http://en.wikipedia.org/wiki/Tegra#Tegra_3
Wasjust reading this and it says the CPU goes to 1.3Ghz in single core mode. does that mean 4 cores run at 1.2 then it shuts 3 down and bumps one of them to 1.3 if it needs it?
also anyone now the memery the gfx uses
4 cores @ 1.2Ghz = 4.8 G-ops/sec
1 core @ 1.3Ghz = 1.3 G-ops/sec.
So, the answer to your question is yes and no. Under an idle condition (not sleep) where there is little processor load or runnable threads, there will be only one CPU running @ 1.3 Ghz Fmax. (It can of course be running at a slower speed than this based on what the rate governor has decided) If the demand changes so that one or more additional cores need to be brought on line, the Fmax is constrained to 1.2 Ghz, and those additional cores are brought on line as needed. It may or may not have actually been clocking at 1.3 Ghz when this happens, but probably it will be.
This answer applies to the STOCK kernels. All of this behavior is under control of the kernel, and some of the dev kernels alter this code with different behaviors.
cheers.
Niice info man. thanks very much. any idea on how much RAM the gfx uses?
-PiLoT- said:
Niice info man. thanks very much. any idea on how much RAM the gfx uses?
Click to expand...
Click to collapse
No clue. Don't even know if the GPU has some private SRAM on the SoC. Consult the Tegra 3 TRM for that. But, a single-layer 24-bit framebuffer @ 1280x800 needs 3MB for a single layer. Add in z-buffering and whatever else the openGL ES (1.0 & 2.0) libraries do with offscreen/onscreen primitive storage in userland memory space and I suppose it can be much much higher than that.
Somehow I sense you are asking about kernel-reserved memory; I don't know if the kernel reserves some minimum amount of general-purpose RAM (e.g. for the framebuffer) or not. Sorry. (Use the force, read the source!)
bftb0 said:
No clue. Don't even know if the GPU has some private SRAM on the SoC. Consult the Tegra 3 TRM for that. But, a single-layer 24-bit framebuffer @ 1280x800 needs 3MB for a single layer. Add in z-buffering and whatever else the openGL ES (1.0 & 2.0) libraries do with offscreen/onscreen primitive storage in userland memory space and I suppose it can be much much higher than that.
Somehow I sense you are asking about kernel-reserved memory; I don't know if the kernel reserves some minimum amount of general-purpose RAM (e.g. for the framebuffer) or not. Sorry. (Use the force, read the source!)
Click to expand...
Click to collapse
No im thinking more a portion of the 1gb system RAM was given to the gfx to play with. so our system RAM is only say 768mb
Related
So I was wondering if there is any hack that would add more ram to the GPU, comparing to RAM hack that takes GPU RAM and gives it to Phones RAM, Could it be done vice-versa, so take ram from Phone RAM and giving it to GPU, could it be done? and would it increase GPU processing power?
i too was wondering this too because im a heavy gamer
I'm just assuming but, the RAM pool is all the same RAM chips. It's just being managed and distributed between the GPU and CPU by software.
But I'm not sure.
Taknarosh said:
I'm just assuming but, the RAM pool is all the same RAM chips. It's just being managed and distributed between the GPU and CPU by software.
But I'm not sure.
Click to expand...
Click to collapse
Same here, but again im not sure
Is there any way to improve the memory bandwidth? Or better, I know that the "formula" for the bandwidth should be [memory frequency * bus width] (that AFAIK is 32bit for Tegra 3) but I didn't understand if the memory frequency is bound somehow to the CPU/GPU clocks that are easily manageable with kernel tools or I have to look somewhere else.
Everything I found was an old post from hamdir where he quoted another guy saying that the memory controller was bound to CPU clock, but I don't understand if that statement was confirmed.
If everything above is confirmed os is it correct to say that the "GPU clock" (listed and OCed in many kernels) affects only the pixel/vertex fillrate?
Thanks
So I've been noticing somethin strange other then the experia z every single other high end device that will come out as of 2013 will be using a 1.7GHZ snapdragon the only high end device that will use a lower one is experia z why????? I mean the phone hasn't come out yet can't they just put the 1.7GHZ instead? why use a lower one 1.5GHZ from phones of the fourth quarter of 2012 its old tech!!!
Sent from my SCH-I510 using xda app-developers app
Calm down man, 1.5GHz quadcore is good for Android. Android really doesn't need such a high speeds
And CPU clock doesn't mean smoother device
Sent from my LT26i using xda app-developers app
1.5Ghz Quad Core is not "only". And they all use exactly the same CPU which operates with 1,5-1,7Ghz speed. Clock speed isn't everything. Look at the iPhone - it uses dual core 1.2Ghz CPU - yet its performance is 5 times smoother than most quad core devices...
Dual core is more than enough for anything a smartphone might need to do - as long as you put good software on it.
gabrielpina4 said:
So I've been noticing somethin strange other then the experia z every single other high end device that will come out as of 2013 will be using a 1.7GHZ snapdragon the only high end device that will use a lower one is experia z why????? I mean the phone hasn't come out yet can't they just put the 1.7GHZ instead? why use a lower one 1.5GHZ from phones of the fourth quarter of 2012 its old tech!!!
Sent from my SCH-I510 using xda app-developers app
Click to expand...
Click to collapse
Seriously? "Only" on a quad core 1.5ghz CPU for a smartphone? What are you gonna run on your phone that would require you more? Or do you just want your battery to drain more for no possible reasons? Some users even just underclock for the sake of saving battery.
If you ask me I buy this because of the features and the RAM not because of processor clock.
If you ever let me choose I would rather go for a 1ghz dual core processor with 4GB RAM than having a 2ghz quad core processor with 2GB RAM.
The only reason why I would want the Qualcomm S4 Pro processor is because of adreno 320.
A dual core is good enough dude .. this is a quad core lol
Sent from my GT-N7000 using xda premium
No offense but you just sound like your trolling. Same goes for your other thread too
AK4TAY7BEN said:
No offense but you just sound like your trolling. Same goes for your other thread too
Click to expand...
Click to collapse
I agree. I think he's trolling.
Riyal said:
Seriously? "Only" on a quad core 1.5ghz CPU for a smartphone? What are you gonna run on your phone that would require you more? Or do you just want your battery to drain more for no possible reasons? Some users even just underclock for the sake of saving battery.
If you ask me I buy this because of the features and the RAM not because of processor clock.
If you ever let me choose I would rather go for a 1ghz dual core processor with 4GB RAM than having a 2ghz quad core processor with 2GB RAM.
The only reason why I would want the Qualcomm S4 Pro processor is because of adreno 320.
Click to expand...
Click to collapse
What would be the purpose of 4GB of RAM in a smartphone are you running Photoshop or a VM? Where did this stigma that more cores is equivalent to less efficient power use. More asynchronous cores allow for more accurate scaling of processing power to needs, thus higher power efficiency because the CPU spends less time in higher power states.
REAVER117 said:
What would be the purpose of 4GB of RAM in a smartphone are you running Photoshop or a VM? Where did this stigma that mire cores is equivalent to less efficient power use. More asynchronous cores allow for more accurate scaling of processing power to needs, thus higher power efficiency because the CPU spends less time in higher power states.
Click to expand...
Click to collapse
Buddy android runs Dalvik VM and our GPU uses dedicated RAM for processing certain graphic tasks. So yes we are running a VM on our mobile phones.
Also with a proper monitoring app you could see for yourself which uses more and needs more resource here. whether processing or memory.
Didn't you even wonder why our phones consume almost 700mb RAM without even a single foreground application open? Yet it sits idle in a single core and remains in between 400mhz+ to 900mhz?
Also I didn't say that more cores consumes more power in my statement but still since each core requires power to run it should also consume more. Also I didn't mention anything about number of core I was talking about the clocks. Since the higher the clock requires more Amperes. Still underclocking your phone by 500mhz would prolly just save about 2% of your battery anyways.
you don't need higher cpu clock to get a pocket heater for this winter
My Optimus G has the same cpu an the thermal throttling is kikin in pretty fast. This 1.7Ghz S4pro will thermal throttle as fast or even faster, rendering numerical advantage meaningless.
I don't want to feed the troll but I also think dual core is good enough.
Riyal said:
Buddy android runs Dalvik VM and our GPU uses dedicated RAM for processing certain graphic tasks. So yes we are running a VM on our mobile phones.
Also with a proper monitoring app you could see for yourself which uses more and needs more resource here. whether processing or memory.
Didn't you even wonder why our phones consume almost 700mb RAM without even a single foreground application open? Yet it sits idle in a single core and remains in between 400mhz+ to 900mhz?
Also I didn't say that more cores consumes more power in my statement but still since each core requires power to run it should also consume more. Also I didn't mention anything about number of core I was talking about the clocks. Since the higher the clock requires more Amperes. Still underclocking your phone by 500mhz would prolly just save about 2% of your battery anyways.
Click to expand...
Click to collapse
I have yet to see my Xperia ZL have less than 800MB of free an extra 2GB would be a total waste. And you simply stated you'd rather have a dual than a quad, peak frequency means very little except for potential processing power. Just because your CPU is capable of a higher clock speed doesn't mean it will idle any higher. Likewise the only difference between a quad with a single utilized core and a dual with a single utilized core is the minute amount if extra leakage current for the quad.
I think you meant to say that higher frequencies may need more voltage, amperage has very little to do with CPU frequency scaling.
And obviously the memory overhead for a virtualized process i.e. Dalvik VM is not even in the same league as a system VM.
REAVER117 said:
I have yet to see my Xperia ZL have less than 800MB of free an extra 2GB would be a total waste. And you simply stated you'd rather have a dual than a quad, peak frequency means very little except for potential processing power. Just because your CPU is capable of a higher clock speed doesn't mean it will idle any higher. Likewise the only difference between a quad with a single utilized core and a dual with a single utilized core is the minute amount if extra leakage current for the quad.
I think you meant to say that higher frequencies may need more voltage, amperage has very little to do with CPU frequency scaling.
And obviously the memory overhead for a virtualized process i.e. Dalvik VM is not even in the same league as a system VM.
Click to expand...
Click to collapse
Buddy that 800mb is without running any foreground applications other than either Settings app or the homescreen. Like I said try running some monitoring apps on the background and try using the browser then maybe try loading some sites with lots of javascript codes in it.
Or if you like try launching 4 types of angry birds and the 2 temple run games simultaneously without killing one of them and let's see if OOM doesn't kick in and kill any of them.
Regarding CPU cores please state something that requires a quad core processor. The smoothness of the UI your experiencing is because of the type of processor your phone is using "The Snapdragon S4 Pro" even if you disable all the extra cores in it you won't feel anything different unless of course you'll run some benchmark tools or video decoding stuffs in there.
And just FYI more cores doesn't mean greater processing power. It's more cores = more processes it can handle.
And on the CPU freq clocks who said t that amperes doesn't increase on each frequency table? Please take a look at qualcomm's document on their site regarding it's processors so you would know how they calculate it. Voltage is just used to provide more electricity to power up the processor but voltage alone won't make a processor active.
This discussion is going to the wrong way.
Thread closed.
Hello,
As you may know in latest 4.4.4 update, there is only 1.1 GB ram available to the system and something around 800MB is dedicated to GPU ! some people argue it is because of high resolution of screen (2560x1440) but still 800MB seems so high to me. my HTC One m7 has a 1920x1080 screen and it has 1.7 GB ram available which makes it only 200MB dedicated ram for GPU. by same proportions nexus 10 should not need more than 300MB of dedicated GPU ram. I know it is not that simple and they use different kind of SOC but I would like to play with this value and check its impact on system.
does anyone know how it is possible to change the dedicated GPU ram in Nexus 10 ? it should be somewhere in kernel but I dont know if it is configurable or a constant value in kernel code.
Regards
fifthelement said:
Hello,
As you may know in latest 4.4.4 update, there is only 1.1 GB ram available to the system and something around 800MB is dedicated to GPU ! some people argue it is because of high resolution of screen (2560x1440) but still 800MB seems so high to me. my HTC One m7 has a 1920x1080 screen and it has 1.7 GB ram available which makes it only 200MB dedicated ram for GPU. by same proportions nexus 10 should not need more than 300MB of dedicated GPU ram. I know it is not that simple and they use different kind of SOC but I would like to play with this value and check its impact on system.
does anyone know how it is possible to change the dedicated GPU ram in Nexus 10 ? it should be somewhere in kernel but I dont know if it is configurable or a constant value in kernel code.
Regards
Click to expand...
Click to collapse
It is not configurable, it is code changes and I never have time to work on this device anymore due to work load on my real job.
I want to raise question is it possible to overclock the CPU, or GPU or memory speed. We all know that ze550kl / z00l performance with msm8916. Is almost in the limit to run apps and games, wanted to know if possible or not, and if there is any development has this respect. I like the device and it is great but it is clear that it is getting weak in the matter of hardware, if someone can work in this respect to give a breath / power to our device
Ermes.mt/brasil said:
I want to raise question is it possible to overclock the CPU, or GPU or memory speed. We all know that ze550kl / z00l performance with msm8916. Is almost in the limit to run apps and games, wanted to know if possible or not, and if there is any development has this respect. I like the device and it is great but it is clear that it is getting weak in the matter of hardware, if someone can work in this respect to give a breath / power to our device
Click to expand...
Click to collapse
i don't think so , if its possible to overclock the processor to its frequency,
you can costumize it too max frequency of it or to lower , as u need
But if is then i would love like to know too
I am already extracting the maximum, from him that I can of him, already put using ex kernel, with governor ordenad, frequency of 200mhz @ 12000mhz, and simple GPU ordered frequency of 200mhz to 400mhz, active zram of 512mb, and entropy of 128/256, then Of course, here with standard frequency of the paddle extracting nothing more than that, the problem and that we have a large bottleneck in memory lpddr3 of 533mhz, that opera and 32bit, there is a single channel giving bandwidth of 4.2gb / s that It has to be split between CPU and GPU. What is clearly one of the pro soc power limiter, I do not know much about the arm architecture, but I can not do over clock in lpddr3 memory, since I did not see any software or tutoring Citing this. Although this would be a solution, if you do over clock in the CPU or ns GPU will force the clock controller up what will increase the memory performance consequently, something else, not good will lock the clock at maximum in 1200mhz and GPU at 400mhz, as there will be a large power consumption and the performance will be limited by the cords of the tempering control and the low frequency of the RAM, # excuse me to have been so far more enthusiastic about hardware,
even with my old htc explorer i could increase cpu clocsk from 600mhz to 900mhz why we cant do it on zenfone 2 just we need good developer and kernel
it's up to the hardware
If you look at the specs from qualcom it has a max speed and not every device is possible to reach the max, but if you have the proper kernel you can reach that speed and "overclock"