Beta Pandora Hotfix 3 - Beta 2


todd said:
I was amused to discover that the "low power mode" switch seems to make a good "pause" button when listening to music. Often when I hit it again, the music will pick back up where it left off (using Exaile). Occasionally it won't.

This reminds me. What does the "Hold Mode" Do? (When you flick the switch to the left) I thought it was supposed to work like the Hold mode on CD Players or the Lock mode on MP3 players, but it does not seem to do that, or anything in fact. Care to share some light on this EvilDragon?

-God Ginrai
 
Last edited by a moderator:
I find that if you , after open the lid and the LCD comes on, simply switch the power switch again to the right and then give it like 5 to 10 seconds it seams to work quite well. I had to reboot one time to get the system back out of it (LCD stayed dark) but that might just be me doing it to fast after each other :) .

I simply move the nub/mouse - if it's slow to react I know it's in "low power mode " , once the mouse reacts fine after switching the power switch the Pandora is back to "normal" :)

If wifi was disabled before going to "Low power" it does not seams to enable it again - not done this to often however.
If wifi was enabled before going to "Low power" it does enable the wifi again but fails (not always but often) to indicate this with the Wifi LED, provoking confusion :)
 
God Ginrai said:
todd said:
I was amused to discover that the "low power mode" switch seems to make a good "pause" button when listening to music. Often when I hit it again, the music will pick back up where it left off (using Exaile). Occasionally it won't.

This reminds me. What does the "Hold Mode" Do? (When you flick the switch to the left) I thought it was supposed to work like the Hold mode on CD Players or the Lock mode on MP3 players, but it does not seem to do that, or anything in fact. Care to share some light on this EvilDragon?

-God Ginrai

Simple: It's not yet included ;)
 
Last edited:
God Ginrai said:
You are calling this a bug, but it is not. The screen turns on because the function for closing the lid is to turn the screen off and the function for opening the lid is to turn the screen on. Even in low power mode, the system runs the same. The only difference is that when you enter low power mode, the system disables some services (like WiFi) and downclocks the CPU.

-God Ginrai

If it looks like a duck and quacks like a duck...

If it is not a 'bug' then it is simply a very poor design choice that should be changed. The screen switches off when pushing the power switch even if the lid is open - so it seems to imply that the lower power mode should mean "the LCD is off". It shouldn't just turn back on without any indication that it is still in low power mode just because I opened the lid. To a user who doesn't know that staying in low power mode while turning the LCD on is intended - it simply looks/feels like my Pandora is unresponsive and not working correctly. Since the Pandora is completely unusable while in low power mode - there is no reason for the LCD ever to be on when it is in this state.

I design software for a living and if it wasn't even intuitive to me that the current behavior is 'correct' - then I'm sure less astute users would be even more likely to think there is a bug.
 
Last edited by a moderator:
It would be more clear, if a warning message is displayed as long as low power mode is active. So if the User has the Pandora in Low Power mode and only turns on the LCD (opening the lid), he instantly sees the message "low power mode" until he switch back to full power mode.
Again some small but nice improvements of usability I guess, no more confusion. ;)
 
MooTheKow said:
If it is not a 'bug' then it is simply a very poor design choice that should be changed. The screen switches off when pushing the power switch even if the lid is open - so it seems to imply that the lower power mode should mean "the LCD is off". It shouldn't just turn back on without any indication that it is still in low power mode just because I opened the lid. To a user who doesn't know that staying in low power mode while turning the LCD on is intended - it simply looks/feels like my Pandora is unresponsive and not working correctly. Since the Pandora is completely unusable while in low power mode - there is no reason for the LCD ever to be on when it is in this state.
Emphasis mine. This is IMO the crux of the problem - there is no feedback whatsoever as to whether you're in low power mode or not when the lid is opened. Just showing the desktop (or menu, or whatever) as it was before suggests it's woken up, when it hasn't. Leaving the screen off would be a better way to deal with it. Maybe later some kind of instructional screen saying how to leave low power mode could be shown. Some feedback that the thing is actually in the process of waking up would be nice too. Just now you can't be sure you held the switch long enough and I've more than once sent it straight back into low power mode by pressing it again.
 
Last edited by a moderator:
SteveM said:
Emphasis mine. This is IMO the crux of the problem - there is no feedback whatsoever as to whether you're in low power mode or not when the lid is opened. Just showing the desktop (or menu, or whatever) as it was before suggests it's woken up, when it hasn't. Leaving the screen off would be a better way to deal with it. Maybe later some kind of instructional screen saying how to leave low power mode could be shown. Some feedback that the thing is actually in the process of waking up would be nice too. Just now you can't be sure you held the switch long enough and I've more than once sent it straight back into low power mode by pressing it again.

Agreed. To me it seems the screen simply should not ever turn back on while in low power mode unless it is in the process of waking. There is absolutely nothing you can do with the Pandora until you take it out of low power mode - so why turn the screen back on at all?
 
Last edited by a moderator:
MooTheKow said:
Agreed. To me it seems the screen simply should not ever turn back on while in low power mode unless it is in the process of waking. There is absolutely nothing you can do with the Pandora until you take it out of low power mode - so why turn the screen back on at all?
Yep - flicking the switch for the first time turns the screen off. How is the user supposed to work out that a) it is in 14MHz clock mode, and b ) that flicking the switch again won't turn the screen off.
Two options:
* Lid open leaves low power mode.
* Power switch leaves screen on, displays 'low-power' dialogue which can be dismissed either by switch or click.

Has anyone else found that the WiFi light stays off after using low-power mode?
 
Last edited by a moderator:
tsh said:
* Lid open leaves low power mode.
I agree with this.
Or just ignore the lid contact entirely. Either way, that's what the rest of the world does.

edit:
What _I_ would prefer if I had a Pandora, is activating the low power mode when closing the lid and configuring the power switch to turn off the device. That's the most intuitive way, imo. There could be a menu item that allows to switch off low power mode when closing the lid for those who want to play music with closed lid.

edit:
It would even be possible to integrate music playing functionality into the low power mode, if the CPU clock was higher and the player got permission to continue playing while other applications are stopped.

edit:
Finally, I don't think a real suspend to RAM should be targeted at all. If too much effort is put into a feature that has its origin in x86 power-wasting technology, a proper low power mode, which is a _core_ feature of ARM, might not get enough attention. You can have neat things on an always-on system which you can't have if you suspend.
 
Last edited by a moderator:
tsh said:
Has anyone else found that the WiFi light stays off after using low-power mode?

Polossatik said:
If wifi was enabled before going to "Low power" it does enable the wifi again but fails (not always but often) to indicate this with the Wifi LED, provoking confusion :)
 
Last edited by a moderator:
mali said:
edit:
Finally, I don't think a real suspend to RAM should be targeted at all. If too much effort is put into a feature that has its origin in x86 power-wasting technology, a proper low power mode, which is a _core_ feature of ARM, might not get enough attention. You can have neat things on an always-on system which you can't have if you suspend.

I'm Sorry, but You don't know what you are talking about. Suspend to RAM is the technology that was used on the Palm devices to give them that "instant on". It's also used on the DS when you close it up. Suspend to RAM is the most useful function a handheld can have, as far as power conservation is concerned, IMHO.

Oh, and this Low power mode we have currently is in the complete opposite direction of what any low power mode we should have using the ARM. For that, we would need probably kernel improvements, which would also most likely enable us to have Suspend to RAM and Suspend to Disk anyway.

-God Ginrai
 
Last edited by a moderator:
mali said:
tsh said:
* Lid open leaves low power mode.
I agree with this.
Or just ignore the lid contact entirely. Either way, that's what the rest of the world does.

edit:
What _I_ would prefer if I had a Pandora, is activating the low power mode when closing the lid and configuring the power switch to turn off the device. That's the most intuitive way, imo. There could be a menu item that allows to switch off low power mode when closing the lid for those who want to play music with closed lid.

IMHO the biggest saving now in "low power mode" comes from disabling the wifi and the screen.
Personally I would not want the pandora to disable also the wifi when closing the lid.

closing lid = turning off screen
using power button briefly = low power mode , turn off screen and turn wifi off if it's enabled
opening lid = turn screen on if it was simply closed before, turn wifi also back on if previous action was "low power" and wifi was at that moment on

the fact to slow down the cpu a lot during "low power" needs to be seen if it actually make such a big difference, I think it would be good to ask Mwestons advice on this if clocking this down so far put the ARM soc in a "super saving mode" or so.
If so, then it might be good if this can be left to the user - "full saving" and " MP3 player" mode of "low power" for example.

edit: I think this and suspend to ram might be simply complimentary, like said it would be nice to cut energy usage while still doing things like playing music.

edit2: Once suspend to RAM there is I think the "super slow cpu" will not be that usefull, but until it is available..
 
Last edited by a moderator:
God Ginrai said:
I'm Sorry, but You don't know what you are talking about. Suspend to RAM is the technology that was used on the Palm devices to give them that "instant on". It's also used on the DS when you close it up. Suspend to RAM is the most useful function a handheld can have, as far as power conservation is concerned, IMHO.

Oh, and this Low power mode we have currently is in the complete opposite direction of what any low power mode we should have using the ARM. For that, we would need probably kernel improvements, which would also most likely enable us to have Suspend to RAM and Suspend to Disk anyway.

-God Ginrai
I can't comment on Palm as I don't own one and I also think older ARM devices were not designed to be always on.
I consider my N800 to be not _that_ old with its OMAP2 and it doesn't have a suspend mode, neither do current smartphones. When you suspend you'd lose too much functionality, imo. A properly optimized low power mode should give a standby time of around two to three weeks and that is enough, imo.
 
Last edited by a moderator:
mali said:
God Ginrai said:
I'm Sorry, but You don't know what you are talking about. Suspend to RAM is the technology that was used on the Palm devices to give them that "instant on". It's also used on the DS when you close it up. Suspend to RAM is the most useful function a handheld can have, as far as power conservation is concerned, IMHO.

Oh, and this Low power mode we have currently is in the complete opposite direction of what any low power mode we should have using the ARM. For that, we would need probably kernel improvements, which would also most likely enable us to have Suspend to RAM and Suspend to Disk anyway.

-God Ginrai
I can't comment on Palm as I don't own one and I also think older ARM devices were not designed to be always on.
I consider my N800 to be not _that_ old with its OMAP2 and it doesn't have a suspend mode, neither do current smartphones. When you suspend you'd lose too much functionality, imo. A properly optimized low power mode should give a standby time of around two to three weeks and that is enough, imo.

And What's wrong with having both? I never said anything was wrong with an actual low power mode. I just said that the "low power" mode that we have now is nothing but a bug-ridden sham.

EDIT: Oh, and IMHO, you don't lose functionality when you suspend to ram because it comes back almost instantly.

-God Ginrai
 
Last edited by a moderator:
^ You can put effort in doing both, then you have a standby mode of 2-3 weeks and a suspend that gives maybe 2 weeks more, if at all. In standby you can still pull your mails/RSS off a server, receive Voip calls, etc. It's much more convenient, imo.

(edit: moved last post to new topic merged with old thread http://www.gp32x.de/board/index.php?/topic/54924-battery-flat-after-3-4-days/ )
 
Last edited by a moderator:
mali said:
In standby you can still pull your mails/RSS off a server, receive Voip calls, etc. It's much more convenient, imo.

Point taken. However, not everyone needs to do those things, so I think effort should still be put forth into making both solutions, so the user can choose the option that best fits them.

-God Ginrai
 
Last edited by a moderator:
Palm did not suspend to ram.

We could exit low power mode on open, or display a 'in low pow mode' box, pretty easily

Jeffphone
 
Suspend to RAM ... isn't everything in RAM already? Since there is no disk cache or virtual memory on disk anyway on the Pandora? :p Since you need power for this, isn't this virtually the same as what the Pandora does in sleep mode?

I think the solution to what the Pandora should do when closing the lid problem is pretty simple. Offer the same options laptops have, or more specifically what windows offers in the power options. It asks if you want it to sleep/standby, shutdown or do nothing.

Give the user the option, instead of forcing it one way! Perhaps even extend that to what the power switch does too! ;)
 
MDave said:
Suspend to RAM ... isn't everything in RAM already? Since there is no disk cache or virtual memory on disk anyway on the Pandora? :p Since you need power for this, isn't this virtually the same as what the Pandora does in sleep mode?

No, that is completely wrong. Suspend to RAM focuses on saving the state of everything into RAM. Then, the CPU and all other parts of the system is shut down, while a steady supply powers the RAM to keep the data until you power back up. All the "low power" or "sleep" mode does is shut down a couple of processes and downclock the CPU. BIG DIFFERENCE.

-God Ginrai
 
Last edited by a moderator:
Back
Top