FAQ: Vietnamese

Nguyen Mai An

Hi,
As some suggestions at the previous thread, I creat this topic for you to ask questions regard adding characters supporting Vietnamese.

A guide for Vietnamese typography written by Donny Truong: http://vietnamesetypography.com

I'm a native Vietnamese speaker, a self-study type design, and I can (hopefully) answer the general questions you might have about designing typeface for Vietnamese, and/or extend your existing typeface so it can support Vietnamese. Nevertheless, you should read the above mentioned guide before asking questions, as it already covers the basics. 

I'm also open for bussiness in extend your typeface for supporting Vietnamese. Thanks for reading!

I'll update the new questions answered at the top of this thread.

Dyana Weissman

Thanks for the offer! I'm curious what your thoughts are on this thread? 

Nguyen Mai An

Dyana Weissman said:

Thanks for the offer, Nguyen! I'm curious what your thoughts are on this thread? 

I believe this topic is already covered in Donny Truong's guide, at section 5: diacritical details. You can read it here. Then if you have any more (specific) question, I'll answer depends on my knowledge.

Thanks for your attention! 

Dyana Weissman

I'm interested, specifically, on the preferred placement of the acute relative to the circumflex. In that thread it is claimed that the acute has been seen placed to the left of the circumflex, albeit rarely. I'm wondering if that is acceptable today, and if so, how often does it happen? 

I had a client once who was certain that the acute could only be on the right, and I didn't believe I had the authority to challenge this notion, even though the placement on the left was a better solution visually. The client was not a native speaker, either, and I wondered if they were simply used to using Times New Roman. It would be great to have as many native speakers weigh in as possible. 

Nguyen Mai An

About your question, yes, the acute could be placed to the left of the circumflex. It does have consequences, however. Firstly, some, although not all of Vietnamese might be uncomfortable reading with the acute placed to the left of the circumflex, as they're more used to the usual placement. I'll explain more clearly in the example below with the word mất, (meaning: losing something), with my Vietnamese version of Montserrat:


As you can see, the 3rd example is hard to read at small size. Even in display size, it's still not easy to read for us. 

BUT, in this case, it depends on the weight of the stroke as well. The acute can be place to the left of the circumflex and still manage legibility for the majority of Vietnamese, as long as the stroke weight is light/thin. As it will be more distinguishable than the bold counterpart.


This is totally acceptable. But should be used in display size instead of text size.

I don't see any reason to keep the acute to the left of the circumflex, because you'll not likely to be able to do that in the bold/heavy weight. Therefore, I recommend stick with the usual placement if your typeface is a text/body one. As for the display ones, you can be free with pretty much any placement I listed above, and even in these photos at http://luuchu.com

Dyana Weissman

Really helpful, thank you so much!

Kent Lew

At TypeCon in Washington DC a few years ago, I had occasion to talk at length with a Vietnamese type aficionado (not a graphic designer nor type designer, but a sincere typophile). He explained a few things that really helped my understanding, and which might not be familiar to other non-native type designers.

When it comes to Vietnamese accents, I think it is helpful to understand the two different roles that the different sets play. This is, of course, explained on Trương’s site, but the distinctions might not have come through clearly to the uninitiated.

One group of accents indicates different vowel sounds, in terms of quality of pronunciation — most notably the circumflex and breve (also the horn).

The other groups of accents indicates tone pattern — a characteristic that most Western languages do not distinguish in their alphabets, per se. These are the acute, grave, hook, tilde, and dot below.

So, when it comes to designing the tricky double accents, I find it helpful to understand the difference and to note that in a letter like ấ, for example, the circumflex is part of the base vowel sound and the acute is the rising tone indicator.

Regarding the placement of the tone mark relative to the circumflex accent, my contact explained to me that in handwriting, most often people write the tone mark either directly above or slightly to the right, in the direction of writing/reading, but rarely to the left.

Also with regard to this question — especially the pros/cons of grave to the left — he explained that “having both on the right presents many more benefits. First, it seems more natural due to the left-to-right flow of reading and writing. Second, the acute-circumflex/grave-circumflex could be preceded by a character with an ascender ("đề"), but is never *followed* by a character with an ascender, so there's much more space to exploit if you place both on the right.”

Mai An — I would welcome your thoughts on these comments.

Ray Larabie

Hopefully some readers can see the correct characters...in Windows Chrome, Vietnamese characters aren't displayed correctly on this site.

How important is accent alignment? For example, Ẳ (A+breve+hook) compared to Ă (A+breve). Sometimes, especially in heavy weights, I drop the Ẳ breve a little bit so the hook isn't too high. Same with ẴẮẰ.

Is it okay to vertically stack the lowercase accents like in your 2nd example and offset the accents on the capitals as in your 1st example? The lowercase accent stack would still be lower that regular capital accents so it would have no affect on line spacing. I've done this before...it seems practical because it reduces the danger of lowercase collisions.

James Puckett

It’s wonderful that Vietnamese designers are writing about Vietnamese diacritical placement in English. But as is too often the case, the information is turning up on forums, facebook, and in small press books. It would be great if this information were consolidated over at the Diacritics Project.

Nguyen Mai An

Kent Lew said:

One group of accents indicates different vowel sounds, in terms of quality of pronunciation — most notably the circumflex and breve (also the horn).

The other groups of accents indicates tone pattern — a characteristic that most Western languages do not distinguish in their alphabets, per se. These are the acute, grave, hook, tilde, and dot below.

So, when it comes to designing the tricky double accents, I find it helpful to understand the difference and to note that in a letter like ấ, for example, the circumflex is part of the base vowel sound and the acute is the rising tone indicator. 

This is totally true. But I think it's more helpful to the people who are learning to speak Vietnamese instead of type designer. Type designers tend to focus on the placement, form, and distance. I'm trying to explain more or less at a type designer's point of view, base to my own knowledge of my native language, and my experience working as a freelancer.

Kent Lew said:

Regarding the placement of the tone mark relative to the circumflex accent, my contact explained to me that in handwriting, most often people write the tone mark either directly above or slightly to the right, in the direction of writing/reading, but rarely to the left.

Also with regard to this question — especially the pros/cons of grave to the left — he explained that “having both on the right presents many more benefits. First, it seems more natural due to the left-to-right flow of reading and writing.

Yes, this is true too. You might consider “the easiest way to read” for us, is following “the way we write”. For example, we read and write from left to right (even with the left handed people), so everything will be placed in that direction. If we are reading something from left to right, and then must go back to the left to read the accent, it'll not be comfortable.

The case with grave is different and more complicated though. Grave is on the left sometimes can be uncomfortable to read, but not like the acute, the direction of the grave is different from the circumflex first stroke's. So it's more distiguishable. Additionally, grave is placed to the left of the circumflex can be consider true to their nature as “the direct opposite” of the acute (if the acute is placed to the right of circumflex), so it can be accepted.

Kent Lew said:

Second, the acute-circumflex/grave-circumflex could be preceded by a character with an ascender ("đề"), but is never *followed* by a character with an ascender, so there's much more space to exploit if you place both on the right.”

Not really, in the cases with “mất” or “bết”, or “ngất”, acute-circumflex is followed by a “t”, which has a short ascender.

Ray Larabie said:

Hopefully some readers can see the correct characters...in Windows Chrome, Vietnamese characters aren't displayed correctly on this site.

Actually due to the webfont of this site doesn't have Vietnamese characters, they will subtitute whatever system font has this character available. In the most cases, it will be either Arial or Times new roman. It's awful, really. 

This is one case I often see across Vietnamese websites. They don't even know they're using the wrong typeface.

Ray Larabie said:

How important is accent alignment? For example, Ẳ (A+breve+hook) compared to Ă (A+breve). Sometimes, especially in heavy weights, I drop the Ẳ breve a little bit so the hook isn't too high. Same with ẴẮẰ.

Is it okay to vertically stack the lowercase accents like in your 2nd example and offset the accents on the capitals as in your 1st example? The lowercase accent stack would still be lower that regular capital accents so it would have no affect on line spacing. I've done this before...it seems practical because it reduces the danger of lowercase collisions.

Surprisingly, the two cases you mentioned is totally possible. Most of the readers will not notice the changes, unless the different is too clear, and be placed near each other for comparing. You might get away with just slightly different alignment. Nevertheless, they often prefer maintaining the consistency across the font. 

Base on my experience, if you're working with a Vietnamese client, they'll most likely want everything's inline. Especially in allcaps headline and/or texts in display size.

Ray Larabie

Which accent collisions do you see the most?

Kent Lew

Mai An — Thank you for your responses.

Type designers tend to focus on the placement, form, and distance.

True. But some text typeface designers also tend to focus on how one reads (even when we don’t actually read the language ;-).

Kent Lew

I’d like to ask for some native perspective on the đồng symbol ₫.

In many common fonts (e.g., most Microsoft fonts), this symbol is merely a đ with an underscore below the baseline.

I have also seen several examples where the glyph is superscripted, with the underscore sitting more or less on the baseline. This includes the example in the Unicode code charts, I think, if you look closely.

Searching for examples “in the wild” (on the internet, anyway), I note that on Vietnamese postage stamps, the currency symbol is noticeably superscripted and the underscore is absent. Perhaps these are not really the symbol, per se, but just a superscripted abbreviation.

What is your impression of an overall preferred form for the đồng symbol, in terms of superscripted or not (and how much) and the presence of an underscore?

Does the average Vietnamese user actually type the ₫, or do they just abbreviate with đ?

Thy Ha

Kent Lew said:

I’d like to ask for some native perspective on the đồng symbol ₫.

In many common fonts (e.g., most Microsoft fonts), this symbol is merely a đ with an underscore below the baseline.

I have also seen several examples where the glyph is superscripted, with the underscore sitting more or less on the baseline. This includes the example in the Unicode code charts, I think, if you look closely.

Searching for examples “in the wild” (on the internet, anyway), I note that on Vietnamese postage stamps, the currency symbol is noticeably superscripted and the underscore is absent. Perhaps these are not really the symbol, per se, but just a superscripted abbreviation.

What is your impression of an overall preferred form for the đồng symbol, in terms of superscripted or not (and how much) and the presence of an underscore?

Does the average Vietnamese user actually type the ₫, or do they just abbreviate with đ?

Hi Kent, I'm another Vietnamese native here. It's weird that I'm replying to this after more than 4 years but I hope this will help.

On a daily basis, we use abbreviate 'đ' the most since it is quick to type or the whole word 'đồng'. To be honest, I don't even know how to type the symbol ₫ on my keyboard (sad face). And in handwriting, most of us write the superscripted đ without the underscore. 

Kent Lew

Thanks for the reply! There’s no statute of limitations on native perspective. Informed input is always welcome. ;-)

John Savard

On the Vietnamese keyboard provided with Windows 10, to type an equals sign (=) you have to use AltGr with the same key as is used for the equals sign on the U.S. keyboard. Without the AltGr key, that key generates the currency symbol being sought.

Piotr Grochowski

It can be seen at https://unicodesubsets.miraheze.org/wiki/Logic_table_of_supersets#Character_sets_more_than_8-bit_comparison that five character sets documented in Unicode Subsets wiki contain Vietnamese: LPTT-1, LPTT-1.1, MES-3B, Subset3, Subset3+. So a font consisting of any of these character sets will automatically contain Vietnamese as well, but not necessarily the other way around.

Some Type Design fonts have Vietnamese. These include: DMCA Sans Serif 9.0 (which contains Subset3+), Custom Font ttf 3.0 (which contains Subset3) and Riglos Mono 1.1 (which contains LPTT-1.1).