Page 2 of 2 FirstFirst 12
Results 21 to 40 of 40
  1. #21  
    Quote Originally Posted by ivhs72 View Post
    This conclusion was based on the data that was in the program. Of all of the applications that were used, Launcher logged more time than any others. Based on the % of use from the program, it used 75%. It logged more that 7000 seconds of access. The next closest one was AppLog with around 1200 seconds.
    You're coming to the wrong conclusion, so please, why do you continue to say this over and over?? It's not the "Launcher", it's something calling the Launcher. This might be like a fire might burn a house down, but something caused the fire.

    Try deactivating Battery Graph, reset the stats in the AppLog program and see what happens. Wouldn't be surprised if "Launcher" stopped mysteriously "launching itself." Battery Graph never worked "well" on my 650.. It caused all kinds of trouble for me, mainly high battery drain as it fought with other applications.
  2. #22  
    Well, I've no serious problems with BG on a 650, other than not reliably reporting accurate on & off switches, and therefore on vs. off time. Problem is, that's supposed to be one of its main features. The other info it provides seems to me to be accurate (almost two yrs on the 650).

    The problem with it, seems to me, is it does not always interpret on/off cycles properly. At least while not charging the phone, it interprets on/off properly, but it also interprets other activity as on/off. I'll check how this compares to how it behaves when charging in a minute.

    FWIW, I believe each record in its DB is supposed to correspond to either an on or an off switch, with each record containing current powered status, time, and batt stats.
  3.    #23  
    Quote Originally Posted by sblanter View Post
    Well, I've no serious problems with BG on a 650, other than not reliably reporting accurate on & off switches, and therefore on vs. off time. Problem is, that's supposed to be one of its main features. The other info it provides seems to me to be accurate (almost two yrs on the 650).

    The problem with it, seems to me, is it does not always interpret on/off cycles properly. At least while not charging the phone, it interprets on/off properly, but it also interprets other activity as on/off. I'll check how this compares to how it behaves when charging in a minute.

    FWIW, I believe each record in its DB is supposed to correspond to either an on or an off switch, with each record containing current powered status, time, and batt stats.
    How do you read the DB?
  4. #24  
    As far as I can tell, ya can't.
  5. #25  
    But you can read the website for info, and arrive at some conclusions by playing with its settings, turning phone on/off, and watching what happens to the db. Open the db with filez and look at a record. There's more bytes in each record than necessary just to store time. If I remember, the first record is some kind of header, and each subsequent record represents an event (eg, on or off).
  6. #26  
    Also note that each time you turn the device on or off, the number of records in the database increments by one record.
  7. #27  
    One more thing. If you do try & use Filez to open & read a DB record, you'll see you cannot readily interpret the data.
  8.    #28  
    Quote Originally Posted by khaytsus View Post
    You're coming to the wrong conclusion, so please, why do you continue to say this over and over?? It's not the "Launcher", it's something calling the Launcher. This might be like a fire might burn a house down, but something caused the fire.

    Try deactivating Battery Graph, reset the stats in the AppLog program and see what happens. Wouldn't be surprised if "Launcher" stopped mysteriously "launching itself." Battery Graph never worked "well" on my 650.. It caused all kinds of trouble for me, mainly high battery drain as it fought with other applications.

    I did what you suggested. I did a hard reset and installed AppLog. The only programs that I have installed (outside of the programs that come with the Treo in ROM) are Palm eReader, MobiPocket 5, Kinoma 3 Player, Ringo Pro, and AppLog. Nothing else. I have my network set to Tzones. I am using stock VersaMail and Blazer. Bluetooth is off, IR is off, send to headset is off. (This is the same way it was set up before, only with BatteryGraph).

    After three hours of use, Launcher shows 3100s of usage, the most of any other program. The closest is AppLog, with 174s. Launcher has accessed the device 136 times. The next closest is AppLog with 11 times. The maximum time Launcher accessed the device was for 1212 seconds, AppLog for 64s. You can add every other program/application that shows in AppLog, add them together, and it doesn't come to 10% of Launcher. All of this within three hours.

    So the question remains: What is going on with Launcher? I've tried to answer that question and share what I have done with everyone here. Why don't you do something instead of just sitting back and saying that Launcher isn't the problem, and I've come to the wrong conclusion. Why don't you do some testing like I have and share the results. I've done my testing with two Treo 680s from Palm, and my own Palm-brand unlocked 650 (BTW, BatteryGraph worked the same in the 650 as it did in both of the 680s). Now it's your turn.

    I don't have an axe to grind and I don't need to feel that I am making other people look bad by demeaning them. I only want my phone to work properly, nothing else. I've been in the electronics repair field for 35 years, and know how to repair and diagnose problems in complex electronic devices. I'm not some schmuck with a paper degree thinking about what is wrong with this phone. I am approaching it in a logical and systematic method, one which will hopefully help all of us get the most out of our phones. I've used this method in the past and it has served me well. I know it will this time.


    Last edited by ivhs72; 01/02/2007 at 12:18 AM.
  9.    #29  
    Quote Originally Posted by sblanter View Post
    Also note that each time you turn the device on or off, the number of records in the database increments by one record.

    I understand that. But I am getting hundreds and thousands of events taking up a lot of memory. That is what led me to start this thread. Someone suggested using AppLog which shows what programs are being accessed, and when, but the data is complete only if you have the full version, not the trial one. Even then it doesn't give running data, only cumulative. I don't know of any other way to see what's going on. Maybe someone else does.

  10.    #30  
    Quote Originally Posted by sblanter View Post
    But you can read the website for info, and arrive at some conclusions by playing with its settings, turning phone on/off, and watching what happens to the db. Open the db with filez and look at a record. There's more bytes in each record than necessary just to store time. If I remember, the first record is some kind of header, and each subsequent record represents an event (eg, on or off).
    Been there, done that, got the t-shirt.....

  11. #31  
    Well, I think the issues with BG may be of the same sort as possible issues with the applog app. I know BG hasn't been updated in a long time, and does not properly interpret the data it sees. Maybe it's the same situation with applog. Still, doesn't mean the data is useless, just means take it with a grain of salt.
  12. #32  
    Wow, I didn't get a T-shirt!
  13.    #33  
    Quote Originally Posted by sblanter View Post
    Well, I think the issues with BG may be of the same sort as possible issues with the applog app. I know BG hasn't been updated in a long time, and does not properly interpret the data it sees. Maybe it's the same situation with applog. Still, doesn't mean the data is useless, just means take it with a grain of salt.
    My confusion comes from having each program installed by itself, and with both of them installed at the same time on multiple phones, I see the same trends in the data. The programs just aren't written to give the amount of detail that one needs when troubleshooting software/firmware issues. This is the sort of issue that Palm design engineers would have the tools to address. They have access to the code, the design concepts, and the people that wrote the software and built the hardware. We don't have that. Yet!

  14. #34  
    Agree. If you have two 680s, identically configured & used, same FW & OS, you should see the same results. But if you have two different pieces of hardware.... probably not.
  15.    #35  
    Quote Originally Posted by sblanter View Post
    Agree. If you have two 680s, identically configured & used, same FW & OS, you should see the same results. But if you have two different pieces of hardware.... probably not.

    They were identical, same firmware, OS, everything. When I got my first one from Palm, I was at the battery life, and called and complained, wanting a new battery. They sent me a whole new phone. (I'm still waiting for them to send me the information on sending the first one back. Maybe they'll forget ). I ran the same apps on both of them, both with and without the SIM card installed, and saw the same results. In fact, I'm running both of them right now and seeing a modest improvement in battery life. I also ran the same apps on my 650 and saw different results using those programs. But then I was seeing far better battery life on the 650.

    One thought about BatteryGraph. It shows the voltage on the monitor, but it doesn't mention the mah of the battery. Since the stock 650 battery has a more mah that the 680, it would be interesting to see how the program calculates the rate of battery running down. My guess is that since the drain is based on current instead of voltage, and you can only measure current in series, any program that shows the current drain would affect the current drain. Think back to Ohm's Law. It has to all be calculated since this is a software program, and there is no current meter in the Palm. Wow, this is giving me a headache I haven't worked with things like this in a few years.

    Tomorrow I'm calling Palm tech support and asking to talk to someone who isn't in India and has an American accent. There has to be some way to talk to a live body who knows something.



    BTW, it was a CHEAP t-shirt.
  16. #36  
    Quote Originally Posted by ivhs72 View Post
    I did what you suggested. I did a hard reset and installed AppLog. The only programs that I have installed (outside of the programs that come with the Treo in ROM) are Palm eReader, MobiPocket 5, Kinoma 3 Player, Ringo Pro, and AppLog. Nothing else. I have my network set to Tzones. I am using stock VersaMail and Blazer. Bluetooth is off, IR is off, send to headset is off. (This is the same way it was set up before, only with BatteryGraph).

    After three hours of use, Launcher shows 3100s of usage, the most of any other program. The closest is AppLog, with 174s. Launcher has accessed the device 136 times. The next closest is AppLog with 11 times. The maximum time Launcher accessed the device was for 1212 seconds, AppLog for 64s. You can add every other program/application that shows in AppLog, add them together, and it doesn't come to 10% of Launcher. All of this within three hours.
    $AppRan = $AppRan + $EndTime - $StartTime;

    You realize Ringo is a background application, yes?
  17.    #37  
    Quote Originally Posted by khaytsus View Post
    $AppRan = $AppRan + $EndTime - $StartTime;

    You realize Ringo is a background application, yes?

    Ringo shows 5s usage for 0.1%, a count of 1 for 0.2%, max usage of 5s, min usage of 5s.

    Launcher shows 5323s for 71.4%, count % of 89.6 (count # is one of the unreg fields), max usage of 1212s, min usage of 2s.

    These are all after being installed for 24 hours.

    So what if Ringo is running in the background?

    Next guess?


  18. #38  
    Quote Originally Posted by ivhs72 View Post
    Ringo shows 5s usage for 0.1%, a count of 1 for 0.2%, max usage of 5s, min usage of 5s.

    Launcher shows 5323s for 71.4%, count % of 89.6 (count # is one of the unreg fields), max usage of 1212s, min usage of 2s.

    These are all after being installed for 24 hours.

    So what if Ringo is running in the background?

    Next guess?


    I have 6 background apps, none of which show significant usage, yet I very well know they're all running.

    I'm done with this conversation. Maybe you can use the ROM tool and delete that nasty Launcher app.
  19. #39  
    I used BatteryGraph to monitor my 680 over a 4-day period and there were several thousand events in my database. According to the stats, my 680 was active on an average of every 17 seconds! This is moderate use during the day, sitting on the charger at night with everything turned off.

    I'm not sure what this means except that it seems my 680 is working really hard when it should be sleeping. Also, perhaps there is a connection between the "hyper" 680 and the innacuracy of the clock. I've NEVER had more difficulty with ANY PDA or phone with accurate timekeeing like I have with this 680. How hard can it be to keep proper time on this thing??
  20.    #40  
    I finally took the plunge and paid the shareware fee for AppLog to get rid of those <unreg> cells. Very interesting results.

    For one thing, Launcher is not showing the usage that it did before. It is still the largest user at 45.7%, but only has a 2095s. I reinstalled BatteryGraph is second at 857s and 18.7%.

    Another interesting item is that the % number do not add up to 100%, only about 75%. Some apps only show 0.1 to 0.3% usage. I'm not sure what all of this means.

    Like BriGuy, I still see a lot of DB access on BatteryGraph, and there are times when it looks like a solid red patch, other times solid white. I am to the point that I this phone is going back, and I'll head to Tmobile and look at a DASH. It's a Smartphone running WM5, but it does have builtin WiFi and BT.

Page 2 of 2 FirstFirst 12

Posting Permissions