Database Info

Overclocking Qualcomm 7201 on Android OS

Hi there!!
Seems to me no one hasn't began this knd of theme. So. Off. info:
Qualcomm MSM7201a
Two cores
ARM11 + ARM9
528 Mhz (one stated speed for two cores...)
So what is G1's heart for real?
This is quotes from here
#1
Jean-Baptiste Queru 26 нояб 2008, 20:37
Indeed, the CPU in the G1 is clocked lower than its maximum rated
speed to conserve battery life. It's running somewhere between 300 and
400MHz if I remember correctly.
JBQ
Click to expand...
Click to collapse
#2
jdc4429 27 нояб 2008, 17:04
Hi Jean,
So your saying that the CPU speed is not controlled by the Android
software?
I was looking through the code and found this in the arch/arm/mach-msm/
clock.c file...
617 #define CPUFREQ_TABLE_MAX 4
618 static struct cpufreq_frequency_table cpufreq_table[] = {
619 { 0, 81920 },
620 { 1, 122880 },
621 { 2, 245760 },
622 { 3, 384000 },
623 { CPUFREQ_TABLE_MAX, CPUFREQ_TABLE_END },
624 };
It looks like the max speed is set to 384mhz and it seems it can be
easily changed.
It also seems that the phone already downshifts the CPU based on this
table and the
screen_open/closed speed setting...
702 if (screen_on) {
703 policy->user_policy.min = cpufreq_table
[2].frequency; // 245mhz
704 policy->user_policy.max = cpufreq_table
[3].frequency; // 384mhz
705 } else {
706 policy->user_policy.min = cpufreq_table
[0].frequency; // 82mhz
707 policy->user_policy.max = cpufreq_table
[2].frequency; // 245mhz
708 }
Sure looks controllable to me through Android. Is it really that hard
to add a setting to allow min/max settings
to be adjusted by the user?
Thanks
Jeff
Click to expand...
Click to collapse
#3
Romain Guy 26 нояб 2008, 20:39
> Can that
> be changed in software on the fly and was it set below maximum speed
> to help with the battery issue?
No and yes
> Also is anyone working on adding hardware acceleration so we can take
> full advantage of the processor?
We have a prototype of SGL running on top of OpenGL (it was actually
shown publicly in the SDK 0.9) but it's not the correct solution at
the moment.
Click to expand...
Click to collapse
Coclusion:
1) Google DevTeam does not know, or don' wanna tell, ca we\how to change CPU's speed or\when it would be possible to get the hardware accelariotion etc.
2) From stated 528Mhz we get 384Mhz maximum, as stated by Jean-Baptiste Queru and the code quote.
Both these I suppose seems not fare for us users)
So, can it be solved through the OS modifing??

Oh yeah that's another good point -- almost all of my experience on mobile
hardware has been that the memory bus speed was far more of a performance
bottleneck than the CPU was. It generally just wasn't useful to run the CPU
at its fastest speed and consume more power, because most of what it would
be doing was sitting there waiting on memory.
Click to expand...
Click to collapse
I don't know what is more important, the memory/cpu ratio working at its best or battery life. We'll have energy plans soon for android me thinks..

interesting... Anyone want to give it a try?

It's very worthwhile to read the entire thread:
http://tinyurl.com/9kme8v
Of particular note, the effect on battery life of clocking the CPU at full speed, and the apparently minimal performance boost.
Of real interest, and the very obvious bang for the buck, is the speed of Dalvik. Note that it's 7-8x slower than comparable JITs. About in line with what you'd expect, but it does imply that if we want to see some serious speed increases - and, I would think, battery life improvements - replacing Dalvik would be the obvious place to start. Or making it JIT, of course.

might be but lets say im at home plugged in or what not... then we could scale our CPU... like BatteryStatus or integrate it into an app somewhat like Locale or Power Manager...
End users look at the End result. my phone is rated at 528mhz and it is running slightly over half its rated operating speed. Battery life? Give me 528mhz and let HTC and Tmobile Recall there $h*tty batterys

The only advantage I can think of for this is for demanding games for instance but any other time its good to have it under clocked to save on battery juice. and the phone is fast enough running at half its rated speed!
Phil

philje123 said:
The only advantage I can think of for this is for demanding games for instance but any other time its good to have it under clocked to save on battery juice. and the phone is fast enough running at half its rated speed!
Phil
Click to expand...
Click to collapse
once you get demanding apps installed such as hello aim. phoneplus dgalerts etc there is constant hiccups and the phone becomes bogged down like i said give me 528mhz and let htc,tmo replace there $h**ty batteries didnt they hear over a year ago there was a huge advancement in nano tech for batteries

diabolical28 said:
once you get demanding apps installed such as hello aim. phoneplus dgalerts etc there is constant hiccups and the phone becomes bogged down like i said give me 528mhz and let htc,tmo replace there $h**ty batteries didnt they hear over a year ago there was a huge advancement in nano tech for batteries
Click to expand...
Click to collapse
True, but *if* it is the CPU which is bogging down (which I am not convinced is the case), the solution is as I stated - speed up Dalvik - not to put the CPU into a mode which drains the battery down even more quickly than it does today. If the fix is to clock the CPU at a higher rate (which again, seems not to be the case), that's only a stepping stone to the ultimate solution, which is going to be removing some apps, since the battery life is marginal right now anyway.

The reason why Android is underclocked is posted in the full thread. After a certain point, the increase in speed you get by bumping up the CPU slows down, because it's limited by FSB speed. An example of this (not real numbers), is that a CPU running at 50% speed could actually be closer to 75% speed.
Sure, the CPU CAN run at 100% speed, but after the FSB slows it down, it'll only be 75%. So after a point, increasing CPU speed isn't worth it.

hmm
i wonder if we would get an app too monitor it or
under and overclock by adjusting a slider
anywhere from 50% to 150% or whatever is possible
that is anywhere from 264Mhz to 798Mhz which would be amazingly fast

gary, might as well just save your breath.

I believe JF has already tested some code to do this, but mentioned that it is highly unstable.

dirr said:
I believe JF has already tested some code to do this, but mentioned that it is highly unstable.
Click to expand...
Click to collapse
Adjusting the speed breaks things, at least so far anyways. But like he said JF is playing with it.

djind said:
I don't know what is more important, the memory/cpu ratio working at its best or battery life. We'll have energy plans soon for android me thinks..
Click to expand...
Click to collapse
Sure, it might not be entirely "worth" it in terms of performance/battery ratio... but that doesn't mean it isn't worth a try. People keep saying "don't bother" but nobody has actually tried an overclocked G1. Maybe it's worth it for demanding gaming or apps while the phone is (gasp) plugged in.

back when i had my wing and so did ttran. Which uses the OMAP processor it was able to overclock over 100mhz faster than it was meant to go. Had real good luck overclocked to 288mhz. Which made the phone run at the speed that it SHOULD have ran at... including faster loading web pages, faster loading everything basically. Which hardly is an issue with the g1, but imagine how much more snappy it would be with an extra 100mhz also? People say its not worth it and it drains the battery more.... umm... if anyone here owns a g1 (surely hope so at least!) you probably know that the battery sucks anyways and needs charging all the time. Would love to see a good OC app made for the g1 =-) (with scaling like battery status was for the wing, which underclocked when screen was off, overclocked when screen was awakened which was amazing for battery life, as the phone doesnt need much power to accept a call, but when the screen is lit.... scaling occurs and it bumps it right up to 288mhz)

I think its well worth a go.
We could start with very small increments.

Gary13579 said:
The reason why Android is underclocked is posted in the full thread. After a certain point, the increase in speed you get by bumping up the CPU slows down, because it's limited by FSB speed. An example of this (not real numbers), is that a CPU running at 50% speed could actually be closer to 75% speed.
Sure, the CPU CAN run at 100% speed, but after the FSB slows it down, it'll only be 75%. So after a point, increasing CPU speed isn't worth it.
Click to expand...
Click to collapse
I think the question is... Are there any circumstances where this may help...
And either way how difficult is it to add a control under settings to allow you to modify the default method?
You could basically apply the same argument to anything... Until someone tries, we will not know for sure. What next? Don't bother trying to add a swap file cause it may not help? Don't bother trying to speed up Dalvik cause it will take too much memory?
That's the fun of open source...
jdc4429

FSB could really be the bottleneck here. From my experience overclocking mobile CPUs (PXA255, PXA263, PXA270, Samsung 300Mhz, TI OMAP 850) several percentage change on FSB speed or Memory speed could make significant performance changes.
The last WinMob device I had - Dell Axim x51v had a PXA270 [email protected] Pumping the bus from 208Mhz to 230Mhz and the CPU speed from 624 to about 700 did abput 50% improvement in several bencmarks (floating point, integer calculations, memory speed, graphics subsystem).
On the other hand pumping the CPU to 1014Mhz, (0.99GHz, the highest 24/7 stable for my device) while lowering the FSB to 185Mhz (23Mhz drop) led to about 10% lower result than at stock frequencies.
700Mhz CPU, 230Mhz FSB was totally 24/7 under stress tests with almost no extra heat, providing aboyt 50% extra performance, while battery life was about 30-40% shorter.
My experience with it showed that it could be worth for short speed bursts, while running many apps. When you are finished you return to normal frequencies.

Wow! Thanks for the info man!!
Thus we can see, that by increasing the CPU up to it's stock 528 Mhz would be perfect?

it will be perfect when the speed of the cpu uses 528Mhz and auto adjusts: high speed - mid speed - low speed - sleep, "ondemand"
android should use ondemand kernel module which does exactly that. (maybe this cpu cant do that?)

CPU Freq: How do you keep yours?

This is a question I've never seen asked, but it's an important one. What are you phones CPU Freqs? What Govs do they have, and what Profiles do you use?
Rom: PyroIce
Main: 192 Min, 918 Max. OnDemand.
Screen Off: 192 Min, 384 Max. PowerSave.
InCall: 192 Min, 540 Max. Interactive.
Time 11pm-7am: 192Min/Max Powersave.
Sent from my HTC myTouch_4G_Slide using Tapatalk

that's cool, i might start using profiles again. from fiddling around i've noticed the difference between the ondemand and conservative governors- with ondemand, the processor seems to go from min. to all the way to max. and back down to min, and doesn't use the middle frequencies as much, while on conservative, the processor doesn't shoot up to max. as easily and seems to meander thru the middle frequencies a lot more. it makes sense too, because i've noticed that the phone seems snappier with ondemand.
i noticed you're using some of the others... i'll have to check them out as well.

ondemand
Available in most kernels, and the default governor in most kernels. When the CPU load reaches a certain point (see "up threshold" in Advanced Settings), ondemand will rapidly scale the CPU up to meet demand, then gradually scale the CPU down when it isn't needed. - SetCPU website
conservative
Available in some kernels. It is similar to the ondemand governor, but will scale the CPU up more gradually to better fit demand. Conservative provides a less responsive experience than ondemand, but can save battery. - SetCPU website
performance
Available in most kernels. It will keep the CPU running at the "max" set value at all times. This is a bit more efficient than simply setting "max" and "min" to the same value and using ondemand because the system will not waste resources scanning for the CPU load. This governor is recommended for stable benchmarking. - SetCPU website
powersave
Available in some kernels. It will keep the CPU running at the "min" set value at all times. - SetCPU website
userspace
A method for controlling the CPU speed that isn't currently used by SetCPU. For best results, do not use the userspace governor. - SetCPU website
Interactive
The 'interactive' governor has a different approach. Instead of sampling the cpu
at a specified rate, the governor will scale the cpu frequency up when coming
out of idle. When the cpu comes out of idle, a timer is configured to fire
within 1-2 ticks. If the cpu is 100% busy from exiting idle to when the timer
fires then we assume the cpu is underpowered and ramp to MAX speed.
If the cpu was not 100% busy, then the governor evaluates the cpu load over the
last 'min_sample_rate' (default 50000 uS) to determine the cpu speed to ramp down
to

so, I bought set cpu about 3 phones ago, but I never really saw a differnence, can it be used with the stock kernel?

yellowjacket1981 said:
so, I bought set cpu about 3 phones ago, but I never really saw a differnence, can it be used with the stock kernel?
Click to expand...
Click to collapse
Yeah Plus there's a huge difference with the Battery. I you use my CPU Settings, you should gain 3 hours Min on your phone. Also dropping the CPU in call is great, helps for longer calls and less overheating. If your S-Off you should use a custom kernel. UV Kernels are great.

AgentCherryColla said:
Yeah Plus there's a huge difference with the Battery. I you use my CPU Settings, you should gain 3 hours Min on your phone. Also dropping the CPU in call is great, helps for longer calls and less overheating. If your S-Off you should use a custom kernel. UV Kernels are great.
Click to expand...
Click to collapse
I actually just got an uv kernel, I am about to flash it, I will use your settings too, thanks.

yellowjacket1981 said:
I actually just got an uv kernel, I am about to flash it, I will use your settings too, thanks.
Click to expand...
Click to collapse
If you run a Sense 3.6 Rom, boost the CPU to 1Ghz. 918Mhz Lags.
Sent from my HTC myTouch_4G_Slide using Tapatalk

yellowjacket1981 said:
so, I bought set cpu about 3 phones ago, but I never really saw a differnence, can it be used with the stock kernel?
Click to expand...
Click to collapse
Yes.
You cannot use anything the kernel doesn't already serve as an option, though.
SetCPU doesn't create options - it allows you user-level access to the ones already existing.
Normally you have no control over your clock speed but with it you can...say... limit the maximum clock speed to only be two-thirds of the stock speed.
Since it doesn't ramp up all the way, it doesn't drain as much power.
If you are playing games and stuff that needs a lot of resources, then the slower clock speeds cost you battery since it takes more effort to get the same work done. But if you don't do much with your phone, and don't need a lot of high-end CPU ability, why let it run flat out?
You can also set it to be like that only sometimes, and maximize battery savings by telling your device to chill out when it doesn't have to be awesome.
Now, if you get into custom kernels where you or a dev has opened up more options, SetCPU will again give you user level control over them.
Especially once you get into overclocking beyond manufacturer(s) recommendations, not having some end-user level control over the processor like this is irresponsible.
If you think I run a 1.7 or 1.9 or 1.5+anything on the device flat-out balls-to-the-wall all the time, I would ask you if you drive your car with the gas pedal smashed to the floor all the time. (it's a fitting analogy)
No, again, that would be irresponsible abuse of the hardware for no good reason.
If not SetCPU, then there would be something else to use, but SetCPU is a great interface that not only works well but is maintaned within the XDA community by a member here.
I will always go for the XDA community member version of any app first - it's usually a problem solved from a perspective very near the way I am perceiving the issue, supports the community, and is always easier to find help on if necessary.
What's even cooler is SetCPU is given away free by the dev to XDA members.
I did go ahead and buy it anyway to support the dev. It is a fantastic tool at my disposal, and I have come to realize just how much time and energy goes into making this kind of stuff.
Beyond all that, though, there are a ton of crap apps and software out there slapped together by people who don't care or didn't put the right level of effort into their work. When I find something that actually works and works well after sifting through a sea of garbage, if there is a paid version of the app i'll usually buy it not only as a thank you but also to encourage that particular person to keep working. They - unlike many, many others (in my opinion) earned it.
I know i'm drifting off topic a bit, sorry for that - I just really appreciate quality work and SetCPU falls in with that crowd nicely.
Even without a custom kernel providing options beyond stock, there is still a lot of value this app can hold for anyone with root level access to their device.

Blue6IX said:
If you are playing games and stuff that needs a lot of resources, then the slower clock speeds cost you battery since it takes more effort to get the same work done.
Click to expand...
Click to collapse
Is this true? in my mind, I feel like it will take longer/not be as smooth, but wouldn't necessarily be worse for battery than a higher clock speed.

On a one-for-one comparison it leans more towards an even playing field, until you add that it's a dual-core processor and it very heavily skews the results towards less battery at higher clock speeds for complex operations.
Faster CPU speeds process the actions more quickly, and when you have to crunch a lot of numbers real quick like you do for most games the higher clock speed (especially on the dual core we are using) will complete the act with less power used. It is able to keep up with and/or outpace the flow of data being thrown at it.
If you are just using a notepad or browsing some forums maybe...sending a few texts? Then the higher clock speeds show a negative return on battery usage - you don't need it and it's wasteful.
But when the data has to stay buffered because the CPU is taking it's time working with it, not only is there more electricity consumed by keeping the processor alive longer but also the juice spent on keeping the data buffered.
The data computed is the same data computed at any speed (unless you lose data along the way...) but the power spent on completing that operation is not equal - not by a long shot.
On single core older processors where you are talking about a couple hundred megahertz one way or another the equation is much different, significantly diminshed returns on electricity invested, but with higher quality and multi-core processors especially they can cut through the data with significant ease when sufficiently powered.
But this is on processor intensive activities, and gaming is the first thing that comes to mind. For most other use of the device it's worth underclocking it and you will see battery gains.

Any Surface Pro mods?

Had my pro 2 about 3 weeks now. Slightly undervolted using XTU (CPU -55, Cache -70, GPU -85). Tweeked a few settings here and there, use a different stylus and wacom drivers, but nothing major. I wondered if anyone had done any serious mods to their device to have it running better?
Could you underclock the turbo therefore while leaving a lower max potential performance increase the amount of time that it is in turbo rather than throttled?
Since it gets so hot under load and some of that heat gets transferred to the case I wondered if putting to smallish heatsinks on the main hot spots would increase the thermal dispersion from the processors. The back of the case near the vents get to about 55 deg from what I can see through searches. Those little fans really seem to struggle under load resulting in fairly regular throttling. Not sure if this would make much of a difference though. I'd seen ones that were 40 x 60 x 6mm which i thought wouldnt be too much of an additional bulk but would increase the thermal conductive area and mass.

Cut back the turbo boost and max turbo boost from 17.5W and 25W to 15W and 18W. The XTU benchmark score was reduced from 307 to 306, so marginal difference but there was 0 throttling and a max temp of 78deg. I think I have my solution to keep the temp under 80 to eliminate throttling and thus large fluctionations in fps in games.
Conclusion, lower max performance and ever so slight reduction in average performance under high load, but much more stable performance for games and other processor intensive activity.

Thermal throttling and hotplug settings

I have been able to fix quite a bit of thermal throttling experienced in multiple benchmark runs by tweaking with the thermal-engine configuration file. With just a few changes and with the copper shim mod I can now run geekbench for as many five times with almost no drop in score. After that I get no thermal throttling in single core but I have it setup to disable two cores after it hits 55C CPU temp in core 4 and 7. I use CPU Z to see actual CPU temp from the sensor readings. Even after 8 Geekbench runs the two cores are still running at full speed and the phone is registering 44C. It's warm but not hot by any means.
If anyone is interested in playing around with it just look for the file in \System\etc\thermal-engine-8994.conf
Most of it is just two adjustments. One for Max temperature at which it will throttle and the one below is the temperature at which throttling stops.

probaina said:
I have been able to fix quite a bit of thermal throttling experienced in multiple benchmark runs by tweaking with the thermal-engine configuration file. With just a few changes and with the copper shim mod I can now run geekbench for as many five times with almost no drop in score. After that I get no thermal throttling in single core but I have it setup to disable two cores after it hits 55C CPU temp in core 4 and 7. I use CPU Z to see actual CPU temp from the sensor readings. Even after 8 Geekbench runs the two cores are still running at full speed and the phone is registering 44C. It's warm but not hot by any means.
If anyone is interested in playing around with it just look for the file in \System\etc\thermal-engine-8994.conf
Most of it is just two adjustments. One for Max temperature at which it will throttle and the one below is the temperature at which throttling stops.
Click to expand...
Click to collapse
Without your copper shim mod have you found any better but safe values we can try?
At least no thermal throttle until 55 celsiius

It should be safe without the copper shim. The thermal pad it comes with works pretty well. It will still thermal throttle to whatever temperature you set it at so it won't cause any harm.
I've set the Big cluster to a max of 65C and a single A57 core to up to 71C and haven't had any problems yet. I used to get horrible performance with PPSSPP emulator. When running Tony Hawk Project 8 in PPSSPP after 30 seconds of gameplay It used to drop to 6fps and now it can sustain over 20fps after reaching 71C on a single A57 core. I only saw a maximum battery temperature of 40C and the device wasn't noticeably warmer than it was with the default thermal settings.

probaina said:
It should be safe without the copper shim. The thermal pad it comes with works pretty well. It will still thermal throttle to whatever temperature you set it at so it won't cause any harm.
I've set the Big cluster to a max of 65C and a single A57 core to up to 71C and haven't had any problems yet. I used to get horrible performance with PPSSPP emulator. When running Tony Hawk Project 8 in PPSSPP after 30 seconds of gameplay It used to drop to 6fps and now it can sustain over 20fps after reaching 71C on a single A57 core. I only saw a maximum battery temperature of 40C and the device wasn't noticeably warmer than it was with the default thermal settings.
Click to expand...
Click to collapse
[SS-LITTLE_CLUSTER_management]
algo_type ss
sampling 10000
sensor xo_therm
device cluster0
set_point 52000
set_point_clr 42000
action_type 10000
I should change for example here set_point to 70000? This is how it works?

kutulu32 said:
[SS-LITTLE_CLUSTER_management]
algo_type ss
sampling 10000
sensor xo_therm
device cluster0
set_point 52000
set_point_clr 42000
action_type 10000
I should change for example here set_point to 70000? This is how it works?
Click to expand...
Click to collapse
Well yes. Except that's the little cluster and they don't get that hot normally. It's the big cluster that needs to be modified. I also tuned my display thermals so the screen wouldn't get so dim when the phone gets warm. Here you can look over my configuration if you want.
https://www.dropbox.com/s/1yve5d0m7uin0ga/thermal-engine-8994.conf?dl=0

probaina said:
Well yes. Except that's the little cluster and they don't get that hot normally. It's the big cluster that needs to be modified. I also tuned my display thermals so the screen wouldn't get so dim when the phone gets warm. Here you can look over my configuration if you want.
https://www.dropbox.com/s/1yve5d0m7uin0ga/thermal-engine-8994.conf?dl=0
Click to expand...
Click to collapse
Thanks i will try them!!

your settings somehow "unlocked" my phones cpu. lol in geekbench I found an increase of performance of 200 points in multicore from 4020 I had before maximum , now nearly 4300 .It is impressive. also if I run 6-7 geekbench in a row there is a decrease everytime ex 4200 4050 3900 3600..... but temperature stays 45 degree.
I will test them more later.
http://browser.primatelabs.com/user/73366

Someone try this?
Wysłane z mojego LG-H955 przy użyciu Tapatalka

Other than the two of us?

kutulu32 said:
your settings somehow "unlocked" my phones cpu. lol in geekbench I found an increase of performance of 200 points in multicore from 4020 I had before maximum , now nearly 4300 .It is impressive. also if I run 6-7 geekbench in a row there is a decrease everytime ex 4200 4050 3900 3600..... but temperature stays 45 degree.
I will test them more later.
http://browser.primatelabs.com/user/73366
Click to expand...
Click to collapse
Thanks for sharing your results. That's the highest multi core score I've seen on our phones. I never gotten that high of a multi core score. Glad to help. My configuration also tweaked the battery thermal settings so now you should see faster charge times as well. The default charge rates had really conservative thermal settings and the high speed charge would drop after a while but now it will 2.2 to 2.4 amp charge through the whole charge.

probaina said:
Thanks for sharing your results. That's the highest multi core score I've seen on our phones. I never gotten that high of a multi core score. Glad to help. My configuration also tweaked the battery thermal settings so now you should see faster charge times as well. The default charge rates had really conservative thermal settings and the high speed charge would drop after a while but now it will 2.2 to 2.4 amp charge through the whole charge.
Click to expand...
Click to collapse
Yes indeed i saw a difference in chargin thanks for your effort in the configuration. I am not very fond of benchmarks but i run some in my devices after magor changes (firmware....). This device should have been speedier because of the full hd snapdragon 810 combination. I wonder if we delete the thermal config is the device gonna be speedier? I am waiting for 5.1.1 hoping for smoother rom. Lets hope soneone unlocks the bootloader!!!! I took my precautions disabling MLT lol

probaina said:
Well yes. Except that's the little cluster and they don't get that hot normally. It's the big cluster that needs to be modified. I also tuned my display thermals so the screen wouldn't get so dim when the phone gets warm. Here you can look over my configuration if you want.
https://www.dropbox.com/s/1yve5d0m7uin0ga/thermal-engine-8994.conf?dl=0
Click to expand...
Click to collapse
My phone gets really hot with those settings
I overwrote the original one ^^ has someone made a backup and can share it with me.?

I guess it depends on what you're doing but I probably wouldn't play any 3d games with those settings. Here's a copy of the original one https://www.dropbox.com/s/yb76q6qwanb9ujr/thermal-engine-8994.conf?dl=0

smouker said:
My phone gets really hot with those settings
I overwrote the original one ^^ has someone made a backup and can share it with me.?
Click to expand...
Click to collapse
When you apply the settings you need to give time to the phone to 'absorve' them.
Run a couple of benchmarks, test a 4k video, reboot the phone 3-4 times, close it also.
I havent noticed higher tdmperature of 45 celsius.

kutulu32 said:
When you apply the settings you need to give time to the phone to 'absorve' them.
Run a couple of benchmarks, test a 4k video, reboot the phone 3-4 times, close it also.
I havent noticed higher tdmperature of 45 celsius.
Click to expand...
Click to collapse
I noticed that haha now its fine

kutulu32 said:
Yes indeed i saw a difference in chargin thanks for your effort in the configuration. I am not very fond of benchmarks but i run some in my devices after magor changes (firmware....). This device should have been speedier because of the full hd snapdragon 810 combination. I wonder if we delete the thermal config is the device gonna be speedier? I am waiting for 5.1.1 hoping for smoother rom. Lets hope soneone unlocks the bootloader!!!! I took my precautions disabling MLT lol
Click to expand...
Click to collapse
I'm not sure what would happen if you just deleted the file. Maybe it'll revert to some default, or perhaps it won't thermal throttle at all which would result in auto shut off when it overheats. Probably not worth risking it without a custom recovery in case it doesn't boot.
I have notice that my phone is way quicker now. It was fast before but now it's a beast when I push it hard. Also things like PPSSPP emulator was useless with the default thermal throttling now it can beat my Tegra note 7 with 4 A15 cores at 2.1ghz at it.

Was there a problem with your settings? I noticed your settings you had shared on dropbox are no longer available.

uh60james said:
Was there a problem with your settings? I noticed your settings you had shared on dropbox are no longer available.
Click to expand...
Click to collapse
The folder was deleted for some odd reason. I was able to restore it so the link should work now. Let me know how it works out for you.

A few questions about these settings.
-At the top where you have SS-Little_Cluster_management and SS-Big_Cluster_management is the set point the shutdown temp for the entire cluster?
-I think this one is self explanatory by the action listed but towards the middle where CPUx_MONITOR for each core is, is the threshold the temperature at which the entire phone shuts down?
-I'm not sure I am noticing the difference between SS-CPUx and CPU_HOTPLUG_MONITOR. Could you explain these?

uh60james said:
A few questions about these settings.
-At the top where you have SS-Little_Cluster_management and SS-Big_Cluster_management is the set point the shutdown temp for the entire cluster?
-I think this one is self explanatory by the action listed but towards the middle where CPUx_MONITOR for each core is, is the threshold the temperature at which the entire phone shuts down?
-I'm not sure I am noticing the difference between SS-CPUx and CPU_HOTPLUG_MONITOR. Could you explain these?
Click to expand...
Click to collapse
I don't want to assume I know everything about this. Everything I know about the thermal configuration file I learned from seeing others modify the LG G3 thermal config file. I also learned some by experimenting with my G Flex 2.
In my understanding SS-Little_Cluster_management and SS-Big_Cluster_management is the temperature at which the Cluster will thermal throttle when all cores in the cluster are being used at once. From what I've seen anything with "action_type 10000" means thermal throttle. The closest thing I can think off that works like that is how Intel I5 and I7 will limit turbo speed when more than one core is being used. Except in our case this is to prevent higher speeds when it reaches a certain temperature. So lets say your phone is running some heavy multithreaded app now it's going to follow the rules set in Cluster_management. The default Big Cluster temperature is set to start slowing things down when temperature reaches 48C and to not go back to full speed until the temperature drops down to 38C. One thing I find interesting is that at about 48C cluster temperatures was about where the Big cores would drop to a max of 1.5 ghz. So I'm just guessing that this setting may play a big part on the big cores max speed.
[CPUx_MONITOR] seems to shutdown the phone when the core reaches a certain temperature threshold of 120 C and I assume will not allow it to turn back on until it drops below 115C.
[SS-CPUx] on a big core seems to slow the specific core down when it reaches a temperature threshold of 85C and to not allow it to go to full speed until it drops below 55C. However this rule can be overwritten by [CPUx_HOTPLUG_MONITOR] which can shut down a core even if it hasn't reached it's maximum temperature threshold.
[CPUx_HOTPLUG_MONITOR] just tells the specific core to turn off once it reaches the temperature threshold specified. On the default configuration core 4 and 7 are set to turn off when their temperature goes above 50C. After that core 5 turns off at 58C and even core 6 turns off at 61C. At which point only the small cores will remain on.
I hope that helps somewhat.

S7 SuperKernel V1.8.5 Feedback (new user, can't post in official forum)

Unfortunately I; being a new user can't post currently in the forum (at least until I rack up a few posts); but hopefully the developer Tkkg1994 might see this.
First off, i'm fairly happy with this Kernel; it seems to be reliable and reasonably fast (at least with a few changes!), and does provide decent battery life (3Days, 15 hours w/ 7 hours screen on time is my record so far). I use it mostly to stream video (Netflix, Crunchyroll) and with netflix especially I get around 7.5 minutes -> 8 minutes per percentage point (1/2 brightness);
This is especially encouraging since when I purchased this phone, I had every intention of rooting the phone (and voiding my warranty!), and I knew at that stage about Samsung's ASV chip binning; I had a lot of fun undervolting my previous S3 to extend the battery life; so I was a bit worried about what quality chip I would get as I handed my hard earned money to the salesman.
See I have terrible luck with chip binning quality for some reason; Ever since I found that my newly purchased Celeron 566MHz could not be overclocked to 850MHz despite the numerous successess of others, I tend to attract bad bin quality chips, so much so I yearn to simply get an average chip that happens to be somewhere in the bulk of that lovely bell curve. but alas...:
ASV_SUMMARY:
big:2, LITTLE:1, INT:10, MIF:7, G3D:10, ISP:10
I think that makes me class E; As I understand it, big:2 is the lowest you can get; and LITTLE:1 is also the lowest. I win!; unfortunately, not the lottery ticket, but the lightning bolt from the heavens
Alright now I got that off my chest, There are some positives that help alleviate this most voltage hungry Exynos 8890 chip. For example, I have managed to undervolt reasonably well (Although to be fair my undervolts are probably equivalent to moving to stock voltages for a good percentage of Galaxy S7 owners out there). I shaved 100mV of the A72 (big) chip's 2.6GHz bringing me down to 1.1v and for most of the other speeds I have managed to drop 125mV. I have managed to drop 100mV on the little(A53) chips speeds.
Now on to some useful information (finally..hopefully?).
I am not an expert, so within synapse, I do not understand all options available to me. But I noticed that unlike the stock Samsung Kernel, getting all 4 of the 'big' cores to work wasn't happening. I found that Geekbench 3 and Antutu weren't really using those extra cores. However, changing the option "Exynos Hotplug" to Disabled fixed this; certainly helped my scores and despite the promised higher power consumption of having those 2 additional cores on all the time, I did not see much change to my battery life.
There is a bug I noticed where Synapse keeps trying to change my big cores minimum speed to 312MHz despite being told not to do it for some reason. I'm always having to hit the 'x" option every time I open Synapse to make sure it leaves my minimum speed at the custom 628MHz. I do not know whether this is a bug with the kernel or with synapse however, so it may not be useful.
In addition, I've noticed that when I first installed Synapse, it seemed to set the Big core voltages to values which while seemingly average, seem far too low for my ASV:2 grade ****ty floor scraping quality chip. It seemed to come right when I attempted to reinstall the system again, but it may be worth checking to make sure the Kernel reads the correct values (if that is indeed a kernel problem rather than a synapse problem.)
I've also noticed this with my GPU voltages. I may be looking at the wrong thing, but here is the ASV_GROUP frequencies and voltages for G3D:
G3D LV3 freq : 650000 volt : 1137500, group: 10
G3D LV4 freq : 600000 volt : 1125000, group: 10
G3D LV5 freq : 572000 volt : 1068750, group: 10
G3D LV6 freq : 546000 volt : 1025000, group: 10
G3D LV7 freq : 455000 volt : 981250, group: 10
G3D LV8 freq : 419000 volt : 931250, group: 10
G3D LV9 freq : 338000 volt : 887500, group: 10
G3D LV10 freq : 260000 volt : 850000, group: 10
Yet without me doing any undervolting, my GPU voltages look like this:
650MHz: 706.25mV
600MHz: 687.5mV
546MHz: 668.75mV
419MHz: 618.75mV
338MHz: 600mV
260MHz: 600mV
It looks like the stock synapse gpu settings are hugely undervolting the GPU...now it seems to be perfectly stable, which could mean that despite having poor CPU's at least the GPU is up to the task, though those undervolts seem rather extreme. What could be happening is that the voltages may be displaying these values, but are actually running at the stock settings; may be worth a look.
In addition for anyone wanting to undervolt there CPU; I highly recommend disabling exynos hotplugging.
The reason why is that I seemed to think I had found a stable voltage configuration with the 2 cores hotplugging only to discover massive instability when forcing all four cores on. I suspect this is because the overall ASV is determined by the lowest grade core of the 4, and so I did not hit any problems since the system was rarely using those 2 additional cores, even within stress testing benchmarks.
I certainly had to add more volts to get it stable, but overall the performance is higher and battery life is probably the most I can hope for out of the Grade E setup I have.
Sorry about the story length first post; but my thanks to Tkkg1994 for creating a very stable, configurable kernel for the Samsung S7