Any knowledge on the DSB format?

[ReallyMadMax]

Not sure if this is the appropriate place to ask but it seems like you are actively developing embroidery software and I have searched far and wide for more info on the .DSB format. From what I have been able to translate from .dst files this is how the bits are aranged:

Each embroidery stitch comes in 3 bytes. The first being machine commands, followed by Y and finally X value(s)

stitch | Y = - | X = - | END | Color | NA | NA | Jump
0 0 0 0 0 0 0 0
Y value
0 0 0 0 0 0 0 0
X value
0 0 0 0 0 0 0 0

X and Y are simple calculated using their literal binary value for a theoretical max stitch length of 255.

Machine commands are activated by having a 1 in its respective position
Here are some direct examples

10001000 or 88 = color switch and single stitch after wards
10000000 or 80 = single stitch
10000001 or 81 = jump
11111000 or f8 = end of machine file

The machine will always move BEFORE it stitches in a dsb file
Not sure if necessary but all existing examples show a single stitch after color change

UNIT CONVERSIONS

10 = 1mm so 1 = 0.1mm

1 inch = 25.4mm or 254

Is there any chance anyone here know what the two NA bits in the command byte could be for? I think one could be a PAUSE command for aplique or something similar, but am unwilling to test that theory live on my machine. Any tips on where to find this missing information would be very helpful. Hopefully this information can also help this project if it eventually wishes to support .DSB aswell.

[robin-swift]

Hi Max,

yeah this is the place. I've been collecting this information but I need help from people who can test on their home machines like you.

We intend to support whatever we have information about: this message is reason enough to include DSB. The following would help us use it:

1. Can you tell us about your machine and why it takes DSB?
2. What designs have you run to get this information?
3. Is it based on running a hexdump on the file or do you have custom tooling?
4. Is the header a DST header?
5. Do you want crediting for your efforts?

Thanks for the contribution. I hope there's more we can collaborate on making.

Cheers,

Robin

[ReallyMadMax]

Hey Robin,

Good to hear back, more than happy to contribute what I have gathered.

1. I have a TAJIMA based machine from china (Not exactly Tajima brand but uses .DSB)
2. I have a small business where I do my own digitizing and have exclusively made .dst files and converted them to .dsb using an app called Data7 Consultancy.
3. I used the converted .dsb file from the app and compared it against the .dst file using hexdump.
4. Yes it is identical to a dst header I believe.
5. A mention to Max Dilger / scfd.ca would be appreciated!

Let me know if there is other information I can provide for you. The data7 conversion app has been a live saver for me as the digitizing app I use currently only outputs to .dst . I think the app could also be used to reverse engineer other formats as-well. I am still looking for those 2 unknown bits of info but will keep searching in the meantime.

Thanks, Max

[bianchifan]

1. I have a TAJIMA based machine from china (Not exactly Tajima brand but uses .DSB)

Did you contact your seller?
DSB is an BARUDAN format, Tajima based machines should use DST, always!
If missing, it should be implemented....

As far as I'm told, pyembroidery (InkSTitch) & libembroidery are able to read DSB but cannot write.

[robin-swift]

I'm moving this discussion to libembroidery since it's about the underlying formats.