Danat

Yes, resolving the issue I posted would serve as help/reference for other users that have similar issues. That's how I usually find answers to my questions - by searching for topics that directly/indirectly address the issue I'm having. I didn't find one here, so I've decided to create it since the question is broad enough (how font size in PDN works). Didn't I say I want this topic to be a good contribution for others ? I've made the (now invalid) workaround in large font for this purpose as well. Btw, I think I'll use strike-through style now there to avoid confusion. You seem to enjoy providing invalid points . Even the thread that I got my quote from in the first post suggests that users are interested in how font size works. That's rather inefficient if you do it every time, when you can just ask about how it works on the forum. The only question that I noticed i missed is "why Arial font". The answer is - its just a standard font that any user of this forum probably have and hence can reproduce my results. I was actually using a non-standard font for my issue but the result is still the same. Nah, you were just insisting that the font size meaning I've found is incorrect/invalid/irrelevant which, if true, would render the point of this topic meaningless, so I kept on defending my statements. Well as a result of me posting here I've learned the right formula today so now I can effectively do the conversions and also shared it in this topic, so I think overall it was a success.

I wasn't referring to 1 pixel difference, but to the fact that "j" or "a" letter size in no way fits your definition of fontsize as "the size of a single letter", otherwise they would also had to be of the 40px height. Even "T" letter in 200 font size in PDN is 192 pixels tall, so 200 can't be interpreted as an approximate size of capital letters either. Yes. That's my goal - to make sure there would not be misunderstandings about font size in PDN in the future. That is the sole reason I created this topic, since I couldn't find any info explaining my issue and thought it would be a good contribution for others to use as well. Yet I have a feeling like I'm portrayed as the bad guy here ;). I've already solved my problem yesterday with my "workaround". Like I said, I rarely work with graphics (and thus with PDN), so its not like I need a program which lets me specify font size in pixels. Given the formula quoted in my post above, figuring out what's the resulting font pixel size is no longer an issue. Eh ? Was I supposed to just guess from that font size is not specified in pixels although that is the word being used in explanation ? Sure, there are hints in Rick's post about it being points, but I understand them only now when I already know the answer: it's units which are equal to "pt" under 96 DPI. I'm a web dev at the moment and I'm used to thinking that "pt" and "em" are DPI dependent, so when I've read "pixels" and "don't take your image's DPI into consideration" the assumption was that it must be pixels that specify the font size.

Hi Rick. Thank you for replying. It's kinda unfair that you cut the second part of my quote saying "unless I'm missing something which is what the topic is about" and said the same thing yourself ;). Also I did try to figure out, as I provided links to my findings about what "font size" should specify. Ok, now its my turn to sound like a jerk . Its 40px tall because T is in uppercase, "t" is 40px tall but thats because it has an ascender, so if we use "ace" as text it will be 30px tall. Why are we even considering descenders as something non-normal in context of font height and ascenders are ok? Also thats not what IRON67 said - he said individual letters are 40px tall and thats what the font-size means which is plain wrong. Yeah my bad on this one. Well I've just figured out that 40 probably means 40pt under 96 DPI which can be easily converted to pixel-based font size using a lookup table, or the following formula from this link FLOAT ConvertPointSizeToDIP(FLOAT points) { return (points/72.0f)*96.0f; } Thats why in my example 30 font size in PDN meant (30/72)*96 = 40px font size. I've tested font with different sizes in PDN and it seems my assumption is correct. The statement I quoted in my first post confused me into thinking "font size 40" actually means "font size 40px" because you referred to font sizes as "following pixels". Though I still don't understand why did not you just say in this topic that "its points duh" ;). It's not complicated at all at this point contrary to what you stated. Sure, I'm a dev myself and I remember the pain I had when trying to use GDI to measure the text size properly, but we're merely talking about units of font size here. Now it feels this thread has gotten way longer than it might have been . Well, unless I'm again missing something.

Yeah I also don't think its critical or smth, I actually was hoping that somebody would explain why it works that way or show a setting that fixes it, because its weird that paint.net exists for so long and yet has so easily discoverable inconsistency. Forgot to say that I did find a simple workaround for this, in case someone stumbles on this as well: Just subtract 10 from font size to get what you need: 50px font in browser = 40px font in paint.net 40px font in browser = 30px font in paint.net ... EDIT: workaround is invalid, read below for the right answer.

well its not about web-design really. Its about text and the commonly accepted definition of "font size". In paint.net you specify font size 40px, yet you get letters that have 48px height - there is no valid explanation for this logic unless I'm missing something which is what the topic is about. I dont see any point continuing this discussion though. It seems after your own definition of font size has been counter-argumented you decided to just keep on saying that "paint.net is as is" which gets us nowhere. I'll try to PM the author I guess after a day or so. It's not that I use paint.net a lot, but I decided to bring this inconsistency to his attention in case he cares.

Well its not that you should, but I've managed to provide a counter argument to what you said about font size, using paint.net itself. I'm sure most web designers think that way otherwise everything web related would prove them wrong anyways. Also that linked forum is dedicated to "Graphic Design" which means this font size definition is probably valid in most graphic-related tools. So I don't get your "hardly anyone thinks". It's kinda funny that you previously said: So now paint.net is not a program that gives me what I want ?

I quoted how "font size" is defined ("The "font size" of a font refers ...", "The size of type ...")

Okay I did another test. Try the same example with "The jar" and measure "j" - it will have 48px height, although the font size is 40px. So, what does the "font size" mean?

Well from that same thread: Which is also how browser works. Seems like the common definition of "font size" is not what you said. So, what did I miss ? So you agree that paint.net uses its own definition of what font-size means right? EDIT: I'll explain what I'm doing. I need to reproduce the text I have in a webpage. The text has 40px font size in CSS. In paint.net I also specify 40px font size but get significantly larger text which is not what I think a user would expect in this situation considering there seems to be a general agreement on what "font size" means.

Hello According to this post: I would expect font size to actually have the height that I've set for it in pixels, yet if you try to type "The dog" in Arial 40px and then try to wrap the entire text in selection rectangle, you would notice the rectangle height to be ~50px. If I try to do the same in CSS, the browser shows the font with the right size in pixels. This thread on stackexchange also suggests that what paint.net does is incorrect. Am i missing something or has this been overlooked?