If it was my reference to the 911 & PW test of the SciVision MAF on a Lux you refer to I think it loss about 0.4bhp. Not significant in other words, but also not a gain. There were also poor running issues as I recall, but also note the supplier sorted it out by doing something undesclosed in their workshop and it showed up with a small gain and improved drivability. Of course the key benefit of the back-to-back nature of the test was lost when there was a period of poorer running in the middle and who's to know if it ran better than stock or just better than badly after it was fixed?
Later cars don't have any more useful triggers than the Lux, but they might well have different ones. I know the Turbo and S2 use different systems for example. My expereince comes from a '90 Turbo, but I can't see how a Lux would be anything other than less advanced, and the key problem is that the DME doesn't use a reference sensor based on the camshaft like the majority of standalone systems need, so you need to fit something on the camshaft as a trigger. It's not the end of the world to do, especially if you go the extra mile and ditch the need for a distributor (as you have the end of the camshaft to play with then), but it still needs to be done. I also had to fit a speed sensor the EMS could use as it didn't support the oddball one the Turbo has as OE.
I tried mapping on the road, using a management system that supports dynamic mapping using a wideband sensor. It's not as easy as it sounds, and it's generally not possible to do the ignition (there are a couple of devices that might help such as a spark plug with a pressure sensor in it, but I doubt it would be worth the expenditure). I got a map on my car that was safe from the perspective of the engine not destroying itself, but that ran like a bag of bolts before I decided to get an expert involved.
We then mapped it "properly" on the road using a wideband in the tailpipe and det cans, and trust me doing that in 4th gear in a car that will hit close to 150mph at the redline of that gear on the public road on a wet day was no fun whatsoever. No way could you do it by yourself and no way would I do it again that way. In fact had I known what we were going to do I wouldn't have done it that way in the first place as it was both highly illegal and arguably dangerous, the only mitigation being the road was very quiet.
That map was good when warm but still ran badly when cold, and it wasn't impressively quick. When it was dyno'd on the "Rollers of Truth" it made 306bhp, which is abysmal for a Turbo with the mods it had. I then removed the ISV and fitted coil on plug CDI and had it mapped truly properly on the Rollers of Truth and it came up to 359bhp and ran & idled sweetly. Quite an improvement over 306 and I think clearly in the main attributable to the map. In truth it was still disappointing but at least I was at last confident it was giving all it was able to without improving the breathing. As an aside it blew itself to bits before any more development happened, and that was due to extreme detonation on #4, but I'm pretty sure that was down to a fuel delivery problem (Lindsay Racing's cheap "uprated" fuel pump?)
Anyway I recount the above to highlight the difference between the map I could get on the car trying to do it myself on the road and what the car was capable of when done properly on a dyno. I'd only map on a dyno in the future, and auto-mapping is a nice idea that doesn't deliver. I guess it got me fuelling that I could trust to get the car to the dyno without detonation or bore-wash, but that's about it.
Lastly, and again not trying to be a wet blanket, I'd caution spending excessive time/effort/cash on Lux tuning. While nothing would ever be achieved without people pushing the boundary and ploughing their own furrow I have to say if it were that easy and gave worthwhile gains it would have been done long ago. Also unless you are resticted by the regulations of a race series there is little point tuning a base model when an off-the-shelf more powerful version is available that will provide far better performance than you'll ever achieve with a Lux, quite possibly for not a lot more money and which will hold most of the value you sink into it.
I can't see John Milledge giving up the source of his camshaft blanks; my indie found out where (also in Germany) Porsche get their cams from and he guards that secret well as it means he can replace S2 and 968 cams for less than pretty much anyone else. If JM gives up his supplier and presumably his profile specs so you can get a cam ground in the UK then he's opening himself up to a competitor of his own creation.
] (doesn't normally stop me, but then I'm a bit thick [
