Dear Sam Wise Gangee.. you are indeed wise for asking for quantification.
Do not let anyone put your questions down. It is always to take a scientific approach to things. No matter what the masses say, there is a scientific approach to problems even at the highest level. With streetcars, everything can and should be a known before proceeding - especially with such a microsystem such as piggybacks.
Truth is that when it comes to small % gain mods, the cost of accurately quantifying becomes a considerable proportion of the total cost (both time and money). This explains why so few people actually quantify it to begin with. The other problem is that large variances, inherent in the level of technology and that the street car enthusiast can get their hands on and the processes used to arrive at the results, the small gains are very easily skewed or even completely masked.
Here are some problems with the different forms of quantification.
Chassis dynos
Before and after dynos must be run. Engineering standards actually call for A-B-A. Dyno without mod, dyno with mod, dyno without mod again, and if gain/loss is equal then results are accepted. The real problem at the enthusiast levels, since no one has a air conditioned dyno cell, is finding days before and after days with the same air pressures, temps, humidity. In addition, how many dyno facilities actually run stabilize engine and differential fluid temps? How many reset ECUs or at least run a simultaneous diagnostic log to watch how fuel trims are being affected?
Now with these piggybacks tricking the stock ECU all over the place, how many go back to the dyno after a week of mixed driving, on a day where atmospheric conditions are the same as the pre-mod dyno, again stabilize all the fluid temps, again run and analyze the simultaneous diagnostic log?
Also, is the customer familiar with the ways the operator (product seller or possibly affiliated with sellers in ways we might not know about) can skew the results? Again this is dyno specific. Have we read about the dyno system to be used? Are we watchful of what the operator is doing?
Timed acceleration tests
Variances in atmospheric conditions, driver technique, condition of car, fluid temps, ECU mode of operation (read on previous threads how modern ECUs are so make and model specific that it takes lots of testing or extremely good code hacking skills to find out what it is programmed to do).
Still I feel this is pretty cheap test to run and can be done many times and quite non abusively. The problem is that people don't run enough of the tests, and keep detailed enough log of conditions. I figure if 5 pre-mod runs and 5 post-mod runs were carried out with conditions as equal as possible, and a trend was shown, that the trend would be accurate, and its degree quite accurate too. 1/4 miles, and gtec or similar device (properly setup) preferred.
Drag races against similar unmodified makes and models
Worst kind of quantification. All the same variables as with the timed acceleration tests - PLUS driver synchronization, inherent variances (from stacked tolerances) even between cars right off the same production line on the same day. Driver sync is the biggest problem. driver motives are also never truly known.
So either..
1. We do things as they should be done and come up with variances small enough to be minor when compared to the % power gain, OR..
2. We go high tech with race level engine dyno cells, etc. OR...
3. We get very technical and make absolutely sure that the theory side at the deepest level we can get to, matches what we are seeing in the above tests (still making sure they are finely controlled).
1 is nearly never done, 2 is not possible on our budgets, 3 is the best option.
FULLY understanding the stock ECUs is a greater task than fully understandinga standalone system since a standalone system's technical specs are fully known, and it is molded into what you want to make it. With piggybacks you are trying to fully understand the stock ECUs and then coming up with ways to sneak things around it. If your stock ECU test processes are wrong and you unknowingly come to a wrong conclusion, the sneakiness will be discovered by the stock ECU and chaos WILL ensue.
My views on piggybacks are listed on numerous previous threads here and on old Delphi. Nearly 100% of the technical arguments have been ignored, and no one has been able to say for certain, for example, "No, the stock ECU does NOT run test cycles." Or even describe how such a test cycle might be defeated.
I have no problem with piggybacks controlling open loop AFR and ignitiong timing. I have no problems with piggybacks controlling closed loop ignition. I have problems with people claiming to be able to permanently control closed loop AFR in cars running these modern ECUs. There has been nothing conclusive, nothing concrete.. just a lot of subjective emotional noise. I am not saying there is no way around it, just that modern stock ECUs have not been fully explored and understood and that current means of getting around them in closed loop are not perfect. This is evidenced by the driving characteristics of the car changing over time (sometimes thankfully noticeably, other times slowly enough not to be picked up by the driver) - the stock ECU is fighting for control and WILL have its way. Run a thorough search and find out how many people have experienced changing driving characteristics with piggybacks initially tuned to control closed loop AFR.
Now would anyone be so kind as to list the concrete operational characteristics (and how he arrived at each conclusion) of the E46 M54 ECU and explain clearly how he seeks to get around each problem with a couple hundred dollars worth of electronics and successfully control closed loop AFR
PERMANENTLY?
Feel? What is feel? The human body is such a huge variable vs claimed change in acceleration. Crossover pipes in V engines used to trick seasoned race drivers into thinking they were flying through laps and they'd come back to the pits all pumped up, only to discover that in reality they were no faster. These are some of the finest human G-sensors in the world, and they can be tricked by just aural changes. What makes us civilian fatasses playing with hp gains in the ones, superior to them?
Yours Fearfully
Smeagol
many Mazda 6 running with it now...
thankx to bernard from BVO n also not forgetting SL2 for generally donating his time slot he booked for my tunning :laughlik: cheers Bro!!! any update on your car yet?? :thinking: hope u get it soon....will be there to have coffee n mee pok with u while u wait... Brother mah( u wait for me i wait for u 2) 
heehee
