You would never capitalize the word “to” in "real life. Zoom1To1 looks funny to me.
Any chance we can have picture frame materials options for:
- multiply mode (whatever is while becomes transparent, no alpha channel needed), other colors multiple.
- knocking out (masking) a sampled color (no need to do create alphas in an image editor)
Thanks!
I don’t understand what you are asking for. I thought we were nearly there. Except the controls don’t actually work right yet.
Hi Bob,
I’m confused too. That UI is very cryptic to me.
1.There’s a field called transparency, but I don’t want to make my image transparent, just change its blend mode to multiply?
2. Inside the transparency the only property being controlled is alpha channel, but my image has no alpha channels? so what does it do?
3. Then there’s a color mask, but that’s a subset of the alpha channel? Not a separate way of controlling transparency?
…So am I using Alpha channel? or color mask? or both? …and what happens if I unchecked the alpha channel, but no the color mask?
changing the material to a basic material, not picture material I was able to get to this…
which looks close to what I want, expect that I have to tell it to be 1% transparent (total hack that I found out by accident).
But when I look at it unselected I can see a disturbing edge on the picture a terrible aliasing on the transparency mask, that only god knows what’s driving it. Where did I tell it to use white as a transparent color?
If I change the image for a .tif version with an alpha channel in it, I finally get the look I want of white pixels going away:
…of course I cannot turn off this effect if even if I want to now, because I do not see a “use alpha” checkbox in this texture settings, anywhere. and I still see artifacts at the image’s edge…
So maybe no, you aren’t that close I’m afraid.
Let me show you what a multiply blend mode does in Photoshop. The top layer is my image, the bottom one is what a viewport would look like:
The reason I want a blend mode instead of using channels, picking colors, etc, it’s because it’s a 100% repetitive approach, that requires no other fuzzing with all the complex controls you have in your material editor, and if I change the image it always still works. even if there’s grayish compression around white areas they would multiply to almost zero so I will not see them, unlike a color picket operation shown above.
here’s a technical explanation from the Photoshop help file of what Multiply does:
“Looks at the color information in each channel and multiplies the base color by the blend color. The result color is always a darker color. Multiplying any color with black produces black. Multiplying any color with white leaves the color unchanged. When you’re painting with a color other than black or white, successive strokes with a painting tool produce progressively darker colors. The effect is similar to drawing on the image with multiple marking pens.”
I would want a global rhino setting where I can set this mode as my default for all pictureframes, since that’s the only thing I would use in 90% of its cases.
Thanks,
G
@gustojunk, I think this bit is completely broken.
The alpha channel and color mask should only be one or the other or neither. What is there does not work right at all.
Also, the transparency slider should standalone an set the overall transparency of the picture.
You might have notice the the alpha or color mask don’t work unless the transparency slider is set to something more than 0… but of course you can’t get to that control if you check the alpha channel.
It is all a bit of a mess at the moment. Once is it is working right, hopefully you will not need any additional controls.
@andy can let me know if I’m confused.
Ahhh, you guys. You guys. 
so let’s see if am not too confused by expecting this to happen:
There will be three separate fields, at an equal level of hierarchy:
- ‘use alpha channel’ chechbox (which should be grayed out if the image being use has no alpha information)
- ‘use color mask’ checkbox
- ‘object transparency slider’ (which should never be grayed out, since this control is completely independent of any masking (color and/or alpha) being applied.
…and the color masking will be tuned up so it includes a range provide results that are as clean as the multiply examples I posted.
G
Yes.
And the main problem at the moment is that the display doesn’t support the new version of transparency that we have defined in the material. So only the object transparency works correctly now.
- Andy
that’s great Andy, I have more thoughts on the PictureFrame UI; I’m realizing that the current properties panel it’s not giving me the information that I really need.
I see the following picture information now, as a material information:
When I work with a picture what I need to know/control is the following:
- dimension of the picture (in the image file)
- dimensions of the picture in Rhino (size of the surface where picture is placed)
- confirmation that the aspect ratio of the picture object (surface) is correct, matching to the image aspect ratio (knowing if it’s distorted or not)
- that the image is embedded in the rhino file (checkbox?)
- Diagonal dimension of the image (most times this is what drives physical dimension in Rhino in my line of work, like making this one image 9.7" diagonal)
- A button to launch and edit the image on my preferred image editor
- be able to add/control rounded corner to the picture frame
The dimensional heads-up and control is a big deal, since there’s no easy way to do this in Rhino right now. Look what happens if I try to place this iPad screen and make it teh correct size of 9.7" diagonal"
When I go scale from center and tell it that my target diagonal dimension is 9.7"/2 (my file is metric) I will get an error because Rhino understands ‘9.7"’ and ‘9.7/2’ but not ‘9.7"/2’
So, I think we need a new Picture Frame UI to control things that are important for picture frames. Should I mock up what I’d like?
G
@gustojunk: no. 7 should be possible by splitting the surface the image is mapped onto (at least in Rhino 5 this works).
Yeah, but that’s not easy, fast or elegant. I want a numeric field where I:
- click on text field
- type “2”
- hit enter
Done!
Rounded corners isn’t really specific to picture frames though - is it? I would have thought rounded corner rectangles was a geometry type that was fairly common.
- that the image is embedded in the rhino file (checkbox?)
All images are (in V6), unless you specifically uncheck the Save Textures checkbox in the save dialog"
- dimension of the picture (in the image file)
This is available from the bitmap properties.
I think these are pretty general:
- dimensions of the picture in Rhino (size of the surface where picture is placed)
matching to the image aspect ratio (knowing if it’s distorted or not) - Diagonal dimension of the image (most times this is what drives physical dimension in Rhino in my line of work, like making this one image 9.7" diagonal)
- be able to add/control rounded corner to the picture frame
My feeling is that all of these should be handled at the general geometry level, not as anything to do with Picture frames.
- A button to launch and edit the image on my preferred image editor
This would be nice additions to the material editor:
- confirmation that the aspect ratio of the picture object (surface) is correct,
This is the only one that seems picture frame specific to me (because it’s guaranteed to be a rectangular surface).
Hi Andy, your thinking make s a lot of sense from a developer’s efficiency POV, but as a user, if I’m dealing with an image, I want to control that all in one place, and no go all over the place in Rhino to do my job. Also I want heads-up confirmation that needs to be up and front at all times, not buried in properties buttons, or requiring manual queries.
So let me go back to the 7 sins:
- dimension of the picture (in the image file)
I gave to go to destination 1, and play Where’s Waldo to try to find that info among 15+ fields of info
-
dimensions of the picture in Rhino (size of the surface where picture is placed)
Go to the destination 2 BoxEdit tab (assuming I even know it exists, why that tab is not on by default?) -
confirmation that the aspect ratio of the picture object (surface) is correct, matching to the image aspect ratio (knowing if it’s distorted or not)
Can’t be done. I could be modeling an entire project with a bad source image aspect ratio. Very dangerous. -
that the image is embedded in the rhino file (checkbox?)
It’s great to know that in V6 that will happen. Will people learn? …by V9 maybe? A quick usability study might give you that answer? …or trust me: they won’t know. -
Diagonal dimension of the image (most times this is what drives physical dimension in Rhino in my line of work, like making this one image 9.7" diagonal)
I have to go to destination 3, use the diagonal dimension tool buried in the dimensions palette. I still risk modeling an entire project with a bad source image size, because I do not have constant confirmation that my 9.7” image is still in fact still 9.7”. Very dangerous. -
A button to launch and edit the image on my preferred image editor
Can’t be done directly. I have to copy the path from the properties, assuming I know I can do that, then paste it in my file browser of my image editor. -
be able to add/control rounded corner to the picture frame.
I have to go to destination 4, and draw a rounded rectangle to split my image with it, assuming I’m clever enough to know that trick. And also assuming that my image plane is parallel to a Cplane, otherwise… more work.
The reason I bring this up, it’s because when I see the picture properties, I cannot help but see a void of gray that in reality is green pastures to make Rhino a more pleasurable tool to use more confidently.
@andy, I’m confused by the latest changes in the Picture material. Shouldn’t it be like this?
I don’t think the word “Use” is needed.
BTW, Why does the object transparency need to be more than zero before the masking works?
ok, I could not help myself, this is what I want… (including Bob’s mockup above)
Added:
quick image flipping
rounded /border as masking elements
a quick pop-up list of recently used images to replace file
image editor launcher
size heads-up (NOT editor, requires standard rhino transform tools to edit) This has various units pop-up since in US images units are usually different that file units.
3 Likes
I like it a lot!
But I question the “frame diagonal”. It only applies to monitors and has nothing to do with pictures in general IMO. Nor does it guarantee that the image isn’t stretched. (Scenario: PictureFrame is stretched in one direction, user sees that it is not 9.7" and 3D scales the PF so it becomes 9.7" with out noticing the distortion)
Aspect ratios for U and V, based on image size vs frame size would on the other hand make a lot of sense.
I don’t get the rounded corner thing. It has nothing to do with a pictuer frame IMO and is just a nice-to-have feature for one who works with monitors with rounded corners. (which is very rare btw) To the majority it would just be a feature that takes up space. It is important IMO to focus on purifying what a picture frame is, and not add other features to it. And it is easy to get rounded corners by trimming or mapping onto a custom surface if needed. So I vote against that option, and say: “Keep it feature rich and pure to the task”.
I also think that giving the reference image feature a secondary set of controls, separate from the default Material Editor was a good thing. Being closely aquainted with the latter is not something one should expect given, for a bunch of reasons; the fact that the Texure Editor is a modal window is a problem too. The issue I see with Gustavo’s mockup is that it now nearly looks as convoluted and intimidating as the Material Editor he didn’t want to use. It should easily be possible to host all this functionality (even including the rounded corner fluff) in half of the screen estate and with half the amount of visible buttons. The overall task is a dead simple one, the setup window should reflect that fact. Who wants to have good looking reference images and even plans misusing them in Layouts etc. will need to open an image editor anyway (which of course give you rounded corners too).
I am surprised about the choice of the new name for the command. One now has exchanged a misleading term with an almost empty generic term which as its predecessor gives no indication of intended usage. When dragging an image into the viewport, which Popup made least of sense to you?
