Hi Everyone,
Took the plunge for the first time and flashed from stock to CM7. I can't get over how fast and stable it is compared to the stock ROM!
I've reading good things about Hey It's Lou (in particular #8 and #9) kernels and would love to get that kind of battery life, but it seems like I can't use HIL with ASOP ROMs like CM7.
1. Am I correct? CAN I use HIL#8/#9 w/ CM7?
2. If not what, other kernels can folks recommend for me? (stability and battery life are more important than performance, though I don't want the phone to crawl.)
Thanks!
A.
Yeah, I'd like to know too. I tried Chad's Incredikernal, and I have to say I wasn't impressed with battery life, perhaps I missed something?
---------- Post added at 04:36 PM ---------- Previous post was at 04:02 PM ----------
I just read a post that said Incredikernal + setting cpu down to half (whatever that means) gave 12-13 hours normal use with battery left(again not specific). I'm all for that! BTW: 10th Post Yeah!!
Hello, I have a launch Evo Shift (almost a year old now), and am running Cyanogenmod Nightly 255 with a2ext and a ton of apps installed (above 200 apps running from a class 10 32gb card, dalvik cache on the card), and ScaryKernel with the smartass governor (max 1497 MHz). However, I sporadically get random reboots, gps not working (requiring a reset), camera reboots, and the rare "turn off" bug where it turns off while sleeping. In addition most recently I have a bug where every time that I reboot into clockworkmod recovery to back up , it resets almost all of my phone preferences.
I absolutely love cyanogenmod and appreciate the hell out of it and the work that has been done, but I'm wondering if these bugs are either due to running cyanogenmod nightlies or if I have a defective phone. I also have tried Sense-based roms such as Mikshifted, but the Sense memory overhead is just too large for me. I have no worries if the phone is defective, since I have sprint insurance and can wipe back to stock to exchange it. However, going through all the exchange, re-root and restore procedure would be a pain if its not necessary. Does anyone have any thoughts or advice?
Thanks in advance.
I don't think this is s shift problem. I experienced this myself and even had my phone replaced by sprint because i thought it was. Try not overclocking and see if that helps. Another way to test if its the phone is to see if you have these same problems on a sense rom
Sent from my PC36100 using xda premium
draven760 said:
Hello, I have a launch Evo Shift (almost a year old now), and am running Cyanogenmod Nightly 255 with a2ext and a ton of apps installed (above 200 apps running from a class 10 32gb card, dalvik cache on the card), and ScaryKernel with the smartass governor (max 1497 MHz). However, I sporadically get random reboots, gps not working (requiring a reset), camera reboots, and the rare "turn off" bug where it turns off while sleeping. In addition most recently I have a bug where every time that I reboot into clockworkmod recovery to back up , it resets almost all of my phone preferences.
I absolutely love cyanogenmod and appreciate the hell out of it and the work that has been done, but I'm wondering if these bugs are either due to running cyanogenmod nightlies or if I have a defective phone. I also have tried Sense-based roms such as Mikshifted, but the Sense memory overhead is just too large for me. I have no worries if the phone is defective, since I have sprint insurance and can wipe back to stock to exchange it. However, going through all the exchange, re-root and restore procedure would be a pain if its not necessary. Does anyone have any thoughts or advice?
Thanks in advance.
Click to expand...
Click to collapse
I would say try using shift deck, U.K.E, or any cm based aosp roms and see if the problem persist. If it does then its your phone, if not then its a bug in the nightly which is the reason its a nightly because bugs could be there
From what you've stated above I don't think it's a phone, or Rom issue. Seems like you have way too many apps. You have to keep in mind a lot of those apps run in the background. Which uses up Memory, which slows the phone. It will also in xtreme cases cause Random Reboots, Lag, even Random app shut down. I'd remove at the very least half of those and see how it performs. Even 100 apps is still a bit much.
Great, thank you for the advice. Even though I have been monitoring my free memory and running programs, I'm kind of a pack rat so I think it's just time to streamline my programs.
Yes definately loose a lot of thoose apps. Might even want to try a fresh reflash, no apps. Then just add what you know you'll definately need. Willing to bet it's not even half.
TEAM MiK
Mik Roms Since 3/13/11
Which governor are you using with Scary kernel? You should be using Scary gov not any of the others.
Min mhz 300+
Check this first.
If that wasn't your problem run fix permissions from ROM manager and reboot.
If that wasn't your problem boot into clockwork recovery, format /system, wipe dalvik cache, wipe cache, reflash cm255, flash crashv2010's gapps, reflash scary kernel, reboot, wait 10 minutes, problems solved.
---------- Post added at 02:10 PM ---------- Previous post was at 02:08 PM ----------
draven760 said:
and ScaryKernel with the smartass governor
Click to expand...
Click to collapse
This is your problem! ONLY USE Scary, Toasty or Batterysave!
Scary = smartass
Toasty = performance
Batterysave! = conservative
ALL OTHER GOVS WILL NOT BE STABLE!!!
Smartass has always been stable for me
Sent from my PG06100 using Xparent Pink Tapatalk
YoungCorruptionV2.0 said:
Smartass has always been stable for me
Sent from my PG06100 using Xparent Pink Tapatalk
Click to expand...
Click to collapse
Depends how large the gap between your min/max cpu is.
What's your min max?
Try widening it with smartass and see how stable it is.
Well I guess I'm the exception . Dodge Kernel v1.6 Ext.4 , smartass setting , 1.92gHz Max , 245gHz Min. Can run that all day, but usually don't. I use the default settings 1.2gHz Max , 245gHz min. The battery life is better at the default settings.
VICODAN said:
Depends how large the gap between your min/max cpu is.
What's your min max?
Try widening it with smartass and see how stable it is.
Click to expand...
Click to collapse
I usually have the min at the lowest and the max at around 1.2 - 1.6 ghz
prboy1969 said:
Well I guess I'm the exception . Dodge Kernel v1.6 Ext.4 , smartass setting , 1.92gHz Max , 245gHz Min. Can run that all day, but usually don't. I use the default settings 1.2gHz Max , 245gHz min. The battery life is better at the default settings.
Click to expand...
Click to collapse
Some shifts can run 1.9ghz some cant. I have 2 shifts here, both hardware rev 004, one can't go past 1.6ghz.
---------- Post added at 01:38 AM ---------- Previous post was at 01:36 AM ----------
YoungCorruptionV2.0 said:
I usually have the min at the lowest and the max at around 1.2 - 1.6 ghz
Click to expand...
Click to collapse
smartass on pershoot is ok
smartass on scarykernel is not ok.
Yes definately know what you mean. Out of six mine is the only one that will do 1.9gHz . All hardware 004. Out of the others two will do 18gHz, but only one is stable. The rest max out at 1.5gHz, after that they're not stable.
TEAM MiK
Mik Roms Since 3/13/11
VICODAN said:
Some shifts can run 1.9ghz some cant. I have 2 shifts here, both hardware rev 004, one can't go past 1.6ghz.
---------- Post added at 01:38 AM ---------- Previous post was at 01:36 AM ----------
smartass on pershoot is ok
smartass on scarykernel is not ok.
Click to expand...
Click to collapse
Probably true, i'm on tiamat v1.1.4 with smartassV2 so that could make a difference but i always use smartass cause i love the battery life i get with it
prboy1969 said:
Yes definately know what you mean. Out of six mine is the only one that will do 1.9gHz . All hardware 004. Out of the others two will do 18gHz, but only one is stable. The rest max out at 1.5gHz, after that they're not stable.
TEAM MiK
Mik Roms Since 3/13/11
Click to expand...
Click to collapse
YoungCorruptionV2.0 said:
Probably true, i'm on tiamat v1.1.4 with smartassV2 so that could make a difference but i always use smartass cause i love the battery life i get with it
Click to expand...
Click to collapse
Thank you for confirming my opinion as fact. I will reference this thread again later if I need to.
i would get new rom manager that fixed some of my problems also try reloading cm7 and put back on you phone with titanium backup
VICODAN said:
Some shifts can run 1.9ghz some cant. I have 2 shifts here, both hardware rev 004, one can't go past 1.6ghz.
---------- Post added at 01:38 AM ---------- Previous post was at 01:36 AM ----------
smartass on pershoot is ok
smartass on scarykernel is not ok.
Click to expand...
Click to collapse
I have rev 0004 and mine clocks to 2ghz just fine...Some kernels just can't handle 2ghz, and most only clock to 1.8 or 1.9
I inadvertently hijacked yugendreams thread a few days ago who was asking for which rom was the best. I stated I did not have good luck with Comi-X on my tablet and recommended going with Cyan Based rom instead. My reasons I stated were crashed reboots/performance of Cromi on my tablet.
This response was my opinion at the time and not very well articulated or well thought out, I admit. After that some discussion ensued and I told sbdags I would try Cromi-X again. In the former post I had stated it had crashed and sometimes I would factory refresh. I believe now (just guessing but pretty sure) the SDcard wasn’t seated properly and with data2sd enabled this was causing my instabilities. I have been able to duplicate the data wiped scenario by removing my SDcard. So this was likely the root cause of my stability issues as I had none my second time around (ran for about 24 hours). My first time through was with 5.1 and now I used 5.2 for what it is worth.
Full disclosure, I have biased toward AOSP roms, let me get that out of the way. I believe that ASUS, Samsung, HTC, etc. are hardware companies and the do a good job at what they do…making hardware. What they fail at is taking something a software company (Google) does and try to improve it. So any ROM that is based on a manufactures image I immediately have a little negative feelings towards it (how horrible the TF700T is stock vs Nexus 7 stock certainly helps solidify my claims). However I understand sometimes it is best to start with something that has been tested on the target platform and make it better.
So long story short, I ran a Androbench and Quadrant on my CMB (CM 10.2 base)/Bryce09 kernel which had better perceived smoothness/speed. I then backed up my rom, wiped in TWRP, flashed Crom-X (5.2a) selecting _Thatsv6 Kernel and mostly stock settings + data2sd. Rebooted and over the next 24 hours ran androbench and quadrant over and over again, I tried without DATA2SD with little to no difference and also Hunds kernel.
Here are my results comma delimited for each recorded run, sorry not formatted the best, I got lazy:
************************************************************************
Bryce/CMB (10.2/4.3 BASED, NO DATA2SD)
AndroBench Micro tests
SR: 14.33MB/s , 16.93, 13.37
SW: 78.37 , 171.93, 126
RR: 4.57 MB/s, 1171.42IOPS(4K), 4.29/1100.14, 4.25/1091
RW 143.63 MB/s, 36769.48(4K), 335.98/86012, 188/48079.03
Quadrant
5488, 5698, 5619
************************************************************************
Cromi-x 5.2/_that6-oc, DATA2SD
Microbench
SR:20.99,
SW:4.22,
RR:4.48/1147,
RW:0.47/121,
Little change on the subsequent test so failed to record, but super slow Random Writes….
Quadrant
4647, 4432, 4554
************************************************************************
Crom-x5.2/_that6-oc, NO Data2sd
No change on AndroBench MicroSD test, should have recorded but basically same as above…limitation in kernel?
Quadrant
4473, 4770 4642
************************************************************************
Crom-x5.2/Hunds Kernel
SR:17, 16, 17.38
SW:6.59, 3.88, 6.79
RR: 1165, 1123, 1107.5
RW: 50.6, 51.62, 47.12
Quadrant:
5432, 5235, 5337
************************************************************************
All of the results were slower than CMB/Bryce09, now I am not trying to say one is better than the other here. In fact I think Quandrant score with CMB is still not as good as I have seen folks get here with Cromi-X. I have since restored my backed up CMB and am back to the faster results. Why are my Random Writes so fast with CMB with nodata2sd?
General thoughts, is there better benchmark tool to use….at this point I think I am going to say Data2SD is not needed regardless of rom….
sjgieson said:
Why are my Random Writes so fast with CMB with nodata2sd?
Click to expand...
Click to collapse
Because the kernel ignores fsync calls. You can do the same with my kernel, see instructions in the kernel thread.
Something still too low on those benchmarks. I'll run androbench now and post up. I can get 1000 to 2000 more on quadrant on that 6.
---------- Post added 26th November 2013 at 12:09 AM ---------- Previous post was 25th November 2013 at 11:53 PM ----------
Cromi-x 5.2/_that6-oc, ROM2SD
Microbench
SR:7.16,
SW:115.49
RR:4.06/1039
RW:250/64000
Obviously rom2sd messing up the numbers. I don't think you have your data2sd setup correctly.
Quadrant
6500 to 6800
---------- Post added at 12:25 AM ---------- Previous post was at 12:09 AM ----------
Ok results on a non rom2sd or data2sd setup
Cromi-x 5.2/_that6-oc
Microbench
SR:9.48
SW:137.87
RR:3.67/940.51
RW:193.04/49418.27
Quadrant
6200 to 6500
-----------------------------------------------
Just realised I forgot to run them in performance mode..... Doh!
---------- Post added at 12:30 AM ---------- Previous post was at 12:25 AM ----------
Now on performance mode
Cromi-x 5.2/_that6-oc
Microbench
SR:20.95
SW:239.87
RR:5.81/1487.83
RW:139.13/35617.5
Quadrant
6400 to 6800
Thanks for response, I seem to get the "I" folder's when I set up cromi with my full access to the card size wise. Will give it another go....so I can trouble shoot some more....
Sent from my ASUS Transformer Pad TF700T using Tapatalk 4
Can anyone help me understand this?
What program are you using? think the antutu benchmark is screwed right now as it CANT detect multiple cpus for our device...
ars0n said:
What program are you using? think the antutu benchmark is screwed right now as it CANT detect multiple cpus for our device...
Click to expand...
Click to collapse
That's what I was using. Any recommendations for a good benchmark?
I believe a lot of people are using the 3dmark for benching... honestly, i benched my phone when I first got it and it was around 58k I rarely have ever seen it higher. I was reading a while back that the antutu program cant see all of our cores correctly and thats why we get such low scores atm. I havent used 3dmark on this phone so cant really give you a comparison to mine. Are you super anal like me and just want to see the max capabilities of your phone vs others or just for piece of mind to know your phone is rock solid/fast? I go for both =D I want to see what my phone is like compared to others, but in the long run i think that does more damage than good. All these notes should be running identical and we dont know what settings and or apps other people have when they run the benchmark, the only for sure would be to make sure that every single person that is benchmarking is on a stock phone with NO tweaks or apps installed other than the benchmarking software..then compare..
---------- Post added at 01:40 PM ---------- Previous post was at 01:36 PM ----------
Also, if you go to the main samsung note 5 page on XDA there is a benchmarking forum you can read through to get a better idea.
http://forum.xda-developers.com/note5/general/antutu-benchmark-t3253521
There is a link to it, but, I think you might end up driving your self crazy like I did going through it all.. lol..
ars0n said:
I believe a lot of people are using the 3dmark for benching... honestly, i benched my phone when I first got it and it was around 58k I rarely have ever seen it higher. I was reading a while back that the antutu program cant see all of our cores correctly and thats why we get such low scores atm. I havent used 3dmark on this phone so cant really give you a comparison to mine. Are you super anal like me and just want to see the max capabilities of your phone vs others or just for piece of mind to know your phone is rock solid/fast? I go for both =D I want to see what my phone is like compared to others, but in the long run i think that does more damage than good. All these notes should be running identical and we dont know what settings and or apps other people have when they run the benchmark, the only for sure would be to make sure that every single person that is benchmarking is on a stock phone with NO tweaks or apps installed other than the benchmarking software..then compare..
---------- Post added at 01:40 PM ---------- Previous post was at 01:36 PM ----------
Also, if you go to the main samsung note 5 page on XDA there is a benchmarking forum you can read through to get a better idea.
http://forum.xda-developers.com/note5/general/antutu-benchmark-t3253521
There is a link to it, but, I think you might end up driving your self crazy like I did going through it all.. lol..
Click to expand...
Click to collapse
Thanks, I really dint cate about the score, I just want to make sure that I don't have a problem. Dunce my multi core, and 3d don't appear to be skirting at all.
My antutu also would NOT run the garden scene correctly with out crashing after a few runs. had to reinstall to work and then eventually said F it.. lol..
Its hard to say when things happen to the phone that slow it down, most of the time it is an app... but the best option if unsure is to master reset and install only that app that is giving you issues and see if it fixes it. Also paying attention to comments on the app to see if any other users might be having the same problems before doing that master reset..
I usually find a new game or a game I know is graphic intensive, like hearthstone, GTA games or other and if the phone plays it fine, then no worries!