Audigy

In the meantime I've been playing around with RWe and the SMBus. It seems that in my system(ICS9LPR363) writing all at once can sometimes randomly crash the system instantly. I've altered the routine in a way that only one byte is changed per run. That way I've stopped this to happening. My script(still in a basic state) consists in calling several bat files, each one with one run. In the first I change the voltages to max and disable the IDA, then run 3 bat files with one byte change each(only need to change 3 bytes). main script bat: call OCVID.bat timeout /t 10 /nobreak call PLL.bat timeout /t 10 /nobreak call PLL2.bat timeout /t 10 /nobreak call PLL3.bat one of the ppl.rw examples(current byte change to 7F in the offset 09): >local0=0x7F00000000000000 >local0=bswap(local0) >smbus write block 0xd2 0x09 0x09 >rwexit

This is not 100% related but I too had some cooling insufficiency on my Clevo M571RU, mainly on the CPU side. The northbridge heatsink was pretty much useless too, there was a 2mm gap between the base of the heatsink and the core, and they were using thermal pads to solve(poorly) the problem. Tried pretty much everything until one day I realized that it was better to made the custom parts myself. Started with the northbridge, I was noticing some discomfortable heat exactly on the area of the northbridge, in the palmrest. At the same time I was having some difficulties regarding overclocking +20Mhz over the stock FSB. Since the RAM timings were quite relaxed and I was certain that I was suffering memory controller instability. Using an old cooler of one ATI X800 I made the necessary modifications for it to fit, including a spring retention system that natively was absent. That way I eliminated the gap between the base and the core, and the dissipation capacity was improved greatly. I started being able to OC to +30Mhz over stock FSB without any problems and without the same heat in the palmrest. Then I organized my efforts to the CPU heatsing. Since little could be done to improve the little and flimsy single heatpipe, I tried to improve the contact between the base and the heatpipe using a shroud that would improve the native one sided transfer to all over the heatpipe. I then realized that I had an old MOSFET heatsink laing around and I tried to alter it to fit the base of the heatsink, spreading the heat to the opposing side since I had some room there. Used some Artic Silver Thermal Epoxy and glued the two pieces. The CPU temps decreased significantly, and I was finally able to use my system without headphones to muffle the noise of the cpu fan(joking). I refitted the cooper heatsinks in the GPU heatsing too, using the same Artic Silver Thermal Epoxy. Before(sry no better one): Now: So my recommendation is, if you have the necessary tools try to add more metal to the heatsing, improve it somehow. Don't fix the problem with more noise, you will be fed up soon after.

I'd like to say thank you for this magnificent work. I've been using RW everything for some years now, mainly to write to my C2D's registers so I could disable IDA and change VIDs(voltages). I tried in the past to figure some way to make RWe write to the PLL registers via commands, but never got it working. Thank you for pointing me in the right direction. I know that nowadays people ditched the old C2Ds and are now running on i7 and so forth, but for people like me that still have that old laptop at home and want to give it a new life, I would like to suggest some addictions to your script. For control and stability purposes, it would be wise to include an IDA disable routine plus VID modification for everyone that still use those models. What we need to do is to write directly to the MSR registers via the address 0x199. Bit 0 of EDX enables or disables IDA, a functionality that is useful performance wise but can lead to instability if used while OCing. Changing that bit to 1 disables IDA. Next is the VID control area which consists in the first two hex values of the EAX. It is useful if you want to fine tune your OC while keeping the temperatures in their lowest possible values, reducing the voltage until the desired frequency is no longer stable. We can even create very low power usage profiles, with lower voltages and if desired lower frequencies(or SLFM) that can easily be toggled when OC is not needed(like doing some light work or just browsing the web). The syntax of the command is this: Rw /Min /Command="CPU 1;WRMSR 0x199 0x00000001 0x00000D25;CPU 2;WRMSR 0x199 0x00000001 0x00000D25;RwExit" Relevant MSR structure: To the date I've been using both my command and your script separately but I think that integrating the MSR command into setPLL.rw is not difficult. Would this syntax be correct? >CPU 1;WRMSR 0x199 0x00000001 0x00000D25 >CPU 2;WRMSR 0x199 0x00000001 0x00000D25 >local0=0xFFFFFF0000000A31 >local1=0x187F15CDC4E92F00 >local2=0xFF6108D534000000 >local0=bswap(local0) >local1=bswap(local1) >local2=bswap(local2) >local33=sub(0x3C , 1) >smbus write block 0xd2 0 local33 >rwexit