Typesetting review: Kore wa Zombie Desu ka – Of the Dead (updated) This entry was posted by herkz.
Kore wa Zombie Desu ka – Of the Dead typesetting review
Groups: Hadena, Commie, Doki-AFFTW, Hatsuyuki
Hadena
No blur is no fun.
Sideways signs are always fun. Again with the lack of blur.
Border color is wrong. Good besides that.
Font choice is a little off and it could use more blur.
Needs blur, font is too thick.
Font choice is bad, border is too thick.
Typesetting overlaps the sign. The color is wrong as well.
No blur.
Not sure what they had in mind, but end of the typesetting is fine except for some reason it has a border. Also, hard not to laugh when they get the border color wrong on their name.
Oh Hadena.
Wrong border color again.
Looks like they forgot to blur the border.
Commie
(tl note: not me)
This is what they did for all the notebook signs.
And this is the dialogue style just so people don’t get confused. Seems like a pretty good idea since most of them move around and don’t have any room to place the sign.
Yeah, this is probably as good as it gets. The effect on the original sign is replicated quite well.
All good here.
Also good.
Yeah, this looks great as well. Especially the clipping around her head at the beginning looks much nicer than Hadena’s.
Not sure why the font choice is not similar to the original but since you can’t see the original it’s not a big deal.
Looks just fine.
This is basically perfect.
Doki-AFFTW
Fades in, changes color, and gets closer together. Pretty well done.
The heart is a nice touch, but the border color is wrong.
I like the movement and that it got blurrier, but the font choice is bad.
This is good.
This is good. Only downside is the movement was done by hand so it jumps around a tiny bit.
I see what you did there.
Font choice could be better but the real problem is that this lags terribly. Way too many moving signs on screen at once.
Border color is a bit off.
Font should be lighter and border should be darker. Fine otherwise.
Notes: For the Eu signs they just put them at the top with the dialogue style.
Hatsuyuki
That’s one way to do it.
Font sucks, movement is off, color is wrong, blur looks bad.
I have no idea what they masked part of the sign and then picked a really bad font and used no blur.
Wrong color, bad font choice, no blur.
See above.
Should be 2 layers so the font can have blur too. Also, the fade ends way too early.
Yes, they really did this. About 7 more times in fact.
This is hardsubbed, although for good reason. It looks nice except the effect goes on for far too long and the font choice sucks.
Grades
Hadena: C+. Some of these are lolworthy and there’s a real lack of blur, but this show has a lot of signs that matter and they actually did them which is nice I guess.
Commie: A-. Every sign except one is really, really good. The episode title sign especially is great. Also, almost every sign has 2 layers which really improves them a lot.
Doki-AFFTW:B+. Most of these were fine. Color was off on a few and the one sign lagged really hard. Also, this release is not sorted by time.
Hatsuyuki C. I’m tempted to give this an H because lolhardsubbing but yeah they actually hardsubbed for good reason. The rest of the signs they did are just there, however. Also, this is not sorted by time.
you sure these reviews are objective herkz?
you’re welcome to point out any errors you think i made
> Talks about layering, yet the dialogue isn’t even layered correctly before sorting.
> Moves drift across the length of the scrolling text.
> Making perfectly hard to read japanese signs easy to read by taking the easy way out.
> Water sign covered for no good reason. Font used isn’t even close.
> Lots of signs missed.
> Next episode title, too much blur and sorta done wrong.
You would get bonus points for the novel approach at doing the episode title, but then lose them for the lack of effort elsewhere.
At best this is still no better than a B- – C+ grade, at least Hadena (while not doing a very good job) tried, it’s a shame I can’t say the same thing for you guys. I would complain about lazy clips, etc, but w/e I don’t really care.
Also herkz, how did you manage to fail rendering that Sage Lupin III tower sign that you claimed was “what”. I tried 3 different vsfilters, and every single one rendered it correctly. Sounds like a case of your own failure, look into it.
> Talks about layering, yet the dialogue isn’t even layered correctly before sorting.
I don’t see how that’s relevant. If you want, I could mention how like 2 or 3 episodes I’ve reviewed so far have sorted by time, but I won’t.
> Moves drift across the length of the scrolling text.
Yeah, that’s kind of hard to avoid without Mocha.
> Making perfectly hard to read japanese signs easy to read by taking the easy way out.
What is this in reference to?
> Water sign covered for no good reason. Font used isn’t even close.
You’re right, one bad sign. My bad.
> Lots of signs missed.
???
> Next episode title, too much blur and sorta done wrong.
It maybe has a little bit too much blur on the bottom layer, but it’s pretty close.
>You would get bonus points for the novel approach at doing the episode title, but then lose them for the lack of effort elsewhere.
I hope you don’t think I typeset this, because I didn’t.
>At best this is still no better than a B- – C+ grade, at least Hadena (while not doing a very good job) tried, it’s a shame I can’t say the same thing for you guys.
Which signs look lazy, exactly?
>I would complain about lazy clips, etc, but w/e I don’t really care.
Pretty sure there’s only one and I don’t see anything wrong with it.
>Also herkz, how did you manage to fail rendering that Sage Lupin III tower sign that you claimed was “what”.
Good question. Though, even if it was aligned correctly, it still doesn’t look very good.
As far as I could see Hadena only did one sign that we didn’t, the truck (which isn’t exactly vital to the plot)
herkz didn’t typeset this show, by the way.
Dear xythar,
Why would it matter whether herkz typeset this show or not?
The only time I mentioned herkz directly was regarding the failed render of a sign in the sage lupin review, that someone brought to my attention. As far as I am concerned, this was done by someone in Commie. Doesn’t really matter who now does it? After all, a typesetting review isn’t taking into account experience, etc and rating people down harder that should know better. It is assuming all typesetters are equal, though I believe anybody using AAE should be criticized more heavily because the software is much more flexible.
Dear herkz,
Actually I think mentioning time sorting, and proper dialogue layering is very important (at least for anything done with ass), because they reflect on the group/persons knowledge of bugs, etc. Especially since you mention blurring of text a lot, and incorrect dialogue layering screws up blur sometimes on playback. Perhaps if you point out time sorting at the very least, it might convince groups to start bothering to do it. Telling them directly doesn’t seem to help.
Your friend,
ar.
Sure, I don’t have any problem mentioning that, but your comment came off as extremely hostile instead of constructive.
The way you said “you” made it sound like you were talking to him directly, with the slight implication that the review was a sham because he was just plugging his own typesetting. Just wanted to make sure I clarified that it wasn’t the case.
I don’t seem that hostile do I? :O
Another thing would maybe be to take into account time constraints, and while that doesn’t excuse pure laziness, you’ll rarely find any typesetting done in After Effects done for speedsubs because it slows releases down, and you probably won’t find much WOWOMG typesetting in speedsubs either, unless it’s stuff that can be done within a reasonable timeframe.
It also might be a good idea to review the second ep of speedsubs instead of the first (or perhaps re-review on the second ep), simply because a lot of the time repeat stuff gets fixed between the first and second episode because it’s not ready in time. There is no really no excuse here for slowsubs though, of course, and obviously if nothing that carries over from ep to ep has changed, then there is no point for re-grading.
Typesetting is very hard to review, and you really need to take a lot into consideration, which is why nobody has really bothered reviewing it in the past. That, and that typesetting can be subjective, what I suppose it comes down to is whether the english is fitting, and done appropriately. I’ve come across many cases where this plays true.
You might want to consider things like blur for zooming/large signs too, since there are times where lag might come into play as a factor as to why it is not present, or has been minimized, same for stroke width (border) and sometimes shadow too.
If in doubt, you can always approach the typesetter for confirmation, or the group. While this sounds like a lot of effort I’m sure, imho it’s one of the perils of soft-subbing, and as such should always be taken into consideration because a sign that lags 50% of the viewers is far worse than a sign that’s been lightened to avoid such consequences.
Of course I will take into account time constraints and such. And yes, I plan on reviewing the second episode for most shows.