Re: Speccy loading noise? - Sinclair

This is a discussion on Re: Speccy loading noise? - Sinclair ; I did wonder if it was from another machine, it being a US show, but I think I've sat around listening to ZX tapes long enough in my time to pick out the "signature" Brian Gaff wrote: > Lots of ...

+ Reply to Thread
Results 1 to 14 of 14

Thread: Re: Speccy loading noise?

  1. Re: Speccy loading noise?

    I did wonder if it was from another machine, it being a US show, but I
    think I've sat around listening to ZX tapes long enough in my time to
    pick out the "signature"

    Brian Gaff wrote:
    > Lots of computers from around the 80s had similar noises of course,
    > but I think the Speccy was quite unique in allowing the user to hear
    > it.


  2. Re: Speccy loading noise?

    lol. Here we go again.

    I've edited the noise and raised the volume, but it still need
    cleaning.

    http://zxspectrum.zidev.com/zx/specc..._cleaning).zip


    I had a routine from Microhobby that could be used to read tape data
    even if it didn't had the leading tone. I'm going to see if I can find
    it.

  3. Re: Speccy loading noise?

    zxbruno wrote in news:9a41084a-4bf8-4435-a32a-
    8cc5870c1ade@u10g2000prn.googlegroups.com:

    > lol. Here we go again.
    >
    > I've edited the noise and raised the volume, but it still need
    > cleaning.
    >
    > http://zxspectrum.zidev.com/zx/specc..._cleaning).zip
    >
    >
    > I had a routine from Microhobby that could be used to read tape data
    > even if it didn't had the leading tone. I'm going to see if I can find
    > it.


    Unfortunately it's such a short sample though, that even if we were able to
    get the speccy to "recognise" it, it'll be almost impossible to decipher.

    Obviously, with it been a screen, it's unlikely there's be any recognisable
    code or text in there.

  4. Re: Speccy loading noise?

    On Mar 16, 9:32 pm, Scribbler wrote:
    >
    > Unfortunately it's such a short sample though, that even if we were able to
    > get the speccy to "recognise" it, it'll be almost impossible to decipher.
    >
    > Obviously, with it been a screen, it's unlikely there's be any recognisable
    > code or text in there.


    Maybe, but it's still worth a try. I found the routine I was looking
    for (this will probably look messy):

    10 ORG 40000
    20 DI
    30 LD HL,#53F
    40 PUSH HL
    50 LD DE,#4000
    60 LBYTE LD A,#7F
    70 IN A,(#FE)
    80 RRA
    90 RET NC
    100 NBYTE LD L,1
    110 LD BC,#B201
    120 LBITS CALL #5E3
    130 JR NC,LBYTE
    140 LD A,#CB
    150 CP B
    160 RL L
    170 LD B,#B0
    180 JR NC,LBITS
    190 BIT 7,L
    200 JR NZ,LBITS
    210 LD A,L
    220 CP " " 230 JR NC,OK
    240 LD L," " 250 OK LD H,0
    260 ADD HL,HL
    270 ADD HL,HL
    280 ADD HL,HL
    290 LD BC,15360
    300 ADD HL,BC
    310 LD C,D
    320 LD B,8
    330 PCHAR LD A,(HL)
    340 LD (DE),A
    350 INC HL
    360 INC D
    370 DJNZ PCHAR
    380 LD D,C
    390 INC E
    400 JR NZ,NBYTE
    410 LD A,D
    420 CP #48
    430 LD D,#50
    440 JR Z,NBYTE
    450 LD D,#40
    460 JR NC,NBYTE
    470 LD D,#48
    480 JR NBYTE

    If this code can be modified, we can:

    1-See if we can make something of what we see, the same way we used to
    try to find sprites inside games (changing the width, height,
    position, etc.) There are some emulators that have this function
    available.
    2-Compare the data with some screen$
    3-Have Specchums listen to it, and maybe someone will recognize it!

  5. Re: Speccy loading noise?

    You need to know if its code or basic, but if its code just save a block of
    code longer then you need, use the header and put the code you have as the
    code, then when you get loading error you can look at the code in ram.
    Remember to save it with ramtop lowered though, and load it back up there
    as well.

    Brian

    --
    Brian Gaff....Note, this account does not accept Bcc: email.
    graphics are great, but the blind can't hear them
    Email: briang1@blueyonder.co.uk
    __________________________________________________ __________________________________________________ __________


    "zxbruno" wrote in message
    news:9a41084a-4bf8-4435-a32a-8cc5870c1ade@u10g2000prn.googlegroups.com...
    > lol. Here we go again.
    >
    > I've edited the noise and raised the volume, but it still need
    > cleaning.
    >
    > http://zxspectrum.zidev.com/zx/specc..._cleaning).zip
    >
    >
    > I had a routine from Microhobby that could be used to read tape data
    > even if it didn't had the leading tone. I'm going to see if I can find
    > it.




  6. Re: Speccy loading noise?

    Brian Gaff did eloquently scribble:
    > You need to know if its code or basic, but if its code just save a block of
    > code longer then you need, use the header and put the code you have as the
    > code, then when you get loading error you can look at the code in ram.
    > Remember to save it with ramtop lowered though, and load it back up there
    > as well.


    Isn't everyone forgetting about the slight problem of bit placement?
    you can't just leap into the audio at an arbitrary point and expect it to
    always be at a byte boundary.

    You'd need to do a lot of shifting to get it into the right alignment.
    If the signal was received strongly enough to have an alignment that is.

    --
    __________________________________________________ ____________________________
    | spike1@freenet.co.uk | "Are you pondering what I'm pondering Pinky?" |
    |Andrew Halliwell BSc(hons)| |
    | in | "I think so brain, but this time, you control |
    | Computer Science | the Encounter suit, and I'll do the voice..." |
    ------------------------------------------------------------------------------

  7. Re: Speccy loading noise?

    On Mar 17, 3:38 pm, spi...@freenet.co.uk wrote:
    > Brian Gaff did eloquently scribble:
    >
    > > You need to know if its code or basic, but if its code just save a block of
    > > code longer then you need, use the header and put the code you have as the
    > > code, then when you get loading error you can look at the code in ram.
    > > Remember to save it with ramtop lowered though, and load it back up there
    > > as well.

    >
    > Isn't everyone forgetting about the slight problem of bit placement?
    > you can't just leap into the audio at an arbitrary point and expect it to
    > always be at a byte boundary.
    >
    > You'd need to do a lot of shifting to get it into the right alignment.
    > If the signal was received strongly enough to have an alignment that is.
    >
    > --
    > __________________________________________________ ____________________________
    > | spi...@freenet.co.uk | "Are you pondering what I'm pondering Pinky?" |
    > |Andrew Halliwell BSc(hons)| |
    > | in | "I think so brain, but this time, you control |
    > | Computer Science | the Encounter suit, and I'll do the voice..." |
    > ------------------------------------------------------------------------------


    Er, the assembler routine I posted can read anything from a Spectrum
    tape, no matter what point you start reading from. It's from
    Microhobby, a Spanish magazine, and it works great! Its original
    purpose was to be able to look for text in code blocks without the
    need to load it in memory. If someone who knows machine code can
    modify it, it could be used to read speccy audio which otherwise would
    be lost forever. I know a regular block of code needs the leading
    tone, flag and sync bit, followed by the actual code, but with a
    slight modification (by someone who knows), we can read just the code.

    I honestly think that this is the 3rd part of a screen$, and by the
    amount of 0s it should mean the last two lines were left blank, but it
    might be possible to recognize something. A gentleman at Speccy.org
    had the same opinion.

    If we can have someone filter the audio the right way (I'm an amateur
    at that) or modifiy the routine I posted (I only know Basic) I think
    we will find something.

  8. Re: Speccy loading noise?

    On Mon, 17 Mar 2008 zxbruno wrote:

    >I honestly think that this is the 3rd part of a screen$, and by the
    >amount of 0s it should mean the last two lines were left blank, but it
    >might be possible to recognize something. A gentleman at Speccy.org
    >had the same opinion.


    I'm pretty sure you're right. This sounds like the same burst of code
    that's been used a couple of times in the various incarnations of CSI
    (Google this group for more info). The CSI clips were longer, including
    an instantly recognisable attribute table loading at the end, but were
    masked by other noise and dialogue. The version from The Wire is the
    clearest by far.

    It would be pretty cool if we could figure out what program this is
    from. Not sure how to go about it, though. I suppose the routine you
    posted could be used as the basis for a literal bit-for-bit comparison
    with EVERY possible SCREEN$ until a match is found, but that would take
    an intolerably long time.

    --
    Kev
    __________________________________________________ ________________________
    "Man wanted to work in dynamite factory. Must be willing to travel."
    Classified advertisement

  9. Re: Speccy loading noise?

    On 2008-03-17, spike1@freenet.co.uk wrote:
    > Isn't everyone forgetting about the slight problem of bit placement?
    > you can't just leap into the audio at an arbitrary point and expect it to
    > always be at a byte boundary.


    If it's a SCREEN$, it doesn't really matter - if you get it to show in
    the Spectrum's frame buffer, you just need to look at what's drawing on
    the screen, then you can probably align it just by looking at it and
    guessing the start byte and start bit for the fragment.

    If there's attributes included in the clip, you could almost certainly
    figure out which byte and bit that starts on with some trial and error.

    --
    From the sunny Isle of Man.
    Yes, the Reply-To email address is valid.

  10. Re: Speccy loading noise?

    Quoting from a post made by Gasman on w.o.s.

    "Here's my attempt at making sense of it:

    I loaded the original audio into Audacity, ran it through a 10000Hz
    high-pass filter, then amplified it by 24dB twice with 'allow
    clipping' ticked - the extreme clipping gave me something
    approximating a square wave. I saved the result out and converted it
    to a raw data file using SoX (sox infile.wav -ub outfile.raw).

    I put together a Ruby script (http://altaria.vm.bytemark.co.uk/
    speccy_wire/decode.rb) to decode this into pulses rather naively - it
    just counts the time taken to rise to 255 and back down to 0 again.
    With a bit of statistics collecting and trial-and-error, I divided the
    pulses into 'zero' bits (16 to 30 samples), 'one' bits (42 to 53
    samples) and 'not-sure' bits (anything else). As suspected, this gave
    me four blocks of (rather dirty) meaningful data, separated by long
    runs of zeros.

    I then fed this into a second Ruby script (http://
    altaria.vm.bytemark.co.uk/speccy_wire/plot.rb) to plot the data in the
    Spectrum screen layout Because each 'not-sure' bit is a potential
    place where we're mis-counting bits, we can't measure the length of
    any chunk with any real confidence, so I did a bit of hand-tuning
    based on the assumption that the first 'one' bit after each long run
    of zeros would occur at the same X position each time. I didn't know
    whereabouts on the screen this would be (although in hindsight it's
    pretty obviously going to be the far left) so I made the image 512
    pixels wide (so the top character row would show the bits at offset
    0...511, the next character row would show the bits at offset
    256...767 and so on). This gave me plot.gif (http://
    altaria.vm.bytemark.co.uk/speccy_wire/plot.gif)

    On that plot I noticed some diagonal bands drifting to the right, and
    the data covered slightly more than 5 pixel rows, which suggested that
    I was over-counting the number of bits. There were a few places where
    we had two consecutive 'not-sure' bits, so on a hunch, I tried
    replacing each of these with a single 'not-sure' bit. The result is
    plot2.gif (http://altaria.vm.bytemark.co.uk/speccy_wire/plot2.gif) - a
    bit straighter, but still not brilliant. However... it may be my
    imagination, or just seeing what I'm expecting to see, but is that the
    rounded corner of an Ultimate logo I see (in the center)? I think we
    may be looking at Sabre Wulf again."

  11. Re: Speccy loading noise?

    zxbruno wrote:
    > replacing each of these with a single 'not-sure' bit. The result is
    > plot2.gif (http://altaria.vm.bytemark.co.uk/speccy_wire/plot2.gif) - a
    > bit straighter, but still not brilliant. [...]

    Hi,

    I've also been working on decoding this sound. I've used your test1 and
    test2.wav because I've had some problems extracting the sound correctly.
    I also used audacity to further process the clips, using normalize to
    remove any dc component and leveler (at a very high setting) to remove
    volume fluctuations. The original clip has an high background noise,
    your clips are better, and I filtered them further.

    I've wrote some python scripts to decode spectrum loading sounds. The
    current script correctly decodes my test signals (generated from .scr,
    not real tapes), but I'm not so sure it is handling the sound from the
    video correctly. Here are my results:

    Original size:
    http://farm4.static.flickr.com/3179/...8456bb_o_d.png

    Same image, just amplified:
    http://farm3.static.flickr.com/2007/...083a08_o_d.png

    I think I can see some structure, and I've seen the same structures in
    the various experiments I've made, however it's hard to see anything.
    Maybe someone can make head or tails out of it?

    Also, can anyone upload a .wav recorded from a tape of a screen of a
    known game? I don't have my tapes (nor a tape recorder, that is old
    tech!) here. A .wav from a tape would be a better test for the decoding
    software than cleanly generated signals.

    I really hope we can solve this.

    Regards,
    Jo„o Silva

  12. Re: Speccy loading noise?

    Hi Jo„o (PortuguÍs como eu?).

    I've started the same thread at w.o.s.
    Here's the link, in case you want to follow both threads.

    http://www.worldofspectrum.org/forum...ad.php?t=19932


  13. Re: Speccy loading noise?

    zxbruno wrote:

    > Quoting from a post made by Gasman on w.o.s.
    >
    > "Here's my attempt at making sense of it:


    [snip]

    > I think we may be looking at Sabre Wulf again."


    He's right. Listen to this - Speccy in left channel, Wire clip in the
    right.

    http://www.divshare.com/download/4076262-791

    This was from a Speedlocked version of Sabre Wulf which I had to pitch
    down somewhat... I assume the sample on The Wire was from the original
    pre-Speedlock release.

    Lee

  14. Re: Speccy loading noise?

    On Mar 21, 10:13 am, qaopm wrote:
    > zxbruno wrote:
    > > Quoting from a post made by Gasman on w.o.s.

    >
    > > "Here's my attempt at making sense of it:

    >
    > [snip]
    >
    > > I think we may be looking at Sabre Wulf again."

    >
    > He's right. Listen to this - Speccy in left channel, Wire clip in the
    > right.
    >
    > http://www.divshare.com/download/4076262-791
    >
    > This was from a Speedlocked version of Sabre Wulf which I had to pitch
    > down somewhat... I assume the sample on The Wire was from the original
    > pre-Speedlock release.
    >
    > Lee


    Thanks!!!!!

    Speccy loading noise on CBS and HBO. Who's next?

+ Reply to Thread