Anand is still using the Dalvik based Linpack test which is absolutely useless in comparing between different phones as the OS and whatever platform code can make huge differences. I told Brian about StabilityTest's native NDK implementation of Linpack several months ago but they still release reviews with the flawed Linpack.
Maybe they fixed the memory bandwidth problem and licensed the IP to Apple. If they can they should run geekbench on it to see how it compares to the previous chips.
I have the same question as Phelerox. It oddly seems that the "single threaded" mode simply uses half the cores available. In most cases, the performance just scales up to double when "multi threaded" mode is used. This holds true for Exynos quad and Tegra 3 as well.
About the International One S... Is it the krait based international model or the "limited" 1.7 Ghz Scorpion based One S? I ask because it wouldn't make much sense to have an international version and a T mobile version in benchmarks if they were using exactly the same hardware.
Regarding your second question, the international One S is not limited to the 1.7 dual core S3. Only certain international markets are stuck with the 1.7 S3.
Kindly ignore the first part of my statement. I has just woken up and had a brainfart while reading those charts. The performance does scale up linearly from one core to 4 making Phelerox's question even more interesting.
I don't know why single-threaded performance is so high but my guess is the SoC is memory bandwidth-limited once it reaches full utilization of all four cores. It's possible that Qualcomm increased the memory cache sizes in the S4 Pro SoC's and that's why we're seeing the big performance boost in the single-threaded run as well. Just my two cents.
After looking for details on the differences between the 8960 and the 8064 I've decided that that's probably wrong and should be disregarded as such. Now I'm stumped too.
On paper the hardware for this phone looks great. I would consider this phone as my next upgrade unfortunately LG does not have a reputation updating it's software very often. I really hope LG does not lock this phone down and allows the modding community to provide updates to the android stack.
Impressive phone, just a shame about lgs crappy software, the sunspider scores should be record breaking, as should browsermark and velamo...I'm pretty sure this beastie carries a 2mb slab of l2 cache....what effect that would have on performance who knows?
Certainly hands on videos ICS looks gun shot fast...faster than my galaxy s3 that's for sure,
First off, excellent spec sheet. Although some of the specs seem to good to be true.
How does the Adreno 320 compare with rogue PowerVR 6 skews? Could Anandtech do an article on the Adreno 320 compared with other low TDP integrated graphics?
It will be very interesting to compare LG Optimus G specs with the iphone 5 (notice it isn't spelled "i-phone") and OMAP 5 based phones. OMAP 5 might have an advantage in battery life because on lower power mode the two A15 Cortex cores sleep, leaving computational activity to the two Cortex M4 cores.
The A6, if Anand is correct, is an ARM Cortex A9 with added functionality taken from the ARM A15 design. The Exynos 5 uses two vanilla A15 cores, mixed with two A7 cores and combined as big.LITTLE configuration with ARM Mali GPUs. It's a typical ARM design. As long as Samsung doesn't design their own cores like Qualcomm did with Snapdragon and Krait, and now probably Apple also did, they can't steel anything important from Apple.
I think Samsung did use a tweaked A9 licensed from a company Apple later acquired. The other phone manufacturers need to get better gpus asap adreno 320 will still be half as fast as the iphone5 if it doubled its performance. I do not understand why they ignore this part of their Soc designs.
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.
20 Comments
Back to Article
Phelerox - Tuesday, September 18, 2012 - link
Any theories as to why quad core Krait performs almost twice the MFLOPS in single-threaded Linpack that the dual-core Krait pushes?AndreiLux - Tuesday, September 18, 2012 - link
Anand is still using the Dalvik based Linpack test which is absolutely useless in comparing between different phones as the OS and whatever platform code can make huge differences. I told Brian about StabilityTest's native NDK implementation of Linpack several months ago but they still release reviews with the flawed Linpack.toyotabedzrock - Tuesday, September 18, 2012 - link
Maybe they fixed the memory bandwidth problem and licensed the IP to Apple.If they can they should run geekbench on it to see how it compares to the previous chips.
dishayu - Tuesday, September 18, 2012 - link
I have the same question as Phelerox. It oddly seems that the "single threaded" mode simply uses half the cores available. In most cases, the performance just scales up to double when "multi threaded" mode is used. This holds true for Exynos quad and Tegra 3 as well.About the International One S... Is it the krait based international model or the "limited" 1.7 Ghz Scorpion based One S? I ask because it wouldn't make much sense to have an international version and a T mobile version in benchmarks if they were using exactly the same hardware.
OCedHrt - Tuesday, September 18, 2012 - link
Regarding your second question, the international One S is not limited to the 1.7 dual core S3. Only certain international markets are stuck with the 1.7 S3.dishayu - Tuesday, September 18, 2012 - link
Yup. When i said "limited" 1.7 scorpion, i meant limited to certain markets.dishayu - Tuesday, September 18, 2012 - link
Kindly ignore the first part of my statement. I has just woken up and had a brainfart while reading those charts. The performance does scale up linearly from one core to 4 making Phelerox's question even more interesting.cmikeh2 - Tuesday, September 18, 2012 - link
I don't know why single-threaded performance is so high but my guess is the SoC is memory bandwidth-limited once it reaches full utilization of all four cores. It's possible that Qualcomm increased the memory cache sizes in the S4 Pro SoC's and that's why we're seeing the big performance boost in the single-threaded run as well. Just my two cents.cmikeh2 - Tuesday, September 18, 2012 - link
After looking for details on the differences between the 8960 and the 8064 I've decided that that's probably wrong and should be disregarded as such. Now I'm stumped too.cmikeh2 - Tuesday, September 18, 2012 - link
Actually the L2 cache for the APQ8064 is 2 MB compared to the 1 MB for the MSM8960 but I don't know what the effect of that would be.phatboye - Tuesday, September 18, 2012 - link
On paper the hardware for this phone looks great. I would consider this phone as my next upgrade unfortunately LG does not have a reputation updating it's software very often. I really hope LG does not lock this phone down and allows the modding community to provide updates to the android stack.french toast - Tuesday, September 18, 2012 - link
Impressive phone, just a shame about lgs crappy software, the sunspider scores should be record breaking, as should browsermark and velamo...I'm pretty sure this beastie carries a 2mb slab of l2 cache....what effect that would have on performance who knows?Certainly hands on videos ICS looks gun shot fast...faster than my galaxy s3 that's for sure,
1008anan - Tuesday, September 18, 2012 - link
First off, excellent spec sheet. Although some of the specs seem to good to be true.How does the Adreno 320 compare with rogue PowerVR 6 skews? Could Anandtech do an article on the Adreno 320 compared with other low TDP integrated graphics?
It will be very interesting to compare LG Optimus G specs with the iphone 5 (notice it isn't spelled "i-phone") and OMAP 5 based phones. OMAP 5 might have an advantage in battery life because on lower power mode the two A15 Cortex cores sleep, leaving computational activity to the two Cortex M4 cores.
1008anan - Tuesday, September 18, 2012 - link
Is there any way the fab Samsung could steel any A6 tech from Apple? I know this is off topic; and sorry if that offends anyone.UpSpin - Tuesday, September 18, 2012 - link
The A6, if Anand is correct, is an ARM Cortex A9 with added functionality taken from the ARM A15 design.The Exynos 5 uses two vanilla A15 cores, mixed with two A7 cores and combined as big.LITTLE configuration with ARM Mali GPUs. It's a typical ARM design. As long as Samsung doesn't design their own cores like Qualcomm did with Snapdragon and Krait, and now probably Apple also did, they can't steel anything important from Apple.
toyotabedzrock - Tuesday, September 18, 2012 - link
I think Samsung did use a tweaked A9 licensed from a company Apple later acquired.The other phone manufacturers need to get better gpus asap adreno 320 will still be half as fast as the iphone5 if it doubled its performance. I do not understand why they ignore this part of their Soc designs.
Aenean144 - Tuesday, September 18, 2012 - link
Intrinsity was the company that did the circuit design for clocked the Cortex-A8 to 1 GHz in the Apple A4 SoC.Aenean144 - Tuesday, September 18, 2012 - link
Anand said the Apple A6 CPU is a custom ARM core, not a tweaked A9 nor a tweaked A15.mytexel - Tuesday, September 18, 2012 - link
Would be nice to compare to iPhone 5....powerarmour - Tuesday, September 18, 2012 - link
Not a bad phone at all, but bear in mind LG's 'support' or lack thereof...Jellybean?, nah... we won't bother with that