Wow. This has a really nice feel -- sort of a mix of a humanist sans and an expressionist form like Pressig, Manuscript or Journal.

The lc "v", "w" and "x" seem a bit heavy. I'd try maybe making them a bit more modulated, with a bit more contrast between thick and thin strokes rather than the more monolinear approach that you have going now. Perhaps this should apply to the lc "k", "y" and "z" as well.

I like the ever-so-slightly wide lc "z", too. I'd leave it as is.

The slight flaring at the terminals is a really nice touch, too.

David

It's a nice start, Nathan.
But I think your descenders are really way too long.
I'd suggest you to shorten them to 1/2 ~ 2/3 of what you have now.

Thanks for the suggestions fellas. I wouldn't've thought to add some contrast to "v" et al, but I think it's working well. I still need to shave some weight from the "x", but otherwise they're fitting in much better as far as color goes.

And I shortened the descenders a bit (thanks for pointing that out, Eduardo) - I hadn't even noticed how incredibly long they were. The "g" still needs a lot of work. When I shortened the descenders, I redrew it with the Beziers to bring it in line, but I think I really need to rework it on paper. It seems too light and perhaps too narrow all the sudden.

Oh yeah, I also fiddled some more with the terminal flares and notches at som eof the joints. I moved the top notch on "p" because the top joint is already tapered - it seemed more logical to trim the bottom joint. And I took a nibble out of the shoulders of "m".

Incidentally, the cyan is the first version, the black is the revision. I guess that's obvious. If you're interested, I posted some of my sketches here - unfortunately my brother's architectural drafting skills didn't run my way in the family...

Great to see a contemperary sans serif with a humanist feel and a touch of a pen - very nice work.

lc g seems a little poor in shape

Nicolaj

Great work that you have here Nathan. I would like advice you about the contrast in some letters l/c (a,o,s,v,w,x). The space is too much close. In general its a very good start and I will be waiting for some new samples with the capitals letters and figures. Good work.

Thanks for the comments, guys. I had actually shelved this to work on some other projects, but perhaps I'll light into it again.

Nikolaj: Yes, the 'g' and the 'a' are the two characters I'm least pleased with. I've got a few more sketches of them that I need to digitize. I'll try to finish them up and post them in the next few days.

Ricardo: Thanks - I've started on the capitals, and I've got the beginnings of the figures. I'll post them soon. I'm a bit confused as to what exactly you mean by:

>The space is too much close.

Do you mean the strokes should have more contrast (thinner gray areas in fig. 1)? Or that the counterforms need to be opened up (larger gray areas in fig. 2)?

Thanks once again.

Haven't had much time to work on this lately. Here's the figures to date.

There are some upper case too, but I'm waiting to post until I have some more.

Minor changes to some of the lower case & finally the start of the upper case. The caps started wider, but they didn't seem to fit with the lower case. So I'm [slowly] working on condensing them.

Cheers.

Thanks Tiff. I see what you mean about the 'y', 'j', & 'g'. Hadn't noticed that before, but I totally agree - I'll try to refine those a bit.

And, frankly, the 'f' I've been struggling with for a while. And now that you say it, I also see an upside down 't'. This is almost like having a shrink ;-)

Hopefully, I'll have some time to work more this week.

Tiffany put it well - the tactile quality is great.

You said you want to work on the 'g'.

One idea. The lc 'g' and 'a' are, as one of the old masters pointed out, the letters where the evidence of the pen is most evident and hard to eliminate in the alphabet.

I am thinking that you might look to your Ardbeg a, particularly the tail of it for a better way to connect the loop of your 'g' to the top. You would then have a shot at consistency with the rest of the font, while coming up with alternative solutions.

To my eyes, the 7 doesn't work.

Thanks for that suggestion William. Sounds like a great idea - I'll give it a shot this weekend.

And I think I might have bitten off too much with that 7. I saw a 7 like that somewhere (Scala maybe?) and I thought I could pull it off. Probably not.

Had some time over the holidays to work on this some more. Finished drawing the uppercase, tweaked the 7, y, g, f. Started on punctuation & diacritics. Still a way to, but have a look.

Seeing your text setting, I am excited that this promises to be a very readable sans even in text. It is narrow but very open because of the angular counters. You might compare Vesta, the new face of Gerard Unger at www.gerardunger.com. It is one of the most readable sans I have seen, and your face promises to accomplish the same thing in a different way. He has rounded counters, and a narrow, humanist structure. He gets the openness partly by very narrow joins between round and straight strokes - a general feature of his designs.

You are approaching this in a different way, which gives a wonderful liveliness as well as good 'color'.

I like your new 7 and y. I don't know whether the f is an improvement; I would like to see more text setting.

The g I think still doesn't work; it looks odd and sticks out, to my eyes. Is the problem that the link between the loop and the top of the g sticks out too far to the left? That the link should narrow more in the joining to the top? Or maybe that the bottom of the loop should be lightened, following the stress in serif fonts? Unger does all of these in Vesta. You have a very different look, but maybe there is a way to make the g more harmonious adapting these patterns to your deisgn.

William, thanks for your comments. I will definitely look at Unger's Vesta. As well as following your suggestions for exploring the 'g'. Hopefully later this week I'll be able to post more - perhaps a PDF of some longer text settings. This last GIF looks pretty atrocious anyway.

Here's a PDF with some text settings at various sizes. The 'g' is slightly modified also: slightly narrower, thinner link, & the link has been moved to the right as per William's suggestion. I've been looking at it so much I can't tell if it's an improvement or not :-/

Shite. Here's the PDF.

TEXTSPECDEC.pdf (15.9 k)

The g looks balanced to me now, and consistent with the look of the whole face. I think the new f may be too stiff, though.

Looking at the PDF with the smaller size settings, I think there are some problems with spacing and rhythm. Also, I suspect your word spaces too large. I am not competent to advise you how to fix this. Hopefully some of the typophiles who are type designers can advise you.

I think you have very interesting direction on this, and with more work this could end up a distinctive and useful face.

Love this...
The counter of the n and subsequently the h, m and u is too generous.
Pulling them in a bit will help the rhythym greatly. I also wonder
if the top exit stems of the n and m couldn't be treated to the pen
like slices many of the other strokes share.

Also, the uppercase hasn't been treated as much as the
lowercase. The B and R are ripe for some of this.

Lastly, the traditional treatment given to the zero doesn't
fit (geometric circle). This face is way too forward thinking
for that.

I am not competent to advise you how to fix this.

Unfortunately, William, I'm not competent enough to space it properly ;-) Thank you for the observations.

Thanks for all your comments Eric. Luckily I've got some time this afternoon, so I'll try to integrate those suggestions and post another sample in the next few days.

I know there are processes to help you space. Briem's web site has some if I remember...

Some other stuff to consider, if you haven't already: Do you want to have the caps the full ascender height? Do you want to shorten the ascenders a tiny bit? Do the round letters need to be tighter relative to the straight ones?

I think you have a wonderful design idea and a good eye. It's now just a question of craft (= tons of work, I suspect).

So, it's been a while since I've posted anything about this face. Because it's been a while since I've done anything to it. But I recently got off my fat arse, and started working on a range of weights.

The PDF is the "first pass" I guess you could call it. The lightest, heaviest, and middle weight are all drawn - the intermediate weights are generated from Multiple Master instances. Obviously they need to be tweaked a great deal.

But for now I was wondering if anyone had any comments on the weights themselves. E.g., are the jumps between the heavier weights too big, the hevaiest weight just too heavy, etc.

I'm not sure about the amount of contrast that shows up in the heavier weights, but I don't know a way to get aorund that. I also hate the big stupid dot on the heaviest 'i'.

Thanks again.

WeightComparison03.pdf (115.2 k)

Try bringing the dot progressively lower as you get darker. In the uberheavy you'll wan't just a white wedge and 3 hairs more between it and the top of the stem. Also, there is no way to turn out a black lowercase without adding contrast. It's a hazard of the territory. That said, you can probably get by with quite a bit less contrast than you have because of your forms.

One final note: check the ascender height on your f. It looks like it lines with the very top of the l. Make it visually line with the lower part of the angled stem. You could probably use a little more length on the top of the f too.

Very suprised at the results at small sizes. Nice!

Randy

Weight progression: just the first two are a bit close.

Dorky dot: try skewing it (vertically).

BTW, your overall spacing is too loose.

hhp

Ok, here are my thoughts on spacing sans serifs. They come only from experience of looking, not designing.

Spacing between letters, according to Walter Tracy and others should relate to the space in the counters of the letters, as well as varying with external letter shape. I think the thickness of the stems is also a factor.

The challenge with san serifs is that absent serifs the spacing looks wide, tending to break up the words. So sans often more closely spaced than serif faces. But this close spacing can hurt readability, IMHO. Here Helvetical/Arial is a prime offender. When you have very circular letters, as opposed to narrower ovals then balancing the spacing with the counters can get ridiculously wide. But putting them very close hurts readibility, as the shapes of the letters become lost in the crush. If you look at a text setting of Meta, which is narrower and has more space in relation to the counters, it is much more readable than Helvetica.

I tend to think that Hrant is just mistaken about the greater importance of the external outline of the word, the 'Bouma' as he calls it, compared to the discernment of the shapes of the individual letters. I think that both are important. Thus wider spacing, to a point, helps, not hurts readibility.

Another factor here is x-height. Another reason Futura is more readable than Helvetica, in spite of its circular letters, is that the smaller x height allows the space between letters to be looser and hence both more balanced and legible in relation to the spaces in the counters. And, yes, taller extenders also help on differentiating more the external outline of the words.

With all these ruminations in mind, my reaction your spacing is that in the more black text setting the spacing is OK, but it is too loose in the case of the lightest weight.

In the lighter weights you might consider also narrowing the letter forms. If I am not mistaken (I don't have the font) I believe that is what Gerard Unger does with his Vesta. That way when you narrow the spacing - which the narrow stems also seem to call for - you still have balance between the inter-letter space and the internal counters.

As I said earlier in this thread, I am really struck by how readable Vesta is as a sans. Unger has written a book about is views, and I am looking forward to the English edition, which I believe his site says he is working on.

As Randy said, and as I noted earlier when you posted a smaller text setting, it is striking how readable Ardbeg is in smaller text settings. I think you have going for you not only the narrower shapes and the open counters that the angularity gives you (a la Dwiggens), but also a more generous x-height than Vesta.

All of this promises to have you end up with an exceptionally readable sans, which I think is great. I think playing with the letter widths as well as spacing with the different weights, as Vesta seems to do, may help you get to this goal.

On the differences in the weights, Unger likes the relatively small differences between different weights, I guess based on the idea that what pairing of weights a designer would want would depend on the design and the paper. James Montalbano also gave a very interesting rationale here.

Oops, I meant to say that you have more generous *extenders* going for you compared to Vesta. This might increase the readability of your type still further. (BTW I think 'Ardbeg' is too awkward and harsh sounding as a fitting name for your excellent font.)

> wider spacing, to a point, helps, not hurts readibility.

Of course. This is because:
1) About 1/3-rd of reading is done in the fovea, where internal details are available as information. (But about 2/3-rds is still done in the parafovea, where internal details are simply unavailable.*)
2) A huge determinant of a bouma is its width. The tighter the spacing the less the absolute difference in width between boumas that are potentially confusable, and the greater the reading errors.

* http://www.themicrofoundry.com/ss_read1.html

On the other hand, too-loose spacing:
1) Prevents the formation of good boumas.
2) Uses up the acuity of the retina faster.

> the smaller x height allows the space between letters to be looser

It's the other way around.

> Unger likes the relatively small differences between different weights

Me too. For one thing, it allows you to make great quasi-fake smallcaps.

> you have more generous *extenders* going for you
> compared to Vesta. This might increase the readability

It depends on the point size. The optimal balance of apparent size versus ideal boumas depends very much on scale.

hhp

Thanks again for all the feedback. I'm implementing the suggestions, and I should have a new sample in the next day or so. The contrast is being dropped substantially out of the heavy master, and I'll re-interpolate the other heavy weights based on that.

>James Montalbano also gave a very interesting rationale

I had forgotten about that post. Thanks for the reminder, William. Based on that, I'm wondering if there should be 3 steps between the regular (fourth row in the PDF above) and the heaviest weight. As opposed to 2. That would allow a Medium weight to function per James analysis. As the weights stand, I can't see them working as a regular + bold and medium + bold.

>according to Walter Tracy and others

I actually just picked up a copy of L.O.C. when I was at the Newberry Library last week. I had forgotten that it was in print again. I have yet to apply his scheme (is it the same as Briem's?) to Ardbeg however. That's this weekend's labor.

As far as the spacing goes right now, I've set the spacing for the regular weight - obviously it is in want of some tweaking. The lighter weights are optical spaced (a la InDesign) and tracked tighter - though not tightly enough. I haven't even begun to touch them for real. And the heavy weights have just some rudimentary side bearings set. Clumsy and horribly un-craftsman-like I know.

That said, I shall attempt to tweak the letterfit of all the weights this weekend, concentrating on tightening the fit of the light weights.

>it allows you to make great quasi-fake smallcaps.

Perhaps I'll use this as my next test for weight distribution...

>In the lighter weights you might consider also narrowing the letter forms.

I may yet give this a try. In my head I was thinking that it would be best to err on the side of wider characters. To keep the counters open and such. Perhaps I have been on the wrong track.

I'm afraid this post is mostly me talking to myself and organizing my game-plan. I apologize for the useless prolixity.

> I'm wondering if there should be 3 steps between the regular (fourth
> row in the PDF above) and the heaviest weight. As opposed to 2.

Not necessarily. Although an even number of weights is indeed generally better, when you get beyond a certain smoothness of gradation people will start pairing a "Bold" with a given base weight more freely.

> The lighter weights are optical spaced (a la InDesign)

Note however that InDesign's optical spacing depends on point size (a good thing).

hhp

>best to err on the side of wider characters

I think it would be worthwhile testing out both options, especially given Eric Olson's suggestion above about narrowing the h,m,n. And unlike me he actually knows what he is talking about. See for example his very well crafted Locator.

>It's the other way around [large x height allows wider letter spacing].

Hrant, why do you say this? Can you cite some examples? When I compare Futura and Helvetica, Futura has the wider spaces compared to the internal spaces of the counters. And I am guessing that this works because of the smaller x-height of Futura. Am I missing something?

>especially given Eric Olson's suggestion above about narrowing the h,m,n.

You know, this has been haunting me since he mentioned it. A while ago, I gave that a shot, but I scrapped it & I can't remember why. I should give it another shot. The 'h' has been bothering me of late.

>InDesign's optical spacing depends on point size

I've noticed that as I've been playing around with it. It's really pretty amazing.

Well, I'm off to a meeting, but later this afternoon I should have some revisions posted. I'm working on a new scheme for the heavy weights, and I've redrawn most of it. The 'g' is killing me though :-/

>>It's the other way around [large x height allows wider letter spacing].
>Hrant, why do you say this?

Here's a game: guess the hrant. :-)
Not a jab, but he is nothing if consistent on this score. Here goes: A larger x-height means a smaller ascender/descender height, means a larger looking face compared to others of the same point size, requiring more spacing at the same optical size... Meanwhile, the shorter ascenders/ascenders gain you vertical space by decreasing the needed leading, which in shorter lines of copy controls the space economy much more than the width of a typeface or looseness of spacing. So you can easily afford a bit more width. As for examples I will leave that up to him! *breath* Monarchy fovea bouma notan. (ok, that was a little boxing wizard jab :-)
How'd I do?

Randy

>If I am not mistaken

Ok, I have checked Unger's PDF and I was mistaken. He doesn't narrow the counters of the lighter letters. The counter of his heaviest weight is 44% smaller than the lightest. Also your n is actually narrower than his, if I am comparing like weights correctly. However, because of Ardbeg's squared-off structure Eric Olsen may be right (or not!) that your normal weight hmnu could benefit by being slightly narrower. I guess one issue to check is the relationship between the counters in the pbod and hmnu, which should balance.

It looks to me like Unger's seven weights run the range between your five middle weights. He says this is deliberate because he doesn't think the lightest and heaviest weights are much used. I like your heaviest weight a lot. Your lightest works the least well to my eyes.

By the way, I agree with Randy about bringing down the height of the f slightly. (The present form works better in the fat weights than the thin weights.)

>Meanwhile, the shorter ascenders/ascenders gain you vertical space by decreasing the needed leading

I was under the impression that faces with shorter extender (due to larger x-heights) were in need of more leading. Because the bloated lowercase eats up the white space. I am, most probably, wrong on this count.

Here's a random guess about x-height/letterfit. It would seem that a larger x-height increases the amount of counter. If you believe Tracy's statement that inter-letter spacing should be in balance with the size of the counters, then that would require a looser fit.

A face with a smaller x-height would work the opposite. Less counter, tighter fit.

OK, back to work.

> [large x height allows wider letter spacing]

Randy's [second-hand] logic probably comes to the same thing, but it's really much simpler: a smaller x-height [generally] means smaller counters. Look at it "absolutely": simply when you set a given font smaller (= smaller x-height) the space between the letters gets smaller too. (But more than "allows" it's "needs".)

Futura has wider spacing than Helvetica because it's designed better.

Leading/x-height:
Randy and Nathan are both right! A larger x-height normalized to the same apparent size needs less absolute leading, but at the same point size it needs more relative leading.

> It's really pretty amazing.

BTW, I've measured it. Its behavior is "bilinear": a line from 4 point to 12; a more shallow line from 12 to 72; and flat below 4 and above 72. I think they would have done a higher-order curve but performance was an issue.

hhp

>normalized to the same apparent size

Ahhh. That's what I was missing. Makes perfect sense now.

>Its behavior is "bilinear":

Cool.

>Futura has wider spacing than Helvetica because it's designed better.

Ok, I can buy that Futura is better designed.
However, I think there is more going on here. When I track Helvetica progressively wider, and the same with Futura, the words of Helvetica seem to fall apart and the whole look too loose long before Futura falls apart.

I think the longer extenders hold together the 'Boumas' (ack! did I say that?) better, so the smaller x-height sans can take looser spacing. In other words, I think the Helvetical designer did the spacing he did for a good reason, even though it messes up the readability at small sizes.

>a larger looking face compared to others of the same point size, requiring more spacing at the same optical size.

Well, Tracy says that the spacing should match the counters, which don't change with varying the extenders. But he didn't discuss sans serifs. If you look at digital Gill sans, and type iio, the space between the ii is much smaller than the counter of the o, whereas in say Times it is very similar. Now I think I read that Gill in metal was spaced much wider, but again I think the digital adjustments were in response to a problem that the spacing looked too loose in the metal.

Ok, the following is just a pet theory of mine, but there may be something in it. My theory is that without serifs you not only have problems visually finding the line of text, but also the spacing problems become substantial. Space a sans like a serif, and the words fall apart. Space it closely, and the legibility and hence readibility decreases in small (text) sizes. Hence sans are usually harder to read in text sizes. Large, the tight spacing is not a problem, as the absolute space between letters is big enough that legibility is not hurt.

I think the x-height and weight of the stem also affect how spacing works, especially with a sans. Also with narrower designs, the problems aren't so severe. I don't know what Unger has figured out about all this, but from the PDF of his Vesta, which mixes sans and serifs impressively, he has some new insights.

So even two Hrants (shudder ) don't convince me.

>Monarchy fovea bouma notan
Democracy, iris, letter, notan (ok, I like notan).

The relevance to Nathan's spacing challenges in different weights is this the following. I don't have a clear theory of how to space a sans. But my point is that there is more in spacing a sans than is usually realized, and that therefore it is worth experimenting with both letter width and spacing, rather than just one or the other, if you have a goal of readability in text for your sans.

> the words of Helvetica seem to fall apart and the
> whole look too loose long before Futura falls apart.

What weights are you comparing?

My impression is that in general Helvetica needs a slight bit of tracking to become less unreadable, while Futura is at its least unreadable by default.

> I think the longer extenders hold together the 'Boumas' (ack! did I say that?)
> better, so the smaller x-height sans can take looser spacing.

That seems to make sense. But why think of what a font "can take"? A text font (at a given size) works best for a given amount of interletter spacing, and since this is all tied together with weight, proportions, etc. a good type designer will use the right amount of spacing, and a good typographer will adjust the tracking (as best he can) depending on any deviation from the "intended" point size.

> Tracy says that the spacing should match the counters, which don't change with varying the extenders.

1) Tracy is not god.
2) A font with longer extenders will generally be used at a smaller size, very much affecting the optimal spacing.

> I don't have a clear theory of how to space a sans.

?
It's much easier to space a sans.

hhp

Well, obviously I have to ponder exactly how I plan on dealing with the spacing - it appears as a pretty major revision of letter widths might be in order in the future.

In the meantime, I've done some changes. This one has an added heavy weight which I'm hoping will make the regular/bold, medium/black combinations work better. As well as diminshing the difference between regular and medium. Unfortunately I don't have the heavy cpitals drawn, so I didn't include any tests for small caps like Hrant mentioned.

The heavier l/c g's are awful. They are in need of more weight it seems.

The contrast in heavier weights has been much reduced (perhaps too much in some places).

The f is lowered (thank you, Randy).

I spent a little time last night starting to space the thin weights, but it's still nowhere near complete. I suppose that's next.

Thanks.

WeightComparison05.pdf (141.8 k)

Decent weight progression. I'd just make the 5th weight a bit darker.

Spacing: the lighters are progressively too looser. (And all your base are belong to us.)

hhp

> the 5th weight a bit darker

My home-made, cobbled-together mathematics pointed me towards 2 weights - I chose the lighter. But upon closer inspection I agree wholeheartedly.

>the lighters are progressively too looser.

Thank you. Your eyes are infinitely more attuned to this sort of thing than mine. As if that needed saying.

>all your base are belong to us

Is that from a book/movie, Hrant? It sounds familiar. Either my mind is playing tricks on me or just not working very well.

>But why think of what a font "can take"?

Perhaps this is off-topic (as it, at best, indirectly concerns this crit) but wasn't there a thread that brought something like this up? Something about how "easy" it was to use certain fonts - i.e., some fonts are easier to use than others.

Is that something that's worth-while (in some instances if not every) to consider in designing a font? I mean, we consider how easy it is to read, what about how easy it is to use?

Although, perhaps the best way to make it easier to use is to minimize the need for tracking adjustments and such things. Which, of course, means impeccable spacing.

I like your old heaviest and next to heaviest weight better. I think the thinner joins go better with the chiseled 'woodcut' look of the face.

The 'easy to use' issue is another way of saying what I been ruminating about in my posts. If your face can work as a text face as well as display, it will be 'easy to use'. This is where Helvetica fails, and Futura works better.

Because it is narrow and open at the same time I think you can make Ardbeg a very good text face, with work on the issues of rhythm and spacing.

> Is that from a book/movie, Hrant?

It's from some computer game. It sort of epitomizes bad English - as in the sentence I wrote right before it.

> If your face can work as a text face as well as display

To me that would mean it's not very good at either.

But the topic of "ease of use" of a font is very interesting.

hhp

>To me that would mean it's not very good at either.

In another thread James Montalbano dismisses your view on this as silly. I wouldn't go so far, but I would think with so many weights, and Nathan's particular design, it should be possible to do both quite well.

It is interesting, though, that Gerard Unger has introduced a new 'Big Vesta', with shorter extenders, for headlines. So he felt a need to have an altered design for display for his text-friendly Vesta.

> Is that from a book/movie, Hrant?
Nathan, Just in case you want to see what popularized that phrase around much of the internet.
http://www.planettribes.com/allyourbase/AYB2.swf

>it should be possible to do both quite well [i.e., work as a text and display face]

When I started drawing Ardbeg back in the summer, that was actually something that was rolling in my head. My hope is (or was) that the corners in the counters would have something to do with that. Perhaps opening the counters or livening the face and small sizes, and adding some sort of 'character' (for lack of a better word) at large ones. So it's heartening that you see some possibility, William.

>To me that would mean it's not very good at either.

And, of course this gives me pause. Because it does make sense. In the same way that a Swiss Army knife isn't the best knife, nor the best screwdriver, or the best scissors, etc.

But, although I don't carry a Swiss Army knife (Benchmade, thank you very munch), I still think they must have some quality that recommends them as useful tools.

Anyhow. My work's cut out for me. I suppose a Titling or Display version might be in the works. In about 10 years ;-)

Thanks for that link, Mitchell. Funny stuff.

[Aaargh. Damnable refresh button.]

>I like your old heaviest and next to heaviest weight better. I think the thinner joins go better with the chiseled 'woodcut' look of the face.

Is it the entirety of the face that bugs you William? Or just some of the letters? As I've been looking at it today, the n, m, h, u really bug me too. That is, I agree with your comment about the joints - way too thick now.

Some of the other letters I prefer the new way. I believe it's because the lighter weights are predominately monolinear, and the higher amount of contrast in the black seems out of context.

>some of the letters?

Yes, I was refering only to the heavy weights. I am wondering is whether it is better to have more differentiation between the way the different weights are handled than is often the case. I suspect that mechanical interpolations, while no doubt saving a lot of work, lead to a better result when they are reworked with both aesthetics and the end use in mind.

On the mnhu, I just haven't spent the time as you have or have the experience, as some others who post here do, to say much helpful. I just think you should not hesitate to change widths or x heights or degree of modulation or spacing with the different weights if you think that will serve the end result. You do need to stay in touch with your orginal inspiration, the 'chiseled' look on this face from the angular counters. But I suspect the family will be better if it is expressed differently in the different weights.

Agreed. Interpolation does save some time initially, but with the re-drawing later on I'm not sure if it doesn't all end up equal in the final analysis.

But since the weights, for the most part, seem fairly close, I believe the time to tweak the crude mechanical output is nigh.

Some interesting ideas in your last post that I'll have to play with this weekend. Perhaps the 2 heaviest weights I've drawn can serve as 2 seperate masters for the intermediate heavies. Or something like that. Blah blah blah. Thanks once more for your insights.

Looking again at the light weights, I am wondering whether the joins on the hmnu on the light weight could be slightly heavier.