Active Topics
-
[ANNOUNCE] MeeShop - new app store (10)
to Applications by aspergerguy - 1 day, 8 hrs ago -
HTTP-401 Sfos-Builder when using /etc/ssu credentials (0)
to SailfishOS by insanelysexy - 2 days, 14 hrs ago -
Longtime users: What device(s) are you using today? (54)
to General by ryu1 - 4 days, 20 hrs ago - more...
 |
|
2012-01-10
, 21:23
|
|
Posts: 1,455 |
Thanked: 3,309 times |
Joined on Dec 2009
@ Rochester, NY
|
#32
|
Originally Posted by marmistrz
This is true, but if you don't understand how they're doing what they claim to be doing it's best to avoid the shinny. All the glitters is not gold. Also, if the author refuses (or can't) explain how it's working, it's snake-oil at best. Karam's patches fall into that category so far.
Yeah, but there may be some apps that look great on first look and are crap, such as karam's patches
Originally Posted by karam
I believe I was that "bonehead". And it wasn't totally proven to be maxcpu. Five people on the thread were complaining about speed issues. All five also happen to have speedpatch installed. Seems like an awful big coincidence to me.
and in other thread as i remember ,there was a call lag problem
first some boneheads said it was by speedpatch
then it turned out the problem was caused by maxcpu app
One of them decided to do the right thing: re-flash and test after each install. He "determined" it was maxcpu, but only after installing speedpatch again, which as we both know tweeks config parameters. He "tested" speedpatch by uninstalling it, but we both know that does nothing to undo what's been done.
That doesn't tell me it's not speedpatch, but rather that there's a conflict with using speedpatch and maxcpu. Since lots of others are using maxcpu without issue without speedpatch, I'm thinking either/both of them are probably not worth the gamble. If he would be willing, I would love for him to re-flash, and just install maxcpu, without speedpatch. That would clarify that it was in fact just that app. My bet: It's something speedpatch is doing to the system, and maxcpu is just amplifying the problem.
Personally, I prefer swapolube for one simple reason: At any given time, I can hit one button and restore every tweak done back to the system stock values. If you add that to your "patch" set, a way to fully uninstall and undo/restore all configurations your patch touches back to stock defaults, and then we'll see what's really going on. Until then, you're tweaking things you don't understand and/or refuse to explain, and making promises that more often than not fail.
Want me to stop suggesting against it? Make it so an uninstall replaces all the values you're touched, and explain how what you're tweaking improves anything at all. (Basically, reply with answers to freemangordon's post below.) Do that, and I'll be happy to tell people go ahead and try it. But without knowing what it's doing, and no way to undo what it's done? It's like playing Russian roulette, and I will advise against it.
| The Following 2 Users Say Thank You to woody14619 For This Useful Post: | ||
|
|
2012-01-11
, 02:51
|
|
Posts: 856 |
Thanked: 1,681 times |
Joined on Apr 2010
@ Aleppo ,Syria
|
#33
|
Originally Posted by freemangordon
well that's a good point
What could be the reason for placebo effect (i.e. speedup of the system reported by some users) is that speedpatch (along with batterypatch) has a new version almost every day and as it requires a reboot, here it is, speedpatch/batterypatch users have uptime no more than 48 hours, so effects like swap fragmentation does not appear.
however the last update was on 3 jan
and the creation if speedpatch was a result of testing
perhaps i have no huge expert to explain how it does that
but according to speedpatch testers .. it appeared that it does something
and i wouldn't take a chance a create a new thread (and it was my first thread i ever create) if i wasn't sure that cgroup patch does something
if you have noticed that the first versions were just scripts
not in devel
but after i got many requests uploading it devel .. had to do it
as for battery patch
Originally Posted by freemangordon
you can add
AIUI those scripts play with two things based on whether device is locked or not:
1. vfs_cache_pressure
2. CPU clocks
3- renice processes
4- changes minfreq to 720 when a call is received (which will improve the response)
5- uses conservative module (still not proven it's the best)
Originally Posted by freemangordon
i think that's why it's called extras-devel ?
if you were playing with your device only, but you are playing with the devices of all users who had this installed from extras-devel
Originally Posted by freemangordon
well it doesn't require it .. it's just compatible with it
Your latest batterypatch requires KP49, ain't? And enables SmartReflex? Well, it is SmartReflex that increases battery life, not your trickery with CPU frequencies and dbus-monitoring whether device is locked or not
the thing about KP49 is that it made it possible to OC with maxfreq 900mhz with vdd1 enbaled
my older versions had 750 as max with vdd1 enabled
now the are 805mhz with vdd1 with KP49 only
but if the user has an older version
then it will use 750 as max
Originally Posted by freemangordon
well here :
in offline mode,device with KP49, dsp profile 125-805 has standby current of:
xx mA without batterypatch
xx mA with batterypatch
in offline mode,device with KP49, underclock profile 125-600 has with vfs 10 and reniced processes
xx>mA with batterypatch than without it
of course the hard ware cannot be edited .. but N900 requires quite less battery life than without batterypatch
Originally Posted by freemangordon
in that one overclock profile with conservative module
the same device running abc background task has current of:
well got nothing to say about that but tests showed me it improves continues time of using
Originally Posted by freemangordon
may i ask you what kind of problem batterypatch may cause ?
Only by making such measurements you can prove batterypatch does anything but problems.
ok i got that speedpatch can cause problems (although not with me)
thank you
__________________
My Projects:
Speed Patch & Battery patch
Task-Swapper
Wifi-Pwner (MDK3 GUI)
Hacking Tools
Tutorials
How To Increase Camera Image Quality
How To Unblock Repositories In Forbidden Countries [Like Mine]
My Projects:
Speed Patch & Battery patch
Task-Swapper
Wifi-Pwner (MDK3 GUI)
Hacking Tools
Tutorials
How To Increase Camera Image Quality
How To Unblock Repositories In Forbidden Countries [Like Mine]
| The Following 4 Users Say Thank You to karam For This Useful Post: | ||
|
|
2012-01-11
, 07:44
|
|
Posts: 3,074 |
Thanked: 12,964 times |
Joined on Mar 2010
@ Sofia,Bulgaria
|
#34
|
Originally Posted by karam
Which proves once again mine and many others point that you have absolutely no clue what your so called speedpatch is doing. And TBH what irritates me most is that you still refuse to google for linux kernel cgroups and educate yourself. Description is several pages long, go read it. Computer science , software development, you name it, is not alchemy or black magic, it is that - science and while six sense helps, one needs knowledge too.
well that's a good point
however the last update was on 3 jan
and the creation if speedpatch was a result of testing
perhaps i have no huge expert to explain how it does that
but according to speedpatch testers .. it appeared that it does something
and i wouldn't take a chance a create a new thread (and it was my first thread i ever create) if i wasn't sure that cgroup patch does something
if you have noticed that the first versions were just scripts
not in devel
but after i got many requests uploading it devel .. had to do it
as for battery patch
.
.
.
in offline mode,device with KP49, underclock profile 125-600 has with vfs 10 and reniced processes
xx>mA with batterypatch than without it
of course the hard ware cannot be edited .. but N900 requires quite less battery life than without batterypatch
in that one overclock profile with conservative module
well got nothing to say about that but tests showed me it improves continues time of using
may i ask you what kind of problem batterypatch may cause ?
ok i got that speedpatch can cause problems (although not with me)
thank you
|
|
2012-01-11
, 07:54
|
|
Posts: 195 |
Thanked: 96 times |
Joined on May 2011
|
#35
|
for me i have 0 expert
i have good six sense
i have horrible english
i use speedpatch and batterypatch
i find them good
i have no problems
now everyone
shut it
this will last forever
who don't like them > don't use them
who likes them > use them
#facepalm#
*legand of the seeker* i love this movie
edit / i will go to sleep now
edit2 / i love you all
Last edited by Seker_94; 2012-01-11 at 08:00.
i have good six sense
i have horrible english
i use speedpatch and batterypatch
i find them good
i have no problems
now everyone
shut it
this will last forever
who don't like them > don't use them
who likes them > use them
#facepalm#
*legand of the seeker* i love this movie
edit / i will go to sleep now
edit2 / i love you all
Last edited by Seker_94; 2012-01-11 at 08:00.
| The Following 2 Users Say Thank You to Seker_94 For This Useful Post: | ||
|
|
2012-01-16
, 15:38
|
|
Posts: 3,328 |
Thanked: 4,476 times |
Joined on May 2011
@ Poland
|
#36
|
Are these scripts also that bad: http://talk.maemo.org/showpost.php?p...68&postcount=1
and
OptimizeN900?
and
OptimizeN900?
|
|
2012-01-16
, 15:52
|
|
Posts: 856 |
Thanked: 1,681 times |
Joined on Apr 2010
@ Aleppo ,Syria
|
#37
|
yes they are bad
an update in batterypatch included them then they were removed because they cause more battery drain than they save
and optimizen900 (not sure about it) because i don't use it
an update in batterypatch included them then they were removed because they cause more battery drain than they save
and optimizen900 (not sure about it) because i don't use it
__________________
My Projects:
Speed Patch & Battery patch
Task-Swapper
Wifi-Pwner (MDK3 GUI)
Hacking Tools
Tutorials
How To Increase Camera Image Quality
How To Unblock Repositories In Forbidden Countries [Like Mine]
My Projects:
Speed Patch & Battery patch
Task-Swapper
Wifi-Pwner (MDK3 GUI)
Hacking Tools
Tutorials
How To Increase Camera Image Quality
How To Unblock Repositories In Forbidden Countries [Like Mine]
| The Following User Says Thank You to karam For This Useful Post: | ||
|
|
2012-01-16
, 16:22
|
|
Posts: 3,328 |
Thanked: 4,476 times |
Joined on May 2011
@ Poland
|
#38
|
one more question: why the titan's ideal profile has 500 as min freq? Won't it consume more power?
EDIT: it'll sit longer on idle. but why will it save power?
Last edited by marmistrz; 2012-01-16 at 16:47.
EDIT: it'll sit longer on idle. but why will it save power?
__________________
If you want to support my work, you can donate by PayPal or Flattr
Projects no longer actively developed: here
If you want to support my work, you can donate by PayPal or Flattr
Projects no longer actively developed: here
Last edited by marmistrz; 2012-01-16 at 16:47.
|
|
2012-01-24
, 22:05
|
|
Posts: 1,455 |
Thanked: 3,309 times |
Joined on Dec 2009
@ Rochester, NY
|
#39
|
Originally Posted by marmistrz
To answer that there are two things:
one more question: why the titan's ideal profile has 500 as min freq? Won't it consume more power?
EDIT: it'll sit longer on idle. but why will it save power?
1> All multi-frequency processors have a performance curve, where they perform best. For the N900 processor, that curve has been mapped a few times, and every time it's a bell curve with a peek right around 500Mhz. It makes sense that this is the case, since it's sold as a 500Mhz processor. That's what TI optimized everything for. So 500Mhz is the most energy efficient when it comes to doing any real solid number crunching. (Clearly sleep mode, or 0Mhz, is more efficient when it comes to not doing number crunching.
)2> Imagine you have a background process that wakes up twice a minute to check some values. Let's say it runs for a quarter of a second at 500Mhz. That's not enough to trigger the system to go up to the next frequency. It wakes up from 0, runs, and the system goes back to sleep. That same process at 125Mhz would run the processor for up to 4 times longer, and probably trigger a move up to 250Mhz. Lets say it triggers the jump at one half second, the 125Mhz/250Mhz version would take 3/4 of a seconds. (250 would take a half second, where 125 would take 1... Half the work is done in .5 seconds at 125, the other half is done in .25 second at 250) After an hour:
500 Mhz = 30 seconds of run time, 59.5 minutes of sleep time.
125/250 Mhz = 1.5 minutes of run time, 58.5 minutes of sleep time.
After a day? The 500Mhz system has run for 12 minutes, where the 125Mhz system has run for 37 minutes. Which one used less energy, assuming 0 mhz takes close to 0 power (when compared to any running state)? Now add up all the little apps running time slices in the background that wake up on occasion to do work. That translates into a reasonable amount (5% to 10%) of battery savings over a day. That's called the "race to idle" effect.
Between those two, choosing your most effective speed for your base speed actually causes less energy use. Thus why most people here who have good battery life are using a 500-X clocking method, be that 500-600 for non-overclockers, or 500-850/900/1100 for overclockers.
Btw: You may note that overclocking does in fact mean you're sliding down the other side of that bell curve for energy efficiency. The more you overclock, the more your battery drain can be when you're using the device for long-term number crunching!
| The Following 8 Users Say Thank You to woody14619 For This Useful Post: | ||
|
|
2012-05-28
, 20:04
|
|
Posts: 3,328 |
Thanked: 4,476 times |
Joined on May 2011
@ Poland
|
#40
|
Hello
right after reflash I installed a lot of apps and the n900 became much laggier. usable but laggierl
Which one of these may cause it: (list from fapman)
right after reflash I installed a lot of apps and the n900 became much laggier. usable but laggierl
Which one of these may cause it: (list from fapman)
Code:
catoriseplus 0.6.0.1 install foreca-weather-applet 0.9.5 install qlshop 0.11 install reactionfaceoff 0.0.2-1 install docstogo 2.001.346 install maemo-applet-tvout 6.0-0+0cssu0 install rootsh 1.8 install libqtm-systeminfo 1:1.0.2-maemo4+0m5+0cssu1 install advanced-power 0.4.2-4 install theme-customizer 0.9-77 install nano-opt 2.2.6-1maemo1 install movie-schedule 0.3.0-3 install fapman 0.7.1-1 install bootmenu 1.11 install libqtm-serviceframework 1:1.0.2-maemo4+0m5+0cssu1 install autodisconnect 0.4.7 install community-ssu-enabler 0.6-6 install scm 0.8.2-3 install easy-deb-chroot 0.9.56-1fremantle1 install sudser 0.2.0-4 install pygtkeditor 3.0.19-1 install cordia-theme 0.23 install simple-brightness-applet 1.4-1 install libqtm-publishsubscribe 1:1.0.2-maemo4+0m5+0cssu1 install ogg-support 1.1.1 install pierogi 1.1.4 install lens-cover-reminder-sp 0.4 install libqtm-location 1:1.0.2-maemo4+0m5+0cssu1 install libqtm-contacts 1:1.0.2-maemo4+0m5+0cssu1 install mp-fremantle-community-pr 21.2011.38-1Tmaemo3.1 install libqtm-messaging 1:1.0.2-maemo4+0m5+0cssu1 install wifi-assistant 1:0.9.1-2 install bigword 1.0.11 install seriesfinale 0.6.9-1 install ytube 1-8 install wifieye 1.0-3 install ati85 1.4.1-1 install qtquickcompat 0.1.0+0cssu1 install libqtm-bearer 1:1.0.2-maemo4+0m5+0cssu1 install n900-fmrx-enabler 1.5-4+0m5 install gtranslate 0.6.1 install kanban 0.7-2 install python2.5-qt4 4.7.5-maemo2 install cell-modem-ui 0.5.1-1 install libqtm-sensors 1:1.0.2-maemo4+0m5+0cssu1 install guitar-chords 1.0-2 install lzma 4.43-14maemo5 install leafpad 0.8.17le-0maemo1 install kernel-power-flasher 1:2.6.28-10power50 install qcpufreq 0.4.4-1 install qnitconverter 0.4-2 install meecalc 1.2-1 install shoplist 0.5.3-0 install kfzcheck 2.5-2 install advanced-power-monitor 0.5.3-3 install swappolube 1.4-2 install socat 1.6.0.1-maemo5 install qtlockscreenconfig 0.1.5 install cssutransitionstuner 0.2.0 install espeak-extra-data 1.46.17 install 3g2g-mode-selection-applet 0.4-2 install currencyconverter 0.8-7 install mstardict 0.6-2 install chessclock 1.1.2-1 install iphone-lockscreen 5.2 install swapset 0.1.1 install scoresheet 1.0-1 install psswitcher 0.2-3 install qtlockscreen 0.1.27 install timerjinni 1.3.20101221-2 install powertop 1.2 install omvoiceserver 0.1-14 install busybox-power 1.20.0power1 install smsfaker 0.1.1-5 install status-menu-applet-tvout 6.0-0+0cssu0 install mypaint 0.9.1-maemo7 install starhash-enabler 0.2 install cssufeatures 2.5-6 install gpsjinni 0.9.20101221-5 install recovery-boot 0.1-1 install notmynokia 1.0+maemo-1 install kernel-power-settings 0.14 install libqtm-versit 1:1.0.2-maemo4+0m5+0cssu1 install espeakgui 0.3-1 install openmediaplayer 20120518-1 install mussorgsky 0.5.2-3 install humanity-icon-pack 1.0-0 install tar-gnu 1.22-2maemo5 install tuner 0.1.3 install tracker-cfg 0.3.2-4 install espeak 1.46.17 install advanced-clock-plugin 0.15.3 install easy-chroot 0.3.5-1fremantle1 install speedtest 1.3.0 install powatool 0.1 install libqtm-multimedia 1:1.0.2-maemo4+0m5+0cssu1 install shopper 0.5.13-1 install
When I try to boot other kernel than omap in multiboot, the only thing I get is the Nokia logo. And its only kernel reflash that helps
If you want to support my work, you can donate by PayPal or Flattr
Projects no longer actively developed: here