> > I still don't believe the problem is in the runtime. Below is a brief >> demo. Do you see any flash when you run it? I have a moderately fast >> machine, so that may be masking it, but it looks smooth as silk when >> I run it here. Check on the sequence of building window, setting >> colors, creating EF. Sure it's great to have a hackable runtime, but >> if it ain't broke... :-) >> >> e-e >> =J= a y >> " >> >> dim as rgbcolor backRGB,foreRGB >> dim r as rect >> dim i >> foreRGB.red = &FFFF 'Red text >> //backRGB.green = &FFFF 'unREM for Green background >> >> setrect(r, 0, 0, 500, 200) >> window 1, "EF on colored BG", @r >> RGBbackColor(backRGB) >> cls >> rgbforecolor(foreRGB) >> for i = 1 to 5 >> edit field -1, "Watch for flash", (50,50)-(450, 100), _statnoframed, >> _centerjust >> setselect 0, 99 >> edit text #1,_sysfont,40,_boldbit%¬ >> //,,foreRGB, backRGB 'unREM for appearance runtime >> delay 1000 >> edit field close 1 >> delay 1000 >> next >> end >> >Jay >Thank you. The demo worked fine but It does not replicate the problem here. >I get the same result in my application when I look at the window at the >point of closing the Edit Field. The problem shows up during the transition >between two functions. The first function contains the edit field and EDIT >FIELD CLOSE, the second function puts text on the screen with no edit field. >Both functions have the same RGBBACKCOLOR. The EDIT FIELD CLOSE statement >must trigger a window update event with a white CLS or white PAINTRECT >because I can slow down (or freeze) the action at the window update portion >of my code and see a white window. I use PG:PRO and tried several >combinations of window settings but got the same problem. Roberts hack gives >acceptable results. However, the first pass always produces a flash but >subsequent passes are fine with no white flash. I can live with that. >Barrie Barrie, Have you set RGBBackColor just before you CREATE the EF? If so, I'll shut up, but I don't think I saw that in the example you posted. I'd still like to have a demo that illustrates your problem. I know Robert's work-around is adequate for your purposes and you'd like to move on to other things, but I'm interested in: A) finding out if there really is a bug B) clarifying how colors in EFs work C) not propagating an impression the runtime is faulty if it's not If you have to move on, I'll understand, but I hope you can spare a little more time to help figure this out. e-e =J= a y "