Develop Bugs

[Q] "Virtual ext filesystem" for desire to speed up I/O?

First of all sorry if this is in the wrong section, i thought it might go in discussion but its more a question:
I own an HTC Desire but was recently looking on the net about the Samsung Galaxy S as a friend of mine was after an android phone, and discovered this article on BriefMobile.com:
Samsung Captivate Reaches New Speeds
So, remember when we told you that Android developers were working on a fix for the Samsung Galaxy S’ I/O read-write problems? Remember when we said that we could see these Galaxy S phones reaching 3000+ Quadrant benchmark scores in no time? Well, buckle up folks, because we’re well on our way.
Today, users tried out a new modification for the Samsung Captivate that “creates a VIRTUAL EXT2 filesystem inside the stock RFS filesystem on the internal SD card” of the Captivate. The so-called lag fix boosts I/O scores by around 800% according to the Quadrant benchmarking application.
And, just days ago, users over at XDA-Developers saw the first overclocked kernel for the Captivate. Made by developer AJerman, this 1.2 GHz overclock kernel will boost your performance by around 15-20%.
With these two fixes in place, we saw consistent 2500+ Quadrant scores, far surpassing the capabilities of any other phone to date. The Motorola Droid X has recently been overclocked to allow for ~1350 scores on Quadrant.
Unfortunately, the overclock kernel contains one major bug: a wake-up lag that stalls the phone for three and a half seconds before waking up from standby. And, the ext2 fix is nowhere near perfect. The ext2 filesystem is far too unstable according to some developers who prefer to execute this fix using ext3 or ext4.
Still, it’s amazing to think that we now have phones that are benchmarking scores twice as high as the Nexus One with Android 2.2 Froyo’s JIT compiler. This phone should fly on Froyo.
Click if you want to learn more about the ext2 hack or the 1.2 GHz overclock kernel.
Click to expand...
Click to collapse
source:briefmobile.com/samsung-captivate-reaches-new-speeds
i thought this was amazing news to be able to boost a phones performance so much and was wondering if this virtual filesystem hack could be implemented on the desire? Or has anyone actually alreaady tried it? I am aware the desire supports ext partitions in hacked roms but i am not aware of any roms that have implemented the ext filesystem in this way

thekeiron said:
i thought this was amazing news to be able to boost a phones performance so much and was wondering if this virtual filesystem hack could be implemented on the desire? Or has anyone actually alreaady tried it? I am aware the desire supports ext partitions in hacked roms but i am not aware of any roms that have implemented the ext filesystem in this way
Click to expand...
Click to collapse
I suspect you will find that this only works on the Galaxy S because of the RFS filesystem. Since the Desire doesn't use RFS, there would be no benefit.
Regards,
Dave

Ah i see that is a bummer! thanks for the reply anyway

[Q]Samsung Galaxy S "lag fix" for Inc?

Has anyone looked in to this to see if it is something that we could possibly port over/modify to work on our phones and benefit from?
http://forum.xda-developers.com/showthread.php?t=749495
I did a search and didn't see any topics started on this so I figured it was worth asking.
It increased the quadrant scores tremendously for the Galaxy S.

Non applicable.
Galaxy class phones have serious lag problems due to installation of apps on slow sd.
Sent from my ADR6300 using XDA App

If someone wants to compile a new kernel with the "fix" implemented, sure. You would need a class 6 sdcard to see any benefit from it though. The EVO guys are doing it and seeing numbers over 2000. There's no reason it wouldn't work. Though, you probably wouldnt notice any practical difference in the phone in regular use. I'm assuming the quadrant increase is due to the read write speeds being much higher using a card that fast, thus increasing the I/O score on quadrant.
What people don't seem to realize is that quadrant score is an average of the different tests, and not just a blanket score to use for comparisons...on the galaxy s it does fix the lag problem they have. Watch one do a quadrant test. The I/O test takes over a minute. A nice aide effect of the fix is that it inflates quadrant scores. The incredible is far from laggy, so you'd basically be doing it just to see a higher quadrant score.
You'd also have to figure out a way to use the Amon-RA recovery instead of clockwork, as only it can partition the sdcard in the correct way to implement this...
I'd actually love to see people try more things with the kernels, the ones we have are nice, but pretty basic. The EVO guys also have one that actively adjusts the cpu voltage based on speed and temperature, it's supposed to really increase the battery life. I guess that's probably due to the EVO not having the same supply shortages, and the fact they gave out a ton of them free to developers at the google conference, lol.

[q] Livewallpapers

Hello!
I-m trying to apply several live wallpapers. Some crash, its acceptable.
Others, plain buggy. Example: Galatic Core.
I select it in the Livewallpapers selector, it does not error, and shows a black screen. i press settings, get the message "buy the app", and then it shows!
If i then select it, i get a black screen... and then the original wallpaper again.
Anyone got more luck than i ?

They were never meant to run on our phones so they will be iffy at best for performance
Also if your using cm6 that's a known issue across the board with them
Sent from my HERO200 using XDA App

thoughtlesskyle said:
They were never meant to run on our phones so they will be iffy at best for performance
Also if your using cm6 that's a known issue across the board with them
Sent from my HERO200 using XDA App
Click to expand...
Click to collapse
Could you guys please stop doing that ?
"" XPTO was not meant to run on our phone ""
What is it there... that our phones have... or have not... that make a given app not run ? Sure... we can agree on "it wasn't designed for 320x240. Ok, i agree too.
But either than that ?? Yeah, our CPU is the slower MSM on the field, at least compared to Diamonds/Raphaels and upwards... but hey!! We have 1/4 of the pixels to take care of. That sould give us some room...
Now, don't get me wrong. It was not designed to run Android. But that doesn't mean it was not meant to. Take for example a simple little phone, sold here on Portugal as Optimus Boston. It has a MSM cpu... it came with 1.6 running at 600mhz... but the latest update to Eclair 2.1... underclocked it to 480. My kaiser usually runs at 550. (official rom, its possible to have custom roms running at the original 600.) and it still is a great phone... look at the simple specs :
http://www.gsmarena.com/gigabyte_gsmart_g1305_boston-3201.php
Now, someone explain... why can't our Kaiser/Vogue/Polaris run Android like that leatle freak underclocked to 480 runs. Please.

The fact that we have absolutely no RAM doesn't play into it ever since sheer clock speed always determines speed, to be honest I haven't even seen them run well on my hero and I have it OC'd to 691
But hey if you want to argue about it there are links to the source around go in and fix the problem, most of the devs for this project have more important things to focus on at the moment
Sent from my HERO200 using XDA App

thoughtlesskyle said:
The fact that we have absolutely no RAM doesn't play into it ever since sheer clock speed always determines speed, to be honest I haven't even seen them run well on my hero and I have it OC'd to 691
But hey if you want to argue about it there are links to the source around go in and fix the problem, most of the devs for this project have more important things to focus on at the moment
Sent from my HERO200 using XDA App
Click to expand...
Click to collapse
Do not feel ofended. It was not my intention.
My point is that if we had accepted the slowness of our devices, we would never had attempted to port Android.
As for the RAM, agree.. it is a bottleneck...
Now, has for speed... i've seen them run well... on other devices of course.

I did many tests on my Kais130 Fresh Froyo, about livewallpapers
I agree with Daedric on the ability of our device.
I managed to run the "Grass" Wallpaper properly. Others, like Mario, Galactic Core, ... Sometimes with bugs and/or too many resources used.
It seems to me that LiveWallpapers work less well in the latest Froyo

Well were not the slowest android phone anymore. But that doesnt mean much, we still have a lot of instability. The ram limitation can be overcome with comp cache (virtual memory) but that brings only more instability and adds overhead that we don't need. Also the performance just isnt there yet. Our phones beat only one phone and thats because that phone doesnt have froyo on it. Otherwise it would beat ours out. Also that was with my kaiser overclocked to 572 mhz with a gpu overclock as well. Any other phone msm 7200 phone at those speeds kick our kaisers ass. I think the resource intense ones are just not meant to work well at all for us. My background works at least lol, I use a background that shows the core of a 486 cpu.

aceoyame said:
Well were not the slowest android phone anymore. But that doesnt mean much, we still have a lot of instability. The ram limitation can be overcome with comp cache (virtual memory) but that brings only more instability and adds overhead that we don't need. Also the performance just isnt there yet. Our phones beat only one phone and thats because that phone doesnt have froyo on it. Otherwise it would beat ours out. Also that was with my kaiser overclocked to 572 mhz with a gpu overclock as well. Any other phone msm 7200 phone at those speeds kick our kaisers ass. I think the resource intense ones are just not meant to work well at all for us. My background works at least lol, I use a background that shows the core of a 486 cpu.
Click to expand...
Click to collapse
Should we assume... that at the same speed, diferent MSM devices offer diferent performance ? Or, are we simply doing still things wrong, i don't know, perhaps a poor schedule, buggy drivers which introduce lag, perhaps the graphic one.
We must remember, they kick our kaiser hard, but they have much more pixels, how can that be ?

V6 Supercharger

I am an Eris user, soon to be DINC2 owner at the end of this week. I've been reading through the threads here getting ready for my purchase and it has been very informative.
I searched the DINC2 forums and did not see this mentioned. Has anyone tried zepplinrox' V6 Supercharger script on the DINC2? I'm not going to regurgitate his post verbatim, but it is a memory optimizer. It was introduced to me over on the Eris forums, in the GSB ROM thread. It really makes a difference on the Eris, but I'm not sure how much difference it would make on the DINC2.
http://forum.xda-developers.com/showthread.php?t=991276
He also has another script called 3G Turbocharger. This will probably make a difference on the DINC2. According to zepplinrox, "the 3g script reconfigures the build.prob and kernel to change the speed settings to a better enhanced setting. Think of it like a TCP optimizer on windows." On my phone I have noticed an increase in speed surfing the web, downloading apps and in other 3g activities. It may be placebo, but it does seem faster. If nothing else, I do plan to add this to my DINC2 when I get it.
http://forum.xda-developers.com/showthread.php?p=15948434#post15948434
There is also a Kernel Tweaks script on the above link as well. I am using it too, but not sure what performance gains I am getting.
Per zepplinrox' post, this should work on stock and rooted ROMS.
I am by no means an expert on this, so don't shoot me if this is in the wrong forum, not valid on the DINC2, or a repost. Just trying to pass along something that works well for me.

For now those scripts are useless for phone that came preloaded with GB 2.3.3.....no root no turbo.

This sounds interesting to me, I am rooted, would it work?
Sent from my ADR6350 using XDA App

This wouldn't really benefit the Inc2. The Eris was a very memory limited device designed to run Android 1.5. Upgrading beyond this was pushing the Eris to its limits, and in some cases, beyond. Running newer versions of Android would make memory management more difficult because of the higher memory requirements and the very limited memory.
The Inc2 has one of the bigger pools of memory available for Android devices. It has 768mb of total RAM when many current generation devices are getting 512mb (others are either getting 768mb or 1gb). Unless you are running a LOT of background apps, or somehow managed to find apps that require massive amounts of memory, this is plenty of room.
What does it do? It tunes memory settings and task manager. That's pretty much it. It could have an effect on the Inc2, but I wouldn't really consider it a necessity.
Edit: BTW, my previous device was an Eris.

Well, Desire HD users with 768mb of ram enjoy improvements.
It even improves performance of an Atrix with 1gb or ram...
So it probably will help.
So "It tunes memory settings and task manager" is an understatement.
It revamps the whole memory thing

ben805 said:
For now those scripts are useless for phone that came preloaded with GB 2.3.3.....no root no turbo.
Click to expand...
Click to collapse
Actually, I think these will work without root. I think the only drawback without root is the scripts do not stick on boot so you have to re-run on each boot/reboot.

nimdae said:
This wouldn't really benefit the Inc2. The Eris was a very memory limited device designed to run Android 1.5. Upgrading beyond this was pushing the Eris to its limits, and in some cases, beyond. Running newer versions of Android would make memory management more difficult because of the higher memory requirements and the very limited memory.
The Inc2 has one of the bigger pools of memory available for Android devices. It has 768mb of total RAM when many current generation devices are getting 512mb (others are either getting 768mb or 1gb). Unless you are running a LOT of background apps, or somehow managed to find apps that require massive amounts of memory, this is plenty of room.
What does it do? It tunes memory settings and task manager. That's pretty much it. It could have an effect on the Inc2, but I wouldn't really consider it a necessity.
Edit: BTW, my previous device was an Eris.
Click to expand...
Click to collapse
It may not, but zepplinrox posted above about the Desire HD with similar RAM and the Atrix with more that reported improvements.
Whether you use the Supercharger script or not, the 3G Turbo is definitely worth trying. Again, it may be placebo, but I think my up/download speeds are faster after the script. Here is a post with some info on speed improvements with the script.
http://forum.xda-developers.com/showpost.php?p=15959134&postcount=1139

I use the Mega memory option with the v6 script. I notice a huge difference in performance.
Sent from my Incredible 2 using XDA App

Has anyone tried it on Virtuous Unity?

Stam2000 said:
Has anyone tried it on Virtuous Unity?
Click to expand...
Click to collapse
I second this... I am intrigued...
Sent from my ADR6350 using Tapatalk

3g turbocharger didnt increase anything to my 3g speeds.

MJL99 said:
3g turbocharger didnt increase anything to my 3g speeds.
Click to expand...
Click to collapse
same here. i've actually tried the v6 superchgarer and saw no improvements. actually, the only thing that happened was my battery life took a hit.

lol, well that's no good.

Why is this? Quadrant score

Hello. I have flashed the Synergy kernel for CM7 and I have CyanogenMod Self-Kang. I am overclocked to 1.8Ghz and have the lag free governor. Why is it that when I run a benchmark I get such low results? Take a look:
I've seen people getting around 5000 on this device! Is there a reason why it is under 4000?
Sent from my SGH-T989 using xda premium
Edit: OK thanks. I am not going to post anymore and I encourage other people not to (by not post I mean in this "Why is this" forum. I'm sorry I posted in the wrong place...

first, wrong forum. Should put this in Q&A
second, Quadrant results means nothing, my phone is giving me 3k and it doesn't lag what so ever.

I've never hit 5000 with my phone I'm always in between 3000-4400. Plus benchmark scores mean nothing. Even if the device is not hardware accelerated it will still be lag free. Not to mention this post should of been in the "General" section. It is not about development. Re-post your question in General, you might get more answers.

Yo just because 1 person got away with posting in the wrong forum yesterday does not mean you can ask questions here. This is why we have different sections in the first place. Follow the rules of the forum and do your research because if you did you would have found out that quadrant means nothing.
_Thursday

STOP USING QUADRANT!!!
Use a program that has been updated like Antutu.
Quadrant has not been updated in over a year. Let it die for the sake of human kind.
Sent from my SGH-T989 using XDA App

Quadrant really doesn't do a very good job of true benchmarking, and it's WAY too easy to fool it. Some very simple hacks can be done to artificially increase the score by a large amount. For example:
1. enabling stagefright by default in the build.prop can do this (on some stock roms.) This will make it so the h.264 encoding test will be falsely inflated.
2. Tricking the system into using a ramdisk for it's IO test instead of the actual nand. This one should be self explanitory.
3. On the old galaxy phones, that used rfs for the filesystem, if converted to EXT4 it would give a HUGE boost, but no real world performance difference. (Quandrant doesn't seem to read rfs correctly.)
And believe me, there are PLENTY of others. No to mention its not optimized for more than one core.
Benchmarks mean nothing. Real world usability does, unless it's just an e-peen thing, my benchmarks are better than yours type of thing. It's kind of like overclocking. I've NEVER seen the point in it (on a cell phone, not a pc.) You end up taking a risk of greatly reducing the lifespan of your cpu for 2-3 extra frames a second in games, drain more battery power, and again, in real world use you won't see a difference at all.
If you're trying to show off to others how cool your phone is, instead of showing them quad scores that mean nothing, why not let them USE it for a few minutes? That's what sold me on this phone when I bought it, using someone's for about 15 minutes

It could have been he was using performance governor and cleared out his ram. That would cause a height quadrant score.
Sent from my SGH-T989 using xda premium