Page 11 of 64 FirstFirst ... 6789101112131415162161 ... LastLast
Results 201 to 220 of 1263
Like Tree3Likes
  1. #201  
    @ GenTarkin Have you tried the advice in codecrum's post #182. This could help you with the UI slowdown. I tried and it greatly improved my program switching performance.
  2. #202  
    Can someone give me a step by step install on #11 of codecrumbs ways for the device to perform better in post 182.

    I changed my I/o scheduler to anticipatory. What is the differecnce between the two?
    Thanks for all your help!
    Thanks and sorry about repost.
    *Patchers STILL make this phone Perfect!
    Palm Devices Currently Owned: Pre, Pre Plus, Pre 2, Pre 3, Pixi, Veer, Touchpads, and too many touchstones!
  3. #203  
    NEW uImage!!!!!
    http://bit.ly/919oL4
    What do u think
    *Patchers STILL make this phone Perfect!
    Palm Devices Currently Owned: Pre, Pre Plus, Pre 2, Pre 3, Pixi, Veer, Touchpads, and too many touchstones!
  4. #204  
    Quote Originally Posted by stu5797 View Post
    NEW uImage!!!!!
    http://bit.ly/919oL4
    What do u think
    In preware yet?
  5. #205  
    Also, I just tried doing the tweaks in post #182, specifically #11 and no change in behaviour =/

    Tried both anticipary and noop (been using noop since got phone)...no changes w/ either one.
  6. #206  
    Quote Originally Posted by GenTarkin View Post
    In preware yet?
    From what I have seen the new uImage is memboot only.
  7. #207  
    codecrumb,
    per your rquest from twitter... I got this a few hundred times.

    2010-10-16T19:55:44.030681Z [8062] palm-webos-device kern.warning kernel: [ 8060.810000] Syncing filesystems ... skipped for target platform ... done.

    and I sent your twitter a pastee link.
  8. #208  
    I got the same. Didn't know what to do. I notice this u21 is loading apps a little faster!!!
  9. #209  
    All posts about UI lag without DETAILED information about how to reproduce this repeatably WILL be ignored.

    Posting every couple hours asking the same question will not change this.

    This is a test kernel. See Rod's postings about what constitutes a good alpha/beta tester.

    /This concludes our public service announcement for the day.
  10. #210  
    thanks codecrumb. On u21 were we supposed to be able to raise the frequency? I can't change the freq but it seems snapier.
    thanks for all your hard work!
  11. #211  
    v22 is the kernel as shipped in preware right now + some additional kernel config changes. It should not be any different performance-wise.
  12. #212  
    You know whats interesting...I tried #11 changes in post 281 on the stock kernel and my performance in UI is lagging when switching pages was horrible, just like it was w/ the OC test kernel.

    I then wrote down stock kernel parameters for those specific settings and injected those into the OC kernel and performance increased...as far as UI lag went. But the UI seemed a bit more "framy" in its transitions in pages...but there were no slowdowns ..
    So, those posts in suggested settings in #11 seem to hurt the performance of my particular pixi rather than help it.
  13. #213  
    Quote Originally Posted by codecrumb View Post
    v22 is the kernel as shipped in preware right now + some additional kernel config changes. It should not be any different performance-wise.
    Is this v22 a iteration of the v21 warthog? or are you referring to something else? Cuz I dont see v22...

    Also, I just noticed something...why is uberkernel like 3.5MB and warthog is around 2.7MB

    Is there any way a kernel can be built with just the OC unlocks in it and not any additional tweaks, except compcache?
    Last edited by GenTarkin; 10/16/2010 at 07:14 PM.
  14. #214  
    Yeah...so even though it helped a bit for those settings from #11, to put those back to stock kernel settings.. it helped but UI eventually starts slowing down back to where it was. So, I am not sure whatever other kernel settings were changed from the stock to current warthog, but I vote only changes needed for OC and compcache to be set during compile.
    Because the stock kernel and uberkernel still both work much better than warthog one. Except for actually opening programs and they run faster.

    --
    I have not tried v21 kernel yet though, I am still comparing current preware one to stock / uberkernel. But for me thats not v22 as you stated above.

    --

    Another note, Ive noticed the test kernels even boot slower than stock / uberkernel does. The palm animation jerks slightly like its slowing down when using warthog. Stock / uberkernel the animation is very fluid and not "bulky"
  15. #215  
    I'm getting lag, I can't pinpoint anything specific, I'm guessing it's cause I may have a ****ty setup. I'm using settings that are good on uberkernel with a couple tweaks for warthog kernel

    Anyone have a good setup working, please post detailed setting if you do, thanks

    -- Sent from my Palm Pixi using Forums Beta
  16. #216  
    Tried the latest v21 image and the lag is actually a tad worse..
  17.    #217  
    Quote Originally Posted by GenTarkin View Post
    Tried the latest v21 image and the lag is actually a tad worse..
    Are you testing this on a freshly doctored device with no other patches installed and no other applications installed or running (e.g. background facebook syncs and stuff) ?

    We need results from controlled environments first when testing new overclocking kernels ...

    -- Rod
    WebOS Internals and Preware Founder and Developer
    You may wish to donate by Paypal to donations @ webos-internals.org if you find our work useful.
    All donations go back into development.
    www.webos-internals.org twitter.com/webosinternals facebook.com/webosinternals
  18. #218  
    Yeah, I dont know what it is guys...I disabled all my bootup modifications to kernel. I tried STOCK, UBERKERNEL, WARTHOG v21 via preware.

    I have to say STOCK and UBERKERNEL are identical in performance, the UI is fluid...the text messages flow fluidly when I flick them up and down really fast. Program transitions are fast and no lag.

    I try the WARTHOG kernel listed above and all that nice smooth performance goes away.
    Scrolling text messages are choppy and laggy, switching between programs is laggy.

    Another thing I notice, lets say I have preware up and downloading its feeds / parsing the feed files.
    With uberkernel / stock ... while this is going on I can easily pull up the menu, open another program with no lag, still smooth.
    This is not the case w/ the warthog. The machine literally slows down to a crawl when its parsing the preware .gz files its downloading when trying to go to another program.

    Now, Ive noticed that the #11 changes instructed on post 182. Well, the stock and uberkernel kernels have IDENTICAL values for these systcl kernel settings. The warthog does not. Which leads to me believe potentially a lot of other kernel parameters have been changed in the experimental kernels...OR MAYBE this is the difference between sprint / ATT kernel??? - not sure..

    But either way I propose we need a kernel w/ just the OC modifications and compcache additions and NO OTHER parameters changed from stock. So we can have a close as possible apples to apples comparison. Because when you throw other tweaks into the mix...we dont really know whats going on.
  19. #219  
    Quote Originally Posted by rwhitby View Post
    Are you testing this on a freshly doctored device with no other patches installed and no other applications installed or running (e.g. background facebook syncs and stuff) ?

    We need results from controlled environments first when testing new overclocking kernels ...

    -- Rod
    NO and this should not matter. The fact that the stock and uberkernels works FINE on my machine and warthogs does not.
    Also fact #2 that there are fundamental differences in kernel parameters in warthogs vs stock / uberkernels...

    TELLS ME, something is wrong with the kernel itself. Not our devices.
    I repeat the kernel needs to be as close as possible to stock condition + OC + compcache ONLY...for this to be a "controlled environmental" test.
  20. #220  

Posting Permissions