Is it possible to save a selected rendered area only?

Zev0Zev0 Posts: 7,046
edited December 1969 in The Commons

If this can be done in studio please let me know. It is really handy. I used this feature a lot on Poser to make corrections to a scene. So instead of re-rendering the whole thing, I just re-rendered the edited part and saved as another png or Jpg and layered it on top of my full render in photoshop.

Comments

  • murgatroyd314murgatroyd314 Posts: 1,440
    edited December 1969

    I don't think there's a way to do it in a single step, but you could try doing a spot render and taking a screenshot.

  • ReDaveReDave Posts: 815
    edited December 1969

    Rob made script examples here: http://docs.daz3d.com/doku.php/public/software/dazstudio/4/referenceguide/scripting/api_reference/samples/start#rendering
    one of those (rendering to viewport) would need to be hacked to make it spit out an area render in a new image. I tried briefly to do that but my DS scripting skill is abysmal. :-S

  • Zev0Zev0 Posts: 7,046
    edited December 1969

    lol so what one is it? There are 3.

  • JimmyC_2009JimmyC_2009 Posts: 8,891
    edited December 1969

    Render to Vviewport as Dave has said :)

  • Zev0Zev0 Posts: 7,046
    edited September 2012

    Good Lord I am blind..LOL Just looked at it. Doesn't really help. If you run the script it renders the entire scene plus only at viewport size. What I want is my area selected area to render only and be able to save that. But at the same time save the image as the exact size as my render settings dimensions ( in a seperate window) so I can crop and overlay it on my full render.

    Post edited by Zev0 on
  • TimmothTimmoth Posts: 108
    edited December 1969

    I used to do the exact same thing in DS3A by spot rendering and taking a screenshot, then copy paste the area.
    For some reason DS4 won't spot render the same, so everything is out by a pixel or more when you try to copy paste a screenshot, the bigger the spot render the more it's out, which sux.

  • Miss BMiss B Posts: 3,071
    edited December 1969

    I usually have two cameras in each of my scenes, one the default camera set up and a render camera that I lock in place once I have it where I want it.

    What I do then when I'm, for example, trying to zoom in close to do hand/finger posing, is click on the hand, switch to my default camera and click on the camera tool that looks like a plus sign inside brackets --> [+]

    That will zoom in where you want it, though you can then tweak the default camera to exactly where you want it, and then render it. Using the spot render does it directly in the viewport and you can't then save the render, but using the default (or any "other" camera you set up) to zoom in, you can then render in that camera.

    Hope that helps. ;-)

  • Zev0Zev0 Posts: 7,046
    edited December 1969

    I think you are miss-understanding what I am asking. I do not want a zoomed in area render. I know how to do all that. I just want to re-render an area of an image EG top corner, and save that as a png. So what I have on my saved picture is my entire scene, but only my selected area is rendered, the rest is still preview but its all on the image and still set as the dimensions in my render settings. So what I have is an image with top corner rendered and the rest un-rendered. Place that on my fully rendered image, cut out the un-rendered part of the re-render and merge in photoshop. This way the edited part overlaps the full render image exactly. Camera and image size must remain the same as original render other wise edit will not match original image. Hope I am explaining clear enough. I used this technique in poser for years, and it saved me a lot of time. Instead of a full re-render, just render the area you made changes to - eg a different person in the background and overlap it in photoshop.

  • Kendall SearsKendall Sears Posts: 2,995
    edited September 2012

    Zev0 said:
    I think you are miss-understanding what I am asking. I do not want a zoomed in area render. I know how to do all that. I just want to re-render an area of an image EG top corner, and save that as a png. So what I have on my saved picture is my entire scene, but only my selected area is rendered, the rest is still preview but its all on the image and still set as the dimensions in my render settings. So what I have is an image with top corner rendered and the rest un-rendered. Place that on my fully rendered image, cut out the un-rendered part of the re-render and merge in photoshop. This way the edited part overlaps the full render image exactly. Camera and image size must remain the same as original render other wise edit will not match original image. Hope I am explaining clear enough. I used this technique in poser for years, and it saved me a lot of time. Instead of a full re-render, just render the area you made changes to - eg a different person in the background and overlap it in photoshop.

    The only way I've found to do it was to make sure that the render area is smaller than your monitor's display. Then you can render at full size by putting the viewport a the correct size. For instance, if the render size is 1024x768 then I'll use my 1280x1024 monitor to put the size of the render frame at 1024x768. Then I can "spot render" and the bitmap sizes will match. Otherwise, that's a no (at least from within DS), there are other options if you're using the standalone 3delight.

    EDIT: Another option, if you're just worried about speed, is to use a mask. Allow only that part of the camera view to be visible through the mask. The render will zoom past the masked parts and only "render" the visible area.

    Kendall

    Post edited by Kendall Sears on
  • Zev0Zev0 Posts: 7,046
    edited December 1969

    Ah. Masking sounds more like a solution. How do you do it?

  • Kendall SearsKendall Sears Posts: 2,995
    edited September 2012

    Zev0 said:
    Ah. Masking sounds more like a solution. How do you do it?

    Set a plane in front of your camera, and then create a transmap.

    EDIT: Or more simply, set the number of divisions appropriately high and set the visibility of the necessary polys via the PGE.
    EDIT2: Or use multiple opaque planes to block the unwanted areas. (like pieces of cardboard taped together :-) )

    Kendall

    Post edited by Kendall Sears on
  • Kendall SearsKendall Sears Posts: 2,995
    edited December 1969

    Zev0 said:
    Ah. Masking sounds more like a solution. How do you do it?

    Set a plane in front of your camera, and then create a transmap.

    EDIT: Or more simply, set the number of divisions appropriately high and set the visibility of the necessary polys via the PGE.
    EDIT2: Or use multiple opaque planes to block the unwanted areas. (like pieces of cardboard taped together :-) )
    EDIT3: Silly, but an option: Use Hex Bridge to cut a hole into a thin cube. This will allow complex shaped masks, and other effects.

    Kendall

Sign In or Register to comment.