Recent changes in get_opengl_mouse_state@ ?

This is another devilry. 30 years this worked perfectly and abruptly it failed. Looks like in 902 the

get_opengl_mouse_state@ (ix,iy,iflag)

(iflag=9) stopped to get any different iflag when you push simultaeously left Ctrl and left mouse buttons in OpenGL screen.

The RightCtrl+LeftMouseDown generates iflag=9 OK. Mouse and keyboard interruptions are handled like this:

   i=winio@('%pv%^og[double,depth16]&', iwid, ihgt,CBmouse) 

I use get_opengl_mouse_state@ and not CALL_BACK_REASON because

clearwin_string@('CALL_BACK_REASON')

does not offer an option when you keep Ctrl and left mouse down simultaneously (it still shows the reason ='MOUSE MOVE' as if Ctrl key is not down)

Both these interrupt functions are called inside CBmouse to handle plot moves.

/* The OpenGL implementation of this compiler was incredible good from the start in 1990th. And shock is that not a single user of this compiler found it in 1/3 of century. This correlates with even more surprising fact that not a single user requested mainstream ways of multi-processing with this compiler. I fail to understand the Fortran users, and find the key explanation of this absurd in the same devilry and if devilty then The Matrix too. Because if not devilry and not Matrix then what? Altzheimer and dementia?

There has been no change to the ClearWin+ code for this function but it is still possible that some other change (e.g. a upgrade to the Windows API) could make a difference.

But we can only investigate when supplied with a demonstrator together with details of the platform (x64 or Win32) and the operating system.

v9.03 is available from this Forum but it is unlikely to make a difference in this context.

Dan,

Why I don't use OpenGL: (1) I can't find a user guide to OpenGL written in plain English (2) Ordinary ClearWin+ graphics are good enough for what I do.

Why I don't crave multiprocessing: (1) My PC is thousands of times faster (and effectively larger capacity) than the mainframes I learnt programming on c. 1970-71 (2) Everything I program runs almost instantly (3) The minute you have a user interface, you slow down to the pace of the user - in my case, it's very pedestrian.

Nevertheless, it would be nice to have that user guide, and to use all the CPUs simultaneously, wouldn't it?

Eddie

Eddie, I am talking about entire pool of users when you can apply statistics and can try to make some reasonable conclusions ( so far i found no reasonable ones though). And at least you are one of those extremely rare individuals here who understand what would be good and bad for the compiler. Talking with others is like hitting with the head thick concrete wall.