Here is my post from vijay555.com site:

Enabling* LumosII on my PPC6700 (and similar units) allows the unit to think its using camera mode, so the autoscaling is disabled, which is how it was designed .
This enables me to run both overclock programs without problems on battery power.
Although this does not make xcpUScalar run better, it makes Pocket Hack Master run when it could not before. PHM makes this unit scream!* It runs noticeably better and much more efficient than xcpuScalar.
Now with PHM, my battery drain has returned to normal levels, even with this max brightness from LumosII.
*
Thank you so very much VJ !!

I just happened to use it while on antontomov (PHM) site which looked dark to me and posted there my discovery as well.

Whoa! That IS good news.

I'm actually trying it now with XCPUScalar: so far no good, like you said. I wonder why?

Running on scaling (204-512mhz) with LumosII knocks me up to an average of 250mAh.

Turning off LumosII brings me down to about 160 mAh.

Not running XCUPScalar I get around 60-70 mAh.

Can anyone post numbers with PHM 3.33? Get the progam to measure here...

Xcpuscalar works different so this prog doen't affect it.
It will still cause cpu drain.

PHM on the other hand works beautifully, and you can still adjust the brightness in settings. wohoo!

Maybe I'm not understanding, but on most other threads about PHM or xcpuscaler it shows that any benefits of underclocking the processor or overclocking it are minimal and that the overhead from the program negates the benefits of underclocking. Does this program fix all of that? I'm just confused as to how all this is coming together.

well, I'd say we're almost there...

True, PHM 3.33 works now and it does run well.

However, there is still no power savings--in fact, I find it uses as much power as XCPUScalar:

PHM 3.33/LumosII (scaling from 104-520; idling at 104 mhz = 165 mAh

PHM 3.33/LumosII: clocked at 104 mhz (no scaling) = 170-180 mAh

No PHM (idling) = 57 mAh


I will say this though:

If you set it to run just a 520mhz--it still idles at about 167 mAh with peaks up to 477 mAh when the processor demands it. It's almost as if the program now allows you to auto-scale to 520mhz when needed but then the system-internal scaling takes back over when idle.

In that sense, you're still using more than not using the program at all but the mAh increase is not too bad--this basically mimicks XCPUScalar's performance/battery usage when set to auto-scale.

Bottom line: both programs still cause you to use nearly 3x as much power than not using either--even when scaling is on and set from 104 (up to 520).

Others are welcome to verify/contribute.

i will extensively test this and report back, anton might have a way to disable the scaling of the phone proper so that phm handles it exclusively and we dont see the ridiculous battery usage, I thought maybe in camera mode the phone locks itself to one speed, obviously from what malatesta is saying that is not the case. But I will test all scenarios myself and report on this as well

I hope anton is able to come up with a way to trick our phones or a way to simply disable the scaling, apparently he says it is some software running on our phones built into the OS or something causing the scaling, and that the scaling engine it uses isnt even that great, I thought it was a hardware thing, bu tinfact it is all software (which I guess makes sense, kinda like the amd cool and quiet feature which is run my a program)

in my tests, PHM drains on average 100+ma less than xcpUScalar at idle and gave me 200% more benchmark speed on TCPMP.

There is a fundamental difference in overclocking that makes PHM better than xcpUScalar in my testing.

you are both right. surpisingly having the speed constantly at 520 via the program lucky said and then having phm handle scaling or set the speed doesnt result in much lower speeds, when nothing is running, it is about 60mah, as soon as you run the brothness program it jumps to about 200mah, and setting any speed within phm after that no matter how low (even 104mhz) doesnt have much impact, I would think setting phm to 104mhz would make the phone consume even less than 60mah (consumption with phm not running and phone on idle)

I really dont get it, I hope whatever Anton comes up with does reduce battery consumption, the ideal situation would be it clocks to very low speed when we are not using it, and goes to medium speed on avergae use and over 400mhz when we are running intensive stuff like emulators.

What I was thinking as a better solution would be to piggy-back over the internal scaling of the device but change it so that it can go higher (e.g. 520 mhz or 600+)...kind of like that SpeedStep program but for our processor/device.

Not sure if that's possible or not but it sounds easier!

My point is not about underclocking to save battery, but about performance when OVERclocking.
Now you can overclock faster without killing the battery.
Overclocked with PHM/LumosII, I achieved average of 200ma (lowest160) on idle, which is exactly what would be used at 416Mhz without overclocking.

at rest, 60ma is the lowest amount at idle on 104Mhz.

Our cpu jumps back to 416Mhz on a screen tap with the autoscaling, so expect at least 200plus ma usage when You use it.