Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
Battle_Korbi

What camera did the guys at id use for making sprites?

Recommended Posts

Gree(n)t(h)ings and salutations,

 

Since I have suddenly acquired a lot of free time, I have decided to try out a new, or should I say, rather ancient way of making sprites for Doom. How do I intend to do this? By making props and taking pictures of them!
 

Yeeess, yeeess, you don't have to lecture me on it. I know I can draw them or render a model of them in Unity perhaps, and then make sprites out of them, but I don't feel like doing what everyone else is doing. But Korbi, if use camera, u r noob and u hav no qualitet.DOES IT LOOK LIKE I CARE, HUH?! DO I!? DO I!? Suddenly, I started to love making props again, so I told myself "why not?" (rhetorical question)

 

Either way, I would like to know what kind of a camera did the guys at id use to make their glamourous sprites. I do not intend to OBTAIN one, oh god no. I would like to know the tech details of them and how could I adapt my smartphone to it. Heck, even if that proves untrue, I could still try and obtain the quality scale and adapt to that.

Share this post


Link to post
On 7/21/2017 at 7:26 PM, Linguica said:

They used a Magnavox Easycam.

 

Mass Effect Andromeda cost $40m and used the latest state-of-the-art technology and still had inferior motion capture to the original Doom.

 

The Easycam was a VHS-C camcorder, so presumably all the sprites were recorded at the NTSC resolution of 333×480 pixels with a TV capture card.

Share this post


Link to post
2 hours ago, Ashley_Pomeroy said:

The Easycam was a VHS-C camcorder, so presumably all the sprites were recorded at the NTSC resolution of 333×480 pixels with a TV capture card.

Probably. Some of the raw captures can be seen in the collection of art Romero dumped a couple of years ago (examples: archvile, revenant, pistol, etc.). Note that it wasn't a TV capture card that was used to get it into the computer - they weren't even using PCs. The NeXT workstations they used had built-in DSP chips that natively supported this kind of thing (shameless plug for my Artwork of Doom article which talks about this a bit more).

 

However, it's probably important to recognize that "which camera they used here" was a very small (mostly irrelevant) part of the process and the work of creating sprites:

  • A large part of the heavy lifting probably came from the Fuzzy Pumper Palette Shop software they wrote. It seems they basically wrote their own video capture software tailored to their needs; in particular the name suggests that the software helped with the palette conversion, which is one of the most difficult parts of working with 256 color images.

    How the palette shop software worked isn't entirely clear, but I find this image in the Romero dump to be particularly interesting as it appears to show the same image imported using various different conversion settings. Palette conversion is inevitably a trade-off of what you want to sacrifice or retain from the original higher-color image: hue / saturation / brightness / possibly others. It's something that probably needs human guidance to choose the right setting for the particular image being imported. Although most image editing software can do palette conversions, I'm not sure I've seen any publicly-available software specifically tailored for doing it really well like this.
     
  • It's very, very clear that image conversion is only the very first step in the process. If you look at those captures I linked above, they're very rough pieces of work - the colors in particular are a washed-out mess. Certainly it seems that some things needed more work than others - this for example appears to be mostly unchanged apart from some colorization and contrast work.

    In other cases it appears that the captures were used for little more than rough outlines, and the actual sprite was essentially redrawn from scratch on top of the original capture. The weapons are one example; the demons are the most extreme one, where photos of a dinosaur toy were drawn on top of to produce something entirely new with little to no resemblance to the original. The other example I'd point to is the Doom title screen - clearly derived from the Punchatz box art, but at the same time it looks like the whole thing has been redrawn almost entirely from scratch. 

    It's worth remembering these were experienced pixel artists who by the time of Doom's development had years of experience making video game sprites (with Wolfenstein 3D, and all the earlier games they'd made together). The video capture no doubt helped them save a lot of time and provided something to work with (easier to edit something than draw from scratch), but these guys could more than hold their own. When you see the game art, you're really not just seeing some lightly touched up video captures - it's a lot more than that.

 

Share this post


Link to post
3 minutes ago, fraggle said:

*cut down to save sphess*

Good points there. I had something in mind for this, and I know video capturing doesn't mean that work is already done, as these props I am building are definitely going to be touched upon further (the whiteness of paper and tape has to be dealt with, plus other stuff as well). But this whole thing fascinates me, dunno why, and so I would like to know the tech aspects behind it, and if I could replicate the effects of it, I guess I would be happy with myself. My idea is to snap some pics with my smartphone, and downgrade and scale down my pictures accordingly, and once the whole readjusting is done, start applying filters and whatnot, and then cut out stuff accordingly and make sprites as usual.

 

Is there any modern software that could be used to "scale down" pictures like that? What kind of "ratio" of stuff I should consider when scaling down pictures (like, lets assume a doom sprite is 128x128 or smtin, which modern resolution should I use and scale down etc etc etc.)? I am basically putting up a sidequest in helping me with providing all the useful tech info :D

 

Share this post


Link to post
16 minutes ago, Battle_Kirby said:

My idea is to snap some pics with my smartphone, and downgrade and scale down my pictures accordingly

I can imagine people laughing at the idea of making sprites from smartphone photos but honestly, I think it's probably good enough. The capture quality that the id guys were working with is clearly several orders of magnitude worse than what you can snap with a modern smartphone, so go for it. If you're making sprites for Doom I don't think you need any special equipment, and even if it's a newer engine with higher resolution it's probably good enough.

 

I feel like I'm rather out of my depth offering advice for this since I'm not really an artist, but I do have a bit of experience. 

  • First off, make sure you get the aspect ratio right. If you're making sprites for Doom, pixels are non-square, when you scale the image down you need to scale down vertically to 83% of original height, otherwise your monsters or whatever will look disproportionately thin.
     
  • Don't use white as a background. Ideally use some bright color so you can do chroma-key. A lot of work can be involved in "cutting out" the thing you're trying to capture from its background. Long ago I did a lot of work in cleaning up some of the original Freedoom sprites, which were drawn on paper. The original artist "cut out" the background in an imperfect way: essentially every single sprite had ugly white speckles around its edges. Every single sprite had to be edited by hand to remove these, one by one. It took a long time; the big downside of sprites is that any change like this takes a lot of work because there are so many sprites to change.
     
  • Put work into fixing the contrast. This was another problem the early Freedoom sprites had - they were drawn on paper with pencil, so all color was skewed towards the white end of the spectrum. You can see that a lot of the id video captures are similar because of distortion from the video format, and obviously needed work (compare untouched vs. retouched). Recoloring in particular seems like an important skill, something far beyond what I'm capable of, and especially important if you're working with a palettized format.

In case you're interested, I experimented myself with doing something similar a few years ago. This is all based on camera-phone photographs of a Doom Reaper imp miniature: original shot (note orange background), cut out from background, edited to remove feet, and example sprite. I got the aspect ratio wrong when making the sprite, and the coloring is a mess. I ended up giving up because I realised I didn't have really artistic skills to develop it further into something that actually looked nice.

Share this post


Link to post

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×