Can anyone please recommend some REALLY power saving Radio rom?
I'm currently using 1.54.30
I like 1.54.07.00. Its a good compromis between power and battery drain.
I think 1.47 was the version with the lowest battery drain for me yet...
1.47.00.10 seems to do the job for me.. moderate usage of voice & data (UMTS 2100)..
thank you so much for responding... i will try 1.54.07...
yes, 1.47 is the best radio I also have used.
i'm trying 1.54.07...using it now...will see how is the battery drainage...
what is so good about 1.47 anyway? just to compare...
2 cents on this topic - the observation of battery drain is somewhat subjective. I'd suggest someone do an extended test where the phone has a current drain logging program and track drain for each phone mode - data connection - transfer / phone use / idle and do a statistical comparison.
With a lack of good data? I'd think that the latest rev would have changes for some reason - hopefully a good one.
I'm using the latest 1.54 without performance issues.
hi all
I am using 1.54.07.00 radio, I use to get call from my gf every night for straight 3 hours. All these time my wifi is also active. Most of the days I get 40-50% of battery after I finish the call. So I think it is pretty impressive, isnt it?
Cheers
Jish
I'm working on a way to evaluate this.. I agree with mattk_r that it will probably vary drastically by region, usage, configuration, network, etc. But at least for my case, wi-fi is almost always off and my phone is in 'sleep'/'suspend' with Flexmail email running an IDLE (PUSH) IMAP email session in the background. Just a note, the way this works, the phone still gets to sleep and anytime a packet/data is received over the GPRS/3G connection the phone is woken up in unattended mode (no backlight) so that the maill app can process the message and notification then the phone returns to sleep state. So my goal is to pseudo-scientifically figure out which radio works best for my use.
To figure that out I have written a pair of programs that simulate the IMAP IDLE connection of an email client. (Why not use the real thing? I want the experiment to be the same everytime I run it, I can't guarantee that with email and software that I don't know the intricacies of.) One half runs on the phone to simulate the phone email client, and one half runs on a server that sends periodic messages to the phone to wake it up. When the phone is waken up it sends its current estimate of the battery life remaining (as reported by Windows API) to the server for logging.
In order to run my tests, everynight I have/plan to do the following procedure.
Flash a new radio to the phone for test
Charge the battery fully
Soft-Reset phone
Run my program
And then in the morning I dump the server log and generate a nice pretty graph of battery remaining v. hours on. This has obvious deficiencies such as
The phone never actually moves, it is on my nightstand so this won't simulate an active day where the radio is having to select the best tower.
When I wake up my battery is not dead, so I am not able to measure time from total charge to total discharge.
Even with these deficiencies and possibly more, I believe that this will give a decent benchmark as to which radios perform better than others. (in terms of battery life) So far I have conducted this for 2 nights and 2 radios, both running in 2G mode. The next two nights I will probably re-run the tests for 3G enabled just to see if that causes a significant change in the result.
So see this pretty graph to see what's happened so far... I'll post updates as I progress.
Oh yeah, I'm on ATT/Cingular Network in USA/TX.
Results in summary:
So far, I have tested radios 1.47.30.10 and 1.54.30.10 and gotten practically identical results for battery life. In my day of use for each radio, I have not noticed any call quality issues with either radio.
Other techniques: ?
There probably are other techniques to accomplish this, such as a program that measures battery discharge current. My gut feeling is to not trust these for this analysis because those programs typically have a low sampling rate and I don't know where they get there information and if that is a reliable source. And the other problem, I haven't found any such programs that can do it when the phone decides to go into the sleep mode and radio is having to maintain an idle but yet present connection to the cellular network.
Feedback anyone? Ideas?
dear experts, i've flashed 1.54.07 radio rom, but i soon discovered that i've lost my HSDPA connection to network. now i can only have 3G connection. Even though i tried to enable HSDPA in Schap's Config...but still no result...
Help please...or please suggest which radio rom which has got HSDPA enabled...
p/s: i think 1.54.07 is power consumption friendly becoz it has disable HSDPA connection...
what do u guys think?
HSDPA is enabled in that rom. Been using it with pandora naked rom with no issues whatsoever.
that's weird...hmmm....I'm using Schap's 3.57a ROM...previously have HSDPA b4 i upgrade to this rom...
mattk_r said:
2 cents on this topic - the observation of battery drain is somewhat subjective. I'd suggest someone do an extended test where the phone has a current drain logging program and track drain for each phone mode - data connection - transfer / phone use / idle and do a statistical comparison.
With a lack of good data? I'd think that the latest rev would have changes for some reason - hopefully a good one.
I'm using the latest 1.54 without performance issues.
Click to expand...
Click to collapse
INDEED !!!
I was go with 1.54.30... then I have a trouble to find "3G" (somethimes it was hard to find "E") then I switch back to 1.50.00... and everything is ok (I mean from the same place I live, I found "3G" successfully). The battery drain should be have relation with the connection and that's vary in other country IMHO
Still on 1.43 and seem to be doing just fine with battery life. Maybe I will try the 1.47
Switch to 1.50.00....HSDPA returns...
So i can presume that 1.54.07 has got no HSDPA enabled for power saving...
What say experts?
There has been so much speculation on battery performance on the Sprint version that I wrote this logging program. This is the first stable Beta Version, Please PM me with bug reports (and as usual use at your own risk).
I know that it works for the CDMA version if someone wants to try it on the GSM version that would be great. Also Using the nuePower Battery driver for the Diamond give better resolution on the battery level.
What it logs:
1. a. Date and Time
b. Is AC Charger Attached
c. Battery Voltage
d. Battery Current (Discharge)
e. Battery Temperature
f. Battery level (%)
g. Pertinent System events:
a. Active Application
b. Speaker Phone Activity
c. Phone Radio Off/On
d. Phone Active Data Call (EVDO/1xRTT Activity)
e. No Phone Service
f. Phone Searching For Service
g. Bluetooth Power State On/off
h. Bluetooth Hands Free Audio in use (BT Headphones)
i. Bluetooth Hands Free Control in use (BT HF Headset)
j. WiFi Power On
k. WiFi Connected
l. Bluetooth Connections Count
m. Bluetooth Connections Descriptions
n. Cellular Connections Count
o. Cellular Connections Descriptions
p. Phone Call (Talking)
I have been unable to locate where HTC has hidden the charging current so if anyone could let me know that I would appreciate it.
Also, It dumps the data into a CSV file that you can pull into excel if you like. Please PM me with Bug Reports. There is a read.me version of the manual in the cab with a caution about logging to a storage card, please be aware of that if you plan on dumping your log to the storage card.
Version 0.2.3.130
New Features:
Support for Windows Mobile 5
Native QVGA Support (Properly Scales the Graph)
Can toggle between F and C for the Battery Temperature
Bug Fixes:
Properly deletes the downloaded file in the Benchmark test
Some additional Exception handling added
VERSION 0.2.1.105+ REQUIRES .NETCF 3.5!
I have uploaded a new version of the utility 0.1.1.58.
Includes the following Bug Fixes:
1) Now throws an update for every event. Previous Logic would drop events if another event occurred prior to the timed update.
2) Now allows the system to drop into Power mode D3 (sleep)
love the idea. will be reporting soon.
I have uploaded a new version with some reported bug fixes...
The following bug fixes are in place:
1) If multiple events occur while saving the log, a file in use exception would be thrown. It is now properly handled.
2) Visibility of the plots is now properly implemented.
After slugging through several long days of debugging and testing I have uploaded ver 0.2.1 of the battery log/diagnostics program.
What's new:
added mAH consumption counters for Data and Phone calls
Added a Battery Benchmark Module
Fixed a few miscellaneous bugs.
Benchmark Module details:
This module places the phone in a standard state (screen and backlight on, data connection active and Bluetooth radio on and discoverable). Then performs a 5 minute continuous data download, followed by a 20 minute telephone call. And saves a final report detailing power usage and AC adapter and charging times.
maybe my comments are silly but after those long discussions about battery draining in Raphael and diamond devices i find curious that nobody has said anything about this tool.
correct me if i'm wrong but this tool seems the best tool i've seen until know to check what the hell is draining our devices when in standby or calling or Data downloading or whatever.
i don't know too much about how it runs but i expect to learn it after some tests i will do.
I think that this tool will let me discover which running process is draining my battery, of course this tool won't tell it to me but after some testsenabling and disabling processes i will be able to see which one/s is/are the guilties.
I guess that the "ma" value that changes over time is per hour right? maybe PalladiumTD, could you explain me how to use this tool as if i would be a dummy, ( you can take out the "as if" if you want...
i don't understand the difference between "average current (ma)" and total usage (mah). The display doesn't change its contents when i switch between these options.
BTW it seems to run in my GSM device with unprotected Proven ROM 1.502. of course i can't install Nuepower driver.
is this tool trustable and accurate?
it doesn't seem to instantly react depending of what is running in the phone. it reacts after a while. ( of course i have it set to one second update)
and apparently it jumps from -78ma to -145ma and to -215ma without doing anything special. The Phone is just flashed with only battlog installed and running and it jumps between these values for no apparently reason.
Also i don't see any difference between having HDSP enabled or disabled (GPRS).
if i turn on Wifi, i see a big difference.
The benchmark test doesn't seem to run. it does not call and it does not download anything. (with wifi turned on)
fourcc said:
maybe my comments are silly but after those long discussions about battery draining in Raphael and diamond devices i find curious that nobody has said anything about this tool.
correct me if i'm wrong but this tool seems the best tool i've seen until know to check what the hell is draining our devices when in standby or calling or Data downloading or whatever.
i don't know too much about how it runs but i expect to learn it after some tests i will do.
I think that this tool will let me discover which running process is draining my battery, of course this tool won't tell it to me but after some testsenabling and disabling processes i will be able to see which one/s is/are the guilties.
I guess that the "ma" value that changes over time is per hour right? maybe PalladiumTD, could you explain me how to use this tool as if i would be a dummy, ( you can take out the "as if" if you want...
i don't understand the difference between "average current (ma)" and total usage (mah). The display doesn't change its contents when i switch between these options.
BTW it seems to run in my GSM device with unprotected Proven ROM 1.502. of course i can't install Nuepower driver.
is this tool trustable and accurate?
it doesn't seem to instantly react depending of what is running in the phone. it reacts after a while. ( of course i have it set to one second update)
and apparently it jumps from -78ma to -145ma and to -215ma without doing anything special. The Phone is just flashed with only battlog installed and running and it jumps between these values for no apparently reason.
Also i don't see any difference between having HDSP enabled or disabled (GPRS).
if i turn on Wifi, i see a big difference.
The benchmark test doesn't seem to run. it does not call and it does not download anything. (with wifi turned on)
Click to expand...
Click to collapse
I guess I really ought to write a manual?? Anyway the top for windows are live readings REPORTED by windows they are (left to right/top to bottom) Battery level, Battery Voltage, Battery Temperature, Current being drawn from the battery. These are realtime numbers.
The display average current and total Usage options on the update log menu apply to the Phone Current and Data Current boxes. (just below the 4 realtime boxes. These all total the current drawn during a voice or data call and allow you to see the average current drawn from the battery and the total power drawn from the battery (in mAH i.e. milliamp hours).
The fluctations in power draw may be real, I see them on the CDMA version when I am in a low signal area. the -78ma is close to what I have seen with the backlight off, the -120 - 150 ma when the backlight is on at 50%, the -215 may be an intermittent radio transmission (just speculation...)
With HDSP or GPRS enabled, you should only see a current draw when the radio is transmitting or receiving (i.e. when you are transfering data).
As far as accuracy, these numbers will be only as accurate as what the phone hardware is reporting. To keep it as generic as possible, I pull the values through the WinMo SystemStatus interface which makes it a bit sluggish in updating if other applications are running.
To run the test you must have .net Compact framework 3.5 loaded, it will not run without it. With that loaded, the benchmark should run (as it call only native WinMo functions). If you have .netCF3.5 loaded PM me and I'll see if we can troubleshoot..
One difference between the CDMA and GSM versions is that the CDMA hardware only reports current draw (as a positive number) I am guessing that The GSM phones report net current draw in the standard WinMo manner ,( i.e. negative for discharge, positive for charging and the number reported is the sum of the discharge and charging currents).
If there is interest (and someone could provide me with a log file from a GSM Phone while not connected to the charger, then when connected to the charger) I will make a GSM flavor of this utility, but will need some willing beta testers as I don't have a GSM Raphael...
Found via freewarepocketpc.net
Your Battlog utility is described on freewarepocketpc.net here: http://www.freewarepocketpc.net/ppc-download-battlog.html
Unfortunately, the download link points to Batti (another good utility) rather than Battlog.
I downloaded Battlog from your link here, and when trying to run Battlog the first time, I got the following error: BattLog.exe MissingMethodException Method not found: get_CellularSystemConnectedEvdo Microsoft.WindowsMobile.Status.SystemState. at System.Windows.Forms.Form.OnLoad(EventArgs e) at System.Windows.Forms.Form._SetVisibleNotify(Boolean fVis) at System.Windows.Forms.Control.set_Visible(Boolean value) at System.Windows.Forms.Application.Run(Form fm) at BatteryStatus.Program.Main() At the time when I ran this, my phone (PPC-6700 running WM5, .Net CF 2.0 & 3.5 installed) was connected to my computer via ActiveSync, the battery was fully charged, and the WiFi was turned on (and connected, I believe). the 1xRTT connection was not connected, as far as I know (no evdo on Cellular South in our area yet).
I'm no programmer (my expertise is in IT: remote access support, hardware, networking, etc.), but I hope this helps you. This looks like a worthwhile utility that has a lot of promise.
*edit*
I verified that I receive the same error if disconnected from ActiveSync, and after a soft reset. I also receive the error if WiFi is turned off and the 1xRTT connection is established.
Maybe I'm digging too deep; is this WM5 compatible, or WM6 only?
Sorry, I guess this could be a "bug report" and should have been PM'd to you. I didn't see that part until after I'd posted this.
*/edit*
TekServer said:
Your Battlog utility is described on freewarepocketpc.net here: http://www.freewarepocketpc.net/ppc-download-battlog.html
Unfortunately, the download link points to Batti (another good utility) rather than Battlog.
I downloaded Battlog from your link here, and when trying to run Battlog the first time, I got the following error: BattLog.exe MissingMethodException Method not found: get_CellularSystemConnectedEvdo Microsoft.WindowsMobile.Status.SystemState. at System.Windows.Forms.Form.OnLoad(EventArgs e) at System.Windows.Forms.Form._SetVisibleNotify(Boolean fVis) at System.Windows.Forms.Control.set_Visible(Boolean value) at System.Windows.Forms.Application.Run(Form fm) at BatteryStatus.Program.Main() At the time when I ran this, my phone (PPC-6700 running WM5, .Net CF 2.0 & 3.5 installed) was connected to my computer via ActiveSync, the battery was fully charged, and the WiFi was turned on (and connected, I believe). the 1xRTT connection was not connected, as far as I know (no evdo on Cellular South in our area yet).
I'm no programmer (my expertise is in IT: remote access support, hardware, networking, etc.), but I hope this helps you. This looks like a worthwhile utility that has a lot of promise.
Click to expand...
Click to collapse
Unfortunately, the many of the systemstate Classes that I use to detect the usage of the cellular radios aren't directly accessible from managed code in WM5. I will have see if I can code around it or remove some of the functionality when run on WM5. Thanks for the Info BTW.
You're welcome!
I know that I'm in a rapidly dwindling minority of WM5 users, but I love my PPC-6700, batter hog though it is, and I'm getting as much use as I can out of it before I choose a replacement from Cellular South's somewhat limited selection. (And I must admit, I am holding out hope that Google Android powered phones live up to their hype.)
Thank you for looking at this, and I'll be watching for a WM5 version!
Palladium, have you taken a look at the HTC-written TBattery app here?
http://forum.xda-developers.com/showthread.php?t=451646
It might help in your further development. Appreciate your work!
Da_G said:
Palladium, have you taken a look at the HTC-written TBattery app here?
http://forum.xda-developers.com/showthread.php?t=451646
It might help in your further development. Appreciate your work!
Click to expand...
Click to collapse
I will take a look at it (in detail that is) on the 7501A chipset it only reads battery level, temp and current. What is interesting is the from what I have read both the 75xx and 72xx chipsets use the PM7500 power management chip which is where those numbers should be coming from anyway. I'll see what I can learn from it... THANKS!!!
TekServer said:
You're welcome!
I know that I'm in a rapidly dwindling minority of WM5 users, but I love my PPC-6700, batter hog though it is, and I'm getting as much use as I can out of it before I choose a replacement from Cellular South's somewhat limited selection. (And I must admit, I am holding out hope that Google Android powered phones live up to their hype.)
Thank you for looking at this, and I'll be watching for a WM5 version!
Click to expand...
Click to collapse
I have just uploaded a new version. This version supports WinMo 5 as well as improved support for QVGA screens. there are a few other bug fixes and the ability to switch between deg F and C for the temperature display.
It works! Nice job; looks good.
A couple of points of constructive criticism (intended in the most positive and complimentary context!):
- I don't think the temperature is reporting correctly, at least on my Apache. It was reporting in C as 8.82 degrees, then when I switched to F it reported 16.94 degrees. As my phone is not currently in a freezer, I'm fairly certain this is off by a bit.
- The readme in the zip file is from the previous version.
I already love this program, and I've only scratched the surface. I've just started it logging and left it running; hopefully this will help me figure out which of the many apps I have is sucking my battery dry.
Thanks again!
TekServer said:
It works! Nice job; looks good.
A couple of points of constructive criticism (intended in the most positive and complimentary context!):
- I don't think the temperature is reporting correctly, at least on my Apache. It was reporting in C as 8.82 degrees, then when I switched to F it reported 16.94 degrees. As my phone is not currently in a freezer, I'm fairly certain this is off by a bit.
- The readme in the zip file is from the previous version.
I already love this program, and I've only scratched the surface. I've just started it logging and left it running; hopefully this will help me figure out which of the many apps I have is sucking my battery dry.
Thanks again!
Click to expand...
Click to collapse
Unfortunately this is a known bug, that has been low on my priority list. Basically the Standard WM Battery Dll report in degrees C*10, However, the Driver provided with the CDMA Touch Pro, reports in degrees C * 100. I just haven't gotten around to having the program check which platform it is running on and adjusting the calculation accordingly. So in the meantime your real temperature is the reported value * 10
on the readme, I'll re-upload with the updated Readme this afternoon.
Multiply by 10, got it. I can do that.
Thanks!
Thanks for this handy little program. I am used to the HTC WinMo battery drain devices but always like to see what might be aiding in the drainage.
HI Palladium,
Thumbs up for nice work.
However I'm not able to run the 0.2.3.130 release on my O2-Graphite w. WM 6. I get: "This app. requires a newer ver. of the microsoft.NET Compact Framework"
Hope this information is of any use to you
Regards
Steen
Sorry, My original post wasn't clear on this point, all versions from 2.1.105 forward require .NET CF 3.5. If you download it here Everything should work fine after you install that.
Palladium TD.....
Thanks to a post on another thread, I found your very nicely done program. I have some 'funny' out of the blue battery drains that your program may help me isolate.
I have v0.2.3.130 beta installed on my US ROM1.93 GSM Diamond. It installed fine, tracks current use and temp (/10) and logs the data. Very easy to pull into excel for analysis. I've only used it for a day now and do note the following issues (which may be me not executing correctly or the beta)
Voltage and temp graphs track exactly together and without a scale value
No response from the two Icons on the left (one looks like a help?)
I don't see any response to the "Display Total Usage mAh
Very excellent work, and if I can support, let me know.
Thanks
I've been testing copilot 8 on my Desire G7 lately. I keep the GPS turned-off, when I'm not planning on using the navigation application and I activate it before i start the navigation app (copilot). At start up, it always takes the app 1-3mins or a longer time to determine it's current location.
Could anyone with off-line GPS experience tell me whether this kind of performance is within the normal boundry? A friend told me that leaving the GPS activated at all time will solve the problem. Is it true? If I do that will the GPS constantly sync satelite signals and causing it to wear out faster than usual?
yrneh122 said:
I've been testing copilot 8 on my Desire G7 lately. I keep the GPS turned-off, when I'm not planning on using the navigation application and I activate it before i start the navigation app (copilot). At start up, it always takes the app 1-3mins or a longer time to determine it's current location.
Could anyone with off-line GPS experience tell me whether this kind of performance is within the normal boundry? A friend told me that leaving the GPS activated at all time will solve the problem. Is it true? If I do that will the GPS constantly sync satelite signals and causing it to wear out faster than usual?
Click to expand...
Click to collapse
if you keep your gps enabled, it will already have a satellite fix before you run your gps apps, so there will be no waiting time, the gps dosent constantly sync satellites, it just keeps a fix, it will use more battery, and it will also use the gps for "my location" but it isnt a noticeable battery drain!! try it, if you think it drains your battery you can always disable it
AndroHero said:
if you keep your gps enabled, it will already have a satellite fix before you run your gps apps, so there will be no waiting time, the gps dosent constantly sync satellites, it just keeps a fix, it will use more battery, and it will also use the gps for "my location" but it isnt a noticeable battery drain!! try it, if you think it drains your battery you can always disable it
Click to expand...
Click to collapse
I'm glad to know that keeping the GPS activated will solve the problem. I'm not too worried about battery drain, since I can charge it frequently and the battery is at least replacable. My understanding is that if the GPS can maintain a "fix" on my current location, it should be synchronizing location data with satelites. My location varies throughout a day. It would mean to me that the "fix" on my location a GPS has is dynamic information. For a machine to maintain dynamic information, it must be processing some type of data at a particular frequency, which means that the GPS will be frequently operating. Is there anything wrong with my speculation?(I'm new to these stuff. Sorry if i'm wrong.)
My biggest concern is that whether keeping the GPS activated 24/7 will cause it to wear out at a abnormal rate. Have you had experience in taking such a practice with your mobile phone for a long period of time while your GPS hardwares still function at normal performance.
Thank you for helping me with my questions.
I've been using Co-pilot with my Desire ever since I got it (end of March) and GPS has always 'fixed' in less than 30 secs.
Recently I was trying out different radio ROMs (for non-GPS related issues) and there was one that took an awful long time to obtain a GPX fix. I switched to another one and all was fine again.
NB: My GPS is off and only switched on if and when an app requires it.
sh500 said:
I've been using Co-pilot with my Desire ever since I got it (end of March) and GPS has always 'fixed' in less than 30 secs.
Recently I was trying out different radio ROMs (for non-GPS related issues) and there was one that took an awful long time to obtain a GPX fix. I switched to another one and all was fine again.
NB: My GPS is off and only switched on if and when an app requires it.
Click to expand...
Click to collapse
May I ask what rom/radio you've been using? Do you still remember which rom/radio didn't work quite well with your copilot? How often do you use your copilot?
I was having the same problem with the stock ROM & radio (2.2 latest OTA). Haven't tried yet with the new ROM (Leedroid 2.2d), but I was wondering if there was an Android application similar to QuickFix, Ephemeris or such on WinMo (downloading some satellite positioning data every few days to keep an approximate mapping of satellites, thus helping for a quick fix).
I also keep my GPS off unless an application requires it (e.g. navigation).
yrneh122 said:
May I ask what rom/radio you've been using? Do you still remember which rom/radio didn't work quite well with your copilot? How often do you use your copilot?
Click to expand...
Click to collapse
Well even with a stock ROM and radio (T-Mobile UK), which I had running for about 1.5 months, Co-Pilot still used to have a fast fix.
Later I upgraded to the earlier MCR ROMs (leaving the stock radio) and it was still fine.
After that I went to Froyo LeeDroid ROM (v2.1a) and upgraded the radio to meet minimum requirement (I forget what version it was). With this same ROM I am now using 32.48.00.32.
I can't remember what radio version didnt work well with GPS, but it was one that many threads stated it had a 'better battery life'.
I don't use Co-pilot that often but whenever I do, the fix is reasonably quick.
I used PhoneWeaver on all my WindowsPhone devices and it was awesome. I use it on my EVO4G and it works great. I tried it on my 2nd EPIC and got the same results as the first one- horrible battery life. I went through my extended battery in about half the time.
Does anyone else have this experience? I really like the app on every other device I've tried it on. I know it's not Android because of the EVO. I know it's not the ROM, because I've tried it on stock & various rooted.
Does anyone have any tips that might make it more keep-able?
You have to make sure you're not using GPS for your locations. It eats up the battery. Use the other option of cell network (can't think of the exact wording).
Verify if PhW appears in the battery usage app list. It shouldn't.
Also activate PhoneWeaver log mode, leave it run for a day and take a look to the log to see if tere are any recurring events that shoulden't be.
As Earldom says, using alert location method and having GPS always on will make the OS to use the GPS for the location system. Ensure you have GPS off or change to Intelligent polling or AP+Cell in location method.
Finally, ensure you have the latest 2.2.8 version installed.