yeah i know, I was just having ideas

You must have been reading my mind.

I was just thinking about this project and voila... an update!


Keep up the great work guys! :thumbup:

im fairly sure that transparency is not hard, i can do it fine in 3 level B,G,W xlib basic. i think that your method may actually make this harder from a coding perspective. what your talking about is masking its not hard, all you gota do is and a mask a negative silouet of the sprite, so like make a silouet and reverse black with white, then whenever your gona output the sprite, 'and the mask and 'or' the sprite. the 'anding' keeps all the detail of the scenery and clears the area of the sprite so it can be 'or'd. this is how it works in basic i would assume itd be the sameish in asm but then again i have no clue what an interupt is or anything. either way, its been done in every good asm game there is, its not a big deal and shouldnt require narrowing of the pallet

This post has been edited by dinhotheone on 10 Dec, 2007, 19:59

So, which would be more efficient? I can create a mask, if you want to go that route. I just want to make sure it's not going to require more memory, or slow the game down any.  

Yeah... As triplea just pointed out... using masked sprites will increase the memory required by the sprite by 50%.

That is one reason I opted to not use masked sprites in UV. The other 2 reasons are: the original does not use masked sprites, and it would not have a profound effect on the visual experience of the game (UV moves a full tile at a time).

If your character sprites move a pixel at a time rather than by a full tile at a time, then masked sprites for them might be a good idea. Else, you would have to redraw the entire screen everytime you move.

This post has been edited by Ranman on 11 Dec, 2007, 7:11

yes/sorta, its not that big of a deal but yes, like how monochrome is more efficient and less memory consuming that gscale, however gscale and masking make things look nice. but even this is not a good analogy because masking will not out any noticeable tax on the system. i think the alternative to masking is that the guy just sits on a white tile, which probably wont look to good. i have not ever heard of transparency masking in calc games. also, you only have one thing that this needs to apply to, some games mask 10 things simultaneously.

at its most basic level, masking is just puting out 2 sprites instead of one. the first prepares the tile so that it looks normal, you dont want to be able to see the scenery behind the sprite right? and the second just puts the sprite out there on the prepared location.

as to your question about slowing the game down, it wont be noticeable. if you read what i wrote before, its not anything that requires an artists touch to create a mask, i would leave this part up to the coders. if you still dont know how calculator masking works, just post and ill try to go more indepth or w/e you need, my first post goes fairly technically into it, and if you dont know what 'and'ing or 'or'ing a sprite onto someplace i can explain that too.

This post has been edited by dinhotheone on 11 Dec, 2007, 16:14

It is difficult be cause that isn't how grayscale displaying works. the second method would still need a masked sprite but it would be generated when the sprite is displayed instead of already being there that is the only difference.  

using a color as transparent is generally a console/handheld only thing because they have graphics hardware displays the color as transparent for them.

since few people use gs, there's no real standard.

you have to keep in mind that grayscale data is stored kinda like this:

.db 00001111
.db 00110011

That right there controls 8 pixels.  so basically, if the bit is 00 in both bytes, it's white.  if the bit is 11 in both bytes, it's black.  then either or gives the two values of gray.  it's kinda difficult to specify one color unless you look at both bytes, which most routines don't.  they kinda process one at a time.  i think what some people might do is generate a mask from one of the other or both layers.  or they might use one of the inverted layers as the mask itself.

i personally like using separate masks.