0

What is the fastest way to capture a screenshot in Windows? A driver? Using DirectX?

Cody Gray - on strike
  • 239,200
  • 50
  • 490
  • 574
Andriy Tylychko
  • 15,967
  • 6
  • 64
  • 112
  • possible duplicate of [is screenshotting by DirectDraw faster than GDI way?](http://stackoverflow.com/questions/4975312/is-screenshotting-by-directdraw-faster-than-gdi-way) – David Heffernan Feb 12 '11 at 00:20
  • 2
    Do you want an answer to this question, or your other identical one http://stackoverflow.com/questions/4975312/is-screenshotting-by-directdraw-faster-than-gdi-way – David Heffernan Feb 12 '11 at 00:21
  • actually I wanted to keep each of them completely clear and well focused. my case is more complicated than both these questions – Andriy Tylychko Feb 12 '11 at 00:23
  • Sweet. I've never done a cyclic "vote to close" before. –  Feb 12 '11 at 00:24
  • Why do you care about the difference between DirectDraw vs GDI if *neither* is the fastest solution? It seems that this question completely encompasses your second one. (No, I didn't post this comment to both questions.) – Cody Gray - on strike Feb 12 '11 at 06:39

2 Answers2

1

A while back I timed the win32 api and the .net wrapper. Here is the complete blog post:

http://blog.bobcravens.com/2009/04/fastest-screen-capture-using-c-vista-vs-win7/

The upshot is that the .net wrapper in the latest version of Win7 is as fast as the native win32 api. I have never tried DirectX or the other options. The .net wrapper was fast enough for my purpose. I have heard that accessing the video drivers (like VNC does) is the fastest. I have nothing to back that up, but it seems reasonable.

Bob

rcravens
  • 8,320
  • 2
  • 33
  • 26
1

The fastest way would be to read mapped region of video card memory I think... Just an idea, never tried that.

mip
  • 8,355
  • 6
  • 53
  • 72