Should I be making bold and light to make a family?

olho's picture

Perhaps the knowledgeable could verify my thinking.

I have a sketched vision for a new typeface. On paper it's quite 'plump', so I could easily start with a bold or black design in FontLab. It's my understanding (mainly from other threads) that to easily interpolate a family I should make a bold and a light. This design may well suit that. Is this an appreciable intention?

Until now I've not really been able to visualise things that aren't a medium/in-the-middle weight. Working from the outside in sounds a bit perverse!

Any thoughts warmly appreciated.

Tim Ahrens's picture

If you feel comfortable with it then designing from the outside in is okay. You will probably find out that the interpolated regular needs some corrections.

The other problem, however, is that during the process of making test printouts and modifying your font(s) you will probably end up making the "same" corrections to the bold and the light, especially when you are testing your regular, so you have double the work. This has happened to me and I have wasted lots of time. That is the only reason why I now tend to develop the regular first.

ebensorkin's picture

Why perverse?

Also, what happens with type is that you always have to make small adjustments.

I suggest that you work from a med to a bold or a bold to the med & then go to a light. The chances of the medium being really good based on the design that arrives from a interpolated light & bold seem small.

But to be fair I don't know this per se. I haven't done it yet myself. But type process are not direct - they always need tweakage.

I am sure other folks here can say more about it.

olho's picture

I'm well prepared for tweakage. The whole process is the tweekage of minutiae!

I only meant it was a bit perverse. I guess I mean that I would usually visualise the regular weight, as that's probably both easier and the most used in a family. Designing the bold (as I'm currently doing) is fine, as I've already drawn it it all out on paper anyway, but I need to visualise the regular while I'm going. I just think the regular weight is supremo and therefore the one most likely to require direct design attention.

Any tips on the whole MM process would be most welcome.

aszszelp's picture

When sketching out characters on paper, I usually find myself drawing a glyph that is too bold for regular and too light for bold... :o) And I always find it hard process to get to the "right" weights from that intermediate one ('cuz that's the one I visualised internally, and I find it hard to deviate i.e. make it "inferior" in my internal eyes :-) ).

olho's picture

My sketches turned out very bold. In, fact they're exactly as you say -- too bold for regular and too light for bold! I'm not tracing them though, but drawing them from my reference drawings in FontLab. I'm deliberately making them consistent and more bold at that stage.

You're kind of onto what I mean. It's a job letting the 'vision' mould itself to weights other than the drawn original.

Inevitably, as I'm digitising my drawings things are changing. The process is very interesting. Here's my 3, sketched and digitised. The contrast on everything has been reduced now I'm in FontLab.

Thomas Phinney's picture

I think I like the sketch better than the digitization. Having the ball terminals thinner than the horizontals looks weird - as does the whole weight distribution on the digitized version, to my eye.

Eben wrote: The chances of the medium being really good based on the design that arrives from a interpolated light & bold seem small.

Interestingly, our experience at Adobe has generally been the opposite. If the masters are really good, the middle interpolations will often be even better. If there's something wrong in the middle, it's almost always indicative of something wrong in one or more masters. (The exception being if you're working with masters that are too extreme, and then you have to do something more complicated to compensate.)

That being said, the question of how early in the design one switches to an MM basis is a tricky one. If sweeping changes are going to happen, then yes, making those changes in several masters is more work. But working directly on a "regular" weight when one is ultimately going to have a light and a bold is also extra work. Different people have different ideas on this score, even folks who have a lot of experience working on MM based designs. Personally, I favor going to MM early, at least the weight axis, but some of my colleagues work differently.



Nick Shinn's picture

It takes very little effort to experiment with tweening.
All you need to do is:
1. Duplicate your font
2. In duplicate font (named Thin) paste the "three" from outline layer to mask layer
3. Draw a path in the outline layer, close to the edge of the "three"
4. Contour > Paths > Make Parallel Path (eg 10 units either side)
5. Tools > Blend
...and see what happens.

dezcom's picture

I began my first family with a medium bold and then did a light and a bold plus some tweens. In my second family, I started with a light and then went to a bold plus tweens. Since then, I have begun with a black and a regular. My point is that you can go any way with it. The advantage of trying several scenarios is that you learn something different each way. There is nothing perverse or magical in it, it is just a method to begin. The key is not to feel compelled to keep using your first attempt even if it is a Master. Try it and learn what happens when you interpolate. It will help you with the next face you begin. I don't think there is a one and best methodology.


ebensorkin's picture

Thanks for your comments everybody! Thomas, didn'y you once say that to get a good middle MM you sometimes have to dow fairly weird things to the very extreme weights? Maybe I imagining or mis-remembering that. What strategy would you suggest for going to bold & lightif you like a medium/regular you already have?

Thomas Phinney's picture

Thomas, didn’t you once say that to get a good middle MM you sometimes have to do fairly weird things to the very extreme weights?

Under "normal conditions" it's not a problem. But....

I had an example of a case where I needed the design space to be as big as possible, using intermediate masters was not practical, plus I needed the axis values to relate linearly to measurable attributes of the font. So I allowed the bold condensed master to turn essentially inside out, and defined a "no fly zone" near that master where the instances would be no good. This allowed me to expand the design space a bit.

This really begs for a more visual explanation. I have some slides on this, but they are intended to go with an interactive demo using the font. Maybe I can get to cleaning this up or recording a video version some time.... No promises, but I'll add it to my list of future blog topics.



ebensorkin's picture

Most cool!

charles ellertson's picture

I imagine if you are too bold, you'll get your face slapped. If you are too light, you'll get ignored. Neither is conducive for starting a family.

Miss Tiffany's picture

Charles cracked a joke. That's hilarious.

aszszelp's picture

ch: :D

dezcom's picture

CHARLES!!! You da man!!! :-)


Syndicate content Syndicate content