"TheEndless" wrote:
That wouldn't allow you to manipulate existing bitmaps, though. A SetByteArray() method would work no matter how the bitmap was created. Given your VNC example, you could create a full screen bitmap, and just call SetByteArray for every 16x16 section that changes on refresh, instead of having to create individual 16x16 bitmaps and draw them to the master. Better yet, you could set the byte array directly on the roScreen, further reducing the overhead... no bitmaps, DrawObjects, or DrawRects needed.
But i don't require to manipulate existing bitmaps. I am a modest man ("with much to be modest about") and can make do with only a little. In this case any of the 4 alternatives i described - or a 5th you propose.
What you describe is not what i thought of
setByteArray(), it is more of
DrawByteArray() - with behavior close to
DrawObject(). If they implement it the way you imagine it - great, i'll use it - thankyouverymuch! I was trying to appreciate the difficulties in implementing the link roByteArray->roBitmap (tough job, thinking of a black box), under the presumption that the simpler something is, more likely it is to happen. Maybe i under-estimate the cost of creating new bitmap by feeding it byteArray (and imagining it won't be copied but only its ref.count will be bumped) and later disposing it. But anything will be better than the current nothing.