• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

How where when Metal slug sprites made?

Such an interesting thread. Lot's of info. SNK pixel artists are the best on the industry. Too bad the new King of Fighters doesn't have 2D art anymore :(
 
Eh, thats the way people visualize the data because it's convenient, and matches the way it'll be output on screen, but that's not usually the case. The NES uses planar graphics, not chunky graphics, to save space. Where chunky graphics could be literally mapped in a grid, planar graphics must be spread across 2 8-bit bitplanes. This means a 4-color pixel can be represented in 2 bytes rather than 3 bytes.


I understand part of what you are saying but not quite all of it. But that is still interesting. The Genesis uses a chunky format, does it not? But it gets broken up into tiles.
 
I understand part of what you are saying but not quite all of it. But that is still interesting. The Genesis uses a chunky format, does it not? But it gets broken up into tiles.

when we are talking about pixel formats like planar (NES, SNES, SMS, Game Gear) or Packed Pixel (Genesis), we're talking about the way pixel data is arranged. A tile itself is, for the contexts of these systems, 64 pixels, drawn in an 8x8 grid.

planar graphics are stored 8 pixels at a time, across n number of planes, where n is the bit depth. What this means is if you want an arbitrary color depth, say 5-bit color, then 1 pixel is only represented by 5 bits, there are no wasted bits. In other words, planar graphics expand by a power of 2. Packed pixel formats must conform to the size of a byte. To put it another way - you want 32 colors? In Planar format, that's 5 planes, with each pixel being 5 bits. You want 32 colors in packed pixel? You HAVE to store it as 8 bits, with 3 bits wasted per pixel.

The advantage of a packed pixel format is that it lends itself to compression better, which makes sense on a console like the genesis which had a fast processor and could handle graphics decompression during runtime. The most dominant type of compression used in genesis titles, LZSS compression, compresses 50% better with packed pixel formats.
 
tumblr_inline_nv56nfSk5j1ri065t_500.gif
tumblr_inline_o3f5qihyJX1ri065t_500.gif
tumblr_inline_o3gzzquo2d1ri065t_500.gif


tumblr_inline_nplcoqCycL1ri065t_500.gif
tumblr_inline_npldbmEzj01ri065t_500.gif
tumblr_inline_nplcgwjixV1ri065t_500.gif

so awesome.

OP, how do you think traditional 2d animation is made? keyframes yes but everything was still hand-drawn. i imagine they had a tablet a la Wacom for drawing, and used that to trace pencil/pen drawings? that is one way to do it. another is digitize a photograph and trace on top of it. maybe not photoshop back then but definitely that kind of functionality (and maybe even, yes, early photoshop).

the reason it looks so good is everything is hand-drawn. instead of breaking an object into a bunch of different parts and animating those via keyframe it has a fluid feel to it because someone painted each pixel by hand. modern tools encourage a more procedural method. i think hand-painted looks more lovely.
 
In case no one posted this yet, here is a direct link to the official NeoGeo Programmer's Guide from SNK in English containing official technical info including about dev tools used. From hardmvs.com.

See around page 91 - 93. Character creation was done on a PC 9801 and undoubtedly done by hand. Editing tools were nothing as powerful as modern PS or whatever.

Here is a (very old and blurry) photo of an actual SNK employee using the NeoGeo dev system (from http://www.neohomebrew.com/neo-geo-development-hardware.php):

snk-neo-geo-development-kit.jpg
 
Well I'm going to go play Metal Slug now after seeing those gifs. I've beaten all of the games up to XX but Metal Slug X was my favorite. Probably because it was the first one I beat.
I'll never forget when I first saw the Metal Slug in the arcade on a beach. Watching the attract mode, For a few moments I thought it was some freaking cartoon and that there was no way you get to control that stuff in the game.
 
Oooh, I love me a Metal Slug thread! Lots of great behind-the-scenes stuff here I hadn't seen before.

As for the OP's question, I'm sure they had a graphic editing program that allowed them to flip between frames or layers. Also, graphics tablets have been around in some form or another for decades, and Wacom (a Japanese company that specializes in graphics tablets) was founded in '83. I don't know if Nazca Corp. was using their tablets, but I'm sure their artists had something similar.

After that it's just a matter of actually doing the animating. Here's a video of someone animating an explosion somewhat reminiscent of the ones seen in Metal Slug. Note that the process is very similar to the tree example Lord Error posted in the first page.

And just for kicks, here's the elephant slug. A sprite that appears in only one game, for roughly 3 minutes and is never seen again, but which has the most lovingly crafted detailed animation of stretching and wrinkling elephant skin ever seen in a videogame.

Qdp1ZQD.gif


EDIT: Holy shit, I hadn't heard of In the Hunt before this, but after seeing it now, it's clear how Metal Slug 3's under water stages were meant to be an obvious callback to fans of the team's earlier work.
 
Eh, thats the way people visualize the data because it's convenient, and matches the way it'll be output on screen, but that's not usually the case. The NES uses planar graphics, not chunky graphics, to save space. Where chunky graphics could be literally mapped in a grid, planar graphics must be spread across 2 8-bit bitplanes. This means a 4-color pixel can be represented in 2 bytes rather than 3 bytes.

Planar graphics work by treating each bit in a byte as a separate component of a binary number that is computed across n number of planes. so like so:

Code:
bitplane 1: 1 0 0 1 0 0 1 0
bitplane 2: 0 0 1 1 0 0 1 1

to see the resulting sprite data, you read top to bottom, to produce 8 pixels:

Code:
b1: 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0
b2: 0 | 0 | 1 | 1 | 0 | 0 | 1 | 1
Re: 1 | 0 | 2 | 3 | 0 | 0 | 3 | 2

where each number of "Re" maps to 1 of 4 palette colors. This means the two bitplanes must be stored as sequential bytes, represented in hex:

byte 1: 10010010 = decimal 146 = $99
byte 2: 00110011 = decimal 51 = $33

thus, every 8 pixels horizontal is represented as two bytes, and are organized into tiles that are 16 bytes long. Our row of pixels above would be $9933. That would be the data they would be inserting into the binary. If we wanted to make a single tile of the row of pixels above, our sprite data would be $9933 $9933 $9933 $9933 $9933 $9933 $9933 $9933. That's one 8x8 tile.

Disclaimer: I know nothing about this.

But regarding the bolded - do you mean bits instead of bytes (or did you mean 4-color pixel row)?

And couldn't a 4-color pixel be represented as packed pixels with no wasted bits as well - 2 bits per pixel, in your example Re: 01001011 00001110 = $4B0E?
 
Metal Slug Team said:
Q: What are the 'secrets' to making such smooth sprite animations?

A: That's effort, spirit, guts and concentration! Also anger, sadness and obsession become energy fueling our sprite animations.

Q: How do you maintain such a consistent visual style for the METAL SLUG series?

A: That is because our chief designer had a mental image for the visuals and it was conveyed to the other staff correctly, concisely and clearly. Additionally, we had to redraw the artwork if it did not meet a certain quality. Under some circumstances, he redid the artwork himself.

Q: Who comes up with the wacky enemy designs?

A: The designers were told to just make plan rough animations and images. Basic art and animation moves for the enemy characters were planned and designed mostly by the chief designer exclusively. That is one big reason why you can feel consistency through the series. After the basics were done, they were sent to other staff to add actual movement and animation patterns. In this process too, the impressive looking movements were born.

My personal favorite animation is the Rootmars energy ball animation.

tumblr_inline_n93qr6Di5O1ri065t.gif

tumblr_inline_n93qrf5CPs1ri065t.gif

tumblr_inline_n93qibfeVR1ri065t.gif


Also, please bring back the Ptolemaics.

Come_on,_work_for_goodnes_sack.gif
 
Nearest neighbor itself actually describes the interpolation method. It interpolates the final pixel output using a single "nearest neighbor" pixel to a considered pixel. This produces a minimal amount of noise (you get 1 final pixel for every pixel considered) but you still interpolate your data.

What I was saying is that there are numerous ways to rotate, including ways that don't interpolate at all. The problem with doing that is your rotated image produces holes when doing this:

RzyOkPa.png


z1ZgrAe.png


This type of artifact is called Speckling. And it might be acceptable in certain situations. Really, the result of your rotation depends entirely on method and the method depends on the needed speed and accuracy.
Wow, that is really bad rotating method.
Most likely rotating the texels as their own points to new locations and some of them hitting the same pixel leaving holes..

Normal nearest neighbor sampling doesn't introduce holes, it samples from destination to the source and when it hits it selects the color of pixel quad it hits. (Without interpolating between the neighboring color values.)
It's quite bad in terms of quality. (bilinear, etc at least give some illusion of subtexel movement.)

There's a nice trick to improve and preserve quality of sprite when rotating it in paint program.
Use at least twice as large source during rotation, there is some real data between single texels and thus you do not need as much post tweaking to fix things.
 
Very obviously not so. The only thing on Huge Hermit that is layered are the effects such as the water and blast. There are no layers on the other enemy objects (for their idle animations anyway). You're welcome to go through Huge Hermit's VRAM data which I've conveniently uploaded here and here, but having used that myself to piece together those animations, I can tell you it's not.

An example of something that does use layers though is the Ohumein-Conga's attack animation, which has a layered arm.
tumblr_inline_o3f5qk0wwi1ri065t_500.gif


Also the Huge Locust's grabbing attack, which requires certain parts to be drawn overtop.
tumblr_inline_nv5kj8s4xZ1ri065t_540.png


EDIT: Also like a million other people have said, the NeoGeo can't do sprite rotation.
I'm not talking about in the game engine, I'm talking about during the animation stage.
 
It's one thing presumably doing this sprites with a mouse, but a ball-rolling mouse.
And also the shading process must of been insanely annoying to animate.
 
Disclaimer: I know nothing about this.

But regarding the bolded - do you mean bits instead of bytes (or did you mean 4-color pixel row)?

And couldn't a 4-color pixel be represented as packed pixels with no wasted bits as well - 2 bits per pixel, in your example Re: 01001011 00001110 = $4B0E?

It's not possible to store a single pixel in two nybbles by themselves, the smallest allocated unit of data is a byte. So despite a pixel being 2-bit or whatever, it is still stored in a byte.

The Genesis packed pixel format isn't mutable, it always wastes a nybble, it is 2 pixels per byte in format XRGB XRGB, where x is an unused nybble. It is always 3-bit color depth
 
In case no one posted this yet, here is a direct link to the official NeoGeo Programmer's Guide from SNK in English containing official technical info including about dev tools used. From hardmvs.com.

See around page 91 - 93. Character creation was done on a PC 9801 and undoubtedly done by hand. Editing tools were nothing as powerful as modern PS or whatever.

Here is a (very old and blurry) photo of an actual SNK employee using the NeoGeo dev system (from http://www.neohomebrew.com/neo-geo-development-hardware.php):

snk-neo-geo-development-kit.jpg

This looks exactly like something OP is actually asking about, thanks for this info. I've known about the various methods mentioned in this thread but I've often wondered myself how exactly did the Nazca team create the graphics and animations, the tools, methods etc.

The manual you posted interestingly mentions a separate "Art Box (Character development unit)" hardware, as well as an option to attach a scanner.

LNJ3jbs.png


So seeing as how elaborate the art of the Metal Slug game is, I wouldn't be surprised if they didn't at least draw some of the background and character art by hand, scanned it and then spent an ungodly amount of time cleaning and retouching it. The characters and smaller enemies are generally small sprites, but they might've at least drawn the basic outlines/silhouettes for some of the keyframes, and done the majority of the work in classic pixel art fashion. EDIT: As well as using graphics boards of course.

The art software must've been either proprietary or something that was popular on PC-98 systems at the time (if there was such a program, not really knowledgeable on the system unfortunately, maybe someone else might pitch in).
 
I love pixel art and Metal Slug is the reason I started using this name so long ago. It's insane the amount of love poured into these games.
 
Sorry to bump the thread a bit, but I'm doing some research and could use some pointers - anyone got any info/interviews/breakdowns of how games such as Castlevania: Symphony of the Night would have been made, graphics wise?
Aside from general animation and art, I'm really interested in how effects such as these were achieved in game.

The info in this thread on Metal Slug has helped a lot, regardless, so thanks to everyone that's posted so far. I can't make threads, so thought this was the best place to ask!
 
How are the Metal Slug ports on Steam?

I've only tried MS3 and it was around launch, but my general advice would be to buy these games and then play them in MAME (especially easy with the Humble Bundle SNK emulator releases).

Given that 1, 2, 3, and X are available via HumbleBundle and those give you the exposed ROMs, I'd say that's easily the way to go over buying the PC ports.
 
I wonder who was doing the sprites in metal slug attack and defense. They just pump out some original chars in metal slug attack

One of the first games I bought for the Wii was Metal Slug Anthology. Now I feel like a dumbass for getting rid of it.

Tsk tsk for shame
 
Drawn and animated by hand which were how they were all done back in the day. they didn't use some program that magically animated it for them.
 
One common approach was / is to create concept art for your "character", draw all animation frames and use both as reference when creating your sprites.

Concept
SH6HSAG.png


Frames
ciUpcbL.png


Sprites
LDPQw6m.gif
 
Searched and saw that shumplations translated metal slug interviews weren't posted: http://shmuplations.com/metalslug/ unfortunately not a lot,of info on the sprite process there however they do note that concept art and pixel art actually differed for a few designs suggesting the pixels might have not been tracings over an original drawing.

Also Undercover cops interview: http://shmuplations.com/undercovercops/
Irem had amazing talent its a shame the r-type team had to be so anonymous along with nazca.
 
Trying to get more info on this "Art Box":

Nobuyuki Kuroki said:
I used the SNK original tool at that time.
The name of the tool is called Art Box.
The Art Box is the origin of all graphic images of NeoGeo games.
This image is an animation pattern of Rock, B. jenet and Tizoc.
It was arranged by Art Box
I created Tizoc with Mr. SoeSoe and other artist.


https://www.facebook.com/wasabi.tenpura/posts/1082093018510090

EDIT: Found some screenshots of the Artbox UI over here (lots more at link): http://neogeocdworld.info/html/aide/programmation.htm

sS14UU4.png

yfJc04E.jpg

o1uCKEL.png

0HO9Zmp.png
 
Talent and work

nice contribution man, way to go.

I thought that was spot on, they worked hard and did meticulous work. If what I read about Japanese game development from the time is accurate, they probably did it the hard way.

How are the Metal Slug ports on Steam?

As far as I can tell, pixel perfect and butter smooth. I have all the games running under Nebula and Rage, but I play MS3 on Steam now, you have online co-op and some extra features.
 
Here's the video where the Art Box shots came from. Shows the sprite artists working (and showering?):

https://www.youtube.com/watch?v=K1yyAA1wgDE

Here is a Sega of America recruitment video from 1995 (the video mistakenly marks this as 1993, but they show off Saturn games) that shows a graphics artist at Sega STI working on the sprites for Comix Zone: https://youtu.be/-M8RIc6Ek0Q?t=10

They don't go into huge detail, but it seems like the artist is also working with an Art Box like program as there is a video output to a CRT TV.
 
Hey, hi everyone!
Can the artbox be dumped into a rom/program or something?
That would be the way to go!

Most of its functions can be ported. This guy created software to mimic the art box, but it looks like his personal toolset: http://www.culturaneogeo.com/entrevistajeffkurtzeng.htm

WaFKL5G.jpg


As far as public releases go, here's a sprite manager for the neo-geo. Combined with a graphics editor, it should do the job: http://furrtek.free.fr/?p=crea&a=atarashi

Other tools: http://web.archive.org/web/20060515202921/http://www.neobitz.com/Pages/DevTools/NeoMapEditor.aspx
 
One of my fave threads of GAF already. Thanks y'all for sharing such amazing stuff about retro pixel creation process. It's something I've have always admired when playing this type of games.
 
They don't go into huge detail, but it seems like the artist is also working with Art Box as there is a video output to a CRT TV.

Sega had its own line of ( SNK "Art Box" esque ) tools, but the animation software used for Comix Zone is "Deluxe Paint Animation" for MS-DOS ( developed by EA ).

OKqvIn9.png
DAewNK0.png
 
Woah this video is amazing. The shower part is hilarious though.

It's weird how limited Neo Geo hardware was but was able to impress through sheer cart size and the effort the artists put into these games.

It could push a LOT of (really narrow and really tall) sprites, but having to use sprites for backgrounds make for a bunch of trade offs. There's some weird design to the hardware. It made porting the games a lot of work because systems like TG16, Saturn, Playstation etc. had very different sprite capabilities than the Neo Geo.
 
Most of its functions can be ported. This guy created software to mimic the art box, but it looks like his personal toolset: http://www.culturaneogeo.com/entrevistajeffkurtzeng.htm

WaFKL5G.jpg


As far as public releases go, here's a sprite manager for the neo-geo. Combined with a graphics editor, it should do the job: http://furrtek.free.fr/?p=crea&a=atarashi

Other tools: http://web.archive.org/web/20060515202921/http://www.neobitz.com/Pages/DevTools/NeoMapEditor.aspx

Amazing, thanks :)
 
It could push a LOT of (really narrow and really tall) sprites, but having to use sprites for backgrounds make for a bunch of trade offs. There's some weird design to the hardware. It made porting the games a lot of work because systems like TG16, Saturn, Playstation etc. had very different sprite capabilities than the Neo Geo.

One thing that always impressed me was the raster effects in games like Sam Sho 3 where it altered the perspective and made it look 3D. It's not perfect but it still looks cool to me. Same with the combination of zooming and perspective in the Super Sidekicks games. I could go without zooming in most games but it was used really good in those soccer games.

If anyone's curious about the effect in Sam Sho 3 you can see it in this video:

https://www.youtube.com/watch?v=w0jxj5F_-iA
 
They certainly didn't re-draw all frames by hand from scratch. Rotation, scaling and other kinds of warping of movable parts were clearly used during the production (and then tweaked/corrected by hand) and it's more obvious in MS3 and onwards.
 
Here's the video where the Art Box shots came from. Shows the sprite artists working (and showering?):

https://www.youtube.com/watch?v=K1yyAA1wgDE

https://www.youtube.com/watch?v=7BR1rE-CqMg

This promotional video for Fatal Fury 3 shows a little bit of development footage. You can see them rejecting an animation frame for Bob and tweaking alignment for one of Terry's animations. The development montage begins at 2:48

*Incidentally if anyone knows where to find the BGM they used in that development montage, I'd appreciate it a lot! I've been looking for that forever.
 
Top Bottom