Fresh out the frying pan and into the fire, I just finished my Nexus One review late last night only to have my iPad preorder show up early this afternoon. I had been preparing for it's arrival not by downloading apps but by figuring out what comparative benchmarks I wanted to run on the iPhone 3GS and Nexus One.

As the first device to use Apple's A4 SoC I wanted to see how it stacked up against the Cortex A8 and Qualcomm's QSD8250. All three chips appear to be dual issue in order architectures with varying pipeline depths, clock speeds and cache sizes.

At 600MHz the Cortex A8 in the iPhone 3GS is the slowest out of the bunch. The Snapdragon is much faster as we just established thanks in part to it's 1GHz clock speed. But what about Apple's 1GHz A4?

There's very little we know about the A4 other than it's operating frequency. It is manufactured by Samsung but on an unknown process node. Jon Stokes recently stated that Apple's secrecy surrounding the chip is because it isn't anything special, just a Cortex A8. If that is true, I suspect that it would have to be manufactured at 45nm in order to reach such a high clock speed.

With a new silicon mask there's also the chance that Apple moved to LPDDR2 to boost memory bandwidth; a change that most SoC makers are planning to make this year.

So how does Apple's A4 stack up against today's favorite smartphone brainchild? Keep in mind that these results are generated by running two different OSes (Android 2.1 and iPhone OS 3.2) and two different browsers. What we're looking at is the performance delivered by the combination of the CPU and the software stack:

Applications Processor Performance
  Apple iPad (Apple A4) Apple iPhone 3GS (ARM Cortex A8) Google Nexus One (Qualcomm Snapdragon QSD8250) % A4 Faster than Snapdragon
Load www.anandtech.com 6.2 seconds 9.3 seconds 8.8 seconds 41%
Load www.digg.com 10.6 seconds 18.0 seconds 11.5 seconds 8.7%
Load www.tomshardware.com 7.9 seconds 13.9 seconds 8.6 seconds 8.7%
Load www.arstechnica.com 7.8 seconds 13.8 seconds 11.0 seconds 39.9%
Load www.legitreviews.com 6.8 seconds 12.3 seconds 8.6 seconds 26%
Load www.techreport.com 3.7 seconds 7.4 seconds 4.2 seconds 11.6%
Load www.engadget.com 13.8 seconds 22.8 seconds 22.0 seconds 59.4%
Load www.gizmodo.com 14.1 seconds 21.4 seconds 16.7 seconds 18.5%
Load m.cnn.com 3.0 seconds 6.0 seconds 2.6 seconds -11.8%

Unless otherwise specified, I loaded the full version of all of the websites above (the exception being CNN, where I used the mobile site). To ensure reliability, I ran all of these tests at least 5 times, threw out any outliers and averaged the rest. The rests were also run at around the same time to ensure that content on the sites was as similar as possible (and thus shouldn't be compared to this morning's Nexus One results). You'll note that the Engadget results are a bit odd. It looks like the iPhone and Nexus One scores are bottlenecked somewhere else (there seemed to be some network issue plaguing the loads, but it wasn't present on the iPad), but if you toss out the very large differences you end up with what I believe to be the real story here. Update: Flash wasn't enabled on any device (not supported on iPad/iPhone, not officially available on Android yet), and all three devices connected to the same WiFi network.  The Apple devices used mobile Safari, while the Android device used the Android Browser.  Both are WebKit based but there are obvious, unavoidable software differences.

Removing the AnandTech, Ars Technica and Engadget loads (which were repeatable, but unusually long) the iPad loads web pages 10% faster than the Nexus One. If you include those three results the advantage grows to 22.5%. I'd say somewhere in the 10% range is probably realistic for how much faster the A4 is compared to the Snapdragon.

I also ran the official WebKit SunSpider JavaScript Benchmark on all three platforms to give us a network independent look at real world JavaScript performance:

If we take the network out of the equation, the A4 in the iPad has a 37.6% performance advantage over the Qualcomm QSD8250. This actually supports some of the larger performance differences we saw earlier. If Apple can manage to deliver this sort of performance in its smartphone version of the A4, we're in for a treat.

The why is much more difficult to ascertain. It could be as simple as the the iPad OS being better optimized than Android, a definite possibility given how much longer Apple has been working on it compared to Google. The advantage could also be hardware. The A4 may boast higher IPC than Qualcomm's Snapdragon thanks to better core architecture, larger caches or a faster memory bus. The likely case is somewhere in between, where the iPad's advantage comes from a combination of hardware and software.

It could also be a power optimization thing. The A4 in the iPad is paired with a much larger battery than the QSD8250 in the Nexus One, Apple may be able to run the SoC at more aggressive performance settings since it doesn't have to worry about battery life as much. Either way the one thing we can be sure of is Apple's A4 SoC is much more like a 1GHz Cortex A8 rather than anything more exotic. Good work Jon :)

I should note that while the performance improvement is significant, it's not earth shattering. Despite the early reports of the iPad being blazingly fast, I found it just "acceptable" in my limited time with it thus far. I'll go into greater detail in my full review later.

This does bode well for the upcoming 4th generation iPhone, which is widely expected to also use the Apple A4 SoC. That upgrade alone should put the next iPhone ahead of Google's Nexus One in performance, assuming that it offers the same performance as it does in the iPad. Pair it with a modernized and feature heavy iPhone OS 4.0 and we might see an Apple answer to Android in 2010.

The A4 is particularly exciting because it combines Snapdragon-like CPU performance with a PowerVR SGX GPU. A much better option than the aging ATI core used in Qualcomm's QSD8x50 series.

With Apple showing its A4 performance this early, Qualcomm also has a target to aim at. The first single-core 45nm Snapdragon SoC due out in 2010 will run at 1.3GHz. That could be enough to either equal or outperform Apple's A4 based on what we've seen here today.

Expect our full review of Apple's iPad as well as more discussion about the A4 next week. Have a great weekend guys.

POST A COMMENT

159 Comments

View All Comments

  • has407 - Sunday, April 04, 2010 - link

    Agree. On identical hardware/OS there's typically a 10:1 range in SunSpider performance depending on browser, but the front-runners are closing up (see e.g., http://ie.microsoft.com/testdrive/benchmarks/SunSp... However, not all WebKit browsers use the same javascript engine (e.g., Safari, which uses a proprietary Apple engine).

    As you suggest, it's a pale reflection of raw CPU performance, but it helps illuminate an area of platform performance of significance and of interest--which for these devices/appliances is an important measure--in particular for javascript/AJAX-intensive apps. Probably a good time to rethink what and how to benchmark for these types of devices. E.g., for synthetics javascript, rendering and network performance tests come to mind.
    Reply
  • calvin80 - Sunday, April 04, 2010 - link

    How about some basic ones too to compare the CPU/SoC platform -- memory read write tests, CPU intensive tests, file system read/writes. Reply
  • Chloiber - Saturday, April 03, 2010 - link

    It's a great preview, thanks for that.

    Of course it's completely valid to compare the iPad to an iPhone or a N1 or any other smartphone as they use nearly identical hardware.
    But as long as you have the difference in battery power, it's still impossible to come to a conclusion like "A4 > Snapdragon". As long as you don't have the same surroundings (or at least comparable) it's impossible IMHO.

    But I still like the short review (and hopefully more thorough review later). Because in the end, what you wanna know is how much faster the iPad is compared to your smartphone and you dont care about "the surroundings".
    Still, you can't extrapolate to a "real" mobile version of the A4 imho. It should say "iPad vs. N1" or "A4 in iPad vs. Snapdragon in N1".
    Reply
  • sxr7171 - Saturday, April 03, 2010 - link

    Did anyone really think the iPad was anything but a hyped up iPod touch with a bigger screen? If anything I'm actually surprised it beats the Snapdragon ever so slightly. Reply
  • sxr7171 - Saturday, April 03, 2010 - link

    I mean they are completely different software platforms. This is a Nexus One vs. iPad comparison.

    I bet the Apple software is highly optimized and as a platform it is faster, but we still have no way of know how SNAPDRAGON VS. A4 compares. So really this article needs a new name. Unless you can put Android on the iPad or iPhone OS on the Nexus you can't call this a CPU or SOC review.
    Reply
  • BreakingStrata - Saturday, April 03, 2010 - link

    Hey Anand, lets say the A4 is just a faster clocked A8. What if the performance scales with clock speed? Then the 3GS should be roughly 60% as fast as the iPad. Well if you divide the test results...

    SunSpider Javascript test:
    iPad: 10475
    3GS: 17360
    10475/17360=0.6033982

    To me it looks like the A4 really is nothing than a faster clocked A8, which would explain the secrecy. Trying to make something magical out of nothing.

    Your thoughts?
    Reply
  • kalster - Sunday, April 04, 2010 - link

    what about the processor in the newly announced samsung galaxy s. that thing is supposed to be (on paper) faster than snapdragon, will be interesting to see how it stacks up against the A4 Reply
  • TemplarGR - Sunday, April 04, 2010 - link

    I agree with all those that said this isn't an apples to apples comparison. And this isn't:

    1) It should be obvious that a much bigger device would be faster. They are not in the same class. Even if they use the same class SoC, there is room for other improvements for the bigger device. For example larger/faster RAM. Some co-processing chips. Faster interconnection etc.

    2) Nothing is said about the die area of both chips. I imagine the ipad's cpu is bigger than Nexus's. I could be wrong of course on this one, but i don't have much information.

    3) The iPad has much more room for battery and thermals, and this gives it a nice advantage.

    4) The iPad OS is much more refined than Android at the moment.

    5) Similarly the browsers are different.

    6) The smartphones never stop to act as phones, they aren't simple PDAs. There is some overhead to that, especially considering small RAM sizes.

    So in the end, this comparison is invalid. I want to believe Anand isn't biased and trying to boost iPads image, but he has made a big mistake here. Please compare the iPad to netbooks and CULVs, not Smartphones. Besides, even if the iPad's Soc finds its way into iPhone, it will be a crippled version due to smaller factor...
    Reply
  • Anand Lal Shimpi - Sunday, April 04, 2010 - link

    It's not directly an apples to apples comparison, but it is the best we can do at this point unfortunately:

    1) Remember these are smartphone SoCs - memory, interconnects, coprocessors are all on the same package.

    2) Die area is unknown (and rarely quoted), but package size appears to be comparable based on the leaked FCC data and ifixit teardown.

    3) The larger battery doesn't give it an advantage in this performance test. It's possible the thermals do, but the device isn't actively cooled and doesn't have a heatsink. At 1GHz it would almost have to built on a 45nm process. A 1GHz 45nm Cortex A8 has already been shown off by both TI and Samsung for future smartphones. I don't think there's any reason to believe that the A4 couldn't make it into an iPhone sized device if it is comparable to the 45nm Samsung/TI SoCs.

    4) This is potentially very true (I didn't write either OS and refined is pretty broad, each platform appears to have its own maturity points). It's because of this that I made the following statement after the Sunspider results:

    "It could be as simple as the the iPad OS being better optimized than Android, a definite possibility given how much longer Apple has been working on it compared to Google. "

    5) Also agreed. Keep in mind that the software advantage can't be *that* great because the Nexus One still manages to step all over the iPhone 3GS.

    6) That's very possible, this goes back to the optimization point earlier. I don't believe any overhead present would make up the gap we saw between the Nexus One and the A4 however.

    I'll be comparing the iPad to both netbooks and smartphones in the review. The fact of the matter is that it bumps into the netbook category in terms of functionality, but has the performance characteristics of a high end smartphone - making a comparison to both very important.

    Take care,
    Anand
    Reply
  • bsoft16384 - Sunday, April 04, 2010 - link

    Anand,

    Remember that the iPad is running a newer version of the iPhone OS than the iPhone 3GS (3.2 vs 3.1). Considering the recent JS performance wars, it's entirely possible that the JS engine shipped with the iPad is faster than the JS engine shipped with the iPhone 3GS (since Apple/Google/etc. are constantly working on their JS engines).

    I would be hesitant to compare JS results between different software platforms and versions.

    As I pointed out, my 2.53GHz Core 2 Duo ThinkPad T400 is 29 times faster than the iPad in the SunSpider benchmark, and that's with one core active. That works out to 11x faster per clock, which is absolutely stupid. Some of the difference is certainly due to Core 2's vastly superior 2 IPC (much bigger caches, out of order execution, higher memory bandwidth, better branch prediction, 4-issue vs. 2-issue, etc.), but I'm willing to bet that at least part of the difference is due to the fact that the Chrome build I'm using (latest beta) has a better JS engine than the iPad.

    The bottom line is that JS benchmarks aren't really useful comparing CPUs when you have different JS engines (or at least different JS engine versions). Chrome, for example, is twice as fast as Firefox 3.6 on my T400.

    I think you're probably right that the A4 beats the Snapdragon, but it's hard to tell. People keep saying that the iPad is "super fast", and, compared with the typical smartphone, they're right. But the iPad isn't a smartphone and it doesn't compete with or replace a smartphone. What it does compete with are netbooks and CULV notebooks, and it's nowhere close to the former - let alone the latter.

    I'm going to go run some benchmarks on my Mom's 1.4GHz Core 2 Solo Acer 1410. I'm betting that Engadget is going to load in way less than 14 seconds.
    Reply

Log in

Don't have an account? Sign up now