Gutenprint / CUPS Dye-sublimation drivers

The 8500 is a rebadged Mitsubishi CP3020, but the 9810 and 8800 is unique.

Both need sniffs to determine if they need an intelligent spooler.

Experimental support is in gutenprint, but this needs testing.

Sony sold their printer line to DNP, which rebranded these two models.

Investigate their Spool format, and find out if they need an intelligent backend.

Code complete, but needs testing.

They're an older-generation of the D70 family's.  Data tables are unfortunately embedded inside the drivers, which makes this a lot more challenging. Sigh.

(If nothing else, simply obtaining the RGB->YMC conversion/gamma tables would make the printers immediately useful..)

No documentation, no driver option, no idea how to drive it

Printer supports combining two or three 6x8 prints into 6x14 or 6x20 panoramas. Figure out how to do this.

The Kodak 6900 is actually the CHC-S2245-6A

HiTi M610 is also a rebadged S2245

We have the USB VID/PIDs already; just need a printer to test with.

P310W is consumer model; the rest are workhorse kiosk printers.

P530D is a duplexing variant of the P52x series. 

X610 is a tandem printhead high-throughput model that can spit out panoramas over 4 feet long.

P510 series has a binary-only driver

P110 series has a binary-only driver

There are also other unknown variations, P518A, P728, P310L.

It would be really nice if we could figure out how to query the UP-DR200's printer status, media counts, flow control, etc etc.

On CP900, CP910, CP1200, and CP1300, a telnet interface is present on port 59223. Upon connecting, it prompts for an unknown password.

DS620 supports 6x14 and 6x20" prints, and DS820 supports 8x18/8x26 and 8x20/8x32 sizes.

The T1/S8145 is a fancy "silver foil" thingey with two print heads.

Aka ColorStream D3

They're the same thing, an 8" wide duplex printer. No support in Gutenprint yet.

It's a card-sized variant of the CS2.  89mm cards of various widths.

And other members of that consumer printer family.  Not sure if they're still relevant commercially.

http://www.mitsubishi-photo.cz/drivers/drivers_printers/cz_en/(%20+firmwaredrivers_printers_digi_en.htm

This references a lot of firmware images, including 'Firmware CP9600 for 13x18,10x15(x2),11.5x15(x2).zip -- None of those sizes are covered in the current Gutenprint code.

The command I know about just returns an endless status stream..

There's a proper serial number returned so that's not so critical.

The P93D supports it via the driver, but the P95D doesn't.  However, it can be set on the front panel..  It's in there somewhere, I bet.

Bonus -- Can we enable/disable iserial support?

User reported the CP800 failed to print when on battery power, turns out one of the status fields changed causing us to fail.  I believe that's resovled for the older SELPHY CP models, but the newer ones (including the CP790) and the ES series are likely to have a similar problem.

Not sure what can be done about this without getting the battery and experimenting...

I have a P120i that is in perfect working order.

K extraction appears to work well, but when printing there's a registration problem.  The black layer appears to print several pixels higher than the color layers.

I'm not sure if this is something that needs a fixed offset applied, or is something that can vary printer-printer.

Duplexing 8x12" models, capable of cuts along both axes.

300 dpi, 4" or 5" wide media.

Not sure if it uses sheet or roll media.

The problem is that, at compile time, we rely on some host-compiled stuff to generate data files and whatnot. We need to compile the same stuff for the target.

I wonder if it may be worth moving to a more unified Make environment..

Right now the backend won't attempt to send anything over if a given deck is printing. However, the printer has 4*num_banks buffers that can be used to pipeline jobs!

So, try to figure out a way to overlap things. We'll have to be smarter about jobids..

HiTi supplied source code, in an "open source all rights reserved" drop

EK8810 in the APEX system supports 8x18/8x26 and 8x22/8x32 sizes (plus 8x14/16/20/36?)

Figure out how to make this happen. And what needs to be done on the overlapping areas..

We can tell if the printer is idle or busy, but don't know how to query active jobs, free buffers, and so forth.

P520/P525 and P720/P750

3.5x5 -> 7x5

4x6 -> 8x6

Combined size is slightly smaller (4 and 23 rows respectively) so we'd have to trim the ends off. Not a big deal.

While I have a P520 printer, I don't have the necessary media to test this.

I'd like to use this as an opportunity to learn ReStructuredText, and the Sphinx documentation system.

How far I take this is another matter..

There's presumably a way to configure this stuff from the host side,  In a FILE output mode the Windows drivers don't allow these settings to be changed, so to make this happen I'll have to obtain sniffs.

The UP-DR200, under Windows, will automatically combine two 4x6 copies into a single 8x6" print.  I have a sniff of this happening; figure out how it works, and implement in the backend if feasible.

I _think_ it's the same algorithm for both.  So use the CP30 to work out what's needed, and test it out.

The library API is already there, so once ready it should just be drop-innable.

Note:  This will also require extracting the data tables from the CP30 driver.