Pre-allocate, MCCSS, Triangular Dithering-Pwr-1 Pwr2-Pwr3, 24bit 44.1 vs 48K or 96K?

QUESTIONS >
Can some one please help me get the best settings for my AUDIO in Sonar 8.5 Producer

Should I enable Read/Write caching or both currently unchecked "Not Enabled" and shows buffer size 128kb, should I change that? I've seen people bring it to 256kb or higher and enable both... why? 
Currently dithering in TRIANGULAR by default and I know Pow-r1,  Pow-r2 , Pow-r3 are more CPU intense (Pow-r3) being the most CPU taxing and the best sounding, if you're serious about your work then you want at least Pow-r1 or 2  but can this affect my projects, does the dithering take place in the analog to digital conversion process, in the playback/final mix only.. how does it work and should I change it? 
Using trigger and Freewheel, as I should and have been told.
 I do have the MMCSS checked and run windows 7,  should I uncheck that? 
Record Pre-allocate file, I run a solid state drive, should I use this feature?


I have more RAM than I'll ever need, top CPU, Solid state drive, everything happens immediately, audio interface with ASIO but for some reason I get better latency using WMD/KS which Sonar recommends you try before using the ASIO anyway, I get no latency and can use live plugins as if it were RTAS, hear it right thru the headphones, autotune, reverbs, compressors, digital delays etc...  
My computer is able to handle so much... but I want to make sure I don't screw with settings that I'll regret in the future, including sharing of projects and exporting to pro-tools or other DAWs.  Not only that, but any changes I make I want to try on brand new projects, obviously not something that is halfway thru with one of my clients. 
I currently use bit depth of 24bit  44.1 but my soundcard can handle more, should I change that too?  I have 64 bit but choose to stay in the 32bit environment because WAVES plugins and Bitbridge not able to work the bugs and make it run.  I have both versions 32 and 64 installed of course, but stick to the 32bit mostly which runs incredibly well and my projects get VERY INTENSE, many tracks, many v-vocal clips, re-wire with reason, use melodyne all at ONCE and my CPU is at most 50-60%  and thats me TRYING TO tax it.
Its time to step it up a notch, lose the fear, make changes, change audio bit depth, better dithering, change any buffers etc...
NOT just try it on a 5 minute test, but the key is to start a real project and get deep with it, and thats my fear.  fear of change and run into problems,...  why mess around with something stable that what works, but I have to move on too.. 
People tell me, max out your settings if you could, its better, consider it...   
 

Dithering triangular? switch to  pow-r3?
MCCSS checked ?
24 bit       44.1K vs 96K ?    or maybe just move up to 48K, some say 96K is too high and can get too intense on any large project and its not needed and the difference is not noticeable except it uses double the processing and size.  
Record Pre-allocate file settings?? 

anyone with a similar setup PC specs and runs all these higher numbers ...  
any input will be appreciated





 

Small but important correction Kalle, Dither is only applied to change in "Bit Depth" not clock rate. 
If all our work is at 24 Bits then dither is not needed or applied until you export to 16 bit for CD compatibly. 
So you are correct that it is not an issue while recording and mixing. 
There are deep readings all about it that will make you chase your tail. For a while I even stopped using it I got so confused. I can't "hear" any difference. But now I'm using Power 1. CJ posted once about each one works better for certain types of music so I'm going with that. Still can't "hear" any difference. 

Very good questions Ace. I often stare at those little settings and wonder too. But like you, my system is running perfectly so I'm not going to touch anything I don't understand. I hope someone explains. Might be better to ask in the X2 forum , lots more people to see this. And it's all the same still ( I think)  

I've read plenty of posts where people recommend changing the buffer size but I don't fully understand it and everything works so I leave mine right where Sonar put it at 256. 
I think a lot of settings and tweaks are left over from the days of crappy CPU and memory specs. We are clear of most of that now. 


I work at 44.1/ 24 and very happy with the quality. I could could change to 48/24 but I doubt if it would change the way my recordings translate into the real world, which is no longer Hi Fi anyway. 
44.1 works best for me as I use a Yamaha 01V which is only capable of that rate. 

Thanks for the correction, Cactus!

Thanks to all. Very good information