iRex 800 quickly discharging

[badbob001]

In sourcecode file sysd-3.0+svnr5463\sysd\src\system.c, there are commands for shutdown, so someone just needs to compile some of those commands together. The initial command appears to be 'do_poweroff(), and it triggers a bunch of processes like to shutdown the inputs, unmount storage, and etc before the final "shutdown -h now".

This part is most interesting:

void sys_standby(void)
{
    LOGPRINTF("entry");

    // shutdown completely when standby switch is toggled during first boot of the device
    if (conf_get_first_boot() && (device_state == STATE_DEVICE_STARTING))
    {
        busy_add_background(0);                    
        sys_spawn_sync("shutdown -h now");
        return;
    }

It's saying that during the first boot of the device, flicking the standby switch will shutdown the device. I guess it's also worth a try to reboot the device and see if flicking the switch again during startup does anything. If not, then one needs to trick the device into thinking it's starting for the first time and then try again.

Note that the sourcecode is for both the dr800 and dr1000 so some things may only be implemented on one device.

[Shaggy]

Originally Posted by badbob001

In sourcecode file sysd-3.0+svnr5463\sysd\src\system.c, there are commands for shutdown, so someone just needs to compile some of those commands together. The initial command appears to be 'do_poweroff(), and it triggers a bunch of processes like to shutdown the inputs, unmount storage, and etc before the final "shutdown -h now".

Good find!

[OrcaBlue]

Perhaps, the code is already compiled and all we need is a registry setting within the dr.ini file to activate it.

Something like "shutdown=true"?

[badbob001]

It's also possible that some sort of process is running in the background that is keeping the device more active than needed. For example, when you open a file, it loads a separate viewer for that file type: pdf viewer for pdfs, image viewer for images, epub viewer for epubs, and etc. It has been shown that viewing a particular file type for the first time takes longer than subsequent times, which seems to imply that for subsequent use, the viewer is still loaded and ready. This may explain why results are so varied.

I don't know how the device handles task management. That is, when does it shutdown a viewer, if ever? At standby? When switching to another viewer? When a certain resource threshold limit is reached? One way to rule out a runaway or crashed process ruining battery life is to reboot the device before testing and don't do anything with the device. For comparison measurements, it may also help to limit usage to particular features to see if one of them is faulty (example scenario: only view one particular file type and don't turn on 3G").

[OrcaBlue]

Originally Posted by badbob001

It's also possible that some sort of process is running in the background that is keeping the device more active than needed. For example, when you open a file, it loads a separate viewer for that file type: pdf viewer for pdfs, image viewer for images, epub viewer for epubs, and etc. It has been shown that viewing a particular file type for the first time takes longer than subsequent times, which seems to imply that for subsequent use, the viewer is still loaded and ready. This may explain why results are so varied.

I think you right. I tried loading "stardict" on DR800sg yesterday. It seemed that it loaded but didn't show up on the screen. I suspended and locked the reader last night. It dropped from 98% to 93% this morning. That's a big drop as compared to the prior night when it dropped only 1%. I think if we reboot prior suspending the reader, it will reduce battery drain.

[Shaggy]

Originally Posted by badbob001

It's also possible that some sort of process is running in the background that is keeping the device more active than needed. For example, when you open a file, it loads a separate viewer for that file type: pdf viewer for pdfs, image viewer for images, epub viewer for epubs, and etc. It has been shown that viewing a particular file type for the first time takes longer than subsequent times, which seems to imply that for subsequent use, the viewer is still loaded and ready. This may explain why results are so varied.

I don't know how the device handles task management. That is, when does it shutdown a viewer, if ever? At standby? When switching to another viewer? When a certain resource threshold limit is reached? One way to rule out a runaway or crashed process ruining battery life is to reboot the device before testing and don't do anything with the device. For comparison measurements, it may also help to limit usage to particular features to see if one of them is faulty (example scenario: only view one particular file type and don't turn on 3G").

Maybe, but we'd need more testing to figure it out. One of the theories about the DR1000 power drain bug is that the PDF viewer is getting hung during shutdown, and prevents the device from fully powering off. This is a completely different PDF viewer than what is in the current DR800 firmware though. It's possible that the problem mgmueller had was similar with his 4 day battery, and there was a viewer/process that didn't let the device go into standby correctly for some reason.

What I can say is that during my initial testing where I got two weeks, I pretty much had a PDF document open the entire time. I didn't do anything special to close the document or reboot before I would put it into standby. So it would appear that, in my case, the PDF viewer acted correctly. Perhaps ePub behaves differently though? It's at least worth testing.

[TGS]

Originally Posted by Shaggy

What I can say is that during my initial testing where I got two weeks, I pretty much had a PDF document open the entire time. I didn't do anything special to close the document or reboot before I would put it into standby. So it would appear that, in my case, the PDF viewer acted correctly. Perhaps ePub behaves differently though? It's at least worth testing.

I also only read PDFs, and seem to get quite good longevity out of a charge. I wonder if there is something in the suggestion that a viewer remains open - even in standby - and that a partial explanation might be that if you have read documents using a variety of viewers they all remain open until rebooting?

[Shaggy]

Originally Posted by TGS

I also only read PDFs, and seem to get quite good longevity out of a charge. I wonder if there is something in the suggestion that a viewer remains open - even in standby - and that a partial explanation might be that if you have read documents using a variety of viewers they all remain open until rebooting?

Whether they remain open or not probably isn't an issue. What's supposed to happen is that they suspend when the OS gets suspended to RAM/Flash/whatever, and resume running when it wakes up. I don't think standby mode will actually close/re-open the applications. What could potentially cause a problem is if for some reason one of the viewers prevents the OS from suspending correctly. I don't know if that is happening or not, but it's a valid theory until we can test it more.

Right now what we know is that there seems to be some discrepancy between the amount of standby time a couple people are getting. I think it's way too early to really know why, yet, but all of these different ideas are good things to look at (viewers preventing standby, stylus proximity, etc). We should be able to test in a controlled way in order to narrow some of them down.

[CoolDragon]

Hmm, I got one issue last night on my 800SG:

I was reading a PDF file before going to sleep, and I forgot to put it into standby. This morning when I woke up, the 800SG is still on, and drained 25% battery. This is the first time it didn't go to standby automatically.

[OrcaBlue]

Originally Posted by CoolDragon

Hmm, I got one issue last night on my 800SG:

I was reading a PDF file before going to sleep, and I forgot to put it into standby. This morning when I woke up, the 800SG is still on, and drained 25% battery. This is the first time it didn't go to standby automatically.

Where was your stylus that whole time?

[CoolDragon]

I was not using the stylus at all.

[mgmueller]

Just got an eMail (proactively, without me contacting them) from Thom, the service manager of iRex. They follow this (and others) thread and opened a service case.
I have to take back my harsh words from post #25 https://www.mobileread.com/forums/sho...0&postcount=25, of course that's excellent service.
I apologise. I guess, my iLiad just dying on me and in parallel iRex 800 suddenly being totally discharged got me "on the wrong foot".

Thom's eMail:
"I'm the service manager at IREX and have been pointed to your post on mobileread.
https://www.mobileread.com/forums/sho...t=76835&page=9

I created this case to give follow up on the problem you and few other DR800 users experience.

I want to let you know that DR800's standby time should be close to two weeks; 4 days is too short.
We are working on this issue and we'll let you know what and when we will do about this behavior."

Again: That's brillant service. Same happened before with some of their developers. Haven't seen that with Amazon or B&N or Apple...

[Dellaster]

That's an encouraging and unusually helpful response, mgmueller. Thanks for sharing it.

[TGS]

Originally Posted by mgmueller

Thom's eMail:
"I'm the service manager at IREX and have been pointed to your post on mobileread.
https://www.mobileread.com/forums/sho...t=76835&page=9

I created this case to give follow up on the problem you and few other DR800 users experience.

I want to let you know that DR800's standby time should be close to two weeks; 4 days is too short.
We are working on this issue and we'll let you know what and when we will do about this behavior."

Again: That's brillant service. Same happened before with some of their developers. Haven't seen that with Amazon or B&N or Apple...

It is brilliant service, (but are you sure it wasn't Shaggy in disguise who sent the email (cf. recent comments on this thread!)).

[superhero]

Originally Posted by CoolDragon

Hmm, I got one issue last night on my 800SG:

I was reading a PDF file before going to sleep, and I forgot to put it into standby. This morning when I woke up, the 800SG is still on, and drained 25% battery. This is the first time it didn't go to standby automatically.

As I stated earlier in the thread, my 800sg is also not automatically going into standby no matter how long I leave it on. It will go into standby if I manually do it. Isn't it likely that this is related to the 4 day drain issue...25% per loss day x 4 days=100% drain.