I know it's too late, but isn't it better using (panel.transform as RectTransform) instead of GetComponent?

@@cannotfigureoutaname a better version will be get the rect transform on awake

@@hop3studio511 by still using a GetComponent?

@@cannotfigureoutaname just one times

@@hop3studio511 Right, my idea was to use ( as RectTransform) as casting, wondering if it's better than GetComponent, but yeah I got tunnel vision of the problem and in the bigger image you just cache the transform lol

You should make a video on how you did it and how others could make their own custom tweens!

But some people like me refuse to not do anything 100% by hemselves

well atleast you can brag about it on your resume and customize it to your hearts content

where could we check the performance cost of animating the UI with animator vs tweening? I find it very interesting as an artist

@@luciorojas4278 Check out the Unity Performances Profiler. There is a few good tutorials about it here on KZfaq. It can tell you where your bottlenecks are and you could create the same animation multiple times using different methods and compare the time taken for each.

You love 12 year old tweens?....

Joanie Pépin FBI open up! Haha

I'd argue it's as much the coder's responsibility to make things look nice as it is the artist! You can go a long way with animating through code. It's not because you're a coder that you should have no feeling for aesthetics or animation!

😂 that's a quick float!

0.3f is not a measurement of time....

@@zyansheep it is if you += Time.deltaTime

@@TheDukeOfReason Hmm... doesn't seem humanly possible...

Maybe that's because we're all really A.I inside a giant simulation, all carrying on our seemingly genuine lives via Time.deltaTime in void Update()...

Also, I was animating UI wrong... :D

I love a good Tween. DOTween is my guy.

Me First Games yikes

@@gavintantleff wait why yikes

​@@bilbostabbins1882 "Tween" is also common slang for a child who is 10-12 years old, which is why most of us have just gone back to calling this "animating," so we don't sound like pedos. 😅

@@mpbMKE I have never heard of that and its suspicious that you do; nonetheless quite weird

Google has you covered man! The omniscient AI knows what you desire even before you do ;)

@@theral056 it certainly did because it changed everything for me. Since leaving my last comment I use LeanTween for everything and recreated it in my job's in-house AR/game dev software because I couldn't live without it

Great tips, Tweening is fast and powerful

Thanks. As a noob, I can't tell you how many times I have used a tutorial for something that ends up being outdated. I end up having to fix pieces of code and trade out elements that are simply not functional in the most recent LTS release.

It fixed performance issues, they not make it as efficient as tweening. You can see a null state in animator when nothing is being animated, showing the animator is constantly looping in the null state.

there are multiple reasons to use leantween: 1. one line of code 2. has easings 3. not a hassle to set up

@@chocolateimage There are multiple reassons to not use tweens: 1- You cant directly preview your tween animations. Time consuming. 2- Complex animations require complex code and this makes a lot of non-codder tilting. 3- It is a hassle to setup.

I came here expecting more of an artistic advice but this is correct. It doesn't matter what engine you use, any sort of change in position, scale, sprite, color, or whatever always requires a redraw. Tweening's no exception. Besides, there's going to be at least one draw call for UI every frame anyway, because ui shares the same explicit window with all the other non-ui moving objects populating the background. GPU is extremely fast and UI draws which are almost always 2D are very inexpensive. Whatever engine you use, batch and cull your ui assets correctly, use atlases to render with few draw calls. Animating has little to no affect to graphic performance.

@@Manas-co8wl Having an animator with even an idle animation that does nothing makes the canvas rebuild and there is a difference between drawing and "batch building" the canvas. I agree that GPUs are fast these days but alot of games are made for mobile and those games usually have a lot of moving ui parts. A tween if i remember it correctly will rebuild it when active but when its done it stops rebuilding the canvas. meanwhile to get that on an animator you have to disable the component fully. One way to get get some performance back if you really want to use animations is to split canvases into sub canvases, so when a element forces a rebuild it only affects that canvas and not the full ui. There might have been some changes in newer unity versions, but i doubt it. some sauce: 24:00 -> kzfaq.info/get/bejne/ld2onNeYyd6RhX0.html forum.unity.com/threads/unity-ui-performance-tips-sharing-my-findings.524916/ learn.unity.com/tutorial/optimizing-unity-ui#5c7f8528edbc2a002053b5a0

This are not 100% correct . watch the end of this video kzfaq.info/get/bejne/ld2onNeYyd6RhX0.html .The problem is that animator dirties the UI even if it's not actually animating anything. But apparently we only want to dirty the UI when there is something "dirty". That's why you should not rely on the animator and write your own transforming code or use a tween library.

@@PredGuy Disabling animator from script isn't any harder than writing tweens, is it?

@@raystenr5828 Its not "hard" to disable the component but it will probobly lead to more code to track when animations are done and disabling the animator after them than actually using tweens would need and it can lead to enoying problems with the animations depending how the controller is set up, for example: say that we have four states "idle_hidden", "show", "idle_show" and "hide". (idle_hidden is the entry state) then we run the show animation and we track the animation finish and disable the animator when we reach "idle_show", all good, now we want to hide it so we enable the animator.. and we are in "idle_hidden" without animating "hide" since it is the entry state. There are of course ways around it but in my opinion its just much more work and makes messy code/anim controllers.

You should not use either exclusively. dotween is great for little things like bounces and procedural stuff but for complex things use animations.(like a title sequence or complex hand made transitions)

he showed the code ? he is using delay on each element so one after the other comes later or what did you mean?

What do you mean? Using the shader to transform the vertices of UI elements like moving them around and scaling them and doing other fancy tricks with them? In DirectX or OpenGL that's damn near the only way to do it ... you can of course use the CPU to do matrix, quaternion and vector math during your update cycle and then construct a new transform matrix and bind that to the shader, but I agree it'd be faster to let the GPU perform all that math ... however, that becomes tricky at a whole new level, and I don't think UIs are generally a performance bottleneck for most games. Now, in a game like a Paradox 4x grand strategy with a super sophisticated UI full of hundreds of buttons and toggles and sliders then hell yeah the UI is a very, very serious performance consideration ... despite the lack of high-poweres 3D graphics, Crusader Kings 3 will all but set an old PC on fire with all the complex calculations, UI rendering and game logic running behind the scenes. I tried to run it on an old laptop once and it took about 30mins to start a new game and then I couldn't even scroll around on the map ... just teleport somewhere after I let the mouse sit still for 5 mins. Needless to say I uninstalled it off that PC and just played it on my good one, lol.

thank you this helped me so much

Thanks I had this exact problem.