Develop Bugs

[DEV]_JIT Libs from XXJH3

For all users that have XXJP3 Froyo release installed. (i don't have Froyo installed) Here is libs from XXJH3 for JIT.
Maybe they are more optimized, only for users that understand what it is.
For instructions read this, you don't need to replace build.prop. The tutorial is for Android 2.1 but it's same for 2.2
vmlibs from XXJH3
Freshly installed Froyo : better run 14.19 MFLOPS

setenza01 said:
For all users that have XXJP3 Froyo release installed. (i don't have Froyo installed) Here is libs from XXJH3 for JIT.
Maybe they are more optimized, only for users that understand what it is.
For instructions read this, you don't need to replace build.prop. The tutorial is for Android 2.1 but it's same for 2.2
vmlibs from XXJH3
Freshly installed Froyo : better run 14.19 MFLOPS
Click to expand...
Click to collapse
have you tried this?

Does this speed up the phone or only the benchmarks?

Ok, I've just tried it.
Unfortunately, it gives me the same result.
Nearly, 14.5 MFLOPS with Linpack.
It was a nice try though.
No problems with these files as well.
I'm gonna have a look if it makes the phone be faster or not.
edit : With Quadrant advanced, it brings me ssssssslightly higher cpu score.
just about... 50 points more, I mean, in total score of 1900 or something (with ext2 fix)
Not really so much good but it's worth to have a try.
At least, we've found it out that we can get some files from Galaxy Tab.

Does it work on 2.1 ?
Sent from my GT-I9000 using XDA App

No, it doesn't. I'm using xxjp3 firmware ATM.
Sent from my GT-I9000 using Tapatalk

Mflops: 14.34

does anybody who tried this notice a difference in stability?
read: still so many FCs with this new JIT?

Maybe my question will sound too lame, but, did anyone try to use JIT libs from Nexus One? The procesor core is quite similar, and the instructions set is the same, so maybe it'll work?

I have. It ran at 15 mflops
Caused a LOT of fc's though
Sent from my GT-I9000 using XDA App

nm ignore my post

14M flops vs 3x Mflops the nexus one ... something goes wrong, do not you think?

braintheboss said:
14M flops vs 3x Mflops the nexus one ... something goes wrong, do not you think?
Click to expand...
Click to collapse
The first jit for nexus had similiar results. So wait for a proper release....

emelie said:
does anybody who tried this notice a difference in stability?
read: still so many FCs with this new JIT?
Click to expand...
Click to collapse
http://forum.xda-developers.com/showthread.php?t=736881&highlight=jit

DocRambone said:
http://forum.xda-developers.com/showthread.php?t=736881&highlight=jit
Click to expand...
Click to collapse
thx for the heads up, didn't notice that thread before.
so in a nutshell the JIT provided in the thread you referenced vs the JIT provided in this thread:
- still low results with benchmarks
i could care less about benchmark numbers. no offense and really thank you for pointing me to that thread!
what i'm looking for though, and this is mentionend in the thread you referenced, is stability of other apps which are used daily compared to benchmark apps.

emelie said:
thx for the heads up, didn't notice that thread before.
so in a nutshell the JIT provided in the thread you referenced vs the JIT provided in this thread:
- still low results with benchmarks
i could care less about benchmark numbers. no offense and really thank you for pointing me to that thread!
what i'm looking for though, and this is mentionend in the thread you referenced, is stability of other apps which are used daily compared to benchmark apps.
Click to expand...
Click to collapse
The current JIT is not stable for daily use. It was same for the first JIT implementations for Nexus and Desire.
Lets hope the will be a better froyo rom soon...

I am sure that linpack score will never be as high as we see today on N1. Maybe 20+ not higher.
If im wrong and with official 2.2 fw update score is 40+ then I will give away my phone for free

fua said:
I am sure that linpack score will never be as high as we see today on N1. Maybe 20+ not higher.
If im wrong and with official 2.2 fw update score is 40+ then I will give away my phone for free
Click to expand...
Click to collapse
My guess we see higher scores than 50+ in linpack

If samsung get their act together we will see 50 mflops. If not then 15 - 20
Sent from my GT-I9000 using XDA App

I'm starting to think the hummingbird might disappoint in performance with the JIT; Samsung might not be able to optimize the JIT as well as Google optimized it for the N1 (and all the other snapdragon-based devices). Look at Motorola, they just released the Droid 2 running a 45nm SoC not too different from the hummingbird and they only managed to get linpack up to 15Mflops with Froyo... I wouldn't think Motorola would do an official release of Froyo before optimizing it (but who knows, maybe pressure from Verizon to have a new device with Froyo before other carriers)
I doubt it, but there's also the possibility that there's something advantageous for the JIT in the Snapdragon architecture that is lacking in the OMAPs and Hummingbird architecture. I guess we'll see soon enough, we just have to hope for the best.

Leedroid but what kernel

Hi all,
I've got Leedroid and would like to know how the hell people get the 3000+ scores on quadrant ?
I've used most of the roms with a lot of the kernels but the most i get is around 2500 - 2600 , i've heard that the kernel by Kamma is a good one but i'm at a loss as to which one i should use.
So if anyone could point out which one i could use for .
1. Stability.
2. Battery saving.
3. Benchmarking (without smoking my device of course)
Any answers would be appreciated.
Thanks.

dladz said:
Hi all,
I've got Leedroid and would like to know how the hell people get the 3000+ scores on quadrant ?
I've used most of the roms with a lot of the kernels but the most i get is around 2500 - 2600 , i've heard that the kernel by Kamma is a good one but i'm at a loss as to which one i should use.
So if anyone could point out which one i could use for .
1. Stability.
2. Battery saving.
3. Benchmarking (without smoking my device of course)
Any answers would be appreciated.
Thanks.
Click to expand...
Click to collapse
the first thing you need to understand is quadrant is ****, it is not an acurate indication of the speed of your device, a few devs have made kernels that trick quadrant into giving it massive scores, but in reality there kernels are not the best that being said, you wont find a better kernel for stability or speed than leedroids one, it is flawless you can download it here
http://www.multiupload.com/U2ECLTRT7V

AndroHero said:
that being said, you wont find a better kernel for stability or speed than leedroids one, it is flawless you can download it here
Click to expand...
Click to collapse
Have you noticed much difference between 2.0.5 and 2.0.6 kernels? The LeeDroid thread is getting very lengthy to read through these days to catch up on new updates lol!

RobSimmo said:
Have you noticed much difference between 2.0.5 and 2.0.6 kernels? The LeeDroid thread is getting very lengthy to read through these days to catch up on new updates lol!
Click to expand...
Click to collapse
Yeah the new one is based on the latest kernel source, with the 1.75 libs,
Sent from my Desire HD using Tapatalk

Tbh I use leedroids newest one an its really good, quadrant gets 2500 +
The device is smooth as silk, bit then again it always was.
I didn't know that quadrant was pants, I thought it was a good benchmarking tool.

Confirm that the newest kernel of Lee is the best!but in a few days he prepare a new awesome BFS kernel...

Mmm badly ****ed scheduler. My favourite.
If you want good quadrant, buy the paid app which gives higher scores than the free one. Do the I/O hack which core and lee droids use. Use Leedroid kernel at max freq, performance governor, and finally run it in airplane mode.
Follow all those steps and you will get around the 3000 score.
You should state if you use quadrant standard or advance when posting results as they do differ. Otherwise it's like comparing oranges to mandarins. They are both orange but one is always bigger.
Sent from my super slick Android device.

dr.m0x said:
Mmm badly ****ed scheduler. My favourite.
If you want good quadrant, buy the paid app which gives higher scores than the free one. Do the I/O hack which core and lee droids use. Use Leedroid kernel at max freq, performance governor, and finally run it in airplane mode.
Follow all those steps and you will get around the 3000 score.
You should state if you use quadrant standard or advance when posting results as they do differ. Otherwise it's like comparing oranges to mandarins. They are both orange but one is always bigger.
Sent from my super slick Android device.
Click to expand...
Click to collapse
Yea it's just the standard version of quadrant, I thought that it would do the job.
Also i don't want to trick it or actually buy an app just so it can give me higher scores, that's not actually helping at all, i'll try the airplane mode to see if that helps, my phone is rapid anyway so it's only for my own satisfaction and to see how high i could get the score to go to.

Low benchmark score after root/custom rom and kernel

Hi!
First of all, I have been reading around and I did not find exactly what I want to know.
Yesterday I managed to root my desire and flash the CM7 ROM (http://forum.xda-developers.com/showthread.php?t=957344). Today I've added a custom kernel, "vork" 720p (http://forum.xda-developers.com/showthread.php?t=782875&highlight=vork).
My quadrant score is still low (around 1100), even though the phone behaves much better than the stock.
I have 2 questions:
1. What can I do to make it perform even better? (i have a friend with a desire hd who just flashed a sense rom, without any added stuff and scores over 2200).
2. should i install a new "radio"? My gsm signal is much stronger than before as it is.
3. Since this is my first time, what do you think of my choices with the ROM and Kernel? I chose them because they were popular and i assumed they are well tested and function well.
Thank you for your time.

Desire HD has better processor and GPU I believe hence bigger difference. Your score is good for Desire. I got around that overclocked on LeeDroid GB ROM.
Sent from my HTC Desire using XDA App

why do you even care bout the quadrant score ? :X
if you really feel excited bout high numbers having nothing to do with ur phone performance, go with data2ext roms, quadrant gives them 3000+ points, but still they perform worse than, for example, LeeDroid.
then again, new GingerBread Sense 2.1 roms have low quadrant scores, but they are really really fast.

I dont really care about quadrant, but it seems like a valid tool for measuring performance. At least it seems that everyone is judging by that. Quite frankly the phone works very well, i am just out of ideas to make it even better. And if the scores are meaningless why do people use it?

Its probably one of those my d...ck is bigger than yours things. Anyway I used data to SD and I also had high scores but overall experience was very laggy and jerky
Sent from my HTC Desire using XDA App

Quadrant scores are mesaurable only when you launch them on stock, unrooted roms. then it can give you and idea on how this phone with this software will perform.
after rooting, some hacks or scripts make quadrant scores not real

So this means my phone is fine as it is? There is nothing more I can do to enhance it?

you can overclock it, it should all be said in your ROMs topic.
like "OC up to 1200Mhz" means you should be able to set processor speed to 1200mhz with SetCpu. it all depends on kernel i think

Ok, thanks everyone for your input! If there are any more suggestions, I dont mind.

placebo effect

this is basically a review on custom roms that ive tried...
my purpose of doing so is one to guide newbiez on sgy like myself
to give a background i was on galaxy 551 before switching to the sgy was running on custom roms myself i am no developer just an end user
currently i am running on stock rom with a couple build prop tweaks , audio tweaks, and customized layouts...following hhubs...
currently my device scores around 16xx on antutu benchmark and around the aame figure on quadrant....
yesterday i tried the hybrid 2.0
ive had high expectations on such custom roms specially the most talked about....and recommended roms...
now...ive installed hybrid which was a light and yeah fast rom....
well i couldnt feel the difference from my stock rom...
so i decided to run a benchmark test antutu and quadrant......
gave me the same score as my stock
16xx
but the thing was the rom was fast and lite almoat the same as my stock rom edited.....
if you have a different review please do post so that others can get a fair.review....
will try to review termas rom next week.....
--------------------------------------------------
termas review will be postponed...pc went down..cant flash his kernel...
--------------------------------------------------
adding aurora v2
pretty decent everyday rom
antutu gave a 1704 on performance governor...
i like the retro look..
would be giving it a rate of 9/10
for acouple of graphic glitches that are and will be fixed soon acocrduing to dev..
will write more later..
Sent from my GT-S5360 using Tapatalk

forgot to post working and bugs
everything was ok..except for the message about my balance...
the rom was overall a good rom with a sense ui feel.....
Sent from my GT-S5360 using Tapatalk

if you think that quadrant will show you your phone's actual performance then you re wrong.If you want hi scores on quadrant try searching for quadrant hack
I cant understand some mobile phone owners like you.
What do you expect from your phone?Magical abilities or something?

<post deleted>

ronnieryan said:
now im confused
Hybrid 2.0 doesnt have sense ui feel
Hybrid 2.1 has the sense ui feel
and then again my antutu score with hybrid 2.0 would be 18xx to max 19xx
pls verify what hybrid did you use 2.0 or 2.1?
Click to expand...
Click to collapse
My apologies 2.1 for the sense ui..2.0 was running on touchwiz
Hmmm...good score for you mate
Mine was disappointing really
Sent from my GT-S5360 using Tapatalk

ronnieryan said:
now im confused
Hybrid 2.0 doesnt have sense ui feel
Hybrid 2.1 has the sense ui feel
and then again my antutu score with hybrid 2.0 would be 18xx to max 19xx
pls verify what hybrid did you use 2.0 or 2.1?
Click to expand...
Click to collapse
The sense UI is not that good! I have tried it on my sgy and i accept that it was fast but I removed it because of its buggy interface especially when the name of the app cuts in the dockbar.
Sent from my GT-S5360 using XDA

<post deleted>

[email protected] said:
if you think that quadrant will show you your phone's actual performance then you re wrong.If you want hi scores on quadrant try searching for quadrant hackI cant understand some mobile phone owners like you.
What do you expect from your phone?Magical abilities or something?
Click to expand...
Click to collapse
i didnt say that i was basing my opinion on quadrant scoring...
i just mentioned it...
for reference i guess not really for anything..
for me i was expecting a lil bit better...
from a custom rom...
not just better aesthetics..
this is not my first phone and certainly not my first rom..
thou it might be true that performance cannot be measured by benchmarking...
but how do you measure or know if something is better that the other without benchmarking...
benchmarking is there for a reason,,,,
as for experience my galaxy 551,stock rom..almost the same edits that i have now,,except with a lower cpu of 600mhz..
scores right about 14xx
and with a custom rom from arkas right about 17xx
with a custom from perseus,,right about 18xx-2xx depending on setting made..
stock is good but a bit slow
roms from arkas are fast..
from perseus,,,hellluva fast..but battery...dead in 1-2 hour,...oc to 729..
why am i saying this,,,
custiom roms to my view should atleast give something that isnt offered in stock..
like a2sd from terma...
as far as cheating the benchmark..
why would you conduct a cheated benchmark in the first place,...like playing a good game but using a cheat...useless..
taking the exams and cheating...
Definition of BENCHMARK
1usually bench mark : a mark on a permanent object indicating elevation and serving as a reference in topographic surveys and tidal observations
2a : a point of reference from which measurements may be made b : something that serves as a standard by which others may be measured or judged c : a standardized problem or test that serves as a basis for evaluation or comparison (as of computer system performance)
See benchmark defined for English-language learners »
See benchmark defined for kids »
Examples of BENCHMARK
a stock whose performance is a benchmark against which other stocks can be measured
<this prize-winning biography will be the benchmark against which all others will be judged in future years>
First Known Use of BENCHMARK
circa 1842
http://www.merriam-webster.com/dictionary/benchmark
btw..it would be cool if the phone transforms

guys i believe that ok maybe benchmarks are not perfect but they can show if a rom is faster than another.if your benchmark shows higher score that mean that something goes faster.but for better resaults you must repeat the benchmark for 4-5 times and from all scores get those that are more "stable" and most repeated.for example:if you run 5 times quadrant and get these results:
1050
1097
1129
1125
1130
take your score as around 1127.most times the first test shows lower result.
as for our custom roms i agree with OP but as long as we cant have a custom kernel,this is almost the best we can do.we can do much much more things and more ads and more tweaks and generally more speed for our roms if we had a custom kernel.but i think that all guys deserve our respect and thanks just only for the time their spend trying to do something better for our phone(this better could be faster phone or more fuctional phone or just more beautiful phone.whatever anyone can do).so i respect the opinion of OP whatever parameters he use.he flashed the roms,so he use them,so he can say his experiences!

Defy+ = high frequency roundkicks or just serious OC

Hi,
Im trying to get a bit of rock ´n roll out of my Defy+ thru serious overclocking,
im running MS2Ginger 3.0 at the moment that doesnt seem to be the perfect rom for getting great speeds, but it works.
Been running 1350Mhz for a while which seems to work rather good, going to 1400 makes my Defy more sluggish than a led slug. So that doesnt work, atleast not for what i can see.
The problem i see is when running Quadrant benchmarks, i only get numbers around 300 on the 2D test, is that normal?
Can someone tell me about a rom that could make me roll even faster and higher up in the result lists of Quadrant benchmark?
Regards.

So 214 views and nobody knows anything? Sounds impoosible on this forum

Don't know much about fast ROMs but everybody brags about CM7, of course. Also there are some people who claim Mokee os (CM7 Based) is the fastest. Check them out and let us know what you find out
Anyway, they say Quadrant isn't everything.
Sent from my Defy+ using xda premium (app needs update!)

fiskenigaten said:
Anyway, they say Quadrant isn't everything.
Sent from my Defy+ using xda premium (app needs update!)
Click to expand...
Click to collapse
exactly...
if you want fast delete all apps except for system ones...
& lob it off a tall building ^^ :chortle:

fiskenigaten said:
Don't know much about fast ROMs but everybody brags about CM7, of course. Also there are some people who claim Mokee os (CM7 Based) is the fastest. Check them out and let us know what you find out
Anyway, they say Quadrant isn't everything.
Sent from my Defy+ using xda premium (app needs update!)
Click to expand...
Click to collapse
Ok thanks for the advice, i'll check them out and post some kind of measurments
King_Rat said:
exactly...
if you want fast delete all apps except for system ones...
& lob it off a tall building ^^ :chortle:
Click to expand...
Click to collapse
Ok they say the Defy is a rugged phone but hell knows if that would be healthy for it. But as they say it aint the speed that kills its suddenly coming to a stop that kills so maybe i should go for a try

I recommend setting max and min to 1350 and also performance governor, on v6 supercharger before benching run an enginge flush and kill all apps using process manager except quadrant
DEFYing the Android planet with ICS

Ok i´ll try that V6 thingie but before that i thought i would tell what i have noticed so far.
MS2Ginger stayed around 1900 score on Quadrant with a top of 2100.
CyanogenMod seems to stay around 2400 with a peak of 2501 so far.
CyanogenMod seems to be better on the hardware tests except on the graphical parts where it seems to drop a couple of FPS, Anyone know any good
advice to maximize the graphical parts of CM?
A fun thing to notice is that with the original settings and MS2Ginger i achieved a score of 1350 and now riding around 2500, pretty nice.
One down thou with CM is that i woke up this morning without a statusbar and when the alarm went off i couldnt shut it off except for rebooting the phone.

mihovil13 said:
I recommend setting max and min to 1350 and also performance governor, on v6 supercharger before benching run an enginge flush and kill all apps using process manager except quadrant
DEFYing the Android planet with ICS
Click to expand...
Click to collapse
Ok now im un-supercharged again, either i did something very wrong or that script only worked in the wrong direction on my Defy+. Installed it and ended up running 1700 - 1800 scores on Quadrant, uninstalled it and is now back to 2300, dont know where the last 100 - 200 points ended up, will do some more tests before moving on to the Mokee OS rom.
Have a safe day/evening

I jujst wanted to say that im in no way is trying to put down the MS2Ginger developers work, this is just me and personal opinions in the hunt for a better performing Defy+
Cheers

in my opinion, oc-ing the defy to over 1.2ghz makes no sense and it only damages your cpu in the long run. i think that managing your ram gives a lot smoother experience than oc-ing your phone, because the only apps that dont run smooth on this little beauty are hd games, but they were never intended to. im using ram manager pro from market (play store), try it, maybe it helps!
sent from my cm7 defy...