The e:cue Patchelor is, in my opinion, one of the best pieces of software out there for doing complex patches. The UI is fairly intuitive and easy to figure out without even looking at the documentation. There are, however, some tricks and tips that will make your life easier. This article is going to focus on patching for video content. As the effects engine is based on the video area, it will also be relevant for that purpose as well.
I’m going to assume at least a basic understanding of patching, DMX512, and Patchelor specifics for this article. You can find out more about the basics through other articles on this site as well as through the official e:cue documentation.
Patchelor Document Layout
First, let’s take a look at the patchelor window.
In the center of the screen, you can see a patch with a single 16 pixel board in it. There are several things to note here.
1) The “Document Area” ecompases the entire white box with the gray grid.
2) The “Video Area” is the violet line that you see that is taking up slightly more than a quarter of the Document Area.
3) The gray grid is for reference only. It’s size can be adjusted and it affects the fixture snapping, but does not have any effects on the patching beyond being a layout tool.
The video area concept seems to be one of the most confusing elements of patching in the e:cue paradigm. The concept is really quite simple if it’s explained well (hopefully I can do that here).
What happens when you load a video into the programmer software (this applies to effects as well) is that the video is stretched to cover the entire video area (the purple outline). No matter what aspect ratio or resolution of the video or that of the video area, it will always be stretched to fit. Then, the video is mapped to each fixture as it is laid out in the video area. For example, if you have a single pixel set dead center in the video area, the pixel at dead center of the video would be mapped to it.
In a few moments we will be getting to how the software deals with translating resolution, but in the mean time…
For all intents and purposes, the document are is simply a container for the video area. Due to, what I beleive to be a bug in the current 5.2 release, the video gets mapped to the document area instead of the video area. Also, any fixtures that lay outside the video area, but within the document area will work properly when simply changing channel values, but since the effects are now based on the video area, effects will not be applied properly.
For these reasons I always like to have the document area set to the same size as the video area, eliminating all confusion and possibilities for weird behavior.
Resolution tends to be another confusing concept, not just in the e:cue world, but when it comes to computer and video in general. It is especially important to understand when working with low and mid-resolution media screens, which is often the case when programming with e:cue and other LED control software.
The basic explanation that will help us here is that resolution is the density of a video file in terms of pixels. It is comprised of the number of pixels there are in each dimension as well as how wide and tall the video is.
Along with the video, the fixtures themselves have a resolution, which may or may not be the same as that of the video you are using. If you have (8) 16 pixel boards that are each .25m square, arranged in a 4×2 pattern, you will have a resolution of 16×8 pixels.
When the programmer software maps a video to your fixtures, it will break down the video to the resolution specified in the Video Area “Width” and “Height” properties. It will then use those video pixels to try to decide the values for each node.
Consider the following setting:
This is a patch file zoomed in close enough that you can see the individual video pixels denoted by the purple squares. As you can see, the pixels on the 16 pixel boards (especially the top one) are not aligned perfectly with the video pixels. The programmer will deal with this by averaging the values of the video pixels a node covers. This will lead to unnecessary blurring of the image when it is finally shown, as all of the nodes are averages of multiple pixels.
In an ideal world you would have all of your nodes line up perfectly with video pixels like this:
In this case, all of the nodes will get exactly 1 pixel worth of data, leading to the sharpest possible image.
This leads to a number of different approaches depending on your video content situation…
Creating a patch for custom video content
In an ideal world, you will have custom content being created for your installation, so you can match the video to suit the needs of the project. There are 2 approaches to creating a patch when you have custom video content.
1. Use a standard video resolution but only use a portion of the screen.
This is similar to the above example, where you would use, say a 640×480 video but use only the top left 16×16 pixel block for video content, leaving the rest black. This makes life easy on the video content producer as they will most likely be used to using standard resolutions. This is also the only one of these two that will work when using a VMC, as you are limited to using standard DVI resoltions with the VMC.
The only major downside to this method is the overhead storage space and processing power used by the area of the video that is not even being used.
To make sure you align the pixels properly, you’ll want to use the following method:
- Figure out the width of each node with NodeWidth = (Fixture Width/Number Of Horizontal Nodes in fixture)
- Set the video area width to NodeWidth * Video Resolution Width
- Do the same for the height
For example, if you were using (16) 16 pixel boards (as in the above image), that are .25m in the Patchelor’s settings and you are mapping to 640×480 video then you would do the following:
- Node Width = .25/4 comes out to 0.0625 (caluclation is the same for height as the fixture is a square)
- Video Right : 0.0625 * 640 = 40
- Video Bottom: 0.0625 * 480 = 30
So, set your right and bottom to 40 and 30 respectively, and your with and height to 640 and 480, distribute your 16 pixel boards, and your pixels will be lined up perfectly.
2. Render the video at the resolution of your patch.
This method is conceptually (and mathmatically) much easier. Instead of trying to match a 16×16 section of video to the pixels being used, we create a 16×16 pixel video and create the patch accordingly.
For this method you simply arrange your fixtures. If you have the same matrix of 4×4 we were using above, you would set your video and document area sizes to 1mx1m (4x.25m) and then your resolution to 16 pixels by 16 pixels, which will end up looking like this:
As you can see, you can barely even notice the purple outline of the video area because everything is perfectly aligned and your patch will line up one to one with your video.
Patching for video not specially formatted.
In most cases, generic effects video will be used for content from places like istockphoto.com. In this case, you will most likely want to set the video and document areas to the outside edges of your layout and then set your video area resolution to be the same number of pixels wide and high as your fixtures cover. Setting it any higher than that will only waste memory and processing power devoted to mapping the video, as no matter how many pixels are being processed, the final resolution is limited by the fixtures themselves.
The major decision to make when using this approach is whether you want the video to stretch all the way across your installation or if you want to just take a specific chunk of the video. If your installation is close to the aspect ratio of the video you are using, it most likely doesn’t matter, but if you are doing something such as a strip of 16 pixel boards 2 high by 20 long, stretching a video to fit the installation may look a little weird as it will be 10 times wider than it is tall. If you want to just take the middle strip of the video, set the video area width to the width of the installation but make the height something that will keep a reasonable aspect ratio and then move the fixtures down to the middle of the video area. This is demonstrated in the image below.