It looks like “Cache” is the new “Bake”…
Sounds great, except to my knowing, it came without much explanation, documentation… nothing.
When you do a search on the forum with “Content Cache” as keywords, you find thread titles like :
-BUG : Content cache (does this or that)
-Content Cache broken ?
-Content Cache indexing issues
-Content Cache not working…
There is ONE post from McNeel saying that the icon is new…great.
Here’s me struggling to bake geometry, just to get it deleted on the next go :
In this thread, @kike says that if no cache name is provided, something goes on internally to avoid overwriting of baked geometry.
Well that does not seem to work.
I can understand that new stuff like this isn’t working perfectly from the get-go ; that’s the Rhino way, and I’ve been dealing with it for over 20 years now, but the lack of ressources to understand how this new tool is supposed to works is un-acceptable.
As I mentioned to you earlier (in another thread), there is a document that is in the works that will help explain all of the different features of the Content Cache component. We do have an updated video that shows how this component works… but this is not annotated and will not really explain it like a full-fledged guide will do. Nevertheless, this video may help: Rhino - Grasshopper: Content Cache
I couldn’t quite understand the exact problem you are running into in your video. Could you please provide more information and/or a simplified example of your issue?
Also, I feel like I should help clarify a bit on how the Content Cache works. First, it’s not exactly a baker. What I mean by that is that the action of baking typically means that it’s simply going to add a copy of whatever object (ie curve, brep, block instance, etc) into the Rhino document. Each time you bake, it simply adds a new copy of that object. So, in previous iterations of Grasshopper, if you were trying to update your geometry, you would need to delete the existing geometry before performing the bake operation (otherwise you’d have many duplicates in your scene).
The Content Cache will try to be “smarter” about baking. If there is no geometry in your document, then the Content Cache component will bake a new copy of the objects into the document. After that, it knows which objects it has baked and keeps a reference to it. On subsequent baking operations it will update that object instead of adding a new copy. So, if the Content Cache has a reference to an object, then it will “update” instead of “adding” a new copy. Does that help. There are other features but that’s the biggest concept to understand with how it works.
Well My definition takes a bunch of intputs and generates from them lines and annotation dots which are Baked, or “Egg-crated” (whatever the new fad is).
If then I select new inputs, which generate new lines and dots, and I “Egg-crate” them, this will delete the previously “Egg-crated” lines and dots, which is a bit silly.
From what I gleaned by rummaging through the forum, there is a new concept of “Cache name”, similar to Elefront’s “Bake name” that allows to retrieve previouly baked objects in GH and do further processing.
I don’t immediately see why I would want to do that, but it’s an interesting concept, as long as it doesn’t wreak havoc on my more traditional workflow.
By this “Cache name” concept, from a post by Kike, I understand that if no “Cache name” is provided, then the objects are baked to Rhino the good’ol fashioned way, and there should be no more referencing and thus possible deletion or alteration to these baked objects from the GH definition.
Well, my video demonstrates that it is not the case, and it is either a bug, or I didn’t check an obscure option somewhere, I don’t know…
Now that the components have become full of hidden drawers, it’s hard to tell, given there is no documentation.
I managed to get my objects to stick by whipping-up a unique “Cache name” for each batch, and at least, I can move on with my work.
So, based on the questions here, there is more that can be done. We are working on more options for this.
Currently, there are 3 ways to do this.
Using the unique Cache Name.
Right click on the Content Cache and use Bake option. This is of course a manual process, but does work.
Use Change something like layer name for each option within a single Content cache. Here multiple layers are created to place the various options in Rhino on thier own layers.
The 3 ways are all well and good, but there should be a way to bake and not worry about some invisible link threatening to erase Rhino geometry.
This is nuts…Please allow for a simple down to earth, honnest to god, good ol’ fashioned bake !
Well, if you can do THIS, you should be able to do a simple bake, right ?
Maybe just bring back the egg icon, with an input for the objects and one for the trigger.
That would be just fine for 99.999% of my bake needs.
Anyway, thanks Scott.
