Printed Amstrad Addict magazine announced, check it out here!
Started by redbox, 19:38, 30 July 13
0 Members and 1 Guest are viewing this topic.
Quote from: arnoldemu on 08:51, 31 July 13Does it mix the sample playing with the music playback routine to get a more correct sample playback?I know that for digidrums it's ok not to mix the samples, but for other sounds it could result in some distortion.
Quote from: arnoldemu on 08:51, 31 July 13Does it mix the sample playing with the music playback routine to get a more correct sample playback?
Quote from: redbox on 11:01, 31 July 13Can you explain what you mean and I'll look at implementing it...?
Quote from: redbox on 11:10, 16 August 13Okay, after looking further I see that both Digitracker and Protracker mix 3 channels of samples into one channel for playback. I expect this is because of the Z80 processor speed limitation (without DMA).So I assume they both use some kind of look-up table (similar to a mask table) to blend the samples? And that this approach could be used with normal AY sounds (e.g. when using digidrums)...?I also see that they both use 4-bit samples at 14.4hz... how is this possible? When I've played samples the CPC could only manage 8hz per frame...?
Quote from: arnoldemu on 11:24, 16 August 13Now, to get it correct, the digi effect should be played on *every* line. The player code then needs to be interleaved so it takes the remaining cycles on the lines it is active on.
Quote from: arnoldemu on 11:24, 16 August 13Next thing I would like to see if you could do it would be to implement the digi play back using DMA
Quote from: arnoldemu on 11:24, 16 August 13It would be nice to have it use AY for 2 channels and 1 channel is digi (drums or otherwise).
Quote from: arnoldemu on 11:24, 16 August 13Next thing I would like to see if you could do it would be to implement the digi play back using DMA That would help with a game I have sitting waiting to be finished off It would be nice to have it use AY for 2 channels and 1 channel is digi (drums or otherwise).
arkos_player.asm - edited Arkos Playerarkos_player_psg.asm - editable file so you can choose the channel you play DMA samples onconvert_sample.asm - converts 8-bit, 15625Hz, Mono, Unsigned, Raw (headerless) PCM into DMA 'sample' formatdma_player.asm - the main player with examples preloaded - just load it into the WinAPE assembler and run (F9)Lxcess_digidrums.aks - Arkos Tracker file for the PC which you can edit (digidrum triggers are in the SpTrk)LXCESSAK.BIN - Arkos Tracker BINary exportsamples/808sn1.pcm - original samplesamples/909Kick.pcm - original samplesamples/sample_dma_808sn1.asm - converted DMA samplesamples/sample_dma_909Kick.asm - converted DMA sample
Quote from: arnoldemu on 17:46, 20 August 13@redbox: Excellent! Exactly what I was after.
Quote from: arnoldemu on 17:46, 20 August 13I would like a pc version of the sample conversion tool though But I can make that.
Quote from: arnoldemu on 17:46, 20 August 13Please can I use this in the MazeZam Plus cart game I mentioned in another thread?
Quote from: Devilmarkus on 16:01, 20 August 13as I'm not a CPC-coder, I'd be happy with a working DSK of this.Could you add it, redbox?
Quote from: tastefulmrship on 15:17, 07 September 13Obviously, all drums can be replaced with tiny samples & a lot of the boring mono-tone tracks can be replaced with SID voice; to properly emulate the original tunes.
Quote from: tastefulmrship on 15:17, 07 September 13I've done some initial stuffs with samples, but when using larger samples (like voices or breakbeats) they were always being played too slowly. I always meant to look into it a lot more... but never got the time.
Quote from: redbox on 18:46, 07 September 13Well it seems the DMA can be used to emulate 'SID voices' quite nicely - there is documentation on the Quasar website. It does sound nice in the Killmax demo and I seem to remember someone on this forum posting something to Soundcloud but I can't remember what now.Still not sure how you'd create the SID voices you wanted though? Unless you rip them from the ST (and even then, how do you get them out of a YM file for example)...?
Quote from: redbox on 18:46, 07 September 13The replay speed is most likely wrong because the sample is at the wrong frequency. If you have any samples you want to play on the CPC, then send them to me in the original format (i.e. WAV etc from PC or IFF etc from Amiga) and I'll convert them to DMA format for you.
Quote from: tastefulmrship on 11:48, 08 September 13Anyway, this is me rambling again... in short; if it's possible to write a tracker/player that has SID sound (for either CPCnorm or CPC+), then I think this community would seriously appreciate that! Top man!
Quote from: tastefulmrship on 11:48, 08 September 13However, from what I gather from the CPC+ DMA player, you can get 4-voices from a 3-voice chip? Is this the case or am I just not understanding? If it is, then almost all .mod files are possible on CPC without external hardware. How cool is that?
Quote from: redbox on 20:37, 08 September 13I had a look at the original Believe MOD and think I know that the problem you're having is.On the Atari ST and CPC, samples are played at one pitch - hence why they are known as 'digidrums' because this type of playback is only really any good for percussion and isn't much use for other instruments which require pitch changes (piano, bass etc).On the Amiga, you can play any sample at any pitch and gotcha is this includes drum sounds. In the MOD you're using the breakbeat is being played at C-6 which is one octave higher than 'standard' pitch of C-5, hence why the exported sample sounds too slow. The 'Believe' vocal is played at C-5 which is the standard pitch and so the sample sounds right.
Quote from: tastefulmrship on 08:34, 09 September 13Ah yeah, my fault... I sent you the original breakbeat .wav file before I had time to resample it to the proper pitch (ie play the sample at C-6 in MODPlug while recording via WaveStudio. Not very high-tech, I know, but it usually works). If I get five minutes after work, I'll attach the new sample here.
Quote from: arnoldemu on 13:00, 09 September 13How are you doing your 8-bit to 4-bit? Are you using a logarithmic table?
convert_4bit_table: defb 1, 1, 1, 1, 2, 3, 3, 4, 5, 5, 6, 7, 7, 7, 7, 8 defb 8, 8, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9, 9, 9 defb 10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,11 defb 11,11,11,11,11,11,11,11,11,11,11,11,11,11,11,11 defb 11,11,12,12,12,12,12,12,12,12,12,12,12,12,12,12 defb 12,12,12,12,12,12,12,12,12,12,12,12,13,13,13,13 defb 13,13,13,13,13,13,13,13,13,13,13,13,13,13,13,13 defb 13,13,13,13,13,13,13,13,13,13,13,13,13,13,13,13 defb 13,13,13,13,14,14,14,14,14,14,14,14,14,14,14,14 defb 14,14,14,14,14,14,14,14,14,14,14,14,14,14,14,14 defb 14,14,14,14,14,14,14,14,14,14,14,14,14,14,14,14 defb 14,14,14,14,14,14,14,15,15,15,15,15,15,15,15,15 defb 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15 defb 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15 defb 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15 defb 15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15
Quote from: arnoldemu on 13:00, 09 September 13There are other ways that the audio could be compressed. You could store the changes in volume - better for Plus DMA with it's pause etc.
Quote from: arnoldemu on 13:00, 09 September 13To improve "audio quality" you may be able to double the frequency and insert an "average" sample between each "good" sample. Perhaps this may make it better?
Quote from: arnoldemu on 13:32, 09 September 13Hmmm.. I wonder what can be done to improve the vocal samples...
Page created in 0.139 seconds with 26 queries.