Today I worked on the gamma resistors. I carefully calculated the best values for each parallel piggyback resistor, and installed piggyback resistors on all three of the originals (150, 82 and 56 ohms). My original addition to the 150 ohm resitor 'converted' that to 27 ohms, so I took R to be 9 ohms (since this resistor was 3R). That left me needing values of 9*1.6 = 14.4 ohms and 9 * 1.3 = 11.7 ohms for the other two.
Since the first had 82 ohms, I calculated 1/14.4 = 1/82 + 1/? (ie: ? = 17.5 ohms). Best I had was a 16 ohm resistor, which I used. Back-calculating, I got 13.4 ohms (vs. 14.4 -- close enough) - a multiple of 1.5R instead of 1.6
The second had 56 ohms, I calculated 1/11.7 = 1/56 + 1/? (ie: ? = 14.8 ohms). Best I had, again, was 16 ohms -- so I used this. Back-calculating I get 12.4 ohms (vs 11.7 -- close enough) - a multiple of 1.37R instead of 1.3
Having done the above, I modified my diffusion filter to be dog food can plastic covers (2) separated by a 5mm air gap. Now there is absolutely no sign of LED spot brightening -- complete uniformity over the view area. This is awesome! I could afford to use the two plastic covers because my brightness is so amazingly high -- they had little effect on the final result as I had to turn my brightness down anyway.
Now I don't get that feeling of 'wrongness' in the images any more -- they seem to have a much softer feeling in the highlights. This is probably because the gamma resistors are now correct.
It was at this point, after I'd redone the gamma (note: all three resistors this time, not just the original two) and installed the new filter arrangement, that I noticed that Graham on track 46 of CD#1 is smoking a pipe. Well, i think he is
This tended to point towards improved clarity and image definition.
Attached image shows the single-board with the recently installed motor control section sitting at the bottom of the board. The blank area bottom-right is probably going to hold the 'negative video handler'.