baseballbert

New member
Aug 14, 2011
1,610
2
0
Visit site
WP SkyDrive Win!!

This morning I needed to edit a PowerPoint but didn't have time to fire up my ancient laptop. So, thanks to WP awesomeness, I put it in SkyDrive yesterday, was able to get it on my Focus this morning, make the edits, save it back to skydrive and then put it on my laptop this morning for my presentation.

All from the phone. How awesome is that?


Mangos taste better than apples!
 

jfa1

New member
Dec 15, 2007
2,567
1
0
Visit site
Because they dont take the time to learn what all the device and os can do before they get frustrated and thro their hands up and complain that the devicve os cant do whateve it is when it probably does if they would actually do something to try and figure it out
 

jimski

New member
Dec 11, 2010
2,253
8
0
Visit site
But, but WP doesn't have dual core, and Brawndo has electrolytes.

Sent from my HTC Surround using Board Express
 

TaliZorah

New member
Sep 27, 2011
649
0
0
Visit site
I hate dual-core because it wastes more battery. And for what? Does WP7 NEED dual-core? Absoutely 100% NO. ****, even single core 1Ghz processors run the OS without a hitch. You can't say that for android. The low-end android's are super laggy and glitchy. So I can see why android fanboys want dual-core. It's because their OS needs it to run smoother. That's just no true with WP7.

Basically if it ain't broke, don't fix it.
 

jeremyshaw

New member
Oct 21, 2011
602
1
0
Visit site
I hate dual-core because it wastes more battery. And for what? Does WP7 NEED dual-core? Absoutely 100% NO. ****, even single core 1Ghz processors run the OS without a hitch. You can't say that for android. The low-end android's are super laggy and glitchy. So I can see why android fanboys want dual-core. It's because their OS needs it to run smoother. That's just no true with WP7.

Basically if it ain't broke, don't fix it.

Dual core wasting power is a myth. Maybe with crud filled nVidia attempts at CPU (by the time they released, they were no longer the first shipping ARMv7 Cortex A9 dual core, and their GPU was laughably slow - it's barely, if even, faster than the Adreno 205, and that's if there is a perfect 50/50 mix of pixel/vertex shader load! Not even unified shaders!).

However, dual core ARM designs normally powergate the second core when it's not being used, so no extra power is expended when it is not in operation. Just the OS has to be written to be aware of this.

If the web browsing improvements (since current modem softcode is still CPU intensive - something Icera is solving with more on-die ASIC, a general GPU design principal nVidia agrees with), are anything to show between the A4 and A5 chips in the iPhone 4/4S (best case example, IMO), then dual core will speed up web browsing so more CPU time is spent idling, and more 3G/4G modem time is spent not transfering data + keeping alive a connection. Though the iPhone 4S still has less standby time, probably a relation of Siri's proximity sensor use, the faster Qualcomm modem, and the massive GPU the A5 has.

IMO, if done well, then dual core = nice to have. With Krait and A15's IPC improvements, I really feel quad core for smartphones is true overkill. Maybe tablet PC, but I don't really care much for that. Considering Qualcomm is only launching a dual core version of their S4 MSM (including LTE), I think it's very likely mobile phones won't see much more than "fandroid MUST HAVE MOAR to finally get 'smooth as butter' GUI (the most improper and overused, IMO, phrase of any android forum) - only on a clean reboot with nothing running, of course," have any real use of quad cores. By the time the rumored A9 quad core phones drop, A15 and Krait dual core will, too.
 
Last edited:

TaliZorah

New member
Sep 27, 2011
649
0
0
Visit site
Dual core wasting power is a myth. Maybe with crud filled nVidia attempts at CPU (by the time they released, they were no longer the first shipping ARMv7 Cortex A9 dual core, and their GPU was laughably slow - it's barely, if even, faster than the Adreno 205, and that's if there is a perfect 50/50 mix of pixel/vertex shader load! Not even unified shaders!).

However, dual core ARM designs normally powergate the second core when it's not being used, so no extra power is expended when it is not in operation. Just the OS has to be written to be aware of this.

If the web browsing improvements (since current modem softcode is still CPU intensive - something Icera is solving with more on-die ASIC, a general GPU design principal nVidia agrees with), are anything to show between the A4 and A5 chips in the iPhone 4/4S (best case example, IMO), then dual core will speed up web browsing so more CPU time is spent idling, and more 3G/4G modem time is spent not transfering data + keeping alive a connection. Though the iPhone 4S still has less standby time, probably a relation of Siri's proximity sensor use, the faster Qualcomm modem, and the massive GPU the A5 has.

IMO, if done well, then dual core = nice to have. With Krait and A15's IPC improvements, I really feel quad core for smartphones is true overkill. Maybe tablet PC, but I don't really care much for that. Considering Qualcomm is only launching a dual core version of their S4 MSM (including LTE), I think it's very likely mobile phones won't see much more than "fandroid MUST HAVE MOAR to finally get 'smooth as butter' GUI (the most improper and overused, IMO, phrase of any android forum) - only on a clean reboot with nothing running, of course," have any real use of quad cores. By the time the rumored A9 quad core phones drop, A15 and Krait dual core will, too.


Its not a myth at all that they take more power. Maybe depending on how you have the software and hardware handle the processing it might make a little difference in battery life but yes, more processing = more power.
 

jeremyshaw

New member
Oct 21, 2011
602
1
0
Visit site
Its not a myth at all that they take more power. Maybe depending on how you have the software and hardware handle the processing it might make a little difference in battery life but yes, more processing = more power.

More instantaneous power, less energy used overall. Finish a task faster = no/little power used after it's done. Epsecially since the CPU would be holding up a lot more than just the CPU threads, it would be holding up the modem (either 3G or WiFi must keep the connection alive longer), keeping the screen active longer (user), AND in WP7's case, keep the GPU active longer.


So overall... if done well, overall energy use is still better, even if it draws more power when actualy doing work. However, even that is debateable, due to HW design. It's two cores, with a shared setup, and normally cohesive design that is more advanced then their stalled single core cousins (which are no longer updated on the design-side, just the fab manuf.). It's not two seperate CPUs with seperate power rails being slobbered into one package, lol.


Case in point, Galaxy SII battery life - not much different from Focus S for my near identical useage (email, web surfing, text, music ~ 256/320kbps) on either device, and I speak as an owner of both (though my Galaxy SII was the TW version, dunno if the American versions are Power Hoggs or not, lol). Identical batteries, too, and both get just short of 16 hours. Not to mention, the Galaxy SII (at least the ones utilizing the Exyos chip) has a significantly faster GPU, too (well, and two faster CPU cores - starting with IPC, and ending in SMP). And that is with Samsung's serious improvements in dual core and GPU utilization over the stock Android (especially noted in the browser, which is GPU accelerated).
 

jeremyshaw

New member
Oct 21, 2011
602
1
0
Visit site
Hey, I'm not trying to pick a fight here, just telling you what I know. And when I was around in xda devs, there were massive complaints on how the Atrix 4G's dual core would cause battery life to / in half. Yet... it didn't. My Atrix 4G manages to hold at least 24 hours of charge after I use it, something I could not say for my Focus S and Galaxy SII (though my old Focus came close - probably because I used it's camera signifcantly less than my other devices, all I can think as for why it's that much different).

To be fair, the Atrix 4G has a monster sized battery (300mAh more than the GalaxyS2/FocusS 1650mAh), though it also had probably the most power inefficient dual core ARMv7-A9 SoC to exist (it cannot voltage gate unused cores, only clock gate, and was synchronus clocked:excl:).


EDIT: IMO, WP7 can gain quite a bit of efficiency from a dual core processor, though I understand with Qualcomm S4 comming up, and a probable desire to "get it right[er]," MS would be well within reason to wait on that 28nm, LTE embedded (instead of a seperate modem) sweetness before making a serious commitment.
 

TaliZorah

New member
Sep 27, 2011
649
0
0
Visit site
More instantaneous power, less energy used overall. Finish a task faster = no/little power used after it's done. Epsecially since the CPU would be holding up a lot more than just the CPU threads, it would be holding up the modem (either 3G or WiFi must keep the connection alive longer), keeping the screen active longer (user), AND in WP7's case, keep the GPU active longer.


So overall... if done well, overall energy use is still better, even if it draws more power when actualy doing work. However, even that is debateable, due to HW design. It's two cores, with a shared setup, and normally cohesive design that is more advanced then their stalled single core cousins (which are no longer updated on the design-side, just the fab manuf.). It's not two seperate CPUs with seperate power rails being slobbered into one package, lol.


Case in point, Galaxy SII battery life - not much different from Focus S for my near identical useage (email, web surfing, text, music ~ 256/320kbps) on either device, and I speak as an owner of both (though my Galaxy SII was the TW version, dunno if the American versions are Power Hoggs or not, lol). Identical batteries, too, and both get just short of 16 hours. Not to mention, the Galaxy SII (at least the ones utilizing the Exyos chip) has a significantly faster GPU, too (well, and two faster CPU cores - starting with IPC, and ending in SMP). And that is with Samsung's serious improvements in dual core and GPU utilization over the stock Android (especially noted in the browser, which is GPU accelerated).


Two different phone architectures with two different operating systems and personal use is hardly any real argument to your statement that more processing power does not mean it needs more power.

I am not saying it's not possible to have the same or possibly even better battery life with a faster / better / more powerful processor. What I AM saying is that in literal terms they need more power.

It is exactly why when I upgrade my computers graphics card or processor / motherboard sometimes I need to upgrade my PSU (Power Supply Unit) as well. If I just replace my graphics card with a newer one it will most likely require more power. Why? more cores, more processing... etc.

If more processing never needed more power than you wouldn't see PSU's on the store shelf because there would never be any reason to add power. I guess cell phones are magically different somehow? Like I said, you may get two phones where one has a beefier processor but the battery life between the two are about the same and that can depend on the OS and hardware architecture. Needless to say phone companies don't usually TRY to have phones with terrible battery life. But as technology goes along things get better.

I remember reading an article about 3 years ago talking about dual-cores in phones and how it's possible but the power consumption wouldn't make it worth it. Give technology time (and it moves fast) and we can now get dual-cores that perform acceptable by today's standards. If people still really wanted to research how to make single cores have better battery life they can. But that doesn't sell products.

However newer single-core / less Ghz phones are seeing an improvement in battery power. I have been researching the T-Mobile HTC Radar and apparently it by far has the best battery life of any WP7 phone out there now (In the USA at least). It's a 1Ghz single core chip but architecture is completely different than any other WP7 phone out there. Thats why the same processor in the Radar and say another phone might get different battery life. Especially if they are a different OS.
 

TaliZorah

New member
Sep 27, 2011
649
0
0
Visit site
Hey, I'm not trying to pick a fight here, just telling you what I know. And when I was around in xda devs, there were massive complaints on how the Atrix 4G's dual core would cause battery life to / in half. Yet... it didn't. My Atrix 4G manages to hold at least 24 hours of charge after I use it, something I could not say for my Focus S and Galaxy SII (though my old Focus came close - probably because I used it's camera signifcantly less than my other devices, all I can think as for why it's that much different).

To be fair, the Atrix 4G has a monster sized battery (300mAh more than the GalaxyS2/FocusS 1650mAh), though it also had probably the most power inefficient dual core ARMv7-A9 SoC to exist (it cannot voltage gate unused cores, only clock gate, and was synchronus clocked:excl:).


EDIT: IMO, WP7 can gain quite a bit of efficiency from a dual core processor, though I understand with Qualcomm S4 comming up, and a probable desire to "get it right[er]," MS would be well within reason to wait on that 28nm, LTE embedded (instead of a seperate modem) sweetness before making a serious commitment.


Oh im not trying to pick a fight either :p I am just speaking from what I know. I have built many, many, many computers in my life. I haven't done so much with hardware on phones so I could be way off. I just don't see how a phone could be so different from a computer in regards to processors / power consumption in this regard. Like I said in my previous post it just doesn't make sense to me here.

Were good. Not fighting :) Good discussion.
 

jeremyshaw

New member
Oct 21, 2011
602
1
0
Visit site
Two different phone architectures with two different operating systems and personal use is hardly any real argument to your statement that more processing power does not mean it needs more power.

I am not saying it's not possible to have the same or possibly even better battery life with a faster / better / more powerful processor. What I AM saying is that in literal terms they need more power.

It is exactly why when I upgrade my computers graphics card or processor / motherboard sometimes I need to upgrade my PSU (Power Supply Unit) as well. If I just replace my graphics card with a newer one it will most likely require more power. Why? more cores, more processing... etc.

If more processing never needed more power than you wouldn't see PSU's on the store shelf because there would never be any reason to add power. I guess cell phones are magically different somehow? Like I said, you may get two phones where one has a beefier processor but the battery life between the two are about the same and that can depend on the OS and hardware architecture. Needless to say phone companies don't usually TRY to have phones with terrible battery life. But as technology goes along things get better.

I remember reading an article about 3 years ago talking about dual-cores in phones and how it's possible but the power consumption wouldn't make it worth it. Give technology time (and it moves fast) and we can now get dual-cores that perform acceptable by today's standards. If people still really wanted to research how to make single cores have better battery life they can. But that doesn't sell products.

However newer single-core / less Ghz phones are seeing an improvement in battery power. I have been researching the T-Mobile HTC Radar and apparently it by far has the best battery life of any WP7 phone out there now (In the USA at least). It's a 1Ghz single core chip but architecture is completely different than any other WP7 phone out there. Thats why the same processor in the Radar and say another phone might get different battery life. Especially if they are a different OS.

That exact same arguement was used back when android was getting sprinked (now almost drowned) with dual core. "Different architecture, different phones, different screens, different batteries, different OS [it was rumored Froyo could not use dual cores, or GPU acceleration, and Gingerbread, the next Android OS version, would be required to utilize both :)]" = non comparible.

Of course, that didn't stop single core devices from having very similar and comparable battery life to their dual core sucessors.

Seriously, if that is the case, your desktop PC comaprison is even less relevant, since motherboards affect the power useage, often more, than "more cores," on a CPU. Though with Intel's tick/tock, outside of some rare server chips, each generation of chips generally draw equal or less power, even with more/equal count of cores ;)


Also, a desktop PC doesn't need more than 300W of power, has been that way for almost a decade now. Off the shelf HP, Dell, Acer desktops are lucky to even have 300W of PSU power. Of course, if you want a massive GPU (which, if you want to know, GPUs in the past decade, since 2001 or so, have been drawing MORE power every generation, something that is opposite of general ARM SoC development), or Crossfire/SLI, then a bigger PSU is recommendable. Which is why I utilized my ownership of a Galaxy SII and Focus S. Both are relatively "this generation" devices, both have almost identical specifications, outside of the SoC utilized and OS. Both have their BSP developed by Qualcomm, and both have their operating system heavily tweaked to better utilize all avalible resources (one by MS, one by Sammy). It's the bigger part of why the Galaxy SII is generally reguarded as the first dual core phone to matter. It's the first to actually put it to good use, outside of gaming.


Back to the desktop example, it's not possible to simply double mobile chipset power consumption every three years. It can only go so far, before battery/size limitations push it right back again. Today, we are hitting a limit around 4.3-4.5" devices being the largest that one hand can reasonably operate. So battery size isn't getting better, and neither is battery tech (if I had a penny for every time a new "10x life, 10x faster charge" tech was developed but never made it anywhere, I'd be about.... $100 or so richer just in this past decade). However.... better utilization of that battery life can be done, and we are seeing that with dual cores, better GPU utilization, new display panel technology, new fabrication processes (for silicon, at least).


And again back to desktop, it's connected to a main power source, where battery life is not a concern. If you note, CPU TDP limits have been staying consistant around 95W - 130W (unless if you are AMD, in which case, you report TDP one way, and actual power consumption every generation will vary wildly), while core counts have been going up (as well with IPC, and ever so slowly, clocks). The reason you have more big PSU on the shelf are due to our GPUs.... comming from a mere ~25W of power draw (iirc) for the "will need a small nuclear reactor ATi 9700 Pro" to the massive 300W++ power draw of Fermi.... and MultiGPU to punch it even higher... well... no guessing required, it's the GPU that's driving PSU capacities up, not the CPU we are discussing.

We do see dual cores being utlized better today (iPhone 4 vs iPhone 4S - the Safari mobile browser is not GPU accelerated in javascript, where the two chips show the largest difference) over single cores. There is no real reason not to include them. ESPECIALLY, since they do not consume more power than their effectively EOL single core counterparts. The last single core SoC designs we will see in smartphones are on the 45nm process. 32nm&28nm dual cores designs are already sampling/shipping out from Samsung and Qualcomm[uses TSMC] (also likely TI, though I don't know what process node they are on right now - TI is a major foundry, 3rd largest after TSMC and GloFo; 4th if you want to count the exclusive, and elusive, Intel).

Single cores being more energy efficient than Dual cores, is simply put, a myth. It's one that plauged Android forums before the Atrix, G2x, Galaxy SII, and later toys, actually were sold and used by real people. It's one that was put to rest a long time ago for Android users. And Android is normally considered a power guzzler of impressive HW.

Oh im not trying to pick a fight either :p I am just speaking from what I know. I have built many, many, many computers in my life. I haven't done so much with hardware on phones so I could be way off. I just don't see how a phone could be so different from a computer in regards to processors / power consumption in this regard. Like I said in my previous post it just doesn't make sense to me here.

Were good. Not fighting :) Good discussion.

I almost didn't want to put up this post, but please, I didn't have enough energy left after to make it seem less... er... attacking? I'll try and edit it later on, but that will require some significant structural changes to my post, so I'll likely never get around to it :( I'm sorry! Just know, I think you're wrong (as do a LOT of android users :)), and it's going to stay that way :p
 

TaliZorah

New member
Sep 27, 2011
649
0
0
Visit site
If you don't think that the processor doesn't use more battery here are some specs:

Radar:
1Ghz CPU
1520 mAh battery
7.7 hours of talktime
575 hours of standby time.

Titan:
1.5Ghz CPU
1650 mAh battery
4 hours of talk time.
336 hours of standby time.

WOW! The titan has a larger battery too! At first I thought hmmm, maybe the screen has something to do with it. Then I realized that when talking or when in standby mode the screen is off.

FACT: Faster processors / multi-core processors use more power than slower / single (lesser) core processors.

Yes, there are ways through software and architecture to make dual-core and faster processor phones have the same life as a single core or less Ghz phone. But most of the time that requires a larger battery or like I have said a thousand times... architectual or software differences. That includes the OS.

From: What's the difference between i3, i5 and i7 processors?
"Power Consumption: A faster processor uses more power. That part's pretty simple, but it's at odds with wanting to extend battery life in notebook computers as much as possible. As a result, many variants of processors are architected to use less power at the cost of some of the processor's other features. It might operate slower or lack other processor features."


7441.png


Power_consumption_graph.PNG


power-3.png


Please notice the number of cores and the speed of the processors related to the amount of power consumption. Yes, these are PC CPU's but its a fact that more / faster processing = more power consumption.

If you and "a LOT of android users" think that more processing power does not mean more power consumption then you and a LOT of android users are very, Very, VERY WRONG!
 
Last edited:

Kahuna Cowboy

New member
May 25, 2011
96
0
0
Visit site
Yep, Skydrive and Office intrigation is a solid feature of WP. It's native on Droid with their Google Docs which no one uses, and requires buying iWork apps on iOS. iOS pull it off nicely too though.
 

kevm14

New member
Nov 6, 2011
554
0
0
Visit site
You need to ignore the 100% load graphs because the faster a processor is, for a given fixed user environment, the more OFTEN it will be idle. And depending on what you are doing, finishing a task faster may result in you putting the phone back to sleep faster, which will REALLY improve battery life.

Also, I think 4 hrs of talk time for the Titan is a misprint. It disagrees with HTC's own spec.
 
Last edited:

TaliZorah

New member
Sep 27, 2011
649
0
0
Visit site
You need to ignore the 100% load graphs because the faster a processor is, for a given fixed user environment, the more OFTEN it will be idle. And depending on what you are doing, finishing a task faster may result in you putting the phone back to sleep faster, which will REALLY improve battery life.

I guess you missed the top chart where it shows the processors at idle. Or the part of my post that shows the Radar vs the Titan where the Titan has faster processor and a bigger battery but still not as much standby time as the Radar.

I don't know why this is so hard for people to understand. It's common knowledge (like the quote I posted) that more processing power / more cores = more power consumption.

I feel like I am going crazy here. This is a pretty basic principal of computer science but for some reason people don't want to believe it. I guess because you all WANT faster processors to use less power so you have more of a reason to want one? I dunno... It's wishful thinking. Unless you get a bigger battery to even it out the faster, more powerful processors are ALWAYS going to use more power.

I also can't believe that somebody actually said: "a desktop PC doesn't need more than 300W of power, has been that way for almost a decade now."

Seriously.
 

Members online

Forum statistics

Threads
322,916
Messages
2,242,890
Members
428,005
Latest member
rogertewarte