Thanks for the edit, but I have to ask, what font did you use? Because it's a little hard to read... The thin lines tend to run together in my eyes. (Also, what color did you use? It doesn't look like pure black; either gray, which doesn't help with the contrast, or black that's been anti-aliased to all hell...)
Thanks for your work, but erm, I have to say it as well... the font isn't very good (I use capitalised Comic Sans). I think it's also because the italics are heading the wrong way...
The font was a font called 'walkway rev oblique' , I used it because I thought it looked like the original writing ... somehow
as for the colour I think it was me compressing the format to jpeg instead of png (which seemed to murder picture quality) , I don't know these kinda things as this was my first time doing this kinda thing.
Strange... I'm sure that .png is always better quality. If you want to continue using this font, have you set it to bold? Doing so might help it stand out more and make it easier to read.
>(which seemed to murder picture quality) Umm. I don't know how you did that, but I'm willing to bet you did something wrong.
First thing, the PNG itself can't ruin quality, as it is a loss-less compression... it can only ruin the filesize if the image is full of gradients and stuff (this one isn't), or if it's in 24 bit RGB, or 32bit RGBA mode when it doesn't need to (and this one doesn't need to, grayscale 8 bits per pixel is more than enough).
For that reason, you should usually* change the mode to 8bit grayscale, or indexed colors, before saving as PNG. That conversion can sometimes ruin the quality, but I don't quite see how it could on a picture like this... but perhaps you did that wrong... somehow?
Funny fact: just taking this image, converting to grayscale, and saving as PNG shrinks the filesize by 10kb. No quality loss (lossless). And probably over 60% of that filesize is wasted on preserving the JPG artifacts (lossless >_>).
Saving the original file, without the artifacts, directly as PNG (greyscale, or 16 colors-indexed) should give you better quality, and filesize about 100kb or less.
About the font... I'm not an expert, but I'd preffer, ALLCAPS, and a font with less italics, and I think you can bump the size up by 1 or even 2. I don't think it's a good idea to try mimicking the 1-pixel-wide-lines of the original Japanese font. And if you really wanted to do that, perhaps a bitmap font actually designed for pixel work would be called for?
>>upload size limits with the png. I see. You probably had it in 24 or 32 bit mode. Convert it to grayscale or indexed colors (16 colors should suffice). I'm not sure where the setting is in Photoshop, but in GIMP it's Image>Mode.
(was about to post about the differences between PNG and JPG, but Easy beat me to it...)
When I'm doing edits, I use Sans Bold for humans and Segoe Print Bold for yukkuri... I'd say that unless you have serious sizing issues, using a bold font is virtually required...
Also, I don't know what editor you're using, but if you're using GIMP, a few tips:
If you're having trouble getting text to fit into a space, don't be afraid to try putting it in at double size onto its own layer and then shrinking the layer that it's on. The scale layer tool does a very good job with resampling.
Also, if the text is aliasing in an odd way, try turning Hinting off (or on, if it's already off).
Finally, don't forget the line separation... Some of these fonts have big line seperations; I routinely have to use separations -10.0 to -15.0 when I'm working with multiple lines of Segoe Print Bold.
On a related note, how is it that you make the text curved/shaped like the biggest Reimu on the left in post #13577 and what's the feature called? I'm looking to see if there's an equivalent inside Paint.Net...
^^ In usual cases I'd recommend using vector fonts, and keeping them as vector objects/ text objects - then you can scale them with no quality loss. Save a backup for easy editing, and then flatten, mode>Indexed, saveAs PNG
Thanks, EasyModo. I don't think Paint.Net has an equivalent, argh... maybe I'll need to switch between the two programs. I actually don't like working in GIMP that much, the interface seems a bit fiddly.
Actually, I used Filters/Distorts/Curve Bend... Text along path would have the text keep the same height throughout... A better example would be post #13779
^^^ I don't generally save the text separately; what I do is have the text in the uppermost layer, then a layer under that to act as "whiteout", and then the original. Depending on the complexity, I might have another layer for sound effects, or I might put that in the text layer. If I need to change something, it's easy enough to erase the text and replace it...
I'm between EasyModo and Toawa. I have the original at the bottom, a cleaned version on top for making drawing corrections, changing constantly its transparence to check the differences between the original and the edited the text as objects on top of that, as it makes it easier to makes changes, as to fit text intead of just resizing the layer you can change its properties (space between letters or between lines, font size, and so on).
Anyway, welcome, its always great to have new people working at having images edited.