web hosting short URLs photo sharing

Introduction

Kokleong had the idea to scan prints from the various Epson (and other brand) printers with a drum scanner to compare there relative qualities. He posted a request for someone with a drum scanner, and I offered to make the scans. I have an Imacon Precision II, which is cpapble of scanning as high as 5760dpi.

We wanted to get down to see the individual dots to compare 3 things:

The dot size. The various printers by Epson will produce anything from a 3 picolitre do tup to a 12 picolitre dot. And the new ones will produce variable dot size.

The ability to render detail. With a smaller dot, finer detail should be achieved. Many small dots rather than a few large ones.

The smoothness of tones. With a smaller dots and the ability to produce more than one size, the prints should be smoother, less blotchy. We wre especially interested in the highlights where large dots, spaced far apart are much more noticeable.

What We Did

The Scans

The scans you see are 2 different areas of each print. They were scanned at 3600dpi. The are tiny images. The scans are all about 3mm x 3mm (about 1/8" x 1/8").

Set 1.Close up of eye

Photo 750

Photo

Photo 750

Photo

You can clearly see the dots (especially in the Photo Stylus prints). The most interesting thing to me is the two scans with the blue line running through the lower right corner. This is the edge of the childs hat. On the 750 output, you can actually see the dots forming the pixels of the image ont the edge of the hat! This is due to the file size of the original print only being about 120dpi. This illustrates why you should try to use a higher resolution when you print, at least 240 dpi. But the point is, this printer can render such fine detail as to actually render the pixels. I was really impressed. When you compare it to the Photo Stylus output, there is no hint of the pixels, just a lot of noticeable dots.  The scans of the child's face also help to show how much smoother the small dots can be.

Conclusions

While we are very encouraged by these results, there are a few shortcomings of this test.

• Image. The file chosen wasn't very good. Probably a scan from a cheap enlargement, or from a low res digital camera. There is a lot of noise in the fleshtones.
• File type. We aren't sure that we printed the exact same file. I printed a jpg from photoshop. Dan printed an Illustrator file from Illustrator. So, there could be some differences in the way the drivers printed the files.
• Size. Dan's image was slightly larger than mine. Not srue why, but it was about 8-10% larger. I still scanned them at the same resolutions, and didn't resample anything, but technically, his print could have a bit more detail because it is slightly larger.
• Paper. Dan printed his print on Epson Photo paper, I think, and I printed on Epson's older Photo quality glossy paper. This probabyl made some difference, but not a great deal. Both papers hold the dot pretty well.  Despite these inconsistencies, I felt that the differences in the prints were so great that I wanted to show everyone these results.

Future tests

We plan on doing this over with the following changes:

Test Image. I will be chosing a better file from my own archives, which will have detail, smooth tones and highlights.

File Type.We will be sure that everyone prints from the exact same file, with the same program, probably photoshop, with the same settings in the printer driver.

Paper. We will make sure that we all print on the same paper.

Other Printers. We will try to get output from the Epson 900 and other printers as well.

We are encouraged by thiese preliminary results, but I'm not ready to conclude anything for sure yet. More results to come. Soon I hope.  Thanks.

 

Greg Scheidemann

My thanks to Greg  Scheidemann and Dan Regenberg.
Any suggestions for future Face Off testing? Goto the Message Board...

FastCounter by LinkExchange