Text archives Help

---

Yeah, I don't see a good reason for allowing -imagedisplay to make several (since it actually makes multiple rendering targets), instead of the MultiImageDisplay which you made. You could use a pair of - imagedisplay to get the same output, but twice as slow ;)

On Feb 14, 2008, at 11:01 AM, Abe Stephens wrote:

Is this a bin/manta problem or something in the MantaInterface implementation? The code that dumps frames to disk (for camera path animation) replaces the OpenGLDisplay with a MultiImageDisplay, where one of the child displays is the original gl display and the other is some type of file display.

It might be the case that a separate channel is being setup for each -imagedisplay on the command line. I'm not sure why this would be useful to bin/manta which only displays one window.

Abe

On Thu, Feb 14, 2008 at 9:31 AM, James Bigler <bigler@sci.utah.edu> wrote:
It may have served some purpose in the beginning (adding more multiple displays), but I don't think it works as intended. I think for the sake of simplicity, this behavior should
be seen as a bug and fixed.

Thanks,
James

Thiago Ize wrote:
> Each time -imagedisplay is passed to manta as an argument, it renders an
> extra image (I think, at the very least, it looks like that according to
> performance). This can make things appear to be slower than they really
> are. I noticed this when I passed -imagedisplay null and - nodisplaybench
> to manta and found my code to run twice as slow. Should this be seen as
> a bug or is this perhaps a required feature?
>
> Thiago

---

• [Manta] -imagedisplay, Thiago Ize, 02/13/2008
  • Re: [Manta] -imagedisplay, James Bigler, 02/14/2008
    • Re: [Manta] -imagedisplay, Abe Stephens, 02/14/2008
      • Re: [Manta] -imagedisplay, Solomon Boulos, 02/14/2008

---

Archive powered by MHonArc 2.6.16.