902 2015-09-21 09:45:49
Topic: 1-Bit Forum Compo: Results (10 replies, posted in General Discussion)
Ok, here are the results.
1-BIT CLASSIC
1 P.F.M. by AtariTufty...............................41
2 new era by irrlicht project........................39
3 Vanvidd by garvalf.................................34
4 my name in neon lights by crash override^hackers...22
1-BIT 1K
1 zrx by irrlicht project............................43
2 Boiled Down by AtariTufty..........................30
3 Water Beeplets by AtariTufty.......................27
4 acid nipple by Byte.Clone..........................24
1-BIT ALTERNATIVE
1 Firabys by garvalf.................................43
2 Glitch! by AtariTufty..............................26
3 Moving Day by krüe.................................25
A total of 5 people voted. As promised, I didn't vote myself as I know the names, but my top choices would have been P.F.M., Water Beeplets, and Moving Day (seriously guys, this one is awesome, just listen with bass properly turned up).
Well, thanks to everybody who participated. Due to low turnout I'm afraid I won't repeat this experiment any time soon.
903 2015-09-21 09:10:18
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Well spotted, thanks. This concerns nanobeep, too. I've updated the docs.
904 2015-09-20 20:41:15
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Cheers guys, I'd love to hear some tunes made with these
There's still some more concepts I want to explore. For example recently I learned a new trick for speeding up sample playback by quite a bit which isn't implemented in either qaop or yawp. So higher quality digi players are definately possible. Also I would love to try and pull off some sort of fake FM synthesis, though that's currently still a bit over my head.
905 2015-09-20 14:09:45
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Ok, one last engine release...
Day 7: nanobeep
- 2 channels of PFM sound
- ~3 octaves range
- 8-bit frequency resolution
- 1 interrupting click drum
- player size 86 bytes
example tune
download (includes XM converter, surprise surprise)
source
Not the most useful engine on the planet, but hey, I had to fill up the 7 days of the week somehow This is my attempt at making an engine with a size below 100 bytes. As the counters are 8-bit, there are of course detuning issues, but unlike with Huby, they occur at both ends of the note range. Notes from the middle of the first octave onwards to mid-4th should be reasonably stable though. Also, a few bass notes (C, D, G) are safe to use.
The keyhandler doesn't check the whole keyboard (only Space,A,Q, and 1), if you find this to be unacceptable it'll cost you 2 additional bytes to fix that.
And thus ends the "7 days, 7 engines" project. Thanks everybody for checking it out
906 2015-09-19 13:20:17
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Getting closer to the finish line of this code run...
Day 6: yawp, aka Yet Another Wave Player
- 3 channel PCM WAV playback
- uses looped 256 byte samples
- 2 bit sample depth, mixed to 4¼-bit output (silence + 9 volume levels)
- 5 octaves note range
- 16-bit frequency resolution
- per-row speed control
- 16.2 KHz mixing
example tune
download (includes the usual converter)
source
Why settle for 2 channels of sample playback when you can have 3? Sound is not as good as in qaop, but still less fuzzy than SampleTracker's output. The 3 channels have slightly different volume levels, and are also not 100% equal in sound.
907 2015-09-18 20:20:55
Topic: A Timeline of 1-Bit Music (1949-1979) (10 replies, posted in General Discussion)
In the 1840s, Ada Lovelace, who studied the relation of maths and music among other things, predicted that one day computers would be used to make music. And right she was.
This timeline provides an overview over the history of 1-bit music (including indirect forms such as Radio Frequency Interference music) from the beginning of the electronic computer era to the advent of affordable microcomputers. It inevitably also gives a glimpse into the the history of computer music in general. In current literature, the early days of computer music are commonly being reduced to a few academic experiments, when actually an active and diverse, though mostly non-professional, computer music scene has existed ever since the advent of digital computers.
If you know about any 1-bit music activity in the years 1949-1979 which is not listed here, or if you have any more detailed knowledge about the events in the list, I would be very pleased to receive your input.
A Timeline of 1-Bit Music
including AM radio interference music
Precursors
1676-1681
Based on conversations with musical theorist William Holder, English scientist Robert Hooke devises a mechanical apparatus that produce tones of varying pitch by continually triggering clicks with a toothed wheel. In the first half of the 19th century, French physicist Félix Savart took up the idea and refined the mechanism. His machine became known as the Savart wheel. source
1799
Scottish natural philosopher John Robinson describes a method of producing sound by opening and closing a pneumatic tube at regular intervals. His musical siren is later refined by Baron Charles Cagniard de la Tour, and to this day all mechanical sirens work based on this principle. source
1940s
1949
Frances E. "Betty" Holberton programs the BINAC to play "For He's A Jolly Good Fellow" to the team who built the machine at the Eckert-Mauchly Computer Corporation in Philadelphia, PA. It is not known for sure on which date it happened, most likely on the occasion of the final verification testing on April 7th. source.
1950s
1950
Alan Turing outlines the theoretical basis for 1-bit music in his manual for the Manchester Mark I computer. source
1951
US Air Force Lieutenant Herb Finney writes a sound routine for the UNIVAC (serial #2) in early 1951, and makes the computer play "The Blue Tango", "The Air Force Song", "The Eyes of Texas", and more via the system speaker (which is not connected to the computer directly, but rather picks up AM radio interference). The program is said to have been distributed with every UNIVAC installation. source, source, source
Frances E. "Betty" Holberton programs the UNIVAC I (serial #1, the one for the US Census Bureau) to play music via AM radio interference at the official "dedication" event of the machine at the Eckert–Mauchly factory in Philadelphia, on either June 14th or possibly March 31st. As of now, this is the first confirmed public performance of computer music in history. Reportedly, the program by Herb Finney was written before the on by Betty Holberton. source
Geoff Hill programs the CSIR Mk1 (CSIRAC) at the University of Sydney to play 1-bit music via the attached hooter. The music is performed in public on August 7th-9th. source, reconstructed recording, more recordings
On December 7th, the BBC records the Ferranti Mark I at the University of Manchester playing 1-bit music. Programming by Christopher Strachey. The recording was previously said to have been done on September 7th, but more recent research indicates that this is incorrect, and Strachey only completed the program around September 27th. There is also a second, unofficial version of the recording which was cut on acetate disk at the request of a certain Frank Cooper. This is the oldest surving original recording of 1-bit/computer music. source/recording, recording (F. Cooper version), source, source (F. Cooper interview), program analysis, further notes
On December 16th, CBS broadcasts an interview with Jay W. Forrester, leader of the team at MIT that built the Whirlwind I computer. Among other things, Forrester demoes the machine playing some music.
1955
Norman Hardy and Ted Ross write a music program for the IBM 701. For this purpose, a control lamp on the console was rewired to the systems' speaker. According to Norman Hardy the music was even recorded on vinyl, but it seems the record is lost. source
1956
Morton Bernstein writes the JOHNNIAC Music Assembler. The program even makes it into the official JOHNNIAC software library. source, source, source, source (p 15f.) Possibly predated by music on the SEAC. source
1957
After the CSIRAC computer is transfered to Melbourne in 1955, Tom Cherry continues where Geoff Hill left off. Cherry writes an improved music driver that interprets score-based input. source (p46 f.)
Chris P. Burton writes a music program for the Ferranti Perseus. It even accompanies the speaker sounds with rhythmic tape loading clicks. (Perhaps the first occurance of interrupting click drums?) According to the author, it was a tradition for Ferranti maintenance engineers to write these programs, and they existed on all Ferranti machines. Perhaps unsurprisingly so, as most of them were designed by a certain Christopher Strachey. source, source
Max Mathews writes a music compiler for the IBM 704 at Bell Labs. He asks Newman Guttman to compose a tune for it, who goes on to write The Silver Scale. This 17-second piece, performed at Bell Labs in May 1957, may be the world's first original 1-bit composition. source, source, recording
1958
At the beginning of the year, a trio consisting of Bob Cooper on oboe, Howard Rumsey on bass viol, and a Bendix G-15 perform Bach at the Bendix Computer plant in Los Angeles. source
Janet Norman plays ‘Noel’ on a Bendix G15 - live on television. recording, description of the "Users' Project No. 41" routine
The SARA computer at Linköping, Sweden plays "Oxdragarsången" by Ever Taube. The SARA is an IAS machine like the JOHNNIAC, though not code compatible with the latter. source
Service technicians have the German Zuse Z22 computers play German folk and popular music. A music program (Programm 47) was even officially distributed by Zuse. source, recording, source code for a Z23 music routine ("Programm 1076"?), reconstruction (Z23)
John Kamena programs the UNIVAC 1103 at the Southern Methodist University in Dallas, TX to play Christmas songs. source
John Nussbaum records Bach played by an IBM 704 computer. Dating uncertain. recording
1959
The MIT Lincoln Lab TX-2 plays music. The machine had a rather large ferrit core memory and two high-quality console speakers, which together provided for about 90 seconds of stereo music. source, source, source
The IBM 704 has a monophonic music program. Keith Reid-Green converts "The Wedding March" for a friend who is about to marry, but the bride doesn't approve. source, source
195x
An unknow programmer at UNIVAC's headquarters in Philadelphia, PA, programs 'Melody Maker' for the UNIVAC 1103. The exact date remains unknown, but according to source it was prior to John Kamena's work in 1958. source
The IBM AN/FSQ-7 is introduced in 1958. According to Doug Elliott, it was also used to play Christmas music, though the exact year remains unknown. Note that the IBM AN/FSQ-7 (aka Whirlwind II) is a direct descendant of the Whirlwind I, which already played music in 1951. source
Music on the Pilot Model ACE, and later on the regular ACE machines. The exact year is unknown, must have been between 1950 (first running tests) and 1955 (Pilot ACE is scrapped). source, source
The English Electric DEUCE (commercial version of the Pilot Model ACE, delivered in 1955) programming manual list a command for sounding the computer's buzzer. Needless to say, the usual course of events ensued. Several people wrote music routines for the machine, among them Denis Brockington, John "Speedy" Denison, Harold Fineberg (who made a piano program which was controlled by console switches), and Richard Young (who even accompanied the machine on his trombone). source
A music program existed on the Ferranti Pegasus, author unknown. The exact year is unknown, probably 1958 or 1959. binary+emu, demonstration of (possibly unrelated) Pegasus music
The IBM NORC military computer plays music during the yearly "open day" of the lab at Dahlgren, VA. (This puts the first possible date at 1955). source
The ARMAC at the Mathematisch Centrum in Amsterdam is used to play the Dutch national anthem whenever a member of the royals would visit. source, source (translated)
The Dansk Aritmetisk Sekvens Kalkulator (DASK) in Kopenhagen, Denmark plays music through the speaker built into the control panel. source
The Telecommunications Research Establishment Automatic Computer (TREAC) in Malvern, England apparantly played music. Sources are very vague, so more research is needed. source source
Gernot Metze programs the ILLIAC I to play music. Exact date unknown, must be between 1953-55. source
1960s
1960
Peter R. Samson develops Music X, a monophonic 1-bit music compiler for the TX-0 (a machine based on the earlier IBM AN/FSQ-7 resp. Whirlwind II). He later wrote another 1-bit compiler that allowed for light pen-controlled editing of music. According to rumour he later expanded that software to 3 channels. source, source, source, source code.
The first CDC 1604 is delivered to the US Navy with a music program written by Charles "Chuck" L. Hawley. The program operates on the machine's built-in 3-bit DAC which was connected to a tube amplifier. source (p 26)
The British LEO III computer performs 1-bit music during a visit to the manufacturer's offices by the Duke of Edinburgh. source, recording
1961
LaFarr Stuart programs the CYCLONE mainframe at Iowa State University to play 1-bit music. The CYCLONE is a clone of the previously mentioned JOHNNIAC, by the way. source, source
Building on his earlier work on the IBM 701, Norman Hardy programs the IBM 7090 to play 3-voice beeper music. source
1962
LaFarr Stuart presents his 1-bit music on the NBC radio network program Monitor. source
Peter R. Samson develops the 4-channel 'Harmony Compiler' for the PDP-1 (the successor of the TX-0). It is later used by Bill Ackerman, Dan Smith, and others to transcribe various pieces of classical music and more. The year is often incorrectly stated as 1964, but the source code reveals the date 10/6/1962. The music is output via speakers connected to 4 flip-flops, which are in turn connected the machine's "control flag" lamps. Some additional filtering is applied to the speaker lines, achieving a rather pleasing sound (so it's a kind of "1-bit plus"). The music is later even recorded on vinyl. source, source, recordings: 1, 2, 3, 4, source code and more, video presentation, recording of the vinyl, new presentation by P. Samson, with technical explanation (2018)
"Rekengeluiden van PASCAL", a 7", 45rpm vinyl with sounds and music from the Philips Automatic Sequence Calculator (aka Philips Akelig Snelle Calculator), is released by Philips' in-house magazine, the "Technisch Tijdschrift". source, recording, discogs
The IBM 704 at the IBM Labs at Mohansic, NY plays 4-voice Christmas music, using accumulator lines connected to a speaker. This is unrelated to the work by Max Mathews, which used a more complex DAC setup. source
IBM's official software catalogue mentions two music programs: MUSIC for IBM 705, written by R. W. Bremer, W. M. Selden, and A. S. Petroulakis at IBM HQ, and "Computer Automated Music" for IBM 650, written by Norman V. Plyer at Univ. of Rochester. The latter appearantly supports percussion via an IBM 407's card puncher. source (p. 181, 241)
A music program is available for the Danish Regnecentralen GIER in late 1961 or early 1962. By 1971, multiple advanced music programs exist for the machine. source, source, recordings, software
Seppo Mustonen develops a music program for the Elliot 803 computer at the University of Helsinki, Finland. The program generates a random score and plays it back in realtime, possibly the first program ever to do so. source, recording
Honeywell EDP publishes a 7" with 1-bit Christmas music played by the Honeywell 800 computer. source, recording
1963
Instead of learning FORTRAN, high school kids at LA City Schools prefer to load up the local IBM 1620 with RFI music programs. source
In November/December, the Regnecentralen GIER plays music at an exhibition in Warsaw, which is also broadcast on TV. This is currently the earliest documented incident of a computer playing music in Eastern Europe. source
T.H. O'Beirne in Glasgow starts to explore the musical capabilities of the Barr&Stroud SOLIDAC mini-computer, giving a first public demonstration in 1965. Despite the machine effectively running at just 30 KHz, he even managed to squeeze some polyphony out of it. O'Beirne published several papers on the subject. He also creates an interactive music program named ORPHEUS. In 1967, Barr&Stroud releases an LP with various permutations of Mozart's Musical Dice Game. source, source, source, recordings, source/recordings
The Laboratory Instrument Computer (LINC) plays music. video of the machine in action in 2007
1964
The Telefunken TR4 mainframe at the Office of Finances in Düsseldorf, DE, is (ab)used to play RFI music source
1965
At its shutdown ceremony, the EDSAC 2 computer plays "The Last Post", using the Music Compiler written by Richard Jennings (probably in 1962/63) source, source
An IBM 1620 Data Processing System at Arlington State College (now University of Texas at Arlington) in Arlington, Texas, USA, is filmed playing "Jingle Bells". video
1966
Peter Samson's 4-voice player MUSCOM ported to PDP-6. Later expanded to 6 channels (1-bit, with optional 4-voice DAC output) as BIG Music Processor, also by P. Samson. source source, source
EUTERPE-LISP, a LISP engine producing 6-voice music on PDP-6 with DEC Precision Display Type 340 as DAC. By Marvin Minsky, for MIT/Project MAC. EUTERPE's output is compatible with Samson's MUSCOM. source, source
1967
Göran Sundqvist records 1-bit music on the SAAB D21 computer. Seems the technique was already applied to the machine's prototype, the D2. Later a complex DAC was added, thus leaving the world of 1-bit music behind. The recordings were released on vinyl in 1970. source, recordings, source, discogs
4-voice music routine on PDP-7, by Ronald F. Brender.source code, binaries
Pietro Grossi at his Studio Di Fonologia Musicale Di Firenze experiments with 1-bit music on the General Electric GE-115. This results in two 7" vinyl releases, GE-115 - Computer Concerto (which even gets two re-releases in Belgium and Sweden, respectively), and Buon Natale 1967 e Felice Anno Nuovo, which was originally only released internally to employees of Olivetti, but later got re-released along with Pietro Grossi's 1972 album Computer Concerto (see the entry for 1972). 1-bit music in Italy very likely dates back even further. Grossi himself mentions hearing an Olivetti Elea 9003 at the Monte dei Paschi computer center in Siena play music in 1962 (source), though the details on this are murky.
1969
Nellie the School Computer (actually an Elliott 405 machine) is recorded by the BBC playing music over the build-in buzzer. Chances are the music program already existed at the computer's previous owner, the Nestlé corporation, who donated the machine to the Forest School in 1965. source, recording.
A program for IBM 1401 that plays music on the 1403 line printer is mentioned in the official software catalogue. Written by M. J. Peskin.source
196x
The Elliott 803 is used to play music. Exact year unknown. source, recording, source code (recreated)
The Elliott 900 series of computers is introduced in 1962. Several music programs exist for these systems, though their exact year of creation remains unknown. Andrew Herbert wrote a new player called AHJ Music in 2014. source, source, source, source, source, software, recordings
RFI music on the IBM 1401 in Iceland and in the US. recordings, reconstruction
A music routine for the Ferranti Sirius, by K.C. Johnson and J.E. Thompson. Possibly other routines were made for the machine as well. source, source, source
Music is made on the Swedish Facit EDB 3 computer. One example tune is later released on a commemoratory 7" vinyl by the computers' manufacturer, Industridata AB. source, recording
The SWAC machine at the University of California in Los Angeles plays 1-bit music. The musical history of the SWAC probably dates back to the early 50s, but so far no substantial documentation regarding this has turned up. source
David Parsons programs the EMIDEC 1100 at Barclays' Computer Centre in London to play music over the console speaker, which is demonstrated to visitors of the centre, and supposedly even featured on the BBC. source
Meanwhile, back at EMI Electronics, someone has the ingenious idea of syncing four EMIDEC 1100 together with long coaxial cables and having them play a string quartet by Mozart. source
1970s
1970
Thomas Van Keuren programs the UNIVAC 1050-II at the US Air Force base in Da Nang, Vietnam to play "Ebb Tide" by Robert Maxwell, and more, in full 1-bit glory. This program might be the first that specifically provides for the simulation of chords by fast arpeggiation. recording
1971
Two RFI music programs for IBM 1620, made by Laura B. Steele resp. Ron Davis, are mentioned in the official software catalogue.source
1972
Pietro Grossi, now working at the National University Computing Center (CNUCE) in Pisa, releases Computer Concerto, a double LP of music made on IBM System/360 and IBM System/1800. Part renditions of Bach and Paganini, part own compositions, he created the music using three tools he made himself, ATP and DCMP for the 360, and PLAY1800 for the 1800. The release also contains Pietro Grossi's two 7" releases from 1967. source, recording (note the sleeve notes)
1974
RFI music routine for the Altair 8080 by Paul Mork. source
1975
Another Altair RFI music routine, this time for the 8800. Made by Steve Dompier, who also demonstrated the program live for the Homebrew Computer Club on April 16th, 1975. The routine takes a mere 28 bytes. source, source code, reconstructed recording, source, source code (better), recording (2009), recording (new songs, 2018)
Richard Wilson programs a 4-voice RFI music player for the PDP-8. recording, recording, recording, recording, binaries, more binaries and sample tracks, source
1977
The "Music System" by Software Technology Corp. (aka "Software Music Synthesis System" by California Software Co.) is released. Developed by Jon Bokelman, the Music System consist of a simple S-100 DAC board and additional driver software, which could be installed on a wide range of 8080 and Z80 based computers. It outputs 3-4 channels of 1-bit music depending on the host system and configuration, making it one of the first true multi-channel 1-bit solutions. software, recording (SOL-20), recording (Altair 8800), manual, manual (SMSS)
197x
908 2015-09-18 13:18:20
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Ah, thanks Tufty, I'll fix the links right away. In the meantime...
Day 5: Octode PWM
- 8 channels of square wave sound, sort of
- 16-bit frequency resolution
- 4 octaves note range
- 3 interrupting click drums
- per-row speed control
- 19 KHz mixing
example tune
download (yes, includes converter)
source
Another, more experimental version of Octode, with a different sound. It's quite noisy though, I guess mixing 8 square wave channels is a little too much for our beloved Speccy. Well, at least it can serve as a bad example
909 2015-09-17 14:32:55
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Day 4: Octode 2k15
- 8 channels of PFM sound
- 16-bit frequency resolution
- 3 interrupting click drums
- per-row speed control
- 15.6 KHz mixing
example tune (think I overdid the EQing a bit on this one, sorry)
download (includes XM converter of course)
source
I realized I hadn't written a PFM engine in ages, so something needed to be done... Octode 2k15 is a rewrite of Shiru's Octode which fixes the detuning issues of the original by using 16-bit counters. The sound has also been improved, and resembles that of Octode XL with all channel volumes set to like ~5-6 or so. Unlike with Octode XL, volumes can't be changed though. To somewhat make up for it, I added per-row speed control.
Another downside is that the song data is huge, so the maximum song length would be about thirty 64-step patterns. That goes for yesterday's engine, too, btw.
You can feed original Octode XMs into Octode 2k15, but you'll need to adjust speed and pitch. XMs from Octode XL will not be compatible, though.
910 2015-09-16 19:11:32
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Ouch... xm template added. I have a hard time staying focussed these days
911 2015-09-16 13:04:40
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Eyyup, here we go again:
Day 3: xtone
- 6 channels of square wave sound
- limited duty cycle control
- 16-bit frequency resolution
- 3 interrupting click drum sounds
- per-row speed control
- 16.2 KHz mixing
example tune
download (includes XM converter as usual)
source
Now we're getting serious xtone is basically like Tritone, but with 6 channels. Inevitably, the sound is quite a bit grittier though, and duty cycle
control is less accurate, though you still get at least 4 different settings.
912 2015-09-16 11:51:43
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Yo Tufty, here's the .tap for the qaop demo track. It was hand-coded in asm so there's no XM for this one
913 2015-09-16 11:21:21
Re: Mewtone (WIP) (7 replies, posted in Sinclair)
What I meant was that it doesn't matter if you define silence as 0 or 1, because as far as I understand the beeper has a heavy DC offset, ie. it doesn't output 0V in either 0 or 1 state. So neither 0 or 1 will produce true silence in the actual meaning of the word.
914 2015-09-16 11:15:03
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
Thanks for the cheers, guys
I've found and (hopefully) fixed a few bugs in the qaop and quattropic XM converters, so please re-download.
Also, I've added a standalone executable converter to the quattropic package, so now you can do things without using Perl. However, I just started to learn C++ yesterday, so it's probably loaded with bugs. But I'd still appreciate if you guys gave it a try and let me know if you run into any problems with it.
New engine for today is coming in a couple of hours, stay tuned
915 2015-09-15 12:47:30
Re: Mewtone (WIP) (7 replies, posted in Sinclair)
Ah, the mysteries of the Speccy beeper. I really don't have a clue what's going on here, maybe introspec can shed some more light.
Regarding the duty comparison, actually the other way around (ld a,h, cp duty) is equally correct. Why? Because if H=0, then H-duty will result in carry. Which means the channel will output 1 when muted.
916 2015-09-15 12:41:16
Re: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
A new day, a new engine.
Day 2: quattropic
- 4 channels of square wave sound
- full duty cycle control
- one of the channels can play noise or fast pitch slides instead of regular tone
- said channel also supports note cut
- 16-bit frequency resolution
- 15.6 KHz mixing
example tune
download
source (includes XM converter)
quattropic is the big brother of ntropic - twice as many channels, and you can even vary the sound of the noise. This comes at a small price, however: There is no per-row speed control in this engine. (It could be easily implemented, but would take another 2 bytes of song data per step.)
quattropic is a little easier to simulate in XM, but still you won't get the full range of sounds out of the template, unfortunately.
917 2015-09-14 13:40:58
Topic: 7 days, 7 (new) beeper engines (65 replies, posted in Sinclair)
As promised, I'll go a bit crazy this week (well, more than usual anyway) and release a new beeper engine every day. So let's get right to it
Day 1: qaop, aka Quite Accurate Overdriven Player
- 2 channel PCM WAV playback
- uses looped 256 byte samples
- up to 3 bit sample depth, downmixed to 3-bit output (silence + 6 volume levels)
- 16-bit frequency resolution
- 3 interrupting click drum sounds
- per-row speed control
- 15.6 KHz mixing
example tune
download (includes XM converter, Perl required as usual)
source
This one I'm quite proud of. It basically fixes all the issues in my earlier rawp engine. qaop fast enough to hide the discretion noise, and has stable tuning due to using proper 16-bit frequency counters. And there is a bonus: When mixing two loud enough samples, the engine will overdrive, which will give you some extra crunch, as heard on the kick drums in the example tune.
The downside is that the XM converter is very basic - if you want to use any additional samples (which is the whole point of this engine), you'll have to hack them in yourself.
918 2015-09-14 13:23:29
Re: 1-Bit Forum Music Compo 2015 (12 replies, posted in General Discussion)
Vote page is up!
Voting deadline is Sunday, September 20th.
919 2015-09-14 08:28:09
Re: 1-Bit Forum Music Compo 2015 (12 replies, posted in General Discussion)
The contest is now closed. Thanks to everybody who submitted!
Vote page/pack will go online tonight or tomorrow at the latest.
920 2015-09-13 11:12:59
Re: Mewtone (WIP) (7 replies, posted in Sinclair)
Hmm, sounds pretty good to my ears I especially like that drums and tones are very well balanced, volume wise.
And yes, the "code" part of the compo will indeed most likely not happen due to lack of entries. So I think I'll just release my own entries one by one over the course of the coming week.
I'm curious about your "realtime saw wave" engine, too, sounds like a very interesting concept.
921 2015-09-10 09:31:51
Re: Youtube dump (11 replies, posted in General Discussion)
Hehe thanks. That video is actually a little misleading as the footage is mostly not from lztek Am playing the track live from actual Speccy, the laptop is from the DJ that was playing inbetween and after the liveacts. And yeah, I did play Crystal Realm, too
922 2015-09-07 11:33:53
Re: The optimization thread (3 replies, posted in Sinclair)
I'd go for option 2. Unless you preshift note values like in the example above, you never need bit 7, so you can use that to signal detune. Option 1 would be not so practical, because with 16-bit frequencies, you typically need to recalculate the detune value specifically for every note anyway.
Option 2 then again boils down to two choices. Either you interpret this and the following byte as a direct frequency value (ie. skipping the table lookup), which would limit frequency range to 0-$7fff, or you interpret it as note value, followed by a byte-length detune value, which you'd then multiply with e.g. (hi-byte of the frequency/16) or so to get a decent detune range.
923 2015-09-02 13:13:50
Re: Youtube dump (11 replies, posted in General Discussion)
Ah, so happy Yeah, that's Whittaker alright.
Ok, here's another one - irrlicht project live @ Ohrbit Duisburg: https://www.youtube.com/watch?v=2ljnBXYOr-Y
924 2015-08-31 15:01:52
Re: 1-bit music on Thomson 8-bit computers (3 replies, posted in Other Platforms)
Wow, didn't know this existed. Though one has to wonder why they don't make full use of the 6-bit DAC. But I assume it's like on Dragon/CoCo - generating 1-bit music might simply be faster than using full 6-bit DAC. In fact in the contest, we'll have a CoCo 1-bit routine
925 2015-08-31 14:53:39
Re: 1-Bit Forum Music Compo 2015 (12 replies, posted in General Discussion)
Deadline extended till Sunday, September 13th, because some people asked for more time. Also, we need more works in the "alternative platforms" category!