• Introducing XDA Computing: Discussion zones for Hardware, Software, and more!    Check it out!

Snapdragon 800 a battery friendly beast?

Search This thread

Turbotab

Senior Member
May 2, 2011
902
819
I've seen a lot of comments, about the possible battery size in the Nexus 5, which is only ~10% larger than a Nexus 4, which had OK, but not great battery life.

So the N5 is rumoured to have 2300 mAh 3.8v battery, the same size as the HTC One, a device with much better battery life than the N4. But what people are overlooking is the improvements in power efficiency from the Snapdragon 800, although it is a much faster SoC than the S4 Pro in the N4, it actually requires less voltage. This is possible due to TSMC's superior HPM 28nm process, which uses a technology called high-K metal gates, both the Snapdragon 600 & S4 Pro were manufactured on TSMC's 28nm LP process.

But, what if you don't believe me, well just look below. I compare the voltage tables for a S800 vs a S600, we must remember that S600 is an improved S4 Pro, these improvements included higher efficiency (battery life), so it's better than the S4 Pro in the Nexus 4.

28nm LP @ 1.7 GHz = 1075mv -- Snapdragon 600

28nm HPM @ 1.7 GHz = 900mv -- Snapdragon 800

So the S800 uses significantly less power, but achieves much greater performance. There is one outlier, the Adreno 330 GPU it of course is also built on the same superior HPM process, but contains more ALU (compute) units than any Adreno 320, so at max 100% usage may consume slightly more power. In the real world, the CPU/GPU do not sit at 100% all time, and the much faster S800 can complete a task eg load a webpage, faster than the Nexus 4, and then power down, whilst the N4 would still be rendering the webpage. This is a double win, faster ultimate performance means a task can be completed quicker, and the SoC put into low power mode, the fact that S800 needs less voltage in the first place magnifies the advantage.

Another possible power efficiency for the S800, is the on-die modem (baseband), unlike the S4 Pro / S600 which had a separate modem, we don't have hard data for this, but logic suggests that this a more efficient solution.

In conclusion using the 2300 mAh HTC One as a guide, which also has a 1080P display, older Snapdragon 600 and more software bloat than vanilla Android 4.4, I'd wager that the Nexus 5 will easily beat the Nexus 4 & HTC One for battery life. Of course we'd all love a a bigger battery, but given the incredible value of the Nexus phones, compromises have to be made, and I think battery life will be sufficient.



TSMC 28nm LP
static struct acpu_level tbl_PVS5_1700MHz[] __initdata = {
{ 1, { 384000, PLL_8, 0, 0x00 }, L2(0), 875000 },
{ 1, { 486000, HFPLL, 2, 0x24 }, L2(5), 875000 },
{ 1, { 594000, HFPLL, 1, 0x16 }, L2(5), 875000 },
{ 1, { 702000, HFPLL, 1, 0x1A }, L2(5), 875000 },
{ 1, { 810000, HFPLL, 1, 0x1E }, L2(5), 887500 },
{ 1, { 918000, HFPLL, 1, 0x22 }, L2(5), 900000 },
{ 1, { 1026000, HFPLL, 1, 0x26 }, L2(5), 925000 },
{ 1, { 1134000, HFPLL, 1, 0x2A }, L2(14), 937500 },
{ 1, { 1242000, HFPLL, 1, 0x2E }, L2(14), 950000 },
{ 1, { 1350000, HFPLL, 1, 0x32 }, L2(14), 962500 },
{ 1, { 1458000, HFPLL, 1, 0x36 }, L2(14), 987500 },
{ 1, { 1566000, HFPLL, 1, 0x3A }, L2(14), 1012500 },
{ 1, { 1674000, HFPLL, 1, 0x3E }, L2(14), 1050000 },
{ 1, { 1728000, HFPLL, 1, 0x40 }, L2(14), 1075000 },
{ 0, { 0 } }

TSMC 28nm HPM

static struct acpu_level acpu_freq_tbl_2p3g_pvs5[] __initdata = {
{ 1, { 300000, PLL_0, 0, 0 }, L2(0), 750000, 72 },
{ 0, { 345600, HFPLL, 2, 36 }, L2(1), 750000, 83 },
{ 1, { 422400, HFPLL, 2, 44 }, L2(2), 750000, 101 },
{ 0, { 499200, HFPLL, 2, 52 }, L2(2), 750000, 120 },
{ 0, { 576000, HFPLL, 1, 30 }, L2(3), 750000, 139 },
{ 1, { 652800, HFPLL, 1, 34 }, L2(3), 760000, 159 },
{ 1, { 729600, HFPLL, 1, 38 }, L2(4), 770000, 180 },
{ 0, { 806400, HFPLL, 1, 42 }, L2(4), 780000, 200 },
{ 1, { 883200, HFPLL, 1, 46 }, L2(4), 790000, 221 },
{ 1, { 960000, HFPLL, 1, 50 }, L2(9), 800000, 242 },
{ 1, { 1036800, HFPLL, 1, 54 }, L2(10), 810000, 264 },
{ 0, { 1113600, HFPLL, 1, 58 }, L2(10), 820000, 287 },
{ 1, { 1190400, HFPLL, 1, 62 }, L2(10), 830000, 308 },
{ 1, { 1267200, HFPLL, 1, 66 }, L2(13), 840000, 333 },
{ 0, { 1344000, HFPLL, 1, 70 }, L2(14), 850000, 356 },
{ 0, { 1420800, HFPLL, 1, 74 }, L2(15), 860000, 380 },
{ 1, { 1497600, HFPLL, 1, 78 }, L2(16), 870000, 404 },
{ 1, { 1574400, HFPLL, 1, 82 }, L2(17), 880000, 430 },
{ 0, { 1651200, HFPLL, 1, 86 }, L2(17), 890000, 456 },
{ 1, { 1728000, HFPLL, 1, 90 }, L2(18), 900000, 482 },
{ 0, { 1804800, HFPLL, 1, 94 }, L2(18), 910000, 510 },
{ 0, { 1881600, HFPLL, 1, 98 }, L2(18), 920000, 538 },
{ 1, { 1958400, HFPLL, 1, 102 }, L2(19), 930000, 565 },
{ 0, { 2035200, HFPLL, 1, 106 }, L2(19), 940000, 596 },
{ 0, { 2112000, HFPLL, 1, 110 }, L2(19), 955000, 627 },
{ 0, { 2188800, HFPLL, 1, 114 }, L2(19), 965000, 659 },
{ 1, { 2265600, HFPLL, 1, 118 }, L2(19), 975000, 691 },
{ 0, { 0 } }
};
 
Last edited:

Peek-A-Boogie

Senior Member
Oct 14, 2013
135
99
Nice thread, i can see that being true. Besides, like it was said above, based on the kitkat video; android 4.4 is said to be an improved project "butter". I expect the 2.300mh battery life to be enough for the phone.I can go as far as to say 3000mh is over doing it and pointless. :silly:
 

topgeardave

Senior Member
Aug 5, 2011
1,430
603
Does anyone know how much screen on time the HTC One gets? I can't find any screenshots in the One forums with the stock kernel and stock ROM.

Sent from my HTC Sensation using xda app-developers app
 

Turbotab

Senior Member
May 2, 2011
902
819
Of course being a Nexus, we are guaranteed excellent dev support, a custom kernel that allows undervolting will boost battery life, even the Nexus 4 isn't bad with a moderate undervolt.
 
  • Like
Reactions: Ensifolk

xXxG0dzRAgexXx

Senior Member
Feb 24, 2013
609
275
Beltsville, Maryland
Does anyone know how much screen on time the HTC One gets? I can't find any screenshots in the One forums with the stock kernel and stock ROM.

Sent from my HTC Sensation using xda app-developers app

Personally for me on stock ROM and kernel I got no less than 4+ hours screen on time. Going up to 6 hours some days.

Sent from my HTC One using Tapatalk now Free
 

BoneXDA

Senior Member
Oct 9, 2012
1,714
1,229
It looks like low power tasks and standby is best in class on the SD800. I'd image however that 100% load would brutalize the battery, although it's still nothing compared to the carpet bombing by the Adreno 330, which according to Erica, destroys the G2's 3000mAh playing NFS in 90 minutes. That's just crazy!
 

topgeardave

Senior Member
Aug 5, 2011
1,430
603
Does that mean by underclocking to around 1.5ghz we will get substantial battery savings?

Sent from my HTC Sensation using xda app-developers app
 

ashirviskas

Senior Member
Jun 13, 2011
303
70
Kaunas
Well, we could downclock our N5s to 1.7 GHz, enjoy good battery and then when the time comes when we need more power we could revert it back to 2.3 GHz. After a few years maybe :cyclops:
 

topgeardave

Senior Member
Aug 5, 2011
1,430
603
That's quite an amusing thought under clocking to 1.7ghz for battery savings, my gnex runs at 1.2ghz which is fine, my 2012 n7 at 1.3 lol

Sent from my Galaxy Nexus using Tapatalk

Haha, same here. My sensation is on 1.2ghz at the moment and it's fine. Only bottleneck is 768mb of RAM which means I have to kill many apps. Once I get the phone I'm going to compare battery life on 1.5ghz (same as N4) and 2.3 and see the difference in battery life. I might try an extreme test and downclock them to 1ghz/1.2ghz for lolz to share with everyone. The main thing I've learnt which determines battery life is mpdecision. I'm going to mess about with it till I can get the best battery life
 

Whyzor

Senior Member
Jul 7, 2011
714
352
Interesting, one thing I want to mention you're assuming the 'race-to-idle' condition for the CPU. And in web browsing, may not apply since the bottleneck is likely the network that it has to wait for, so even though the CPU may finish local tasks faster, it may spend extra (wasted) cycles waiting for the network to fetch the data. Or even RAM which are slower than local CPU cache.

Also some brief googling for the N4's S4 Pro (8064) chip voltage tables came up with something like this, for more accurate comparison

Code:
static struct acpu_level acpu_freq_tbl_fast[] __initdata = {
        { 1, {   384000, PLL_8, 0, 2, 0x00 }, L2(0),   850000 },
        { 0, {   432000, HFPLL, 2, 0, 0x20 }, L2(6),   875000 },
        { 1, {   486000, HFPLL, 2, 0, 0x24 }, L2(6),   875000 },
        { 0, {   540000, HFPLL, 2, 0, 0x28 }, L2(6),   900000 },
        { 1, {   594000, HFPLL, 1, 0, 0x16 }, L2(6),   900000 },
        { 0, {   648000, HFPLL, 1, 0, 0x18 }, L2(6),   925000 },
        { 1, {   702000, HFPLL, 1, 0, 0x1A }, L2(6),   925000 },
        { 0, {   756000, HFPLL, 1, 0, 0x1C }, L2(6),   975000 },
        { 1, {   810000, HFPLL, 1, 0, 0x1E }, L2(6),   975000 },
        { 0, {   864000, HFPLL, 1, 0, 0x20 }, L2(6),  1000000 },
        { 1, {   918000, HFPLL, 1, 0, 0x22 }, L2(6),  1000000 },
        { 0, {   972000, HFPLL, 1, 0, 0x24 }, L2(6),  1025000 },
        { 1, {  1026000, HFPLL, 1, 0, 0x26 }, L2(6),  1025000 },
        { 0, {  1080000, HFPLL, 1, 0, 0x28 }, L2(15), 1075000 },
        { 1, {  1134000, HFPLL, 1, 0, 0x2A }, L2(15), 1075000 },
        { 0, {  1188000, HFPLL, 1, 0, 0x2C }, L2(15), 1100000 },
        { 1, {  1242000, HFPLL, 1, 0, 0x2E }, L2(15), 1100000 },
        { 0, {  1296000, HFPLL, 1, 0, 0x30 }, L2(15), 1125000 },
        { 1, {  1350000, HFPLL, 1, 0, 0x32 }, L2(15), 1125000 },
        { 0, {  1404000, HFPLL, 1, 0, 0x34 }, L2(15), 1137500 },
        { 1, {  1458000, HFPLL, 1, 0, 0x36 }, L2(15), 1137500 },
        { 1, {  1512000, HFPLL, 1, 0, 0x38 }, L2(15), 1150000 },
        { 0, { 0 } }

I've seen a lot of comments, about the possible battery size in the Nexus 5, which is only ~10% larger than a Nexus 4, which had OK, but not great battery life.

So the N5 is rumoured to have 2300 mAh 3.8v battery, the same size as the HTC One, a device with much better battery life than the N4. But what people are overlooking is the improvements in power efficiency from the Snapdragon 800, although it is a much faster SoC than the S4 Pro in the N4, it actually requires less voltage. This is possible due to TSMC's superior HPM 28nm process, which uses a technology called high-K metal gates, both the Snapdragon 600 & S4 Pro were manufactured on TSMC's 28nm LP process.

But, what if you don't believe me, well just look below. I compare the voltage tables for a S800 vs a S600, we must remember that S600 is an improved S4 Pro, these improvements included higher efficiency (battery life), so it's better than the S4 Pro in the Nexus 4.

28nm LP @ 1.7 GHz = 1075mv -- Snapdragon 600

28nm HPM @ 1.7 GHz = 900mv -- Snapdragon 800

So the S800 uses significantly less power, but achieves much greater performance. There is one outlier, the Adreno 330 GPU it of course is also built on the same superior HPM process, but contains more ALU (compute) units than any Adreno 320, so at max 100% usage may consume slightly more power. In the real world, the CPU/GPU do not sit at 100% all time, and the much faster S800 can complete a task eg load a webpage, faster than the Nexus 4, and then power down, whilst the N4 would still be rendering the webpage. This is a double win, faster ultimate performance means a task can be completed quicker, and the SoC put into low power mode, the fact that S800 needs less voltage in the first place magnifies the advantage.

Another possible power efficiency for the S800, is the on-die modem (baseband), unlike the S4 Pro / S600 which had a separate modem, we don't have hard data for this, but logic suggests that this a more efficient solution.

In conclusion using the 2300 mAh HTC One as a guide, which also has a 1080P display, older Snapdragon 600 and more software bloat than vanilla Android 4.4, I'd wager that the Nexus 5 will easily beat the Nexus 4 & HTC One for battery life. Of course we'd all love a a bigger battery, but given the incredible value of the Nexus phones, compromises have to be made, and I think battery life will be sufficient.
 
Last edited:

The Gingerbread Man

Senior Member
Sep 4, 2010
8,097
3,945
Gloucester
Also the much lower voltages will really help save the power to if you saw that earlier in the thread which also should help not make these things heat up so much.

Sent from my Galaxy Nexus using Tapatalk
 

Top Liked Posts

  • There are no posts matching your filters.
  • 39
    I've seen a lot of comments, about the possible battery size in the Nexus 5, which is only ~10% larger than a Nexus 4, which had OK, but not great battery life.

    So the N5 is rumoured to have 2300 mAh 3.8v battery, the same size as the HTC One, a device with much better battery life than the N4. But what people are overlooking is the improvements in power efficiency from the Snapdragon 800, although it is a much faster SoC than the S4 Pro in the N4, it actually requires less voltage. This is possible due to TSMC's superior HPM 28nm process, which uses a technology called high-K metal gates, both the Snapdragon 600 & S4 Pro were manufactured on TSMC's 28nm LP process.

    But, what if you don't believe me, well just look below. I compare the voltage tables for a S800 vs a S600, we must remember that S600 is an improved S4 Pro, these improvements included higher efficiency (battery life), so it's better than the S4 Pro in the Nexus 4.

    28nm LP @ 1.7 GHz = 1075mv -- Snapdragon 600

    28nm HPM @ 1.7 GHz = 900mv -- Snapdragon 800

    So the S800 uses significantly less power, but achieves much greater performance. There is one outlier, the Adreno 330 GPU it of course is also built on the same superior HPM process, but contains more ALU (compute) units than any Adreno 320, so at max 100% usage may consume slightly more power. In the real world, the CPU/GPU do not sit at 100% all time, and the much faster S800 can complete a task eg load a webpage, faster than the Nexus 4, and then power down, whilst the N4 would still be rendering the webpage. This is a double win, faster ultimate performance means a task can be completed quicker, and the SoC put into low power mode, the fact that S800 needs less voltage in the first place magnifies the advantage.

    Another possible power efficiency for the S800, is the on-die modem (baseband), unlike the S4 Pro / S600 which had a separate modem, we don't have hard data for this, but logic suggests that this a more efficient solution.

    In conclusion using the 2300 mAh HTC One as a guide, which also has a 1080P display, older Snapdragon 600 and more software bloat than vanilla Android 4.4, I'd wager that the Nexus 5 will easily beat the Nexus 4 & HTC One for battery life. Of course we'd all love a a bigger battery, but given the incredible value of the Nexus phones, compromises have to be made, and I think battery life will be sufficient.



    TSMC 28nm LP
    static struct acpu_level tbl_PVS5_1700MHz[] __initdata = {
    { 1, { 384000, PLL_8, 0, 0x00 }, L2(0), 875000 },
    { 1, { 486000, HFPLL, 2, 0x24 }, L2(5), 875000 },
    { 1, { 594000, HFPLL, 1, 0x16 }, L2(5), 875000 },
    { 1, { 702000, HFPLL, 1, 0x1A }, L2(5), 875000 },
    { 1, { 810000, HFPLL, 1, 0x1E }, L2(5), 887500 },
    { 1, { 918000, HFPLL, 1, 0x22 }, L2(5), 900000 },
    { 1, { 1026000, HFPLL, 1, 0x26 }, L2(5), 925000 },
    { 1, { 1134000, HFPLL, 1, 0x2A }, L2(14), 937500 },
    { 1, { 1242000, HFPLL, 1, 0x2E }, L2(14), 950000 },
    { 1, { 1350000, HFPLL, 1, 0x32 }, L2(14), 962500 },
    { 1, { 1458000, HFPLL, 1, 0x36 }, L2(14), 987500 },
    { 1, { 1566000, HFPLL, 1, 0x3A }, L2(14), 1012500 },
    { 1, { 1674000, HFPLL, 1, 0x3E }, L2(14), 1050000 },
    { 1, { 1728000, HFPLL, 1, 0x40 }, L2(14), 1075000 },
    { 0, { 0 } }

    TSMC 28nm HPM

    static struct acpu_level acpu_freq_tbl_2p3g_pvs5[] __initdata = {
    { 1, { 300000, PLL_0, 0, 0 }, L2(0), 750000, 72 },
    { 0, { 345600, HFPLL, 2, 36 }, L2(1), 750000, 83 },
    { 1, { 422400, HFPLL, 2, 44 }, L2(2), 750000, 101 },
    { 0, { 499200, HFPLL, 2, 52 }, L2(2), 750000, 120 },
    { 0, { 576000, HFPLL, 1, 30 }, L2(3), 750000, 139 },
    { 1, { 652800, HFPLL, 1, 34 }, L2(3), 760000, 159 },
    { 1, { 729600, HFPLL, 1, 38 }, L2(4), 770000, 180 },
    { 0, { 806400, HFPLL, 1, 42 }, L2(4), 780000, 200 },
    { 1, { 883200, HFPLL, 1, 46 }, L2(4), 790000, 221 },
    { 1, { 960000, HFPLL, 1, 50 }, L2(9), 800000, 242 },
    { 1, { 1036800, HFPLL, 1, 54 }, L2(10), 810000, 264 },
    { 0, { 1113600, HFPLL, 1, 58 }, L2(10), 820000, 287 },
    { 1, { 1190400, HFPLL, 1, 62 }, L2(10), 830000, 308 },
    { 1, { 1267200, HFPLL, 1, 66 }, L2(13), 840000, 333 },
    { 0, { 1344000, HFPLL, 1, 70 }, L2(14), 850000, 356 },
    { 0, { 1420800, HFPLL, 1, 74 }, L2(15), 860000, 380 },
    { 1, { 1497600, HFPLL, 1, 78 }, L2(16), 870000, 404 },
    { 1, { 1574400, HFPLL, 1, 82 }, L2(17), 880000, 430 },
    { 0, { 1651200, HFPLL, 1, 86 }, L2(17), 890000, 456 },
    { 1, { 1728000, HFPLL, 1, 90 }, L2(18), 900000, 482 },
    { 0, { 1804800, HFPLL, 1, 94 }, L2(18), 910000, 510 },
    { 0, { 1881600, HFPLL, 1, 98 }, L2(18), 920000, 538 },
    { 1, { 1958400, HFPLL, 1, 102 }, L2(19), 930000, 565 },
    { 0, { 2035200, HFPLL, 1, 106 }, L2(19), 940000, 596 },
    { 0, { 2112000, HFPLL, 1, 110 }, L2(19), 955000, 627 },
    { 0, { 2188800, HFPLL, 1, 114 }, L2(19), 965000, 659 },
    { 1, { 2265600, HFPLL, 1, 118 }, L2(19), 975000, 691 },
    { 0, { 0 } }
    };
    22
    Besides all these, I believe android 4.4 is more battery friendly than 4.3

    Sent from my Nexus 4 using xda app-developers app

    Can I borrow your crystal ball? I need to get tonights lotto numbers
    7
    Here's a more accurate comparison based on real battery tests done by GSMArena on Nexus 4 and LG G2. The N5 should have a slightly thriftier LCD panel than G2's 5.2", so the projected battery life may even be better.

    http://blog.gsmarena.com/google-nexu...-galaxy-nexus/

    Nexus 4:
    32h Endurance Rating
    14:17h 3G Talk time
    4:34h Web Browsing
    4:55h Video Playback

    http://blog.gsmarena.com/lg-g2-batte...e-the-results/

    LG G2 (3000 mAh battery):
    62h Endurance Rating
    25:01h 3G Talk Time
    11:22h Web Browsing
    11:51h Video Playback

    Nexus 5 (Projected w/ 2300 mAh battery, 77% of LG G2's):
    47h Endurance Rating
    19:11h 3G Talk Time
    8:43h Web Browsing
    9:05h Video Playback

    Most people, including me, get about 4.5 hrs of screen on time on N4, that translates to about 8+ hrs on Nexus 5 with same usage patterns.
    7
    4 8 15 16 23 42

    :D

    Sent from my Nexus 4 using xda app-developers app

    Wouldn't we all be devastated if these actually turn out to be the correct lotto numbers haha

    Sent from my Nexus 4 using XDA Premium 4 mobile app
    5
    I just don't get it when some people here say that's ok if the battery is small,you can always underclock,keep screen brightness at 10%,use your phone less etc,etc.
    What's the point of getting a device with a 2.3GHz processor if you are going to underclock it?
    What's the point of having a great bright screen if you have to use it at lower brightness just to get you through the day.
    And generally speaking what's the point of having a smartphone if you have to do all these and also turn sync off,greenify apps,stop watching videos,stop playing games,stop browsing the web and have 2-3 chargers in every place you go just to keep your device running?
    Smartphones were made for "power" users and should last at least a day of ''heavy'' usage..