Hello everybody! To get the ball rolling, let's see how spread out is our community around the World in this new platform!

Hello everybody In this era of everything AI - and some backlashes starting to happen - I thought sharing a very interesting interview in here about the topic. This interview features Eric Schmidt, who is someone that was/is at the forefront of the development of this subject. Hope you enjoy!

This happen more than 2 years ago but is always current. Watch it here

Hello everyone, so happy to be here with all of you, like-minded peeps! I posted a methodology piece on Linkedin last Friday which unexpectedly got a fair amount of attention, and Monica suggested that I post a version of it here so that we could discuss the topic among ourselves as well, I thought it was a great idea! I elaborated a bit here with feedback from Linkedin that I found were great points missing in my original post. So without further ado, here it is! Last week at work, my lovely colleagues and I had an interesting conversation about picking up in-progress Nuke scripts that are difficult to read, and it made me feel like elaborating on the subject here with all of you. So if you let me, I will entitle this post "Nuke script ethics, or how to show your undying love and respect to your colleagues who may have to pick up your work sometimes." ;-) In my opinion, here are the most important things to ask yourself when you build a comp script, so it's easy to read and fast to process: 1- Do I have a main pipe? If I have a main pipe using mainly merge > overs, that flows from top to bottom in depth order as much as possible (far up top, closer at the bottom), my colleagues will easily understand my script in 3 seconds. If I don't have a main pipe, my script will be hard to read and will probably take a long time to decipher. This will be a waste of time for everybody. Chop chop. 2- Do I have diagonal connections that go all over the place or tons of hidden inputs? Now that's a touchy subject. Many studios literally forbid the use of hidden inputs. In my opinion, if they are not used to making a poorly structured script look clean, I say go for it as long as they are labelled clearly. Also, if used to cleanly link one camera node or one reformat node to many projections, that's also the right way to do it: it speeds up the script and doesn't impede the readability. For example, say I need the alpha of an element to stencil an alpha on another element. I usually make a dot at the source and write a note saying "hidden output to <specific location>" and make another dot at the target with a label saying "hidden input from <specific location>". Boom! Clear. And I try to limit my usage of that method as much as possible to keep my script easy to read.

Hi Everyone, I hope you're all well 😀 The other day I realised that a lot of people were sharing with me things that they were struggling with at their workplace and I found some common things in them. That made me think about our community and that we could share a bit more about what's happening and see if others have some input on things to help or just to share the feeling that we're not alone. so... What is the latest challenge you are experiencing or experienced at work? And how did you overcome that challenge if you did already! Happy sharing Monica