Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Scott5114]

This is some very good information, and useful. Thank you. I can definitely see a lot of the things you mention with some experimentation in Inkscape. It would be trivial to ensure the X/Y pair match (just lop off the bottom half of an X and add a stem for a new Y, which doesn't look too out-of-place).

this is actually the case with BPR 1926 (the old block font).

FHWA "K" is downright quirky; I've never seen a K that looks like that any other typeface.

see again: BPR 1926, and some of the actual implementations.  I think 1948 is identical to one of the more popular 1926 variants.  I will have to check.

good point-a lot of the FHWA Series fonts' quirks probably trace from the BPR fonts.

and here's where I lose you.  am I the only one that finds Helvetica to be positively garish?  never mind the tiny-to-moderate differences between Helvetica and Arial and the other ones in that family - I even think Clearview looks a lot better than Helvetica.

it is far from my favorite; mostly the comparison arises to use Helvetica as a generally recognized cromulent typeface. Helvetica is useful as a boring default when you want to display text with no extra connotations from the font.

As mentioned before, though, FHWA cannot do many of these things because they do not create a software implementation of the fonts. They merely offer the characters in PDF form and spacing tables, and it's up to professional type foundries to create a TTF/OTF file from this specification.

I have seen lengths-and-radii specifications of all the uppercase series (A-F), and the lowercase EM.  so there is definitely would be a canonical implementation - if it weren't for the fact that the lengths and radii are occasionally overspecified and contradictory.  the only glyphs I've ever constructed from these definitions are Series A uppercase and numbers: I had a hell of a time with "8" and used the drawing to resolve the issues to a best approximation.

These days the specs are just vector objects in a PDF.

[vtk]

These days the specs are just vector objects in a PDF.

Actually, the specs are text objects with an embedded font — the Saa version, I think — in a PDF.  I don't think they intended for vector data to be directly extractable from the PDF, as helpful as that may be.  The intended purpose of those pages is probably to be printed and used in a low-tech fashion by the few people who need official font specs and haven't embraced the digital age.

[Bobby5280]

It's easy for any agency to go way overboard on mandating specifications for elements in a design or even a typeface. It's also easy for people who aren't actually doing the design work to request things that don't add up mathematically.

If the FHWA was to mandate some specifics for a new type familiy for road signs they should mostly stick to the basics, such as ratio of lowercase characters in proportion to the uppercase characters, approximate stroke thickness (this varies in most typefaces) and not too much else.

am I the only one that finds Helvetica to be positively garish?  never mind the tiny-to-moderate differences between Helvetica and Arial and the other ones in that family - I even think Clearview looks a lot better than Helvetica.

A lot of it is a matter of taste -and also which version of Helvetica is being used. The original 1950's version has some odd features. The "Neue" version, created in the 1980's, cleaned up a lot of things and added a bunch of new weights. The latest "Pro" version of Helvetica Neue added some OpenType features and some new compressed weights.

Helvetica can be a hazard to deal with in sign work. The different versions may all look the same, but they do have substantial differences. That can be a nightmare if you're having to replace trim-capped acrylic faces on a channel letter sign. If you don't use the exact same font file your finished sign parts won't fit the existing sign. There's not only varied versions of Helvetica. Clones like Swiss 721 and Nimbus Sans have subtle differences. These days too many people love squeezing and stretching fonts out of their original proportions. In the end it really helps to have the original art files used to make the sign in the first place.

Helvetica is a clean, neutral (if not plain looking) typeface. It looks a lot better than Arial simply because it is a lot more consistent looking than Arial. The problem with Arial is Monotype just couldn't figure out what Arial needed to be visually. They tried to fit all the metrics of Helvetica and mimic a little of its style. But then they tried copying visual cues from Akzidenz Grotesk and Univers. In the end Arial would up being a hodge-podge of ugly crap.

Not everyone likes Clearview, but at least Clearview has a specific visual style -unlike Arial.

[agentsteel53]

good point-a lot of the FHWA Series fonts' quirks probably trace from the BPR fonts.

correct.  I know for a fact that they intended to preserve the spacing exactly between the block and round variants of the various widths.  in some cases this didn't quite happen (series B "S" takes up just a bit more space in round than block, for example) but for the most part it is true.

Helvetica is useful as a boring default when you want to display text with no extra connotations from the font.

that's actually what Highway Gothic represents to me.  a simple, basic font, with none of the connotations of "we are a high-faluting design firm, and/or a subway agency that thinks FHWA 1948 is too boring".

These days the specs are just vector objects in a PDF.

are these vector objects plus the spacings sufficient to describe a font file?  I feel like they are.  thus, it's just a matter of a script to do the translation from one vector description language (PDF + spacings) to another (TTF/OTF).  no need for a "professional type foundry"; just someone with coding ability and a day or two of time. 

[agentsteel53]

These days too many people love squeezing and stretching fonts out of their original proportions.

*shudder*

[PHLBOS]

These days too many people love squeezing and stretching fonts out of their original proportions.

*shudder*

Man, that's even worse than some of the elongated Series D numerals on some 3dI-shields in PA I've seen.

Exhibit A

While not terrible, but with it've killed PennDOT to just simply use Series C font for these shields?

[JoePCool14]

ISTHA and IDOT also do that for BGS signs with 3DI's.

[Scott5114]

These days the specs are just vector objects in a PDF.

are these vector objects plus the spacings sufficient to describe a font file?  I feel like they are.  thus, it's just a matter of a script to do the translation from one vector description language (PDF + spacings) to another (TTF/OTF).  no need for a "professional type foundry"; just someone with coding ability and a day or two of time. 

Yes, this is exactly what Sammi did. (See the "Roadgeek 2014 fonts" thread). All she had to do was copy the characters out of the PDF and into a font editor and apply the spacing. The result is a mostly-usable font file; some adjustments to the official spacing specs to make it nicer-looking could be done, but her current release is basically the same as the Roadgeek 2005 fonts that we all have installed on our systems.

[Bobby5280]

Oh yeah, that one is a classic. I've seen that one in person. Tulsa had a few more of those horrible things spread around town, even on overhead signs along I-44. I know a couple of the overhead signs were replaced last year with new sign panels bearing correct Interstate shields and Clearview legends. Still, some of those shield disasters may still be lingering around the Tulsa landscape. There's some other odd-ball I-35 Interstate shields on overhead signs just a couple or so miles before the Turner Turnpike runs into I-35. At least the lettering and numerals are correct; the shield shape is the only thing screwed up.

[hbelkins]

I actually like those Oklahoma route markers pictured above precisely because they're different.

And the odd-shaped interstate route markers on the overheads are what you get when you take the standard three-digit wide shield and compress it into a two-digit shield. Conversely, when you take a standard two-digit shield and stretch it to three-digit width, you get the bubble shield. Which is why I like the bubble shield, because I think it's foolish to have two different shield shapes for the same type of route. The bubble shield is the logical extension (pun intended) of the standard two-digit interstate route marker.

[vtk]

because I think it's foolish to have two different shield shapes for the same type of route.

Stretching makes it a different shape.

[Scott5114]

Yes. The reason why the shields are different is because the Interstate shield is composed of circular arcs:

(the circles don't match up exactly because I was eyeballing it, but you get the idea)

When you create a bubble shield, those circles are stretched out to ovals. The proper 3di shield is a wider shield that is still composed of circular arcs.

Why is it so important that the arcs be circular? Because, in general, humans perceive circles to be more aesthetically pleasing than ovals. I found this out myself when I was doing a series of playing card images for my company—the arcs were even less noticeable than they were in the Interstate shield, but when I stretched things out into ovals, they looked cheap and 'wrong'. Keep stuff circular, and it looks fine.

There's also the fact that in 1956 when the Interstate shield was designed, there was no stretching of anything. There was no CAD that allowed you to squish everything. Instead, you had the high-tech tools of a compass and straightedge. And a circle is a lot easier to make with a compass than an oval.

[mrsman]

Yes. The reason why the shields are different is because the Interstate shield is composed of circular arcs:

(the circles don't match up exactly because I was eyeballing it, but you get the idea)

When you create a bubble shield, those circles are stretched out to ovals. The proper 3di shield is a wider shield that is still composed of circular arcs.

Why is it so important that the arcs be circular? Because, in general, humans perceive circles to be more aesthetically pleasing than ovals. I found this out myself when I was doing a series of playing card images for my company—the arcs were even less noticeable than they were in the Interstate shield, but when I stretched things out into ovals, they looked cheap and 'wrong'. Keep stuff circular, and it looks fine.

There's also the fact that in 1956 when the Interstate shield was designed, there was no stretching of anything. There was no CAD that allowed you to squish everything. Instead, you had the high-tech tools of a compass and straightedge. And a circle is a lot easier to make with a compass than an oval.

I wonder if you came across a similar diagram (intersecting circles) for the US highway cutout sign (like in California).  There are enough curved edges that I imagine the sign is also the conjunction of several intersecting circles.

[J N Winkler]

I wonder if you came across a similar diagram (intersecting circles) for the US highway cutout sign (like in California).  There are enough curved edges that I imagine the sign is also the conjunction of several intersecting circles.

There are actually multiple designs of the US route shield and I am not aware of any that are officially approved and use non-circular arcs; the arcs I have seen are all circular.  California used to have a particularly interesting guide-sign design where the sides were designed to be slid out as needed to accommodate different numbers of digits.  If memory serves, there were four separate widths available.

[J N Winkler]

Breaking news:  We have a first defection from Clearview.  Iowa DOT, which has been using Clearview for several years, just advertised a signing contract (Call 351 in the letting of July 15, 2014) that uses the FHWA series instead.

[jakeroot]

Breaking news:  We have a first defection from Clearview.  Iowa DOT, which has been using Clearview for several years, just advertised a signing contract (Call 351 in the letting of July 15, 2014) that uses the FHWA series instead.

Could you provide a link? I'm wanting to read the contract but can't find it.

[J N Winkler]

Direct link:

http://www.iowadot.gov/contracts/biddocuments/july_2014_letting/15JUL351.zip

It is accessible from this page (may have to click "All" in the dropdown box to expose the full listing):

http://www.iowadot.gov/contracts/biddocuments/july2014.html

And, actually, I discovered an even earlier contract using Series E Modified after I made my last post.  It was Call 353 in the June letting.

http://www.iowadot.gov/contracts/biddocuments/june_2014_letting/17JUN353.zip

http://www.iowadot.gov/contracts/biddocuments/june2014.html

[jakeroot]

Thanks! I thought those ZIPs were the contracts but I got lost really fast.

On a side note, I'm wondering what made the Iowa DOT switch back to FHWA Series? I am aware of the FHWA's notice to Grays Harbor WA about an end to interim approval, but Iowa ending their usage of Clearview seems to imply that either A) interim approval has ended and thus the approvals are no longer approved (!) or B) Iowa DOT decided they didn't like Clearview anymore? I must have missed something in the last few pages.

[J N Winkler]

My suspicion (and it is only that--I haven't actually tried to contact anyone there to obtain the inside line) is that they are just going ahead and going back to Series E Modified without waiting for FHWA to revoke the Clearview interim approval, since the handwriting is already on the wall for Clearview.

I think that when FHWA does actually revoke the Clearview IA, it will be a big enough deal that a notice will be posted on the top page of the MUTCD subsite, which I am sure many members of both the professional and enthusiast communities monitor using Watchthatpage.com.  There will very likely be a race to start the first thread about it on this forum.

[jakeroot]

Unrelated to everything you've said thus far, J N Winkler, what would you think if the FHWA granted interim approval of the Transport typeface? I see you've spent time in England.

[Scott5114]

Can't speak for J.N. Winkler, but I think that it would come as a major shock...there has been little study comparing the two typefaces, and I would expect that after the Clearview experience, FHWA would want to see such studies before approving another interim approval.

Personally, I wouldn't mind it if they did investigate its use in the US. Transport is a perfectly fine typeface and has proven itself at least as capable as FHWA Series throughout the world.

[on_wisconsin]

Speaking of Transport, does anyone know if there is a way to get an official version of that typeface, for free.

[Zeffy]

Speaking of Transport, does anyone know if there is a way to get an official version of that typeface, for free.

Did http://www.roadsuk.com go down? I remember getting "Transport Medium" and "Transport Heavy" typefaces from there for free. If someone wants to tell me if they are actually the real deal (I believe they are), then I'll re-up them.

Can't speak for J.N. Winkler, but I think that it would come as a major shock...there has been little study comparing the two typefaces, and I would expect that after the Clearview experience, FHWA would want to see such studies before approving another interim approval.

Personally, I wouldn't mind it if they did investigate its use in the US. Transport is a perfectly fine typeface and has proven itself at least as capable as FHWA Series throughout the world.

I am also supportive of testing Transport for US based road signs. I would suggest DIN 1451, but there needs to be thicker weights I think.

[sammi]

Speaking of Transport, does anyone know if there is a way to get an official version of that typeface, for free.

The only 'official' version is in the form of drawings on the government website (which by the way uses a web version of Transport; it looks nice on web pages). If you want a pre-made version, the best quality is by URW for $35. The best free ones are from CBRD.

[on_wisconsin]

Quick & dirty Transport mock up:

(The control city legend size turned out a little too big.)