So as we all know with our devs pushing the cpu like up to these ridiculous speeds and our benchmarks not really showing the type of jump that we were really expecting, Is it possible that the way the CPU works is causing these scores not to score as high but not affect it's speed in the same way?
So after reading a bit on our CPU, from what I got from it ours does has about the average amount of IPS(instructions per second) compared to the snapdragon and it's custom architecture. However ours is said to be designed in a way that it requires 25-50% less instructions for 20~% of it's functions. So if thats the case we should be doing a little better in different areas.
In turn my real question is...
Because these benchmarks were built prior to our CPU being out, will that affect the numbers considering they probably aren't built for the way ours functions. Or am I just thinking way to far into this and have no idea what i'm talking about .
Which benchmarks are you referencing?
MFLOPS?
lqaddict said:
Which benchmarks are you referencing?
MFLOPS?
Click to expand...
Click to collapse
Anything from Linpack to SetCPU native benchmarks.
xplanowestx said:
Anything from Linpack to SetCPU native benchmarks.
Click to expand...
Click to collapse
You should probably read a bit on LINPACK - it was originally designed in the late 70's, 1979 to be precise, it was a set of Fortran procedures to perform numeric linear algebra. It is quite affected by the OS, and that's why you see LINPACK scores on 2.2+ systems are quite "improved" over 2.1 or earlier systems running better or the same hardware.
As far as SetCPU is concerned - there is no point running the benchmark on the CPU's that do not support NEON instructions, NEON is a Cortex-A specific (our phone and iPhone4 are among the few that have it); running it on a non-Cortex-A CPU, like a G2's GenII Snapdragon/Sciorpion from Qualcomm, does nothing but confuse people with bunch of numbers (the numbers are correct, they show how long it took that CPU to ignore the instruction instead of execute it, ignoring takes a lot less time than executing ).
And the C number in native SetCPU, I would assume it is a condition set on the instruction set to Carry on even if the previous instruction resulted in NOP (No Operation), so in the case of the glorified G2 for instance (you prolly saw the pretty screenshots) after ignoring the NEON instruction, it carries on ignoring the next NEON instruction throwing at it, so I am not sure what it is indicative of though, have to ask the SetCPU dev.
I hope this helps.
lqaddict said:
You should probably read a bit on LINPACK - it was originally designed in the late 70's, 1979 to be precise, it was a set of Fortran procedures to perform numeric linear algebra. It is quite affected by the OS, and that's why you see LINPACK scores on 2.2+ systems are quite "improved" over 2.1 or earlier systems running better or the same hardware.
As far as SetCPU is concerned - there is no point running the benchmark on the CPU's that do not support NEON instructions, NEON is a Cortex-A specific (our phone and iPhone4 are among the few that have it); running it on a non-Cortex-A CPU, like a G2's GenII Snapdragon/Sciorpion from Qualcomm, does nothing but confuse people with bunch of numbers (the numbers are correct, they show how long it took that CPU to ignore the instruction instead of execute it, ignoring takes a lot less time than executing ).
And the C number in native SetCPU, I would assume it is a condition set on the instruction set to Carry on even if the previous instruction resulted in NOP (No Operation), so in the case of the glorified G2 for instance (you prolly saw the pretty screenshots) after ignoring the NEON instruction, it carries on ignoring the next NEON instruction throwing at it, so I am not sure what it is indicative of though, have to ask the SetCPU dev.
I hope this helps.
Click to expand...
Click to collapse
It does help very much so which is sort of what i was seeing regarding SetCPU scores, they didn't seem correct based off the newer frequency's of the TeamWhiskey kernel they are developing. This would make sense.
So regarding linpack and 2.2, why is it that the Snapdragon is able to pull out these 55+ linpack results? It doesn't make sense that our CPU wouldn't be capable to pull off the same with what were are able to OC our CPU too, were getting much lower results compared to other phones moving to 2.2, if you can please explain this a little further. Thanks!
xplanowestx said:
It does help very much so which is sort of what i was seeing regarding SetCPU scores, they didn't seem correct based off the newer frequency's of the TeamWhiskey kernel they are developing. This would make sense.
So regarding linpack and 2.2, why is it that the Snapdragon is able to pull out these 55+ linpack results? It doesn't make sense that our CPU wouldn't be capable to pull off the same with what were are able to OC our CPU too, were getting much lower results compared to other phones moving to 2.2, if you can please explain this a little further. Thanks!
Click to expand...
Click to collapse
It's the OS limitation, and I am trying not to make an excuse for it, but RFS as well.
LINPACK score is highly affected on what your OS is doing, and how it is doing what it does , given a fair chance, with Galaxy S hopefully getting 2.2 this year (LOL), and team z4mod developing the overhaul of the cripple RFS you will see comparable LINPACK scores.
lqaddict said:
It's the OS limitation, and I am trying not to make an excuse for it, but RFS as well.
LINPACK score is highly affected on what your OS is doing, and how it is doing what it does , given a fair chance, with Galaxy S hopefully getting 2.2 this year (LOL), and team z4mod developing the overhaul of the cripple RFS you will see comparable LINPACK scores.
Click to expand...
Click to collapse
You made some realy great points, it is always good to have knowledgeable people around to answer technical questions. However, I don't understand how RFS, a file system affects arithmetic calculation? (from what I read, Linpack is basically a really difficult math problem).
It's the OS limitation, and I am trying not to make an excuse for it, but RFS as well.
LINPACK score is highly affected on what your OS is doing, and how it is doing what it does , given a fair chance, with Galaxy S hopefully getting 2.2 this year (LOL), and team z4mod developing the overhaul of the cripple RFS you will see comparable LINPACK scores.
Click to expand...
Click to collapse
Thanks for the very in depth answers. Hopefully fixing the filesystem would fix the speeds and such of the CPU. I'm hoping your right on that idea. If thats the case, do you think we will be comparable to the G2's Extremely high scores?
You made some realy great points, it is always good to have knowledgeable people around to answer technical questions. However, I don't understand how RFS, a file system affects arithmetic calculation? (from what I read, Linpack is basically a really difficult math problem).
Click to expand...
Click to collapse
+1
On a second note, what would compel Samsung to use such filesystem? Is it because of the proprietary side of it?
xplanowestx said:
Thanks for the very in depth answers. Hopefully fixing the filesystem would fix the speeds and such of the CPU. I'm hoping your right on that idea. If thats the case, do you think we will be comparable to the G2's Extremely high scores?
+1
On a second note, what would compel Samsung to use such filesystem? Is it because of the proprietary side of it?
Click to expand...
Click to collapse
I might've gone there on a limp, but from my understanding is that floating point operations stressed out by computing NxN matrix (LINPACK benchmark) is optimized to take an advantage of the L2 cache on the processor. Well our processor has 512K L2 cache if I am not mistaken, and from what I read on Qualcomm CPU in G2 - it has no cache at all. So where to keep the data? The next fastest media is memory and then disk I/O, and idsk I/O is where the weakest link is, given the fact that we have a decaying FAT32 implementation the weakest link is even weaker.
Remember your system is as good as your weakest link.
Did I make a plausible excuse ?
lqaddict said:
I might've gone there on a limp, but from my understanding is that floating point operations stressed out by computing NxN matrix (LINPACK benchmark) is optimized to take an advantage of the L2 cache on the processor. Well our processor has 512K L2 cache if I am not mistaken, and from what I read on Qualcomm CPU in G2 - it has no cache at all. So where to keep the data? The next fastest media is memory and then disk I/O, and idsk I/O is where the weakest link is, given the fact that we have a decaying FAT32 implementation the weakest link is even weaker.
Remember your system is as good as your weakest link.
Did I make a plausible excuse ?
Click to expand...
Click to collapse
It COULD be possible, considering how freeing up RAM increases linpack scores (to an extend). But how about people who have voodoo? Isn't that ext 4, so linpack on those system should fly off the chart but they don't/
PaiPiePia said:
It COULD be possible, considering how freeing up RAM increases linpack scores (to an extend). But how about people who have voodoo? Isn't that ext 4, so linpack on those system should fly off the chart but they don't/
Click to expand...
Click to collapse
Voodoo only addresses /data, I would assume instructions for LINPACK calculation will be punted to /cache
So then the new z4mod should (hopefully) raise our linpack scores fairly higher since thats a 100% complete conversion correct?
xplanowestx said:
So then the new z4mod should (hopefully) raise our linpack scores fairly higher since thats a 100% complete conversion correct?
Click to expand...
Click to collapse
I believe with mature z4mod and a stable froyo build the Galaxy S will shine, hopefully it is not a wishful thinking, and these things happen sooner than later for us.
lqaddict said:
I believe with mature z4mod and a stable froyo build the Galaxy S will shine, hopefully it is not a wishful thinking, and these things happen sooner than later for us.
Click to expand...
Click to collapse
+1 to that.
Related
How high do you think we can clock the processors on the EVO 3D? I recall they are 1.5 ghz chips underclocked to conserve battery life. Think these can hit that magical 2.0? Or at least 1.8?
I could see maybe 1.6 but honestly nothing over 1.4ghz is worth it... (batter>speed)
And nothing currently requires anything over 1.2ghz or 1.5ghz for that matter, other than peoples e-penis.
Id like to see a 1.4ghz uv kernel over 1.8ghz 1 hour battery killer but I will use and test all of them
sent from anything but an iPhone
nate420 said:
I could see maybe 1.6 but honestly nothing over 1.4ghz is worth it... (batter>speed)
And nothing currently requires anything over 1.2ghz or 1.5ghz for that matter, other than peoples e-penis.
Id like to see a 1.4ghz uv kernel over 1.8ghz 1 hour battery killer but I will use and test all of them
sent from anything but an iPhone
Click to expand...
Click to collapse
Well that's your opinion. I highly doubt a overclocking the processor to 1.8 would bring the phone down to one hour of battery life. It's not like it would be constantly running at that speed. I would prefer speed over battery life as I charge my phone every night and have plenty left over even overclocked to almost 1.3 on my EVO.
nate420 said:
I could see maybe 1.6 but honestly nothing over 1.4ghz is worth it... (batter>speed)
And nothing currently requires anything over 1.2ghz or 1.5ghz for that matter, other than peoples e-penis.
Id like to see a 1.4ghz uv kernel over 1.8ghz 1 hour battery killer but I will use and test all of them
sent from anything but an iPhone
Click to expand...
Click to collapse
I think this is less about practicality and more about pushing our phone to the limits. overclocking on an already fast enough processor on a device which runs for the most part on battery, is not needed. however it is fun and nice to see the benchmarks soar.
I say 1.8ghz-2ghz
If they're anything like the EVO 4G, then it wont be a very high overclock
But assuming all are capable of 1.5 GHz, then it would be at least a 400-450 MHz overclock!
freeza said:
If they're anything like the EVO 4G, then it wont be a very high overclock
But assuming all are capable of 1.5 GHz, then it would be at least a 400-450 MHz overclock!
Click to expand...
Click to collapse
My g2x was overclocked to 1.6ghz and its only a 1ghz dual core phone...
Id say we could see maybe 1.8ghz if this phone is really 1.5 dropped down to 1.2
sent from anything but an iPhone
fmedina2 said:
Well that's your opinion. I highly doubt a overclocking the processor to 1.8 would bring the phone down to one hour of battery life. It's not like it would be constantly running at that speed. I would prefer speed over battery life as I charge my phone every night and have plenty left over even overclocked to almost 1.3 on my EVO.
Click to expand...
Click to collapse
Again for e-penis and bragging rights on benchmarks nothing more...
As for saying 1.8 oc would kill it in a hour I was joking...
And I bet dollars to donuts you don't see a change in "speed" past 1.6ghz other than a hot battery.
Ginger bread can't fully optimize dual cores it does the job but untill a new os is out
no point ruining a battery for "speed" you won't see
sent from anything but an iPhone
While performance is key, I'd say this phone is well above the bar of expectations for most Android Apps at the current time. I'm more interested in squeezing the most battery life I possibly can via Underclocking. It will be nice to see how far this can be pushed with Two Cores to spread the workload across.
nate420 said:
I could see maybe 1.6 but honestly nothing over 1.4ghz is worth it... (batter>speed)
And nothing currently requires anything over 1.2ghz or 1.5ghz for that matter, other than peoples e-penis.
Id like to see a 1.4ghz uv kernel over 1.8ghz 1 hour battery killer but I will use and test all of them
sent from anything but an iPhone
Click to expand...
Click to collapse
btw the way i have the bigest e penis lol it is googolplex inchs
why are people saying such low numbers the second gen snapdragons can go to what 1.9? if ours is 1.5 stock dropped down to 1.2 then i think we can at least hit 2
I'd bet that the chips in these phones will be those that were unstable at 1.5 ghz. That's how chip makers do these things. They make them all the same, then those with unstable silicon are sold as a lower clock speed. Not sure I'd expect over 1.5 and that might require higher voltage. Hope I'm wrong. We'll see I guess.
hdad2 said:
I'd bet that the chips in these phones will be those that were unstable at 1.5 ghz. That's how chip makers do these things. They make them all the same, then those with unstable silicon are sold as a lower clock speed. Not sure I'd expect over 1.5 and that might require higher voltage. Hope I'm wrong. We'll see I guess.
Click to expand...
Click to collapse
Wrong, to lazy to explain for now.
toxicfumes22 said:
Wrong.....
Click to expand...
Click to collapse
Hope so!
10char
toxicfumes22 said:
Wrong, to lazy to explain for now.
Click to expand...
Click to collapse
OK, a little less lazy right now. But simply the way that manufactures choose the speeds for processors is actually simple. In the case of the 3D it IS underclocked. The processor is an asynchronous dual core with clock speeds initially set at 1.5 by Qualcom and is used in Qualcom's phone they produce for developers. It is underclocked by HTC because of battery problems listed from the 4G and the unnecessary need of 1.5GHz in a F*ing phone. Manufactures for the most part do not underclock the CPU. The reason it is set at the level it is, is because it is most stable, efficient and meets the heat extraction needs (People forget CPUs are just circuits and produce heat with more voltage). OK lets back this up shall we. OK.
That is why I'm too lazy to post thing, I have to search up a link cause most of this is my general knowledge. Anyways, the QSD8650 found in the EVO 4G is clocked at 1GHz and has been posted to a stable 1.3GHz I believe by a recent post. Now the MSM8660 is posted to be a 1.5GHz CPU, so its overclocking potential is more near 2GHz but I would suspect it to get a little warm(sweaty palms anyone?) and I wouldn't know how stable it would be either (I don't know phones the best). Why is it underclocked? Because people kept *****ing at how much battery the EVO used and as technology improves so does the efficiency of CPUs so they go with the most recent and just underclock it. I've seen a comparison graph somewhere by Qualcom but I spent about 10minutes looking for it and couldn't find it but it was really nifty. If someone finds it plz post it, it shows the energy vs Clock speed and it is very cool.
Anyways, to respond to whoever said that the 1.5GHz is the max and that all manufacturers underclock the CPU based upon the silicon is WRONG, wrong WrOnG and Rong/wong (Im sorry I dont remember the exact response). Anyways, its the heat extraction and the silicon hurts it because it doesn't let all the heat through, which is one of the reason your PS3 may have yellow lighted on you(Yes its because of the CPU disconnecting from the Motherboard, but why do you think this extra heat was generated?).
Sorry this is so long and I got distracted a few times while writing it so it I messed up or something doesn't make sense I apologize but being lazy is really a pain in the ass.
hdad2 said:
I'd bet that the chips in these phones will be those that were unstable at 1.5 ghz. That's how chip makers do these things. They make them all the same, then those with unstable silicon are sold as a lower clock speed. Not sure I'd expect over 1.5 and that might require higher voltage. Hope I'm wrong. We'll see I guess.
Click to expand...
Click to collapse
That would be the case if this wasn't an MSM 8660. You're thinking like when AMD makes chips for the HD 6970 and some are found not to be stable at 880 mhz so they bin it to use in the HD 6950 which runs at 800 mhz. These are actually sold as two separate products. In the case of the processor in the Evo it's an MSM 8660 which is sold by qualcomm to be run at speeds as high as 1.5 ghz. If they wanted to sell chips binned for lower speeds they'd have to sell it as a different model since it wouldn't be capable of the 1.5hz.
jersey221 said:
why are people saying such low numbers the second gen snapdragons can go to what 1.9? if ours is 1.5 stock dropped down to 1.2 then i think we can at least hit 2
Click to expand...
Click to collapse
1.9?
No sir it was 1.19stable...
Sent from my PC36100 using Tapatalk
donatom3 said:
That would be the case if this wasn't an MSM 8660. You're thinking like when AMD makes chips for the HD 6970 and some are found not to be stable at 880 mhz so they bin it to use in the HD 6950 which runs at 800 mhz. These are actually sold as two separate products. In the case of the processor in the Evo it's an MSM 8660 which is sold by qualcomm to be run at speeds as high as 1.5 ghz. If they wanted to sell chips binned for lower speeds they'd have to sell it as a different model since it wouldn't be capable of the 1.5hz.
Click to expand...
Click to collapse
Can you explain this to me please.
toxic and donatom,
Your explanations make perfect sense. So I hope to be wrong. Does qualcomm sell a processor with that same architecture and a lower clock advertised?
Just seems like they're not gonna throw them away if they are stable and 1.2 or 1.4 but less stable at 1.5+. The 3vo seems like a good way for them to unload those processors.
hdad2 said:
toxic and donatom,
Your explanations make perfect sense. So I hope to be wrong. Does qualcomm sell a processor with that same architecture and a lower clock advertised?
Just seems like they're not gonna throw them away if they are stable and 1.2 or 1.4 but less stable at 1.5+. The 3vo seems like a good way for them to unload those processors.
Click to expand...
Click to collapse
To my knowledge, if this happens it gets recycled. But.....if this happens a lot then they need to change their manufacturing process or that the technology isn't there yet. Like now we have the technology to do 64GB MicroSD, but why do it because most devices can only do 32GB. For the companies that do sell them, well....I don't have good words for them, I also don't know of this happening. I can understand that it could be useful for donations to universities or others that could use them for damn near free prices, but not resold even under a different name.
toxicfumes22 said:
Can you explain this to me please.
Click to expand...
Click to collapse
Well in the case of AMD with many of their chip lines they produce a higher end chip. The ones that don't fully pass the tests at the higher speed get sold as a different model with a lower clock and voltage.
I have the most experience with the HD 6970 and 6950. They both use the same GPU, but the ones in the 6950 didn't pass AMD's tests at higher speeds so they are set at a lower clock and voltage than the 6970 (they also have some shaders disbaled). They are sold as two different models even though they were made the exact same way with the same silicone. This is not new chip manufacturers have been doing this for a while.
Think of it this way I make 100k chips out of those 100k I'm going to have a percentage that can't perform at their top performance, so instead of throwing them away I make a different model and underclock it and still make money on the chips that didn't pass at the higher speed. Now sometimes I will sell more of the lower end model so I actually have to take some chips that probably would have passed as the higher end model and sell them at the lower end. In this case the user gets lucky and can unlock their chip to the performance of the higher priced model.
EDIT: What HTC is doing here is buying a 1.5ghz chip but purposely underclocking it to save battery, since they figured most users wouldn't see the .3 ghz difference but would see the difference in battery life. Again in video cards you see this but usually the other way around. A manufacturer such as Asus, gigabyte, whomever takes the best of their chips they bought and overclocks them because again some were made even better than the standards set by AMD or Nvidia.
I guess what I'm trying to say here is that ALL these chips should do 1.5 ghz stable without question, unless there isn't enough space inside for the cooling requirements at 1.5ghz (which I doubt), and most should easily go above 1.6.
Edit again since I just saw this post:
toxicfumes22 said:
To my knowledge, if this happens it gets recycled. But.....if this happens a lot then they need to change their manufacturing process or that the technology isn't there yet. Like now we have the technology to do 64GB MicroSD, but why do it because most devices can only do 32GB. For the companies that do sell them, well....I don't have good words for them, I also don't know of this happening. I can understand that it could be useful for donations to universities or others that could use them for damn near free prices, but not resold even under a different name.
Click to expand...
Click to collapse
This is something that happens mostly in higher end processors because their tolerances at those speeds are less forgiving. No manufacturing process is perfect, you're going to have some that won't perform at those very high speeds, and recycling would cost more to the company and environment then simply selling them at lower speeds. These chips are not bad, and not defective, just found to not be stable at those highest speeds, but are perfectly fine at the speeds they are being sold at, so why throw them away. If they don't meet the standards at the lower speed then yes they would be recycled.
I've been looking for a more technical analysis of these SOCs and I have been trying to learn how the async CPU setup on the MSM8660 affects performance.
Nvidia claims that the power saving feature of our CPU (async) will inevitably cause a decrease in performance:
http://www.intomobile.com/2011/03/2...ed&utm_campaign=Feed:+IntoMobile+(IntoMobile)
Does anyone have any comments on this? If this is the case, I am wondering if through software we can force both cores to run at the same voltage/frequency. I wonder if it would cause an increase in performance (at least in benchmarking). Many claim that the Evo 3d only gets medicore benchmark scores due to having asynchronous cores that are not being accurately benched. It would be interesting to verify this claim.
Also, does anyone know which SOC between the three I listed in the title is the highest in performance (not talking about useless benchmarks like quadrant)?
So....there is possibly a 10–15% decrease in performance.....that's fine with me. Most of the time you won't even notice until you run benchmarks and looks at the numbers.
SetCPU + Performance mode are all you should need
DarkManX4lf said:
So....there is possibly a 10–15% decrease in performance.....that's fine with me. Most of the time you won't even notice until you run benchmarks and looks at the numbers.
Click to expand...
Click to collapse
Well the 10-15% slower is nVidia's claim, not sure if its true.
Does that make both cores run at the same time or running cores at the same time not possible due to the processor
xHausx said:
SetCPU + Performance mode are all you should need
Click to expand...
Click to collapse
Sent from my PG86100 using XDA App
ttieder said:
Does that make both cores run at the same time or running cores at the same time not possible due to the processor
Sent from my PG86100 using XDA App
Click to expand...
Click to collapse
It will keep the cpu running at full speed. Which core gets used for what depends on a lot of things but it mostly depends on how the apps and kernel are programmed.
xHausx said:
It will keep the cpu running at full speed. Which core gets used for what depends on a lot of things but it mostly depends on how the apps and kernel are programmed.
Click to expand...
Click to collapse
Yes, but is it possible to keep both cores at their full frequency? Setting the exynos or tegra 2 on performance mode makes both cores stay at their maximum frequency since they are synchronous. I think setting performance mode on the Evo 3d would only guarantee that one of the core will remain at its full frequency.
Not sure about this of course. Anyone have any insight into this?
Second Core wouldnt kick in if ur not heavily multitasking or running multithreaded apps and u wouldnt need second core for minor multitasking or single threaded operations as single core is enough
i will tell you that on paper the msmx60 should beat out all, but in real world use, the exynos hammers everything. the s2 is a beast
The Exynos is the better SoC, plain and simple. If we get into GPU discussions, the Adreno 220 is the best, as in better than Mali 400.. Go to Anandtech, and watch them use a Qualcomm device for the benches.
Sent from my PG86100 using XDA App
Is it a "for sure" thing that ICS will use the GPU acceleration in the OS? Or is that just everyone's hopes and dreams
Sent from my PG86100 using XDA Premium App
You could program the kernel to keep both cores at max frequency. Im not a developer but am sure something like this could be done
Sent from my HTC Sensation Z710e using Tapatalk
bballer71418 said:
Is it a "for sure" thing that ICS will use the GPU acceleration in the OS? Or is that just everyone's hopes and dreams
Sent from my PG86100 using XDA Premium App
Click to expand...
Click to collapse
ics will include all of the features that honeycomb has and honeycomb has 2d acceleration so yes
Sent from my PG86100 using XDA Premium App
Maybe we should make some real world benchmarks and get some SGS2 people in on it. Like how fast a particular app opens(say angry birds), How many fps a game plays at, Convert a file to another format, complete a 5 step plan to take over the world things like that. Alot of things like that are how reviewers rate and test things like new video cards and cpus plus all the benchmark programs.
I used to use a program called fraps to see how many fps my pc games were playing at so I could tweak stuff( long live unreal tournament!!!!). It would just display the fps in the top corner of the screen.
Also comparing the 3vo and SGS2 will really heat up when we get root and CM7. 400mb less roms have gatta make a huge difference on performance. I dunno about you guys but I haven't been able clog up my 3vo yet(and I've been trying!), I'm pretty impressed with the hardware so far.
Drewmungus said:
Maybe we should make some real world benchmarks and get some SGS2 people in on it. Like how fast a particular app opens(say angry birds), How many fps a game plays at, Convert a file to another format, complete a 5 step plan to take over the world things like that. Alot of things like that are how reviewers rate and test things like new video cards and cpus plus all the benchmark programs.
I used to use a program called fraps to see how many fps my pc games were playing at so I could tweak stuff( long live unreal tournament!!!!). It would just display the fps in the top corner of the screen.
Also comparing the 3vo and SGS2 will really heat up when we get root and CM7. 400mb less roms have gatta make a huge difference on performance. I dunno about you guys but I haven't been able clog up my 3vo yet(and I've been trying!), I'm pretty impressed with the hardware so far.
Click to expand...
Click to collapse
Fraps tends to lie with FPS.
Sent from my PG86100 using XDA App
GPU acceleration will be nice. Hope we see ICS soon.
Sent from my EVO 3D w/ Tapatalk
It is known that the MSM8660 can achieve higher clock frequencies than the Exynos, though clock for clock the Exynos has better IPC.
As of right now the GSII beats the 3VO in both benchmarks and real world tests, but I suspect this is because Sense is a pig that takes far too much ram and system resources. HTC also seems to have poorer unoptimized drivers. In addition to this, the async CPUs of the 3VO may not be properly tested by current benchmarking tools.
I think comparing a rooted 3VO and a rooted GSII should be much closer. Imagine the MSM8660 at 1.8-2.0 Ghz both cores running full frequency with no Sense and other bloat to slow it down. Combine that with a hardware accelerated GUI and this phone should be amazing.
The Adreno GPU will get better over time... and will develop much faster than before. Since Qualcomm purchased the branch from AMD (ATi), there has been much improvement in a reasonably small amount of time. There are various claims that the Adreno 220 outperforms the Tegra 2. I havent seen a solid comparison of the Adreno 220 vs the Exynos although I have read that the Exynos is a very capable processor.
As they both stand in stock offering, the Samsung GS2 will be faster; it has tremendous less resources to move. I agree with what has been said about root & rom options: CM7 on the EVO 3D will likely result in unprecedented (real world) benchmarks .Also note that the current Android releases are not yet optimized for dual/quad core management. But rest assured, it is well under development and the Sprint EVO 4G4D (hypothetical name) will behold a treasure trove of menacing capabilities.
HTC + Qualcomm + Android = Future
I think we should just wait until we can do a head-to-head AOSP CM 7 benchmark/real world test to see what happens. I'm confident the SGSII will get shredded by the E3D.
It seems unfair to compare anything within the phone itself now, because of what each phone has to run. Sense is pretty tasking on our phones and I can't say as much for the opposition.
It's funny to see NVIDIA make snide comments about Qualcomm when their phones are getting bested. Although I must say it is impressive to see that Tegra 2 phones are over a year old and keeping up with the E3D's dual-core deliciousness.
Just my thoughts.
Personally I don't believe Nvidia, plenty of benchmarks contradict their statement. That and whoever said "Additionally, the operating systems like Android and many apps aren’t set up for an asynchronous architecture." is an idiot because 99% of apps in the market don't support dual core lmfao.
Just an epiphany..
The Evo3D has an asynchronous dual core, meaning as of right now, it unloads tasks to the other core when the 1st is being over burdened with things to do (am I essentially right on that one?)
So, basically, the second core only pops in to say hello once in a while right?
with they way that cpu's work, the harder you push it the higher amount of energy it needs and the higher amount of heat it throws off and that begins to multiply. So, overclocking the cpu even .2ghz increases heat and power usage by 1.3x rather than just 1.0x (not actual values, just stating it is MORE than a slight bump the higher it clocks.)
So, wouldn't running both cores more often than not, keeping them from reaching their peak speeds be way more beneficial to battery life than letting one core hit it's peak and than have the slacker core go "oh, dude... yeah, I'll take bag of breads while you carry the remaining rack of pop, bag of canned veggies, frozen pizzas and laundry detergent."
I dunno, just a thought.. now that we have s-OFF, is there any way we can drop the threshold so the second core will kick on sooner, rather than later?
Before the phone came out I read somewhere that one core operates the android system and radios while the other does user initiated tasks.
Im not really sure but that makes sense to me. It would be nice to have a definite answer on this though.
It has asynchronous cores lol. Not asymmetrical.
doctor ladd said:
Before the phone came out I read somewhere that one core operates the android system and radios while the other does user initiated tasks.
Im not really sure but that makes sense to me. It would be nice to have a definite answer on this though.
Click to expand...
Click to collapse
Nah..but in a synchonous cores both run at the same time. Asychronous cores one handles all the tasks and stuff, and only calls on the second one as needed. I don't know if this can be changed though. This is how the 3vo is.
Asynchronous just means each core can run at a different speed. It's much more efficient.
Product F(RED) said:
It has asynchronous cores lol. Not asymmetrical.
Click to expand...
Click to collapse
*facepalm* god... Yeah, I should have proof read it. Duh.
Anyways. from what I have read the Evo3D only really uses the second core when the first is over taxed. That's what I was getting at. Wouldn't it still be more effective to call on the other core sooner?
Are We Not Phones?
We Are D3VO!
Are we not phones?!
3.DeE.V.O.!!
Here, this is from my answer to another persons question.
View Full Version : asynchronous dual core vs others
18th June 2011, 08:46 AM
I have a question about the 3D's dual core that I'd like more clarification on the vague answers I'm getting by searching this site and google. So I've read that the core is asynchronous so basically meaning the second core doesn't do much work unless needed as others like the tegra 2 and exynos have both cores running or something similar to that, and that this is affecting the benchmark scores. I also read that one would basically double the score of the 3D to get a more accurate reading. Can anyone confirm or further explain this?
18th June 2011, 09:18 AM
Yes, asynchronous is when something operates on another thread whereas the main thread is still available for operating. This allows for better performance in terms of managing tasks. Now just because it doesn't score high on a benchmark, it doesn't mean it is going to perform. Also this allows for better performance for the battery.
I haven't slept for the past 12 hours so if this doesn't help you, just let me know and I will fully elaborate on how the processor will operate on the phone. Now time for bed :'(
In short, asynchronous operation means that a process operates independently of other processes.
Think of transferring a file. A separate thread will utilized for doing so. You will then be able to do background things such as playing with the UI, such as Sense since you will be using the main thread. If anything were to happen to the transferring file (such as it failing), you will be able to cancel it because it is independent on another thread.
I hope this makes sense man, kind of tired. Now I'm really going to bed.
Like my help? Thank me
Hi guys! I was just checking out this thread in the Bionic section and they have an overclock without that many prerequisites. http://forum.xda-developers.com/showthread.php?t=1439013All we need is a rom with init.D support and tekahuna's "symsearch.ko" and "opptimizer.ko" modules. DroidTh3ory already has init.D support in this Bionic rom so if we could get that in our roms and just find a way to apply tekahuna's modules, our device may be able to get an overclock. Let me know what you guys think. I'm not sure if this is possible but I just want to try and help the Razr XDA community. Sorry if this is arbitrary and completely unapplicable to our device.
Intriguing - conveniently, there's a thread on enabling init.d support that just popped up: http://forum.xda-developers.com/showthread.php?t=1444545
Sent from my XT910 using xda premium
Works on the atrix2
From what I read in original thread on rootzwiki Razr can gain only 50MHz addition with this.
theEnzy said:
From what I read in original thread on rootzwiki Razr can gain only 50MHz addition with this.
Click to expand...
Click to collapse
Ha, funny. Given the RAZR is factory overclocked to 1.2ghz, I suppose it's to be expected. I've underclocked mine to the 1ghz the chip is rated for to save juice anyway.
Sent from my XT910 using xda premium
razr is not overclocked versus the other omap4430 1.0Ghz devices. Parts are tested in the fab, and the ones that can do 1.2g without crash andd without heating a lot are selled to be 1.2-compatible. Other are only 1.0G approved.
It means also that overclocking a 1.0 to 1.2 may cause crash and excessive heat.
For the 1.2G parts of the razr, don't expect great overclock. the omap is more enclosed than in bionic and is already reaching the thermal policy barriere that underclock it when intensive use. If you clock the opp-max over this frequency, you will hit the barrier faster and then finish by using lower opps most of time...
greg_mp said:
razr is not overclocked versus the other omap4430 1.0Ghz devices. Parts are tested in the fab, and the ones that can do 1.2g without crash andd without heating a lot are selled to be 1.2-compatible. Other are only 1.0G approved.
It means also that overclocking a 1.0 to 1.2 may cause crash and excessive heat.
For the 1.2G parts of the razr, don't expect great overclock. the omap is more enclosed than in bionic and is already reaching the thermal policy barriere that underclock it when intensive use. If you clock the opp-max over this frequency, you will hit the barrier faster and then finish by using lower opps most of time...
Click to expand...
Click to collapse
Interesting, so not all 4430s are born equal? Is there an actual variant thereof rated at 1.2, or do they actually test every single chip individually? I wasn't implying it's unstable at 1.2ghz, and hardware design does of course come into it as you've said, but as far as I'm aware it's the exact same chip that TI rated at 1.0.
Perhaps 'clocked higher than the chip has been generally rated for by its manufacturer' is a more accurate description than 'factory overclocked'?
Sent from my XT910 using xda premium
hmmm.. this is all very interesting. I feel like the OMAP 4430 in the Razr should be at more capable than 1.2ghz. My OG Droid ran stable at 1.275ghz and it had a 3430 which was built on the 65nm chip process vs the 45 in the Razr. I must say 1.25gz as a max is quite underwhelming. I was hoping this bad boy would at least be capable of 1.5ghz.
MeNaCe2s0cieTy said:
hmmm.. this is all very interesting. I feel like the OMAP 4430 in the Razr should be at more capable than 1.2ghz. My OG Droid ran stable at 1.275ghz and it had a 3430 which was built on the 65nm chip process vs the 45 in the Razr. I must say 1.25gz as a max is quite underwhelming. I was hoping this bad boy would at least be capable of 1.5ghz.
Click to expand...
Click to collapse
I feel the same, but reality is cruel
I read some time ago on a tech website that the 4460 (1.5) and the 4430 were made from the exact same materials. The only difference is that the ones that hit a very high threshold at 1.5 were stamped 1.5. 4460's. It's not that the 4430's couldn't do 1.5ghz , most could. Just that they couldn't run the benchmark all the way to the end the amount of times needed to qualify. Odds are that any 4430 could do 1.5ghz and probably stable. We will never stress it out the way these benchmarks do.
Does anybody know if this hack could actually work and allow us to overclock our device. I think we can worry about how far we can overclock after we get an actual overclock lol. Would anyone be able to rewrite the "symsearch.ko" and "opptimizer.ko" modules for our device so we can get an overclock on our device?
MeNaCe2s0cieTy said:
Does anybody know if this hack could actually work and allow us to overclock our device. I think we can worry about how far we can overclock after we get an actual overclock lol. Would anyone be able to rewrite the "symsearch.ko" and "opptimizer.ko" modules for our device so we can get an overclock on our device?
Click to expand...
Click to collapse
No rewriting necessary. It was developed on a RAZR. Available in the download section of the OPPtimizer site. 0.1 modules only adjust top frequency. 0.2 modules adjust top frequency and voltage. As stated on the homepage, symsearch.ko is the creation of Skrilax_CZ. In case you don't know who that is, he's the guy who got 2nd-init functioning for all of us.
http://opptimizer.googlecode.com
---------- Post added at 08:21 PM ---------- Previous post was at 07:54 PM ----------
greg_mp said:
razr is not overclocked versus the other omap4430 1.0Ghz devices. Parts are tested in the fab, and the ones that can do 1.2g without crash andd without heating a lot are selled to be 1.2-compatible. Other are only 1.0G approved.
It means also that overclocking a 1.0 to 1.2 may cause crash and excessive heat.
For the 1.2G parts of the razr, don't expect great overclock. the omap is more enclosed than in bionic and is already reaching the thermal policy barriere that underclock it when intensive use. If you clock the opp-max over this frequency, you will hit the barrier faster and then finish by using lower opps most of time...
Click to expand...
Click to collapse
Very true, and something to consider the when overclocking OMAP4 devices. They don't clock as high as some would think. Although, I'm not so sure that the lag that most people are describing is the thermal policy kicking in. I've noticed that this lag can be cleared up with additional voltage, and not exhibit the behavior of getting clamped down on by the thermal policy. For example, 1250MHz on my Droid RAZR at stock 1.375V, is flaky, bump the nominal voltage to 1.388V, and it holds, and up to 1280MHz without the lagging condition.
You appear to be well informed on this subject. Good points you raise!
Lol. I'm still trying to understand the fact that we had a overclock this entire time. I never found one on the forums but I guess we had the tools available all this time. I honestly didn't know we had one.
Sent from my XT910 using xda premium
MeNaCe2s0cieTy said:
Lol. I'm still trying to understand the fact that we had a overclock this entire time. I never found one on the forums but I guess we had the tools available all this time. I honestly didn't know we had one.
Sent from my XT910 using xda premium
Click to expand...
Click to collapse
The kernel modules were only recently release. Some early testing was done in the LG Thrill community, as that was my original test device. Development was stalled a little by issues which I now realize were with the LG Thrill's kernel itself, rather than what I was doing. Got a RAZR on opening day and was able to come up with some code I feel comfortable releasing. It's only been available to the public for about a week now. I think some RAZR's will hit low 1300's... Don't know about much more.
Does this work? Is there a script for it? I have init.d support on stock from a bionic init.d cwr zip from droidrazr.com that works. What would a script look like to change voltages/ect?
Nevermind i see...its in the download section on link.
How do you run it or setup?
frostincredible on RootzWiki has created some flashable zip for Bionic, but they only work for my 0.1 modules that only support frequency manipulation. I have 0.2 modules that control voltage as well for Droid 3. You can open up his zip file and get an idea how to get things going with init.d... You can actually use his script, and just modify the echo line to say "echo 1122000000 1388000 > /proc/opptimizer", to send a voltage value, along with frequency.
http://rootzwiki.com/topic/14698-in...or-tekahunas-omap4-overclock-modules-1-10-12/
Page 2/3/4 on the official thread also has some info on init.d scripts for this.
http://rootzwiki.com/topic/14511-op...ng-kernel-modulesofficial-thread/page__st__20
CDjones over on DroidForums.net also started a good thread with some useful information:
http://www.droidforums.net/forum/dr...imental-voltage-control-support-tekahuna.html
Not the same device, but same idea.
orateam said:
I read some time ago on a tech website that the 4460 (1.5) and the 4430 were made from the exact same materials. The only difference is that the ones that hit a very high threshold at 1.5 were stamped 1.5. 4460's. It's not that the 4430's couldn't do 1.5ghz , most could. Just that they couldn't run the benchmark all the way to the end the amount of times needed to qualify. Odds are that any 4430 could do 1.5ghz and probably stable. We will never stress it out the way these benchmarks do.
Click to expand...
Click to collapse
This is not true. OMAP4460 and OMAP4430 are different chips. E.G. there are [email protected] and [email protected] the fact that the 2 products exists should be enough to say there are different.
I reckon they are not drastically different, most of the things are common.
So has anyone gained any noticable performance improvements with this? Also so it can 0nly be OC'd to 50mz more so 1.25ghz is the max? At this point is this mod developed enough for the razr for it to be a usable worth while tweak? Not dogging any devs just want to see if its worth it yet to start messing with this...
Also in the download section there is now this "MO_simple_spyder_1.5-beta-01.zip Milestone Overclock ported to Motorola Droid RAZR"
Whats that?
It depends on your particular processor, because every one is little bit different. I was able to run mine at 1300 MHz quite stable with stock voltage. But I don't see any need for more CPU power now, maybe someday
More interesting is undervolting to achieve longer battery life, which can be done with OPPtimizer too.
theEnzy said:
It depends on your particular processor, because every one is little bit different. I was able to run mine at 1300 MHz quite stable with stock voltage. But I don't see any need for more CPU power now, maybe someday
More interesting is undervolting to achieve longer battery life, which can be done with OPPtimizer too.
Click to expand...
Click to collapse
Has anyone had any good success with undervolting and battery life?
I do not yet have a Nexus 7 to test this with. You use this at your own risk!
This is the first real kernel work I've done, so don't be surprised if it doesn't work. I've only provided a boot.img as fastboot is easy enough to use on the Nexus 7.
Features (If it works):
CPU OC to 1.7GHz maximum
CPU over volt to hopefully reach 1.7GHz
GPU OC to 600MHz
Simple IO scheduler
SmartAssV2 CPU governor
The boot.img is attached. Source can be found at my github.
If anyone here has a Nexus 7 it would be very helpful to know if it works. I should have mine soon though. If it works well, enjoy. Feedback is always welcomed, as are benchmarks. Thanks.
Removed link until fixed!
This is scary looking, an untested Overclock that's never been run on the hardware before.
I'm guessing that the T30L is just a speed binned T30 and as such this shouldn't damage it. The same overvolt (and higher) has been successful on the T30 to get even higher clocks (1.8GHz). I would test this given hardware, however I don't yet have my Nexus 7.
ben1066 said:
I'm guessing that the T30L is just a speed binned T30 and as such this shouldn't damage it. The same overvolt (and higher) has been successful on the T30 to get even higher clocks (1.8GHz). I would test this given hardware, however I don't yet have my Nexus 7.
Click to expand...
Click to collapse
This sounds interesting. Especially the 600mhz gpu OC but may I ask if you are thinking of implementing some kind of app interface to control gpu clock and voltages etc like Extweeks on google play please? As I am guessing a lot of people won't be able to go to 600mhz stably, so a way to change the OC to something like 520mhz (to bring it to T30 speed) would be a good option
I've seen voltage tweaks controlled from userspace on other devices but not the GPU clock. I'd like to get it working first, then I guess I'll look at such things, especially if there is interest.
Cel1084 said:
This is scary looking, an untested Overclock that's never been run on the hardware before.
Click to expand...
Click to collapse
you go first!
bencozzy said:
Glados kernel on the galaxy nexus allows gpu oc control.
Click to expand...
Click to collapse
Siyah kernel for the galaxy s2 and galaxy s3 both let you control gpu clock speed and voltages, might be cool to have something similar
I don't even have my Nexus yet, and i'm already downloading things to flash to it, hahah. Will report back once Google ships to the US!
if this kernel works can we control the clocks with antutu or similar?
The CPU clock should be controllable, and I'm working on making the overvolt controllable. The GPU clock is not yet controllable, and I'm not so sure where to start on that.
ben1066 said:
The CPU clock should be controllable, and I'm working on making the overvolt controllable. The GPU clock is not yet controllable, and I'm not so sure where to start on that.
Click to expand...
Click to collapse
you should put the gpu clock to 520mhz arnt t30l 413mhz stock and kai would be even slower assuming its the budget tegra3 soc
Right, here's the thing. I've spoofed the SoC speedo ID to be that of the standard T30, however, without looking through with a fine tooth comb, it seems that the top that that id goes is 600MHz. In usage, it may be 520MHz, but I'm not sure. In addition I'm fairly sure these are just speed binned, and can probably run at the higher clocks if we just add a bit more voltage, or they get a little hotter. If anyone can tell me if this actually works, then I can adjust either way.
Look at tegra3_dvfs.c, line 256-262. It seems to indicate a maximum of 600MHz.
This should be helpful to you. tegra3 technical reference manual. everything there is to know about all variants of the chip. how it works, what its capable of, schematics, diagrams, chip layout, etc,,
http://db.tt/vWWou2Fu
Thanks but I already have access to NVidia's Tegra portal, which includes the TRM for Tegra 2 and Tegra 3. I'm hoping I shouldn't have to mess with it that low level
I don't understand why anyone would want to overclock a Tegra3, which is plenty fast enough already, especially when they have never even touched the device.
Also, I don't understand why anyone with any sense would use Simple IO scheduler, which has a higher latency and lower throughput than deadline, or even the bloat that is CFQ for that matter.
And finally, I don't understand why any real 'developer' would release something like this without testing it, especially with possibly dangerous overclocking and overvoltage settings. Only on XDA...
With all due respect, you should remove it until you have tested it *yourself* and confirmed that it doesn't make your Nexus 7 vanish in a cloud of smoke.
When I feel the need the need for speed owww.
_thalamus said:
I don't understand why anyone would want to overclock a Tegra3, which is plenty fast enough already, especially when they have never even touched the device.
Also, I don't understand why anyone with any sense would use Simple IO scheduler, which has a higher latency and lower throughput than deadline, or even the bloat that is CFQ for that matter.
And finally, I don't understand why any real 'developer' would release something like this without testing it, especially with possibly dangerous overclocking and overvoltage settings. Only on XDA...
With all due respect, you should remove it until you have tested it *yourself* and confirmed that it doesn't make your Nexus 7 vanish in a cloud of smoke.
Click to expand...
Click to collapse
Our Tegra 3 CPU is a lower clock version that the normal T30, it's the T30L. I have no doubt that this will not damage your device, the voltages used are still less than used by some TF201 ROMs (the TF201 uses the T30). I included Simple IO scheduler since it is something that seems popular, latency isn't the only thing that matters (read http://www.vincentkong.com/wiki/-/w...42041E#section-Android+IO+Schedulers-Deadline). I have seen benchmarks that show both SIO and deadline as better than each other, it depends what metric you record. I didn't remove CFQ, it's not that I've added it. The scheduler can be changed if you so desire anyway.
I have not provided a simple flash package and I've clearly stated in red writing that this is UNTESTED. I do not have the device, and it is yes untested however I didn't see the point on keeping something potentially useful private. If you have the knowledge to use fastboot to flash a boot.img, you probably know how to flash back the old one too.
_thalamus said:
I don't understand why anyone would want to overclock a Tegra3, which is plenty fast enough already, especially when they have never even touched the device.
Also, I don't understand why anyone with any sense would use Simple IO scheduler, which has a higher latency and lower throughput than deadline, or even the bloat that is CFQ for that matter.
And finally, I don't understand why any real 'developer' would release something like this without testing it, especially with possibly dangerous overclocking and overvoltage settings. Only on XDA...
With all due respect, you should remove it until you have tested it *yourself* and confirmed that it doesn't make your Nexus 7 vanish in a cloud of smoke.
Click to expand...
Click to collapse
seriously harsh man, just because you don't understand doesn't mean its wrong, or right for that matter
ben1066 said:
Our Tegra 3 CPU is a lower clock version that the normal T30, it's the T30L.
Click to expand...
Click to collapse
I take it you understand why similar chips are rated at various speeds for different devices? Because they are designed with a lower thermal output and / or the cooling characteristics / power characteristics of the device are different. The T30L has lower speed apps processors, lower speed GPU and lower speed memory. All in all, it will pump out much less heat than a T30.
I have no doubt that this will not damage your device, the voltages used are still less than used by some TF201 ROMs (the TF201 uses the T30).
Click to expand...
Click to collapse
You don't *know* that it won't damage someones device, you are assuming that it won't. The likelihood is that it probably won't, but would you stake your life on it? I wouldn't, and I've been doing Android kernel development for some time.
Also, this isn't the TF201, and it isn't the T30. It is a different device with different thermal characteristics and a different SoC, you can't compare them like that.
I included Simple IO scheduler since it is something that seems popular, latency isn't the only thing that matters
Click to expand...
Click to collapse
Latency of reads and writes and throughput are the only 2 things which matter (and I mentioned both), and SIO is poor at both of them. Justin Bieber is popular, but he's still ****, so including something which is popular isn't really a good reason.
---------- Post added at 06:44 PM ---------- Previous post was at 06:33 PM ----------
foxorroxors said:
seriously harsh man, just because you don't understand doesn't mean its wrong, or right for that matter
Click to expand...
Click to collapse
Harsh perhaps, but I prefer honest. Necessary, most certainly.
It is stupid and irresponsible to release something which is untested and potentially dangerous as it isn't fair on the poor muppet that flashes it and then f**ks their device up.
It has only been released because some 'developer' wants to make his epenis bigger by releasing something for a brand new device on XDA. Not that I am saying that he is the only one, there's plenty of others that do it, but as I have one of these on order I am taking an interest in these threads and was quite surprised with what I saw.
As someone who has done kernel development for some time now, I would never dream of releasing something I haven't tested thoroughly myself, or which I have got a trusted tester to thoroughly test, but hey, this is XDA and the standards are low.
ben1066 just out of curiosity may I ask how the gpu scales frequencies on the Tegra 3 t30l please? As I am used to the galaxy s2 and s3 where you have numerous frequency steps like 166mhz, 260mhz, 350mhz and 440mhz and you have an up and down threshold to govern whether you jump up or down the available frequencies, is this similar to how the gpu in works on the tegra 3 please?
Also when you say overclock the gpu, is it replacing 416mhz with 600mhz or is it adding an extra gpu frequency step after 416mhz, so 416mhz is still available to be used if needed? Sorry one last question, if the gpu does have frequency steps like other gpus, what ones are available for use please?
I am sorry to ask, I am just so curious about these questions, and I can't find them anywhere on the internet, so any help would be greatly appreciated. Thank you so much