huh? you're post about my post makes no sense whatsoever.

I understand it works for some rather particular scenarios... I clarified that it did not work for my particular needs. And gps has nothing to do with it... sheesh.

Raging Idiot

You initially called XCPUScalar a sugar pill - do you want me to explain in depth how you did that?

You then followed that up by explaining that it was worthless to you because it didn't speed up your slowest apps.

Can you not understand how someone could see your second statement as validation for the first?

So I gave some EXAMPLES where the program wouldn't speed up those applications, while still accomplishing what it's supposed to. I took a shot in the dark at GPS because you mentioned navigation from the memory card in your post. Copy understood the same, because he ran a benchmark of XCPU and navigation.

This discussion is totally off topic, but it started when you called this performance app & XCPUScalar 2 sugar pills.

thanks... now i'm running from 400 to 520 on my mogul like nothing, but i'm not even gonna try it on 624 due to the risks.

You're an ass.

Seeing you decided to overlook my post with the clarification:
"I did very strict tests with these overclockers...my results were different than yours, perhaps because I ran the navigation testing from storage card or maybe setup differences- I don't know."

Also this thread is still right on target IMO as these overclocking details are being clarified.

As for these apps...well I don't have stock in them, and although they didn't fit my needs maybe they fit yours.

LOL at your reply and the namecalling. The sad thing is that I had originally replied to you with no animosity, I just debated your conclusions.

No, I did not overlook that statement about your testing criteria, since both I and another member DIRECTLY addressed it.

Apparently a "sugar pill" is any program that doesn't directly meet your needs, in spite of doing exactly what it says. Gotcha.*

*edited to note that yes, this last sentence was unnecessary and written out of spite over the namecalling.

So now I guess it would be nice to get the CPU's autoscale turned off some how. I remember when I upgraded my thinkpad laptop's processor to a p3 proc on a p1 board the autostepping on the p3 was "disabled" due to the p1 incapability of utilizing this feature. So. Either a debug session to turn it off or a cut trace on the mainboard.

Any takers?

Now that I've pulled the wool from my eyes I really would like to tackle this autostep situation with some hands on experimenting... Any direction from anyone would be appreciated. Even if it means opening up my referb'd POS, I'm fine with that.

So. Another question. If I lock my proc @ 520 (like it already is) is it still speed stepping with the built in CPU's stepping (my guess is yes) and if so why doesn't it do it effectively enough to actually save power? Does it only step one down or does it actually alter the frequency in a non-multiplier fashion?

----------------

I've got the gun to my 6700's display. I'm ready to pull the trigger.

Copy

dude.. that's why you're an ass. Taking sh!t out of context via half quotes and misdirection such as the gps ass-umption, and saying the clarifications were "off topic"...not to mention your misconception of chronological order. bleh

Nobody gives a crap anyway.

Bottom line this debate did fish out some useful details which might not have appeared had not certain egos been at stake.

Raging Idiot

So be it. I'll let readers determine who the ass is in this discussion.

Copycounslr, good luck with actually getting the phone to shut off it's own scaling. I don't know if anyone's tried a hardware solution like you're suggesting . . . I definitely don't have the experience. I know Anton's looked at the programming end for a long time, and PHM hasn't been able to solve it either.

If your phone is locked in at 520, and you're not autoscaling in XCPUScalar - then it's definitely autoscaling at the phone's original rate. I thought you were talking about a savings versus the stock phone frequency. Everytime you're actively running the phone - it'll be at 520, which means it's burning more energy than a 6700 out of the box. However, otherwise - it'll be burning the same power as you would otherwise see on standby. Did I explain that clearly?

You're not going to be saving more energy than a stock phone, but you would only be burning more power when actually using the overclock. That's why I love to use my phone at 624 100% of the time, I've been doing it for over the last year. I don't think the power loss is significant at all. With a 2600 lithium polymer battery, I can still go 2 days of heavy use without charging - though my phone will be close to dead by the end.

sounds good to me...

too bad scaling will likely never work on dual processor phones such as mine...

I mentioned long ago in both Xcpu & PHM my findings on the 6700, to disable autoscaling & also increasing the "Bus" speed as being more effective than just CPU speed. :P

For those of you who want the best performance on a 6700:


Turn on "VJLumosII" (to disable auto-scale)
Open Pocket Hack Master.
Set speed to either:
520Mhz,260B,130m,65lcd
546Mhz,273b,137m,68lcd
572Mhz,286b,143m,72lcd
Goto settings/system/backlight.
Lower brightness.
Goto PHM & exit application.
Open "CPU-ID".
Verify unit still running overclocked without PHM.
Open "acbPowermeter" to verify unit using less Ma draw.
Open TCPMP player & open a video.
In TCPMP select file/benchmark and verify HUGE increase in performance.

There you have it...
Your unit will revert to normal when you turn off.
:)