Old Dog New Tricks

1111214161722

Comments

  • Thanks Worlds_Edge, in photography everything not in the focal plane is blurred, the amount of blur increases the farther away from the focal plane a subject is. I took some artistic liberty here, I rendered the background and Darius with a really shallow depth of field, but rendered the Reptilian separately with out it, then blurred him is PS. He was too blurry with the DOF. 

  • That's very nice! I love the action. Postwork, as always, looks fantastic.

  • Thanks KM

    Merry Christmas Everyone! 

     

     

     

     

  • IceDragonArtIceDragonArt Posts: 12,548

    Merry Christmas!

  • deathbycanondeathbycanon Posts: 1,227
    edited December 2017

    I'm going to post a couple before and afters because as you all know I postwork EVERYTHING. lol  CC, and suggestions are always welcome. So are questions if anyone wants to know how something was done. 

    Here is Aisling that I posted in the gallery. The 1st image is the rendered png image - I didn't even use that. I rendered to a beauty canvas and 3 light groups and combined in PS. 

      

    You can click the images to see larger if you want, kept them small in the preview you you could see them side by side. 

    AslingBefore.jpg
    618 x 1000 - 152K
    Post edited by deathbycanon on
  • deathbycanondeathbycanon Posts: 1,227
    edited December 2017

    I rendered this one using just the default HDRi, rotated the dome to give interesting light and shadow. I forgot to render to a beauty canvas so all I had to work with was the png file, I always think they look sort of dull in comparison to a beauty canvas. Did a bit of post work to bring out detail and contrast, that caused my skin to go way too yellow so I tried an old photography trick using the eye dropper tool and tried to bring the color tones into a more normal range. I'm still not sure if they are good but better then the pumpkin orange trimming the histogram did to it. :) 

      

    CharlieBefore.jpg
    769 x 1000 - 366K
    Post edited by deathbycanon on
  • Great Santa image! I saw the Aisling image in the gallery. I thought that was superb. I really don't play with canvases enough since I learned how to use them. I love the effect that you got in your final image.

  • deathbycanondeathbycanon Posts: 1,227
    edited December 2017

    Thanks KM, it's fun going back to portraits now and then.... and now back to dragons. lol 

    This one was a composite piece. At 1st I was just going to keep it simple and do the dragon on a black background and mirror it. When RL babies were sitting up, around 6 moths old, one of my most popular sets was to do black and white photos of them on a mirror with a black background. Anyway, that's what started the idea, but then I thought maybe it was too simple... Here's how this image started, through to the finished image. 

     

     

    RedHouseChristmas1.jpg
    385 x 500 - 121K
    christmas-dragon.jpg
    500 x 500 - 59K
    christmas-dragon1.jpg
    1000 x 1000 - 255K
    Post edited by deathbycanon on
  • IceDragonArtIceDragonArt Posts: 12,548
    edited December 2017

    Oh I really love that one! 

    Post edited by IceDragonArt on
  • Thanks Sonja

  • I love how that turned out. It looks great with just the black background and mirror. However, I love what you ended up with at the end. That just took it in a whole other fantastic direction.

  • Worlds_EdgeWorlds_Edge Posts: 2,142

    Santa is so realistic and friendly looking :)  I love the Xmas dragon, wonderful work as always.  I'd never heard of "beauty canvas."  I'll be poking around to find out what it is.  Learn something new every day!

  • Thanks KM, I liked the simplicity of the black but the others in the house told me it had to have more. lol 

    Thanks @Worlds_Edge there's some conversation and links about it earlier on in this thread, somewhere in the 1st 3 pages I think. Mostly from TabasccoJack and alogovincian they had me trying it out fairly early on and now I hardly ever do a render without it and often use light canvases too. 

  • deathbycanondeathbycanon Posts: 1,227
    edited December 2017

    Here's a composite piece I did. I rendered the 1st Santa then decided I wanted his face more towards the camera with a light source coming from in front of him, so I did a 2nd Santa. I couldn't get the hat to bend the way I wanted it too, so I still used the hat from Santa #1.  I did the snowy hill in Bryce. The tress are PNG files off DA, they were really way too small & pixilated, but I knew I was going to blur them so it worked out. I need to buy my own snowy trees though. :) 

    The finsihed piece: All In A Days Work

    Noctice that I didn't just erase parts of Santa's feet, I added shadows and clumpy stuff around his legs to apear like Santa's feet broke down into the new snow. This is one way to ground an image and make it look more like it is attached to the background, not just a sticker sitting on top of it. 

    SantaStorm1.jpg
    500 x 500 - 69K
    SantaStorm2.jpg
    500 x 500 - 76K
    snow2.jpg
    500 x 281 - 95K
    Post edited by deathbycanon on
  • Sven DullahSven Dullah Posts: 7,621

    Beautiful work, death! Stunning pictureyes

  • IceDragonArtIceDragonArt Posts: 12,548

    Love the Santa!  I have got to go back and figure out the beauty and light canvas's. I had started to try and figure it out but didn't have a huge amount of luck with it.

  • Thanks Sven and Sonja

  • deathbycanondeathbycanon Posts: 1,227
    edited December 2017

    Night Stalker 

    Rendered a beauty canvas, an environment canvas,  and a light canvas, combined in PS

    added on top of photos from pixabay and moon from DA - all images link to source.

       

     

     

    Chapter1.jpg
    417 x 500 - 131K
    milky-way-2341762.jpg
    500 x 304 - 144K
    starry-sky-1946936.jpg
    500 x 333 - 139K
    Post edited by deathbycanon on
  • IceDragonArtIceDragonArt Posts: 12,548

    Gorgeous. I have got to make time to figure out the canvases.

  • SaphirewildSaphirewild Posts: 6,648

    What a beautiful and creepy render Death!!

  • Thanks @IceDragonArt if I get a quite day I'll do a video on how I do canvases. 

    Thanks @Saphirewild

     

  • It would be great if you're able to create a video on how you do canases, also if it's not too cheeky a video on how you combine it all in PS? Thank you.

  • deathbycanondeathbycanon Posts: 1,227
    edited December 2017

    The canvases video is uploading now, it says it's going to take just over two hours. I just did the night stalker piece, not a huge change in it canvas wise, but if I do another piece where I use the canvases more I'll record it as I do it. At least it shows you how to get the canvases from Daz and into a working format in PS. 

    Two new pieces. Same background on the tree house as the the night stalker one, as they take place in the same area. Also i was redoing the background so it would be a little different but my PS crashed at the very last steps in the image and I didn't want to redo it all over again so I just pulled the background from the Night Stalker. lol 

     

    I used "local adaptation" when converting this next room scene from 32 bits to 8 bits. You'll know what I'm talking about if you watch the video, I'll post the link when it's done uploading!

    Here's the before:

     

    treehouse1.jpg
    417 x 500 - 123K
    TheOldMansParty.jpg
    500 x 375 - 128K
    Post edited by deathbycanon on
  • algovincianalgovincian Posts: 2,562

    Rendering out 32-bit really opens up a whole new world, eh? Looking forward to watching your vid and learning something!

    - Greg

  • Rendering out 32-bit really opens up a whole new world, eh? Looking forward to watching your vid and learning something!

    - Greg

    Your probably not going to learn anything, but you can give me suggestions! lol I really don't use the 32 bit parts much. I really dislike the hdr type look on people, which is what I primarily work with. Love it on landscapes and architecture, but I personally don't like it for most portrait work, or just about anything with a person in it. Unless I'm doing someone with lots of wrinkles or rough textured skin and want those details really strong. Beings I do mostly people I put my 32 bits into 8 bit mode using exposure and gama so I don't get the funky hdr look. Now if I could open them in camera RAW....... That would be more useful to me, then I could work them like any raw file without getting the weird hdr effect, I don't think this can be done in CS5 though. Can it be done with CC? 

    Anyway I don't really know what I'm doing, just what I have learned through experimenting on my own, so if you have any suggestions to try after watching the video please chime in. You have way more experience at this then I do! lol  This video I just showed how to render canvases, and I explained parts of it wrong, but I get nervous doing videos and I don't do tech speak. lol Then how to convert to 8 bits, and a tiny bit on how to combine them to get lighting effects from more then one canvas. It wasn't the best one to use as an example because I didn't do a lot on this one. It was the most recent one I had done and still had available though. 

  • algovincianalgovincian Posts: 2,562
    edited December 2017

    There's always something to learn! Especially from somebody who brings so much experience to the table . . .

    I don't like the typical HDR look, either. Personally, I prefer a more subtle effect, and definitely one that doesn't violate the tone curve like most of the exxagerated HDR effects you see on the web.

    Still running CS2 myself (once scripting was included many years ago, stopped paying for upgrades). I've written many actions to automate the process of converting the 32-bit output from DS to 8-bit images in a predictable fashion. Here's an example where the differences are subtle, but to me make all the difference in the world:

    I don't think I've ever used a tool exactly as it was designed to be used, though I do like to check out what others are doing (but only after I've had a chance to freely explore without any influence).

    Anyway, your images are fantastic, and your quite prolific! Definitely something for all of us to learn here.

    - Greg

    vincia-compare-400.png
    400 x 450 - 246K
    Post edited by algovincian on
  • Here's the video link to one way I use canvases. 

    https://youtu.be/xEe2r5m7Fto

    @algovincian there is such a huge difference in tone, brightness, and clarity between the png image and the beauty canvas that yea, just rendering to a beauty canvas already makes the image better. I would love to see your process (I actually still have a copy of CS 2 on my system as my cs 5 is an upgrade from CS 2.) There is very little information out there that I could find on the subject so I have just been playing around as I go. I couldn't even figure out what to do with the darn files at first and was using Photomatrix pro to open them, so they were REALLY HDR then. lol Jack told me to use exposure adjustment in PS, and I am still always playing around with other settings but for this one I just did what I knew worked. :) 

    Seriously anyone out there can tell me what I'm doing wrong or tell me how I can do it better and I will give it a try, I'm always open to learning and have no problems taking CC. 

     

     

  • algovincianalgovincian Posts: 2,562
    edited December 2017

    Attached is Here is a link to the first action (CS2) I run in my HDR pipeline for processing 32-bit output from DS:

    https://drive.google.com/file/d/1NvqIoVu3qSjh0149xjpBxXdD96nXMkYl/view?usp=sharing

    It assumes an appropriately lit scene in DS, and a 32-bit beauty canvas having been rendered out.

    When you run the action on a beauty canvas rendered out of DS, the first thing that happens is you'll get prompted for an exposure. The default is -14.4, which is what I usually use. It's a bit under-exposed (on purpose) compared to the DS default of -13, but it typically works given the subsequent processing that is going to happen. The exposure that you choose will be used as a reference point throughout the rest of the action.

    The action will take a few moments to run. Once it has completed, your history tab should look something like this:

    You can click on any of the history states to quickly access them in your processing of the image. Here's a brief description of what they all are:

    base: whatever exposure you chose (32-bits)
    0: whatever exposure you chose (16 bits)
    +1: whatever exposure you chose + 1 (16 bits)
    etc.

    current: a quasi-HDR image (16 bits)

    This initial processing is designed to be subtle, and provide a starting point for further processing - it's not meant to produce a finished image. As was discussed before, I prefer a light hand when it comes to HDR processing, which is why I only look at +/- 4EV. You could modify the action if you wish to extend the dynamic range.

    Typically, additional processing is done while the image is still 16-bit to avoid banding (most filters and tools support this), and you can convert it to 8-bit before compressing, etc. for distribution. 

    This style of processing may not be for everyone, but at the very least, it may serve as an introduction to actions and the power of the history palette.

    Hope this helps someone!

    - Greg

     

    history.png
    212 x 787 - 27K
    Post edited by algovincian on
  • algovincianalgovincian Posts: 2,562

    Just finished watching your video . . . not sure what you were nervous about - you did a great job explaining everything! I've never actually used the nodes before, so it was useful to see somebody actually go through it. Thanks for taking the time and posting it. 

    - Greg

  • @algovincian thank you so much for the action! I can't wait to play around with it. My photography colleagues and I would always share actions so I am really familiar with them. They help to cut work flow time WAY down! Also I do sometimes work in 16 bit and only have a few filters that won't accept a 16 bit image, I didn't know that about the banding. I use a lot of photography with my pieces so I transfer to 8 bit so I'm not getting the "this image is in a different mode" warning. If I'm working on a full 3D piece I'll have to remember to stay in 16 bit. 

     

    Just finished watching your video . . . not sure what you were nervous about - you did a great job explaining everything! I've never actually used the nodes before, so it was useful to see somebody actually go through it. Thanks for taking the time and posting it. 

    - Greg

    LOL so funny! When I 1st started doing canvases the information I was linked to told me to use nodes, so I just always have. If I forget something in my node then the beauty canvas renders it as transparent. Now are you telling me that I didn't even need a node for the beauty canvas??? I have never even tried to render without one. lol Like I said I couldn't find a lot of information on the subject and some of what I could find was very high in tech speak and that is just way over my head. I did learn that you can not render multiple beauty canvases at the same time with different nodes/items assigned to each beauty canvas. That was one of the 1st things I tried to do, so I could have separate items on different layers in PS and only have to do one render. Couldn't get that to work. :) They work fantastic for separating out your lighting though! 

Sign In or Register to comment.