I'm really talking about how a tuner decides what to do next, not so much the programming or specific tuning software a guy is using. As a basis, a tuner is trying to do 2 things: 1. maximize power, and 2. an even bigger priority in most circumstances is to keep the motor safe. Essentially, the goal for a race car is to lean the AFR and advance the timing to MBT the limit of detonation, which can usually be seen on a dyno or by improved results at a track. For a street car, the goal may be to stay away from detonation altogether. These two goals are not, despite popular opinion, the same goals. So the old way before dynos, widebands, datalogging, etc... was to get a motor running smooth on a safe rich low timing tune and take it for a blast down the strip. Dial the fuel/air mixture by changing carbs/jets/carb fine-tuning until you're not getting any faster, then start dialing in the timing advance. We all know how to listen for knock, and some of us still know how to pull the plugs and read the subtle indicators of what's happening in each of the cylinders. Little, by little, your car gets faster, and the tune gets better. Along comes the dyno that significantly shortens this procedure and removes driver inconsistency as a variable. It has it's own disadvantages like not having proper load as if the car were actually driving on the street accelerating its own mass, etc... But the major advantage here is a document showing how much power the car is making at every point in the RPM band. Suddenly, you can more accurately fine-tune specific parts of the band. You can see power drops associated with too much detonation. Let's not forget the biggest disadvantage with a dyno: money! A pretty reasonable $75 dollars an hour means that most backyard guys are only going to pay for a WOT tune and we'll try to work out any major driveability kinks either on our own or with a street tune. Next come the widebands and dataloggers that DIYers could afford. Suddenly, the DIYers can see the AFRs across not only the RPM band at WOT but even at partial loads and idle. We know, as a rule, NA AFRs will range from 14.7:1 at idle and very light throttle (or leaner if your motor likes it), 14:1 to 13:1 at part throttle, and around 13:1-12.5:1 at wide-open throttle. Boosted AFR targets are normally the same as naturally aspirated engines outside of boost, but with boost we run a lot more rich to leave extra fuel in the chamber for cooling and preventing lean spots reducing detonation. We usually shoot for 11:1 to 12:1. So.... awesome. With widebands, we can get pretty close to optimum AFRs. But how do we do timing? We're basically stuck to the old-fashioned method of advancing timing slowly and listening for detonation, reading plugs, etc... This is really problematic, though. My car is so loud I'm never going to hear it. Reading plugs, is time consuming because it requires you to literally shut the car off at a given load/rpm and the plugs tell you what was going on in only the few seconds before shut down. Pull all 8 plugs at every RPM and load? Not so feasible, but again doing this once or twice at particular points will get you in the ballpark. Want to tune the timing on the dyno? You'll more easily find MBT (mean best torque, or max brake torque, most best timing or whatever), but that doesn't mean you're safe. It just means that under those load conditions, at that temp/humidity you made the most power. This also typically happens after a bit of detonation too. So, we get somewhere in the neighborhood and then we pull 2-3 degrees back out of the tables to pull us out of detonation and to account for moderate changes in the weather. Man!....With so many variables, this is getting complex.... we're adding extra dimensions of complexity. So far, I've only scratched the surface. RPM, load (MAP or MAF typically), fuel, spark, IATs, engine temps, EGTs, tip-in. You don't really get an appreciation for the complexity until you try going through the tune on a stock EEC IV from almost 20 years ago. God knows what the factory is doing now. The stand-alones are simpler, but then again you're never going to invest the time, expertise, resources, and money that the factory engineers could into your car's tune. Now, I'm sure we've all heard about knock sensors in factory cars to protect the motors from poor gas, an unexpected problem, etc... But if this is reliable, the tuning help would be awesome! Now you could tune with relative confidence you're going to be safe and at the same time, you'll be able to see if you push the car over the edge. I personally don't want to be at MBT. I want to be out of detonation, even though a little detonation is supposedly ok. If you can use this thing to actually see what's going on in each cylinder and you can tune individual cylinders on the street/track as a DIYer, this is the holy grail. As an extra measure of security it's nice, but as a tuning tool it could be amazing! The pessimist in me is sure it's too good to be true, but I'm very interested in pursuing this if it's reliable. Effective knock sensors that retard timing on individual cylinders and allow for datalogging would be amazeballs! Then you will quickly be able to see where to add/remove timing to get to what I would consider optimal: the most timing and least fuel possible prior to detonation. I would probably still factor in an extra degree or 2 across the board, in case this system failed AND some other factor would push the motor dangerously toward detonation. As an engineer, I like a double-redundant system. I also like how easy it seems it would be to be able to easily and efficiently see/know that I've properly tuned all the way across the RPM band. There are probably plenty of them in the aftermarket, but this is the one I've been reading about: http://www.jandssafeguard.com/NewUniversal/UniversalVersion.html I hope it works for my motor with with the MSD digital 7+. If it's as good as I hope, then tuning should be a breeze and it should be safe.