Loading...
> >

Best Practices - Workflow

Db3dDb3d Posts: 248
in New Users

I suppose I am like many in that I am self-taught to use DAZ3D using documentation and online tutorials.  As with anything self-taught, you also self-teach yourself bad habits because you don't know any better.

For this reason, does anyone know of a "best practices" for workflow?

For example, is it better to start by adding the environment, then the props to build the scene, etc.?

Or, is it all simply a matter of preference?

 

Comments

  • BeeMKayBeeMKay Posts: 7,019

    It's a matter of preference, though in some cases, an environment will want to reset your scene, and merging it into the existing scene will not always work with render settings, etc.

    On the other side, if you are tinkering around with the character and go through several test renders until you have found the perfect combination, having an environment might slow down you render times.

  • ChezjuanChezjuan Posts: 515

    My basic workflow for Iray renders, which I won't call a "best practice," is below. I find it the quickest way to get the end result.

    1. pick scene elements (characters, environment, props, etc.).
    2.  Load the figures.
    3. Pose the figures (naked).
    4. Add clothes and figure-based props. This prevents the poses from undoing things like foot poses for shoes or hand poses for weapons.
    5. Set the camera and turn off the headlamp and the "preview Lights" setting.
    6. Load the environment and position it around the characters if needed.
    7. Set the Aux viewport to the camera so I can see what the scene should look like.
    8. Load other props (background items).
    9. Turn on Iray preview in the aux viewport.
    10. Set the lighting. It could be HDRI lighting, or a 3-point light system, or both, depending on the scene specifics.
    11. Run a quick test render at at least 1/2 of the final planned res so I can check for things that may not be evident in the aux viewport (like characters floating above the floor, poke through, etc.). I don't run it to completion all the time, it depends on how long it is taking.
    12. Make any necessary adjustments.
    13. Do the final render.
    14. Postwork as needed.

    That gives me the most flexibility, and generally allows for quicker work IMO. One reason (other than test renders mentioned above) is that larger environments can slow down the scene even just using the texture shaded viewport if it is very large or has a lot of elements. Though as BeeMKay says, some environments try to reset things, or there may be other reasons for loading the environment.

    If I am still trying to work out a scene in my head, sometimes I will load the environment before I do anything else and explore it a bit so I can figure out where to keep the characters. Then I delete it and follow the above.

  • OstadanOstadan Posts: 1,125
    edited October 2018

    I think it depends on your own creative process for the particular scene.  Sometimes you imagine the character(s) first, and can work them and then save them as subsets.  Other times, you are more interested in the setting, and might want to work on that first, including the lighting, using cubes as stand-ins for the eventual characters to set up approximate camera positioning, etc.  This is art, after all, and trying to establish a 'best practice' may be missing the point.

    Post edited by Ostadan on
  • Db3dDb3d Posts: 248
    Ostadan said:

      This is art, after all, and trying to establish a 'best practice' may be missing the point.

    My question deals more with the technical aspects of what order makes the most sense.

    I agree to a point.  While it is art, it is also dealing with technology, and sometimes technology functions best under certain conditions.  Case in point is Chezjuan's number 4 on his list.  His #4 is an example of why I asked the question.  I too have discovered doing certain things out-of-order can result in time wasted.  (In Poser's Cloth Room I learned this the hard way.  There's a case where you dress the character in their zero-pose position, and then go to another frame and pose them.  Do that out of order and you've made extra work for yourself.)

     

  • handel_035c4ce6handel_035c4ce6 Posts: 460
    brent_81ed1cee said:

    I suppose I am like many in that I am self-taught to use DAZ3D using documentation and online tutorials.  As with anything self-taught, you also self-teach yourself bad habits because you don't know any better.

    For this reason, does anyone know of a "best practices" for workflow?

    For example, is it better to start by adding the environment, then the props to build the scene, etc.?

    Or, is it all simply a matter of preference?

    Most important - do something for real, not for learning. Or everything quickly becomes boring and there always will be something "more important" to do.

     

  • Db3dDb3d Posts: 248
    handel_035c4ce6 said:
    brent_81ed1cee said:
    Most important - do something for real, not for learning. Or everything quickly becomes boring and there always will be something "more important" to do.

     

    I didn't think anything in DAZ3D was real.   ;)

  • JonnyRayJonnyRay Posts: 1,744
    edited October 2018

    Because most of my renders these days are actually of the same two characters in different settings & poses, I have a different workflow. Also one point I may slightly disagree with Chezjuan about.

    1. I have Conall and Simi saved already as "Scene Subsets" with their morphs, textures, and hair already applied. They are in the "zero" pose for Genesis 8.
    2. I load the character (say Conall) and dress him in the zero pose.
      • This is where I differ from Chezjuan.
      • There are some clothing items and smart props which don't load well if the character is already posed in something other than the default.
      • I would rather adjust the feet in the final render than pull what's left of my hair out trying to get a sword to fit properly in Conall's hand
    3. I almost always adjust material settings during this process as well. I'm rarely 100% happy with the textures the artist provided.
    4. I'll do some test renders with a simple HDRI lighting setup to see how the clothes are looking
    5. Once I'm happy with his outfit, I will save a new version called something like "Conall - Jeans and Sweater"
    6. Create a New Scene and repeat above for Simi
    7. Create a new scene and load the environment I'm going to place them in.
    8. Find a rough camera location, create the camera (setting Aux viewport to use that view) and frame the shot.
    9. If it is an especially heavy environment, I may at this point hide some of the setting as I'm about to work with my characters and don't need all the details just yet.
    10. Load each character and pose / position them in the scene.
    11. Get the rough lighting for the scene worked out. This lets me work with light and shadow as I get to the final steps.
    12. The last step is filling in the props and details of the scene.
      • At this point I know enough about the setting to identify where there may be large "blank" areas that I could fill with furniture or plants or something to make it more interesting.
      • By waiting until this point I don't load a bunch of props that aren't going to show in the final image
    13. I make any final adjustments to lighting and materials, then switch the Aux port to Iray to see an approximation of my final image.
      • This also pre-compiles material settings and such making my final render start more quickly.
    14. Throughout the above process, I've probably fired off many little test renders to see how things are looking in the full Iray engine.
    15. My "final" rendering process may involve multiple additional tweaks. All in all, it is not uncommon for me to have 10-20 draft renders and revisions before I get to a final one I'm happy with.
    16. I render the final 50% larger than I want my actual final image to be.
    17. Postwork in GIMP lets me adjust exposure and colors if the render wasn't quite right. It also lets me reduce the resolution which helps eliminate some of the "noise" that can be left with an Iray render.
    Post edited by JonnyRay on
  • ChezjuanChezjuan Posts: 515
    JonnyRay said:

     

    • There are some clothing items and smart props which don't load well if the character is already posed in something other than the default.
    • I would rather adjust the feet in the final render than pull what's left of my hair out trying to get a sword to fit properly in Conall's hand

    I've been there as well. I've also had clothing items reset some or all of the pose I set up when they are applied. I'd say the "put clothes on a posed character" works about 75% to 80% of the time for me.

    I've also recently (maybe 6 months ago) started using scene subsets for characters that I use often in the same base configuration. It saves a lot of time.  

  • JonnyRayJonnyRay Posts: 1,744
    Chezjuan said:
    JonnyRay said:

     

    • There are some clothing items and smart props which don't load well if the character is already posed in something other than the default.
    • I would rather adjust the feet in the final render than pull what's left of my hair out trying to get a sword to fit properly in Conall's hand

    I've been there as well. I've also had clothing items reset some or all of the pose I set up when they are applied. I'd say the "put clothes on a posed character" works about 75% to 80% of the time for me.

    I've also recently (maybe 6 months ago) started using scene subsets for characters that I use often in the same base configuration. It saves a lot of time.  

    Yeah, I think it's more of a "preference" than a "best practice" when it comes to whether clothes and smart props are loaded on a posed or default character model. I'd suggest people play with the content they have, try both, and see what works best for them. 

  • IceDragonArtIceDragonArt Posts: 12,548

    I have a blank scene set up to load when I open Studio, that has a camera loaded with the headlamp off as I never use it.

    Unless I am specifically doing a landscape or environent I always start with the character. 

    Add clothes, pose and dforce if necessary. 

    Add make up and hair. 

    Add environment and/or props.  If the character is sitting on something, I add that at the posing stage. I rarely move the character from center, I will move the environment to put the character where I want it. 

    I do my lighting last and its at this point that I start to run the iray preview, which also lets me see if someone is not touching the ground or any other issues that might crop up

    Render and move to photoshop, which is a whole nother workflow lol.

  • KitsumoKitsumo Posts: 1,216
    brent_81ed1cee said:
    handel_035c4ce6 said:
    brent_81ed1cee said:
    Most important - do something for real, not for learning. Or everything quickly becomes boring and there always will be something "more important" to do.

     

    I didn't think anything in DAZ3D was real.   ;)

    My debit card would disagree with that thought. Especially after this month.crying

  • IllidanstormIllidanstorm Posts: 655

    For iray renders:

    1. Add character with clothes hair etc.

    2. Pose character

    3. Add lighting just for the character

    4. Add enviroment

    5. Lighting for the enviroment

    6. Adjust render settings

    7. Render

Sign In or Register to comment.