Jump to content

Short and sweet OP centric lessons


kleer001

Short OP videos, yay or nay?   

222 members have voted

  1. 1. Would you watch short (5-10 min) OP centered how-to videos?

    • Yes! Where are they?
    • Yes, even though I'm pretty good I can always be better.
    • No, I know enough already.
    • No, but I know someone who would.


Recommended Posts

Nice work, kleer01 - I think the latest iteration was perfect: "short and sweet" as advertised. The video compression was clean, I thought the speed was great, and I appreciate the fact there is no talking - concise and direct, perfect for the busy CG artist. At under five minutes, it's a very small investment of time to re-watch if anything was unclear.

I definitely agree with pclaes' suggestion regarding hash tags for the SOPs used as well, that will prove very useful in the long run.

Great to see this plan coming together, I look forward to the contributions you will be bringing to the community!

  • Like 1
Link to comment
Share on other sites

  • 2 months later...

Excellent discussion. Thank you peeps for putting so much thought into this. I had no idea people were so interested.

Here you go, even added attribute promote to keep the faces that are on the edges

Shinjipierre, thank you so much for putting up your own version of the camera frustum culling method. I had not thought of using the projected texture coordinates to cull geometry. So, I opened up your scene and checked it out.

Overall my personal preference is for as few nodes as possible. It's probably because I think I'm kinda lazy and dumb. Taking into account that some of the default ops are actually digital assets and not fully optimized (I'm looking at you Color SOP!) this is not always the case, and sometimes two nodes are faster than one node. So, I looked very hard at your setup.

And, I am sad, to say it's not something I would use in production or teach other people as it is slower and more complicated than the volume method, or even the slightly simpler double delete method. Please see attached hip file and test it yourself with the Houdini Performance monitor (under Widows->Performance Monitor). Btw I cranked up the resolution of the grid for an overall higher performance hit and greater contrast between the methods.

I got for times: 0.744 seconds for the UV method, 0.326 seconds for the double delete method (which could have been one delete), and 0.178 seconds for the volume method. Also, timing aside, the UV method leaves geometry that is behind the camera as you can see when you pull out from the camera view. Which kinda means it doesn't really work at all.

Again, please please test this file for yourself and let me know if you get different results. I can't thank you enough for stepping forwards with an alternate method and showing us another way to get 'er done.

cameraFrustrum_Test.hipnc

Edited by kleer001
Link to comment
Share on other sites

Sorry, it was far slower 'cause of the group node I made out of lazyness. (did that setup quickly)

For the point not removed behind, I forgot to test if the UV is not 0 (meaning, behind the camera)

I've removed the attribute promote trick (it's just to keep the faces that are in cam) as well and I get about the same time to compute, the difference is in the result, the volume method is not as precise since the volume would need more detail to be precise (making it slower), and the far of the volume has to be changed, that's a bit annoying.

But, yeah, most of the time, we don't need to be that precise when removing points out of bounds of the camera.

cameraFrustrum_Test_corr.hipnc

Link to comment
Share on other sites

  • 3 weeks later...
  • 8 months later...
  • 5 months later...

Tutorials that don't vocally explain what is happening on the screen I often skip, especially those that just have music in the background. I would rather listen to an author explain something than just a quick flyby of someone using the app over a music track that I turn turn off. I don't even mind if the explanation is in another language that I don't fully understand. The intention will still be there.

 

I think 10-15 minutes is the 'sweet' spot for a tutorial length. I know that is hard to do on more complex topics but that is generally the goal I shoot for in my tutorial work. It also helps refine the process to it's essential steps. Highlighting pitfalls and mistakes is also a good thing as long as it does not become the focus of the tutorial.

Edited by Atom
  • Like 1
Link to comment
Share on other sites

.

 

I know that is hard to do on more complex topics but that is generally the goal I shoot for in my tutorial work. 

 

 

Do you have a link to your tutorial work?

 

Thanks!

Edited by tar
Link to comment
Share on other sites

Sure, here are a few links that are Houdini centric. I also have some Blender tutorials on Vimeo as well.

 

Animated Crack Through Ground Plane:

 

Shadow Matte Composite:

 

Import A 2D Logo:

 

How To Tech Sphere:

 

These tutorials represent my steps in learning Houdini so if information is wrong or misleading please feel free to post comments with better or easier solutions.

Edited by Atom
  • Like 2
Link to comment
Share on other sites

  • 1 month later...
  • 1 month later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...