2.5D/3D printing brings new perspectives in many fields, such as manufacturing, architecture, and arts. These new applications also bring out the issues that were barely considered in traditional 2D printing and go beyond color accuracy. Given the recent possibilities of surface construction, one may wonder how the texture details affect the reflection properties of a relief print. In this paper, we attempt to address this question by designing and testing on two groups of relief printed samples: 25 textured surfaces with different roughness levels and 40 visually flat surfaces with different gloss levels. We will first describe how reflection data (in the form of BRDFs) was acquired with our test device. Then we fit the data with some current analytical BRDF models. The results show that, in terms of accuracy and speed, the Ward model fitting outperforms the other models we use. On the other hand, we see that for our rough surface samples, there is a weak positive correlation between input roughness and the roughness parameter in microfacet based Cook-Torrance model. This parameter, however, shows a stronger negative correlation to surface gloss level for the flat samples. These results provide a useful insight to the reflection and texture properties of relief prints, and can be further embedded in the printing pipeline.