Can I overclock my Bionic?

Bear in NM

Active Member
Joined
Nov 12, 2009
Messages
1,389
Reaction score
1
Ex,

Yes, a pretty good place to be right now, especially if you are not having data connection problems. Since the the OTA will have a modified kernel over stock and over several of the test upgrades, it kinda makes sense for someone like Prime to wait until the drop. It is pretty obvious the rommers are doing this to some extent. Just prepare yourself for some ugly threads here once the OTA hits. There are so many gyrations of configurations out there with kernels, radios and webtops that I am half afraid it is going to take a new FZX to get everyone squared away properly. I would expect the FZX to be available to the stores in a timely fashion, just no idea how long it will take to get "leaked". I do remember having one of those "woa boy this is going to get ugly" thoughts when I first saw the Cheesecake efforts, and to some extent it has.

Craig
 

prime

Kernel Developer
Premium Member
Developer
Joined
Apr 26, 2010
Messages
1,939
Reaction score
70
Location
Florida
Website
github.com
Ex,

Yes, a pretty good place to be right now, especially if you are not having data connection problems. Since the the OTA will have a modified kernel over stock and over several of the test upgrades, it kinda makes sense for someone like Prime to wait until the drop. It is pretty obvious the rommers are doing this to some extent. Just prepare yourself for some ugly threads here once the OTA hits. There are so many gyrations of configurations out there with kernels, radios and webtops that I am half afraid it is going to take a new FZX to get everyone squared away properly. I would expect the FZX to be available to the stores in a timely fashion, just no idea how long it will take to get "leaked". I do remember having one of those "woa boy this is going to get ugly" thoughts when I first saw the Cheesecake efforts, and to some extent it has.

Craig

I didn't know a new kernel was eminent, I will definitely need to wait as the memory addresses will change along with the kernel vermagic.

Already being helpful per normal, thanks for the information.
 

sjflowerhorn

Member
Joined
May 26, 2010
Messages
99
Reaction score
1
Location
Collingswood
I am running 1.2 ghz currently check rootzwiki for info, d3 oc method applied very experimental.
Have fun ladies and gentz!

Sent from my DROID BIONIC using DroidForums
 

Bear in NM

Active Member
Joined
Nov 12, 2009
Messages
1,389
Reaction score
1
Prime,

I do not know of what changes may be in the works, but the latest OTA leak does have a new kernel number. This would be the latest 94, which is different than the various 93's which are all different than the original where most folks are sitting. Whether or not there are enough changes in these various kerenles to affect addresses, I have not a clue. What we are awaiting is the first official OTA, the typical fix the major bugs from the initial release of the phone. Things should in theory settle down, as the next OTA is probably going to be ICS. At least that is my edumacated guess from following the official bug trackers with moto/google. The serious issues have been targeted for this first OTA, with lesser bugs being relegated to no official word of fix date. Probably just roll the whole kit and kabootle into ICS and be done with it (wag).

You probably have not followed the leaked status of our Bionic, but essentially the folks at XDA are using a nifty app to poll and pull all of the moto servers. Hence we have 4 or 5 versions of images that have been treated as possibly the final first OTA. And we have folks all over the place with kernel versions, radio images and webtop images. If you do get to feeling frisky, you will need to make sure whomever is helping to test is clear about what they are running.

SJ, it might be helpful to know what images you are currently sitting at with your phone. Stock and root v. one of the various leaked updates. Folks could get seriously fouled up, especially if they try and do any of the persist methods of sticking the values through a boot. Not so dangerous if they (and you) are running stock and rooted, more so if they have mods that might preclude the current fzx from working. Cool to know, though. The D1's were underclocked by something close to 100%. It looks like we are more in the 50% range, but that is still a bunch. And seriously tempting:icon_evil:

Craig
 

sjflowerhorn

Member
Joined
May 26, 2010
Messages
99
Reaction score
1
Location
Collingswood
At first I was running stock rooted 5.7.893 now I am running liberty 3 2.0 rc1 both are working fine but lose values on boot.

Sent from my DROID BIONIC using DroidForums
 

prime

Kernel Developer
Premium Member
Developer
Joined
Apr 26, 2010
Messages
1,939
Reaction score
70
Location
Florida
Website
github.com
The vermagic will definitely change (vermagic example being 2.6.38 to 2.6.39 or sub versions).
 

ex0rcist

Member
Joined
Sep 12, 2011
Messages
328
Reaction score
6
At first I was running stock rooted 5.7.893 now I am running liberty 3 2.0 rc1 both are working fine but lose values on boot.

Sent from my DROID BIONIC using DroidForums

Must overclock... show us howwwww. :)

Sent from my DROID BIONIC using Tapatalk
 

Bear in NM

Active Member
Joined
Nov 12, 2009
Messages
1,389
Reaction score
1
[OVERCLOCK]XT860/XT862 Overclock - xda-developers

This is the XDA link. I did not see it on Rootzwiki, but probably not looking correctly. In this link is a zip file which appears to work for the D3. The link also contains the link to the person who actually figured this out, along with the code he/she is using. I am assuming the zip file contains this code. I am on a windows machine, and it appears that you need Linux to see the code in the zip file. When unzipped in windows, they show as zero size files, but with a hex editor program I can see something. They are not empty files.

It appears that in order for this to work, you have to set Setcpu to full throttle, min and max. And Setcpu does not report accurately, but the folks are using benchmark programs to "confirm" overclocking. So when using this hack, your phone is going to be forced to run wide open, no native underclocking. And it would be harmless to try to check for functioning, as it does not survive boot. May not be harmless in use for long periods, but that needs to be tested on each phone. And it appears this is a work in progress.

As to surviving the reboot, probably the easiest way would be to use a program like Gscript and just one button push to start after boot. You could set up an autorun program as one poster mentions, but this could be very dangerous unless you have the backups and system setup to restore from recovery.

At issue here is the overclocking and heat and speed, but just as important is voltage. Folks are reporting unstability at higher clock speeds, but this could easily be a voltage issue, and not just the pure speed. It would be my best guess that as moto is using our omap at reduced clock speed, that they have matched the system voltage to work with the reduced clock speed. If this guess is correct, then until the slots and voltages can be addressed together, there is a bit of dark stabbing going on, as undervolted conditions are just as unstable as excessive speed, and undervolting the CPU may be as harmful as the excess speed. At least with excess speed and voltage you usually get high temps, and the system will shut itself down with high temps. At least to date this has been the case.

I only mention this stuff, as it would be easy to say to yourself, hey it is working and I am good to go because my system is not bootlooping or shutting down. With the D1, Prime got us to be able to set our speeds and voltages so we were able to ensure that we were not in undervolting scenarios, and we could balance performance and battery life. So this is some good stuff, but does need to be approached with a bit of understanding, as all things root.

Craig
 

ex0rcist

Member
Joined
Sep 12, 2011
Messages
328
Reaction score
6
[OVERCLOCK]XT860/XT862 Overclock - xda-developers

This is the XDA link. I did not see it on Rootzwiki, but probably not looking correctly. In this link is a zip file which appears to work for the D3. The link also contains the link to the person who actually figured this out, along with the code he/she is using. I am assuming the zip file contains this code. I am on a windows machine, and it appears that you need Linux to see the code in the zip file. When unzipped in windows, they show as zero size files, but with a hex editor program I can see something. They are not empty files.

It appears that in order for this to work, you have to set Setcpu to full throttle, min and max. And Setcpu does not report accurately, but the folks are using benchmark programs to "confirm" overclocking. So when using this hack, your phone is going to be forced to run wide open, no native underclocking. And it would be harmless to try to check for functioning, as it does not survive boot. May not be harmless in use for long periods, but that needs to be tested on each phone. And it appears this is a work in progress.

As to surviving the reboot, probably the easiest way would be to use a program like Gscript and just one button push to start after boot. You could set up an autorun program as one poster mentions, but this could be very dangerous unless you have the backups and system setup to restore from recovery.

At issue here is the overclocking and heat and speed, but just as important is voltage. Folks are reporting unstability at higher clock speeds, but this could easily be a voltage issue, and not just the pure speed. It would be my best guess that as moto is using our omap at reduced clock speed, that they have matched the system voltage to work with the reduced clock speed. If this guess is correct, then until the slots and voltages can be addressed together, there is a bit of dark stabbing going on, as undervolted conditions are just as unstable as excessive speed, and undervolting the CPU may be as harmful as the excess speed. At least with excess speed and voltage you usually get high temps, and the system will shut itself down with high temps. At least to date this has been the case.

I only mention this stuff, as it would be easy to say to yourself, hey it is working and I am good to go because my system is not bootlooping or shutting down. With the D1, Prime got us to be able to set our speeds and voltages so we were able to ensure that we were not in undervolting scenarios, and we could balance performance and battery life. So this is some good stuff, but does need to be approached with a bit of understanding, as all things root.

Craig

Seen any reports of high temps or screwy voltages?

Sent from my DROID BIONIC using Tapatalk
 

ex0rcist

Member
Joined
Sep 12, 2011
Messages
328
Reaction score
6
I have not. Warms up under gaming or tethering. . . Like usual but nothing out of the blue.

Sent from my DROID BIONIC using DroidForums

Same. I don't notice anything unusual either.. same old warm feeling while gaming or tethering... what freq u using? 1.5 for me. And this thing is super super fast. Running neck n neck to similar clocked sgIIs in cfbench not to mention roaring past stock rzrs. Very happy with the developments on the bionic in the past week... now another team on db rom dev groups...very happy i didn't go to rzr. Very happy to see people still interested in our device... shes a beast :)

Sent from my DROID BIONIC using Tapatalk
 
Top