Conclusion
Costco
Looking at all the data Costco can be said to produce a print most true to the original digital images. Its spatial frequency print was very good, with measurable periods matching the test image even at the highest frequency. In the color resolution test it produced the most linear progression through the 256 blocks when compared to Wal-Mart and Fotki. Costco's color context prints also displayed the highest degree of color consistency when compared with the other two companies.
Wal-Mart
Wal-Mart's prints were not as true to the original images as Costco's prints were. Wal-Mart's spatial frequency print aliased at the high frequency end of the spatial pattern. The colors that Wal-Mart produced in its prints during the color consistency test were highly dependent on the surrounding colors. And although Wal-Mart's color resolution images produced continuous shades of all seven test colors, it saturated very early on in about half of them.
Fotki
Fotki's prints seemed to be about on par with Wal-Mart's. Their data was very similar, and had many of the same apparent weaknesses in attempting to reproduce the test images. Fotki's color reproduction was highly dependent on surrounding colors. Fotki's color resolution prints also saturated early on in the shade gradient pattern, although, generally not as early as Wal-Mart. What can be said for Fotki is that it produces results comparable to Wal-Mart at a much lower price at 9 cents a print, compared to Wal-marts 16 cents a print.
It may be a fair question to ask if exact reproduction of the original digital image is the golden standard. We will not attempt to answer that question in this project, but do concede that perhaps many of the aforementioned weaknesses in Wal-Mart's and Fotki's data, could be deliberate attempts to improve the appearance of the image as compared to the original data. But as we set out to conduct quantitative tests, we will only make evaluations on how true the printed data was to the test images.
One aspect that we did not touch on in this report is the reasons behind the differences in output from the different companies. As a continuation of our research, we might suggest that more analysis be done of our data, specifically in attempt to discover the causes of the differences.