I did most of my font collecting when I was really running my business full bore, with offices and employees. We’d get 30-50 artwork jobs a week. A lot of them were reproducing designs that had already been printed on shirts. We’d get these disgusting, grimey t-shirts that were sometimes 10-20 years old and the prints were all worn and faded. Our specialty was reproducing crystal clean artwork that could be used by screenprinters to print new shirts for their clients. So Dennis (my main employee) and I became very proficient at identifying old fonts by appearance. Usually if I didn’t know one, he would, and vice-versa. Sometimes we’d get totally stumped and have to resort to some of our various identification tools. The first was a giant binder in which I had printed out hundreds of pages of sample fonts from my collection - all alphabetized by name. Many times we would recognize a font and knew it began with the letter “R” for example, but couldn’t recall the full name, so we’d just start thumbing through all the pages of “R” fonts until we’d find it.
When that didn’t work, there were a couple of online resources. The first was a font identification website called What The Font. It used an uploaded bitmap sample of the font to match to it’s database of commercial fonts.
https://www.myfonts.com/WhatTheFont/
Obviously they hope that if someone identifies a font through their tool and don’t have it, they could sell it to them. This was very rare in my case. Usually, I either had the font, or had “ways” of obtaining it. But this tool didn’t work that well because it was limited to matching uploaded samples to fonts they sold.
A much better tool came along a few years later.
https://www.whatfontis.com/
This would scan the entire web and try and match the font with any font it could find, both free and commercial. We had a lot of success using this tool.
In rare cases, when we couldn’t identify a font ourselves, nor could these tools, there was a forum called Typophile.com where you could upload sample images to the boards, and the community would often be able to identify it. A lot of times, however, some fonts couldn’t be identified because they weren’t actually fonts, they were logotypes that were created by design agencies specifically for clients who wanted a non-reproducible logo. The NY Yankees font is one example. For years people would ask, “what is the font used in the NY Yankees logo?” and the answer was, it’s not a font. Although I think since then, someone has gone on and created a (non-official) font based on their logo. This would often happen with movie titles as well. Like Harry Potter had a uniquely drawn font for the movie and people created full font sets to match the Movie poster lettering.
As an absolute last resort, I had a little gem of a program for Windows called fontmatcher
Which is very old and used to work by scanning a .bmp file of a letter, and then it would scan all the Truetype (and only Truetype) fonts on your computer and try and match it based on the shape and percentage of pixels that matched. It occasionally worked when nothing else did.
One of the real challenged in identifying fonts using software-based tools is a lot of time the fonts we’d scan in to upload were stretched in width or height by the original designer, and that would totally throw off the software’s ability to match. Sometimes we’d do a scan and get nothing, then take the image in Photoshop and stretch it or squeeze it as little as 10% and “bang” the matching site would get a match. It was a real combination of art, technology, and having a good eye and memory for font names.
In 2011, a font management program called TypeDNA came out that was lauded as a revolution in font matching. It used custom algorithms to scan letterforms and then come of with similar style fonts from a database of your choosing (which would typically be your own font collection.) The idea being, it could both match a font, and give you a whole bunch of alternates that looked very close (hence the “DNA” part of the software’s name.)
I downloaded the trial and the results were pretty impressive, but the program kept choking when it tried to build a database of my font collection. Probably had to do with the fact that a lot of my fonts were acquired through years of grabbing whatever I could find, wherever I could find them, and undoubtedly many of them were bootlegs with scrappy code or just outright corrupt. Corrupt fonts were a HUGE problem for us back in my pre-press days. They could wreck an entire file and make it un-openable by the software that created it (like Quark or InDesign.) As the years went by, font quality got better, and programs were better able to avoid loading up poorly coded fonts.
Anyway, TypeDNA still exists, and it’s not terribly expensive ($49), and it’s software you OWN, which is always a plus in my book, but aside from a bunch of early updates that were done around 2012-2013, it doesn’t appear as though there’s been much effort on the part of the developers to continue improving the program. Maybe they feel it’s a finished product and does what it does as well as it can do it. IDK. I may buy it one day just to have, but it appears it hasn’t have any development it at least 6 years. Even their Twitter and FB pages have been silent for as many years. But I do like the underlying concept.
The thing is, I really don’t use fonts to anywhere near the degree I did 10 years ago. Mostly I use Google fonts on websites, and even then I tend to stick to about 2 dozen popular fonts that just work well. On rare occasions I’ll get a job creating and ad or a flyer and I might get creative and go into my collection to look for something unique, but more often than not, I just use stuff I’ve used in the past.
Still, having acquired all the experience was valuable at the time. But you know how it goes… if you don’t use it, you lose it, and my identifying skills are a mere shadow of what they were ten years ago.
