Cubase Audio Dongle Clone

[TXG/MNX]

And a quick update on the Cubase Audio Falcon dongle reversing project... the analysis of the die image was completed, a schematic created and then an emulator was created from that, but it didn't work. Not really very surprising considering the complexity of manually analysing the die image and that even a single error in that or the emulator would result in total failure. So everything is being carefully done again using improved methods and although some errors have so far been found and corrected the emulator still doesn't work, so further checking is required. As was known from the start there is no guarantee of success but it's still being worked on.

I wish you all the luck to finally crack the holygrail on cubase for Atari

[Zippy]

Does anyone have a broken (non-working) CAF (Cubase Audio Falcon) dongle they'd be willing to sell or donate, please?

We have an offer of another working CAF dongle but as this will be a complete, destructive de-cap down to the bare die it seems a shame to destroy a working dongle if it can be avoided.

[Zippy]

Thanks to a very generous donation of a 2nd CAF dongle we were able to 100% confirm the bondwire connections by doing a more complete de-cap. Pic attached.

Turns out we had some connections out by 1 position, but unfortunately even with that corrected we're still not getting the expected response to their initial 128 bit dongle test. So we now have the schematic simulated in Proteus and also emulated in 68000 code, both giving the same result, but it's not quite right. So it seems there must still be at least 1 error in the manual analysis of the gates and therefore further checking will be needed. Slightly better quality pics of the die will be possible now that it's fully exposed so hopefully that will help track down the problem.

It's very close now and we're not giving up.

[mikro]

Fingers crossed Zippy, for me this is pure wizardry! :)

[Cyprian]

Fingers crossed

[TXG/MNX]

Man this is great news this would be the best thing ever created if you succeed. Fingers crossed.

[shoggoth]

Here's the source for the dongle emulator crack if you want to take a look.
...
CBCRKSRC.zip

I browsed it a bit, it's so cool! I just came to think of one thing - does it make sense to add a little jsr to a cache flush rout? Code has been changed in RAM, the 030 caches are so tiny it probably never fails, but is that the case on the 040-060?

Once before Pexec(GO) and once in the 'patch' subroutine just before RTS.

[shoggoth]

On this topic - reverse engineering the dongle for Notator Logic/Audio would have some importance for software preservation. The hardware timing is so sensitive so that it doesn't even work on certain machines, so being able to construct a dongle emulator like the one Zippy made would bring this truly great application back to life.

I used to have a LOG3 dongle but could never get any version of Logic to run on any of my machines. If I knew where it was I'd happily donate it for decapping, but so far no dice.

[Zippy]

does it make sense to add a little jsr to a cache flush rout? Code has been changed in RAM, the 030 caches are so tiny it probably never fails, but is that the case on the 040-060?

Not sure if it's needed but no one has said (yet!) that it doesn't work on Falcon or other machines. Has anyone even tested it on a Falcon?

Regarding the Notator dongle, we have one and the offer of another one, so it may get looked at if (when?) we complete the CAF dongle hack. The initial look at the Notator dongle shows that it's encapsulated in epoxy inside the case, so at this point even the chip type is unknown.

[shoggoth]

Not sure if it's needed but no one has said (yet!) that it doesn't work on Falcon or other machines. Has anyone even tested it on a Falcon?

It's theoretical. If you generate code in memory, the data cache will keep track of this, but not the instruction cache. When the CPU jumps to the generated code, there's a theoretical *kaboom*. If the caches happen to be small enough and/or the amount of code executed prior to the generated code is large enough, it works by chance. If not, the CPU gets lost.

It's just something I noticed, and I thought the patch/crack looked so nicely implemented that I figured why not make it a work of art!

Additional questions:
- the Cubase binaries are packed with PFX (IIRC), and you patch the unpacked to patch the binary. Not questioning that, it looks cool, but is there a reason for not unpacking the binary?
- is there a reason for not patching the binary itself, i e the data on disk; does Cubase do some sort of integrity check on itself?

Regarding the Notator dongle, we have one and the offer of another one, so it may get looked at if (when?) we complete the CAF dongle hack. The initial look at the Notator dongle shows that it's encapsulated in epoxy inside the case, so at this point even the chip type is unknown.

Fingers crossed. Cubase is an iconic application on the platform, and Notator Logic is on par with that thing. Yet very few people got the opportunity to try, including those who bought it.

Anyways, amazing work. I've had a great time patching MROS/Cubase and getting it to run on the Vampire V4 because of your crack for Cubase 3.10. I have a dongle, but obviously I can't use it on that hardware. So the crack opens up new opportunities.

[frank.lukas]

Regarding the Notator dongle, we have one and the offer of another one, so it may get looked at if (when?) we complete the CAF dongle hack. The initial look at the Notator dongle shows that it's encapsulated in epoxy inside the case, so at this point even the chip type is unknown.

Maybe start a new thread about the notator / logic dongle...

I have a faulty C-Lab Combiner. I think it has a Notator dongle inside?

EP600.jpg

[ldv-01]

The C-Lab Combiner is just a dongle expander. Dongles are incorporated in Unitor C (Creator), Unitor N (Notator) and Log 3 (Logic and Notator).

I have a faulty C-Lab Combiner. I think it has a Notator dongle inside?

[Zippy]

Additional questions:
- the Cubase binaries are packed with PFX (IIRC), and you patch the unpacked to patch the binary. Not questioning that, it looks cool, but is there a reason for not unpacking the binary?
- is there a reason for not patching the binary itself, i e the data on disk; does Cubase do some sort of integrity check on itself?

I left it packed and patched it like that so the whole thing would still fit on a single floppy disk which is how I was testing it on my real ST. With the main program file unpacked it didn't leave enough space for much else on the disk and I don't have any HD for my ST. I do nearly all the work in STEEM on my PC but for final testing and debugging it had to run from floppy on my ST.

Cubase is an iconic application on the platform, and Notator Logic is on par with that thing. Yet very few people got the opportunity to try, including those who bought it.

From the picture of the C-Lab Combiner that frank.lukas just posted it has an Altera EP600 on the board, which is the same PLD chip used in the Cubase 3.x red dongle, so that looks quite promising if that is indeed the chip for the Notator dongle.

Anyways, amazing work. I've had a great time patching MROS/Cubase and getting it to run on the Vampire V4 because of your crack for Cubase 3.10. I have a dongle, but obviously I can't use it on that hardware. So the crack opens up new opportunities.

I saw that, looks like a great development.

[frank.lukas]

LOG3 has also a EP600PC and a EP330PC too ...

EP600PC.jpg

[troed]

I have depoxy:d a Creator dongle. It's indeed an EP600.

[Zippy]

Progress, we found final 2 errors in die schematic in last 24 hours.

But testing doesn't work in STEEM or even my real ST even with real dongle , reporting that it needs a real Falcon030 and then that Audio engine can't start. Just doing basic test of going to file menu, closing down open file and discard changes, then selecting "About" menu (Diamond at left side menu option) it just sticks in loop of "Internal Error" like old error for bad crack, but same thing happens with real dongle on my ST.

Is there someone with real Falcon who could help with testing, maybe someone who already has CAF working? I'm using the file CAF206 downloaded from here: https://drive.google.com/drive/folders/ ... ibvVU-N1fh

Thanks!

[Greenious]

Is there someone with real Falcon who could help with testing, maybe someone who already has CAF working? I'm using the file CAF206 downloaded from here: https://drive.google.com/drive/folders/ ... ibvVU-N1fh

Thanks!

Wouldn't the official archive directly from Steinberg be a better choice?

[Zippy]

^^ Maybe, but I just tried opening it with Opera, Brave and Chrome and none of them worked. Has been years since I used an FTP link but have they removed FTP support from browsers recently or something?

[Greenious]

^^ Maybe, but I just tried opening it with Opera, Brave and Chrome and none of them worked. Has been years since I used an FTP link but have they removed FTP support from browsers recently or something?

Yes, many modern browsers have... I've noticed. But I guess that means FTP sites are back to being a leet thing... lol

Luckily, you got a hardware guru on your side...

[Zippy]

Thanks!

Turns out the CAF_206.PRG from that archive is just a packed version of the CAF_206.PRG from the google drive link, so basically the same. Actually the google drive one may be better as it doesn't sit for a while depacking.

[Greenious]

As mentioned in this thread:

As you probably know there is no Cubase Audio titled 'Atari Cubase Audio CBX' per se, only Cubase Audio vx.x or Cubase Audio Falcon vx.x. The manual you are after would be for Cubase Audio v1.1, also referred to as Cubase Audio Falcon v1.1, this is the version that works on ST, STe, TT to utilise the Yamaha CBX, as well as Falcon.

It might be a good idea to test with v1.1?

[Zippy]

^^ That one doesn't seem to work at all, gives address error during depack in STEEM and also on my real ST.

[Greenious]

Which is causing the trouble, the one from google drive or the one from Dteinbergs FTP? Both?

*Edit* Ooopsie! If you read the entire thread linked above, it says the v1.1 version on Steinbergs FTP is corrupt! Didn't see that until now!

A working version can be found in that thread...

But for simplicity, I'll add it here

[troed]

Wouldn't Hatari be an option considering it does emulate Falcon?

[Atari74user]

Progress, we found final 2 errors in die schematic in last 24 hours.

But testing doesn't work in STEEM or even my real ST even with real dongle , reporting that it needs a real Falcon030 and then that Audio engine can't start. Just doing basic test of going to file menu, closing down open file and discard changes, then selecting "About" menu (Diamond at left side menu option) it just sticks in loop of "Internal Error" like old error for bad crack, but same thing happens with real dongle on my ST.

Is there someone with real Falcon who could help with testing, maybe someone who already has CAF working? I'm using the file CAF206 downloaded from here: https://drive.google.com/drive/folders/ ... ibvVU-N1fh

Thanks!

This one is my archive and is fine on a real Falcon. In fact, all of the ones in my archive work, as they have all been cleaned up where necessary, and tested on real hardware before I posted them up. For the record, I have CAF 2.0, and used the CAF 2.06 upgrade from the Steinberg site to create a complete 2.06 'package' which I use on my Falcon.

EDIT: I just realised, if you are trying to launch CAF on an ST, it won't work, unless you use the Left shift key, hold it down before you double click to launch, and it disables the audio features.

Kind regards