A most perculiar undervolting result

[furyoo]

I just bought a new Macbook, and before that was using a Fujitsu laptop. One of the best tweaks I had ever done to it was undervolting the processor core, using a free application called Notebook Hardware Control. RM Clock is another alternative.

Here's where the interesting part comes in: No matter how low I put my voltage in XP under BootCamp, there is no instability issue. I have my 12x multiplier set at 0.9500V, the lowest it can possibly go, and I've not had one BSOD yet. However, my fans still go into overdrive at high speeds, my CPU cores' temperatures also sore up. This hints to me that perhaps Bootcamp is blocking direct modification of the processor core, and I've not actually undervolted anything.

Does anyone out there have the same results? I have tried with both NHC and RM Clock. Perhaps if I had a way of undervolting in OS X I could tell if I could really push the VCore all the way down, but I don't want to pay $10 for Coolbook when the Windows equivalent is free.

 

[EspressoLove]

Whoa, that's crazy if Apple blocked even that, but on the second thought, I believe BootCamp just emulating BIOS for your Windows, maybe that's the reason you have no cool results

As for me, recent MB & MBP on my hands (both too hot for my laps ) so I was hoping to find something like HNC for Mac.
At some point I was ready to buy CoolBook, but now it seems like abandonware, and it doesn't work on SR MBP, not even reading CPU/voltage.

I had high hopes for SpeedIt, but it also sounds like abandonware now, they actually have a posted solution, but I never was able to install it properly.

But the biggest problem I see is ignorance & arrogance of general Mac Public, people usually steer clear from not Apple/3rd party hacks, even if they claim to help your pains, hence zero awareness, and low interest for such solutions like NHC or RMC. They are probably taken as too unholy for a Mac.

 

[furyoo]

Thanks for your reply. I think you'll find my recent investigations very interesting though. I'm using Coolbook on my laptop, and guess what my voltage is at 12x? 0.9500v. It's been at that setting for the past few weeks, and I've never had a hint of instability at all.

I do think it's working, because my core temperature in OS X used to rise to 80+ degrees celsius when fully loaded. It hits a maximum of 75 now, but very rarely. Needless to say, I'm very happy.

 

[EspressoLove]

Oh, that's good news, I also was thinking if OS X own implementation of throttling is active and helps keep MB/MBPs cool and happy.
Also there is a possibility that this controls were implemented in Bootcamp, that could explain that you didn't have tangible results, but still managed to control voltages & frequencies.

I wonder if we can have some solution which will let us go beyond permitted undervolting thresholds. I can remember running under 0.7 volts on my Centrino Dell lappy, that way I was able to keep it quiet, when it's fans started to spin a tornado, even under moderate load. (this has developed over time, past warranty expired as always)
And I was quite happy with 800-1000Mhz most of time, and easily available max 1,6Ghz.

That's what I look for my SR'MBP now, this time to simply keep it cool enough .

Can you please elaborate on your experience with CoolBook, do you get any difference if you strap it down with FIXED minimum Voltage & Frequency, under NORMAL use, with browser and few basic apps on.
(total CPU usage in 10-15% region)

As I said I don't have it working for my SR'MBP right now (all up to date), yet it's active (reads Voltage, Temp, Freq) on my gen2,1 MB (on 10.4.8)

 

[furyoo]

Apologies for taking so long to reply. I've been besot with work the past few weeks.

I wasn't aware that there was the possibility of undervolting below the minimal treshold set by Intel. I had always stuck to 0.700v for my old Centrino chip.

I'm using my Macbook under minimum voltage and frequency now, but with the CPU usage hovering about 15% or less there is no discernible difference in the performance from when it is locked at minimum voltage and maximum frequency. If I tax it further, I should experience performance hits. The temperature readings are also comparable for both cases.

I cannot say much about battery life because thus far, I've never had the misfortune of having to recharge when I bring my notebook out. I will do some quantifiable tests when I'm done with my work in a week or two.

Cheers!