Looks great! I’m sooo going to steal this once it’s done 
5 Likes
As I already mentioned: You can not steal it 
4 Likes
Just a dot is hard to see. A mini cross-hair would work, I think.
Do you have suggestions on how to label? There’s an (old) example of the vectorscope from the pull request. I think moving the labels outside the circle will reduce clutter.
Absolutely.
I think clipping indicators would be great.
Like this? You can change the “brightness” by clicking and dragging the scope.
3 Likes
Their locations don’t matter[1]. The font size could be smaller. The coloured lines are kind of distracting. I have seen an implementation use a ring of colour on the outside. (I still wish the inner rings are labelled.)
[1] I noticed that the colour axes are equidistant from one another and the rings are perfectly circular and equidistant from the origin. My colour science hat isn’t on at the moment… but it makes me wonder: Is this a correct representation of colour distances? Or is it okay for it not to be?
I am not sure this translates well into text and static images. This may have to change to make it easier to document and explain. Otherwise, you will have people asking the same question repeatedly in the future.
Perhaps, include a vertical slider in the remaining space where the icons are. I don’t know if GTK3 can do it elegantly.
Thanks so much for this. In addition to video uses, it would be great for getting an intuitive sense of vignetting and uneven illumiation going on in an image, removing the guesswork of manually correcting vignetting on unprofiled lenses.
Note the boxes in the Davinci vector scope…when you look at color grading with Davincin the boxes are angular targets for hue but also limits for saturation. They use a rule of thumb that saturation should not extend in any direction beyond a line that could be drawn between any adjacent boxes …not sure of the exact science but its a good double reference…
Found this interesting…https://www.peachpit.com/articles/article.aspx?p=2161673&seqNum=3 Also I cropped a bit close above but Davinci allows a scope for shadows mid highlights or all as well…not sure how hard this is or if it would be worth the effort
Why not ask for a 3D gamut viewer then 
On a more serious note: do you expect to do any measurement on scopes? That’s not really what they are intended for afaik. Though, the outer ring in Resolve shows some angular markings, so maybe there could be a use case for that.
@priort While comparing to Resolve can be useful, what tools would you expect to use in a photography oriented scope? Do you need such fine-grained control?
@afre
I wouldn’t say the axes and circles represent color distances well, but they do represent the color space accurately. I want to keep the markings simple and avoid excessive labeling. It’s not that hard to conclude what each circle means in the H-S vectorscope. The H-C vectorscope is a bit different. I think one label for the outermost circle should be enough.
That would make the feature easier to discover 
. I chose the current implementation because there isn’t much room in the histogram panel. If the slider is in the collapsible column, that wouldn’t be an issue.
@priort
DaVinci Resolve’s vectorscope is a U-V vectorscope. Unfortunately, there’s no way to show the same targets on the H-S and H-C vectorscopes. As for the low, mid, and high ranges, I’m not sure if the usefulness justifies the added complexity.
1 Like
What does a point on the vector scope represent? The angular and radial coordinates represent the hue and saturation (or chroma), respectively, together showing the distribution of colours.
Since we are currently given concentric circles going from 0-100°, are we to assume that for any hue (and implicit brightness), the saturation (or chroma) is normalized to each concentric circle?
OMG, YES!
Thank you!
Might just do it in Davinci and create a lut……
I don’t entirely understand what you are asking, but there is no normalization. In the H-S vectorscope, the circles represent 25, 50, 75, and 100% saturation (or 0.25, 0.5, 0.75, and 1.0 on a scale of [0-1]). On the H-C vectorscope, the circles should represent 32, 64, 96, and 128 chroma, although it is currently wrong and I need to fix it.
I may be able to plot those colors in the H-C vectorscope…
1 Like
That is okay. I don’t think it matters to most people. 
Honestly, that answer is not really helpful.
Sorry, I have to deal with more serious matters. (Usually is the case but I try to stay positive here.) I could continue this another time.