Migration Sans Project

sim's picture

I begin a new projet. I plan to use it as a text typeface and I also want to draw some weight and their italic.
Your criticisms would largely be appreciated to pursue my project.

Thanks
After some tips from Chis here the necklace test I've done.
See my most recent Necklace Test v2

Migration_v10a
Migration_v10b
Migration_v11a
Migration_v13a
Migration_v13-7
Migration Bold v1
Migration Medium_v15-2
Migration Medium-italic-bold-1a
Migration (5 weights)
Necklace Test v1
Necklace Test v2

hrant's picture

This has more character than one might expect, considering how overcrowded this genre is. The "a" is quite charming, and the "g" is encouraging. Plus I like the fact that it's a bit narrow.

One thing I don't get though is the right legs of "h" and "m": is that on purpose? If so, the intention (adding character?) might be good, but it's hard to see this feature working out well in the end.

The caps are a bit too dark.

The spacing needs a lot of work! :-/ But overall it has a nice tightness - I'd keep that.

And some macro stuff: the color is a bit darkish (which I actually like), but if you're shooting for mainstream usage, it implies optimal setting at smaller sizes. If that's what you're shooting for, I would: loosen the spacing; and pull in the descenders a bit. Otherwise I would: lighten the color; and lengthen the ascenders a bit.

BTW, it would be nice to see a character set too.

And one more thing: good luck!

hhp

sim's picture

Hrant

BTW, it would be nice to see a character set too.
Did you see this Character set in my previous post?.

I don’t get though is the right legs of “h” and “m”: is that on purpose? If so, the intention (adding character?) might be good. I thought to use this inslant only for the h and the m. Only to give something different for this genre.

The caps are a bit too dark. Do you mean, I have to reduce the weight of the caps or reduce the width of the letter (stem, bowl, etc.)?

The spacing needs a lot of work! :-/ But overall it has a nice tightness - I’d keep that. I dont know how to do this. How to reduce the spacing and keep the tightness. My not enough experience in this domain, prevents me from knowing how to make it. Some advices on this subject would be appreciated.

mainstream usage Could you explain to me what is a mainstream usage versus Otherwise usage?

Thanks for your generous comments.

André

hrant's picture

You mean that GIF? Too low res.

The caps: reduce the weight.

Spacing: a good first start is the scheme in Walter Tracy's "Letters of Credit". As for the tightness issue I mentioned, I meant that I think you should keep it slightly tight overall (partly because the color is darkish).

Mainstream versus idiosyncratic color: at any given time in type design history, most designers will have a preference for lighter or darker fonts. In the Middle Ages people liked dark fonts, while in the 70s they liked very light fonts. That's been moving more towards dark lately, but some people (like Fred Smeijers) still prefer a color that's darker than most other designers agree is good. I like darkish fonts too. A designer can justify a preference for something outside the mainstream either by simply saying that's the way he likes it (not too hot), or by believing that users benefit (very good).

hhp

sim's picture

Here a better Migration Character set

Hrant,
Your comments have a big value for me. I pursue my work and I shall return with modifications.

Thanks

sim's picture

(partly because the color is darkish).

Hrant

It took some day to react to that comment. Could you elaborate a bit on that. I don't know what the word darkish mean. In french we say gris typographique. Is that you want to say?

BTW, I worked a lot on my Migration typeface. I hope I will be able to send a second version of it soon.

Thanks

hrant's picture

"Darkish" just means slightly dark. So I was saying that your font is a little bit darker than fonts like this typically are (and I for one like darkish fonts).

hhp

sim's picture

After several good comments I put aside Migration to see how I'd be able to improving it. I made several changes at all glyphs. For instance, I completely redraw the weight of the caps, I rebuilt the bottom of the a (it was to dark), the e, B, P, R, æ, œ, all the number was also redraw. I also rework on the ?, &, the currency. I have some difficulty with the comma, I'm not sure where I am for now with that glyph. I also work on the spacing.

Your comments will be really appreciate. Here is the most recent result.

You could see here the character set .

Mark Simonson's picture

André, the most recent PDFs are coming out set in the Adobe Sans substitute font for some reason. (The previous ones worked correctly.)

Mark Simonson's picture

It's working now.

I would agree with Hrant's comments about the weight of the caps, spacing, the h and m, and so on.

I think the revised version shows a lot of improvement over the earlier one. One thing that stuck out to my eye was that there was not enough space on the left side of the lowercase a. It looks fine now.

Some things I'm seeing in the new version:

Certain characters--the A, a, G, s, V, 4, for example--look slightly bolder than the rest. It's not just these characters, but those are the ones that stand out to me.

The X appears to lean to the left, and there is too much space on its right side (or the left side of the Y).

The curve on the bowl of the R is too thin.

I liked your previous non-connecting fi and fl ligatures better. It's traditional, but I don't think it necessarily makes sense to connect the characters in a sans serif. You're really just trying to correct the problem of the f colliding with other characters. Connecting the characters is not the only way to do this.

I think you need to spend more time with the ampersand. I can't tell for sure from your sample, but I would guess that you have drawn it with the counters as separate paths. During development, you don't have to draw your characters this way. You can make one continuous path with overlapping elements. That way, you get a nice flow to the strokes. You can remove the overlaps anytime before you generate the fonts. I do it as one of the last steps.

hrant's picture

Some things:
- The "s" has a malformed bit at the top-left of the spine.
- I might make the slashes in Oslash/oslash less slanted.
- Try to relieve the clotting in the joins of "OE", maybe by thinning from the inside of the "O".
- In the "ae" the "a" is too small.

Spacing: Uneven, but can't tell much else from justified paragraphs.

Mark, nice to see you've warmed up to type crits! Your eye is most welcome.

hhp

sim's picture

I ask some questions on my Migration «g». I worked on some new models. The first one is the one who is a part of my family typeface. His lower storey is not sitting on the baseline, 3 others are new possibilities and the most left one has his lower storey sitting on the baseline.

Could I have your opinion on those Migration «g».

hrant's picture

I think the 2nd and 4th are good candidates.

hhp

Mark Simonson's picture

It's hard to compare the new g's out of context. Perhaps if you set a word or two with each variation?

sim's picture

Here some words with the Migration «g» in context.. That will be helpfull to give your advise. Thanks

Mark Simonson's picture

Honestly, I like the first one the best.

I think I see what you're trying to do with #4, but aligning the bowl geometrically with the baseline, because of optical effects, makes it look like it's below the baseline.

What do you feel is wrong about your original g that you want to change it?

sim's picture

Mark

I'm glad to ear that. The # 1 was also my first choice. I checked a lot of typeface and I found there are a majority of them sitting their lower storey on the baseline. I though it was a part of the response to have a good readability. So I tried with mine but I was'nt satisfied. For now I'll keep my first choice. Thanks

hrant's picture

Of itself, #1 is better I agree. But it's not... "dorky" enough, which this face is, although in a decidedly endearing way. The slight disproportion of #4 makes it more in character I think. And the fact that it aligns better is a bonus.

hhp

sim's picture

Great Hrant. You gave me a new view on that glyph. I love challenge, I'll try to get what you're taking about. Thanks

Mark Simonson's picture

And the fact that it aligns better is a bonus.

That's just it--it doesn't align better. It looks like the top of the lower counter is below the baseline because there's no optical compensation. If apparent alignment is not the goal, then it wouldn't be a problem. Nobody says it must align there (or appear to).

hrant's picture

I actually meant optical alignment... except I didn't
actually look to see which was better that way though. :-/

One thing about anything optical however: it depends on scale; you can't know the amount of optical compensation needed by just looking at magnified glyphs out of context. Most notably, this applies to basic things like the round overshoot of the "o" (as explained by Harry Carter in his translation of Fournier). What size is this design meant for? The larger the size, the less optical compensation you need, and there's a threshold above which it's worth just snapping to the exact alignement.

hhp

sim's picture

Mark & Hrant:

Both comments are really interesting and usefull to me.

So I've to ask a question to me. What size my typeface will be use the most often? Do I have to choose if I'll align my lower storey g on the baseline or not in relation with that question?

Is the reader could be bother with an apparent alignment?

hrant's picture

What I was trying to say is that if you decide that some part of a glyph should look aligned to something, the degree of optical compensation you need to make it appear aligned (as opposed to being aligned Cartesially) varies depending on the size the font is used at.

> Is the reader could be bother with an apparent alignment?

Generally he's "bothered" by close-but-not-exact [apparent] alignment. But only during deliberative "reading". During immersive reading I feel misalignement can actually be used to good effect - because information comes from contrast, and the subconscious doesn't care about things lining up (at least not in terms of the mechanics of reading, if not a type design's "subliminal" effects).

hhp

sim's picture

I worked on my g to give him something more...«dorky» Here the result.
Thanks for your helpfull comments.

hrant's picture

I think that Garamond form is mighty fine*,
but to me the second one still works the best.

* I might make an entire font just based on that!

hhp

sim's picture

On all of them I changed the ear from the earlier version. I worked with the a tail. Looks better to me for now and I agree with you Hrant, the second one is still the better choice. We have sometime to go far away and come back to find our first idea was the better one. I'll take in count your suggestion to start a new family on the «g garamond form» Thanks

sim's picture

I worked on my ampersand (esperluette or perluette in french) and it's hard to me for now to choose one of them if one of them should be choosen. May be I will have to work more.
Thanks for your generous comments.

Let's get ready for the Migration ampersand.

Mark Simonson's picture

I think the last one works the best. The others are more unconventional and call undue attention to themselves, I think. They also seem a bit off balance. You may want to try some other forms if you think the last one is too conventional. If you want go with the open design you are using in the first four, you'll need to pull the strokes more into that open space on the right.

sim's picture

I also worked on my number 4. I tried to open the top to avoid to have this sector too much bolder as in the previous version.

Time for the new Migration 4.

hrant's picture

Personally, I actually tend to think the "fully closed" ampersand tends to disrupt more - it's too complex a shape (like the [traditional] binocular "g"). I think only some type designers have a problem with the "open" style, probably due to habit. Considering the other shapes in your font, I would go with the second one. But you might try asking laymen.

As for the new "4", I think that tight space is too distracting. You might try making the vertical stem much shorter. But really, the closed one you had before seems fine.

hhp

sim's picture

As for the new “4”, I think that tight space is too distracting

…And after a second closer look, the two open 4 look smaller than the other number. I'll keep the older one.

Mark Simonson's picture

Personally, I actually tend to think the “fully closed” ampersand tends to disrupt more - it’s too complex a shape (like the [traditional] binocular “g”).

As far as that goes, it's less complex than the g he's using. I say, if the binocular g is okay, the closed & is okay.

sim's picture

Hard to say which Migration «g» could be having the best impact, the longest life or the best readability. May be my question is superfluous.

Your advice will help me. Thanks

dezcom's picture

André,

The binocular g seems as though it may be a dark spot at text sizes. Perhaps it needs to be opened up some to get the space in it to feel like the rest of the glyphs. The single story g is the opposite, it looks a bit too large. You might pull up the counter a bit so that the lower horizontal stroke is a bit above the baseline. In general, the double boweled g is supposed to be a bit more readable in text because it has greater differentiation but it may not always be the best fit for a particular typeface.
Do you have a PDF of Migration in text size and quantity? this would help evaluate the effect.

ChrisL

dezcom's picture

Part of the difficulty with a low contast face is that you have fewer options in making adjustments to make complicated recurves like the binocular g. This is easy in a serif face with contrsat and decender room. I always struggle with this myself, the chalenge forces you to try different ways--(which is why some people say I "make wierd g's" :-)

ChrisL

sim's picture

I worked on my migration by taking in count of the comments of Hrant, Mark And more recently of those of Christ. I redrawn all the caps weight, the 4, the Ss, the æ, the Œ and I also tried to give in my g more character. Here the result.

Your comments are largely appreciated. Thanks.

dezcom's picture

André,

Mostly, the color is not uniform enough yet. The a, e, and g are dark spots. You may have to decrease the horizontals in those glyphs and open the counters a bit (Look hard at the upper bowl of the g.). I think your horizontal strokes overall need to be optically adjusted thinner. Remember, you are trying to give the appearance of monoline, you are not actually making it measured as monoline. This shows up most as the size gets smaller. If you intend this to be a text face, it is more of an issue. You may also consider putting some traps in to make it optically balanced in weight. In some cases, I can't tell if it is a metrics problem that contributes to unevenness but I think it is as well. I would decrease the sidebearings on curved sides of glyphs like b, d, p, q. At this stage, I would not set samples justified, stick with ragged right for the moment. It is too hard to separate out what spacing issues are due to your justification algorithm and which are inherent to your metrics.
Set an "nu" necklace (nuanubnucnudnuenuf...) and then an "o" necklace (oaobocodoeofogohoio...) to test your sidebearings. Leave kerning off. Print strings of these at text sizes and make adjustments to sidebearings until the color is even. If you cannot get even color with this setting, it indicates that some of the glyphs need adjustment. Those will stick out like the g does now. This is a difficult stage. For me, the first drawing of the glyphs is quick. The longer time is spent going back and forth with spacing and glyph adjustments to turn a set of glyphs into a harmonious family.
I think you have to be sure you are consistent in spacing as well. Look at page 2 column 2 at the word "playwright" and then at the word "works" on the next line. Look at the "ri" and then the "rk" The r seems farther from the i than it does to the k.
Mostly, just keep at it. You will get a good result for your efforts. I think it is coming along well. You just need to push through the hard part. You will see. I am confident you will end up with a good typeface for your efforts.

ChrisL

sim's picture

Chris,

You give me many tips to keep on and the encouragement to continue. Your explanations are really clear.

Thanks so much.

André

dezcom's picture

Glad it helped André, keep working :-)

ChrisL

hrant's picture

> necklace

Chris, did you make that up yourself? :-)

--

André, some of the changes you've made are quite interesting - it makes the font more unique, not to mention harder to evaluate - but I do think in some spots you've gone too far...

The "g" for one needs a lot of work all of a sudden - although in the context of some of things you seem to have done, it probably won't work to simply go back to one of your previous forms... Instead, I would: put less weight in the head, as well as the join between the head and lower bowl; and put more weight into the left side of the lower bowl. The UC "S": bad curves throughout. The florin: should slant. The Brit Pound: smaller head. Quotes: too much stroke contrast.

The cap height: I think it's too close to the ascender height.
I would either make them the same, or more clearly different.

The traps: I think they're cool, but inconsistent. Have you
seen this? http://www.themicrofoundry.com/ss_trapping1.html
It helps you quickly put consistent trapping in a font.

hhp

sim's picture

Hrant,

I know I went far with my g, and I'm not satisfy yet. Also thanks for the link on the traps. I did not know it.

Your advice will help me. Thanks again

André

dezcom's picture

"> necklace
Chris, did you make that up yourself? :-)"

Should I have said string of pearls? :-)

Actually, the term "necklace" was used regularly by Arnold Bank, my calligraphy teacher from the early 60s. I don't think he invented it though. It was probably a long-standing calligraphy thing.
Arnold had us do many n and o necklaces as practice for calligraphy. It was more about developing consistency in writing than measuring side-bearings. It seemed like a good term and it stuck with me low these many years.

ChrisL

hrant's picture

It IS a nice term!
For one thing it certainly sounds nicer than "control string"...

hhp

dezcom's picture

"Control String" sounds like what Cheney has wrapped around old "Scooter" :-)

ChrisL

dezcom's picture

André,
How is it going? You must be up to your ears in it by now:-)

ChrisL

sim's picture

You know what it is. I work hard on my Migration and I've an hard time with my lower case g. I also work on necklace as you suggested and I find it's gaving me good result. That's why it takes longer time to post something. I also redraw all caps.

I hope to post something new soon.

I've an other problem, I tried but I dont't know how to really work with the Trapping Flower as suggest by Hrant :

http://www.themicrofoundry.com/ss_trapping1.html

Thanks to take care of me ChrisL,Hrant, Mark, Norbert, and of course my wife, my daughter, my friend.

dezcom's picture

Keep at it but family always comes first!

ChrisL

dezcom's picture

André,
You asked me about text to test your typeface in. I thought I would post an answer here. You might consider longer sections of text in different languages to see the different letter combinations that come up and may need attention. In case you are not already aware of it, a good site to find multilingual text is the United Nations Human Rights Charter which has the text set in many languages. Here is the link if you don't know it :
http://www.unhchr.ch/udhr/index.htm

ChrisL

Nick Shinn's picture

That's neat Chris.
Here's one I found for Greek polytonic, Smyth's Grammar.

http://perseus.mpiwg-berlin.mpg.de/cgi-bin/ptext?doc=Perseus%3Atext%3A19...

However, there is something a bit strange about its polytonic, as the plain tonos-accented characters are taken from the monotonic character-set.

dezcom's picture

Nick,

Maybe they were having some PolyGin with their PolyTonic :-)

ChrisL

dezcom's picture

Seriously though Nick, thanks for the link! I was not aware of that site.

ChrisL

Syndicate content Syndicate content