Looking for techniques and tips for 'LED wall' digital signage content design
Hi all !
I wonder if anybody here could comment and/or share their experiences and methods used when designing static and video content destined for "LED wall" digital signage display systems.
To clarify : I'm not talking about 'Video Wall' type of systems, which consist of multiple high-resolution monitor displays.
I'm talking specifically about displays comprised of a number of low-resolution square (or rectangular) LED panels joined together to form larger display surfaces.
They're ubiquitous and are used in advertising and digital signage of all sorts.
In this message I refer to this type of display as "LED wall”.
LED wall displays have some unique characteristics that differentiate it from a regular digital monitor.
I found these differences to be important considerations when designing content for an LED wall display.
Number one is a relatively low resolution and pixel density compared to a modern computer display.
Another is the fact that the individual "pixel" LEDs are large in diameter (anything from 2 though 8mm) and are spaced apart from one another, often significantly.
Pixel pitch can be anything between 2 and 30mm.
This gives the content displayed on the LED wall a very specific "look".
With that in mind, the final "look" of any design displayed on an LED wall is a function of the display resolution, pixel size, pixel pitch and the viewing distance.
I was approached to design some content for such LED wall.
The LED wall I'm designing for is "only" 640 x 160 pixels (a rectangle comprised of 16-wide x 4-tall 40 x 40 pixel square tiles)
However, with a pixel pitch (the distance between individual LEDs) of 10mm, the physical wall dimensions are huge (6.6 meters x 1.8 metres).
I have not done design for this type of display system before.
I'm faced with several challenges.
One is the destination pixel resolution.
Anyone who designed a Web banner will know how having so little space
really makes you refine the design to best use available canvas size.
So far so good, though.
Limitation often stems creativity
The biggest problem is :
What I see on my Photoshop computer monitor looks nothing like itself when displayed on the LED Wall.
On a computer monitor the individual pixels are all crazy close together (irrespective of any canvas magnification applied).
Yet, when viewed on the actual LED wall, they're very visibly spaced apart.
From any reasonable viewing distance I can clearly see all individual pixels.
This makes for a lot of jagged edges and makes me want to return to original design and tweak it.
Of course, I don't have a 6.6 X 1.8 metre LED Wall to-hand in my office to proof the design.
When back in Photoshop I'm trying to make a best guess as to what the revised design will look like on the "real thing".
So I tweak and, to proof, I go back to the client's shop where the LED wall is.
That itself is a pain.
It's a two-hour round trip journey and I have to book time for proofing as their shop is busy during the day.
It's all a very frustrating, time consuming and "hit-and-miss" experience.
I tried two ways of approaching the design process :
First, I started with the design canvas the exact size of the destination LED Wall.
It's very tough to see what I'm doing when working on a 640x160 image on a 2560x1440 monitor.
I found myself wanting more resolution to finesse the details.
I worked mostly at high magnification (300 - 400%) but that just makes the pixels larger without actually giving me any more resolution.
To boot - when proofing the design on the actual LED Wall it always threw me just how different it looks due to its considerable pixel pitch.
For another approach, I would start with my canvas at multiples the pixel size of the LED Wall.
With my canvas sized much larger I had more real resolution to play with.
Working on a design was a more pleasant experience.
With my Wacom tablet I was able to use basic tools like brushes and selection tools with much more accuracy.
Sadly, when downsizing the finished design to the destination resolution the result was very disappointing when viewed on the actual LED Wall.
No matter what resizing algorithm I used the resultant images had a "smudgy" and "blurry" appearance when viewed on the LED Wall.
This was particularly evident with any type.
Again, I found myself having to re-visit the design.
In comparison - working at the destination resolution from the off (rather than down-resizing the larger designs) produced a better-looking final image on the destination LED Wall.
Things looked crisper and more "punchy".
The design process was much more painful, though.
It seems it doesn't matter which design approach I take.
There are pros and cons to either.
Overall, I find that a lot of guess-work and/or experience of the destination medium is needed to even "get close" between what I see on the computer monitor and the LED Wall "look" of the same design.
It drives me mad.
With all above in mind, my questions are :
- What are your guys' experiences with this type of design process ?
- Am I the only one getting frustrated and putting more time into it than should be necessary ?
- Is there an easier way to do this (ie : to better "see" what you're doing at the design stage without needing to do so much "back-and-forth" to use the actual LED Wall for proofing ?
- Are there any specific design techniques that would help the artwork "translate" better to this destination medium ?
Any comments/tips will be much appreciated.
I don't have experience with this exact issue, but the first thing I would do is take a good quality photo of the LED wall when one of you images is displayed. Then you can compare it with your original of the same image with both of them displayed at the same time on your computer monitor. This way you'd be able to tell what the actual differences are.
I would go with a one-pixel-to-one-LED ratio while designing, as that is going to be much more accurate.
Then you would want to have a method of converting that original to a "preview" of how it will look on the wall. This is where your reference images come in.
My feeling is that you'll want to resize the image, add a black "grid" overlay to add the gaps between the LEDs, then duplicate and blur the result and curve/saturate until it looks as close to the photo reference as possible.
Which brings up the question: is one LED RGB or are there three separate LEDs? That would make things even trickier.