Alleged 'A12' benchmark for 2018 iPhone with 4GB RAM pops up

24

Comments

  • Reply 21 of 77
    wizard69 said:
    Sadly if Apple doesnt lower proces significantky i wont be seeing an iPhone in my future.  
    Lower than $350? Góôd lück.
    StrangeDays
  • Reply 22 of 77
    mjtomlinmjtomlin Posts: 1,821member
    bsimpsen said:
    At 8-11% faster for the CPU, I think this would be the smallest generational improvement in the A-family since its introduction. Eventually Apple will run up against the wall or simply not need much additional horsepower, but I was expecting bigger gains, particularly in light of the rumored ditching of x86 in the next couple years.

    The "A-family" is not a CPU. It's an SoC. There are many, many different areas that "performance" can be gained. These benchmarks test very specific areas of a system... CPU, GPU, memory and storage. They do not account for targeted processing units, such DSP, ISP, AIP, etc. Apple's macOS was the first to move far away from CPU and towards GPU for processing in their OS. Benchmarks can only test what they're looking for and not what the system is actually doing. It does not matter how close the competition "catches up" to these specs... Apple optimizes their software for their hardware and no benchmarks can reliably test that against another platform.
    edited July 2018
  • Reply 23 of 77
    KITAKITA Posts: 163member
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
  • Reply 24 of 77
    tmaytmay Posts: 3,567member
    mjtomlin said:
    bsimpsen said:
    At 8-11% faster for the CPU, I think this would be the smallest generational improvement in the A-family since its introduction. Eventually Apple will run up against the wall or simply not need much additional horsepower, but I was expecting bigger gains, particularly in light of the rumored ditching of x86 in the next couple years.

    The "A-family" is not a CPU. It's an SoC. There are many, many different areas that "performance" can be gained. These benchmarks test very specific areas of a system... CPU, GPU, memory and storage. They do not account for targeted processing units, such DSP, ISP, AIP, etc. Apple's macOS was the first to move far away from CPU and towards GPU for processing in their OS. Benchmarks can only test what they're looking for and not what the system is actually doing. It does not matter how close the competition "catches up" to these specs... Apple optimizes their software for their hardware and no benchmarks can reliably test that against another platform.
    I'll wait for more information, especially die area, as an indicator of where Apple is going with the A Series for the iPhone. Lots of complexity that isn't necessarily going to be measured by Geekbench.

    I'd like to see a final bifurcation of the A Series to split off the iPad SOC as it's own architecture specifically optimized for an increased TDP.
  • Reply 25 of 77
    wizard69wizard69 Posts: 12,720member
    Apple never designs for spec performance.   They ignore it in favor of user experience. 

    I wouldn't be at all surprised if they use this year to perfect, stabilize and fine tune the A series processors just as they are doing with iOS.   That doesn't mean they aren't delivering.  It just means it won't be as immediately visible, glitzy and flashy.
    I think what people are missing here are the advances in the GPU.   People often underestimate the importance of the GPU in modern software and operating systems.   Also it kinda explains Apples WWDC focus over the last couple of years on GPU compute via Metal.

    Given that I'm not sure why people are upset that they are only getting 10% that is actually not that bad given similar clock rates.
  • Reply 26 of 77
    wizard69wizard69 Posts: 12,720member
    melgross said:
    wizard69 said:
    melgross said:
    Rayz2016 said:
    bsimpsen said:
    At 8-11% faster for the CPU, I think this would be the smallest generational improvement in the A-family since its introduction. Eventually Apple will run up against the wall or simply not need much additional horsepower, but I was expecting bigger gains, particularly in light of the rumored ditching of x86 in the next couple years.
    You’re assuming that Apple will put the fastest chip they can make in the phone? If it has more memory and a bigger screen then they might step it down to preserve battery life. The same wouldn’t apply to a processor fitted inside an iPad or a laptop since they have much larger batteries. 
    Yes, we do. Unless Apple has totally changed what they’re doing, we should be expecting at least a 25% boost for the individual core, and at least a 40% boost for the GPU cores. Of course, we don’t know what else Apple may have done for specialized purposes, but this would be disappointing.
    I would t call these numbers disappointing until we know more, even then these are bigger gains than Intel is seeing from year to year.  

    Things to consider:  


    1.  The cores could very well be 3 + 3 that is geekbench may be confused as to the number of high performance cores. 

    2.   The GPU resilts are fairly impressive and are supportted with sound growth in compute.   This is a very good thing in my mind because GPU performance is critical for midern apps.  

    3.   In all likelyhood a prototype would be running heavily instrumented code slowing the benchmarks down.  

    4.   There is good reason for Apple to spend engineering time on other sections of the SoC.   They could have easily place priority on the GPU, AI/ML hardware, and other sections that have or are starting to see increased importance.   Im as guilty as the next guy when it comes to looking at CPU performance first but the reality is that the CPU is only part of the game in modern hardware.   Iphone couldnt even exist without the capabilities of the GPU and that applies to many apps.  

    5.   As others have suggested Apple may have heard from the sheep and decided that battery life is important.    This could be a huge win for some users as the process shrink is rumored to do very well when it comes to power.  

    Seriously im going to wait for real figures.  Sadly if Apple doesnt lower proces significantky i wont be seeing an iPhone in my future.  
    You know, I’m just going by the claimed results here, as is everybody else. If these results are correct, then it’s disappointing, unless Apple has something else happening in the chip that more than makes up or it, but otherwise, yes, it’s disappointing. We can’t assume that Apple has done other things to make this performance seem ok, because that just wildly guessing at this point. I’m just hoping that these tests aren’t the real thing. But in previous years, results coming out about now have been pretty close to actuality.

    lets not stick our heads in the sand folks.
    I don't think we are sticking our heads in the sand, rather I see it as focusing on what has advanced strongly and that is the GPU.   The GPU performance can make a huge difference in many apps, where it doesn't the A11 is already fast enough in many cases.

    I will not be disappointed until we see what the entire package is like.   If Apples goal was a significantly extended battery life over the A11 based phones I'd be all for it.   Performance can be measured various ways and frankly I see a lot of good in these numbers.   It is too bad a power profile doesn't get posted with these benchmark numbers.
  • Reply 27 of 77
    melgrossmelgross Posts: 31,475member
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    igohmmm
  • Reply 28 of 77
    wizard69wizard69 Posts: 12,720member
    wizard69 said:
    Sadly if Apple doesnt lower proces significantky i wont be seeing an iPhone in my future.  
    Lower than $350? Góôd lück.
    I'm thinking here of any phone supporting the A12 SoC.   I'm simply not willing to offer Apple such large margins for its top of the line hardware.    Last year Apple effectively tried to rais the prices on the entire line up by $100 and that is just plain greedy.
  • Reply 29 of 77
    melgrossmelgross Posts: 31,475member
    wizard69 said:
    melgross said:
    wizard69 said:
    melgross said:
    Rayz2016 said:
    bsimpsen said:
    At 8-11% faster for the CPU, I think this would be the smallest generational improvement in the A-family since its introduction. Eventually Apple will run up against the wall or simply not need much additional horsepower, but I was expecting bigger gains, particularly in light of the rumored ditching of x86 in the next couple years.
    You’re assuming that Apple will put the fastest chip they can make in the phone? If it has more memory and a bigger screen then they might step it down to preserve battery life. The same wouldn’t apply to a processor fitted inside an iPad or a laptop since they have much larger batteries. 
    Yes, we do. Unless Apple has totally changed what they’re doing, we should be expecting at least a 25% boost for the individual core, and at least a 40% boost for the GPU cores. Of course, we don’t know what else Apple may have done for specialized purposes, but this would be disappointing.
    I would t call these numbers disappointing until we know more, even then these are bigger gains than Intel is seeing from year to year.  

    Things to consider:  


    1.  The cores could very well be 3 + 3 that is geekbench may be confused as to the number of high performance cores. 

    2.   The GPU resilts are fairly impressive and are supportted with sound growth in compute.   This is a very good thing in my mind because GPU performance is critical for midern apps.  

    3.   In all likelyhood a prototype would be running heavily instrumented code slowing the benchmarks down.  

    4.   There is good reason for Apple to spend engineering time on other sections of the SoC.   They could have easily place priority on the GPU, AI/ML hardware, and other sections that have or are starting to see increased importance.   Im as guilty as the next guy when it comes to looking at CPU performance first but the reality is that the CPU is only part of the game in modern hardware.   Iphone couldnt even exist without the capabilities of the GPU and that applies to many apps.  

    5.   As others have suggested Apple may have heard from the sheep and decided that battery life is important.    This could be a huge win for some users as the process shrink is rumored to do very well when it comes to power.  

    Seriously im going to wait for real figures.  Sadly if Apple doesnt lower proces significantky i wont be seeing an iPhone in my future.  
    You know, I’m just going by the claimed results here, as is everybody else. If these results are correct, then it’s disappointing, unless Apple has something else happening in the chip that more than makes up or it, but otherwise, yes, it’s disappointing. We can’t assume that Apple has done other things to make this performance seem ok, because that just wildly guessing at this point. I’m just hoping that these tests aren’t the real thing. But in previous years, results coming out about now have been pretty close to actuality.

    lets not stick our heads in the sand folks.
    I don't think we are sticking our heads in the sand, rather I see it as focusing on what has advanced strongly and that is the GPU.   The GPU performance can make a huge difference in many apps, where it doesn't the A11 is already fast enough in many cases.

    I will not be disappointed until we see what the entire package is like.   If Apples goal was a significantly extended battery life over the A11 based phones I'd be all for it.   Performance can be measured various ways and frankly I see a lot of good in these numbers.   It is too bad a power profile doesn't get posted with these benchmark numbers.
    As I said, I’m just goi g by what we see here, and that’s not impressive. We can’t talk about chip feature improvements that we know nothing about.i would have thought that moving to 7nm, with the large claimed advantages, would have given us at least a 25% improvement, still less than before, but nevertheless significant.

    until Apple announces the new phones, and details the new chip for us, we can just look at this, unless some other test comes out. I’d hope, after seeing this, that when they do announce it, we’ll be saying wow! I never expected this!
  • Reply 30 of 77
    KITAKITA Posts: 163member
    melgross said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    That is from Anandtech.

    This is also from Anandtech:

    The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle.

    - Andrei F.

    edited July 2018 muthuk_vanalingam
  • Reply 31 of 77
    irelandireland Posts: 17,547member
    sflocal said:
    ireland said:
    Battery life gains are everything people want, at this stage. iPad Pro would be 15+ hours on light use would be ideal.
    People who buy the iPad Pro aren't "light" users for sure... 
    Ok, but it was an example. "Two days I used my iPad Pro light and got 15 hours, and yesterday I used it quite heavy and got over 9 hours". That'd be awesome.
  • Reply 32 of 77
    I wouldn’t worry about a leaked benchmark this early. The A10X leaked benchmark showed a multi core score of 6,600 compared to around 9,300 in shipping versions. That’s a huge difference. There was also a leak for the iPhone 8 at 4,500 single (300 points faster than actual) and 8,900 multi (over 1,000 points slower than actual).

    The most accurate leak I recall was the iPhone X last year. But that leak was right before the keynote, not months in advance.
    smalm
  • Reply 33 of 77
    tmaytmay Posts: 3,567member
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    That is from Anandtech.

    This is also from Anandtech:

    The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle.

    - Andrei F.

    You took that "thermally constrained" out of context:

    "When we’re looking at competitor devices we see only the the iPhone X able to compete with the last generation Snapdragon 835 devices – however with a catch. The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle. Unfortunately this also applies to current and last generation Exynos and Kirin SoCs as both shed great amount of performance after only a few minutes. I’ve addressed this issue and made a great rant about it in our review of the Kirin 970. For this reason going forward AnandTech is going to distinguish between Peak and Sustained scores across all 3D benchmarks. This however needs to be tested on commercial devices as the QRD platform isn’t a thermally representative phone for the SoC, so until that happens, we’ll have to just estimate based on power consumption where the Snapdragon 845 ends up.

    This article was a preview on the Snapdragon 845, not a device test. 

    Even at that, the only deficiency that the A11 had was the 3Dmark tests, which actually weren't that far off of the 835's.


     
    edited July 2018 igohmmm
  • Reply 34 of 77
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    That is from Anandtech.

    This is also from Anandtech:

    The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle.

    - Andrei F.



    Always hilarious when people intentionally leave something out. This is literally the next sentence after your quote:

    Unfortunately this also applies to current and last generation Exynos and Kirin SoCs as both shed great amount of performance after only a few minutes.”

    Curious why you’d leave that bit out. Never mind, I already know. For the same reason you quoted a Physics score and left out the numerous other benchmarks. Gotta cherry pick the right quotes to suit your narrative.
    radarthekatigohmmm
  • Reply 35 of 77
    KITAKITA Posts: 163member
    tmay said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    That is from Anandtech.

    This is also from Anandtech:

    The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle.

    - Andrei F.

    You took that out of context:

    "When we’re looking at competitor devices we see only the the iPhone X able to compete with the last generation Snapdragon 835 devices – however with a catch. The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle. Unfortunately this also applies to current and last generation Exynos and Kirin SoCs as both shed great amount of performance after only a few minutes. I’ve addressed this issue and made a great rant about it in our review of the Kirin 970. For this reason going forward AnandTech is going to distinguish between Peak and Sustained scores across all 3D benchmarks. This however needs to be tested on commercial devices as the QRD platform isn’t a thermally representative phone for the SoC, so until that happens, we’ll have to just estimate based on power consumption where the Snapdragon 845 ends up.

    This article was a preview on the Snapdragon 845, not a device test.

    Show me a device with a Snapdragon 845, in production, that does better than the A11 and doesn't throttle, and I'll give you a "like".

     
    I was never talking about the 845.

    The Snapdragon 835 does better and doesn't throttle nearly as much.

    For the Pixel 2 using OpenGL ES 3.1 API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2907
    Sustained: 2864

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4390
    Sustained: 3591
  • Reply 36 of 77
    tmaytmay Posts: 3,567member
    KITA said:
    tmay said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    That is from Anandtech.

    This is also from Anandtech:

    The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle.

    - Andrei F.

    You took that out of context:

    "When we’re looking at competitor devices we see only the the iPhone X able to compete with the last generation Snapdragon 835 devices – however with a catch. The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle. Unfortunately this also applies to current and last generation Exynos and Kirin SoCs as both shed great amount of performance after only a few minutes. I’ve addressed this issue and made a great rant about it in our review of the Kirin 970. For this reason going forward AnandTech is going to distinguish between Peak and Sustained scores across all 3D benchmarks. This however needs to be tested on commercial devices as the QRD platform isn’t a thermally representative phone for the SoC, so until that happens, we’ll have to just estimate based on power consumption where the Snapdragon 845 ends up.

    This article was a preview on the Snapdragon 845, not a device test.

    Show me a device with a Snapdragon 845, in production, that does better than the A11 and doesn't throttle, and I'll give you a "like".

     
    I was never talking about the 845.

    The Snapdragon 835 does better and doesn't throttle nearly as much.

    For the Pixel 2 using OpenGL ES 3.1 API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2907
    Sustained: 2864

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4390
    Sustained: 3591
    Okay, but what about all of the other tests that the A11 scored better than the 835?
  • Reply 37 of 77
    KITAKITA Posts: 163member
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    That is from Anandtech.

    This is also from Anandtech:

    The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle.

    - Andrei F.



    Always hilarious when people intentionally leave something out. This is literally the next sentence after your quote:

    ”Unfortunately this also applies to current and last generation Exynos and Kirin SoCs as both shed great amount of performance after only a few minutes.”

    Curious why you’d leave that bit out. Never mind, I already know. For the same reason you quoted a Physics score and left out the numerous other benchmarks. Gotta cherry pick the right quotes to suit your narrative.
    Those are regarding the Exynos and Kirin. Not the Snapdragon. 

    The point of my "narrative", as you like to put it, is proving that the A11 does considerably throttle. And in this case, more than a flagship Snapdragon 835.

    If you bothered to read the conversation, you'd know this.
  • Reply 38 of 77
    KITAKITA Posts: 163member
    tmay said:
    KITA said:
    tmay said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    That is from Anandtech.

    This is also from Anandtech:

    The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle.

    - Andrei F.

    You took that out of context:

    "When we’re looking at competitor devices we see only the the iPhone X able to compete with the last generation Snapdragon 835 devices – however with a catch. The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle. Unfortunately this also applies to current and last generation Exynos and Kirin SoCs as both shed great amount of performance after only a few minutes. I’ve addressed this issue and made a great rant about it in our review of the Kirin 970. For this reason going forward AnandTech is going to distinguish between Peak and Sustained scores across all 3D benchmarks. This however needs to be tested on commercial devices as the QRD platform isn’t a thermally representative phone for the SoC, so until that happens, we’ll have to just estimate based on power consumption where the Snapdragon 845 ends up.

    This article was a preview on the Snapdragon 845, not a device test.

    Show me a device with a Snapdragon 845, in production, that does better than the A11 and doesn't throttle, and I'll give you a "like".

     
    I was never talking about the 845.

    The Snapdragon 835 does better and doesn't throttle nearly as much.

    For the Pixel 2 using OpenGL ES 3.1 API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2907
    Sustained: 2864

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4390
    Sustained: 3591
    Okay, but what about all of the other tests that the A11 scored better than the 835?
    We don't have sustained performance tests for most other tests, and for the ones that we do, the A11 throttles hard.

    Hence, we circle back to my original post. Geekbench uses pauses to prevent thermal throttling, so all we see is the peak performance, and not the sustained performance.

    Melgross was claiming that the current A12 scores in the iPhone 11,2 aren't very impressive, but contrary to that, the A12 scores might better represent sustained performance as it's on a new 7nm process. As such, it's likely a considerable improvement over the A11, an SoC that was known to throttle.
    edited July 2018 muthuk_vanalingam
  • Reply 39 of 77
    tmaytmay Posts: 3,567member
    KITA said:
    tmay said:
    KITA said:
    tmay said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    That is from Anandtech.

    This is also from Anandtech:

    The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle.

    - Andrei F.

    You took that out of context:

    "When we’re looking at competitor devices we see only the the iPhone X able to compete with the last generation Snapdragon 835 devices – however with a catch. The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle. Unfortunately this also applies to current and last generation Exynos and Kirin SoCs as both shed great amount of performance after only a few minutes. I’ve addressed this issue and made a great rant about it in our review of the Kirin 970. For this reason going forward AnandTech is going to distinguish between Peak and Sustained scores across all 3D benchmarks. This however needs to be tested on commercial devices as the QRD platform isn’t a thermally representative phone for the SoC, so until that happens, we’ll have to just estimate based on power consumption where the Snapdragon 845 ends up.

    This article was a preview on the Snapdragon 845, not a device test.

    Show me a device with a Snapdragon 845, in production, that does better than the A11 and doesn't throttle, and I'll give you a "like".

     
    I was never talking about the 845.

    The Snapdragon 835 does better and doesn't throttle nearly as much.

    For the Pixel 2 using OpenGL ES 3.1 API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2907
    Sustained: 2864

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4390
    Sustained: 3591
    Okay, but what about all of the other tests that the A11 scored better than the 835?
    We don't have sustained performance tests for most other tests, and for the ones that we do, the A11 throttles hard.

    Hence, we circle back to my original post. Geekbench uses pauses to prevent thermal throttling, so all we see is the peak performance, and not the sustained performance.

    Melgross was claiming that the current A12 scores in the iPhone 11,2 aren't very impressive, but contrary to that, the A12 scores might better represent sustained performance as it's on a new 7nm process. As such, it's likely a considerable improvement over the A11, an SoC that was known to throttle.
    So, you're confident that a couple of synthetic tests of a single aspect of an SOC, the GPU, are a reliable means to predicting the performance of a SOC in general use? 

    edited July 2018 ericthehalfbee
  • Reply 40 of 77
    KITAKITA Posts: 163member
    tmay said:
    KITA said:
    tmay said:
    KITA said:
    tmay said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    KITA said:
    melgross said:
    If true, not an impressive boost from last year, particularly since they say it’s got six cores. I hope this is off.
    The A11 was known to throttle considerably under sustained load.

    Geekbench 4 has pauses built in to avoid thermal throttling.

    These numbers might actually be quite impressive if they're closer to the actual sustained performance of the A12.
    No, it did not. All chips, particularly mobile chips, throttle under load, but the A11 throttles significantly less than any competing chip.
    For the iPhone X using Metal API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2523
    Sustained: 1895

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4428
    Sustained: 2884

    The Snapdragon 835 throttled much less than the A11.
    You should look at the reviews of phones on aRstechnica and anandtech, instead of choosing one spec, which is the one in which Apple doesn’t seem to care much about, which is physics. You need an overal test for this.
    That is from Anandtech.

    This is also from Anandtech:

    The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle.

    - Andrei F.

    You took that out of context:

    "When we’re looking at competitor devices we see only the the iPhone X able to compete with the last generation Snapdragon 835 devices – however with a catch. The A11 is severely thermally constrained and is only able to achieve these scores when the devices are cold. Indeed as seen from the smaller score of the iPhone 8, the SoC isn’t able to sustain maximum performance for even one benchmark run before having to throttle. Unfortunately this also applies to current and last generation Exynos and Kirin SoCs as both shed great amount of performance after only a few minutes. I’ve addressed this issue and made a great rant about it in our review of the Kirin 970. For this reason going forward AnandTech is going to distinguish between Peak and Sustained scores across all 3D benchmarks. This however needs to be tested on commercial devices as the QRD platform isn’t a thermally representative phone for the SoC, so until that happens, we’ll have to just estimate based on power consumption where the Snapdragon 845 ends up.

    This article was a preview on the Snapdragon 845, not a device test.

    Show me a device with a Snapdragon 845, in production, that does better than the A11 and doesn't throttle, and I'll give you a "like".

     
    I was never talking about the 845.

    The Snapdragon 835 does better and doesn't throttle nearly as much.

    For the Pixel 2 using OpenGL ES 3.1 API:

    3DMark Sling Shot 3.1 Extreme Unlimited - Physics (CPU bound):

    Peak: 2907
    Sustained: 2864

    3DMark Sling Shot 3.1 Extreme Unlimited - Graphics (GPU bound):

    Peak: 4390
    Sustained: 3591
    Okay, but what about all of the other tests that the A11 scored better than the 835?
    We don't have sustained performance tests for most other tests, and for the ones that we do, the A11 throttles hard.

    Hence, we circle back to my original post. Geekbench uses pauses to prevent thermal throttling, so all we see is the peak performance, and not the sustained performance.

    Melgross was claiming that the current A12 scores in the iPhone 11,2 aren't very impressive, but contrary to that, the A12 scores might better represent sustained performance as it's on a new 7nm process. As such, it's likely a considerable improvement over the A11, an SoC that was known to throttle.
    So, you're confident that a couple of synthetic tests of a single aspect of an SOC, the GPU, are a reliable means to predicting the performance of a SOC in general use? 

    Yes, the physics test is CPU bound, the graphics test is GPU bound. Regardless of the results compared to other SoCs, it's very easy to see the difference between peak and sustained performance on the A11.

    That, and we also know each one of the A11's big cores uses 3~3.5 W at load. Having all 6 cores at load (2 big and 4 small) would simply not be sustainable in an iPhone. If the A12 brought that TDP down, with potential adjustments to architecture and the move to a 7nm process, we'll likely see better sustained performance.
Sign In or Register to comment.