thanks for the warm welcome. i hope it is. it has caused many sleepless nights to some guys, including me.
But isn't the problem with the ripping rather than the image file format? From the way I am reading this, the new device will be able to read copy-protected formats without having to go to the effort of manually figuring out the track lengths, or whatever you have to do to achieve it now. Then you can stick it in an IPF.
yes and no. first, the software lets you image just ANYTHING. as long as we can "talk" to the drive, we get the whole stream that's coming in and stuff it into stream files. if it's on the disk, it's in there. that means we can read any copy protection, whatever. the "problem" is to represent data in a way that it appears to the (virtual) computer. you can do this to some extent with raw images, but more advanced copy protections will need scripting added, which is what we use our analyser for. therefore, the "consumer" will not be able to create ipfs (might come someday, though), but we can use dumps made by users to turn them into ipfs...
And considering, this should work for my Amiga, ST, Mac, C64, Apple II and possibly eventually my Coleco Adam (Thinking that's all I have that use floppies).. seems like it would be a bargain compared to getting multiple other solutions.
yes, this is one solution intended to dump them all.
There's a point - does the software support Linux currently?
If so, is there any chance of a kernel driver for it being made?
If so, we could mount it as if it were a normal floppy drive
(or it could be done via FUSE if you don't like the GPL...)
I seem to remember reading that the initial release for the software was command line for Windows, but that it was portable, so it should be able to be converted over.
Of course, going from a command line program to a driver might take a bit of work, but I'm betting there will be some people who jump on that once it's released..
two things here... please be aware that this is not free software, it's free for anyone in private, but we do exclude commercial use and other things. i think you know and understand why... there are other commercial solutions that try to do the same or less. we don't want to fix their problems (we encourage competition, we just don't want to see our code in competitive products). so there will be some restrictions in place. i am not sure if these might be a showstopper for the kernel driver. please note that this is meant as a gentle and warm hint, it's not intended to sound like there's a lawyer talking. we just have to see what we do have to "enforce" to make sure this thing does not get ripped apart.
making a kernel driver is also morally depreceated. i did state the reasons for this. we're coming from the preservation approach, so this is like asking someone at a museum to produce something that could harm valuable paintings... it just feels odd. this solution never was intended for realtime and interactive floppy use. if you want to simulate a real floppy, look at the hxc project. it does the same and i think even write support is coming or planned. floppies get older, and their surface will become weaker and reach the point where it will all fall apart. they should not be messed with more than you have to. if you really have to... why don't you use a real machine for this as well?
you will also run into problems with emulation because the usb-drive will be slower overall. you have this layer inbetween, where data is read and then needs to be interpreted, somehow, to be presented to the virtual host... this is not trivial. this will cause latency, which might break some more advanced protections. you could halt the emulation while you do this, but hey, we've got ipfs for that, right? of course, this will not be a problem for e.g. normal ados disks, but i do not see any benefit in using a real floppy compared to an image. in fact, this is the way i do work with disks today. i work in emulation with nearly no slowdowns for disk access. i prepare masters and other things and then write the image back to disk to use it in a real machine.
best,
chris
