The Cambridge ABug software hacking/archiving thread

discussion of beeb/electron applications, languages, utils and educational s/w
User avatar
leenew
Posts: 3329
Joined: Wed Jul 04, 2012 3:27 pm
Location: Doncaster, Yorkshire

Re: The Cambridge ABug software hacking/archiving thread

Postby leenew » Sun May 14, 2017 3:29 pm

Thank you all for taking time to do this.
It really is appreciated. =D>
Maybe it will be quicker next time :wink:

Lee.

User avatar
billcarr2005
Posts: 1055
Joined: Fri Sep 09, 2005 3:01 pm
Location: UK
Contact:

Re: The Cambridge ABug software hacking/archiving thread

Postby billcarr2005 » Sun May 14, 2017 4:22 pm

Mathemagic Land

Using viewtopic.php?f=2&t=9061&p=115830#p115830 as a starting off point as potentially what the value of F07 and F0F were, I've managed to get this working... although the instructions on line 260 of START seem to be somewhat garbled!
Attachments
Mathemagic Land.zip
Mathemagic Land
(8.34 KiB) Downloaded 11 times

User avatar
CMcDougall
Posts: 5534
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: The Cambridge ABug software hacking/archiving thread

Postby CMcDougall » Sun May 14, 2017 4:25 pm

Westquest Audio:

Code: Select all

Date (D/M/Y) and time:
14/05/2017   17:22

Expected phase shift is +0.
Found :  WESTQUEST 00   Checksum ok
Found :  WESTQUEST 01   Checksum ok
Found :  WESTQUEST 02   Checksum ok
Found :  WESTQUEST 03   Checksum ok
Found :  WESTQUEST 04   Checksum ok
MISSED A BLOCK.
Found :  WESTQUEST 06   Checksum ok
Found :  WESTQUEST 07   Checksum ok
MISSED A BLOCK.
Found :  WESTQUEST 0D   Checksum ok
MISSED A BLOCK.
Found :  WESTQUEST 14   Checksum ok
MISSED A BLOCK.
Found :  WESTQUEST 1E   Checksum ok

all i get with a quick look on laptop, only left channel & only 4mins 1sec of look good waves :x

EDIT; but sound, as you say, knackered, needs rewound /fforward /banged on table /in another shell / tape>tape HiFi decks to pull!
ImageImageImage

User avatar
pau1ie
Posts: 269
Joined: Thu May 10, 2012 9:48 pm
Location: Bedford

Re: The Cambridge ABug software hacking/archiving thread

Postby pau1ie » Sun May 14, 2017 10:00 pm

What I learned:

Before -
  • get the adt2rom (Earlier in the thread) on a disc or disc image
  • Get a solid state recorder
  • Get all the required cables.
  • Get some cassettes which are screwed together so they can be used to replace the enclosure of dodgy ones.
  • Datacentre
  • USB stick
  • Cassette player.
  • 3.5" disc drivee
  • 3.5" discs
  • 5.25" disc drive (I don't have one of these, and it would have been useful)
  • Manual for the recorder you are using
  • ADT manual
  • Datacentre manual
  • DFS manual
  • ADFS manual

Load adt into sideways RAM with

Code: Select all

*srload adt2rom 8000 4

CTRL-BREAK to get the computer to read the ROMs

If the software is on cassette, use

Code: Select all

*xfer tape disc

If this doesn't work, then visually check the speed that the cassette is running at.
If it looks normal, use the recorder to record the audio.
If it looks slow, try fast forwarding and rewinding a few times.
If it is still slow, try to get permission to change the enclosure
Otherwise record it to capture the audio a couple of times.

The software ends up on a 3.5" disc (Assuming that was the default drive). Now transfer it to an image as below.

If the software is on disc.

Put the disc in the drive.
Change to the drive with the disc in it, and export the disc as follows:

Code: Select all

*export -d0 image.ssd


This will write the image to the USB in the datacentre. Also do side 2 if there is one.
If this doesn't work, then adt comes in again. Assuming there is an ADFS hard drive (CF card in the datacentre). Do:

Code: Select all

*disc
*drive 0
*.

Check that this is the correct disc

Code: Select all

*adfs
*cdir image
*dir image
*xfer disc adfs
*ram
*drive 0
*xfer adfs ram
*export 0 image.ssd


Make copious use of CTRL-BREAK, especially if something goes wrong.
It might be possible to skip some of the intermediate steps, but when I tried to do
*xfer tape ram
it didn't work, though maybe it is possible and I was doing something wrong.

What I am not clear on then is what to do if you can catalogue the disc, but there is an error when reading files. Is it possible to do a copy of the parts of the disc that work while skipping the parts that don't? I think the disc controller automatically retries a few times on error, so I suspect that if there is an error retrying the command is likely to fail again.

Does anyone have suggestions for improvements to the procedure above?
I'm working on http://bbcmicro.co.uk

User avatar
pau1ie
Posts: 269
Joined: Thu May 10, 2012 9:48 pm
Location: Bedford

Re: The Cambridge ABug software hacking/archiving thread

Postby pau1ie » Sun May 14, 2017 10:06 pm

Oh yes, sorry all the recordings I did are left only, as I only had left plugged in. I think it would have been possible to mirror them, but I got fed up with the manual. I hope it doesn't cause too many problems.
I'm working on http://bbcmicro.co.uk

Coeus
Posts: 407
Joined: Mon Jul 25, 2016 11:05 am

Re: The Cambridge ABug software hacking/archiving thread

Postby Coeus » Sun May 14, 2017 11:10 pm

pau1ie wrote:
lurkio wrote:Can't get the audio file to load.


Yes, that is what I found, and is why I uploaded the audio. I copied the audio in the hope that someone clever could get something out of it.


if it played too slowly, but at a consistent speed, it would be possible to resample it to get it back to the right speed, i.e. so the frequencies are correct. Correcting for speed varations would be much harder. One possible first step might to to open it in an audio editor with a spectrum analyser plugin. What I'd hope to see if two bug peaks are what should be 1200 and 2400 so by working out what actual frequency these have ended up with it should be possible to come up with an adjustment.

User avatar
davidb
Posts: 1824
Joined: Sun Nov 11, 2007 10:11 pm
Contact:

Re: The Cambridge ABug software hacking/archiving thread

Postby davidb » Sun May 14, 2017 11:20 pm

Coeus wrote:if it played too slowly, but at a consistent speed, it would be possible to resample it to get it back to the right speed, i.e. so the frequencies are correct. Correcting for speed varations would be much harder. One possible first step might to to open it in an audio editor with a spectrum analyser plugin. What I'd hope to see if two bug peaks are what should be 1200 and 2400 so by working out what actual frequency these have ended up with it should be possible to come up with an adjustment.

I think it's inconsistently slow, which makes it harder to read - it sounds awful to listen to! Having said that, the waveform looks very clean, so there's hope for this one. :)

User avatar
pau1ie
Posts: 269
Joined: Thu May 10, 2012 9:48 pm
Location: Bedford

Re: The Cambridge ABug software hacking/archiving thread

Postby pau1ie » Mon May 15, 2017 8:47 am

davidb wrote:I think it's inconsistently slow
Yes, I think it starts off OK and eventually stops. I imagine it is horrible.

billcarr2005 wrote: I've managed to get this working

Thanks, this really cheered me up!

lurkio wrote: I do seem to have got the audio playing into my Master now

Again, thanks, this was really encouraging.
I'm working on http://bbcmicro.co.uk

User avatar
leenew
Posts: 3329
Joined: Wed Jul 04, 2012 3:27 pm
Location: Doncaster, Yorkshire

Re: The Cambridge ABug software hacking/archiving thread

Postby leenew » Mon May 15, 2017 9:44 am

There are programs that can "straighten out" bad audio.
If anyone uses the expensive DJ/Producer app Ableton, this does it by looking at the peaks and then making them equidistant I think.
It is used for making old live recorings (where the drummer did not keep consistent time) more friendly for mixing, but whether it would work for beeb audio is anyone's guess!

Lee.

User avatar
CMcDougall
Posts: 5534
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: The Cambridge ABug software hacking/archiving thread

Postby CMcDougall » Mon May 15, 2017 4:03 pm

^Lee, already got that Sony Soundforge prog too, it's utter pish for to re-press RARE LOL PRICE vinyl, hence why we use WaveLab6 (again ive got all the whites /illegal presses /DJ Promos that went no further from BITD :mrgreen: ) & back now on everyones wheels of steel Technics SL1200s =; 8-[
ImageImageImage

User avatar
CMcDougall
Posts: 5534
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: The Cambridge ABug software hacking/archiving thread

Postby CMcDougall » Mon May 15, 2017 4:54 pm

Trivial Pursuit data tapes now as .CSW, seem to show up on TapeControl/s
Attachments
b1.zip
csw
(681.48 KiB) Downloaded 10 times
b2.zip
csw
(701.5 KiB) Downloaded 9 times
r1.zip
csw
(703.42 KiB) Downloaded 9 times
r2.zip
csw
(748.36 KiB) Downloaded 11 times
ImageImageImage

User avatar
CMcDougall
Posts: 5534
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: The Cambridge ABug software hacking/archiving thread

Postby CMcDougall » Mon May 15, 2017 7:26 pm

sydney wrote:I've got Wolfpack on the pc too just not had time to upload

did it work Simon? can't wait to fly another plane! (if it is this type of game as per DocSoft usual...)
ImageImageImage

User avatar
sydney
Posts: 1964
Joined: Wed May 18, 2005 9:09 am
Location: Newcastle upon Tyne

Re: The Cambridge ABug software hacking/archiving thread

Postby sydney » Mon May 15, 2017 7:32 pm

CMcDougall wrote:
sydney wrote:I've got Wolfpack on the pc too just not had time to upload

did it work Simon? can't wait to fly another plane! (if it is this type of game as per DocSoft usual...)

It is on my mother's pc and it's back at her house so it'll have to wait until the weekend before I get it.

User avatar
sydney
Posts: 1964
Joined: Wed May 18, 2005 9:09 am
Location: Newcastle upon Tyne

Re: The Cambridge ABug software hacking/archiving thread

Postby sydney » Tue May 23, 2017 1:02 pm

At my mothers house now so here is wolfpack.
I think it's only WOLF and WOLFPACK - Ignore the 3 other files.
Attachments
wolfpack.zip
(26.49 KiB) Downloaded 12 times

User avatar
CMcDougall
Posts: 5534
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: The Cambridge ABug software hacking/archiving thread

Postby CMcDougall » Tue May 23, 2017 3:30 pm

^hope she made you a nice cup of tea :D

I had my mum's car in bits the other week, seems car dealers don't actually service cars at all, so done all the usual for £30 , I even got a biscuit with my cup of tea for doing it 8) (now it's like a RS turbo, as garage ddone nothing for 4yr!)
ImageImageImage

User avatar
CMcDougall
Posts: 5534
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: The Cambridge ABug software hacking/archiving thread

Postby CMcDougall » Tue May 23, 2017 6:25 pm

mum just replied to an email :
"The car is going much better now, I didn't even notice I was doing 80 on the way home from Asda on Friday." \:D/

Wolf Pac III , has lots of *FX15,1 & *FX9,40 etc that all need taken out to so can run from disc, but it also makes MCode from &D00 to &DFC
so put it back to tape:
wolf.jpg
in game

so to go:
*TAPE
PAGE=&E00
CH.""
Attachments
wolfpac-docSoft1983.zip
uef
(11.66 KiB) Downloaded 11 times
ImageImageImage

User avatar
lurkio
Posts: 1141
Joined: Tue Apr 09, 2013 11:30 pm
Location: Doomawangara
Contact:

Re: The Cambridge ABug software hacking/archiving thread

Postby lurkio » Wed May 24, 2017 12:26 am

Thanks, sydney and Col!

CMcDougall wrote:Wolf Pac III , has lots of *FX15,1 & *FX9,40 etc that all need taken out to so can run from disc

Eh? I thought *FX15,1 cleared the current input buffer. And *FX9 changed the blink rate of flashing colours or summat?

Here's a transfer to .SSD disc-image:

    WOLFPACK.ssd.zip
    .SSD DFS disc-image of Wolfpack
    (7.96 KiB) Downloaded 13 times
Play online:

:idea:

User avatar
davidb
Posts: 1824
Joined: Sun Nov 11, 2007 10:11 pm
Contact:

Re: The Cambridge ABug software hacking/archiving thread

Postby davidb » Wed May 24, 2017 2:32 pm

After some messing around trying to adjust the speed of West Quest, here's the UEF and a screenshot. No guarantees about the UEF, of course, but it appears to work.

Can you guess which famous game this is a clone of, children? Here's a hint. ;)
Attachments
westquest.uef.zip
(8.43 KiB) Downloaded 15 times
westquest.png

User avatar
lurkio
Posts: 1141
Joined: Tue Apr 09, 2013 11:30 pm
Location: Doomawangara
Contact:

Re: The Cambridge ABug software hacking/archiving thread

Postby lurkio » Wed May 24, 2017 3:46 pm

davidb wrote:After some messing around trying to adjust the speed of West Quest, here's the UEF and a screenshot. No guarantees about the UEF, of course, but it appears to work.

This is great! Thanks, David. Did you make a constant speed adjustment or did the speed vary across different sections of the audio?

:?:

User avatar
davidb
Posts: 1824
Joined: Sun Nov 11, 2007 10:11 pm
Contact:

Re: The Cambridge ABug software hacking/archiving thread

Postby davidb » Wed May 24, 2017 8:17 pm

lurkio wrote:This is great! Thanks, David. Did you make a constant speed adjustment or did the speed vary across different sections of the audio?

I initially tried to use various tricks to convince my program that pulses of 1200 Hz and 2400 Hz could be wider or narrower than normal, then I tried playing the audio at a faster rate than it was recorded at. These were only partially successful, but some blocks steadfastly refused to load. When I tried to transfer block 2B, I realised that I would have to deal with a variable playback rate.

What I did in the end was to change the playback speed when encountering high tones to gradually try and compensate for the mismatch between the frequency being read and an ideal wave at 2400 Hz. I'm actually surprised that it worked! :shock:

User avatar
leenew
Posts: 3329
Joined: Wed Jul 04, 2012 3:27 pm
Location: Doncaster, Yorkshire

Re: The Cambridge ABug software hacking/archiving thread

Postby leenew » Wed May 24, 2017 8:51 pm

Impressed! =D>

Lee.

User avatar
davidb
Posts: 1824
Joined: Sun Nov 11, 2007 10:11 pm
Contact:

Re: The Cambridge ABug software hacking/archiving thread

Postby davidb » Wed May 24, 2017 8:58 pm

Thanks. :) It only worked because the signal was so clean - kudos to everyone who had something to do with that. :D

User avatar
pau1ie
Posts: 269
Joined: Thu May 10, 2012 9:48 pm
Location: Bedford

Re: The Cambridge ABug software hacking/archiving thread

Postby pau1ie » Thu May 25, 2017 8:31 am

Thanks davidb, this is amazing. I really thought the only hope was to try and get a better recording of the cassette! Can you share how you did what you describe? Do you have some expensive professional audio processing software, or can it be done cheaply?
I'm working on http://bbcmicro.co.uk

User avatar
davidb
Posts: 1824
Joined: Sun Nov 11, 2007 10:11 pm
Contact:

Re: The Cambridge ABug software hacking/archiving thread

Postby davidb » Thu May 25, 2017 10:45 am

I used a tool I wrote called recordUEF.py which really needs me to add more comments because I tend to forget what it's doing at times. :roll:

The tool is really simple in principle, but not quite to simple to use as it should be. What it does is try to filter out frequencies above and below the two frequencies used to represent data, clamps the resulting waveform to be between values of 0 and 1, then tries to figure out the spacing and widths of the pulses that it sees. The inspiration for this came from this article, which actually takes a simpler approach.

What I did yesterday was to look at one of the blocks in Westquest in Audacity and examine the spacing between peaks of high tone signal at different points in the block, finding that they differed quite a bit. I realised that it would probably be better if I changed my tool to check the frequency of the high tone periodically and compensate for variations by changing the time step between samples. So, if it "sees" high tone at 2200 Hz then it makes the time step smaller so that the frequency of the wave being read "increases" to 2400 Hz; if the tone was higher then it makes the time step larger, stretching out the wave. One side effect of this is that the tool loses track of the original time in the sample but I think I can fix this.

As I mentioned, the tool isn't easy to use, and the only cost is the time spent getting it to do what you want. You have to be comfortable with running it from the command line and tweaking the options. I also needed to use an audio editor to fix recordings of some old cassettes before feeding them to the tool because it doesn't handle glitches in the signal very well. :( But the Westquest audio was very clean, so that step wasn't required in this case. :)

User avatar
richardtoohey
Posts: 3354
Joined: Thu Dec 29, 2011 5:13 am
Location: Tauranga, New Zealand

Re: The Cambridge ABug software hacking/archiving thread

Postby richardtoohey » Sat May 27, 2017 12:43 am

Thanks for sharing. =D>

I get the general principle of how this is meant to work but every time I try and really understand it, it dribbles out my ears. :oops: ](*,)

So when I get that nice quiet day, all to myself #-o :roll: :-({|= I'll have a look at the Python code.

User avatar
davidb
Posts: 1824
Joined: Sun Nov 11, 2007 10:11 pm
Contact:

Re: The Cambridge ABug software hacking/archiving thread

Postby davidb » Sat May 27, 2017 10:04 am

I'm not sure my code is a great way to decode the data, though as long as I can get something off cassettes then I'm happy. :) I'm still considering reimplementing the Electron's cassette hardware in software at some point.


Return to “software: other”

Who is online

Users browsing this forum: No registered users and 5 guests