I'm going to ramble on a bit with this because I've been thinking about this for a while.
I think you have to tread carefully. Is this a +T 8 valve? Before I had ms3x I used det cans exclusively to monitor knock.
I had a friend drive the car while I listened in on a set of det cans. At the time, the car was really loud, so "hearing" knock even with the det cans was a bit like listening to the ground for hoof-beats while a cargo train rolled by. I would advance the timing a bit, make a pull on a deserted road and both listened for knock and tried to see if the car felt faster. I had no way of estimating how much power I picked up so one method I used to kind of get an educated (maybe not) guess was to use the torque feature in megalogviewer and guesstimated a torqueincrease every time I advanced the timing. (it was a really sketchy way to go but it was the only thing I had to work with. More on this below) The road was flat, I was starting the run in the same place each time etc. I advanced the timing until I heard a bit of knock, then backed it off a couple degrees. I was never really comfortable with the 8v head. With conversations with others much more knowledgeable and experienced than I, the 8v head didn't start really making power until it was knocking a little bit. I deliberately avoided knock at all costs around the torque peak (about 3500rpm for the 8v), but past the torque peak into the 4's and 5000's rpm a little bit of knock was ok.
I recommend the friend to do the driving for you so you can listen in. Driving, tuning, and listening in for knock all at the same time is a little nerve wracking. You're concentrating hard to listen to see if you're engine's about to blow up and miss the red light and hit the kid on his cellphone standing in the middle of the street.
The 16v was a different animal. (16V +T...) First thing is the torque peak is higher (4-4500rpm). Cylinder pressure is at a maximum at torque peak so that's why it's nice to keep it from knocking there. With e85, I found I could add an additional 10 degrees beyond my already lame and probably too conservative pump gas map. This time I'm using the ms3x knock chip to help sort out the knock from the noise. I've forced the car to knock just to know what it "sounds like" or rather looks like when I'm observing the logs by using cheapo 87 octane. It's a possibility that I've added enough timing to actually go past MBT... I don't have a way of doing a spark hook test on the street, but the 10 degrees of extra timing keep making more power. I will experiment more with it this summer at the track. The one thing I found at least with e85 is that being too rich will cause the thing to knock
Obviously, the best way to really know how much timing it use is to take it to a dyno. Assuming you already have a decent base tune with your afr's well sorted, it should be a simple matter to add the timing and figure out what it'll do with a spark hook test.
I get the idea that dyno time is expensive. Which is probably why I haven't been to the dyno since 2007 and I've replaced 5 redblocks that I've done blowed up.
The other way to do this is just take it to the track. Make a pull, verify that it's not knocking on what you've currently got, advance the timing a smidgen, make another pull and see if the car's faster.
From earlier... what we really need is a system that gives a reliable estimate of engine output so we can advance the timing and get a realistic/reliable/repeatable measure of engine performance while street tuning. I tried a couple things with megalogviewer... observing RPM/sec, m/s2, torque, horsepower and the data was always... ALWAYS too noisy to make any REAL determination that that particular run was an improvement over the previous if you catch my drift.
I then got the idea to use the fancy GPS speed thing that tunerstudio sells... maybe I could get a speed vs time reading... never worked.
Some future ideas... rig up a VR sensor to one of the front wheels to calculate wheel speed vs rpm. Tunerstudio offers this as an input for the torque calculation. I haven't gotten around to it yet but I think that's my next avenue to explore... because basically I'm cheap and want to go fast but don't have the money for dyno tuning.
EDIT.... if this is for a b21FT then everything I said (most of it) doesn't apply. With 7.5:1 compression ratio you probably don't even need the 100 octane. You probably haven't reached MBT yet on pump gas.