a very lucid post. The biggest variable in the degree of the let's call them errotrs vary accross the screen are do to lens shift and thius are CA and not convergence. The constant error component would be convergence.
The appearance of misconvergence (constant) can be hidden by the misnamed convergence controls which of course do not adjust convergence at all but merely hide convergence errors by scaing the red and blue colors to change there position on the grid (the subpixel adjustments). Full pixel adjustments are done by changing the red and blue timing and are a good thing which no adverse effects. The subpixel adjustments have lots of bad effects which fortunately most are unaware of. a loss of focus and resolution.
A see the mention of 3 feet being offered as a metric as to whether convergence and CA errors are acceptable . What about 4 ft or 5 ft or 6 ft. Anything is rather arbitrary, of course the larger the error, the further away one would have to sit to make the misconvergence and CA not visable to the viewer. What should be your test. Not 3 ft unless that is the distance you normally sit from the screen. The test is whether you see it from your viewing distance. Duh? Its not brain science. If you don't see it, don't worry about it. Hard to accept, but good advice.
M
It also is interesting when you see say green on eitherr side of the primary color of the line, white.Think about this. Is the green misconverged? If the green was shifted to one side of the white, that would be misconvergence and/or CA. The gren of course can not be electronically shift by those mislabeled convergence controls. What does it show. The lens is more tightly focusing the red and the blue.
repeating earlier coments. If you see blue convergence error close up. Just ignore it. It won't be visable to your eyes at normal viewing distances. Be happy. If you can't be happy, get a single chip DLP projector and never worry about convergence because there is nothing to converge, it takes two or more chips before anything can be converged.
seriously, the JVCs and any projector have let's call them faults which lead to less thsan pefect test patterns. Just like anything, the JVC panels have design limitations and the lens has certain tolerances which like any lens can be observed. Mainly the problem with the JVCs lies with the panels. They are great in someways, not so great in others.
But these on balance are very fine projectors that one should be quite happy with if one just watches normally and not worry about photos of cse up test patterns. Enjoiy your very fine JVCs.
The appearance of misconvergence (constant) can be hidden by the misnamed convergence controls which of course do not adjust convergence at all but merely hide convergence errors by scaing the red and blue colors to change there position on the grid (the subpixel adjustments). Full pixel adjustments are done by changing the red and blue timing and are a good thing which no adverse effects. The subpixel adjustments have lots of bad effects which fortunately most are unaware of. a loss of focus and resolution.
A see the mention of 3 feet being offered as a metric as to whether convergence and CA errors are acceptable . What about 4 ft or 5 ft or 6 ft. Anything is rather arbitrary, of course the larger the error, the further away one would have to sit to make the misconvergence and CA not visable to the viewer. What should be your test. Not 3 ft unless that is the distance you normally sit from the screen. The test is whether you see it from your viewing distance. Duh? Its not brain science. If you don't see it, don't worry about it. Hard to accept, but good advice.
M
It also is interesting when you see say green on eitherr side of the primary color of the line, white.Think about this. Is the green misconverged? If the green was shifted to one side of the white, that would be misconvergence and/or CA. The gren of course can not be electronically shift by those mislabeled convergence controls. What does it show. The lens is more tightly focusing the red and the blue.
repeating earlier coments. If you see blue convergence error close up. Just ignore it. It won't be visable to your eyes at normal viewing distances. Be happy. If you can't be happy, get a single chip DLP projector and never worry about convergence because there is nothing to converge, it takes two or more chips before anything can be converged.
seriously, the JVCs and any projector have let's call them faults which lead to less thsan pefect test patterns. Just like anything, the JVC panels have design limitations and the lens has certain tolerances which like any lens can be observed. Mainly the problem with the JVCs lies with the panels. They are great in someways, not so great in others.
But these on balance are very fine projectors that one should be quite happy with if one just watches normally and not worry about photos of cse up test patterns. Enjoiy your very fine JVCs.
Enjoy! I have absolutely loved my 40/45 for 2d.
I think it's generally agreed not to use horizontal shift but if it's level with the top of the screen and the right way up like mine it is still as sharp as when dead centre as far as I can tell (long throw setup though). If it's inverted I don't know if this would result in more or less shift than I have, my feeling is probably less, so at least the nay sayers should be happy. 
