ubuntu Linux as the ARM box.

While the Nano and the Athlon are close to performance parity with each other, the ARM Cortex-A8 is less than one-half as fast as either x86 chip.  Moreover, the ARM CPU is much weaker on floating point calculations, providing lower that 25% of the performance of either x86 chip.

Worse, the ARM system frequently becomes unresponsive for several seconds at a time.  And even though I only ran the ARM system at 16-bit XGA resolution while the x86 systems ran at 24-bit 1080p resolutions, both x86 systems trounced the video performance of the ARM box.

After using the ARM Cortex-A8 Ubuntu system, it is safe to say that people migrating from x86 netbooks (which typically use 1.3-1.6GHz x86 processors) — or, heaven forbid, thin-and-light notebooks — will be very disappointed with the performance compromises they will encounter from stepping down to ARM.

So if Apple is deploying a 1GHz ARM Cortex-A8 in the iPad, overall performance will be worse than a 500MHz x86 chip — sometimes much worse.  And that level of performance won’t make many people happy.  For Apple’s sake, if the company is indeed using an ARM design in its iPad, let’s hope that they at least chose a multi-core Cortex-A9.  BSN* confidently claims that this is indeed the case.

I’m trying to get my hands on an ARM Cortex-A9 system.  The Cortex-A9 will boost performance over the A8 because it adds an Out-of-Order engine which reduces pipeline stalls.  However, I expect a 1GHz Cortex-A9 to be no faster than a 600-650MHz x86 counterpart at best.

These results don’t surprise me because several years ago when I was working at Centaur I compared the performance of an Intel XScale chip against a VIA C7.  The C7 creamed the XScale even though the ARM chip ran at a slightly higher clock speed.

I’m convinced Intel sold off XScale to Marvell because the chipmaker recognized that this performance deficit would be untenable once ARM chips inevitably began competing with x86 designs.  Of course, Intel’s response was to create the highly successful Atom.

I’m trying to get permission to publish the ARM Cortex-A8 results.  Check back in a few days if you are interested.

ARM CPUs certainly have their strengths, but raw performance is not one of them.  They will face very stiff competition as they go head-to-head with much more powerful x86 designs.