i noticed the first chart at the top of your first page has 1ghz for the cpu speed all the way down when the paragraph right below it states that it the 3d receives a 200mhz bump.
Font changing is also available on my Galaxy S i9000. Samsung ships the phone with 3 extra fonts and you can download more from the Market by searching for "FlipFont".
I'm not sure I agree with the wording in this article about TBDR. The reason PowerVR didn't need to slap DDR RAM onto the Kyro II cards is because they simply didn't need it thanks to the reduction of traffic that comes from deferred rendering. The unknown element at the time was hardware T&L because it simply wasn't available and was thought to be impossible, however as this is yet again being performed on-die, wouldn't that also result in a marked reduction in traffic? Might need some clarification on this one.
I've never seen it confirmed that the Adreno GPU performs TBDR; some clarification would be appreciated on this one as well! :)
Onto the option for changing fonts... my vendor-agnostic Galaxy S has such a feature called "Font style" under the Display settings, allowing you to choose the "Default font", "Choco cooky", "Cool jazz" and "Rosemary", with the option of getting more online.
Qualcomm bought out AMD's mobile GPUs and hence the Adreno 200 was a rebrand of AMD z430. Here's a little more background on the tiling nature of z430:
(since apparently I can't post a link without being flagged as spam just google for this: gdc2008_ribble_maurice_TileBasedGpus.pdf - it's the first hit)
Is the camera really too thick to fit the width of the phone? I heard of many Evo users who have cracked the glass covering. While I haven't had this problem, it makes me wonder if it's really necessary in the first place.
I cant wait until you can wirelessly shoot your phones display to your tv ala intel WiDi. This HDMI out is cool, but the cable kind of kills it. With wireless display your phone becomes a game controller :) or a remote when shooting movies to your tv. Plus how cool would it be if your phone can fetch movies from your home network and you shot the playback to your TV. The Sony NGP should have implemented this. Ah well.
May I suggest x264 encoding as a test of the CPU power? There's a version of x264 available for ARM chips, along with NEON optimizations. Should be interesting!
What is the point in having a high performance video processor when you cannot do the two things that actually make use of it? Those two things are: 1. Watch any movie in your collection without transcoding? (FAIL) 2. Play games. No actual buttons = FAIL. If you think otherwise then you dont actually play games. Just stick with facebook flash trash.
There are some issues I've found with some information in this article:
1) You mention that Cortex-A8 is available in a multicore configuration. I'm pretty sure there's no such thing; you might be thinking of ARM11MPCore.
2) The floating point latencies table is just way off for NEON. You can find latencies here: http://infocenter.arm.com/help/index.jsp?topic=/co... It's the same in Cortex-A9. The table is a little hard to read; you have to look at the result and writeback stages to determine the latency (it's easier to read the A9 version). Here's the breakdown: FADD/FSUB/FMUL: 5 cycles FMAC: 9 cycles (note that this is because the result of the FMUL pipeline is then threaded through the FADD pipeline) The table also implies Cortex-A9 adds divide and sqrt instructions to NEON. In actuality, both support reciprocal approximation instructions in SIMD and full versions in scalar. The approximation instructions have both initial approximation with ~9 bits of precision and Newton Rhapson step instructions. The step instructions function like FMACs and have similar latencies. This kind of begs the question of where the A9 NEON DIV and SQRT numbers came from.
The other issue I have with these numbers is that it only mentions latency and not throughput. The main issue is that the non-pipelined Cortex-A8 FPU has throughput almost as bad as its latency, while all of the other implementations have single cycle throughput for 2x 64-bit operations. Maybe throughput is what you mean by "minimum latency", however this would imply that Cortex-A9 VFP can't issue every cycle, which isn't the case.
3) It's obvious from the GLBenchmark 2.0 Pro screenshot that there are some serious color limitations from Tegra 2 (look at the woman's face). This is probably due to using 16-bit. IMG has a major advantage in this area since it renders at full 32-bit (or better) precision internally and can dither the result to 16-bit to the framebuffer, which looks surprisingly similar in quality to non-dithered 32-bit. This makes a 16-bit vs 16-bit framebuffer comparison between the two very unbalanced - it's far more fair to just do both at 32-bit, but it doesn't look like the benchmark has any option for it. Furthermore, Tegra 2 is limited to 16-bit (optionally non-linear) depth buffers, while IMG utilizes 32-bit floating point depth internally. This is always going to be a disadvantage for Tegra 2 and is definitely worth mentioning in any comparison.
Finally I feel like ranting a little bit about your use of the Android Linpack test. Anyone with a little common sense can tell that a native implementation of Linpack on these devices will yield several dozen times more than 40MFLOPS (should be closer to 1-4 FLOP/CPU cycle). What you see here is a blatant example of Dalvik's extreme inability to perform with floating point code that extends well beyond an inability to perform SIMD vectorization.
It is mostly FP64 calculations done on Dalvik. While this may not be the fastest way to go about doing linear algebra, it is a fairly good representation of relative FP64 performance (which only exist in VFP).
And let's face it, few app developers are going to dig into Android's NDK and write NEON optimized code.
Then let's ask this instead: who really cares about FP64 performance on a smartphone? I'd also argue that it is not even a good representation of relative FP64 performance since that's being obscured so much by the quality of the JITed code. Hence why you see Scorpion and A9 perform a little over twice as fast as A8 (per-clock) instead of several times faster. VFP is still in-order on Cortex-A9, competent scheduling matters.
Maybe a lot of developers won't write NEON code on Android, but where it's written it could very well matter. For one thing, in Android itself. And theoretically one day Dalvik could actually be generating NEON competently.. so some synthetic tests of NEON could be a good look at what could be.
Linpack as it currently exists on Android probably doesn't tell very much at all. But if you're just going to slap together an FP heavy app (pocket scientific computing anyone?) and aren't a professional programmer, this likely represents the result you see.
I wouldn't mind seeing SpecFP ported natively to Android and running NEON. But alas, we'd need someone to roll up their sleeves and do that.
I did do a native compile of Linpack using gcc to test on my Evo, though. It's still not SIMD code, of course, but native results using VFP were around the 70-80MFLOPS mark. Of course, it's scheduling for the A8's FPU and not Scorpion's.
1) You're very right, I was thinking about the ARM11 - fixed :)
2) Make that 2 for 2. You're right on the NEON values, I mistakenly grabbed the values from the cycles column and not the result column. The DIV/SQRT columns were also incorrect, I removed them from the article.
I mentioned the lack of pipelining in the A8 FPU earlier in the article but I reiterated it underneath the table to hammer the point home. I agree that the lack of pipelining is the major reason for the A8's poor FP performance.
3) Those screenshots were actually taken on IMG hardware. IMG has some pretty serious rendering issues running GLBenchmark 2.0.
4) I'm not happy with the current state of Android benchmarks - Linpack included. Right now we're simply including everything we can get our hands on, but over the next 24 months I think you'll see us narrow the list and introduce more benchmarks that are representative of real world performance as well as contribute to meaningful architecture analysis.
I'm surprised to hear that shot was from IMG, given that it was an IMG employee who made the comment originally about Tegra's 16-bit banding being evident on it, from the screenshot. Whoops. I do wonder what could be causing this, then.
Nonetheless, while that definitely makes my 16bit color claim invalid the depth buffer one should still hold. We might need to wait and see how much of a difference this actually makes, or rather how effective nVidia's 16-bit depth space is.
I'm glad to hear that you're as concerned about benchmarks on Android as I am. It's especially frustrating when I see people using them to try to indicate Atom being substantially better clock for clock than Cortex-A9.
Great review as usual but i'm disappointed about the lack of details on the Sound Quality of the phone. A comparison of the sound quality (via headphone jack) alongside the iPhone 4 and the Galaxy S (with Voodoo kernel preferrably) would give a good idea about the SQ of the phone.
It plays back from the phone properly, and most of the playback software just does a crop. A ton of devices actually produce 1088 and don't make note of it, it should playback fine.
Pages and pages of (apparently not very acurate, too) perf data, and not even one line on sound quality, which is one of my key buying points for a phone.
No info on whether I'll be able to stream PC-resolution videos off my server to my bed over wifi.
Overall, not a very useful review. More like a dick size contest.
Do you have any idea if that audio chip will make it into all future Tegra 2 phones, or is this just a lucky phone? Droid Bionic with a Wolfson DAC would make me a happy camper.
Those specific choices are ultimately left up to the handset makers. It seems to be a relatively common component to select however. I have no idea whether the Bionic or Atrix are getting that same audio IC.
We can definitely run RightMark audio analyzer on phones from now on. ;)
This is probably the most comprehensive Optimus 2x review I've seen yet, so big kudos.
What still bugs me about phone reviews is they don't seem to measure how well GPS receivers work in the phones. Even a simple stick it outside at the same location and see how many satellites show up and how quickly (GPSStatus has a feature that measures time to first fix) would help. E.g., my crappy Moto Cliq sometimes flakes out and even if it claims it sees 10 satellites, it never gets a fix. It's important for those of us who are using our phones for navigation...
There is a whole section titled "GPS Performance" on page 9, basically comes down to that the hardware seems fine but since it is still preproduction A-GPS wasn't set up.
Good performance but the back of the phone is the ugliest I have seen in a while ... And why the plastic ... Isn't this supposed to be a premium phone?
Solid review, but please at least use spell check. I'm not a grammar or typo freak, but there were way too many simple typos that spell check wouldn't even let you get by with. At least have someone proof read it before you publish to the public.
Yeah, and until then those who manage to come up with some decent power management will be the richest...
Seriously, every improvement on the battery front almost always just leads to devices drawing more power. It's somewhat ironic that last year's iPhone still leads the pack when it comes to battery life. Power management (that is: don't draw more power than absolutely necessary by throttling or shutting down components that aren't needed or aren't fully needed in a given moment) is hard and boring design work nobody seems to care for. And with devices and software getting replaced with the next iteration every few months this is even understandable, it's just not worth the effort, especially when nobody seems to care and benchmarks are so much more important to the crowd.
Tegra 3 has 4 1.5GHz Cortex-A9's according to a leaked slide.
That was a great article! A few minor corrections: The ARM11 VFP is fully pipelined (so it can beat the A8 on FP performance). Like the A8, Scorpion is 2-way in-order, not limited out-of-order. In-order cores issue instructions in-order but may complete them out-of-order. On the other hand, OoO cores use register renaming to issue instructions out-of-order but complete them in-order.
Note none of the micro benchmarks used emits Neon instructions. JIT compilers don't have enough time to generate high quality code, let alone autovectorize! For proper benchmarking you will need to run native code compiled with a quality compiler (not GCC - it is still far behind the state of the art on ARM, especially Thumb-2).
I would argue with that definition of OoO. A design does not need register renaming in order to issue any arbitrary instruction OoO. It's simply a trade-off of whether to centralize hazard tracking on register accesses or on retirement.
"NVIDIA tells us that the Tegra 2 SoC is fully capable of a faster capture rate for stills and that LG simply chose 2MP as its burst mode resolution. For comparison, other phones with burst modes capture at either 1 MP or VGA. That said, unfortunately for NVIDIA, a significant technological advantage is almost meaningless if no one takes advantage of it. It'll be interesting to see if the other Tegra 2 phones coming will enable full resolution burst capture. unfortunately for NVIDIA, a significant technological advantage is almost meaningless if no one takes advantage of it. It'll be interesting to see if the other Tegra 2 phones coming will enable full resolution burst capture. meaningless if no one takes advantage of it. It'll be interesting to see if the other Tegra 2 phones coming will enable full resolution burst capture."
LG have probably made this decision based on the sensitivity of the invariably minuscule sensor they will have used. Having 6 frames of 12mp is pointless if they are 12 incomprehensible megapixels due to the lacklustre sensitivity of the pixels in their chosen part.
The kind of sensor you find delivering a meaningful burst in something like a 5D mk2 is enormous and power hungry, in comparison to an operating environment such as a phone.
Just curious because I've heard rumors that HP will use the Qualcomm chipset and I've also heard rumors that they will stick with Ti for their new tablets/phones. I just wondered if you know for sure since I know that you met with folks at CES. I hope that we all find out tomorrow at the HP event. Great review.
Why is there no Incredible on 2.2? I could understand if you had both 2.1 and 2.2, like the Evo, but as it is now does not show the full/current performance.
"It’s a strange dichotomy that LG sets up with this launcher scheme that divides “downloaded” apps from “system applications,” one that’s made on no other Android device I’ve ever seen but the Optimus One. The end result is that most of the stuff I want (you know, since I just installed it) is at the very last page or very bottom of the list, resulting in guaranteed scrolling every single time. If you’re a power user, just replace the launcher with something else entirely."
You are not right there. First you can create additonal categories (aside from system applications and downloads) and move applications between them. Secondly you can rearrange the ordering of the applications inside a category (allowing you to have those on top which you access most frequently). You can also delete applications right away in this edit mode.
Anand, A question on the statement in the benchmarking section, "the SunSpider benchmark isn't explicitly multithreaded, although some of the tests within the benchmark will take advantage of more than one core. "
My understanding was that all of the tests within sunspider are single-threaded, but a dual-core processor can run the javascript engine (and the sunspider tests) in a separate thread from the main browser thread when you call sunspider from a browser window.
Can you clarify which tests support multi-threading in sunspider if that is in fact what you meant?
On the topic of multi-threading, we've used moonbat, a multi-core variant of sunspider, to explicitly test multi-core performance with javascript code. I wonder if you have any other benchmarks under investigation that measure multi-core performance? Thanks
I now own a optimus 2x. The first was dead on arrival, but this one is perfect. The LG software is innovative and pleasing to the eye. In various places they made real improvements to the UI that are just brilliant,ie. the ability to sort and categorize apps. At times the UI is not as fast as you would expect, especially when adding apps/widgets to one of the 7 pages. It seems LG generates a list of widgets for you, so you can see what apps support this mode, and that takes about a second. As I recall, on HTC devices you are just presented with a list of apps and u have to try and see if you can widget it.
The LG keybord has a brilliant feature, you tab the side of the phone to move the cursor. Sadly in other respects the keyboard is lacking, ie. when you long-pres you do not get the alternates you might wish, such as numbers.
The batterytime is superb, using the UI consumes much less power than on my desire.
Copy/paste in the browser does not activate via long-pres, you have to hit menu button, but on the plus side its easier to use than what HTC made.
During 2 days of very intensive use i have had 1 app (partially) crash and that was the marketplace. No other issues so far, its my verdict that the unstability issues are overrated.
No problems with wifi using stock ISP (TDC) supplied router. (sagemcom)
To engadget: How on earth (!!?!!?) can you state there is no use for dualcore. When browsing one loads flash the other the rest. Its so fast you cant believe it. Try loading www.ufc.com on a non dualcore phone and you get my drift.
I do not hesitate to give the optimus 2x my warm recommendations.
Very nice article ... and I'm just a layman about this kind of information technology. But maybe someday I should know about complicated things like the above. I can only make the review at a glance (and maybe subjective, try looking at this page: http://www.bestdealscomputers.net) on products that use chips, such as NVIDIA, on a tablet computers. Move on .....
So Smartphone can do all that can be done of a tablet computer? The only restriction the size of the screen just ... look at the specifications on the following pages ... http://www.bestdealscomputers.net/tablets/motorola... Wow ....
An interesting story, I recently purchased a HTC Panache (aka MyTouch 4G) and was really disappointed by the battery life when compared to my wifes iPhone, my iPad and my work BlackBerry Torch - but looking at the stats here it appears that it's about average when it comes to Android devices and other phones.
I am still looking at getting a higher capacity battery.
Optimus 2X uses Duel Core so the speed is very fast. When playing a game with my friend, Optimus 2X loads faster than his smartphone. If users with an LG smartphone want to download various apps or games, I recommend to visit uk.lgworld.com
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
75 Comments
Back to Article
andrewsolid - Monday, February 7, 2011 - link
"Tegra 2 uses a uses a" 6th para from the baseAnand Lal Shimpi - Monday, February 7, 2011 - link
Fixed!Take care,
Anand
NCM - Monday, February 7, 2011 - link
Call for effective editing...Adobe "flash" or just plain "flash" used in multiple instances should be "Flash."
MonkeyPaw - Monday, February 7, 2011 - link
"Android itself already is multithreaded naively, in fact, that’s part of delivering speedy UI. "Naive Mutlithreading? Sounds a little scary to me! :D
chinkgai - Monday, February 7, 2011 - link
hey guys,i noticed the first chart at the top of your first page has 1ghz for the cpu speed all the way down when the paragraph right below it states that it the 3d receives a 200mhz bump.
nice article btw.
chinkgai - Monday, February 7, 2011 - link
i meant fourth pageAnand Lal Shimpi - Monday, February 7, 2011 - link
Thanks, fixed :)Take care,
Anand
ssj4Gogeta - Tuesday, February 8, 2011 - link
Anand,Font changing is also available on my Galaxy S i9000. Samsung ships the phone with 3 extra fonts and you can download more from the Market by searching for "FlipFont".
mongo lloyd - Monday, February 7, 2011 - link
The review, as far as I could tell, didn't mention which audio chip was in the phone, or have any audio fidelity benchmarks for that matter.A top-of-the-line Wolfson chip, like the Galaxy S I hope?
Brian Klug - Monday, February 7, 2011 - link
It's got a Wolfson WM8994 - audio sounds great. I'll definitely consider audio benchmarks as well going forward.-Brian
djgandy - Monday, February 7, 2011 - link
It'll be interesting to see how all the other SoC's perform with DDR2.DanNeely - Monday, February 7, 2011 - link
Where can I find more information on this?Anand Lal Shimpi - Monday, February 7, 2011 - link
Here's a link to the immediate mode vs. tbdr discussion in our old Kyro II review:http://www.anandtech.com/show/735/2
Take care,
Anand
silverblue - Monday, February 7, 2011 - link
I'm not sure I agree with the wording in this article about TBDR. The reason PowerVR didn't need to slap DDR RAM onto the Kyro II cards is because they simply didn't need it thanks to the reduction of traffic that comes from deferred rendering. The unknown element at the time was hardware T&L because it simply wasn't available and was thought to be impossible, however as this is yet again being performed on-die, wouldn't that also result in a marked reduction in traffic? Might need some clarification on this one.I've never seen it confirmed that the Adreno GPU performs TBDR; some clarification would be appreciated on this one as well! :)
Onto the option for changing fonts... my vendor-agnostic Galaxy S has such a feature called "Font style" under the Display settings, allowing you to choose the "Default font", "Choco cooky", "Cool jazz" and "Rosemary", with the option of getting more online.
Exophase - Monday, February 7, 2011 - link
Qualcomm bought out AMD's mobile GPUs and hence the Adreno 200 was a rebrand of AMD z430. Here's a little more background on the tiling nature of z430:(since apparently I can't post a link without being flagged as spam just google for this: gdc2008_ribble_maurice_TileBasedGpus.pdf - it's the first hit)
silverblue - Monday, February 7, 2011 - link
Interesting... I'd like to see the differences between their approach and that of Imagination Technologies.silverblue - Monday, February 7, 2011 - link
Thanks by the way... answered a lot of questions. :)AndroidFan - Monday, February 7, 2011 - link
should be 300mhz (=600/2)Zaitsev - Monday, February 7, 2011 - link
Is the camera really too thick to fit the width of the phone? I heard of many Evo users who have cracked the glass covering. While I haven't had this problem, it makes me wonder if it's really necessary in the first place.Thanks
MeSh1 - Monday, February 7, 2011 - link
I cant wait until you can wirelessly shoot your phones display to your tv ala intel WiDi. This HDMI out is cool, but the cable kind of kills it. With wireless display your phone becomes a game controller :) or a remote when shooting movies to your tv. Plus how cool would it be if your phone can fetch movies from your home network and you shot the playback to your TV. The Sony NGP should have implemented this. Ah well.GoodRevrnd - Tuesday, February 8, 2011 - link
TV link would be awesome, but why would you need the phone to bridge the TV and network??aegisofrime - Monday, February 7, 2011 - link
May I suggest x264 encoding as a test of the CPU power? There's a version of x264 available for ARM chips, along with NEON optimizations. Should be interesting!Shadowmaster625 - Monday, February 7, 2011 - link
What is the point in having a high performance video processor when you cannot do the two things that actually make use of it? Those two things are: 1. Watch any movie in your collection without transcoding? (FAIL) 2. Play games. No actual buttons = FAIL. If you think otherwise then you dont actually play games. Just stick with facebook flash trash.TareX - Wednesday, February 9, 2011 - link
The only reason I'd pay for a dual core phone is smooth flash-enabled web browsing, not gaming.zorxd - Monday, February 7, 2011 - link
Stock Android has it too. There is also E for EDGE and G for GPRS.Exophase - Monday, February 7, 2011 - link
Hey Anand/Brian,There are some issues I've found with some information in this article:
1) You mention that Cortex-A8 is available in a multicore configuration. I'm pretty sure there's no such thing; you might be thinking of ARM11MPCore.
2) The floating point latencies table is just way off for NEON. You can find latencies here:
http://infocenter.arm.com/help/index.jsp?topic=/co...
It's the same in Cortex-A9. The table is a little hard to read; you have to look at the result and writeback stages to determine the latency (it's easier to read the A9 version). Here's the breakdown:
FADD/FSUB/FMUL: 5 cycles
FMAC: 9 cycles (note that this is because the result of the FMUL pipeline is then threaded through the FADD pipeline)
The table also implies Cortex-A9 adds divide and sqrt instructions to NEON. In actuality, both support reciprocal approximation instructions in SIMD and full versions in scalar. The approximation instructions have both initial approximation with ~9 bits of precision and Newton Rhapson step instructions. The step instructions function like FMACs and have similar latencies. This kind of begs the question of where the A9 NEON DIV and SQRT numbers came from.
The other issue I have with these numbers is that it only mentions latency and not throughput. The main issue is that the non-pipelined Cortex-A8 FPU has throughput almost as bad as its latency, while all of the other implementations have single cycle throughput for 2x 64-bit operations. Maybe throughput is what you mean by "minimum latency", however this would imply that Cortex-A9 VFP can't issue every cycle, which isn't the case.
3) It's obvious from the GLBenchmark 2.0 Pro screenshot that there are some serious color limitations from Tegra 2 (look at the woman's face). This is probably due to using 16-bit. IMG has a major advantage in this area since it renders at full 32-bit (or better) precision internally and can dither the result to 16-bit to the framebuffer, which looks surprisingly similar in quality to non-dithered 32-bit. This makes a 16-bit vs 16-bit framebuffer comparison between the two very unbalanced - it's far more fair to just do both at 32-bit, but it doesn't look like the benchmark has any option for it. Furthermore, Tegra 2 is limited to 16-bit (optionally non-linear) depth buffers, while IMG utilizes 32-bit floating point depth internally. This is always going to be a disadvantage for Tegra 2 and is definitely worth mentioning in any comparison.
Finally I feel like ranting a little bit about your use of the Android Linpack test. Anyone with a little common sense can tell that a native implementation of Linpack on these devices will yield several dozen times more than 40MFLOPS (should be closer to 1-4 FLOP/CPU cycle). What you see here is a blatant example of Dalvik's extreme inability to perform with floating point code that extends well beyond an inability to perform SIMD vectorization.
metafor - Monday, February 7, 2011 - link
According to the developer of Linpack on Android:http://www.greenecomputing.com/category/android/
It is mostly FP64 calculations done on Dalvik. While this may not be the fastest way to go about doing linear algebra, it is a fairly good representation of relative FP64 performance (which only exist in VFP).
And let's face it, few app developers are going to dig into Android's NDK and write NEON optimized code.
Exophase - Monday, February 7, 2011 - link
Then let's ask this instead: who really cares about FP64 performance on a smartphone? I'd also argue that it is not even a good representation of relative FP64 performance since that's being obscured so much by the quality of the JITed code. Hence why you see Scorpion and A9 perform a little over twice as fast as A8 (per-clock) instead of several times faster. VFP is still in-order on Cortex-A9, competent scheduling matters.Maybe a lot of developers won't write NEON code on Android, but where it's written it could very well matter. For one thing, in Android itself. And theoretically one day Dalvik could actually be generating NEON competently.. so some synthetic tests of NEON could be a good look at what could be.
metafor - Monday, February 7, 2011 - link
Well, few people really :)Linpack as it currently exists on Android probably doesn't tell very much at all. But if you're just going to slap together an FP heavy app (pocket scientific computing anyone?) and aren't a professional programmer, this likely represents the result you see.
I wouldn't mind seeing SpecFP ported natively to Android and running NEON. But alas, we'd need someone to roll up their sleeves and do that.
I did do a native compile of Linpack using gcc to test on my Evo, though. It's still not SIMD code, of course, but native results using VFP were around the 70-80MFLOPS mark. Of course, it's scheduling for the A8's FPU and not Scorpion's.
Anand Lal Shimpi - Monday, February 7, 2011 - link
Thanks for your comment :)1) You're very right, I was thinking about the ARM11 - fixed :)
2) Make that 2 for 2. You're right on the NEON values, I mistakenly grabbed the values from the cycles column and not the result column. The DIV/SQRT columns were also incorrect, I removed them from the article.
I mentioned the lack of pipelining in the A8 FPU earlier in the article but I reiterated it underneath the table to hammer the point home. I agree that the lack of pipelining is the major reason for the A8's poor FP performance.
3) Those screenshots were actually taken on IMG hardware. IMG has some pretty serious rendering issues running GLBenchmark 2.0.
4) I'm not happy with the current state of Android benchmarks - Linpack included. Right now we're simply including everything we can get our hands on, but over the next 24 months I think you'll see us narrow the list and introduce more benchmarks that are representative of real world performance as well as contribute to meaningful architecture analysis.
Take care,
Anand
Exophase - Monday, February 7, 2011 - link
Thanks Anand.I'm surprised to hear that shot was from IMG, given that it was an IMG employee who made the comment originally about Tegra's 16-bit banding being evident on it, from the screenshot. Whoops. I do wonder what could be causing this, then.
Nonetheless, while that definitely makes my 16bit color claim invalid the depth buffer one should still hold. We might need to wait and see how much of a difference this actually makes, or rather how effective nVidia's 16-bit depth space is.
I'm glad to hear that you're as concerned about benchmarks on Android as I am. It's especially frustrating when I see people using them to try to indicate Atom being substantially better clock for clock than Cortex-A9.
Exophase - Monday, February 7, 2011 - link
Managed to miss this:"The test ramps from around 3k vertices to 15k vertices per frame, and 190k to 250k triangles per frame"
That line doesn't make any sense. How would you have hundreds of times more triangles than vertices? You must have meant something else.
sid1712 - Monday, February 7, 2011 - link
Great review as usual but i'm disappointed about the lack of details on the Sound Quality of the phone. A comparison of the sound quality (via headphone jack) alongside the iPhone 4 and the Galaxy S (with Voodoo kernel preferrably) would give a good idea about the SQ of the phone.ScentedKandle - Monday, February 7, 2011 - link
Related to this, the audio codec lists "lossless" but doesn't mention what format. Can the audio chip natively decode FLAC?teldar - Monday, February 7, 2011 - link
The order of buttons if the same as my droid x.Pjotr - Monday, February 7, 2011 - link
Does it really record 1920x1088? Does this unorthodox resolution play well on TVs, if you put it on a USB stick, for example?Brian Klug - Monday, February 7, 2011 - link
It plays back from the phone properly, and most of the playback software just does a crop. A ton of devices actually produce 1088 and don't make note of it, it should playback fine.-Brian
unmesh - Monday, February 7, 2011 - link
For active aka switching transistor power consumption, C*V^2*f (C is capacitance and f is frequency) is a better proxy than V^2/R.The conclusion that operating voltage has a huge effect remains the same.
Kevin098 - Monday, February 7, 2011 - link
Hey, can you make a video comparison between the iphone 4 retina display and Optimus 2x ?StormyParis - Monday, February 7, 2011 - link
Pages and pages of (apparently not very acurate, too) perf data, and not even one line on sound quality, which is one of my key buying points for a phone.No info on whether I'll be able to stream PC-resolution videos off my server to my bed over wifi.
Overall, not a very useful review. More like a dick size contest.
SoulShadow - Monday, February 7, 2011 - link
Brian,Do you have any idea if that audio chip will make it into all future Tegra 2 phones, or is this just a lucky phone? Droid Bionic with a Wolfson DAC would make me a happy camper.
Brian Klug - Monday, February 7, 2011 - link
Those specific choices are ultimately left up to the handset makers. It seems to be a relatively common component to select however. I have no idea whether the Bionic or Atrix are getting that same audio IC.We can definitely run RightMark audio analyzer on phones from now on. ;)
-Brian
LostPassword - Monday, February 7, 2011 - link
I travel. Limited bands is disappointing. Waiting for comparisons to HTC dual core. Screw Motorola.HAVM - Monday, February 7, 2011 - link
There is one SoC missing in the 2011 roadmap...Nokia and likely also Sony Ericsson will start to ship smartphones based on the ST-Ericsson U8500 platform in the second half of 2011.
softdrinkviking - Tuesday, February 8, 2011 - link
is that going to be dual-core, or competitive with the tegra/omap/scorpion level socs?kenyee - Monday, February 7, 2011 - link
This is probably the most comprehensive Optimus 2x review I've seen yet, so big kudos.What still bugs me about phone reviews is they don't seem to measure how well GPS receivers work in the phones. Even a simple stick it outside at the same location and see how many satellites show up and how quickly (GPSStatus has a feature that measures time to first fix) would help.
E.g., my crappy Moto Cliq sometimes flakes out and even if it claims it sees 10 satellites, it never gets a fix. It's important for those of us who are using our phones for navigation...
strikeback03 - Tuesday, February 8, 2011 - link
There is a whole section titled "GPS Performance" on page 9, basically comes down to that the hardware seems fine but since it is still preproduction A-GPS wasn't set up.vision33r - Monday, February 7, 2011 - link
Is the iPhone 4, all the other phones all had bulges, plastic, and weird ergo shapes.No wonder the iPhone 4 is sought after more, metal, optical glass, highest resolution, and pixel density count.
azagall - Monday, February 7, 2011 - link
Good performance but the back of the phone is the ugliest I have seen in a while ... And why the plastic ... Isn't this supposed to be a premium phone?texanstc - Monday, February 7, 2011 - link
It looks like I'm not missing much with the LG 2X.rpmrush - Monday, February 7, 2011 - link
Solid review, but please at least use spell check. I'm not a grammar or typo freak, but there were way too many simple typos that spell check wouldn't even let you get by with. At least have someone proof read it before you publish to the public.zowie - Tuesday, February 8, 2011 - link
who can create a new type battery, who will be the richest man in the worlduhuznaa - Tuesday, February 8, 2011 - link
Yeah, and until then those who manage to come up with some decent power management will be the richest...Seriously, every improvement on the battery front almost always just leads to devices drawing more power. It's somewhat ironic that last year's iPhone still leads the pack when it comes to battery life. Power management (that is: don't draw more power than absolutely necessary by throttling or shutting down components that aren't needed or aren't fully needed in a given moment) is hard and boring design work nobody seems to care for. And with devices and software getting replaced with the next iteration every few months this is even understandable, it's just not worth the effort, especially when nobody seems to care and benchmarks are so much more important to the crowd.
DanNeely - Tuesday, February 8, 2011 - link
How is is typically played back: Cropped, or vertically resampled?Wilco1 - Tuesday, February 8, 2011 - link
Tegra 3 has 4 1.5GHz Cortex-A9's according to a leaked slide.That was a great article! A few minor corrections: The ARM11 VFP is fully pipelined (so it can beat the A8 on FP performance). Like the A8, Scorpion is 2-way in-order, not limited out-of-order. In-order cores issue instructions in-order but may complete them out-of-order. On the other hand, OoO cores use register renaming to issue instructions out-of-order but complete them in-order.
Note none of the micro benchmarks used emits Neon instructions. JIT compilers don't have enough time to generate high quality code, let alone autovectorize! For proper benchmarking you will need to run native code compiled with a quality compiler (not GCC - it is still far behind the state of the art on ARM, especially Thumb-2).
metafor - Tuesday, February 8, 2011 - link
I would argue with that definition of OoO. A design does not need register renaming in order to issue any arbitrary instruction OoO. It's simply a trade-off of whether to centralize hazard tracking on register accesses or on retirement.PWRuser - Tuesday, February 8, 2011 - link
Excellent review. Please, in your future reviews don't stop including gems like this one:"Generally while browsing I can feel when Flash ads are really slowing a page down - the 2X almost never felt that way."
That's what matters! Including hands on observations along with a full volley of synthetic benchmarks.
This review comes as close as humanly possible to portraying a handset's ability to readers without the said readers trying it out.
Your attention to detail puts other reviews to shame. Keep up the good work.
sarge78 - Tuesday, February 8, 2011 - link
Don't forget about ST-Ericsson's U8500 A9. They could be a major player in 2011/2012 with potential design wins from Nokia and Sony Ericsson.warisz00r - Tuesday, February 8, 2011 - link
What equipments do you use to test the phone's audio quality with?phut- - Tuesday, February 8, 2011 - link
"NVIDIA tells us that the Tegra 2 SoC is fully capable of a faster capture rate for stills and that LG simply chose 2MP as its burst mode resolution. For comparison, other phones with burst modes capture at either 1 MP or VGA. That said, unfortunately for NVIDIA, a significant technological advantage is almost meaningless if no one takes advantage of it. It'll be interesting to see if the other Tegra 2 phones coming will enable full resolution burst capture. unfortunately for NVIDIA, a significant technological advantage is almost meaningless if no one takes advantage of it. It'll be interesting to see if the other Tegra 2 phones coming will enable full resolution burst capture. meaningless if no one takes advantage of it. It'll be interesting to see if the other Tegra 2 phones coming will enable full resolution burst capture."LG have probably made this decision based on the sensitivity of the invariably minuscule sensor they will have used. Having 6 frames of 12mp is pointless if they are 12 incomprehensible megapixels due to the lacklustre sensitivity of the pixels in their chosen part.
The kind of sensor you find delivering a meaningful burst in something like a 5D mk2 is enormous and power hungry, in comparison to an operating environment such as a phone.
matt b - Tuesday, February 8, 2011 - link
Just curious because I've heard rumors that HP will use the Qualcomm chipset and I've also heard rumors that they will stick with Ti for their new tablets/phones. I just wondered if you know for sure since I know that you met with folks at CES. I hope that we all find out tomorrow at the HP event.Great review.
TareX - Wednesday, February 9, 2011 - link
I'd like to see Tegra 2 on the Xoom compared to Tegra 2 on the Optimus 2X.Why? Well, simply put, the only Android version that seems to be optimized for dual-core is Honeycomb.
Dark Legion - Wednesday, February 9, 2011 - link
Why is there no Incredible on 2.2? I could understand if you had both 2.1 and 2.2, like the Evo, but as it is now does not show the full/current performance.Morke - Thursday, February 10, 2011 - link
"It’s a strange dichotomy that LG sets up with this launcher scheme that divides “downloaded” apps from “system applications,” one that’s made on no other Android device I’ve ever seen but the Optimus One. The end result is that most of the stuff I want (you know, since I just installed it) is at the very last page or very bottom of the list, resulting in guaranteed scrolling every single time. If you’re a power user, just replace the launcher with something else entirely."You are not right there.
First you can create additonal categories (aside from system applications and downloads) and move applications between them.
Secondly you can rearrange the ordering of the applications inside a category (allowing you to have those on top which you access most frequently). You can also delete applications right away in this edit mode.
There is a youtube video demonstrating this:
http://www.youtube.com/watch?v=Dvvtl6pSNp8
See time index starting with 4:21.
Maybe you should correct your review on this?
Morke - Thursday, February 10, 2011 - link
The correct youtube URL demonstrating application launcher management is actuallyhttp://www.youtube.com/watch?v=lDo-1-jwLko&fea...
brj_texas - Thursday, February 10, 2011 - link
Anand,A question on the statement in the benchmarking section, "the SunSpider benchmark isn't explicitly multithreaded, although some of the tests within the benchmark will take advantage of more than one core. "
My understanding was that all of the tests within sunspider are single-threaded, but a dual-core processor can run the javascript engine (and the sunspider tests) in a separate thread from the main browser thread when you call sunspider from a browser window.
Can you clarify which tests support multi-threading in sunspider if that is in fact what you meant?
On the topic of multi-threading, we've used moonbat, a multi-core variant of sunspider, to explicitly test multi-core performance with javascript code. I wonder if you have any other benchmarks under investigation that measure multi-core performance?
Thanks
-Brian
worldbfree4me - Saturday, February 12, 2011 - link
Thanks for another thorough and in-depth analysis. But I have a question to ask,Should we upgrade (break our 2 year contract agreement for this phone) or ride out our contract?
We trust and value your our opinion. Tom’s hardware does a GPU hierarchy chart every few months, can you do a phone hierarchy in the future?
lashton - Sunday, February 13, 2011 - link
They have a really good idea and lead the market but it falls short because its not quite righttnepres - Tuesday, April 5, 2011 - link
I now own a optimus 2x. The first was dead on arrival, but this one is perfect. The LG software is innovative and pleasing to the eye. In various places they made real improvements to the UI that are just brilliant,ie. the ability to sort and categorize apps. At times the UI is not as fast as you would expect, especially when adding apps/widgets to one of the 7 pages. It seems LG generates a list of widgets for you, so you can see what apps support this mode, and that takes about a second. As I recall, on HTC devices you are just presented with a list of apps and u have to try and see if you can widget it.The LG keybord has a brilliant feature, you tab the side of the phone to move the cursor. Sadly in other respects the keyboard is lacking, ie. when you long-pres you do not get the alternates you might wish, such as numbers.
The batterytime is superb, using the UI consumes much less power than on my desire.
Copy/paste in the browser does not activate via long-pres, you have to hit menu button, but on the plus side its easier to use than what HTC made.
During 2 days of very intensive use i have had 1 app (partially) crash and that was the marketplace. No other issues so far, its my verdict that the unstability issues are overrated.
No problems with wifi using stock ISP (TDC) supplied router. (sagemcom)
To engadget: How on earth (!!?!!?) can you state there is no use for dualcore. When browsing one loads flash the other the rest. Its so fast you cant believe it. Try loading www.ufc.com on a non dualcore phone and you get my drift.
I do not hesitate to give the optimus 2x my warm recommendations.
VERDICT: 9/10 (missing 4g)
Sannat - Thursday, May 12, 2011 - link
gsmarena sound benchmark for optimus 2x isnt great...could it be a s/w issue...??dendysutrisna - Saturday, August 6, 2011 - link
Very nice article ... and I'm just a layman about this kind of information technology. But maybe someday I should know about complicated things like the above. I can only make the review at a glance (and maybe subjective, try looking at this page: http://www.bestdealscomputers.net) on products that use chips, such as NVIDIA, on a tablet computers.Move on .....
dendysutrisna - Saturday, August 6, 2011 - link
Sory I mean at here: http://www.bestdealscomputers.netdendysutrisna - Saturday, August 6, 2011 - link
So Smartphone can do all that can be done of a tablet computer? The only restriction the size of the screen just ... look at the specifications on the following pages ... http://www.bestdealscomputers.net/tablets/motorola...Wow ....
mohamedbhimji - Tuesday, September 13, 2011 - link
An interesting story, I recently purchased a HTC Panache (aka MyTouch 4G) and was really disappointed by the battery life when compared to my wifes iPhone, my iPad and my work BlackBerry Torch - but looking at the stats here it appears that it's about average when it comes to Android devices and other phones.I am still looking at getting a higher capacity battery.
sarah sm. - Tuesday, October 4, 2011 - link
Optimus 2X uses Duel Core so the speed is very fast. When playing a game with my friend, Optimus 2X loads faster than his smartphone. If users with an LG smartphone want to download various apps or games, I recommend to visit uk.lgworld.com