Q) what is a better way to correct fuelling? percentage change using rpm x map or tps x gear or percentage change using rpm x rpm derivative?

background :

Ok - so I run a fuel gear correction table x and y are rpm and map and there is a table for each gear

This is because if I run through the gears on the road errr track - and the car having had its main tune done in 4th - then revving through the gears I found that 1st was lean - 2nd less so - 3rd less than second and 4th was spot on - but as that's slightly less than 1:1 gear ratio - so 5th and 6th are slightly rich when under full load - under light load and steady state (cruising along) its fine in any gear

So use a gear correction fuel table in the haltec that does a percentage change for the fuelling from -20kpa to 0kpa (na engine) in first by adding 7% fuelling , 6.5% in 2nd and 3.5% in third etc etc so that the full load revving through the gear the fuelling tracks the targeted af ratio correctly - and this has been working pretty well.

If I don't use this correction by gear setup - then there is a noticeable seat of pants loss of power in 1st gear compared to running the gear corrected percentage change enabled - and when its disabled the gears that the car was not tuned in are in some cases (1st) wildly out of sync with the desired target af ratio.

Today I saw a screenshot on the Haltech forum somewhere where someone was correcting based on rpm on one axis and rpm derivative (rate of rpm change) on the other

That got me thinking - wouldn't this allow me to just run one correction table to sort out the fuelling so it tracks more accurately and elegantly than the hamfisted method of blunt force correction by percentage with a table for every gear that I'm using at the moment?

Yes - it could take a bit more effort to setup and I'd need to datalog a few things and spend some time to see if its going to actually work - but I'm kinda thinking that in theory, it should allow "better" control over fuelling in different gears / transients than the method I'm using at the moment and only need one x,y table with no extra tables for each gear.

Has anyone tried this themselves or is the idea incorrect or am I only now realising something that's been obvious to other tuners out there for a while?

As each gear has a different rate of rpm change over time - 1st being the most change - 6th being the least change, a single nice x,y table of percentages to correct by should be able to handle all conditions and corrections on the one table for any gear and not need a table per gear?

I hope to hear some advise from you all - should it work - have you tried it - is the theory ok etc etc - or why its a good / bad idea