Xanadu II Translation Development Blog

[Black Tiger]

If that's what you really want to do, that's cool. But I can totally do an alternate version with a normal sized X and push the words closer together for LoX1. I don't think it will look as faithful or quite as nice, but it could even look better for all I know. As I've said, I'd prefer to keep things as authentic as possible, but whatever you gotta do within what you're up for doing.

[elmer]

If that's what you really want to do, that's cool.

Am I amazingly happy to have the prospect of all the extra work dumped onto me unexpectedly?

Nope. And last night I'd have said "no way can this be done".

But it looks like it's going to be possible in Xanadu 2, and it may be possible in Xanadu 1 (IMHO it needs to be the same logo in both).

At that point it is a cost/benefit calculation ... and the answer is that after a night to sleep on it, I can really see the benefit that allowing a 64-pixel height could bring.

Now, there are definitely some things that jar about the current iteration of your bigger logo, particularly the blank space between the 'a' and the 'n', and between the 'd' and the 'u'.

Those serifs are just too darned big, causing the letter spacing to result in those eye-distracting blank spaces.

There's also the question of how much space is avaiable inside the stems ... is there going to be enough room to put in the internal shadow that's in the original Japanese logo ... and then the next question, do we care about that?

Phase has done a wonderful job on his 32-pixel height logo, and the anti-aliasing work that he's done is excellent.

But it does begin to remind me of why I don't always love anti-aliasing at low resolutions on modern display devices.

It looks brilliant as a small image on my monitor while reading this forum, and I'm sure that it would look absolutely fantastic on a TV with the natural softness and blurring that that would give it ... but the when it comes to displaying it at a "playable" size on a modern high resolution display, all that anti-aliasing is too visible IMHO, and I need to squint to make it really look good.

But that's just my current reaction ... would I just get used to it over time? 

There's also a fairness issue in that Phase has been doing some great work on his logo that basically takes the same source inspiration ... it only seems fair to see if he wants to try doing a 64-pixel logo, too.

Either way ... I need to investigate whether I can get a 240 pixel wide logo working in Xanadu 1 (I think that the answer was no the last time this came up).

[SamIAm]

If you use that really slow super-compression, will it possibly stall the progression of the intro sequence to the point that transitions start taking a long time and/or it no longer finishes with the music?

In the prologue scenes that are done within the normal game-engine, each screen transition takes 5 or so frames longer in the translation than the original because of elmer's new compression. A few frames each time isn't a big deal, especially when there are only a few transitions in each scene, but if there are more transitions and more frames of delay added each time in that intro sequence, we're going to have us a syncing problem.

If we can get 64 pixels in height in both games, then I'd give a tentative thumbs up to the box-art cursive style. However, I wouldn't mind seeing a quick mock-up more like the original e.g. Black Tiger's first logo for comparison.

[Black Tiger]

Here is a 240 x 32 pixel version.










Even as I was saving these I noticed that I'd like to move the "L" a pixel closer to "egend".

Still lots of leeway to get it narrower for the first game.

[johnnykonami]

This last version of the title screen looks really professional, good job guys.

[SamIAm]

That is some skilled work on BT's part. I know I couldn't do that.

I'd like to draw people's attention to something, though:



You might look at that and think wow, the English looks even better than the Japanese! And that would be understandable, because the English here is very exciting to look at.

However, keep in mind that the original Japanese graphic, in relation to the art in the rest of the game, is not dull. Tonally, it's very well-matched, with just a little touch of its own style like a good logo should have.

At the very least, I for one would like to see BT's work de-glitzified a bit, e.g. using milder coloring and maybe making letters like the X a bit tighter, along with slimming the background II at least a little.

However, in 240x32 especially, where it's all forced to get kind of fat, and even in 240x64, I'm afraid that that box-art-style cursive might be a little over-the-top. I felt this about Phase's rendition of it, too. It's really abandoning the subtle sensibility of the original logo entirely. If we use it, then in Xanadu 2's case especially, it will be by far the most ostentatious graphic in the game.

Xanadu 2 is not a wild party or a swashbuckling adventure, nor is it a particularly glamorous affair. It's a story of the somewhat mild-mannered characters from the first game wandering almost helplessly through an unknown continent that is full of ruin and dark, beautiful nature. Even the towns in the earlier parts of the game that are clean and pretty are also very subdued.

My work is not the height of perfection, but I do think it's a lot more in line with the tone of the game itself. I'd love for someone who can draw to do what it does even better.

[Phase]

I see what SamIAm is saying, and thats why I like his version as well. It has a simplicity to it.
Definitely something to keep in mind.

In a way I like BTs smaller logo better than the bigger one, maybe the big one is too big?, so is something in between possibly the answer like 48height? Not to make you guys do more work just theoretically

When SamIAms mention anti aliasing a few post back I was going to post that I was thinking the same thing that elmer was talking about later, the AA may deviate too far from the pixel shading look even though it has its benefits. About the extra colors on my last pic even though i added 4 colors, I didn't use some of the original colors so perhaps just picking some nicer shades is the key. Anyways went back to a pixel shading look.

I think this may be my final entry for 32 height as I'm pretty happy with it, though Im always down for changing stuff based on others input. The 2 was really bothering me and after I saw BTs version I thought I'd go back and take a look at the artwork again
see that little 2 thats the new plan 

Also reading what SamIAms was saying on his gradient and lighting he made a good observation, and his is looking super nice ..so I borrowed his Idea there O:). My outline is heavier than BTs so I defo need to have a nice color gradient. also added a drop shadow back in (the first one I posted had one) but also because SamIAms looks nice :-" it can always be knocked out if its too much. Well, Im having fun and learning some techniques so don't quote me on this being my final entry. 


Edit: updated had a few odd pixels and had to lose the shadow because it was looking 3d
 

[elmer]

If you use that really slow super-compression, will it possibly stall the progression of the intro sequence to the point that transitions start taking a long time and/or it no longer finishes with the music?

The "super-slow" part is only on the compression side, it has no effect on the decompression speed (except to make it a tiny, tiny, tiny bit faster).

It's not something that you have to worry about ... except that that it makes previewing translation changes grindingly-slow when enabled, i.e. 10 to 20 times as long to do an insertion pass ... and that's just doing one file.

I really need to look at what LZ4 for is doing for its "optimal" processing and steal it!

You might look at that and think wow, the English looks even better than the Japanese! And that would be understandable, because the English here is very exciting to look at.

However, keep in mind that the original Japanese graphic, in relation to the art in the rest of the game, is not dull. Tonally, it's very well-matched, with just a little touch of its own style like a good logo should have.

Well, if we follow that train of thought, then I think that we get back to Black Tiger's original logo idea.

IMHO, that matches the feel of the original Japanese logo, but with just a touch of it's own style.



BTW ... I find this logo easier to "read" than any of the current cursive logos (except SamIAm's) because there's a clear distinction between the letters (<EDIT> when viewed at game-display resolution, but less so at web-browser resolution).

The biggest problem that I have with SamIAm's logo in it's current form is that it has too little style of its own.

The original Japanese logo seems (to my eye) to have a "feel" that's lacking from the simple text in Sam's logo.

[Phase]

After rereading SamIAms post, I get what hes saying even more, so didn't mean to brush it aside, its just I was posting my latest. Its something worth discussing.
edit: doh elmer beat me to the next post lol

[Phase]

For the sake of argument, many RPGs aren't super happy or humorous and still have heavily stylized logos and title screens?

 Not to be too biased :-" but imo I like both the cursive logos so far, and think they look a little more RPG like than the original and SamIAms.  The original still throws me off with the smaller words and also the font doesnt seem very rpg like and when you add the italic to it, it kinda makes me think of something speedy imo. The original english artwork doesn't really seem too crazy to me and looks like an RPG.

[SamIAm]

Thanks a million, Phase! If we keep experimenting in 240x64, I hope you'll stick around and throw in some ideas. You do good work. 

EDIT: I just saw your next post. I'd be happy to talk more about why I hesitate to embrace cursive soon, but I need to start doing real-life work.

The "super-slow" part is only on the compression side, it has no effect on the decompression speed (except to make it a tiny, tiny, tiny bit faster).

It's not something that you have to worry about ... except that that it makes previewing translation changes grindingly-slow when enabled, i.e. 10 to 20 times as long to do an insertion pass ... and that's just doing one file.

I really need to look at what LZ4 for is doing for its "optimal" processing and steal it!

Well, that's good. I mean, it's good and bad, but for our purposes, it's one less barrier to making things work.

Well, if we follow that train of thought, then I think that we get back to Black Tiger's original logo idea.

IMHO, that matches the feel of the original Japanese logo, but with just a touch of it's own style.

BTW ... I find this logo easier to "read" than any of the current cursive logos (except SamIAm's) because there's a clear distinction between the letters.

The biggest problem that I have with SamIAm's logo in it's current form is that it has too little style of its own.

The original Japanese logo seems (to my eye) to have a "feel" that's lacking from the simple text in Sam's logo.

I totally agree. One could criticize my logo in a word as "forgettable" and without better art skills, I probably won't be able to fix that. If we are in 240x32, however, I honestly think it's the best compromise of the logos currently existing; it's easy to read and stylistically appropriate.

What we really need to know is, are we going up to 64 pixels in height? Because if we are, the possibilities go into the stratosphere. That would open up a whole new round of experimenting.

I just tried blowing up the font I was using and putting the title in two lines. Without touch-ups, it's not nice to look at, but it shows a lot of promise. With a different font, we could get that slightly-chiseled look going in lowercase, plus we could do the shading exactly like the Japanese logo does it (plus external drop-shadow if we like). 

[SamIAm]

Re: Why I'm not crazy about the cover-art cursive for Xanadu 2's title screen.

Let's start with a question: if they hadn't seen the cover art, would Phase or BT ever have thought to make the X like that?

I rather doubt it.

If the only reason to go with that whole style is because Falcom has used it elsewhere, I think that that's probably not enough.

On the cover itself, the cursive is thinner, the color is much lighter, the whole thing is more contained as one unit, and even though it is prominent, there are other strong and prominent visual elements to go with it. On these title screens, however, this writing gets dark and splayed out, with very noticeable outlines, and it's basically the only major thing the player looks at. At the very least, I think we have to acknowledge that it's working in a different way once we bring it over, and I'd say that anytime you do that, there's a very good chance you're just forcing it.

One of the reasons why I liked Phase's straight chiseled logo is because it emerged organically: Phase looked at the background, assessed what would fit, and proceeded from there. Something tells me it just doesn't quite strike the right chord for these games, but I feel almost sad to let it go. I still like that one more than the cursives, actually.

There's no accounting for taste, they say. My taste, for what it's worth, tells me that the cursive is simply too gaudy. Make no mistake, it also tells me that mine is nothing to get excited about, either. However, think about it like this: if you're going to a formal event, and you've only got two suits - one that's appropriate but a little dull and one that you're pretty sure is too flashy, wouldn't you pick the former? If you were choosing between two items at a buffet, and one was too salty and the other a little bland, wouldn't you go with the blander one?

I'd love to have one that hits the bullseye, but we just don't right now.

One more thing along the same lines: I'd rather undersell Xanadu 2 with the title screen than oversell it. I'd rather that players be pleasantly surprised that the game itself offers richer art than the title screen would suggest than have them be disappointed that it didn't deliver as much dazzle.

With Xanadu 1, it doesn't matter as much. That game is not only a lot brighter and peppier, but the logo only appears for a few seconds during the cutscene between the third and fourth chapter.

By all means, if you disagree with me let know how and why. 

[Phase]

Sorry if my last post was overly argumentative this is yours and Elmers project.
I was also curious though because I would have probably went with the cursive style on a 64height concept. (at least with the actual word Xanadu)

Good points and thanks for going into more detail. I don't entirely agree, but I see what your saying.
Did you guys want to pursue some 64 height concepts now?

[elmer]

Did you guys want to pursue some 64 height concepts now?

Can you just give me a little longer?

I just need to finish something off before looking at the Xanadu 1 sprite situation to see if 240-wide is achievable.

[SamIAm]

Sorry if my last post was overly argumentative this is yours and Elmers project.
I was also curious though because I would have probably went with the cursive style on a 64height concept. (at least with the actual word Xanadu)

Good points and thanks for going into more detail. I don't entirely agree, but I see what your saying.
Did you guys want to pursue some 64 height concepts now?

No no! I don't think you're being argumentative at all! On the contrary, I'd love to hear more of your perspective.

I'm happy to have this conversation with you.

Black Tiger's 64-high cursive is probably the prettiest thing we've got going. There's something to be said for that, and at that proportion, I think works well enough that it might be worth going for. Of course, with such a drastic spec change, I think we really should see another round of mock-ups before we decide. Even an additional 8 pixels in height would make the possibilities explode.

BT's 32-high cursive, as I feared it would, looks a bit cramped and fat. It's lacking the grace of the 64-high version.

By the way, I was going to ask, could you give me some examples of some of the 16-bit RPGs you were thinking of when you mentioned other games with exaggerated styles? I'd honestly love to know.

-------------

By the way, I just now had the unfair privilege of being able to try mine out on a real system and CRT in RGB.

You all might be interested to know that on my setup at least, the lightest blue outline color is less of a blue and more of an off-white, especially against the dark-green background before it fades to black. Also, the two medium blues are so close to each other that it's hard to distinguish them, although that might just be due to where the boundary is on my graphic in particular.

My first reaction to my logo was that I should probably raise the ceiling of the second band and reduce the first. What I might try is putting the brightest color on the upper right corners of all the short letters only, allowing a little more for the taller letters, and that's it.

I also think I can get away with using a fourth band on the tails of the f and g, and that it will look better. Right now, they look very blue.

In case you're wondering what I thought as a whole, here's how I'd sum it up: It doesn't really work as a logo at all. However, as a stylized title, like what you might see on the cover of a novel, it works very well.

A proper logo is admittedly preferable, though. If it's a choice between this and a moderately dysfunctional logo like elmer's mod of BT's old version, well, that's going to be tough.