Wow. WOW. I keep trying to relate this to After Effects and just how many DAYS it would take to set something like this up and render it. Bless you all for creating this tool. Can't wait for the final release.
I started following... then couldn't pick up my jaw and I was in pure awe with the ease, depth and sheer thought that went into developing Cavalry. Bravo!
I wouldn't say Cavalry could replace Ae completely for me (or some others), but it definitely has potential to, in the future, be a must-have tool for motion design aside from other great ones, such as the earlier mentioned After Effects, to do things either is good at, reasonably quickly. Truly a fantastic job and finally some major progress in the world of 2d motion design software. It seems like it will give Adobe a run for its money for many motion designers
@cavalry As a salty old After Effects user Cavalry is currently rocking my world! Loving all of the procedural tools and interconnectivity. Query: is there a way to implement a fixed border between elements of different sizes within the quad tree- instead of having the border increase as scale of element increases?
We've been waiting for something like this! bug: UI doesn't refresh properly while playhead is moving aka if you add a deformer/filter while it's playing it looks like nothing happened. once you stop the playhead the interface updates. Look forward to seeing this tool grow; I'll put out a few tutorials on it.
Thanks Mike, glad you like it. The UI thing is not so much a bug, just the way it has to work. Adding elements is changing the node graph behind the scenes. Doing that during playback can cause issues with the calculations to we 'save' those actions until playback stops. You can of course adjust attributes in real time. That should work for you.
16:25 seems like there should be an obnoxious colored dotted outline when stuff is going out of bounds. Or a setting|tool for Domain of Definition, frame mask, constrain to area|bounds,etc.
I started playing with this yesterday and absolutely! Fantastic work! One question: I'm trying to key frame the position of a falloff but I don't see how, any hints please? Thanks
Double click your Falloff (either in the Project Window or from the list on a Behaviour) to load its UI into the Attribute Editor and then use the little diamond icon within the Position inputs to set a keyframe :)
Awesome ! Definitely a game changer in procedural FUI creation ! I have a request for you : please update the Quad Tree Shape with one more addition : for the moment, we are limited to a square ratio since, even if we change the "Size" of it in its parameters, we can't change the number of "divisions", separatly for each axis. Let say you want to achive a kind of large and thin "band" of data, like a "data ribbon", I would have need to add more subdivisions in the x axis (colunms if we talk in "tab cells" language) than in the y axis, in order to keep a visual "square" ration of the cells. would it be possible ?
Glad you like it! I think what you're asking for there is probably a different type of distribution. Quad Tree is based on maths that's explained here - th-cam.com/video/jxbDYxm-pXg/w-d-xo.html so it's slicing in a very specific way that's probably not suited to your request :( We do have an 'Image to Shapes' element that uses a Shader to subdivide based on the image/ gradient etc. That might help for now but we also have some thoughts here about the kind of slicing you're suggesting :)
@Cavalry For FUI's is there an L-system or way to add one? (aka pattern drawer, aka turtle graphics) Basically a way to procedurally make basic geometry or more complicated shapes with corners and borders have ,or lack, snazzy angles. Also used for fractals like the Dragon curve
I made it, apparently in the newer version you have to connect the text to the linear, and add the rectangle to the Text "group" and then add the Noise and Align behaviors.
@@lanzsantos_ Wow, yes - you're right. So the way to 'fix' this is to follow most of the steps but, rather than add the Rectangle to the Duplicator, first Group the Rectangle and the Text Shape and then add that Group as the Input Shape to the Duplicator. Basically, make sure the Text is not a child of the Rectangle (to stop it inheriting the rectangle's transform). Your hierarchy should look something like this: Group - Rectangle - Align - Noise - Text
![โรงเรียนห้ามรัก โดนจับได้ตาย...!! [เอิร์นไดเม่]](http://i.ytimg.com/vi/4u02Sbr556Q/mqdefault.jpg)
![ตีสิบเดย์ [FULL] | เปิดใจ "อาจารย์ เบียร์ ฅนตื่นธรรม" นักสนทนาธรรมสายดุดัน!](http://i.ytimg.com/vi/vqaBI_mnoDk/mqdefault.jpg)
Wow. WOW. I keep trying to relate this to After Effects and just how many DAYS it would take to set something like this up and render it. Bless you all for creating this tool. Can't wait for the final release.
Thanks for the kind words!
I started following... then couldn't pick up my jaw and I was in pure awe with the ease, depth and sheer thought that went into developing Cavalry. Bravo!
This is so incredible. I've been painfully cobbling this kind of stuff together manually for decades. Unbelievably wonderful. So exciting
This is profession changing tutorial here. Magical for people like me who didn't do 3d or AE.
I wouldn't say Cavalry could replace Ae completely for me (or some others), but it definitely has potential to, in the future, be a must-have tool for motion design aside from other great ones, such as the earlier mentioned After Effects, to do things either is good at, reasonably quickly. Truly a fantastic job and finally some major progress in the world of 2d motion design software. It seems like it will give Adobe a run for its money for many motion designers
I'd love to make some tutorials for this software. There is some really powerful stuff here.
Yes Please!
Do it!
AE killer product. Just keep it up. Great job
Nice I look forward to giving this one a try
I'm really looking forward to Linux. Thank you for your work.
I'm looking forward to trying this software out. This looks very cool!
This is a Great software, hats down
@cavalry As a salty old After Effects user Cavalry is currently rocking my world! Loving all of the procedural tools and interconnectivity. Query: is there a way to implement a fixed border between elements of different sizes within the quad tree- instead of having the border increase as scale of element increases?
Great to hear. Dust that salt off! You could try adding a Path Offset to the Quad Tree as a Deformer and then giving it a negative value.
We've been waiting for something like this! bug: UI doesn't refresh properly while playhead is moving aka if you add a deformer/filter while it's playing it looks like nothing happened. once you stop the playhead the interface updates. Look forward to seeing this tool grow; I'll put out a few tutorials on it.
Thanks Mike, glad you like it. The UI thing is not so much a bug, just the way it has to work. Adding elements is changing the node graph behind the scenes. Doing that during playback can cause issues with the calculations to we 'save' those actions until playback stops. You can of course adjust attributes in real time. That should work for you.
So powerful!
16:25 seems like there should be an obnoxious colored dotted outline when stuff is going out of bounds.
Or a setting|tool for Domain of Definition, frame mask, constrain to area|bounds,etc.
Obnoxious ;) Yes, we have plans for this and the comp bounds is already in here internally.
I started playing with this yesterday and absolutely! Fantastic work! One question: I'm trying to key frame the position of a falloff but I don't see how, any hints please? Thanks
Double click your Falloff (either in the Project Window or from the list on a Behaviour) to load its UI into the Attribute Editor and then use the little diamond icon within the Position inputs to set a keyframe :)
Good Software
Awesome ! Definitely a game changer in procedural FUI creation !
I have a request for you : please update the Quad Tree Shape with one more addition : for the moment, we are limited to a square ratio since, even if we change the "Size" of it in its parameters, we can't change the number of "divisions", separatly for each axis. Let say you want to achive a kind of large and thin "band" of data, like a "data ribbon", I would have need to add more subdivisions in the x axis (colunms if we talk in "tab cells" language) than in the y axis, in order to keep a visual "square" ration of the cells.
would it be possible ?
Glad you like it! I think what you're asking for there is probably a different type of distribution. Quad Tree is based on maths that's explained here - th-cam.com/video/jxbDYxm-pXg/w-d-xo.html so it's slicing in a very specific way that's probably not suited to your request :( We do have an 'Image to Shapes' element that uses a Shader to subdivide based on the image/ gradient etc. That might help for now but we also have some thoughts here about the kind of slicing you're suggesting :)
@Cavalry For FUI's is there an L-system or way to add one? (aka pattern drawer, aka turtle graphics)
Basically a way to procedurally make basic geometry or more complicated shapes with corners and borders have ,or lack, snazzy angles.
Also used for fractals like the Dragon curve
No L-systems yet I'm afraid :( We do have some basic Solvers in the current beta though.
I am trying to follow this tutorial but the texts change the position following the scale of their respective rectangle. How do I fix this!!??
I made it, apparently in the newer version you have to connect the text to the linear, and add the rectangle to the Text "group" and then add the Noise and Align behaviors.
Okay, this is cool...
when i try to add the text to the rectangle layer it automatically becomes it's parent! how can i avoid that?
Which bit of the video is this? You can enter the time in the comments like 1:30
Goodnight! In version 1.3 this does not work correctly in the chat compose. text moves left-right mod by exposure to noise. Has the algorithm changed?
Which section of the video are you referring to? (you can reference a time via e.g. writing 0:30 )
@@cavalryapp This is true. At 11:38. When you drag the text under the rectangle shape. The position of the text is also influenced by the Noise :/
@@lanzsantos_ Wow, yes - you're right. So the way to 'fix' this is to follow most of the steps but, rather than add the Rectangle to the Duplicator, first Group the Rectangle and the Text Shape and then add that Group as the Input Shape to the Duplicator. Basically, make sure the Text is not a child of the Rectangle (to stop it inheriting the rectangle's transform). Your hierarchy should look something like this:
Group
- Rectangle
- Align
- Noise
- Text
Lots of FUIn! This seems like an "app to watch"
Will Cavalry remain free after BETA? ❓
how to add sounds?
Check out docs.cavalry.scenegroup.co/-LU1VVdL61zw4e2kW3m1/elements/behaviours/sound
Already ditch premiere for davinci resolve. Yes, please! Release a linux build will be welcome too.
I have to recover and repair my jaw by being dropped too much along this video...
You're a loud eater, aren't you