Image Vs Rect - Socoder

-=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- (c) WidthPadding Industries 1987 0\|157\|0 -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=-
Socoder -> Blitz -> Image Vs Rect
Sun, 03 Feb 2008, 16:08
Forklift_Fred	Odd little speed question for you. One that probably isn't relevant in anything I'll ever make but one that came to me just now as I'm about to use one or the other... Is it quicker to draw a solid rectangle or use a preloaded image? Baring in mind it would be no bigger than 32x32 and could be 300 of them each cycle (possibly more, that's just a quick estimate) So, Rect(x,y,32,32,1) or DrawImage(blank,x,y)? -=-=- Come rain or shine...
Sun, 03 Feb 2008, 16:11
steve_ancell	I'm not sure. But I do know that DrawBlock(image,x,y) is faster than DrawImage(image,x,y).
Sun, 03 Feb 2008, 16:15
Forklift_Fred	Hadn't even considered that! Makes sense as it isn't thinking about masking. I've never used it before but now you've mentioned it, it's worth remembering -=-=- Come rain or shine...
Sun, 03 Feb 2008, 16:17
Jayenkai	I was going to say Images, but then I tested it.. Graphics 640,480,32,2 SetBuffer BackBuffer() rec=CreateImage(32,32) MaskImage rec,255,255,255 ClsColor 255,255,255 Color 0,0,0 avg=0 For m=1 To 10 time=MilliSecs() Cls For n=1 To 5000 Rect Rand(0,640),Rand(0,480),32,32,1 Next avg=avg+(MilliSecs()-time) Flip False Next avg2=0 For m=1 To 10 time=MilliSecs() Cls For n=1 To 5000 DrawImage rec,Rand(0,640),Rand(0,480) Next avg2=avg2+(MilliSecs()-time) Flip False Next Cls Color 0,0,0 Text 1,1,avg Text 1,20,avg2 Flip WaitKey() --v Turns out it rects.. -=-=- ''Load, Next List!''
Sun, 03 Feb 2008, 16:19
Jayenkai	Graphics 640,480,32,2 SetBuffer BackBuffer() rec=CreateImage(32,32) MaskImage rec,255,255,255 ClsColor 255,255,255 Color 0,0,0 avg=0 For m=1 To 10 time=MilliSecs() Cls For n=1 To 5000 Rect Rand(0,640),Rand(0,480),32,32,1 Next avg=avg+(MilliSecs()-time) Flip False Next avg2=0 For m=1 To 10 time=MilliSecs() Cls For n=1 To 5000 DrawImage rec,Rand(0,640),Rand(0,480) Next avg2=avg2+(MilliSecs()-time) Flip False Next avg3=0 For m=1 To 10 time=MilliSecs() Cls For n=1 To 5000 DrawBlock rec,Rand(0,640),Rand(0,480) Next avg3=avg3+(MilliSecs()-time) Flip False Next Cls Color 0,0,0 Text 1,1,avg Text 1,20,avg2 Text 1,40,avg3 Flip WaitKey() --v Version 2, with DrawBlock. Block is surprisingly not all that good. Maybe 'cos there's no actual mask on the blocky image. So, still rects. -=-=- ''Load, Next List!''
Sun, 03 Feb 2008, 16:22
Forklift_Fred	Thanks for that. I feel so lazy now for not making a test myself... but while you've been doing that I've run into another little hurdle that needs a head scratch so at least there's one less thing to worry about Cheers. -=-=- Come rain or shine...
Tue, 05 Feb 2008, 16:23
mindstorm8191	Well Fred, you know what NASA says. A test is worth a thousand expert opinions. -=-=- Vesuvius web game
Tue, 05 Feb 2008, 18:46
Forklift_Fred	Yes but why devise a zero gravity pen when a pencil will do? I think I know what I mean by that... -=-=- Come rain or shine...
Tue, 05 Feb 2008, 19:04
Jayenkai	Because the pens sell quite well! (See also "Vista") -=-=- ''Load, Next List!''
Tue, 05 Feb 2008, 19:20
power mousey	Now, if Nasa can decide if there was that face or building structure that has a face in Cydonnia on Mars. Rectangular structure sorta of and images were beamed back showing it.