My fourth filter - please comment

Woops, forgot to attach it - here it is.

Fold and Tatter.ffxml

Hi Carl,
That’s the kind of detached eye I need. I do not feel like I am doing my blending smart. I am missing something but I have not been able to figure it out. I don’t think multi-blend is the answer and I don’t like switch because in many applications they can be confusing for the user. The problem is that I had to tie those last three stages together while still allowing for the option of the folds with no effects – I thought that was important.
I will take a serious look at it after I get some rest but I probably would not do an update for eloquence – I would for speed though.
BTW, you have a lot of wonderful filters. I would spend more time looking at them but I am afraid they will colonize my mind – I mean that in a complimentary way.

Thanks,
Ron

ronviers,

wow. that's a lot of curves! i'm attaching a filter snippet you might find useful.

also, those two stones components could probably be replaced by two perlin noises. the perlin's run much faster.

Crossed Lines 1.ffxml

Hi Kraellin, Thanks for the snippet - it was very useful.
I could not have even told you which kind of noise I used. Two weeks ago I didn't even now noise came in different kinds. I like Perlin because it reminds me of Phothshop clouds - which I'm used to. It's good to know Perlin is faster - I will remember that.

The number of curves makes it look like that was the difficult part but to be honest I laid those out in a couple of hours. The real trick for me was the gradient profile that gives the appearance of a paper fold. It turned out to be simple, just two curves with a discontinuity but I must have tried more than a hundred before I learned to see it in my mind. It’s getting easier to look at a scene and understand the value gradients that make it up.

I am including my original notes from the planning stage. Look at how few of the features I was actually able to pull of. I worked for hours on things like paper clip and rotate but decided it was just taking too long. Another big problem I was having is with all the maps associated with surface filters. What the hell is a reflectivity map? Metallic map? I didn’t even know these kind of things existed. Presumably the maps have a different relationship to the bitmaps than I am used to with blended layers but I have not figured out what it is.

Oh well, it’s not that bad considering how few features I was able to implement and how little I know about what I’m doing. Just seeing Persidio use it made it worth while.

Here are the notes – it was originally going to be called Poster but that changed like almost everything else.

Poster characteristics
Fold directions (invert profile?)
Crease depth – slider
Tatter – slider (corner and fold)
Wrinkles – slider
Crumple – slider (refract with macro noise)
Edge curl
Corner curl
Number of folds (type)
Rotate (with clip or stretch)
Fastener types:
Thumbtacks, Paper clips, Staples, Tape
Shininess
Paper types:
Construction paper – Black, Orange, Red, Yellow
Typing paper
Poster board

I was dreaming.

Thanks Carl – the big gripe I have with the switch component is that it does not pass remap information. I really could have used it for simplifying the user access to the fold control but since it stripped the mapping information it would not work. It would be nice if they passed remap data and could be daisy chained.
Of course I may be mistaking about them stripping remap data but if they don’t I am missing something and I hope someone will correct me.

Thanks for the filter, positive feedback and other helpful information.
Ron

Quote
ronviers wrote: I really could have used it for simplifying the user access to the fold control but since it stripped the mapping information it would not work. It would be nice if they passed remap data and could be daisy chained.

Well you could use 2 switches that pass the information..Taking the filter snippet that Carl posted and putting a little more in..You can daisy chain the switches, Its just the pattern of how to arrange them to work is the interesting part...

On _ Off.ffxml

Hi Carl and CFandM,
The dilemma I had with this filter was that each fold needed to be defined and controlled independently but not necessarily set individually. For example, say I wanted the appearance of a road map; these folds would be similar (but not congruent) horizontally and reverse alternately vertically. My intention was to give this configuration a preset called road map. The ideal way (from the user’s POV) would have been to use checkboxes where each checkbox selected a different switch setting where each setting’s input was a control input pre-configured through remapping to represent a different type of fold – roadmap et. al. The problem is that the switch does not pass (or allow for) remapping information from the control input to the output side. I think remapping switch outputs for purposes defining user’s presets would be very helpful. How else can we make it easy for a user to select between, for example, a newspaper fold or a roadmap fold? As it is I had to break out each control even at the risk of intimidating the user.

Thanks,
Ron

Something like this..It may get interesting if the filter has a lot of components already but its one way..


test.ffxml

I guess then what would be needed component-wise since there is this switch a switch for curves, maybe now a switch for the sliders to pass the information through..

Here is one more. Imagine the output from the switch on the left connected to the input of the switch on the right. This configuration would allow for the simplified checkboxes to be used to select between the more sophisticated mappings of the sliders. This would also be nice for users because the check boxes are individually labeled.

Preset selection II.ffxml

Hey Ron, neat effect!!! Looks like it is coming along nicely. I really like your approach to how you are learning this.....which I think will really pay off.....

Quote
ronviers wrote: I just wish I had a way to make the process go faster.

You and me both