DF-21 Forums

[Tarsus]

I'm still new to this process: I've studied the BMPDF tool and am attempting to create a WAX. I've set up the makefile properly, and upon executing 2WAX, I get the error "Error opening file TEST001.BMP line: 78" (Where line 78 is, indeed, the first line of my cell declarations.)

So what don't I know about how to create a BMP for this process? All my BMPs are 40x68 pixels, have completely black backgrounds, and don't have pure black anywhere else in the image. What else can be wrong with the BMP files?

As far as the makefile, I'm not declaring a PAL file, and I've tried both explicitly declaring the INDIR/OUTDIR and not doing so. I can paste in my makefile if I need to.

Any help would be appreciated. Thanks!

[The MAZZTer]

Are you running the tool using the directory with the BMP files as your working directory?

If you don't know what that means, try using absolute paths in your makefile.

_________________
http://www.mzzt.net/ | I am a respectable admin with a respectable sig.

[Tarsus]

Yes, I have tried it both ways; my BMPs are in the same directory as 2WAX.EXE and my makefile. Therefore, I have tried running this with only the filenames set in the makefile. And I have also tried an absolute directory in the makefile:

SET INDIR C:\bmpdf

And at your suggestion, I've now tried using absolute paths for the cell declarations themselves (without the INDIR line above):

0: bmp=C:\bmpdf\test001.bmp

Same result in every case. As a final test, I grabbed the Thrawn WAX from this site, WAX2'd the BMPs, and replaced the reference to the first cell in my makefile with thraw000.bmp. (Everything is in the same directory.) Sure enough, now the error that I get is for the second BMP, meaning thraw000.bmp is being opened correctly. It seems certain that the error is with the BMPs themselves.

In comparing my BMPs with those from the Thrawn WAX, I did notice that mine were RGB color mode instead of indexed, but changing that was to no avail either.

[Tarsus]

If someone's willing to take a little time to look at the images and makefile, I've placed them here:

http://tarsus.freehostia.com/smiley.zip

(These are, of course, horrible quality drawings that I'm just using to test out the process!)

Again, I'm placing the BMPs and makefile in the same directory as 2WAX.EXE when trying this, and I've also tried using absolute paths in the makefile.

[Tarsus]

Okay, after some studying I thought the problem was that even when indexed my BMPs didn't have the correct colormap (or at least correct number of indexed colors).

So I used my graphics tool, the GIMP, to meticulously create a custom palette from the colormap of one of the original DF WAX cells. Then I converted my BMP to indexed using that palette. Here's the result:

http://tarsus.freehostia.com/smiley001.bmp

It's an indexed file with 256 colors in the index, corresponding exactly to the colormap of a DF file! And yet I still get the error.

I'm officially stumped. I can't think of anything else to try.

[Tarsus]

Wow. I mean, wow. Know what the problem turned out to be? My filenames were too long.

Now, all that work I did to re-create the 256-color palette and index all my images with it was necessary, because I got that error just as soon as I fixed the filename length. It's just bizarre that something this simple was what was causing the error all along. Glad I tried one more thing!

[klasodeth]

How long were your filenames?

[Tarsus]

Nine characters. (smiley001.bmp) Guess that'll teach me to use a generic filename in the forum post!

[The MAZZTer]

Well they are DOS programs, they can only handle up to 8.3 characters. Longer than that (or if they contain spaces or extra dots) and Windows will create a special unique 8.3 name for DOS programs and only show you the long name.

Next time you create an example like "test0001.bmp", make sure it doesn't work first.

_________________
http://www.mzzt.net/ | I am a respectable admin with a respectable sig.