Optimizing Art: How to Create the Best Art for Direct-to-Garment Printing

jerid hill

Throughout the years, Jerid Hill has written articles, blog and forum postings, created informative videos and taken the time to speak to many direct-to-garment (D2) users to assist in moving the D2 industry forward. Hill has been a thought leader in the D2 industry pursuing new technologies for better quality and output. In 2018, BelQuette merged with the ColDesi group of companies in Tampa and Jerid now serves at DTG Product Manager for ColDesi Inc. You can rely on Jerid’s unique perspective in the garment industry, focusing on the end user’s experience. He has extensive technical knowledge of D2 printing as well as details and tips for preparing artwork, pre-treating garments, expert level tips, and best overall practices for the garment printing realm.

When printing T-shirts with direct-to-garment printing, taking a couple of minutes to optimize art before printing can turn a typical print into something amazing. No matter which direct-to-garment printer you use, here are some basic rules that will help you achieve the best results possible.

Color model

Even though your printer uses CMYK (cyan, magenta, yellow, and keyline/black) to print, designing your artwork in a RGB (red, green, and blue) color model can push the boundaries of direct-to-garment prints.

Digital textile inks used in direct-to-garment printers may not necessarily be pure CMYK values, which gives RIP software the ability to intensify colors that might otherwise be dull when printed on paper. When designing in RGB, your files will be smaller but will not lack the quality needed.


If your printer has white ink capability, it’s best to design your artwork with a transparent background. All software used to print your design may have differences, but generating transparencies for a white layer is common throughout most programs.

Where there is a transparent background, there is no data and since there is no data, no white pixels will be generated. This is the most accurate way to generate a white ink layer, and simply the most forgiving. If you are printing a photo or rectangular design, then transparencies won’t be needed.

If your graphics are built from a vector-based program like Adobe Illustrator or CorelDRAW, then exporting the file as a PNG or TIF file will include transparencies with the correct export settings. A PNG file is a great file structure to use because it reduces your file size and does not add unwanted pixels to your image.

One main difference to note when choosing between a PNG or TIF file is that PNG files are only used with an RGB color model, whereas TIF can be saved in either RGB or CMYK. It is better to save your file with a CMYK color model when using TIF files since some RIP programs may not be able to handle an RGB TIF file properly.


While low resolution files may be able to achieve good results, it’s wise to consider the print head’s resolution. Since print heads vary in resolution, look to see how many nozzles per channel are in a linear inch. In a common direct-to-garment print head, there are 1,440 nozzles in a linear inch with eight channels or colors. Most of the time, four channels will be dedicated to CMYK and four to white ink. It is a 1" print head so there would be 180 nozzles per linear inch, per channel (180 x 8 = 1,440). The optimal resolution to use is 180 dpi to match the nozzles per inch. Anything more than this and you wouldn’t notice a difference, anything less and you could lose detail and quality.

There are other factors that the printer uses to achieve optimal results, so the print head nozzle count per inch doesn’t necessarily mean one printer is greater than another. For instance, just because one print head has only has 150 nozzles per channel in a linear inch doesn’t mean that it can’t achieve the same results as the 180 nozzle per linear inch print. The purpose of the nozzles in this instance is to have a baseline to generate your artwork resolution.

Lights, darks, and mid-tones

One of the biggest challenges of inkjet printing on fabric is in relation to mid-tones. Mid-tones are simply the tonal value between your darkest and lightest areas of an image. If there isn’t a great difference between the dark areas and mid-tones or the light areas and mid-tones, your print will blend too much and not stand out. Until you begin working with correct mid-tones, you may not even notice this is an issue. Certain RIP programs can correct some of the mid-tones for you, but it’s always better to visually see the results before the print. Depending on what program you choose, the settings may differ slightly, but the premise is the same: darken your darks, lighten your lights, and bring out the mid-tones.

For a visual, if you were to print the portrait at XX using Photoshop, the final result should be more than acceptable. However, if someone has worked with mid-tones enough, they’ll instantly spot areas where the results could be better. For instance, the tonal values in the woman’s hair, although moving from dark to mid-tone to lights, are very close together. When printed, the hair won’t to stand out and will blend together too much.

If someone were to hand illustrate this portrait, artists are trained to not worry about every strand of hair but rather accentuate areas to show texture. In distinguishing dark from light areas, we will lose detail, so it’s important to know how much detail is safe to lose without making the image look odd. This is done with experimentation for each design; there is no formula you can apply across the board. To help see these changes better, it’s a good idea to duplicate layers and make those changes per layer instead of the original image. You can always delete layers, but depending on how many changes you make, it could be more difficult to get back to your original design.

To duplicate a layer in Photoshop, choose Window > Layers from the menu bar. In the open Layer panel, simply click and drag the layer to the page icon at the bottom. You should now see two layers—the original and a copy denoted as such. For the purpose of this demonstration, let’s duplicate this layer again to create a third layer. We will also want to turn this layer off since it’s on the top of the other layers and will cover up anything we may be doing to the lower layers. To turn off this layer, click the eye in the Layers window, then click the layer you want to work with. In this instance, choose the copy layer.

In order to control the dark, mid-tone, and light areas, use an image feature called Levels. To access this area, choose Image > Adjustments > Levels from the menu bar. In the panel that opens, there are various buttons we can use, but we are only focusing on the three sliders for 1: Darks, 2: Mid-Tones and 3: Lights. As we use these sliders with our preview button checked, we can see the instant effect on the image. The order of sliders to work from is darks, lights, and then mid-tones.

Here we’ll look for the level where the tone begins to “ramp” up. According to the levels window, this image already has very high darks and lights indicated as “A” and “B”, but everything in between isn’t balanced. At the point where it begins to ramp up, this is where we will set our sliders. After these are set, we want to set our mid-tones. The middle slider will do this for you. Again, every image is different but the general idea is the same. Take the slider and move it to the left, just where it begins to “ramp” up. When we toggle this layer on and off by clicking the eye, you will be able to see a major improvement.

Because this image already had very dark and light areas, if we solely relied on this method by sliding the values to where they ramp up begins again, we may create too drastic of a change. We lose too much detail in the dark and light areas. To combat this, turn on the top layer and select it, then open the levels window again and choose a distance about half way from what we chose the prior instance. Simply toggle the layers on and off and see the difference. Chances are, moving half the distance will give you a better result. Again, this is because there were already very dark and light areas but since there were only a few of these areas, the mid-tones were still too close and needed distinguished. Levels can be a very powerful tool and when used properly, can enhance your images with a couple clicks.


Beyond enhancing tone levels, images can also be sharpened. With direct-to-garment printing, it is possible to over sharpen your image, but there is a little more forgiveness than what you see on screen. In other words, the image can look like it’s been sharpened too much on the screen, yet still print well. In the menu bar choose Filter > Sharpen > Unsharp Mask. When using this tool, we typically keep the amount less than 100 percent and the pixels at a radius of two or less. You can play around with these settings with the preview button selected to see the results. When the preview button isn’t selected, this image will appear softer. Take some time, sharpen your image more than normal, and print. You may be surprised at the result.

Color control

In Photoshop, there are many ways to control and enhance color. Many of these are global settings which effect the entire image. From the Image > Adjustments menu selection, we see there are quite a few color controls to choose from, including Curves, Vibrance, Hue/Saturation, and Color Balance. All of these settings will affect the entire image.  

If we only want to effect a background, we can use a feature called Selective Color. In the drop down menu for Colors, choose the color that needs to be enhanced. In this example we’ll use the green background. Since green is made up of cyan and yellow, if we increase these values, we may or may not see a tremendous difference. If we instead remove other colors from the greens, this is where we will most likely see the greatest results. With greens selected and the Preview button checked, slide magenta all the way to the left. Toggle your preview on and off and notice the intensity of the green areas compared to the original file.

All of these settings and tools can be used effectively on most any art you receive, but they are not meant to be one size fits all. Every file is unique and may require different variables, but the basics are the same.

With just the few adjustments described here, the image looks 10 times better than what was originally submitted. This is just a testament that playing with your art files will only require a minute or two of your time, but with a little practice and the right know how, you can create better direct-to-garment designs than you ever expected.