Rediscovering Monaco

Much to my sur­prise, I ended up de­sign­ing a new bitmap font.

I wanted a small, monospaced font, and I as­sumed (ap­par­ently, erro­ne­ously) that copy­right would pre­vent me from us­ing Mo­naco. So I spent a day with a pen­cil and some graph pa­per, try­ing to make some­thing new. Af­ter­ward, I took a closer look at Mo­naco, and was fas­ci­nated by dif­fer­ences.

So, with­out fur­ther ado, here’s what I made:

My bitmap font

vs. the orig­i­nal Mo­naco 9:

Susan Kare's Monaco 9

and its lat­est ver­sion:

The latest version of Monaco 9

If you’re in­ter­ested in all the gory de­tails, read on.

Shared Glyphs

Shared glyphs

I was try­ing to fit at least 24 rows of 80-col­umn text on a 320×480 pixel screen, so it’s not sur­pris­ing that I ended up with sim­i­lar met­rics to Mo­naco, since it aimed to do roughly the same thing on a 342×512 pixel screen. Specif­i­cally, The cells are 6×11 with a five-pixel x-height and a seven-pixel cap height.

As such, sev­eral of the iden­ti­cal char­ac­ters were thor­oughly ex­pected. Given those di­men­sions, how could low­er­case X or cap­i­tal T and H be any­thing other than what they must be? Cap­i­tal S, the dol­lar sign, and both O’s and Z’s are sim­i­larly ir­re­sistible and seem­ingly in­evitable. Sim­i­larly for cap­i­tal B, C, J, K, L, P, U, the ex­cla­ma­tion point, pipe, plus sign, hy­phen-mi­nus, and equal sign.

Surprisingly shared glyphs

Oth­ers were a pleas­ant sur­prise. Af­ter try­ing sev­eral vari­a­tions on cap­i­tal A, I landed on what turned out to be the same glyph. The same was true for cap­i­tal E and F, low­er­case M, N, R, and U, and es­pe­cially low­er­case C and S. We also share the nu­mer­als three and eight, but I re­peat­edly at­tempted al­ter­nates be­fore set­tling on mine.

Possible Improvements

My preferred glyphs Susan Kare's six, nine, capital D, Q, and R glyphs

In some cases, I ac­tu­ally pre­ferred my glyphs to those in Mo­naco. The nu­mer­als six and nine, cap­i­tal D and Q, and a let­ter near and dear to my own heart: R.


My inferior glyphs Susan Kare's superior glyphs

In many other cases, Mo­naco was clearly bet­ter. Most no­tably the as­ter­isk, cap­i­tal M and N, and low­er­case E and F. Mo­naco also has a re­ally clever trick that I didn’t dis­cover on my own, demon­strated in cap­i­tal M, X, and Y, as well as both W’s.

Also, my Y is far too V-ish.

Modernish Glyphs

My modernish glyphs Susan Kare's original glyphs The latest version of easily confused glyphs

I al­ready knew to look out for things that were even­tu­ally ad­dressed in later ver­sions of Mo­naco, like punc­tu­a­tion weight and eas­ily dis­tin­guish­ing be­tween cap­i­tal O and zero or cap­i­tal I, low­er­case L, and the nu­meral one. I also ended up with the later-gen­er­a­tion quo­ta­tion mark, apos­tro­phe, and caret.

In­ci­den­tally, later-gen­er­a­tion Mo­naco also fixes the baf­fling hor­i­zon­tal po­si­tion­ing of the nu­meral one. How­ever, it in­tro­duces a hor­i­zon­tal po­si­tion­ing er­ror in the ASCII caret, de­spite get­ting it right with the cir­cum­flex mod­i­fier glyph. I also don’t un­der­stand why the pe­riod and colon are po­si­tioned fur­ther right­ward than the comma and semi­colon.


My glyphs with descenders Susan Kare's glyphs with descenders

Of course, where things di­verge strongly is with the de­scen­ders. For my us­age, I could af­ford to have three-pixel de­scen­ders. De­spite the greater ver­ti­cal res­o­lu­tion on the Mac­in­tosh, Mo­naco had shorter de­scen­ders. Why? My sus­pi­cion: be­cause they needed to make room for the menu bar and win­dow ti­tle bar. Mo­naco’s sym­me­try be­tween de­scen­der and as­cen­der height is also nice.

Ef­fec­tively, this means that my font lands some­where be­tween Mo­naco 9 and 10. Mo­naco 10 kept the two-pixel de­scen­ders, and used the ex­tra height to make the cap-height and x-height taller. Given the choice be­tween longer de­scen­ders and taller caps, I think Mo­naco made the right call.

I have ab­solutely no idea why the de­scen­der on my low­er­case J is so com­i­cally small.

Mediocre Glyphs

My tossup glyphs Susan Kare's tossup glyphs

Oth­ers seem like a closer call. My low­er­case A is too heavy, but I’m still in­ter­ested in the dou­ble-story ap­proach. Given our shared low­er­case N, R, and U, I’m sur­prised Mo­naco didn’t end up with my low­er­case H. Both fonts have nearly iden­ti­cal oc­tothor­pes and un­der­scores, but I slightly pre­fer mine, which are nudged one row fur­ther down.


My remaining numerals Susan Kare's remaining numerals

The re­main­ing nu­mer­als are a di­verse bunch. I ever-so-slightly pre­fer my two, but Mo­naco’s seven is def­i­nitely bet­ter. I also pre­fer the style of Mo­naco’s four. Again, I would prob­a­bly have to pick Mo­naco’s five as the bet­ter one, es­pe­cially given how read­ily it dis­tin­guishes it­self from the cap­i­tal S, but it tends to en­cour­age the cor­re­spond­ing sixes and nines, which I don’t like nearly as much.

My terrible special characters Susan Kare's awesome special characters

The spe­cial char­ac­ters are where Mo­naco re­ally puts me to shame. I’m pretty proud of my am­per­sand, but Mo­naco’s is bet­ter. My ques­tion mark is garbage; it ex­tends above the cap height be­cause I couldn’t bring my­self to clip the tail. Sim­i­larly, I knew that Mo­naco’s tilde was far su­pe­rior the in­stant I saw it.


My brackets Susan Kare's brackets Brackets from the latest version of Monaco

The paired char­ac­ters are a bit of a mixed bag. My curly brack­ets are grotesque com­pared to ei­ther ver­sions of Mo­naco, and my paren­the­ses are far too heavy. My an­gle brack­ets prob­a­bly go too far to­ward use as de­lim­iters with­out con­cern for their use as op­er­a­tors. At least I landed on the mod­ern heights for most of these though. I still don’t un­der­stand the hor­i­zon­tal po­si­tion­ing of the open­ing square bracket. Per­haps it should have moved left­ward when it grew ver­ti­cally.


My curious glyphs Susan Kare's curious glyphs Curious glyphs from the latest version of Monaco

I’m still a fan of my low­er­case at sign, and I’m re­as­sured by the lat­est ver­sion’s foray be­low the base­line. It would be in­ter­est­ing to in­ves­ti­gate do­ing the copy­right and reg­is­tered trade­mark sym­bols in a sim­i­lar style.

Cu­ri­ously, both of my V’s came close to the lat­est ver­sion, but that’s likely be­cause I hadn’t dis­cov­ered the won­der­ful trick used else­where.

My per­cent sign is heavy-handed and unin­spired, but the orig­i­nal Mo­naco’s is a work of art. It in­cludes a top con­nect­ing bar and ob­long ze­ros. The cen­tral pixel isn’t even filled in! Un­for­tu­nately, it’s also a bit hard to read, but I’d bet that the mod­ern per mil glyph is more re­spon­si­ble for its later ver­sion.

Given the dif­fer­ences in our de­scen­ders, I’m not sur­prised that the slashes in the lat­est Mo­naco sit higher on the base­line than my glyphs, but I can’t for the life of me un­der­stand the ben­e­fit of po­si­tion­ing the for­ward slash so far to the right. Per­haps it’s an­other er­ror like the caret.


If vec­tor type­faces are prose, bitmap fonts are po­etry.

I’m clearly not a great poet, but I had a ton of fun go­ing through this ex­er­cise. The end re­sult was de­cid­edly mixed. I think I ended up with a few ar­guably bet­ter glyphs, and I even found one or two ap­par­ent er­rors that were in­tro­duced in the later ver­sion of Mo­naco. But there were far more ex­am­ples of Mo­naco be­ing clearly su­pe­rior. Fur­ther­more, I was only try­ing to make the print­able ASCII char­ac­ters, but Mo­naco has nu­mer­ous ad­di­tional glyphs, which also show­case a tremen­dous amount of va­ri­ety and cre­ativ­ity.