OBJECTIVE: Comparison the quality of a digital shoulder x-ray using a compensation filter and without using a compensation filter. METHODS: 100 digital shoulder x-ray were made, 50 without a compensation filter and 50 with a compensation filter. Measurements of the image quality were performed in the number of pixels, i.e. shades of gray scale, within the selected measurement area on the image. RESULTS: Primary measurements gave the results of the number of pixels in favor of images without compensation filter 2016.41 ± 214.92, as opposed to images with compensation filter 1971.88 ± 173.12, with P = 0.04, which confirmed the statistical relevance of the measurements. Other obtained results show the ratio of exposure time in images without compensation filter 10.88 mAs ± 4.19mAs and images with compensation filter 13.33 mAs ± 5.70 mAs, and correlations between other measured variables such as gender, age, BMI. CONCLUSIONS: By using a compensation filter in digital shoulder radiography, we cannot achieve a better image, measured by shades of gray scale. In clinical practice, the use of a compensation filter for shoulder radiography is at the discretion of the radiological technologist and/or radiologist, for the purpose of better presentation of anatomical structures and calcifications of the shoulder girdle, always guided by the ALARA principle.