Related
Does any dev know if it is possible on Gingerbread to use this "technology" to create "buttersmooth" refresh rates using vsync and videobuffer?
Thank
This is what i know accurate as per my HW knowledge multiplied by driver programing experiences.
Phones' LCD hw as per resolution doesnt support Vertical Sync more than 65Hz. Although LG has used 75Hz to all LG devices. Increasing Vertical Sync only gives high benchmarks by false 2D/3D results. Increasing HW Vertical Sync HZ rather than supported by phone actually worsens performance, needless to mention in longer run it damages LCD controllers. Phone LCD supports max 65HZ refresh rate thats what called VSync(Virtical Sync).
FramBuffer refresh rate is another story. We can setup kernel kill fd signal(SIG KILL) to specific ms to refresh forcefully kill and restart framebuffer artificially but that will gives blinking display. I am already done with FB refresh rate
Sent from my LG-P990 using Tapatalk 2
But spica surely it works in a similar way to PCs? I.e the Vsync is only set to the max possible frame render of the display, e.g. a 60HZ monitor will sync at 60FPS where Vsync will kick in if enabled... you can also do this for any refresh rate like the newer 120hz monitors.
Wouldnt it simply require changing the variables to acknowledge this extra hz in our displays?.
Just curious, probably wrong but it would be nice if it were true as it would be even easier .
Project butter also uses both the CPU and the GPU to render frames. Its switches between the 2 and one can "prepare" while the other one is still being used to render
Sent from my LG-P990 using xda app-developers app
Chris4evernoob said:
Project butter also uses both the CPU and the GPU to render frames. Its switches between the 2 and one can "prepare" while the other one is still being used to render
Sent from my LG-P990 using xda app-developers app
Click to expand...
Click to collapse
Hmm I dont know much about this project.
Wider LCDs for CPU as per the resolution, can support hight refresh rate but our phone LCD cant:/ it works(increasing vsync and it doesnt damage to LCD controllers in short run) but it will cheat benchmark(as softwares detects wrong info and muliplication) but in realtime performance will be worsened. And its not good for LCD controller HW. We will find every phone with 65hz max except LG. As because phone's LCD controller HW can draw only max 65 HZ sync. Increasing it will draw more frames but it wont be used by HW resulting in wasted frames. FrameBuffer refresh is completely different thing unrealated to vsync. If I can find more detailed refrence link about it than I will post the link when i get PC access soon
This project might be using combination of FB and vSync, not only vsync.
Can anybody rdirect me towards link of this project?
Sent from my LG-P990 using Tapatalk 2
I previously had phone Samsung Galaxy Spica((I-5700).
I began Horse Power Kernel Development 1st with these kernel. Reffer this link:
http://forum.samdroid.net/f53/development-kernel-horse-power-4973/
Click on Previous Changelogs History spoiler.
You can see in older changes i have increased Vertical Sync rate to 80Hz. But believe me it gives only massiv e 2d/3d score. But the performance of games were worsenedd incl. Of UI smoothness. I did revert it in newer version.
Yes its just matter of changing values in kernel display driver
I dont anything about butter project, can anubody provide me link of it?
Sent from my LG-P990 using Tapatalk 2
http://www.google.com/search?hl=en&...3.0.0.0.8612.2.0.0.2.2.0.0.0..0.0...0.0...1ac.
Don't know what you are looking for Spica?
And another link with some info.
http://www.androidpolice.com/2012/0...e-responsive-apps-that-anticipate-your-touch/
Thank you for providing links.
Edit: oh sorry w/o reading it I completely misunderstood before. Its different
Sent from my LG-P990 using Tapatalk 2
Okay I read overly not in deep. Its not related to kernel-HW's refresh rate that I was discussing, hence its safe. Okay its related to android framework I guess, not read in deep. Not related with Kernel and kernel needs no changes. Its google's work I guess
It seems good. I will look more deeply tomorrow as soon as I get PC access
Sent from my LG-P990 using Tapatalk 2
IF this would ever be ported to Gingerbread for our phone, it would only be found on CM7 ROMs, not on Stock ROMs.
Android has kinda bad design concerning the scrolling stuff in its framework which causes it to never be as smooth as other Smartphone OS like iOS or WP are or will be.
This base design can't be change easily anymore, in order to NOT break the (available) Android apps.
It's also clear that even though Google can't change the whole base, they will put efforts in creating and finding workarounds for providing a smoother UI anyway.
This and faster hardware will make Android smoother every year.
They have already proven with both ICS and Jelly Bean that they are able to improve the smoothness. But the changes they had to perform for this surely (without knowing further details) affect big parts of the Android Framework.
This means in order to get a smoother scrolling experience on Gingerbread, ROM DEVs would need to backport features and changes of the new OS revisions to GB framework. And I'm sure this is way too much work (for our crappy little phone ).
Also it can only be done if the whole framework source is available, so no smoother scrolling for STOCK. And there is nothing kernel DEVs could do in this case by their own (besides the things they are doing already), unfortunately.
But you guys could ask Temasek to backport smoother scrolling to his CM7 ROM *jokingly - won't be that easy*
Answering to the OP, it's quite impossible... We need to backport lots of changes in GB framework (for example hw acceleration) and to do a lot of work that it doesn't worth it. And the only rom that would support these changes is cm7, and it isn't so stable.
If you like project butter, get a galaxy nexus. We have doubts to get ics update on our 2x...
markop90 said:
Answering to the OP, it's quite impossible... We need to backport lots of changes in GB framework (for example hw acceleration) and to do a lot of work that it doesn't worth it. And the only rom that would support these changes is cm7, and it isn't so stable.
Click to expand...
Click to collapse
Yeah that's exactly what I wrote - just in short But I wouldn't call CM7 unstable.
markop90 said:
Answering to the OP, it's quite impossible... We need to backport lots of changes in GB framework (for example hw acceleration) and to do a lot of work that it doesn't worth it. And the only rom that would support these changes is cm7, and it isn't so stable.
If you like project butter, get a galaxy nexus. We have doubts to get ics update on our 2x...
Click to expand...
Click to collapse
Cm7 is one of the most stable roms on our phone...
Sent from my LG-P990 using xda app-developers app
Stefan Gündhör said:
IF this would ever be ported to Gingerbread for our phone, it would only be found on CM7 ROMs, not on Stock ROMs.
Android has kinda bad design concerning the scrolling stuff in its framework which causes it to never be as smooth as other Smartphone OS like iOS or WP are or will be.
Click to expand...
Click to collapse
Actually scrolling in Android is very well implemented. The reason for the lag is twofold:
1) Android started as keyboard driven OS (a better blackberry), with no touch interface in mind, so the UI thread is of normal priority. That means that touch events don't have a higher priority than any other process that runs on your phone at the same time, thus any animation or transition can be interrupted (in cpu cycles) by another process. Say for example that you are scrolling a long list. Every other process that does something at the same time (updating your widgets, checking your email etc) will get equal time with the cpu as the scrolling animation. That means allot of interrupts during that animation that cause frames to drop and thus lag to occur. It also means that the ui thread must wait for it's time whenever there is a touch event (user interaction through the touchscreen) to respond. Changing the UI thread priority would solve most ui problems with android but it would also indeed break compatibility with older apps.
2) The other problem is that most devs do a very poor implementation of lists. Android lists are actually very efficient but badly documented, and many developers implement them in ways that are sub-optimal. In order to save memory, the list only uses as many views (items) as fit the screen, and when the user scrolls it simply recycles them by changing their content. That means that the objects displayed in the list are cleared from the memory as soon as they leave the screen. So every time you return to the place where those objects where, an interrupt followed by IO with the filesystem takes place (the worst kind of interrupt). That's not such a big problem when the only thing you want to display is text, but what if you use bigger stuff like bitmaps? Then you get allot of lag. To avoid that you have to keep the list objects in memory. If you have too many such objects and want to avoid filling all the heap you can create a cashe with a limited size, or simply load object to the list only after scrolling has stopped (look to the cm9 music player cover art for one such example). Anyway, to make a long story short you can implement very complex and content heavy lists that run like butter on a arm6 600mhz gingerbread phone, but you can also make those same lists lag on OneX with the wrong implementation.
And btw, project butter DOES use the kernel, it makes use of a custom governor that immediately rumps up the cpu freq to the max whenever there is a touch event.
Stefan Gündhör said:
Yeah that's exactly what I wrote - just in short But I wouldn't call CM7 unstable.
Click to expand...
Click to collapse
I didn't say that cm7 is unstable, but it still has some freezes and random reboots... For example, yesterday I flashed cm7 again, after 2h of runtime I opened the browser and the phone froze and restarted himself... It is sufficiently stable, but I cannot use it as daily rom... And it is also battery hungry...
Ricardo has done a really great job on this phone, but cm7 isn't a rom for my business.. I prefer max stability, so I had to flash a 2.2 rom... GB memory leak is very annoying..
Supupa said:
Cm7 is one of the most stable roms on our phone...
Sent from my LG-P990 using xda app-developers app
Click to expand...
Click to collapse
"one of the most stable"? Maybe you just haven't tried a good stock based rom... I used cm7 for 7 months and i had a lot of problem...that's because of the closed Tegra2 sources...
markop90 said:
"one of the most stable"? Maybe you just haven't tried a good stock based rom... I used cm7 for 7 months and i had a lot of problem...that's because of the closed Tegra2 sources...
Click to expand...
Click to collapse
What do the closed sources have to do with cm7? It's GB as well and they are fully compatible.
I used CM7 for over half a year, never had any real problems.
ICS and GB are very diferent in the way they interact with the hardware.
Butter smoth stuf on GB is not possible.
This is what Ricardo Cerqueira answered about this:
"Project Butter is mostly about synchronizing gfx composition with the display's actual refresh rate... and Gingerbread didn't have a hardware compositor at all."
If you have undervolted your Nexus 7 or any other tegra 3 based device please post the results here so I can throw them into a Excel sheet for development.
Using advanced mathematics and complex formulas I will make a undervolted kernel that's stable for everyone. For that, I need ideally 30 results or more. Less is ok but more or equal is better.
FORMAT 1:
CPU
- [FREQ1] [VOLTAGE1]
- [FREQ2] [VOLTAGE2]
- [FREQ3] [VOLTAGE3]
- ,,,
Click to expand...
Click to collapse
FORMAT 2:
CPU GLOBAL: [OFFSET]
Click to expand...
Click to collapse
I also accept 1 frequency result. You don't have to undervolt everything. I favor the last known voltage that crashes as well if you can send this I am even more happy.
Undervolted mine by 50mv
Sent from my Nexus 7 using Tapatalk 2
djjonastybe said:
If you have undervolted your Nexus 7 or any other tegra 3 based device please post the results here so I can throw them into a Excel sheet for development.
Using advanced mathematics and complex formulas I will make a undervolted kernel that's stable for everyone. For that, I need ideally 30 results or more. Less is ok but more or equal is better.
Click to expand...
Click to collapse
In order to use "advanced mathematics" and "complex formulas".. You might want to provide a format for "the results" you want people to provide.. Without a stream of formatted data to input your "complex formulas" and "advanced mathematics" is impossible..
"Advanced mathematics" and "complex formulas" sounds like you are doing an infomercial.
Pure waffle.
Sent from my GT-I9300 using xda premium
djjonastybe said:
..I will make a undervolted kernel that's stable for everyone
Click to expand...
Click to collapse
ha! good luck. considering some devices arent stable at all when undervolted just a tiny bit, and otbers are stable with a lot of undervolting, you have a tough road ahead of you.
simms22 said:
ha! good luck. considering some devices arent stable at all when undervolted just a tiny bit, and otbers are stable with a lot of undervolting, you have a tough road ahead of you.
Click to expand...
Click to collapse
I actually came back to post exactly this but Simms22 was too quick, as usual!
Sent from my GT-I9300 using xda premium
nodstuff said:
"Advanced mathematics" and "complex formulas" sounds like you are doing an infomercial.
Pure waffle.
Sent from my GT-I9300 using xda premium
Click to expand...
Click to collapse
I see you went with the more direct and honest approach. I tried being more subtle with my response.. :laugh: But yeah.. Marketing hype for a kernel.
styckx said:
I see you went with the more direct and honest approach. I tried being more subtle with my response.. :laugh: But yeah.. Marketing hype for a kernel.
Click to expand...
Click to collapse
Might as well be honest, set him on the right track early in his dev career rather than him rely on buzzwords and fud like some people do.
These are forums for android devices, not Apple devices, the reality distortion field doesn't work here, fancy marketing terms should be left at the door.
Sent from my GT-I9300 using xda premium
My Galaxy Nexus is proof of undervolting being unstable using software while being stable with manually programming the voltages. So I am going to give this a good shot anyway.
My Galaxy Nexus uses SmartReflex to undervolt.
Look at this table I made:
https://docs.google.com/spreadsheet/ccc?key=0AuqacKmGLKJ6dEdJZG5QTmNtQnoyYk0zbm5IanluMHc#gid=1
You see the core or gpu was already crashing at 974mV, which was the reason why I stopped undervolting using the thread. But don't give up, I tried again using SmartReflex because I felt I should be able to get lower. And now I am at 911mV but I did not try to get lower yet.
I plan on undervolting by actually programming it. I don't know how you guys undervolt for the Nexus 7, since I am fairly new with this device.
READ FIRST POST for formatting
nodstuff said:
"Advanced mathematics" and "complex formulas" sounds like you are doing an infomercial.
Pure waffle.
Sent from my GT-I9300 using xda premium
Click to expand...
Click to collapse
http://en.wikipedia.org/wiki/Student's_t-distribution
Using that I can calculate which voltage to take that will work for 90% of us or 99% of us or maybe 100% ?
For that I need some results. The more the better.
The only voltage settings that will work for 100% of people are stock voltages.
you will run into endless problems if you release an undervolted kernel because what works for 75% of people won't work for the rest.
If you adjust voltages to include more people the people that can handle the lower voltages aren't getting the full savings their device can handle. Which defeats the point of releasing an undervolted kernel to save battery.
Undervolting should be up to the individual, just release with stock voltages and make a guide showing people how to undervolt.
Sent from my Nexus 7 using xda premium
nodstuff said:
The only voltage settings that will work for 100% of people are stock voltages.
you will run into endless problems if you release an undervolted kernel because what works for 75% of people won't work for the rest.
If you adjust voltages to include more people the people that can handle the lower voltages aren't getting the full savings their device can handle. Which defeats the point of releasing an undervolted kernel to save battery.
Undervolting should be up to the individual, just release with stock voltages and make a guide showing people how to undervolt.
Sent from my Nexus 7 using xda premium
Click to expand...
Click to collapse
^^ Truth
I've done a bit of pc overclocking over the years and we'd always seek out the best steppings to find cpu batches that would give a better clock speed for less volts. What would work on one person's cpu wouldn't work on another sometimes even within the same stepping. (My current machine is getting long in the tooth but it's running a 33% OC below stock volts and stable as a rock.)
When the N7 cpu's are tested the stock voltage is the point where 100% of them will run without issue. Certainly there are many, maybe even most N7's that will run undervolted but until you try it you just won't know. Undervolting needs to be done on a case by case basis.
I uv by 100mv on every step on Franco kernel and I've never had a sod or reboot. Maybe you could release different versions like light, medium and heavy to cater to different users that have sets than can uv more or less. To address all these people naysaying I say just go for it, don't let these people dishearten you from what you want to do. I'm sure there will always be users who are willing to try
sorry to break the fun lol
Opened a thread with undervolting settings in different situation. For now i think its the lowest possible
bervin said:
I uv by 100mv on every step on Franco kernel and I've never had a sod or reboot. Maybe you could release different versions like light, medium and heavy to cater to different users that have sets than can uv more or less. To address all these people naysaying I say just go for it, don't let these people dishearten you from what you want to do. I'm sure there will always be users who are willing to try
Click to expand...
Click to collapse
That's exactly the same as what I have done. 100mV right across the board, and I have also never had a reboot or any other negative effect from doing it. Just a bit more battery life.
I too UV 100mv across the board. Never had an issue. And I OC up to 1.6ghz.
phonic said:
I too UV 100mv across the board. Never had an issue. And I OC up to 1.6ghz.
Click to expand...
Click to collapse
Same here, no issues with 100mv UV on stock speeds. What would happen if we went too low, would it fail to boot at all or is there some safety mechanism that would allow us to boot into boot loader?
davidoff59 said:
Same here, no issues with 100mv UV on stock speeds. What would happen if we went too low, would it fail to boot at all or is there some safety mechanism that would allow us to boot into boot loader?
Click to expand...
Click to collapse
When either OCing or UVing, you should ALWAYS test out the changes prior to saving them as a persistent boot state. This way, if you set something too high/low and it proves to be unstable, if you restart the device it will revert back to normal.
In the event that you OC/UV too much, yes, you would see signs of instability. In some cases it could show up as soon as you applied to change, in other cases it could take a while to notice. For example, some games or other high processing powered activities could FC/crash. Or your tablet might freeze or reboot. But again, as long as you didn't save your changes as a boot setting, it will be fine once it reboots.
In the event that you did save it, and the settings were unstable for your device, you could boot up into recovery and then manually erase them from taking effect. Recovery doesn't use those settings, as it's technically a different OS. But that can be complicated a process for some. So test test test save. Or don't save and just apply them manually.
Two other points:
Yes, most (all?) devices do have safeguards to protect itself against permanent hardware damage from OC/UVing. If it gets too hot, they will often shut down to protect themselves.
And just because one device can handle being OC'd or UV'd to a certain level doesn't mean they all can. Very minor differences in each component can impact one devices ability from another. The manufacturers only test to see if they can handle the published speeds/etc., but in many cases they can go higher if set to. In others, not so much.
Thanks for that phonic. Given the range of ROMs and kernel's available and the improvements already built in, under volting may not help a lot but every bit helps. Eg this screen on time is pretty good.Close to six hours screen on time with iirc 35% battery remaining.
Paranoid android with motley kernel. I don't over clock the nexus 7 as its fast enough. In fact I under clocked it to 1000 MHz max before and it was still fast but the screenshots are at stock volts and speeds.
it's actually plain simple. If everyone posts results. I can make a kernel with voltages that work for 75% of all people or maybe for 95% of all Nexus 7 owners.
It's called mathematics. But to do that I need as much results as possible.
Easy to do if you know what you are doing, right ?
http://en.wikipedia.org/wiki/Confidence_interval
Anyone used this app? Some users reported to have significant lag reduction when starting apps but in my case my O2X was already snappy so the difference is not night and day with this app running but it might have improved a little. I am still on stock GB (Spica's V20q rebuild) + Gueste Kernel 2.2.1 HOC SC
You can refer to the original thread for more information and get the APK : http://forum.xda-developers.com/showthread.php?t=1987032&nocache=1
Please share your results here! (mention your ROM / Kernel)...
I tried it and it should be working, but I don't seem to see any dramatic improvement.
Sure, there is a bit of improvement but it might be subtle or just placebo.
I'm running the CM10 on new bootloader btw.
arcee said its a placebo in his gplus
lopiop said:
arcee said its a placebo in his gplus
Click to expand...
Click to collapse
Well, that explains why I still lag out in Beach Buggy Blitz, and why that's no noticeable difference on my phone.
Ricardo Cerqueira said that it doesn't work, look at his post in G+: https://plus.google.com/115049428938715274412/posts/GWr72W9zmY2
Saw RC's comment about it before even knowing it existed, still tried it and might have experienced a very small difference which I still don't kown if it's just placebo or some part of the script which somehow speed things up a little.
This following discussion is pretty interesting though:
https://code.google.com/p/android/issues/detail?id=42265#c114
I tried and I can see some difference, opening times are different, not instant opening, but better. Placebo? Placebo effect is documented to function... Anyway, still discussion goes ahead, and something new still comes out.
gramaglia said:
I tried and I can see some difference, opening times are different, not instant opening, but better. Placebo? Placebo effect is documented to function... Anyway, still discussion goes ahead, and something new still comes out.
Click to expand...
Click to collapse
arcee said it has something to do with maintaining higher cpu frequencies, thus making apps launch quicker
It works great on my one. One word that is satisfied.
Sent from my LG-P500 using Tapatalk 2
lopiop said:
arcee said it has something to do with maintaining higher cpu frequencies, thus making apps launch quicker
Click to expand...
Click to collapse
i was thinking about it, so i changed minimum frequency to 1000mhz instead of 216 or 312 mhz. It's definitely snappier when i unlock the phone and start using it. but switching between apps, is still faster w/ seeder on.
What this app is gonna do?
This will give you some boost that is really required to play games that are demanding on GPU front.
What this app require?
You need to be ROOTed and for that there is a really great app with one touch facility available! The link is http://forum.xda-developers.com/showthread.php?t=2387507
After you get ROOTed just install the Tegra Overclock OUYA EDITION app from attachment and click on OK for one click optimization.
You can also change the actual resolution to somewhat lower to again get a massive boost in GPU intensive games or shall I say those games that scale with GPU clocks very well. (I can confirm nearly 2X boost in RipTideGP2 with all settings maxed out when comparing with 1080p to 720p option provided in the app :laugh
Any kind of hazardous warning before use?
If anything goes wrong then you know where the power switch is! Close the console and cut the power and start again. We have not implemented the on Boot settings for safety reasons...
What about future updates?
I just wish that 4.2.2 gets for the OUYA and that will be a game changer for the console as that support full screen scaling on resolution changing vs screen compensation provided by current android 4.1 version....
NOTE: The app is fully free to use one click solution for OUYA owners and don't require you to get involve in any of cracking stuff Still the app does costs us in development and support front so any kind of donation will be greatly appreciated. If you are planning to buy any Tegra device in future then you can support us by purchasing the original Tegra Overclock app from the market that will in turn benefit you as well..
BTW don't forget to hit the thanks button
EDIT: Version 1.1 Change log
*Nothing new just made the interactive governor as a default rather than performance governor due to heating concern.
*Checked the code so that 2D GPU is set to 520 MHz too, in case it still reports half let me know..
EDIT: Version 1.2 Change Log
***Added requested LBCMT (LastBootCpuMaximumTemperature!) This shows the maximum temperature your device went during last boot.
***Added apply on boot (Works only when the USB device with at least a single file is detected to ensure you don't intentionally mess up anywhere )
The LBCMT will works as soon as you boot once after you open the app for the first time. Also note that YOU SHOULD NOT KILL THE APP manually by task killers or the reporting will be fake. BEWARE....
EDIT: Version 1.3 Change LOG
***Full screen stretched resolution support for CM11 Rom users. Use the special newly added resolution (will only work for CM users currently as no other 4.3 or 4.4 Roms available)
Fixes for temp not showing on custom kernel like EPRJ Kernel.
Thank you for being patient.
Kingshoo said:
What this app is gonna do?
This will give you some boost that is really required to play games that are demanding on GPU front.
What this app require?
You need to be ROOTed and for that there is a really great app with one touch facility available! The link is http://forum.xda-developers.com/showthread.php?t=2387507
After you get ROOTed just install the Tegra Overclock OUYA EDITION app from attachment and click on OK for one click optimization.
You can also change the actual resolution to somewhat lower to again get a massive boost in GPU intensive games or shall I say those games that scale with GPU clocks very well. (I can confirm nearly 2X boost in RipTideGP2 with all settings maxed out when comparing with 1080p to 720p option provided in the app :laugh
Any kind of hazardous warning before use?
If anything goes wrong then you know where the power switch is! Close the console and cut the power and start again. We have not implemented the on Boot settings for safety reasons...
What about future updates?
I just wish that 4.2.2 gets for the OUYA and that will be a game changer for the console as that support full screen scaling on resolution changing vs screen compensation provided by current android 4.1 version....
NOTE: The app is fully free to use one click solution for OUYA owners and don't require you to get involve in any of cracking stuff Still the app does costs us in development and support front so any kind of donation will be greatly appreciated. If you are planning to buy any Tegra device in future then you can support us by purchasing the original Tegra Overclock app from the market that will in turn benefit you as well..
BTW don't forget to hit the thanks button
Click to expand...
Click to collapse
now a simple question....can you tell me which options i need to use for get a best performance in ouya games and emu?
This app has a one click solution (found in the op). All you need to do is click the button and your ouya is optimized. The op already has this information.
On another note, this app rocks. Thank you.
Awesome, finally the OUYA version, im about to jump on this right now and give it a spin!!!:good:
darkknight1812 said:
This app has a one click solution (found in the op). All you need to do is click the button and your ouya is optimized. The op already has this information.
On another note, this app rocks. Thank you.
Click to expand...
Click to collapse
i asked before install...now i saw it very nice and user friendly
.. one click optimization .. nice job .. cheers
sasuke1981 said:
now a simple question....can you tell me which options i need to use for get a best performance in ouya games and emu?
Click to expand...
Click to collapse
Just Press the OK button and wait a few seconds as the settings getting applied. After that Run your Emulator Games and you should notice a difference . Note that Resolution options also cuts the screen area so use them only if a game is unplayable!
is there no way to allow users to set advanced profiles so that they can configure a medium and max overclock setting?
i saw the temps on my core rise up pretty quickly just from fiddling with the UI, not sure if I want to subject this little guy to that much abuse with no really efficient cooling solution under the hood.
Careless_ said:
is there no way to allow users to set advanced profiles so that they can configure a medium and max overclock setting?
i saw the temps on my core rise up pretty quickly just from fiddling with the UI, not sure if I want to subject this little guy to that much abuse with no really efficient cooling solution under the hood.
Click to expand...
Click to collapse
it's understandable for the lil guy to get hot quickly when you are overclocking the GPU+CPU, can't expected to run cool unless you put a better cooling fan into you console, i don't mid my ouya running a lil hot, well mines don't get hot enough cause my ouya got a acrylic custom shell with enough vents and airflow that keeps my console way cooler then the ouya stock/factory shell...
its obvious that it would run hotter if the clock speed is higher. no one is disputing that. i just don't see how not allowing the user to specify speed as much of the other CPU overclocking programs like SetCPU allow the user to do.
It's a needed feature IF it can be supported and implemented.
Excellent!
I'll give this a bash tonight and see what improvements there are!
Careless_ said:
is there no way to allow users to set advanced profiles so that they can configure a medium and max overclock setting?
i saw the temps on my core rise up pretty quickly just from fiddling with the UI, not sure if I want to subject this little guy to that much abuse with no really efficient cooling solution under the hood.
Click to expand...
Click to collapse
Actually Temperatures are no problem for this chip as it can bear temperatures of 70 C with great ease as the manufacturer mentioned. Don't worry about that also let me know if they are above this temperature so that we can do something about that. If you ask me I am comfortable with 85C as that is what the throttle temperatures (Default mind you!). Just Relax and crush the games already
Careless_ said:
its obvious that it would run hotter if the clock speed is higher. no one is disputing that. i just don't see how not allowing the user to specify speed as much of the other CPU overclocking programs like SetCPU allow the user to do.
It's a needed feature IF it can be supported and implemented.
Click to expand...
Click to collapse
The custom CPU clock feature conflict the one click feature and not to mention lower clocks are not required as there is no trouble with the highest speed since the cooling is active and with sink. Just don't worry about the temps too much as everything is normal here and the purpose of console is to bear high temperatures with high load even with little heat management..
I have both the old app and the Ouya specific one and it looks to me that 2D GPU clock is set to 260MHz (1/2 of 3D clock). @Kingshoo, can you tell us what the "one click" settings are (governor, CPU clock, GPU, etc).
Much appreciate the work you've done.
Ipse_Tase said:
I have both the old app and the Ouya specific one and it looks to me that 2D GPU clock is set to 260MHz (1/2 of 3D clock). Can you tell us what the "one click" settings are (governor, CPU clock, GPU, etc).
Much appreciate the work you've done.
Click to expand...
Click to collapse
Oh really If its working on Half then I will bump that to full to in next iteration
Settings are nothing but final settings in one click. Meaning Performance governor (Because this one has no polling overhead as opposed to interactive). GPU to 520Mhz, CPU clock to 1.7Ghz (1.6 on all 4 cores as set by OUYA). Ram at 800Mhz and also no throttling on Ram and GPU so that performance is reduced only due to H/W limitation or by bad coding practices and not due to some power saving logics
Kingshoo said:
Oh really If its working on Half then I will bump that to full to in next iteration
Settings are nothing but final settings in one click. Meaning Performance governor (Because this one has no polling overhead as opposed to interactive). GPU to 520Mhz, CPU clock to 1.7Ghz (1.6 on all 4 cores as set by OUYA). Ram at 800Mhz and also no throttling on Ram and GPU so that performance is reduced only due to H/W limitation or by bad coding practices and not due to some power saving logics
Click to expand...
Click to collapse
Cool...thanks! Looking forward to v2
BTW...for reference, I had my governor set to performance for a while and it drove the fan to run almost at all times. To the point I ended up shelling 20 bux to buy the Noctua quiet fan. Maybe others have better luck with the stock fan - but if noise is a concern, I found that the "Interactive" governor scales pretty fast with no visible impact on performance vs "Performance".
My 2 cents.
After reading several reviews about Noctua 40x10mm A-Series Blades with AAO Frame, SSO2 Bearing Premium Retail Cooling Fan NF-A4x10, I decided to buy it. I replaced the noisy fan of my OUYA, and the difference in sound is amazing. I had to check if the fan was actually running because barely hear it. Also, the temperature of my equipment dropped ~ 10°C, so overheating issues could be easily fixed, if you decided to overclock. I recommend this product without a doubt. Check on eBay or Amazon.
If you want to replace the fan, here's a guide how to do it. youtube . com /watch?v=rra0FwlHvBA
Well honestly I wasn't able to play games on OUYA but since you people are playing games what the temps of CPU are anyways after say an hour? I am asking this because I don't see a need for custom coolers as 50C or 70C it really don't make any difference to your hardware. Yes the fan noise may be concerning to many but I have cotton plugs in my ears to keep it silent (these really costs cents and not $$!). OK so newer version will have interactive on one click..
My stock fan is totally silent anyway
Sent from my ADR6350 using Tapatalk
Ipse_Tase said:
Cool...thanks! Looking forward to v2
BTW...for reference, I had my governor set to performance for a while and it drove the fan to run almost at all times. To the point I ended up shelling 20 bux to buy the Noctua quiet fan. Maybe others have better luck with the stock fan - but if noise is a concern, I found that the "Interactive" governor scales pretty fast with no visible impact on performance vs "Performance".
My 2 cents.
Click to expand...
Click to collapse
i will also try this, i run my OUYA on performance all the time and it's pretty loud...
Greetings,
This driver is obtained from Qualcomm's 07 Aug 14 drivers for "Qualcomm Adreno 3xx GPU on Nexus 4, 5 & 7 devices running Google Android 4.4.4 KitKat".
All credits go to the folks at Qualcomm for making the developer version of these lib files available for us to play with.
This is for testing purposes only. Please don't flash without backing up first and then start screaming if something happens to your phone!
The package only contains files for Adreno 300 and 330. I cooked the packaged to be flashable on our SM-G900T or any SM-G900 with the Adreno 330 chipset.
I'm on a "XtreStoLite_G900T_ROM_v1.3_UVU1BNG3" ROM, which is a super clean/slim version of the stock rom. This is also Android 4.4.2, my AnTuTu benchmark results dropped from 37400 to 35800 after flashing this package.
I'm requesting someone with a KitKat 4.4.4 ROM to test this and report back their before and after benchmarks.
I appreciate your feedback,
Hmm, interesting.. I'm only on 4.4.2 or I would help out..
Turned out the 3D performance on antutu version 5 with this driver puts our phone one head and shoulder above everyone else!
Sent from my SM-G900T using XDA Free mobile app
Commodore 64 said:
Turned out the 3D performance on antutu version 5 with this driver puts our phone one head and shoulder above everyone else!
Sent from my SM-G900T using XDA Free mobile app
Click to expand...
Click to collapse
And this means....... I'm sorry but I don't understand the purpose... Is this supposed to make Tw snappier?
Thanks in advance!
It is updating the video (graphics) drivers. It probably improves the touchWiz performance as well (don't have evidence to support this), but what i noticed was a huge performace increase in the 3D graphics score. This will make games run alot smooter and more efficiently.
To clarify, this file resembles how your Mac/PC has a video card with say NViDIA or ATi video cards and they have their own driver releases. Our phones graphic chipset is designed by Qualcomm and it gets its own drivers as well. Hope I answered your question, and didn't make it worse!
Sent from my SM-G900T using XDA Free mobile app
In order to get any real "TouchWiz/Nova/Apex/etc" performance increase from it you'd have to flash this, then venture into Developer's Options and enable Force GPU rendering and Disable Hardware Overlays. And even then our device is so performance driven that you likely would likely not even realize the difference other than giving the CPU a break for a few cycles and possibly seeing a battery life increase depending on whether the CPU or GPU pulls more juice. These drivers are largely made to increase the 3D performance of your device when gaming and while it wasn't meant for us I'm glad to see someone took the time to let us benefit from it.
I'm going to flash it on an NG4 (4.4.2) release and see how it holds up to GPU rendering no hardware overlays. I don't game so if you want an objective test on that someone will have to pick up my slack. I'll report back when I know if it truly impacts it.
Tried flashing the update and ran into a couple random fcs for a few apps. Nothing device-breaking or anything, but I also didn't notice any real difference in performance or framerate in games. But I do appreciate you posting this, I'm always up for trying a little experimentation on my device to find something new. Thanks man.
AlkaliV2 said:
In order to get any real "TouchWiz/Nova/Apex/etc" performance increase from it you'd have to flash this, then venture into Developer's Options and enable Force GPU rendering and Disable Hardware Overlays. And even then our device is so performance driven that you likely would likely not even realize the difference other than giving the CPU a break for a few cycles and possibly seeing a battery life increase depending on whether the CPU or GPU pulls more juice. These drivers are largely made to increase the 3D performance of your device when gaming and while it wasn't meant for us I'm glad to see someone took the time to let us benefit from it.
I'm going to flash it on an NG4 (4.4.2) release and see how it holds up to GPU rendering no hardware overlays. I don't game so if you want an objective test on that someone will have to pick up my slack. I'll report back when I know if it truly impacts it.
Click to expand...
Click to collapse
Anything?
Thanks in advance!
Slicktune said:
Anything?
Thanks in advance!
Click to expand...
Click to collapse
Don't worry, I didn't forget about you. It takes a few days to really put it through its paces. But here is what you want to know: It works in 4.4.2 Touchwiz (Alliance Rom B4 for me)
Performance Increase in ROM function and control with "Turn Off Hardware Overlays" and Force GPU Rendering: Check
So I tested this by underclocking my CPU to the point where lag was perceivable in things likes app launching, keyboard typing, app draw opening, etc. I dropped it down to about 1200Mhz on all cores and then disabled HW Overlays. It was night and day. After the GPU took over rendering my CPU clock speed hardly moved all of the items previously lagging went right back to being fluid.
So does it help with Rendering in the ROMs? Yes, it definitely can.
The other concern people have is with battery life that goes with GPU rendering and the claims are that the GPU uses more power than the GPU. In my preliminary testing, what I noticed was lower overall CPU temps while Hardware Overlays was off, lower CPU clock speeds since it didn't need to render apps, and a small decrease in battery life.
In the power draw category I need more time with BetterBatteryStats and I need to adjust some variables. I want to underclock the GPU to the point of perceivable lag (if that is possible) and then I want to adjust the governor with KTweaker and see if I can't get better battery life. I will simultaneously lower CPU speeds to keep my core temperatures down and keep the battery cool. If these things are successful and battery life is interchangeable between the two, I may run the GPU full time for rendering the ROM. Keep in mind though, only custom kernels allow this kind of tweaking. If you use a stock ROM you are stuck with CPU and GPU clock speeds where they are so battery will drain faster.
Give me about a week to test my theory on this and see if I can find an equilibrium for CPU and GPU temperature and speed. I'll post the profile I used with KTweaker when I have my results. Until then, if you want to help and find your own settings and post them here along with battery life that would be awesome.
Edit: For what it is worth, to test application launch speeds and UI fluidity with the GPU I completely turned off system animations, I set Nova Launcher Animations Speed to Faster than light and scroll speed to "Fast" (personal preference). I got the same, or better response time using the GPU to render the UI.
AlkaliV2 said:
Don't worry, I didn't forget about you. It takes a few days to really put it through its paces. But here is what you want to know: It works in 4.4.2 Touchwiz (Alliance Rom B4 for me)
Performance Increase in ROM function and control with "Turn Off Hardware Overlays" and Force GPU Rendering: Check
So I tested this by underclocking my CPU to the point where lag was perceivable in things likes app launching, keyboard typing, app draw opening, etc. I dropped it down to about 1200Mhz on all cores and then disabled HW Overlays. It was night and day. After the GPU took over rendering my CPU clock speed hardly moved all of the items previously lagging went right back to being fluid.
So does it help with Rendering in the ROMs? Yes, it definitely can.
The other concern people have is with battery life that goes with GPU rendering and the claims are that the GPU uses more power than the GPU. In my preliminary testing, what I noticed was lower overall CPU temps while Hardware Overlays was off, lower CPU clock speeds since it didn't need to render apps, and a small decrease in battery life.
In the power draw category I need more time with BetterBatteryStats and I need to adjust some variables. I want to underclock the GPU to the point of perceivable lag (if that is possible) and then I want to adjust the governor with KTweaker and see if I can't get better battery life. I will simultaneously lower CPU speeds to keep my core temperatures down and keep the battery cool. If these things are successful and battery life is interchangeable between the two, I may run the GPU full time for rendering the ROM. Keep in mind though, only custom kernels allow this kind of tweaking. If you use a stock ROM you are stuck with CPU and GPU clock speeds where they are so battery will drain faster.
Give me about a week to test my theory on this and see if I can find an equilibrium for CPU and GPU temperature and speed. I'll post the profile I used with KTweaker when I have my results. Until then, if you want to help and find your own settings and post them here along with battery life that would be awesome.
Edit: For what it is worth, to test application launch speeds and UI fluidity with the GPU I completely turned off system animations, I set Nova Launcher Animations Speed to Faster than light and scroll speed to "Fast" (personal preference). I got the same, or better response time using the GPU to render the UI.
Click to expand...
Click to collapse
Thanks for the detailed response, can I flash this over a cm rom or only "touch wiz" and/or after I updated/flash this no Mauer what rom I flash will it stay. ..?
Thanks in advance! I'm still a bit confused...
Slicktune said:
Thanks for the detailed response, can I flash this over a cm rom or only "touch wiz" and/or after I updated/flash this no Mauer what rom I flash will it stay. ..?
Thanks in advance! I'm still a bit confused...
Click to expand...
Click to collapse
I haven't tried it on CM yet, I'm about to flash the 09/04 build of CM11 by Albinoman with KT's kernel. But I won't flash these drivers prior to running the build 24 hours or so to make sure there aren't bugs that existed before flashing the files. So, if you've been running CM for awhile and want to test this build just nandroid, flash, and report any issues.
AlkaliV2 said:
I haven't tried it on CM yet, I'm about to flash the 09/04 build of CM11 by Albinoman with KT's kernel. But I won't flash these drivers prior to running the build 24 hours or so to make sure there aren't bugs that existed before flashing the files. So, if you've been running CM for awhile and want to test this build just nandroid, flash, and report any issues.
Click to expand...
Click to collapse
Ooo that sounds like a killer combo cm + kt kernel + adreno update = project slip n slide lol nah but I'm really curious of what results you might get [emoji4]
And once I flash this, can I go back or will it erase the update once I flash a different rom?
Thanks in advance!!
Slicktune said:
Ooo that sounds like a killer combo cm + kt kernel + adreno update = project slip n slide lol nah but I'm really curious of what results you might get [emoji4]
Click to expand...
Click to collapse
I'll keep you posted. I wasn't impressed overall with it's performance in Touchwiz on battery life with the CPU underclocked, so I am branching out to a more AOSP-like build. I'll keep you posted to what I find out.
Commodore 64 said:
Turned out the 3D performance on antutu version 5 with this driver puts our phone one head and shoulder above everyone else!
Sent from my SM-G900T using XDA Free mobile app
Click to expand...
Click to collapse
This can't be right because right now I haven't flashed the driver update and look at My score
Now imagine that's without kt kernel just stock kernel, I'm on xtralite 3.5a t-mobile, so what if I flashed the kt kernel AAAAND the adreno update.... would my phone explode?!?!
I'm thinking twice now whether I should flash thisor not. ..
You have a good point. I think antutu v5 also rescaled their scoring system. Anyway, post your new score if you decided you want to give it a shot.
Sent from my SM-G900T using XDA Free mobile app
This maybe off topic, but does this works on other devices like Z2 or HTC M8 using the same GPU?
43718 before
43292 after
All I did was reboot into recovery, flash the ZIP and then run the test, same clock speed etc..
92drls said:
43718 before
43292 after
All I did was reboot into recovery, flash the ZIP and then run the test, same clock speed etc..
Click to expand...
Click to collapse
Lower score, what about gaming?
Slicktune said:
This can't be right because right now I haven't flashed the driver update and look at My score
Now imagine that's without kt kernel just stock kernel, I'm on xtralite 3.5a t-mobile, so what if I flashed the kt kernel AAAAND the adreno update.... would my phone explode?!?!
I'm thinking twice now whether I should flash thisor not. ..
Click to expand...
Click to collapse
Commodore 64 said:
You have a good point. I think antutu v5 also rescaled their scoring system. Anyway, post your new score if you decided you want to give it a shot.
Sent from my SM-G900T using XDA Free mobile app
Click to expand...
Click to collapse
92drls said:
43718 before
43292 after
All I did was reboot into recovery, flash the ZIP and then run the test, same clock speed etc..
Click to expand...
Click to collapse
HD-man said:
Lower score, what about gaming?
Click to expand...
Click to collapse
You guys make me sad. Benchmarks are not now, nor will they ever be, an acceptable means for telling you how capable your device is. Your scores in AnTuTu are about as capable of telling you your phone's speed as your ESN is at telling people who your phone belongs to. It is just arbitrary numbers that will fluctuate up and down based on the slightest inconsistencies.
Drivers, just like in Windows/Linux/Mac, are meant to optimize the performance of your hardware. That's it. This driver likely optimizes the GPU cycles to make it more efficient and lower heat output while performing better in games. You know why hardware review sites start out with benchmarks saying what it can do in a perfect world but still ends the review by telling you it gets 75 FPS in Borderlands 2? They do that because while synthetic tests are nice, what would be the point if our games don't play better? It is the same concept here.
I'm begging you, uninstall AnTuTu, re-install your favorite mobile game and just base performance off of whether it plays better or not.
Side Note: Drivers didn't appear to do anything in AOSP CM11 when I gave it a try. Sorry I am so late on reporting in, college takes up a lot of my time. Also, using the GPU to draw your applications does eat more battery whether you use these drivers or not. So there is no marked improvement from lowering CPU speed and using GPU only to render.
92drls said:
43718 before
43292 after
All I did was reboot into recovery, flash the ZIP and then run the test, same clock speed etc..
Click to expand...
Click to collapse
AlkaliV2 said:
You guys make me sad. Benchmarks are not now, nor will they ever be, an acceptable means for telling you how capable your device is. Your scores in AnTuTu are about as capable of telling you your phone's speed as your ESN is at telling people who your phone belongs to. It is just arbitrary numbers that will fluctuate up and down based on the slightest inconsistencies.
Drivers, just like in Windows/Linux/Mac, are meant to optimize the performance of your hardware. That's it. This driver likely optimizes the GPU cycles to make it more efficient and lower heat output while performing better in games. You know why hardware review sites start out with benchmarks saying what it can do in a perfect world but still ends the review by telling you it gets 75 FPS in Borderlands 2? They do that because while synthetic tests are nice, what would be the point if our games don't play better? It is the same concept here.
I'm begging you, uninstall AnTuTu, re-install your favorite mobile game and just base performance off of whether it plays better or not.
Side Note: Drivers didn't appear to do anything in AOSP CM11 when I gave it a try. Sorry I am so late on reporting in, college takes up a lot of my time. Also, using the GPU to draw your applications does eat more battery whether you use these drivers or not. So there is no marked improvement from lowering CPU speed and using GPU only to render.
Click to expand...
Click to collapse
I figured since the OP stated he got less points in Antutu, I would go ahead and use the same app and test to compare...
F*ck me right?
Regardless, I don't notice anything different either, my stats are in the sig.