@Goatmeal: Thanks. As you can see from the creation date, I started this a long time ago. Back then it was just a concept with just an N in the character set. When this competition came around, I went looking through my backlog to see if I have something unpublished that can be used. Almost every day, I just played around with the pencil tool, roughing out a character or two and then refining it brick by brick. Until a day before the deadline, I didn't think it was good enough to be shared. One of the last characters I drew was the 0. It had a single pixel curved line in the middle. Looking at it over and over, that line seemed too weak. So I thickened it at one end. That gave it a flair that the glyph was lacking before. Once that happened I went back to other characters and flared the open ends of other characters. Now the characters had a distinct character (npi). However, the curves still weren't correct. Leaving them aside, I did the punctuation marks and brackets and stuff. While drawing the {, something clicked and I figured out how to do smooth curves in pixels. For some reason, the brackets looked better thinner than the letters, so had to go back to each letter and reduced the stem thicknesses. The curves of the C and other round letters was still just a tad off. One of the last thing done was the ampersand which gave me clue as to how to fix the curve of the C. Only then did I feel it was good enough to share. It's funny how the mind keeps coming up with solutions and discovering the right way to do something if you keep at it. Although, only partly figuring out now what pixel artists figured out in the 80s shows the dimness of my mind. Anyway, thanks to the love of the letters and the ease of use of fontstruct, I might actually do something worthwhile one of these days. Until then...

Sorry that this got so long. Thanks for the appreciation. 

@Aeolien: Skill is the easy part; figuring out what to do takes perseverance. I wish I can say that like a true artist, I could see the result before the letters emerged, but the reality is that it started with just a basic idea and evolved into what it is now. Thank you for appreciating my meager effort. 

In trying to figure out kerning pairs to test, it got confusing fast. So I wrote a small Excel macro to determine them all. For a combination of 59 glyphs, it turns out 3477 kerning pairs must be checked. Yikes! I've done maybe a 1000.

Thanks, Goatmeal and dpla. I don't know if I got them right or wrong, but I checked over 4000 kerning pairs.