As you probably know, Chippy is in the trenches at IDF as we speak type. In addition to a solid hands-on with the devices, he’s also doing testing and keeping us up to date with some audio logs. Let’s first have a look at the blazing speed of the Galaxy Tab 7.7 as indicated by the Sunspider benchmark:
This score, which is thanks to a fast dual-core processor and the latest build of Android, puts the Tab 7.7 at the number 1 position on our Sunspider benchmark chart:
Chippy faily notes that Chrome has improved in performance since then so just I ran Sunspider on a fairly modern Atom based netbook (1.6GHz HP Mini 311 with Nvidia ION) and the scores are comparable with the Mini 311 scoring 1336ms, putting the Tab 7.7 only about 8% away from that score.
This is mighty impressive; here we’ve got the Tab 7.7 which is rated for around 10 hours of battery life, not to mention much longer on standby, and it has browser performance similar to a much larger netbook that has considerably less runtime. Intel might be in trouble…
Chippy Chimes In
Chippy was able to make a little time to share with us a quick audio log with some thoughts on the Tab 7.7 and the Galaxy Note. I’ll let you listen for yourself, but I will say that I share many of his sentiments:
As you probably know, Chippy is in the trenches at IDF as we speak type. In addition to a solid hands-on with the devices, he’s also doing testing and keeping us up to date with some audio logs. Let’s first have a look at the blazing speed of the Galaxy Tab 7.7 as indicated by the Sunspider benchmark:
This score, which is thanks to a fast dual-core processor and the latest build of Android, puts the Tab 7.7 at the number 1 position on our Sunspider benchmark chart:
Chippy faily notes that Chrome has improved in performance since then so just I ran Sunspider on a fairly modern Atom based netbook (1.6GHz HP Mini 311 with Nvidia ION) and the scores are comparable with the Mini 311 scoring 1336ms, putting the Tab 7.7 only about 8% away from that score.
This is mighty impressive; here we’ve got the Tab 7.7 which is rated for around 10 hours of battery life, not to mention much longer on standby, and it has browser performance similar to a much larger netbook that has considerably less runtime. Intel might be in trouble…
Chippy Chimes In
Chippy was able to make a little time to share with us a quick audio log with some thoughts on the Tab 7.7 and the Galaxy Note. I’ll let you listen for yourself, but I will say that I share many of his sentiments:
On the last day of MWC I just had one more task to do and that was to spend as much time as possible with the Galaxy Tab 10.1. Contacts at MWC had already told me that it was good so I headed over to Samsung with time to spare.
Yes, the Galaxy Tab is a quality bit of industrial design. The back feels slightly disappointing if you’re looking for appeasing materials but it’s practical and I’m sure that it’s going to be of long-lasting quality. The camera is good not awesome, but good enough. The speakers disappoint. The other part of the product that disappoints, and you might hear this from other reporters is that the UI was a bit underwhelming. Beware, this is incomplete software so we should’nt use it as any sort of indicator. In its current state it’s worse than that Galaxy Tab so I feel certain it can’t go out of the door like that. It was missing Market, some BT features and the browser had a lot of trouble with Google Mail and iGoogle.
In a Sunspider test I managed to get some idea of how Honeycomb and Tegra 2 are going to perform together. The score you see below is 4x better than the iPad and a lot better than results I have from Android 2.1 on Tegra2. A 7 inch Galaxy Tab returns about 7000 ms
Yes that’s 2256. Over 2x faster than the Galaxy Tab 7 inch-er!
As I begin a series of tests on the Aava prototype phone today, the first thing I wanted to get out of the way was a Sunspider javascript test. It’s CPU-intensive and forms an important part of the chain of events that take place to get a web application displayed on your browser. Yes, it’s one of many variables but it’s a good indicator of CPU performance.
I’ve been recording SunSpider tests for a long time now (feel free to contribute to that list) and have been very impressed by the way that both hardware and software improvements have brought the figures down. In the ARM-based world, the best-in-class devices are producing SunSpider results in less than 10 seconds (iPad, iPhone 4) with some new devices even reaching down to sub 6 seconds [See title reference.] In the world of Intel, netbooks are producing Sunspider results of about 2 seconds (using the latest Chrome build.)
Click to enlarge.
Above is the result for the Aava phone I’m testing at the moment. 4215ms. It runs on Intel’s Moorestown platform at 1.5Ghz and uses the same CPU as you find in netbooks. Running Sunspider on the latest Firefox build on the netbook and the Firefox Mobile build on the MeeGo-based Aava phone you can see the difference is marginal. 4.2 seconds for a prototype phone is quite impressive. The figures confirm the CPU speed and also confirm that the sunspider score is likely to drop to around 2 seconds with the latest Chrome or Android browser build. Try Sunspider on your phone or PC browser to see what score you get. For reference, I use a ultra mobile PC as my desktop PC. It runs a 1.6Ghz Atom CPU and returns a score of about 4.1s. True desktop PCs will come in at below 0.5s
We really are talking PC-power here but there are two things we must not forget. 1) Multi-core ARM-based devices are round the corner too. A dual-core 1.2Ghz Snapdragon platform is likely to reach right down to the 2-3s range. 2) We must also remember that if you were to run this test continuously on these high-power platforms, you are likely to have a dead phone in just a few hours!
At an ARM event in Taipei this week I was fortunate enough to have a few minutes with a Tegra 2-based tablet prototype known as ‘Harmony.’ It’s a fairly standard tablet with a capacitive touchscreen and weighing in at the 700gm iPad-weight. You’ll see more details in the video below.
The interesting thing was the SunSpider javascript benchmark I did. As you might know, SunSpider is a well-recognized test of a browser’s ability to run javascript and it serves as a good data point for working out how fast a processor is. As far as I know, most, if not all, the javascript processing is done by the CPU.
You’ll see a result of 9.8 seconds on the video which is about the same as you’d see on the iPad. The iPad uses what we believe is an ARM A8 core, or at least something very similar. However, that doesn’t mean that dual-core Tegra 2 is only as powerful as a single-core A4 CPU because the two browsers are vastly different. The javascript engine on the iPad is super-fast where the engine used on the Android 2.1 browser isn’t.
Android 2.1 on Tegra 2 9 seconds (Tegra prototype shown in video)
Android 2.2 on Snapdragon 1Ghz 6 seconds. (Nexus One, Google V8 engine)
Android 1.6 on Snapdragon 1Ghz 24 seconds (Xperia X10)
Chrome on Atom 1.6 2 seconds (average netbook, Google V8 engine)
As you can see there, the Android browser is at least 4 times less efficient in processing javascript on version 1.6 than it is on 2.2. I suspect that version 2.1 is close to 1.6 in its efficiency too which means that the Tegra 2 is over 200% faster than Snapdragon. Given that it has two cores, it’s not surprising. A single A9 core (in my estimate) brings about 20% more raw CPU performance over A8 so the figures look correct.
Add in the 4X improvement that you’ll see with Froyo’s Google V8 processing engine and there you go! Tegra2 with Froyo will be able to process javascript as well as a 1.6Ghz netbook. Just imagine what the 1.2Ghz dual-core Snapdragon will do!
We’ve already seen that ARM can beat Atom in certain low-end Atom scenarios and cross referencing to some other figures I have seem to confirm that ARM is going to challenge Atom in raw CPU. If only they could sort the operating system out and get some productivity apps going, we’d have some interesting smartbooks out there.
Chippy Sidenote: 2 cores needs 2x power! It would be interesting to work out the CPU drain figures under these javascript test conditions. Also note that system performance is not directly related to CPU performance. A lot of work has to go into a lot of other hardware and a lot of other software to get a system running ‘fast.’ Please remember that when Intel and ARM enter the clock-speed game over the next few years!
