Certain people that use your work, trust you guys.
faux's 2.1ghz kernel is really 1.8
KT's kernel @ 1.5 is really running @ 1.86
he is making it seem his kernel is faster than it really is by
making it display x speed when it is really clocked higher.
Why faux does the opposite. I have no idea. But this is grimy.
here is KT747's code - you can see this for yourself at his github
here is the link to this part of his github -> https://github.com/ktoonsez/KT747-J...be41bf639e/arch/arm/mach-msm/acpuclock-8960.c
Code:
{ 1, { 1512000, HFPLL, 1, 0, 0x3C }, L2(16), 1200000 },
{ 1, { 1512000, HFPLL, 1, 0, 0x3E }, L2(16), 1200000 },
{ 1, { 1512000, HFPLL, 1, 0, 0x40 }, L2(16), 1200000 },
#ifdef CONFIG_SUPER_CLOCKED
{ 1, { 1512000, HFPLL, 1, 0, 0x42 }, L2(16), 1200000 },
{ 1, { 1512000, HFPLL, 1, 0, 0x44 }, L2(16), 1200000 },
#endif
{ 0, { 0 } }
};
this is part of his "static struct acpu_level acpu_freq_tbl_8960_kraitv2_stock[] = {" table
now lets have a look here
Code:
{ 1, { 1512000, HFPLL, 1, 0, 0x42 }, L2(16), 1200000 },
{ 1, { 1512000, HFPLL, 1, 0, 0x44 }, L2(16), 1200000 },
what he have here is some additional 1.512 entries on top of the stock samsung one
lets take this apart
1512000 is the clock speed displayed in sysfs, or apps such as system tuner, or set cpu - it has NO impact on cpu speed - you can put ANY number here
HFPLL is the reference clock, for most MSM based devices, HFPLL = 27 MHz, or 27000 Hz
the "1" following HFPLL is the divider used - mainly for lower clock speeds, as any number divided by 1 is the number
now on to the good part
the 0x42
and 0x44
are the multipliers used to adjust the clock
so
convert 42 hex to decimal, we get 66
we multiply 66 by 27,000 and we get 1782000
we divide by 1, we get 1782000, or 1.782 GHz
now - since he also uses 0x44 - lets do some math
0x44 hex = 68 dec
68 x 27000 = 1836000
1836000 divided by 1 = 1836000
so, his kernel will display 1.5GHz, even though it is really running 1.836
here are some pics for further proof. because faux has either deleted or removed his git. or made it hidden. who knows. he's a fox (pun). these are of ktoonz' however, but they use the same tricky mathematics nonetheless.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
dont have to be a genius to see it. its simple math.
All respect. Lost. Until explanation is provided.
Pretty sheisty when you take donations for your "superior" work.
So faux kernel was set to.. .... wait for it..... a "faux 1.5 GHz".... ??
Looollll
It's 2:57 am I'm sure this thread will closed.by 3:00 am
Sent from my SGH-T999 using xda app-developers app
OP updated with screen shots... since faux decided to hide or delete his entries about ten minutes ago. i figured he would do this, because he is a faux
cobraboy85 said:
Did you cure cancer or something?
Click to expand...
Click to collapse
iirc he's a douche... (lol.. Google it)
Holy crabs, what is going on? That wins pic of the year in any case. "There's a cobraboy in my boot." Let's just hope the boot doesn't come and find you!
Also, to quote, you really have no ****ing clue who I am? Really? Son of a ***** you're re...diculously uproarious, son.
cobraboy85 said:
i found this rather funny. kind of depicts everything this thread stands for.
Click to expand...
Click to collapse
I'm so sleepy but waiting to see when they close this lol
Sent from my SPH-L710 using Tapatalk 2
oh i promise, it will be closed. and deleted as well. any type of truth around here about what people are really doing gets stifled out quickly.
i know people that use their kernels, and i have before. and this is flat out dishonest when you take donations from people for your "superior" work.
cobraboy85 said:
oh i promise, it will be closed. and deleted as well. any type of truth around here about what people are really doing gets stifled out quickly.
i know people that use their kernels, and i have before. and this is flat out dishonest when you take donations from people for your "superior" work.
Click to expand...
Click to collapse
It is..I have only ran one of those kernels it was great for a couple days then it went all weird on me..plus didn't like the app that forcefully was added. But this is shinning light to a point, but at least from sprint section there are limited kernels, therefore his creates traffic to that section even other kernels I've seen with his "governor" so doubt any action would be taken mods will point towards the disclaimer, probably hand out warnings if not bans and **** will remain dude
Sent from my SPH-L710 using Tapatalk 2
well hopefully that is all that happens, but we'll see. history of this site says otherwise, however.
i just want an explanation. and not some slimy excuse, which unfortunately faux is known for from what i have observed before.
Man wish it would happen but I seriously doubt it...another thing that will get blamed on "anthrax" people starting ****. Which is not even the case but the disclaimer and past actions here I guess will be used to scapegoat into no explanation
Sent from my SPH-L710 using Tapatalk 2
That escalated quickly!
Sent from my SGH-T999
chrisram88 said:
That escalated quickly!
Sent from my SGH-T999
Click to expand...
Click to collapse
And died quickly, I was kinda hoping for a response from one of the devs :-\
we need/deserve some answers. do we need to start a petition or something. if we dont get some answers im going to all the android news sites to get the ball rolling with these saved pages. thanks for pointing this out cobraboy85
So faux still doesn't get his clock
tables right?
Or another innovation?
Sent from my SGH-T889 using Tapatalk 2
imneveral0ne said:
And died quickly, I was kinda hoping for a response from one of the devs :-\
Click to expand...
Click to collapse
I'm guessing that's not going to happen around here. But maybe we will get lucky..
Sent from my SGH-T999 with using Tapatalk 2
Closed due to flaming and disrespect to each other
It also became more about dev bashing than getting answers
Some posts had to be edited including the OP for name calling
The kernel source has ALWAYS been available as per GPL
And just NOW someone decides to examine it with a microscope?
We have too few developers making S3 kernels and you want to try and run them off?
XDA is about development and nothing here is aiding development
The developers in question have always responded to my PMs.
Perhaps you should have asked these questions there instead of this
Had this been an intelligent discussion politely requesting the developers to help understand what happened, it would have remained open.
Instead it became a circus and witch hunt. You might want to question the source of the information as well as the source of the code used as a base.
Related
What's your profile setup?! Here's mine:
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
(Using Kingx's BFS kernel #1, btw)
Why did you change all of the "on demand" defaults? I never understood the difference, even after reading the release notes.
this will be interesting to follow, just installed this app...
I use conservative.. It's a good medium for performance and battery life.
And for screen off... powersave.
Sent from my HTC Evo 4G using XDA app.
I really don't know the difference between all of the governors, is there a list and a description of what each do?
engagedtosmile said:
I really don't know the difference between all of the governors, is there a list and a description of what each do?
Click to expand...
Click to collapse
this helps a bit ... http://www.pokedev.com/setcpu/
http://forum.xda-developers.com/showthread.php?t=758655
i've just started mussing with it.
due to the version of the -snap kernel, i am using conservative as per dev notes.
Using Netarchy-Toastmod 4.1.2 I only have Performance and Conservative.
main: Max: 998 Min: 245 (Conservative)
Profile: Screen Off: Max: 245 Min: 128 (Conservative)
Quadrant: 1281
DraginMagik said:
this helps a bit ... http://www.pokedev.com/setcpu/
i've just started mussing with it.
due to the version of the -snap kernel, i am using conservative as per dev notes.
Click to expand...
Click to collapse
Seems to be missing some. Like "interactive". Hmm. I tried doing a search for it but its just release logs of adding it.
I don't use profiles because they seem to slow my phone down like mad.
ALL credit goes to shift_
temp > 50c = 422/128 powersave (priority 100)
screen off = 384/245 powersave (priority 98)
battery < 20 = 499/245 powersave (priority 91)
battery < 50 = 768/245 ondemand (priority 84)
charging/full = 1152/768 ondemand (priority 82)
battery < 100 = 1152/768 ondemand (priority 76)
This is the current on i use on cm6 rc2
Heres mine. I just got setcpu, and im still tweaking it, but this seems to be working well.
and to the op, you know your 65.2 degrees C is like 145 degrees F right? I might be wrong, but wouldn't the phone be getting heat damage well before 145 degrees?
And also, I downloaded JuicePlotter, it really helped in telling me how the battery life was affected.
Anyone else going to chime in?
here's what i got, had this setup for a while and so far it's working great:
the governors have to come from the kernel right? bakedkernel 2 doesnt have conservative or powersave. is there any way to add them?
Using 4.1.8 bfs/bfq fresh compatible kernel so far smooth, no issues, satisfactory battery life.
This is not mine. At 1000Hz (52Vsel) I got only 2000.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Which rom did you use?
Hi ABC Universal,
can you tell us how did you reach this score ???
The difference is so big...!!??
Looks like you're based on a chinese version as I see some chinese in the status bar.
Thx.
Really impressive!!!
How is it possible?
My stock rom eclair 2.21 with jit and deodex keep an highest value of 1550 on quadrant...
Sent from my MB525 using XDA App
1- A fake.
2- Sense Kernel or something similar.
tomeu0000 said:
1- A fake.
2- Sense Kernel or something similar.
Click to expand...
Click to collapse
Agree
Sent from my MOTO Defy
It is genuine output from quadrant running on defy. But yes there is no miracle performance boost.
Actually it demonstrated a limitation of defy's (and many others') CPU and a problem of quadrant benchmark. There is a detailed Chinese article explaning it but I understand only a little. Essentially, with some settings in android 2.2 system, defy (and many others) will fail the H264 decoding test, but quadrant takes it as completed in mini second, and thus the exceptional high cpu mark.
It was originated by some low end Chinese android phones which surprised people with exceptional quadrant benchmark. It turned out they had these settings enabled by default (/system/build.prop):
media.stagefright.enable-player=true
media.stagefright.enable-meta=true
media.stagefright.enable-scan=true
media.stagefright.enable-http=true
So this simply means the guys over at Aurora need to update their software to read the fail and drop the score to where it should be in reference to high end phones.
Does it mean that the Nexus and Evo 2.2+ scores are also wrong for the same reason?
sandresimon said:
Does it mean that the Nexus and Evo 2.2+ scores are also wrong for the same reason?
Click to expand...
Click to collapse
According to the discussions, these two models enabled the settings by default. But I don't have them to examine myself.
That's explains why you install the sense rom, you obtain amazing Quadrant scores...
But really? no better performance? i thing this processor can do more than now...
not fake, i can do that too http://forum.xda-developers.com/showthread.php?t=976015
I think people getting 2700+ from defy now with this hack
hohoho, i go to try this....
EDIT.: For me, 1750 1Ghz 57vel
I've got 1231 @ 1000 mhz/53 vsel and 1303 @ 1100 mhz/55 vsel.
What benchmark app is this?
Whooooaaa!!! I don't believe it....!! Look awesome!
Down the line if the Defy can stay at average/consistant score around 1300 to 1500 should be more than enough running the daily apps/games available in the market.
Hit it more will eventually fried the poor Defy... huhuhu..!
mine got [email protected] on cm7 beta6, so you don't need to "cheat" to get a score that high out of the defy
this is my best score:
2617 on my defy with cm7 beta 6, cpu set to 1200 mhz and 80 vsel
but I think that there is something wrong: I usually get about 2200 (1950 with 1000 mhz, 1700 with 800 mhz), but every 5-6 tests i get similar score
LeoD7K said:
What benchmark app is this?
Click to expand...
Click to collapse
Quadrant
I've always gotten lower scores (barely 2000, sometimes a little more than 3000) and was wondering if I was going crazy or this is normal. I've tried so many rom/kernel combinations but no luck.
I'm on AOKP and I'm getting 2068 and lower.
Comes from the Epic and the GS2, I was kinda expecting a lot more.
Sent from the future.
SocialReject said:
I've always gotten lower scores (barely 2000, sometimes a little more than 3000) and was wondering if I was going crazy or this is normal. I've tried so many rom/kernel combinations but no luck.
I'm on AOKP and I'm getting 2068 and lower.
Comes from the Epic and the GS2, I was kinda expecting a lot more.
Sent from the future.
Click to expand...
Click to collapse
The updated quadrant reports much lower for all devices. The GN isn't as much of a "benchmark queen" as the SGSII has potential to be. Unless your phone is reporting a score far under what Quadrant reports as typical, don't worry about it.
I'm benchmark app free
There are a few posts around with people showing around 4000 on quadrant i think. Dont know which roms though. Search the benchmark thread.
ive found the BEST stability and over all best scores without over clocking with the official cyanogenmod 9 nightly i get 3000+ with that... but like the dude before said quadrant scores are all hype... just use a rom with good speed that works best for u.
I don't always choose a ROM based on 1 stat...
but when I do, its for Quadrant.
Quadrant doesn't really translate to real world use.. why do people obsess over quadrant scores so much? If the phone runs well it runs well I don't need third party software to tell me if my phone is running fine.
Sent from my GT-N7000 using xda premium
My best is ~3700 using AOKP B25 & Imoseyons 3.0.20 kernel. Undervolted, stock clocked and using Ondemand.
Never even used quadrant...
#Likeaboss B-)
Sent from my Galaxy Nexus using XDA App
user experience > quadrant score
idc about quadrant, i just want a clean phone w/o bloatware with lot of resources and support
I owned a Sprint SGS2 and while quadrant scores were in the 3000's consistently and the GN is in the 2000's I cannot tell the difference in speed between the 2 of them.
Now with quad cores, they will even score higher but how much smoother can a device run then smooth as butter?
If you don't think your phone runs smooth, then try AOKP and turn your swag on. Let's see if the iPhone 4s can do that sh.t!
I get around 3200-3300 give or take a hundred running AOKP build #26 + franco.Kernel 18.6 oc'ed to 1350mHz (minimum 700mHz) and using conservative governor. The UV has been left as it came with the kernel or whatever.
I have to add that I just got this phone a couple of days ago, and I am totally and utterly giving this the title of best phone I have ever had. I can't believe the person I got this off of actually switched to an iPhone because of battery issues xD
Aokp build 25
franco.kernel @ 1.65ghz (the new releases don't go this high because it's unstable for some)
interactive governer
Close:
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
http://review.cyanogenmod.com/#/dashboard/4343
Please log in at gerrit and vote for my commits to have them included faster. There are a few issues which need to be adressed fast.
Why are you saying urgent hotfix ? Afaik, adding BFQ and the deadline patches for flash are far from being urgent.
Sent from my Galaxy Nexus using xda premium
chadouming said:
Why are you saying urgent hotfix ? Afaik, adding BFQ and the deadline patches for flash are far from being urgent.
Sent from my Galaxy Nexus using xda premium
Click to expand...
Click to collapse
I removed them if you look closely. These are the commits you should vote for :
I5f878001 OMAPDSS: DISPC: more precise scale limits checking
Iabeac2ad OMAPDSS: DISPC: fix pclk divisor limits+1
If417fabc Correctly show battery usage for MPU OPP 1350MHz
I78377898 MPU 1350MHz: improve stability
djjonastybe said:
http://review.cyanogenmod.com/#/dashboard/4343
Please log in at gerrit and vote for my commits to have them included faster. There are a few issues which need to be adressed fast.
Click to expand...
Click to collapse
Gerrit isnt a voting system.
Even if you get hundred +1 it doesnt mean it will be included.
how i can vote?
djjonastybe said:
I removed them if you look closely. These are the commits you should vote for :
I5f878001 OMAPDSS: DISPC: more precise scale limits checking
Iabeac2ad OMAPDSS: DISPC: fix pclk divisor limits+1
If417fabc Correctly show battery usage for MPU OPP 1350MHz
I78377898 MPU 1350MHz: improve stability
Click to expand...
Click to collapse
First like the above poster said gerrit is not only about voting. Second if you want users from here to vote on something you should explain why and what those patches are aimed for. Third those two OMAPDSS patches are not from you, the authors are engineers from TI, so please add the author signature to the patches.
franciscofranco said:
First like the above poster said gerrit is not only about voting. Second if you want users from here to vote on something you should explain why and what those patches are aimed for. Third those two OMAPDSS patches are not from you, the authors are engineers from TI, so please add the author signature to the patches.
Click to expand...
Click to collapse
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Their inclusion, or exclusion, will be based on technical merit. I don't think that lobbying for +1's here will help your cause.
espenfjo said:
Gerrit isnt a voting system.
Even if you get hundred +1 it doesnt mean it will be included.
Click to expand...
Click to collapse
Especially if the patch is just plain WRONG like the power_profile patch is. What's so "urgent" about a patch that is identical to one you submitted a month ago, received a comment on, did not even reply to the comment, and then abandoned after a month?
Also, every single one of your patches I've seen only affects overclocking, or other "non-stock modifications"... Not exactly "urgent".
This does not belong in Development
There is no release here
Thread moved
CMNein said:
Their inclusion, or exclusion, will be based on technical merit. I don't think that lobbying for +1's here will help your cause.
Click to expand...
Click to collapse
In fact, it will be more likely to make the patches ignored or inclusion delayed, since it makes it obvious the +1s are bogus.
Sorry but
LOL
Sent from my i9250
Sorry bud... but kernel patches break build for me
Sent from my Maguro
Hello Everyone!
I have sample of Mugen Power 3000mAh Battery Case for LG Nexus 5 available for independent objective testing. Please leave a note here if you are willing to test it and write a review about it.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
I have 1 sample available, so the final choice of tester will be based on Your contribution to xda-developers community: number of posts, number of thanks, thanks/posts ratio and years on forum, using following formula:
p = number of posts
t = number of thanks
y = years on forum
score = p/100 + t/20 + (t/p)*10 + y
Thank You & Have a Great Day!
More info on the battery: http://mugen.co/mugen-power-3000mah-battery-case-for-lg-nexus-5.html
Well. I can give it a try.
I'll attempt this. Battery case would be handy for RSP.
---------- Post added at 02:00 PM ---------- Previous post was at 01:58 PM ----------
When is the deadline date?
dmitrymugen said:
Hello Everyone!
I have sample of Mugen Power 3000mAh Battery Case for LG Nexus 5 available for independent objective testing. Please leave a note here if you are willing to test it and write a review about it.
I have 1 sample available, so the final choice of tester will be based on Your contribution to xda-developers community: number of posts, number of thanks, thanks/posts ratio and years on forum, using following formula:
p = number of posts
t = number of thanks
y = years on forum
score = p/100 + t/20 + (t/p)*10 + y
Thank You & Have a Great Day!
More info on the battery: http://mugen.co/mugen-power-3000mah-battery-case-for-lg-nexus-5.html
Click to expand...
Click to collapse
As someone who owns both the limefuel nexus 5 case and the zerolemon, I'd be happy to review this unit.
Let me know if I can help.
-Rob
I've been looking for more than Amazon reviews on this for weeks now. I would give it a full detailed write up!
Looks like a nice case, and I'm particularly interested in how it attaches.
I have one of the Limefuel cases which I almost never use, because it tends to pry up the glass screen protectors I use.
Looking at the photo of this one, am I right in thinking it's partly making use of the headphone jack to hold itself in place
seems pretty cool, I wanna try it out
I wanna try it too
Sent from my Nexus 5 ?
Currently using a cheap Ebay bumper cover, I'd love to review the unit, I've pretty obsessed with battery life!
Looks cool, i would like to try it
I would love to give this a try, currently testing a portable charger but having a backup attached to my device would make things even better Also I need a new case and this looks like it would be perfect for my red Nexus good luck everyone also thanks for doing this, glad to see people giving back
Hi! Could I get it? P
CyberScopes said:
Looks cool, i would like to try it
Click to expand...
Click to collapse
*sigh* guess we have a winner
I might be ineligible but I'd like to try it.
BRiANj64 said:
*sigh* guess we have a winner
Click to expand...
Click to collapse
Lol sorry. If i do win however and someone else really needs it i might let them have it instead. Sorry again :good:
CyberScopes said:
Lol sorry. If i do win however and someone else really needs it i might let them have it instead. Sorry again :good:
Click to expand...
Click to collapse
It needs proper testing so that he can produce more. Go for it dude. I can't wait to get one of these bad boys.
iKlutz said:
It needs proper testing so that he can produce more. Go for it dude. I can't wait to get one of these bad boys.
Click to expand...
Click to collapse
Yeah it would be great trying one of these and review it.
Hello. I've never had a battery case and this may sound stupid but, does the battery work without the case itself? Thanks.
Aww guess I can't test it then .
I'd be willing to test also.