My Latest Hit .....
Post by hitsware on Jun 6th, 2015, 01:33am
ON CLOSE PROC_cleanup : QUIT
ON ERROR SYS "MessageBox", @hwnd%, REPORT$, 0, 0 : PROC_cleanup : QUIT
SYS "midiOutOpen", ^hMidiOut%, -1, 0, 0, 0 TO ret%
IF ret% ERROR 100, "Failed to open MIDI output device"
SYS"midiOutShortMsg",hMidiOut%,(192)+(34<<8)
SYS"midiOutShortMsg",hMidiOut%,(193)+(89<<8)
DIM root(11), drum(15), perc(15): a=0
FOR x=0 TO 11: READ root(x): NEXT x
FOR x=0 TO 15: READ perc(x): NEXT x
FOR x=0 TO 15: READ drum(x): NEXT x
FOR x=0 TO 3
SYS"midiOutShortMsg",hMidiOut%,(153)+(33<<8)+(127<<16)
WAIT 44: NEXT x
REPEAT
FOR x=0 TO 11: IF root(x)=a THEN 10
SYS"midiOutShortMsg",hMidiOut%,(177)+(123<<8)+(0<<16)
SYS"midiOutShortMsg",hMidiOut%,(145)+(root(x)+69<<8)+(50<<16)
SYS"midiOutShortMsg",hMidiOut%,(145)+(root(x)+64<<8)+(50<<16)
10 FOR y=0 TO 7: z=((8*x)+y) MOD 16
SYS"midiOutShortMsg",hMidiOut%,(153)+(perc(z)<<8)+(80<<16)
SYS"midiOutShortMsg",hMidiOut%,(153)+(drum(z)<<8)+(127<<16)
SYS"midiOutShortMsg",hMidiOut%,(144)+(root(x)+33<<8)+(90<<16)
WAIT 22
SYS"midiOutShortMsg",hMidiOut%,(176)+(123<<8)+(0<<16)
a=root(x): NEXT y: NEXT x
UNTIL FALSE: END
DEF PROC_cleanup
hMidiOut% += 0 : IF hMidiOut% SYS "midiOutClose", hMidiOut%
ENDPROC
DATA 7,5,0,7, 0,0,0,0, 5,5,0,0
DATA 0,0,0,0, 0,0,64,64, 0,0,0,0, 0,0,63,63
DATA 35,0,76,35, 35,0,0,0, 35,0,76,35, 35,0,0,0
Re: My Latest Hit .....
Post by rtr2 on Jun 8th, 2015, 10:47am
on Jun 6th, 2015, 01:33am, hitsware wrote: Code: IF root(x)=a THEN 10
SYS"midiOutShortMsg",hMidiOut%,(177)+(123<<8)+(0<<16)
SYS"midiOutShortMsg",hMidiOut%,(145)+(root(x)+69<<8)+(50<<16)
SYS"midiOutShortMsg",hMidiOut%,(145)+(root(x)+64<<8)+(50<<16)
10 |
|
I think you should win a prize for the most unnecessary use of an (implied) GOTO! To eliminate it just reverse the test:
Code: IF root(x)<>a THEN
SYS"midiOutShortMsg",hMidiOut%,(177)+(123<<8)+(0<<16)
SYS"midiOutShortMsg",hMidiOut%,(145)+(root(x)+69<<8)+(50<<16)
SYS"midiOutShortMsg",hMidiOut%,(145)+(root(x)+64<<8)+(50<<16)
ENDIF
Remember, GOTO (with a line number destination) is by design incredibly slow in BBC BASIC. It works by searching the program line-by-line, from the start, until the specified line is found - even if the program is tens of thousands of lines long! For the same reason, GOTO does not work at all in a library or CALLed module. Just don't do it. Ever. :)
Richard.
Re: My Latest Hit .....
Post by hitsware on Jun 9th, 2015, 10:49pm
>I think you should win a prize
>for the most unnecessary use
>of an (implied) GOTO!
Duh ...........
And the prize is ? ! ?
Re: My Latest Hit .....
Post by hitsware on Jun 12th, 2015, 10:47pm
Code:
CALL @lib$+"HQSOUND"
*TEMPO 133
ENVELOPE 1,0,0,0,0,0,0,0,63,0,0,0,40,0
ENVELOPE 2,0,0,0,0,0,0,0,127,0,0,-2,100,0
ENVELOPE 3,0,0,0,0,0,0,0,127,-127,0,0,126,0
ENVELOPE 4,0,0,0,0,0,0,0,0,0,0,0,0,0
ENVELOPE 5,0,0,0,0,0,0,0,127,-1,0,0,60,0
DIM orgn(11), bass(7), drum(7)
FOR x=0 TO 11: READ orgn(x): NEXT x
FOR x=0 TO 7: READ bass(x): NEXT x
FOR x=0 TO 7: READ drum(x): NEXT x
FOR v=0 TO 3: SOUND v,4,0,30: NEXT v
FOR x=0 TO 3: FOR y= 0 TO 3
SOUND y,5,148,8: NEXT y: NEXT x
REPEAT
FOR x=0 TO 11: FOR y=0 TO 7
SOUND 3,1,orgn(x)+48,4
SOUND 1,1,orgn(x)+76,4
SOUND (4098-(bass(y)*4096)),2,orgn(x),1:SOUND 4098,2,0,3
SOUND 0,drum(y),140,4
NEXT y: NEXT x
UNTIL FALSE: END
DATA 32,24,4,32, 4,4,4,4, 24,24,4,4
DATA 1,0,0,1, 1,0,0,0
DATA 4,4,3,4, 4,4,3,4
Re: My Latest Hit .....
Post by hitsware on Jun 14th, 2015, 10:07pm
Code:
ON CLOSE PROC_Cleanup: QUIT
ON ERROR PROC_Cleanup: REPORT : END
SYS "midiOutOpen",^hMidiOut%,-1,0,0,0 TO ret%
IF ret% ERROR 100,"Failed to open MIDI output device"
DIM c(19),n(3),d(7)
FOR x=0 TO 19: READ c(x): NEXT x
FOR x=0 TO 3: READ n(x): NEXT x
FOR x=0 TO 7: READ d(x): NEXT x
DIM p(3),v(3)
p(0)=034: v(0)=120
p(1)=107: v(1)=080
p(2)=108: v(2)=080
p(3)=050: v(3)=050
dv=127
REPEAT
FOR x=0 TO 19
n2=45+INT(LOG(c(x)*4/20)*(12/LOG(2)))-5
n3=45+INT(LOG(c(x)*6/20)*(12/LOG(2)))-5
SYS"midiOutShortMsg",hMidiOut%,(193)+(p(3)<<8)
SYS"midiOutShortMsg",hMidiOut%,(145)+(n2<<8)+(v(3)<<16)
SYS"midiOutShortMsg",hMidiOut%,(194)+(p(3)<<8)
SYS"midiOutShortMsg",hMidiOut%,(146)+(n3<<8)+(v(3)<<16)
FOR y=0 TO 3: dc=((4*x)+y)MOD 8
r=RND(3)-1:o=2^r: jn=o*c(x)*n(y)
n1=45+INT(LOG(jn/40)*(12/LOG(2)))-5
SYS"midiOutShortMsg",hMidiOut%,(153)+(d(dc)<<8)+(dv<<16)
SYS"midiOutShortMsg",hMidiOut%,(192)+(p(r)<<8)
SYS"midiOutShortMsg",hMidiOut%,(144)+(n1<<8)+(v(r)<<16)
WAIT 22
SYS"midiOutShortMsg",hMidiOut%,(153)+(d(dc)<<8)+(00<<16)
SYS"midiOutShortMsg",hMidiOut%,(144)+(n1<<8)+(00<<16)
NEXT y
SYS"midiOutShortMsg",hMidiOut%,(145)+(n2<<8)+(00<<16)
SYS"midiOutShortMsg",hMidiOut%,(146)+(n3<<8)+(00<<16)
NEXT x: UNTIL FALSE
DEF PROC_Cleanup
IF hMidiOut% SYS "midiOutClose", hMidiOut%
ENDPROC
DATA 16,12,09,12, 12,12,09,12, 16,12,09,09, 12,12,16,12
DATA 16,12,18,12
DATA 2,3,4,5
DATA 35,00,76,00, 00,35,76,00
Re: My Latest Hit .....
Post by hitsware on Jun 16th, 2015, 1:12pm
Code:
CALL @lib$+"HQSOUND"
*TEMPO 133
ENVELOPE 1,0,0,0,0,0,0,0,63,0,0,0,40,0
ENVELOPE 2,0,0,0,0,0,0,0,127,0,0,-2,100,0
ENVELOPE 3,0,0,0,0,0,0,0,127,-127,0,0,126,0
ENVELOPE 4,0,0,0,0,0,0,0,0,0,0,0,0,0
ENVELOPE 5,0,0,0,0,0,0,0,127,-127,0,0,70,0
DIM orgn(29), bass(7), drum(7), octv(7)
FOR x=0 TO 29: READ orgn(x): NEXT x
FOR x=0 TO 7: READ bass(x): NEXT x
FOR x=0 TO 7: READ drum(x): NEXT x
FOR x=0 TO 7: READ octv(x): NEXT x
SOUND 0,4,0,40
SOUND 1,4,0,40
SOUND 2,4,0,40
SOUND 3,4,0,40
FOR x=0 TO 3: FOR y=0 TO 3
SOUND y,5,1056,10: NEXT y: NEXT x
REPEAT
FOR z=0 TO 119: x=z DIV(4): y=z MOD(8): v=z MOD(4)
SOUND 3,1,FNfreqout(orgn(x)*22*octv(y)/2),5
SOUND 1,1,FNfreqout(orgn(x)*33*octv(y)/2),5
SOUND (4096-(bass(y)*4096)),2,FNfreqout(orgn(x)*11/2),1:SOUND 4096,2,0,4
IF v=2 THEN SOUND 2,3,FNfreqout(1056),5 ELSE SOUND 2,5,FNfreqout(RND(2)*180),5
NEXT z: UNTIL FALSE: END
DATA 10,12,9,10, 10,12,9,9, 10,12,9,10, 8,9,10,10
DATA 8,10,9,9, 8,10,9,9, 8,10,9,9, 8,9
DATA 1,0,0,1, 1,0,0,0
DATA 4,4,3,4, 4,4,3,4
DATA 1,1,2,2, 1,2,2,1
DEF FNfreqout(f)
LOCAL I% : PRIVATE ftab%, indx&
IF ftab% = 0 THEN
LOCAL base%
SYS "GetModuleHandle", 0 TO base%
FOR I% = base% TO base% + 65534 STEP 2
IF !I% = &03550354 IF I%!4 = &06060606 EXIT FOR
NEXT
IF I% > base% + 65534 ERROR 100, "Cannot locate frequency table"
ftab% = I% + 8
SYS "VirtualProtect", ftab% AND -&1000, &2000, &40, ^I% TO I%
IF I% = 0 ERROR 100, "Cannot make memory image writable"
ENDIF
indx& += 1 : IF indx& = 0 indx& = 1
I% = 2 * indx&
ftab%!I% = ftab%!I% AND &FFFF0000 OR INT(f * &10000 / 22050 + 0.5)
= indx&
Re: My Latest Hit .....
Post by dfeugey on Jun 16th, 2015, 5:56pm
Really very good 
Re: My Latest Hit .....
Post by hitsware on Jun 16th, 2015, 9:26pm
on Jun 16th, 2015, 5:56pm, dfeugey wrote:Really very good
|
|
Thank You ! 
You are obviously a person of rare taste and refinement 
Tweaked some here :
https://home.comcast.net/~mnjmiller/bbcgyp.bbc
Re: My Latest Hit .....
Post by David Williams on Jun 16th, 2015, 9:44pm
on Jun 16th, 2015, 5:56pm, dfeugey wrote:Really very good
|
|
I used two of hitsware's algorithmically-generated tunes in one of my games, SubZap II:
http://www.jeroengroenendaal.com/repository/bb4w/progs/zip/subzap2.zip
I'd say the music makes the game worth playing!
hitsware: I haven't listened to your latest hit yet (since I haven't downloaded the library HQSOUND), but I did listen to the previous MIDI-based one, and I thought it was excellent.
David.
--
http://www.proggies.uk
Re: My Latest Hit .....
Post by hitsware on Jun 16th, 2015, 10:21pm
WOW !
Thank You Dave. I had forgotten about that. That piece was done with ZEL .....:
http://zelsoftware.org/
A main difference between BBCmidi and ZELmidi is that ZEL produces midi files. I.E. songname.mid, while BBC addresses the synth directly. I hadn't yet discovered BBCbasic at that time. I prefer the BBC approach because the tunes can go on forever as opposed to the finite length of a midi file ...........
ZEL is worth a look though ..........
Re: My Latest Hit .....
Post by rtr2 on Jun 17th, 2015, 1:49pm
on Jun 16th, 2015, 9:44pm, David Williams wrote:| I haven't listened to your latest hit yet (since I haven't downloaded the library HQSOUND) |
|
HQSOUND is not a requirement, it is simply a way of improving the quality.
Incidentally over at the Yahoo! group (but not at the direct link I gave) HQSOUND has been updated to v1.2 which supports 127 amplitude steps, rather than 16, to reduce the problem of being able to hear the steps in a slow amplitude envelope.
Richard.
Re: My Latest Hit .....
Post by hitsware on Jun 17th, 2015, 6:47pm
If I rename the 2 versions, put them both in LIB, call 1 from a routine, run the routine, then call 2 and run the routine .......... Will the second version over-write the first ?
......In order to test the effectiveness.........
Re: My Latest Hit .....
Post by hitsware on Jun 17th, 2015, 10:47pm
1) No matter the order entered I get an error on second iteration
2) Must listen some more. At first I was hearing clearly the noise I speak of, but when I simplified the routine as much as possible it seemed to go away (both HQ versions) Seems perhaps other things can aggravate the problem.
Code:
*TEMPO 133
ENVELOPE 1,0,0,0,0,0,0,0,127,0,0,-1,127,0
CALL @lib$+"HQS_1"
FOR x=0 TO 7
READ n
SOUND 0,1,n,1: SOUND 4096,1,0,10
NEXT x
RESTORE
CALL @lib$+"HQS_2"
FOR x=0 TO 7
READ n
SOUND 0,1,n,1: SOUND 4096,1,0,10
NEXT x
END
DATA 52,60,68,72,80,88,96,100
Re: My Latest Hit .....
Post by rtr2 on Jun 18th, 2015, 05:40am
on Jun 17th, 2015, 6:47pm, hitsware wrote:| In order to test the effectiveness |
|
So long as you run them in separate sessions of BB4W (i.e. separate processes) they will of course be entirely independent. That's how I have been comparing different versions; commonly I have three or more copies of BB4W running simultaneously.
The alternative is to compile your programs and run the EXEs, which of course also ensures that they are independent.
Here's a good test of the difference (also worth trying without the HQSOUND library at all); it is very important that you listen with headphones:
Code: ENVELOPE 1,5,0,0,0,0,0,0,1,-1,-1,-1,126,0
SOUND 1,1,148,252
Richard.
Re: My Latest Hit .....
Post by hitsware on Aug 8th, 2016, 12:58am
REM:m = 12*log2(fm/440 Hz) + 69
REM: key(),D(144), C(160), A(96), E(128), G(108)
key=144: DIM mn(700)
FOR jn=1 TO 700
mn(jn)=69+INT(12*LOG((jn)/key)/LOG(2)+0.5)
NEXT jn: PROC_midistart
FOR ch=0 TO 4:READ pa,le,pn
PROC_env(ch,pa,le,pn): NEXT ch
DATA 034,127,064, 107,060,127, 11,100,0
DATA 127,064,000, 127,127,064
DIM cm(29),cn(3),cr(3),da(7),db(7),d1(7),d2(7)
FOR x=0 TO 29:READ cm(x):NEXT x
FOR x=0 TO 03:READ cn(x):NEXT x
FOR x=0 TO 07:READ da(x):NEXT x
FOR x=0 TO 07:READ db(x):NEXT x
DATA 5,6,9,5, 5,6,9,9, 5,6,9,5, 8,9,5,5
DATA 8,9,5,5, 8,9,5,5, 8,9,5,5, 8,9
DATA 2,4,3,4
DATA 035,000,076,000, 035,035,076,000
DATA 042,000,042,042, 042,000,042,042
REPEAT
FOR x=0 TO 29:FOR y=0 TO 3
rr=INT(RND(3)-1): oc=2^rr
nn= mn(cm(x)*cn(y)*oc)
PROC_playnote(rr,nn)
NEXT y: NEXT x: UNTIL FALSE: END
DEF PROC_playnote(rr,nn)
SYS"midiOutShortMsg",hMidiOut%,(144+rr)+(nn<<8)+(127<<16)
WAIT 22
SYS"midiOutShortMsg",hMidiOut%,(144+rr)+(nn<<8)+(000<<16)
ENDPROC
DEF PROC_env(ch,pa,le,pn)
SYS"midiOutShortMsg",hMidiOut%,(192+ch)+(pa<<8)
SYS"midiOutShortMsg",hMidiOut%,(176+ch)+(07<<8)+(le<<16)
SYS"midiOutShortMsg",hMidiOut%,(176+ch)+(10<<8)+(pn<<16)
ENDPROC
DEF PROC_midistart
ON CLOSE PROC_Cleanup: QUIT
ON ERROR PROC_Cleanup: REPORT : END
SYS "midiOutOpen",^hMidiOut%,-1,0,0,0 TO ret%
IF ret% ERROR 100,"Failed to open MIDI output device"
ENDPROC
DEF PROC_Cleanup
hMidiOut% +=0 :IF hMidiOut% SYS "midiOutClose", hMidiOut%
ENDPROC
Re: My Latest Hit .....
Post by michael on Aug 8th, 2016, 4:16pm
nice mix.