Release Hotfix 7 Beta 1 released!


Big Problem:


I can't reboot any Hotfix.


I'm using HF7a3.


This is what I get:



Code:
reading boot.scr

5045119 bytes

## Executing script at 82000000

Wrong image format for 'source' command

script failed.


I've tested 2 different Sd-cards (Fat32).


I wanted to get Beta1. As this didn't work I tried out to reflash alpha3 but the same problem occures.


What's wrong. I'm already using Alpha3. Why can't I reflash?
Any progress? I've had occasional troubles downloading from the German servers. If you're willing to try downloading it again, I've re-uploaded it here temporarily: https://www.box.com/...h0h3lytjdjaa0m1


Try installing from a freshly formatted SD card too, if you aren't already.
 
I now tried an third, reformatted Card. That works now but doesn't solve the problem with the other two cards.


The other cards don't have any other problems.


Thanks.
 
Last edited by a moderator:
I downloaded it and gave it a shot. Works wonderfully for me. I really like the touch screen now, it works so smoothly.


Thank you everyone who had a hand in this, im simply amazed at how the product i purchased just keeps getting better and better.
 
Yeah, it seems to happen. Good to hear you got it updated eventually.


Maybe try this formatter on the other cards: http://panasonic.jp/...load/index.html

I'll try it on one card (SDHC 32 GB). There's not much space used.


The other one is SDXC 64 GB. 32GB are used (more than 100000 Data because I downloaded some modrana maps). This will take a while.


I'll report the result of the first card as soon as I've tried it.
 
This problem was encountered whilst using the last stable Hotfix 6, flashing to Hotfix 7 Beta 1 did not resolve the issue, so I flashed back to 6 and the problem was resolved.


Pandora boots up OK into MiniMenu, then When trying to switch from MiniMenu to Xfce I get this message.....


Could not look up internet address for xxxxx xxx.


This wll be prevent Xfce from operating correctly.


It may possible to correct the problem by adding xxxxx xxx to the file/etc/hosts on your system


Been in there and xxxxx xxx exists in hosts.


I can continue through into Xfce at that point.


------------------------------------------------------


Then once in Xfce I cannot switch GUI from Xfce to MiniMenu, message recieved is.....


Failed to receive a reply from the session manager


Session manager must be in idle state when requsting a shutdown


Also cannot logout from Xfce same message as above.


So have to force shut down, then reboot to get to MininMenu.
 
I dont know why but for some reason I vaguely recall something similar if you gave your pandora a name with spaces in it?


Probably barking up the wrong tree here?
 
as far as that charging bug thats fixed in hf7a3/b1, is that the bug that wont let some pandoras charge to 100%, or the one that wont let some pandoras discharge below 10%?
 
as far as that charging bug thats fixed in hf7a3/b1, is that the bug that wont let some pandoras charge to 100%, or the one that wont let some pandoras discharge below 10%?
It is supposed to be a "two in one" issue...
 
isnt working for me, my pandora still forces shut down at an average of 17%, never a single sign of low or critical battery. if i quickly turn pandora back on, it does read it at 1% or something though, and i was always able to charge to 100%
 
isnt working for me, my pandora still forces shut down at an average of 17%, never a single sign of low or critical battery. if i quickly turn pandora back on, it does read it at 1% or something though, and i was always able to charge to 100%
This is a physical problem, I'm almost absolutely certain. The resistance being reported to the bq chip (which measures the capacity and discharge rate of the battery to arrive at a really good estimate of the remaining battery) is incorrect. When I asked if corrosion on the contacts could cause this, the TI tech I was talking to said yes. The result is that it measures a smaller rate of discharge than is actually happening, say around 15-20% which means after 4000mah have been used, it has only measured 3400 or so. It thinks there's still 600mah left when in fact the battery is completely drained.


I've got two Pandora's, one very old (one nub busted) and one fairly new. The old one has no problem handling either battery. The new one dies at about 20% for both batteries. All four cases (each battery in the old and the new) lasts for approximately the same length of time, which means it isn't actually a problem of the battery being bad or one Pandora using more power than the other, it's a reporting issue. This happens on HF4, HF5, and HF6, consistently. I haven't tried HF7 yet. My testing took a very, very, very long time.


There are two ways to fix this: recalculate the resistor value by estimating what the current should be and working backwards, a value that will differ for every Pandora that is affected (or might not be affected at all); alternatively, turn the auto adjust function which will recalculate the capacity of the battery based on the consumption. The second one is very simple: a quick command, discharge the battery, and let it sit for several hours to "normalize" or whatever. In the second case, it continues to mismeasure the discharge rate, but now it's got an equally mismeasured total capacity, so the percentage reported is at least accurate. That's what I did. Now when my battery is dead, it actually reads 0% and I still get my full usage out of it.


Anywho, the real reason I came to this thread was to ask if anyone is able to read from /dev/dsp. I used to be able to "cat /dev/dsp > /dev/dsp" to echo the microphone back to the speaker, but now attempting to cat /dev/dsp gives me an Input/Output error. arecord -L gives me only the null device. Can't seem to find the microphone at all. Is this just me, or can no one access the microphone anymore?
 
Anywho, the real reason I came to this thread was to ask if anyone is able to read from /dev/dsp. I used to be able to "cat /dev/dsp > /dev/dsp" to echo the microphone back to the speaker, but now attempting to cat /dev/dsp gives me an Input/Output error. arecord -L gives me only the null device. Can't seem to find the microphone at all. Is this just me, or can no one access the microphone anymore?
try arecord/aplay -l


And proper way to do loopback is "arecord -f cd -D hw:0,1 | aplay".
 
No errors at least, but still no sound. Maybe I broke another microphone, somehow. Could'a sworn at least on HF4 that I was reading directly from /dev/dsp, but my memory has been faulty before.
 
After Installation of the HotFix (7b1) the Pandora booted up fine, I was greeted by the FirstRun Wizard, set everything up, only to discover it did not recognize my SD card anymore (from which it had just flashed).


It shows up as "_PNG", PNDs are recognized or rather desktop files are created, but when I click on the icons on the XFCE desktop all it does it open the actual .desktop files in the text editor.


I also cannot open the SD card through the file manager (it gave me an error before about some unrecognised media type, but it does not do that now after a restart and after I deleted the Hotfix image from it).


Currently backing up my data through Windows (which still opens the card fine) to format the card...


Apart from that no problems with the new HF so far.


EDIT: Formatting did the trick, card works well again, still a mysterious bug.
 
Last edited by a moderator:
There are two ways to fix this: recalculate the resistor value by estimating what the current should be and working backwards, a value that will differ for every Pandora that is affected (or might not be affected at all); alternatively, turn the auto adjust function which will recalculate the capacity of the battery based on the consumption. The second one is very simple: a quick command, discharge the battery, and let it sit for several hours to "normalize" or whatever. In the second case, it continues to mismeasure the discharge rate, but now it's got an equally mismeasured total capacity, so the percentage reported is at least accurate. That's what I did. Now when my battery is dead, it actually reads 0% and I still get my full usage out of it.
And what is this "quick command"? Is it listed somewhere and do you have the respective link?
 
Last edited by a moderator:
And what is this "quick command"? Is it listed somewhere and do you have the respective link?
No, it requires installing i2c tools which could be dangerous in the wrong hands. It's a very simple "execute this one line" and it'll work as long as you copy it exactly, but get one character wrong and it could require the bq chip to be reflashed. I also just discovered this morning that if you don't turn off the feature that the command turns on once everything is recallibrated, it will keep recallibrating, sometimes in very weird ways. :p


So I need to figure out how to turn off the auto-recallibration first before I tell anyone how to turn it on, and then I need to come up with an absolutely fool proof method so people don't accidentally break something.
 
Back
Top