[XabaRus-1]

I found this line in the conf.ini as I was seeing what IL2:MLR had changed in my conf.ini

;ProcessAffinityMask=1

What is this for?

[Vacillator]

It supposedly tells the program which core of your processor to use (1 or 2 for dual core or 1, 2, 3 or 4 for quad). There has been debate about whether it makes any difference - perhaps if one core is 'busy' with something in the background it might help? Personally mine is set to 1 and setting it to 2 made no noticeable difference.

EDIT: Oops got the numbers wrong! See below.

[JAMF]

It supposedly tells the program which core of your processor to use (1 or 2 for dual core or 1, 2, 3 or 4 for quad). There has been debate about whether it makes any difference - perhaps if one core is 'busy' with something in the background it might help? Personally mine is set to 1 and setting it to 2 made no noticeable difference.

Removing the ";" makes it active. AFAIK core naming starts at "0"? But you can force windows to assign IL2 processes to a core of your choosing. If you have a quadcore and find most processes run on core 0 and 1, you can set IL2 to 2 (The third core).

Probably going to see more benefit from other tweaks, but it could reduce some stutter, duration of low fps moments, things like that.

[Boris The Spider]

I found this line in the conf.ini as I was seeing what IL2:MLR had changed in my conf.ini

;ProcessAffinityMask=1

What is this for?

Use care when enabling.

;ProcessAffinityMask=1 results


;ProcessAffinityMask=2 results

[chas1963]

So let me see if I got this right...If I have a "Duo Core" chip I could remove the ";" from ";ProcessAffinityMask=1", and I could see if it makes any difference in performance???

Is that right?

[RRuger]

So let me see if I got this right...If I have a "Duo Core" chip I could remove the ";" from ";ProcessAffinityMask=1", and I could see if it makes any difference in performance???

Is that right?

eeeeeerrr, yup, that just about sums it up, try the number "2" if you don't see any performance improvements with what you have now, meaning "1"

[chas1963]

So let me see if I got this right...If I have a "Duo Core" chip I could remove the ";" from ";ProcessAffinityMask=1", and I could see if it makes any difference in performance???

Is that right?

eeeeeerrr, yup, that just about sums it up, try the number "2" if you don't see any performance improvements with what you have now, meaning "1"

Mmm, cool...don't know how I missed that one.

[JAMF]

I was wrong in my previous explanation about the functions of the numbers.

This is how it works, if IL2 were programmed for multi cores:

=0 - OS decides
=1 - core 0
=2 - core 1
=3 - core 0+1
=4 - core 2
=5 - core 0+2
=6 - core 1+2
=7 - core 0+1+2
=8 - core 3

=15 - core 0+1+2+3

So I guess the options for a quad core user and IL2 are 0, 1, 2, 4 and 8.

[XabaRus-1]

But IL-2 was created before multi core CPUs were common or even coming into the PC world.

I wonder if it is to do with HT capable CPUs. When I do Ctrl-Alt-Del and look at CPU usage history I can select on graph for all CPUs or one graph per CPU. Now I have a P4 3.2Ghz not sure what the name of the CPU is but I think XP sees 2 cores. So I wonder if it would have any effect.

[RRuger]

But IL-2 was created before multi core CPUs were common or even coming into the PC world.

I wonder if it is to do with HT capable CPUs. When I do Ctrl-Alt-Del and look at CPU usage history I can select on graph for all CPUs or one graph per CPU. Now I have a P4 3.2Ghz not sure what the name of the CPU is but I think XP sees 2 cores. So I wonder if it would have any effect.

I played with this a bit and that's all I ever saw, so I said hell with it. I have a dual core 3.06 and when I do A-C-D I see what you see, no matter the combination of
=1 - core 0
=2 - core 1
=3 - core 0+1
=4 - core 2
=5 - core 0+2 etc...

[JAMF]

Like I said:

Probably going to see more benefit from other tweaks, but it could reduce some stutter, duration of low fps moments, things like that.

OS should do the job of spreading the workload over different cores. If you want to be sure, do framerate measurements in a simple track. Do not benchmark with BlackDeathTrack, as it varies much and is dependant on graphics card, overall system and a number of other factors.