Emulating Voyager disks

Hi everyone,

We’ve run accross two use cases of optical media from the Voyager collection that we are unable to emulate. The two titles we are trying to get working are Photograph to Remember and Beethoven’s Ninth Symphony CD Companion. I believe the first title was created using “Director?” and the Beethoven disc is a Hypercard application.

The challenge we’ve encountered with both of these optical media titles is that the media contains both a data track for the application and an audio tracks encoded on the same disc. When launching this application in an emulated envirnment our mounted disc all we see is the audio tracks. The application data is not showing up in the emulator. Is this a limitation of the emultor? Here’s a screen shot of what the disc looks like in OS 9.2. All that is mounted is the audio and we can’t get the emulator to recognize the data track. Does anyone have any ideas how we might solve this problem?

Hi Michael,

I suspect it might be related to the way the CD was imaged. These mixed-mode CDs are particularly challenging to capture.

@johanvanderknijff has a great post about CD imaging here with a long section on such discs: Preserving optical media from the command-line and I know there are more resources he’s shared at https://www.bitsgalore.org/ that would be of use.

If we’re lucky Johan may jump in with a recommendation also.

Hopefully that helps, please update if not or if I’m way off-base.


Hi Euan,

Thanks for the pointer, this could be the issue. This gives me something to see if I can troubleshoot, thank you and I’ll let everyone know how it goes. Michael

@klaus.rechert also directed us to this paper about prior efforts here with Voyager discs specifically:


A couple of ideas that might help to identify the culprit:

To rule out that the behavior you’re getting results from the emulator, you could try to open the image file with a virtual drive application. I’ve successfully used CDEmu (Linux-only) for mixed-mode images in the past:


From what I recall, if all goes well the audio and data tracks should show up as 2 separate devices.

If that works, try to add the virtual drive to your emulator (whether this’ll work probably depends on the Mac emulator you’re using; I’ve only done this with VirtualBox myself).

Also, I’m probably stating the obvious here, but with bin/cue images make sure to always open the cue file in the emulator or virtual drive, NOT the bin!

To rule out some weird/unusual CD layout I’d recommend to run the cd-info tool on the physical disc, as explained in my blog (“Identifying mixed-mode CDs”). Feel free to post the output here if you’re not sure what to make of it.

Finally, if it’s an imaging problem you might give IsoBuster a shot (Windows-only though), which is also capable of bin/cue output.

Thank you Johan, Just a heads up where we are with our attempts to fix this. We are having to relocate the discs in our collection to reimage. We’ve got them, its just not super easy to get into our collections right now particularly as this is an unprocessed collection. Hopefully we’ll have discs in hand soon so we can see what fixes this…I’m excited to try. Thank you!

I just wanted to return to this thread after spying on a Twitter exchange involving @johanvanderknijff and Paul Wheatley at DPC - Paul pointed out this “Flashback” optical imaging workflow from the British Library that I find very useful/interesting, particularly the steps for recreating “access images” from multi-track and multi-session discs using cdrdao, toc2cue, bchunk, dd and genisoimage:

Workflow:Flashback project - optical media imaging workflow - COPTR

At this time these extra steps would probably be critical for getting working images into EaaSI. BIN/CUE files can work in local emulators using the virtual drive technique Johan mentioned, and in theory we can get them working in EaaS as well, but neither the EaaSI or Demo client/interface are currently set up to import this image type properly. So some conversion back into ISO image(s) may be required.

@ethan.gates The extra steps listed there are only for multisession discs (“enhanced” audio / “Blue Book”). See also this blog post I once wrote.

Mixed mode discs only have one session (which contains both audio and data tracks), and these can be captured by (and accessed from) a single BIN/CUE pair (you might need the TOC2CUE step if you use cdrdao). Oddly the diagram doesn’t cover this case at all.

1 Like

Thank you for the clarification, @johanvanderknijff - my interpretation of the diagram was that mixed-mode discs would fall under their “Sessions? =1” → “Tracks? >1” path (then “Console Mode? = N”, which, I’m not sure what “Console mode” refers to here), which they send down the same toolchain as multi-session/enhanced/“Blue Book”.

Substituting IsoBuster for cdrdao as the imaging tool for mixed-mode would allow skipping the cdrdao and toc2cue steps, as you’d go right to BIN/CUE. But I guess from there is where I get confused in the difference between multisession and mixed-mode re: creating a mountable ISO image from the BIN/CUE (which, as I mentioned above, may not be necessary for emulation in general but will be necessary for EaaSI/EaaS users limited by ISO-only import.

Specifically: is it still necessary with a BIN/CUE from a mixed-mode CD to calculate the track offset for the data track and generate an ISO with appropriate padding? Otherwise, will generating an ISO (or treating the BIN as an ISO) just wind up with the emulator reading the audio track, as @mgolson and co. seem to be observing in the screenshot at the top?

@ethan.gates actually you’re right, I had overlooked that (probably got confused by the “Console Mode” thing).

is it still necessary with a BIN/CUE from a mixed-mode CD to calculate the track offset for the data track and generate an ISO with appropriate padding

For all examples I’ve seen the start offset was 0, so this shouldn’t be necessary.

Would be useful (and simpler) if EaaS would add support for BIN/CUE files, either directly or using some virtual drive emulator.

1 Like

Gotcha, thanks @johanvanderknijff! Yes, this would definitely be a work-around and not ideal. I think EaaS can technically support them on the back-end (e.g. manually uploading and manipulating BIN/CUE resources on the server), but it’s more the matter of making the pathway available through the client/web interface. Hopefully that’s a relatively quick UI/front-end piece that can go up the development queue.