Sleep Mode, Diagnostic/maint. Menu, Boot Off Usb?

[DaveC]

The battery is removable if you really wanted to change them out. The battery compartment will have a cover held down by a screw (like the DS). One thing to note is that you will have to re-set the clock each time you do this.

Bad news. Batteries should be easy changeable without screw or resetting the clock.

Yeah if they are AA's. Wow, I think the expression "you can't please everyone" can be changed to "you can't please anyone!".

The PSP doesn't use AAs and you can swap the PSP battery out easily without removing screws.

I guess with the Pandora if the battery dies in the field you just can't use it until you get to a charger, just put it away. Kind of limits it's useability. Maybe something more user-friendly can be done.

 

[Squidge]

Maybe we could do like a PDA - have a coin cell as a backup battery which keeps the unit in sleep mode for a few minutes when the battery is removed?

You can do that with a capacitor (which is what I asked about higher up in the thread), and no need to worry about the state of the backup battery.


It would have to be a big capacitor, or you would have to change the battery in nanoseconds unless the only thing you providing power for is the RTC because I was thinking more along the lines of getting a battery warning, then closing the shell to put the thing in sleep mode, changing the battery and then carrying on where you left of. I'd very much doubt a capacitor would be able to store enough charge for that in the space we have.

I guess with the Pandora if the battery dies in the field you just can't use it until you get to a charger, just put it away. Kind of limits it's useability. Maybe something more user-friendly can be done.

You could always do one of the psp emergency solutions, such as those that contain a battery of there own plus a dc jack to plug into the machine to be powered. A little bulky, yes, but it lets you continue playing/whatever without even shutting the lid

 

[b._.o._.b]

Can you remove the battery and keep the device powered when it is plugged in? This will at least make it possible to change batteries without resetting the rtc.
If that is possible and you can swap the battery without unscrewing that would be very nice.

Best would be the solution of Squidge, but I can imagine that will have some more impact on the current design.

Also interested what will happen when the battery is almost empty. How long will it keep the rtc fed and is it still possible to wake the system up a couple of times for an alarm. This would be very practical when using it as a PDA. People using the PDA functionality will give that a higher priority than 10 or 20 minutes of extra gaming for sure.

[Edit] damn typos.. changed them before I need to leave as well

 

[cappuchok]

Can you remove the battery and keep the device powered when it is plugged in? This will at least make it possible to change batteries without resetting the rtc.

I'm thinking it would be possible to run the RTC off a lithium cell (or even a capacitor) when the device is not powered from another source (battery and/or wall outlet), similar to how it's done on most other computers. That would let you change the battery without having the device connected to the mains, and safeguard the RTC against a possible power outage whether it be an empty battery or a mains power failure...

 

[Karel Jansens]

Maybe we could do like a PDA - have a coin cell as a backup battery which keeps the unit in sleep mode for a few minutes when the battery is removed?

You can do that with a capacitor (which is what I asked about higher up in the thread), and no need to worry about the state of the backup battery.


It would have to be a big capacitor, or you would have to change the battery in nanoseconds unless the only thing you providing power for is the RTC because I was thinking more along the lines of getting a battery warning, then closing the shell to put the thing in sleep mode, changing the battery and then carrying on where you left of. I'd very much doubt a capacitor would be able to store enough charge for that in the space we have.


Really? There was one in the Newton MessagePad that kept the RTC and RAM powered for 40-odd minutes, and I don't remember it being a huge monster (then again, the Newton's architecture is vastly different from what we have in these latter days).

Obviously I was not advocating that the machine would be kept running of the capacitor while changing the battery, just that it would be possible to change batteries in the stand-by state. Five to ten minutes ought to be enough to switch any battery, even when screwdrivers are needed.

 

[pupnik]

Re: "Can't please anyone". This is just the time and place for bitching and whining. When pandora is released people will realize it's the best thing in the world, and purchase it.

 

[Squidge]

Really? There was one in the Newton MessagePad that kept the RTC and RAM powered for 40-odd minutes, and I don't remember it being a huge monster (then again, the Newton's architecture is vastly different from what we have in these latter days).

Yes, there is a bit of difference. I don't know which model your talking about, but the first version only had 640KB of RAM, whilst the most advanced one (the 2100 I believe) had 8MB. Also, apple states that the backup battery is a round disposable 3-volt Duracell DL2032 lithium battery, not a capacitor (again, they don't state the actual model number).

 

[Karel Jansens]

Yes, there is a bit of difference. I don't know which model your talking about, but the first version only had 640KB of RAM, whilst the most advanced one (the 2100 I believe) had 8MB. Also, apple states that the backup battery is a round disposable 3-volt Duracell DL2032 lithium battery, not a capacitor (again, they don't state the actual model number).

The backup battery was in the earlier models, the 2X00 has a capacitor. The 2100 is the model I have.

 

[atomicthumbs]

It would have to be a big capacitor, or you would have to change the battery in nanoseconds unless the only thing you providing power for is the RTC because I was thinking more along the lines of getting a battery warning, then closing the shell to put the thing in sleep mode, changing the battery and then carrying on where you left of. I'd very much doubt a capacitor would be able to store enough charge for that in the space we have.

For 49 cents each in batches of 1000:

22000 microfarads. 5.5v DC. 0.45"Dia. x 0.30"Tall.

 

[Squidge]

Yes, there is a bit of difference. I don't know which model your talking about, but the first version only had 640KB of RAM, whilst the most advanced one (the 2100 I believe) had 8MB. Also, apple states that the backup battery is a round disposable 3-volt Duracell DL2032 lithium battery, not a capacitor (again, they don't state the actual model number).

The backup battery was in the earlier models, the 2X00 has a capacitor. The 2100 is the model I have.


Do you have any pictures by any chance? I'm curious to see what kind of capacitor it is, as all the internal pictures I remember seeing looked just like coincells, but with metal strips "stapled" to them in a fashion to make them non user replaceable.

 

[b._.o._.b]

Where is MWeston when you need him
It can be frustrating for these guys I guess, they are doing a great job and sometimes it looks like it is never good enough.

on topic: in laptops or desktops a coin cell can work many years without a problem. A capacitor doesn't prevent problems after longer power loss, but if it can be used it is certainly better than nothing.

 

[Squidge]

For 49 cents each in batches of 1000:

22000 microfarads. 5.5v DC. 0.45"Dia. x 0.30"Tall.



Instead of just posting pictures, how about you do the maths as well as I really can't be bothered - tell us how it will last at a typical discharge (say, 10mA @ 3.7V).

 

[Karel Jansens]

Yes, there is a bit of difference. I don't know which model your talking about, but the first version only had 640KB of RAM, whilst the most advanced one (the 2100 I believe) had 8MB. Also, apple states that the backup battery is a round disposable 3-volt Duracell DL2032 lithium battery, not a capacitor (again, they don't state the actual model number).

The backup battery was in the earlier models, the 2X00 has a capacitor. The 2100 is the model I have.


Do you have any pictures by any chance? I'm curious to see what kind of capacitor it is, as all the internal pictures I remember seeing looked just like coincells, but with metal strips "stapled" to them in a fashion to make them non user replaceable.


I'm not going to open my Newtons again, it was stressing enough the last time, but here are pics of disassembled MessagePads (2X00). Smarter minds than mine might probably distinguish a capacitor and/or backup battery.

If you go to the Newton wiki and do a full text search, you'll find several posts referring to (problems with) the capacitor, also referred to as "supercap". Bear in mind that indeed the Newton 2X00 was the only model ever fitted with the thingie. Even the eMate didn't have one (causing people cursing at all too frequent time and date resets following a dead battery).

 

[Squidge]

Thanks for the info, I've done some research armed with this and noticed the following:

NewtonTalk.net March 2002 archive: 'The MP2x00 has "only" a supercap, capable of buffering the RTC for max. 4 hours.'
NewtonTalk.net October 2002 archive: 'Even if the battery is low the Supercap in the Newt (not a backup battery) should hold the voltage up to 4-5 hours. '

and some other similar posts. It seems that they used to use a backup battery to save the contents of RAM and the RTC, as all the user data plus system updates/etc was stored in RAM. In the 2100, they replaced the battery with a cap which only backed up the RTC, as the user data/etc was no longer stored in volatile memory.

 

[Blah]

Well the memory could be backed with the regular battery, and a capacitor used so you can swap batteries.

 

[Squidge]

Well the memory could be backed with the regular battery, and a capacitor used so you can swap batteries.

Someone obviously isn't reading the thread, or doesn't understand what we are talking about...

Maybe you should re-read?

 

[Peter R]

I haven't tried out the math in detail but from my general area of thinking, I don't think you would be able to get a capacitor with the required capacitance to provide anywhere near the required current for anywhere near long enough in any reasonable size.

 

[Perky_Goth]

I am in awe. It's perfect. The only other thing that might be interesting is having a way for the sleep process to tell the apps it's going to sleep so it can choose what to save and what to discard. Say, an emulator would save it's state and shutdown, freeing memory that doesn't have to go to SD.

OTOH, that might be too complex...

 

[Karel Jansens]

Thanks for the info, I've done some research armed with this and noticed the following:

NewtonTalk.net March 2002 archive: 'The MP2x00 has "only" a supercap, capable of buffering the RTC for max. 4 hours.'
NewtonTalk.net October 2002 archive: 'Even if the battery is low the Supercap in the Newt (not a backup battery) should hold the voltage up to 4-5 hours. '

and some other similar posts. It seems that they used to use a backup battery to save the contents of RAM and the RTC, as all the user data plus system updates/etc was stored in RAM. In the 2100, they replaced the battery with a cap which only backed up the RTC, as the user data/etc was no longer stored in volatile memory.

Well, sorta. The Newton uses a funky combination of SRAM and flash which, unless you're in for a MAJOR redesign, isn't really applicable to the Pandora.

Still, my Fujitsu Stylistic 3400 tablet uses a similar supercap (maybe a slightly bigger one) to allow for battery switching in standby mode. The Fujitsu's supercap keeps both RTC and RAM juiced up for at least 10 minutes, but then it's battery as basically a click-and-snap affair. Still, it can be done with an X86 computer, so...

Oh, and: no, I'm not disassembling my Fujitsu.

I haven't tried out the math in detail but from my general area of thinking, I don't think you would be able to get a capacitor with the required capacitance to provide anywhere near the required current for anywhere near long enough in any reasonable size.

Not if you want to keep using the computer while switching batteries, I agree. The point is that the capacitor provides juice for a computer in standby mode, just enough to keep the RAM's capacitors from draining. It can and has been done, see my other post.

 

[chris_r]

Have we heard yet what kind of drain to expect from the panda yet? Say at the worst case full frequency, full screen brightness, full volume, full wifi what would be the absolute maximum drain.