Mobile Game Image Type Question

brett-norton

Ok so i know all the basics to file types and optimisation etc.

But im using adobe illustrator for my game art which is going to be used as a mobile game 2d platformer.

There are many options to consider when saving to web in adobe illustrator i just want to make sure i get this right it would be a huge backtrack if i got it wrong. and i cant seem to find detailed answers on this.

Here is to my understanding and things to consider.

So far i have built my vector character at 1024 x 1024 i will now resize that to fit into a 512x512 while considering the size and gap surrounding the character when in a 512x512 box/image to allow for plenty movement when animating and making sure he doesn't extend past 512 pixels.

when i have my size i will break this up into separate pieces ready to animate in my animation software "Spriter"

But before that i need to export the files, and i read its best to always export as a save for web???? I assume.

Now when exporting for save for web i have many options. I will be using Png-24 as i have semi transparent parts to my character.

Next option.. is where i am stuck.

Should i use

Selective, Perceptual, Adaptive, Restrictive???

Should i use Diffusion, Pattern, Noise, No Dither???

Should i tick transparency check box???

Thanks

Socks

@brett-norton said:
But before that i need to export the files, and i read its best to always export as a save for web???? I assume.

Any export/save/conversation method that lets you produce a 24/32bit PNG will work.

@brett-norton said:
Now when exporting for save for web i have many options. I will be using Png-24 as i have semi transparent parts to my character.

Next option.. is where i am stuck.

Should i use

Selective, Perceptual, Adaptive, Restrictive???

Should i use Diffusion, Pattern, Noise, No Dither???

I've never saved (rasterized) directly from Illustrator, but what you're saying kinda' doesn't make sense, if you're saving as a 24bit PNG then why would you need to create a LUT, all the colours are already available in the 24bit image !?

So, I would say don't use Selective, Perceptual, Adaptive or Restrictive at all ! Not sure why it would even ask you to choose with a 24bit image ?

Same deal with Diffusion, Pattern, Noise, No Dither, unless you specifically want to reduce the colour table and have your image dithered because you like the look (I've always liked 'Noise' dithering) you should ignore this choice (again it's weird that you are being asked to choose one).

But, if I misunderstand your question, and you specifically want to reduce the colour depth to 8bit, then all these choices (Selective, Perceptual, Adaptive or Restrictive / Diffusion, Pattern, Noise, No Dither) are largely aesthetic, if you like the look of Perceptual with a Pattern dither then go for it, if you think Adaptive with a Diffusion dither better preserves your image then go for that one . . .

The only thing to bear in mind is that the file size will change deepening on your selection (and your source image), for example dithered images will never compress as well as undithered images, so expect dithered file sizes to be significantly larger.

And . . . (just re-read your post) . . . should you even be importing reduced colour palette / dithered images into Spriter instead of the full quality PNGs ?

I hope some of that makes sense !

@brett-norton said:
Should i tick transparency check box???

Like everything else, that's your choice, do you want transparency ?

brett-norton

Sorted thanks socks, wasnt looking to achieve anything complex just wanted to make sure i was exporting correctly.