Page 17 of 56 FirstFirst ... 7121314151617181920212227 ... LastLast
Results 321 to 340 of 1118
  1.    #321  
    Quote Originally Posted by st0921 View Post
    Try the "performance" setting with maximum at 800. Not sure why there is a "minimum" frequency assigned in this setting, but mine seems to run at 800 consistently with this setting. I've been running the battery drain testing for about an hour with this setting at idle, interestingly the drain per hour is at 2.11% now, much better than the screenstate of 3-4%...will keep posted.
    Google 'Race to Idle' and you'll find one of the theories on why this might be so.

    That is why we provide that option for those people who have a usage pattern which better matches with that governor.

    -- 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
  2. #322  
    Quote Originally Posted by rwhitby View Post
    1) Unixpsycho has done the overclocking patch for every single overclocking kernel ever released for the Palm Pre. So when you are making comparisons between kernels (which is fine to do, if it's done in an objective and non-emotional manner) remember that you're actually comparing the same basic code, just with different default settings which you haven't changed to suit your circumstances yet.

    -- Rod

    What if we are using the same settings and seeing a very different over all difference between kernels? With Uber's config set the same way as SPK is/was, and Uber is very noticeably the laggier of the two, do we continue to post here even if we have to use the acronym "SPK" to do so?

    Not trying to be a wise ***, I'm asking an honest question since (as far as my posts and several others I've read today), SPK gets mentioned as something to compare Uber to.

    M.
  3. #323  
    If you use any kind of scaling, even the screenstate governor, it will appear slower than the kernel which ran at 800mhz as there is always a delay while the processor throttles up. As things stand currently, I don't think anything other than screenstate scaling is worthwhile. All the other governors degrade both performance and battery life for me.

    I have used the Performance governor with minimum and maximum both set to 800 and noticed zero performance difference between that and the old 800mhz kernel.

    I currently use screenstate 500/800 just for the peace of mind of knowing the device will not overheat when I am not there to monitor it.
  4. #324  
    I am. I'm using 800/500/screenstate and am seeing a lot more lag then I ever did with The Kernel That Has Been Sacked.

    I think I'm done with talking about this though. I either choose to use the Pre the way Palm wants me too (slow and laggy), or choose to use it the way WebOS Internals / PreCentral offers me to (fast, and kinda laggy). I guess I'll go with the speed.


    Thank you for some choice back, on my hand-held Linux device, WOSI!


    M.
  5. #325  
    qustion, if i use this uber 3rd gen. , and use cpuscalerultimate with it, will my pre goto 500 while on touchstone ? thanks
  6. #326  
    Quote Originally Posted by rwhitby View Post
    1) Unixpsycho has done the overclocking patch for every single overclocking kernel ever released for the Palm Pre. So when you are making comparisons between kernels (which is fine to do, if it's done in an objective and non-emotional manner) remember that you're actually comparing the same basic code, just with different default settings which you haven't changed to suit your circumstances yet.

    2) If you want to replicate the configurarion of previous overclocking solutions, then use the 'performance' governor, not 'screenstate' and do your own long-term objective comparison

    -- Rod
    Rod, a technical question then.

    The earlier overclocking kernel solutions operated at specific frequencies and the kernel installed contained that information in the package you chose to install. It would only change if you installed a scaler application of some sort or used the linux commands.

    Uber operates at a default 500 unless the commands are entered or an app is used.

    So what in the kernel package makes one run at a given frequency?

    It seems to me if all the other things are equal then whatever makes one kernel default to 800 all the time would be different. Could it be that this difference produces different dynamics when installed?

    Also since I don't understand how the scaling process works, it would seem like double work for the scaling app to undo what a given kernel set as default and reconfigure to something else.

    Could it also be that since the pre booted up with the static overclocked kernels it somehow had different performance characteristics than one that is overclocked after the boot is complete?

    I mean we know that the order in which certain patches are installed make a difference in whether or not those patches function properly (named or tabbed launcher pages for example must be installed before any other launcher patches or all of your icons disappear from your pages).

    Again, I don't know anything about kernels and Linux so I am just asking some quasi-technical questions that if possible need to be answered by putting the cookies on the lowest possible shelf.

    Thanks in advance.
  7. #327  
    Can someone explain me how to add the testing feed in Quickinstall?? (wich link should i use?)

    Or can someone upload the uber.ipk file ?

    regards
  8. #328  
    FWIW...

    I've used the SPK and I am currently using Uber Kernel. I have not noticed any difference between the two. Under uber kernel my Pre is just as fast as the old .sh file

    My Pre has no lag anywhere, I can scroll through my launcher, I can breeze through webpages, apps open quickly... so for me everything is as it was, just IMO 'safer' and 'more stable' (again NOT starting any flamewar.. just giving my PERSONAL experience)

    Yes, I get luna reboots, but I was getting that regardless of what overclock method I was using, and it has been said MANY times that it is NOT related to overclocking, but a bug in 1.4 and 1.4.1.1 and hopefully Palm is able to knock it out. (to be clear I was getting the reboots without any type of overclock)

    Also, I doctored my Pre because there were leftover files from SPK *before* I went to the Uber Kernel. I am NOT pointing fingers, but maybe if people are truly seeing lag with uber kernel after upgrading from SPK, maybe there are leftover files affecting the Pre/webOS??
    Last edited by fastamx79; 05/04/2010 at 06:29 PM. Reason: edited some stuff
  9. #329  
    Quote Originally Posted by mebe View Post
    Can someone explain me how to add the testing feed in Quickinstall?? (wich link should i use?)

    Or can someone upload the uber.ipk file ?

    regards
    the first post in this thread.
  10. #330  
    After how hot my Pre ran yesterday, I, too, am giving Performance at 800 a try. So far today it seems more stable and not running nearly as hot. Maybe my particular phone didn't like scaling...each of the kernel modifications will affect each of our phones differently due to binning of the CPU's as has been explained elsewhere.
    Last edited by Blaize; 05/04/2010 at 06:31 PM. Reason: grammar oops
    Blaize, Mistress of Verbosity



    Be nice until it's time to not be nice.--Dalton, "Roadhouse"
  11. #331  
    Quote Originally Posted by mebe View Post
    Can someone explain me how to add the testing feed in Quickinstall?? (wich link should i use?)

    Or can someone upload the uber.ipk file ?

    regards
    it's an entrance test. they only want testers that can figure that out (grin)
  12. #332  
    Quote Originally Posted by fastamx79 View Post
    FWIW...
    Also, I doctored my Pre because there were leftover files from SPK *before* I went to the Uber Kernel. I am NOT pointing fingers, but maybe if people are truly seeing lag with uber kernel after upgrading from SPK, maybe there are leftover files affecting the Pre/webOS??
    Right after the 1.4.1.1 update I decided to give Uber a try with Govnah. At the time I was allowing Govnah to run while I charged my Pre overnight on my TS so I could track the temp range for an extended period. I didn't realize the memory build up would cause the crash and restart thing at the time, but I also was getting some really high temps. It would cruise along really steady around 32 then start climbing and hit 52 or higher and just stay there. So I thought maybe there was something left over from the earlier SPK. So I doctored to a clean 1.4.1.1 and reinstalled Uber and Govnah clean. Same thing continued. So when the updated SPK app came out, I switched to it with no scaling at all, just straight 800 and used it with no issues until a few days ago. When the screenstate update in Govnah came out I switched back over to it and Uber.

    I don't seem to be having the heat issues with the screenstate on the touchstone. Just the other governor stability problems I mentioned in my earlier post. I have switched to performance with min and max set to 800 and will see how that goes.

    So, after a long post... starting clean didn't alleviate any of the issues I had.
  13. #333  
    I started clean and it is still is a bit lagy in launcher pages.
  14. #334  
    Quote Originally Posted by pastorrich1 View Post
    Right after the 1.4.1.1 update I decided to give Uber a try with Govnah. At the time I was allowing Govnah to run while I charged my Pre overnight on my TS so I could track the temp range for an extended period. I didn't realize the memory build up would cause the crash and restart thing at the time, but I also was getting some really high temps. It would cruise along really steady around 32 then start climbing and hit 52 or higher and just stay there. So I thought maybe there was something left over from the earlier SPK. So I doctored to a clean 1.4.1.1 and reinstalled Uber and Govnah clean. Same thing continued. So when the updated SPK app came out, I switched to it with no scaling at all, just straight 800 and used it with no issues until a few days ago. When the screenstate update in Govnah came out I switched back over to it and Uber.

    I don't seem to be having the heat issues with the screenstate on the touchstone. Just the other governor stability problems I mentioned in my earlier post. I have switched to performance with min and max set to 800 and will see how that goes.

    So, after a long post... starting clean didn't alleviate any of the issues I had.

    Like I said... FWIW.. I guess I forgot the YMMV tag too

    But I've used a few different settings since Uber Kernel came out and I haven't had too much of a heating issue.. Maybe 43c was the highest I saw streaming internet radio and leaving my Pre plugged into my car charger. TS issues I have not had also.

    But this proves every Pre is different, and everyone will go though different growing pains, and what works for some, will not work for others.

    But for me, the most stable would be 'performance' settings leaving it pegged at 800mhz, and now the 'screenstate' settings having it at 500mhz and 800mhz. Not one issue of any overheating, or crashes where govnah was directly involved (besides playing with settings and discovering anything below 500mhz would make my Pre lock up.
  15. #335  
    I am looking forward to giving this a try. So I can now use Preware to download this kernal....correct? If I am having issues I can just use Preware to remove it and in theory that should eliminate any issues I was having. Is this correct?

    Thanks
    There is no spoon
  16. #336  
    I am using userspace,,500,800,800. And battery life is unbelievible now.heat never gets above 40c and phone is blazing fast. I used for an hour strait,,heavy use,,and only burned 11 percent. I am loving this.
  17.    #337  
    Quote Originally Posted by Blipstein View Post
    I am looking forward to giving this a try. So I can now use Preware to download this kernal....correct? If I am having issues I can just use Preware to remove it and in theory that should eliminate any issues I was having. Is this correct?
    That is correct. Make sure you read the description for the package in Preware.

    -- 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. #338  
    Quote Originally Posted by rwhitby View Post
    That is correct. Make sure you read the description for the package in Preware.

    -- Rod
    thank you sir
    There is no spoon
  19.    #339  
    Quote Originally Posted by pastorrich1 View Post
    The earlier overclocking kernel solutions operated at specific frequencies and the kernel installed contained that information in the package you chose to install. It would only change if you installed a scaler application of some sort or used the linux commands.
    That is correct.

    Uber operates at a default 500 unless the commands are entered or an app is used.
    That is also correct. That principle is in place specifically so that if your device is unable to reliably handle the overclocking settings, you will not end up in a reboot loop and need to visit the webOS Doctor.

    So what in the kernel package makes one run at a given frequency?
    The previous overclocking solutions did this by making a binary modification to the Palm LunaSysMgr program in the same package which deployed the kernel. The bundling of a kernel and a non-OTA-safe binary patch together into one package is one of the items to which I had (and still have) a strong technical objection, because it causes problems when you get a new webOS version arriving. Note that this modification is *not* removed if you simply install the recovery Palm Kernel, which is why I ask people to go back to the maintainer of their previous overclocking solution to get it completely removed before using the Uber-Kernel.

    It seems to me if all the other things are equal then whatever makes one kernel default to 800 all the time would be different. Could it be that this difference produces different dynamics when installed?

    Could it also be that since the pre booted up with the static overclocked kernels it somehow had different performance characteristics than one that is overclocked after the boot is complete?
    It is quite possible that is the case. I hope Jason R is able to create an AUPT-4 binary patch package which replicates the modification to LunaSysMgr in an OTA-safe way, so we can all find out for sure.

    -- 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
  20. #340  
    Quote Originally Posted by rwhitby View Post
    It is quite possible that is the case. I hope Jason R is able to create an AUPT-4 binary patch package which replicates the modification to LunaSysMgr in an OTA-safe way, so we can all find out for sure.

    -- Rod
    Thanks for the thorough answer that still placed the cookies within reach of my stumpy nubs.

    I too hope J R can do that. I also wish I had an extra Pre so I could work things like this side by side and get some not so subjective measurements.

    That would be helpful for measuring all kinds of apps and patches and their impact on performance / function. Maybe I'll pick one up on ebay or something?

    I can't develop but I could provide objective measurement for development.

Posting Permissions