Active Topics

 


Reply
Thread Tools
Posts: 4 | Thanked: 0 times | Joined on Jul 2011
#241
even after the update my camera interface is showing only high low and fine resolutions options, no option of hd or dvd recording tab.
i understand i have to install codecs for hd recording but isnt the camera interface supposed to show the options?
my update was successful
am i missin something?
 
Posts: 85 | Thanked: 36 times | Joined on Jun 2010 @ Berlin/Germany
#242
I have severe battery drain problems since updating...
Batterygraph shows a 45 degree falling curve in standby right after booting!
I had this previously but could magically solve this by starting conky for some seconds then quit it again. After this the batterygraph curve was almost horizontally in standby.

I have kp49 installed.

Anyone else having the same problem?
 
Posts: 336 | Thanked: 129 times | Joined on Jan 2011 @ portugal
#243
no..just the problem i mentioned before...no matter how long i leave n900 charging it never goes more than 85% of charge, in both batteries i tried it on...something is not right for sure in this matter...i too have kp49, dsp profile, and recently purged batterypatch and speedpatch...
 
Posts: 1,203 | Thanked: 3,027 times | Joined on Dec 2010
#244
I did, but as stated earlier I reflashed to try single out the cause and its been okay since.

10hrs so far. Latest CSSU with 3G (Dual Mode) connected, light-medium browsing over 3G and Wifi, MfE running at 30mins intervals. Its only dropped one bar so far. Haven't got batterygraph installed because I'm trying to reinstall gradually to find the culprit, so I can't give any more details at the moment.

Please note I haven't got KP installed yet. I'll give it a few days at least before testing that.
 
Posts: 569 | Thanked: 462 times | Joined on Jul 2010 @ USA
#245
Originally Posted by ed_boner View Post
and recently purged batterypatch and speedpatch...
Purging batterypatch and speedpatch is not enough - a re-flash is needed to get rid of them, according to this post:
http://talk.maemo.org/showpost.php?p...&postcount=834
Originally Posted by woody14619
For an answer as to why not to use batterypatch or speedpatch, just do a power-search from the column on the right for either. You will find one announce thread, and about 500 "problem" threads where people installed one or both of these and it screwed things up.

These "patches" make permanent changes to your device, with no way to undo them except a re-flash. An uninstall does not undo the changes! These changes can lead to anything from reboot-loops to the device just being flakey, including problems with camera use, calls, TV-out, USB use, and just about everything else.
 

The Following 3 Users Say Thank You to rotoflex For This Useful Post:
Posts: 560 | Thanked: 422 times | Joined on Mar 2011
#246
Firstly, thank you to the CSSU team for persisting with your efforts. The N900 is ever more useable with each release.

It's impressive to see, apart from other things, the near-full capability of the device in portrait mode, as well as landscape mode! Even if to type more than a few characters I find I need a hardware KB there are two clear bonuses of these improvements: firstly when the device is needed to behave like an old N-series (i.e. a phone +) it now does so with minimal hassle. More importantly though, switching between portrait and landscape modes has become increasingly reliable over the last couple of updates. It's not absolutely error-free - but we're talking about software that's in 'testing'! Big up to the developers and error-checkers!

OS version: 21.2011.38-1Tmaemo1.2 (available 17th Jan 2012)
Language: en-gb

I apologise if any of the points below are discussed elsewhere, I couldn't find them readily. Please direct me if this is the case so I can read answers/discuss in the appropriate place. Should any of the following warrant a proper bug report, please advise and I'll post it. I want to make sure it's not just me!

A few of questions and suggestions...

General
- Please would someone place a direct link to the CSSU Wiki page, on the Wiki main page?

- There is an error on the CSSU Wiki page: the TMO link in the Contribute goes to the old thread.

On both these matters, although I'd be happy to edit these pages, I don't feel it's quite my place to do so.

- If there is an announement for a release. It would be very helpful if one of the first few items in that post/email/article were a links to the bug tracker (to see how much work's been done!) and the changelog. I believe this might reduce confusion amongst over eager users?

CSSU team
It would be helpful to know who's contributed to/responsible for which areas. Why? Well, maemo.org is a community of developers, testers and interested users. Therefore, being reminded that behind each line of a code is a person who has been buring the midnight oil, definitely assists in community cohesion. It might even result in the transfer of a few coppers! The sort of thing I'm talking about shouldn't breach any online anonimity...
"Hi I'm " + <TMO_username> + ". My role in the CSSU team is " + <areaOfInterest> + <activeFrom> + <activeTo> + <otherVolunteeredInfo> .
As well as developers, formal testers and documenters could be named and get some credit too? I also believe the quality of reporting and documentation might improve if it's loosely formalised by some community recognition, thereby helping both developers and users (because reporting means developers know what to fix and docs mean users know how to use software / what to expect).

On a more practical note, knowing who does what would allow for more appropriate direction of questions regarding how-tos, bugs and extensions.

It could be on a wiki page or a link from the wiki to a "Meet the cssu team" thread?

... just an idea.

Re: Calculator
The good...
The ability to use this in portrait mode makes life a whole lot easier - a proper touch screen calculator!
Thanks for redesigning the divide key to resemble a normal calculator. Much prefer it. Are any other keys due to be redesigned, multiply and power spring to mind.

The less good...
My experience is that the calculator is currently a (slightly unintuitive) mix of two of the frequently found input styles for calculators (procedural entry and instant). The values and most operators are declared in line with what one might expect from a calculator that requires procedural input. However, the output doesn't show procedurally; furthermore, some operators function in an instant manner (i.e. operate on just the content of the input box, rather than operating once the whole procedure (calculation) has been declared).

Are these bugs or intentional behaviour?
- the values persist in the input box until another numerical value is introduced. This is confusing. Can value-persist be toggled?
- some operators function in an instant manner, not procedural. e.g. The root operator.
- there is no 'help' or 'about' button. I thought these were Maemo design obligations?

Both the input-persist and illogical operation order are exposed in the following example: (keys in square brackets).

Aim: calculate √((23 + 4) ÷ 3)
Attempt 1: enter data procedurally
Code:
>> [√]			// press root symbol
<< Error		// message on till roll
Attempt 2: another approach
// calculate: result_1 = (23 + 4) ÷ 3
// calculate: result = √result_1
Code:
>> [(]			// open bracket
<< (			// current content of tillroll
>> [2]			// input number
>> [3]			// 23
>> [+]			// number still in input box
<< (23 +		// still in input box 
>> [4]			// now 23 is overwritten by 4
>> [)]			// '4' stays in input box
<< (23 + 4)		// tillroll output for calculation
>> [÷]			// Thanks for redesigning this key
<< (23 + 4) ÷		// '4' still in input box
>> [3]			// input number
>> [=]			// GO!
<< (23 + 4) ÷ 3		// now '3' is shown
<< = 9			// [M+] not needed. Result written to input box
>> [√]			// try this...
<< √9			// tillroll shows √ inserted before value
<< = 3			// executes without pressing [=]
Expected I/O:
Code:
>> [√]			// permit root operator at start
<< √			// display operation as entered
>> [(]				
<< √(				
>> [(]				
<< √((       		
>> [2]				
<< √((2		// display numbers as entered
>> [3]			// 23
<< √((23		
>> [+]			// clear input box on pressing operator
<< √((23 + 			 
>> [4]     		    
<< √((23 + 4		
>> [)]			// again, clear input box
<< √((23 + 4)      	
>> [÷]			// operator
<< √((23 + 4) ÷		// empty input box
>> [3]			// input number
<< √((23 + 4) ÷ 3	// display
>> [)]			// operator
<< √((23 + 4) ÷ 3)	// clear input area
>> [=]			// operator ==> parse & calculate
<< = 3			// result displayed.  only one [=] execution
Suggestion/request:
- Some buttons could quite happlily have 2nd functions, without comprimpising usability. If anyone wants a full on graphing calculator, there are a couple available but even in 'scientific mode', trigonometric or logarithmic functions are absent.
- Can [enter] be made to work like pressing [=]? [enter] is an intuitive input to finalise an operation.

Browser (microB)
The protrait mode is really useable, including those UI things like swiping in from the left and right. Thank you! From my use so far, the browser is the application that rotates most appropriately - that is, it's neither over-, nor under-sensitive to movement, so it pretty much rotates when I desire, as opposed when it desires!

Again, if this has been mentioned elsewhere, I apologise - please direct me to the answer and I'll be quiet.
- Is there a means of taking a screenshot in portrait mode? Obviously, opening the KB to press Ctrl+Shift+P is not an option because it triggers the orientation to switch.
- input problems:
- - Some textboxes do not accept input when using microB. Not sure if this is CSSU related or slightly old underbelly or something else on my device.
- - Extra characters (Fn + Ctrl ...) do not always get displayed on screen. Again, usure of cause. Anyone else notice this? I was checking something on the w3schools 'tryit' page.
- I thought a portrait VKB popped up when an input field has focus? For me it's switching to the landscape VKB - am I doing something wrong, or just imagining the existence of a portrait VKB?

PDF reader
I think this has had some attention from the CSSU programme but not sure if it's still being updated.
- Is it me or is this a bug? There's something funny going on with the text and layout of the bottom-right button in the drop-down menu; the text is all one word and the checkbox masks some of it.

Telephone
I haven't had an incident yet where I couldn't answer with this release of the CSSU. This application has a bit of a mind of it's own when it comes to rotation. It is much easier to make and take calls in portrait mode because it feels more like using a phone. The problem is that some times the answer/reject/slient/text-reject buttons (red and green) can be partially obscured when the application rotates - they are cut off by the edge of the screen. When this happened with previous CSSU versions, the phone was inert (no interaction possible) until it stopped ringing. From the outside, it's as though the ringing takes so much juice the device can't orientate and organise the UI. With the newest CSSU, those buttons that are still visible are push-able but some are still trimmed from view.

Voice mail
This may have nothing to do with the CSSU but I've only tried this since installing the CSSU. On introducing the dialing code to toggle the answerphone - command to disable voicemail or change trigger-time seems to have been accepted. However, in the device settings the voicemail it's permanently set to trigger at 20s and the voicemail won't turn off.

Hildon Desktop / Device UI
One of the best features of the CSSU, right from the beginning has been the ability to toggle UI editing state. The default state is editable so it's really easy to accidently move or delete things on one of the desktop screens. The CSSU made it possible to "lock-down" the UI, which was a massive improvement.
There are some things I'd like to know...
- When the state is un-editable, is it possible to turn off the little gear-wheel that pops up when a blank bit of the desktop is clicked?
- Is it possible to lock the desktop scrolling, so it stays on one screen?
- As an aside, I think it would be worthwhile if the CSSU posted an alert when first installed on a device to remind the user to rearrange desktop buttons for each desktop in portrait mode. I thought I'd lost some shortcuts but found that some buttons had become stacked on top of each other.

Camera & Camera UI
More improvements. Good job! I'm looking forward to using the night camera mode. If anyone can direct me to information about frame compositing that works on the N900, I'd be very grateful. There are a couple of things that may have been pointed out already but I can't see them at the moment...
- There are two "General Settings". Perhaps "Application Settings" should be from the top as a drop-down settings menu, while "General Settings" is the name for the menu that appears from the spanner?
- There is a zero in the top-left of the screen, next to the big A. What is it?
- There used to be a distance meter on the last release, but I can't find it. Was it disbanded?
- Is there a means to hide all icons?
- Please confirm: the HD record can only be used once Kernel-Power and freemangordon's scripts have been installed?

TV-out
The new UI for this is so much more complete than the original. The available functions, all in one place. Nice one.
- Can the content being sent to the AV cable be specified? Or is it only what's being drawn on the screen? Is it the same as the device's frame buffer? It would be nice to guarantee output RGB888, not RGB565 colour because it's so much more noticable on a TV.
- Might it be possible in later releases to scale the AV output to (s > 100%) in case the user wishes to view a cropped output? Central pixels only to start with.

Apologies for length of post. A summation of thoughts!
... if you've read all of that Thank-you some more!

Last edited by demolition; 2012-01-20 at 21:26.
 

The Following 20 Users Say Thank You to demolition For This Useful Post:
Posts: 560 | Thanked: 422 times | Joined on Mar 2011
#247
Originally Posted by Android_808 View Post
I did, but as stated earlier I reflashed to try single out the cause and its been okay since.

10hrs so far. Latest CSSU with 3G (Dual Mode) connected, light-medium browsing over 3G and Wifi, MfE running at 30mins intervals. Its only dropped one bar so far. Haven't got batterygraph installed because I'm trying to reinstall gradually to find the culprit, so I can't give any more details at the moment.

Please note I haven't got KP installed yet. I'll give it a few days at least before testing that.
Apart from the standard, beware of -devel and to some extent -testing software, it's worth looking at something like HTOP or Conky to see what's guzzling the resources. If batteries dry up fast - either you need a new one or a number of programmes are chewing cycles and or leaking memory, thus wasting electricity.

There are cases in which only a reflash will sort out the corruption of poor devel software. Or at least, it's significantly more practical to do this than rummage around trying to extract the problem and all its tenticles.

Edit: On re-reading this I see that neither the quoted comment, nor my reply have any mention or reference to the CSSU!

Last edited by demolition; 2012-01-20 at 22:00.
 
Posts: 336 | Thanked: 129 times | Joined on Jan 2011 @ portugal
#248
Originally Posted by rotoflex View Post
Purging batterypatch and speedpatch is not enough - a re-flash is needed to get rid of them, according to this post:
http://talk.maemo.org/showpost.php?p...&postcount=834
Sh**...thanks..still i was wondering..i am very happy with my battery in dsp profile, can it be that this patches only screwed my battery measurement system (sorry dont know the technical term) or the battery doesnt really charge totaly?
 
woody14619's Avatar
Posts: 1,455 | Thanked: 3,309 times | Joined on Dec 2009 @ Rochester, NY
#249
Originally Posted by ed_boner View Post
Sh**...thanks..still i was wondering..i am very happy with my battery in dsp profile, can it be that this patches only screwed my battery measurement system (sorry dont know the technical term) or the battery doesnt really charge totaly?
Could be either/both. Just FYI: Most LIon battery chargers don't even try to charge above 95%, since there's a rapid loss on the return above that level. You never want to fully saturate or fully discharge a LIon anyway, as both tend to make them unstable and/or destroy the battery.

As for the 85% mark, it could be that the guestimator is off because of some recent past event. That number is a guess from the internal system, which sometimes gets out of sync with reality. If so, it should just drift back toward reality over time. Or, your battery is degraded, which happens. Worst case, buy a new one. Third party batteries with higher capacity than the stock Nokia can be had for quite cheap. There's a thread on it here in the forums for where to get the best ones, well summarized in the first 2 or 3 posts by a very active community member.
 

The Following 2 Users Say Thank You to woody14619 For This Useful Post:
Posts: 336 | Thanked: 129 times | Joined on Jan 2011 @ portugal
#250
not the battery, i tried on 2 different ones...so i guess i am gonna wait a while to see if it drifts back as you emntioned...tks woody!
 
Reply

Tags
cssu testing


 
Forum Jump


All times are GMT. The time now is 14:01.