Kalleheikki Kannisto
1
COW Coin
Forum Replies Created
-
I’m by no means an expert on the matter, but my suspicion is that it has to do with the bit-depth of the file format as well as the monitor.
ProRes 4444 supports up to 12 bits per channel, so storing 8 bits per channel without losing information would not be a problem but 16 bits would.
Similarly, most monitors show 8 bits per channel which poses no problem to display an 8-bit image as intended, but a 16-bit image — or a 16-bit image converted to 12-bit for storage and then converted to 8-bit for display — has some definite possibilities of losing some accuracy along the way.
But that’s an educated guess at best.
Kalleheikki Kannisto
Senior Graphic Designer -
You’re welcome. The methodology is not all that complex, but there were some challenges to the implementation for sure.
The main trick is the “Separator” layer, which derives its height from the “cage” height minus the total of all the text box heights, which is then divided by the number of gaps between the texts (i.e. number of text layers minus one). If you followed that, you see that the resultant shape height is the distance that needs to be inserted between the lines.
The rest of the code is just adding up those text heights and multiples of the separator height from top down for each three of the moving text layers (the top text layer stays in place).
Plus that one checkbox to choose whether to include 3 or 4 text layers in the separator height calculation.
Kalleheikki Kannisto
Senior Graphic Designer -
Sorry, haven’t been here for a while…
If the shape in question is on it’s own layer, you can use the “new layer-based slice” function.
Kalleheikki Kannisto
Senior Graphic Designer -
In essence, you will have to apply the filter to the individual layers, not to the smart-object composite.
Whether the image will actually look the same in the end depends on the filter and the blend modes of the layers concerned.
Kalleheikki Kannisto
Senior Graphic Designer -
The question is, how would Photoshop know the conditions in the environment you’re bringing the image into?
There’s only one reliable way (that I know of) to achieve this, which is to get a video color calibration test image which includes any critical tones you plan on including (like skin tones) and go to the place where the image will be displayed along with a laptop (which has Photoshop on it) with the same reference image looking like you want it to look and actually project the image under the same conditions as it will be in real life. Then create versions of the image with adjustment layers (a single Levels adjustment layer is the simplest approach) and re-project, comparing the result to your unaltered reference image (on the laptop screen) until you are as close to the reference image as the projection system allows. Then you will have the needed adjustment layer(s) that you can add to the top of the layer stack in the image when you’re ready to export.
If you’re generating the image on a different computer (not the laptop you’re using for matching) you also have to make sure the image on laptop looks as close to the original on the source computer before doing the above.
Kalleheikki Kannisto
Senior Graphic Designer -
High quality printing starts at 300 dpi, so you can resize your file (without resampling) to 300 dpi, which will give you 10 inches for the width.
Now, since you’re in Photoshop, you have at least one trick up your sleeve: Detail-preserving Upscale. In general I use it when the scaling is 200% at the maximum, beyond that you will start to get into a painterly style, at least with photos. So that would give you a 20-inch wide image. Considering you’re working with an illustration, you may get away with an even larger enlargement.
An alternative way to enlarge, one which I prefer, is to use the LetsEnhance website. It gives you a 400% enlargement with AI-driven image enhancement. You can run, I believe, 5 test images through it for free. So that would give you a 40-inch image, if you’re happy with the outcome. You can shrink the image back down in Photoshop if you want something in-between.
Kalleheikki Kannisto
Senior Graphic Designer -
The same ones as in the Illustrator forum…
Kalleheikki Kannisto
Senior Graphic Designer -
Appears the conversion program has interpreted full black on the red and green channels as white.
One way to fix in Photoshop is to use a “Selective color” adjustment layer and with Blue colors targeted and the mode set to Absolute crank the magenta, yellow and black channels way up to the top. This may shift some other places in the image that are in the the blue ranges, but you can clean it up so that just the problem areas are affected by painting them out on the layer mask.
Kalleheikki Kannisto
Senior Graphic Designer -
Here’s one way to lessen it (see attached). I had to reduce the image size to 25% to send you a layered PSD, so you will have to adjust the gaussian blur size for a larger image.
13181_13171dsc0424small.psd.zip
For a different look, you can use the Camera Raw filters Clarity and Dehaze, compensated with shadows and highlights and decrease in contrast. Will look rather different from your original, but quite good. To get the color closer to the original, you can overlay the original image in Color blend mode.
Kalleheikki Kannisto
Senior Graphic Designer -
Create a character format that uses a western font. Find and replace with a GREP expression to select any numeral and replace with your newly created character format.
Kalleheikki Kannisto
Senior Graphic Designer