Jump to content


  • Content Count

  • Joined

  • Last visited

Community Reputation

105 Excellent

About dr.flo

  • Rank
  1. :):):) try this libhardware_legacy.so.zip
  2. 1. Do you have both cameras working when you use an official 2.1 rom? no -> your phone is broken -> nothing can be done yes -> continue with 2 2. Only if you can say 'yes' to 1: Do you have working front cam with .29 kernel based CM + camswitch (e.g. styx .29)? no -> no idea what's wrong (because kernel is almost like stock, cam should work there if it did in 1.) yes -> you possibly use a cam sensor that is not supported by kernel .32 ** ** if you end up here you might need to ask pier11 for further help. I've sold my pulse yesterday so there's not much I can experiment any longer...
  3. I don't know anything specific to that problem, but one could do a brave experiment and build a hybrid rom with ingredients from TomG's port and CM6.1.1 - take everything from TomG (or maybe even better, use TXG07?) - take only system/lib from CM6.1.1 I would expect this to result in a bootable rom. The idea of this is to revert as much as possible to the "last known good configuration" while keeping the new libs with camera / gps / etc fixes...
  4. what happens if you try to take a picture? any hints in logcat?
  5. Then you do not have (working) front camera. At least none that the kernel drivers can detect. Your cam sensor might be broken or it is of another type.... Otherwise msm_camera1 should be there...
  6. what are the contents of /dev/msm_camera after a fresh reboot? (it will be re-created on boot, so camswitch cannot really do any harm there ;)) I was using WhatsApp without issues on CM 6.2... Installed from market.
  7. I would guess libmmjpeg.so needs to be replaced for taking pictures. At least this was needed for u8150 & co... Video recording works without crashing....? In HQ? That would be the first working gingerbread environment....
  8. The u8500 cm7 rom has a lot of animations enabled by default, they appear "normally" fluent. Not lighting fast but it looks ok. I will check on this rom if I get slower animations when replacing the lib with our cm7 version of the lib. Of course this is a very subjective test. Another idea would be fps2d benchmark. We will know more in a few minutes, I'm currently booting that u8500 rom again... it takes quite some time on first boot... I cannot confirm libhardware_legacy.so being related to UI smoothness. The rom performs the same to me regardless of lib origin. This is true for both, perceived smoothness as well as Fps2d benchmark (29 fps on the u8500 cm7). I did not look at wifi because the rom generally is very unstable on my phone. I cannot reboot it - it will always freeze. I need to do a factory reset every time I want to start the phone a second time.
  9. Ok I will investigate this tonight. If it's true, it should be possible to patch u8500-lib to work with our wifi (in a similar fashion as it was possible for ideos x3 rom), so we could use that lib if it really gives better animations...
  10. So you are saying that with libhardware_legacy.so from u8500-cm7 the animations are good and with libhardware_legacy.so from pulse-cm7 the animations are slow?
  11. Yes, I agree on that. Blade prices on ebay are 1.5-2 times higher. I simply decided that x3 is good enough for me. After all I had been more or less happy with the pulse which makes me a low end smartphone user... But I wanted to have a reliable, somewhat current android os without issues. And without too much work on my own. After all I payed for the phone - I should not need to put development effort into it to make it functional. The discussions regarding sound lost in calls on froyo shows that for the pulse at the moment no stable rom exists above eclair - sadly - after years of work started by TomG and even after newer work from your and my side - even after support from experimenters like desalesouche who is willing to try millions of combinations of audio libs.
  12. I payed EUR 69 (x 1.3something for US $) on ebay for a new device (official price ranges are 89-99 EUR, the latter unlocked) It is said to have 256 mb ram - not too much but it runs its own stock rom absolutely fluent and is more than enough for me because I do not need to play games and such... Battery lasts 2-3 days (2g data always on, but device in standby most of the time) It should also be capable of running ICS ports, although none exist yet. But since it has msm7227 I believe sooner or later a port will exist with similar state of functionality as in zte blade. But I'm fine with gingerbread - after I hacked stock launcher to remove the ugly tiles behind application icons ;)
  13. I found a picture of ascend pcb some time ago. Ours looks quite similar. After all, ascend is the cdma version of the u8230 which is quite the same as the pulse. On this PCB I don't see any IC that can be resoldered easily as most of them are BGA type (ball grid, where the solder is under the IC) - re-soldering of these is very difficult. It also might need a firmware mod to recognize and use additional ram.... this might reside in some factory part of the rom (bootloader / baseband / nvram ???) where we do not have access to. The cost of doing this will be higher than getting a u8230 or even an ideos x3.... I recently bought the latter and will sell my pulses soon, as long as they still worth some money... btw. x3 gets around 60 fps in Fps2D - similiar score to what others have reported even for msm7200a devices - so something seems to be really broken with the pulse....
  14. Maybe this weekend I have a chance to try getting wifi on your rom, but I can't promise. Regarding the original pulse-x3 rom: Has anyone every tried if using libril.so from cm72 helps with the phone call issue? Also in call settings there is an option "Proxymity sensor" enabled by default - I don't know if the pulse has such... it might help to disable that option and see if it makes any difference....
  15. the kernel in cm72 does not support this, but you need its ramdisk for wifi to work. you could google for split_bootimg.pl, a perl script (will run in linux out of the box usually) that can split boot.img into these components. Then take kernel from a boot.img where usb thethering works (e.g. um840) and ramdisk from cm72 bootimg. then use mkbootimg to re-combine both like outlined here in the section "Alternative method", but there the link to split_bootimg.pl is broken...
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.