2 days ago new music disk from Moods Plateau which called "Hot Affairs", was released. You can grab it from the site, or directly here

Executables are 68k, but everything system friendly and works fine on aos4.1u1 (tested). There is 2 screenshots of that music disk in action on my os4setup:

screenshot1
screenshot2

To enjoy, just downlod, unpack, copy necessary libs from install/ directory to your LIBS: Then run prefs, choice WB mode (or full-screen with 640x512x32bit mode), and press launch.

There is 7 nice modules, have fun.

ps. There also some comments on Pouet.net about.

Last edited by kas1e on 26-Jan-2010 at 11:27 AM.

_________________
Join us to improve dopus5!
zerohero's mirror of os4/os3 crosscompiler suites

@kas1e

Ups, didn't know that my friends inside moods put out the musicdisk so fast. We collected many pix last weekend. fixed some stuff allso. That the disk worked under OS4 I knowed of course our Dr.K coded the Disk in C code and systemfriendy as it can be. Could not test the last Beta on Sam for myself course my Sam went to Italy and I hope I get a replaycement for it soon, but our DaFreak has allso one and it worked fine.
We inside moods have allso totally different Amigas running, our main coders use classic stuff, one codes demos in ASM and the other in C systemfriendly code, but we try to make musikdisks as system friendly as possible so anyone on classics, emulation and Ng Amigas can enjoy the productions. The disk itself is AGA 8bit graphic so it works on all main systems equal good. The tunes where Protracker made, but we had to convert them to DBM course the existing AHI PTreplayer routines where not fast enough on all systems, so it was better this way. Anybody now on 030 or higher CPU's can use the disk without audio problems.
Perhabs when you find some bugs or whatever, contact our ppl on the moodsplateau.net site so we can make our Musikdisk engine even more compatible on all systems.

@Psycreator

Quote:

AHI PTreplayer routines where not fast enough on all systems,

Does it play the modules properly in an interrupt with the right mode (Like Paula)? Or just render them as audio to AHI?

@kas1e

sound doesn't seem to work at all here.

SAM-Flex OS4.1 Update 1

_________________
BroadBlues On Blues BroadBlues On Amiga Walker Broad

@broadblues

Did you copy all the necessary libs from archive to LIBS: ? (expectually dbplayer.library must be in libs: , because that music disk based on DBM modules).

I only copy (manually of course, not by scripts) necessary libs to LIBS: (os4 versions), and the run it.

@Hypex
Not sure what you try to ask from Psycreator, but they just use public libs which avail for os3/os4/mos, etc (ptplay.library, ptreplay.library, dbplayer.library, etc).

Related to how PTplay/Ptreplay works - it works fine (we use in our diskmag ptplay.library for depacking mod's data, and then play it as raw data over ahi). And also ptreplay.library, which have 3 versions too:
-- old one (68k), very old, play mods over CIA (so, hw baning)
-- new ones (os4/mos), done by Salas00, which plays mods over AHI in end by using of ptplay.library (depacking).

It mean that we have 2 different libs in general:
-- ptReplay.library (that what now avail everythere, but before was only for 68k/cia/paula).
-- ptPlay.library (that library only depack mod data to raw data, and then coder play that raw data over ahi).

In case with ptReplay.library on os4/mos, it's in end use ptPlay.library for decode mods too, but for coder it does not matter, because API of ptReplay.library absolutly the same.


So, I think Psycreator just a bit messed all of this, and say not so correctly about "works not fast enough", because plain PTReplay (REplay, not Pt_play) library works madness fast, and even on 020 cpu it fast and nice (because HW baning code).

But if, he try to say about PTPlay.library (not pt_replay.library), then, of course it will be a bit slow on classic hardware, because it's CPU based depacker from mod to raw data and then, coder plays that raw data over ahi himself.

All in all, speedup problems can be only detected on classic hardware, and 060/50mhz are enough for making even ptplay.library works in realtime (example it's our diskmag from my signature).

The solution here, just use PTReplay.library (not ptplay.library) which for now already avail for all platform , and for os3 use that old one, for os4/mos use new one. And it will be works over the same fucntions from the coder side

I just think that coder does not know about it at momnet.

That how i organize code in our diskmag in case to support all:

data:libs/

data:libs/ptreplay.library - for paula plaing (68k only)

data:libs/ptplay.library - for depack/play over ahi (68k)

data:libs/mos/ptplay.library - for depack/play over ahi (mos)
data:libs/mos/ptplay.library.elf

data:libs/os4/ptplay.library - for depack/play over ahi (os4)

data:libs/aros/ptplay.library - for depack/play over ahi (aros)

But for now it's better just use ptreplay.library at all, because when i code our mag, ptreplay.library for os4/mos was not available, but now, it's here, and it will be very-very easy to use it.

It's even can be not in data, but refer to LIBS: , users yourself can download these libs and install.

Last edited by kas1e on 26-Jan-2010 at 03:44 PM.
Last edited by kas1e on 26-Jan-2010 at 03:24 PM.
Last edited by kas1e on 26-Jan-2010 at 03:17 PM.
Last edited by kas1e on 26-Jan-2010 at 03:13 PM.
Last edited by kas1e on 26-Jan-2010 at 03:10 PM.
Last edited by kas1e on 26-Jan-2010 at 03:07 PM.

_________________
Join us to improve dopus5!
zerohero's mirror of os4/os3 crosscompiler suites

@Hypex
As far as I know, all Module formats the engine can handle renders the audio to an AHI stream and then it goes to the soundcard at the end. I cannot say for sure what the problems was with those ptreplayer.lib, but even on 060 it was to slow and sometimes you got terrible breaks between tunes, only on PPC it was just fine. So using those old dbplayer.lib was the only sollution which was acceptable for all of us. Or we had to use p61 routines, but then no OS4 user could here the sound, even worst the chipram would be to small to handle those bigger modules together with the graphics on AGA machines.
We are trying to get those engine better and better, perhabs in the future we find better sollutions for all tracker formats we use internaly. But all in all finnaly we got it stable enough I think, this is the right direction. And the engine is open end, means you could allso make an highend looking musikdisk with that in 24bit colours and whatevee with downrenering to other screenmodes while AHI stayes the same and the qualit of the sound is only depending of the soundcard you use. It is an compromise at the end.

But this time I wannted to make a musikdisk for all Amigas, course of the 4channel music inside, so we keeped it 256 colours so classic users and AmigaNG users have together something to play with :O))
And for me it is a goodby to my old A4000TE and al the good times I had with that machine :O)

@kas1e
Hello, about the ptreplayer.lib. We tested it many times without success at all, something went wrong anytime our coder tried to fix problems. But I hope soon he finds out what the problem was at the end. The biggest problem is he uses an 060 AGA only Amiga for coding, and while on my little 030 Amiga anything was fine, his computer went crazy wiht the same libs. So until we find the correct sollution for the MOD format, we use dbplayer.lib instead which seems to run fine on all systems so fa. But I know, for me it would have been better to keep the tunes in MOD format instead of dbm now. But better this way that an unstable way ;))

@Psycreator
If coder in interst, i can help with that libs/mods/etc crap (kas1e [[at]] yandex.ru if he in interest)

_________________
Join us to improve dopus5!
zerohero's mirror of os4/os3 crosscompiler suites

@kas1e
I will Dr.k know about that, thx a lot :O)

Hi Amiga OS4 Community,

thanks for your feedback here!

@Hypex

The Hot Affairs Musicdisk renders the audio as 16 Bit PCM data by DBPlayer Library and does output via AHI.

PtPlay Library is too slow on 68k Platforms if running other stuff at the same time.

PtReplay Library 6.6 seems not working correctly. I got heavy drop outs and system stability problems. That's at least my experience. Maybe I did something wrong in calling the library functions. Besides that the modules have to be loaded into chip ram which is a knock out if the sizes are above 1 meg.

Greetings,
dOc.K

@dOc.K

Yep, you are right. 68k version of PtReplay use chip memory (what suck inded). But related to PtPlay - it works fine with propper routines on 060/50 and 66mgz (its with Warp3D usage at the same time, input handling (keys, mouse), assign/deassign memory blocks and alt). For example for our crapomag i run my music_player in different process (maybe that give me speed boost).
Anyway, that is only about 060, all what is lower will be slow like hell (and on 030 it will slow 100%). But you always can use for example PtReplay for the 68k (just load and unload modules when its need it, not load it at one push, but just load one , play, unload, load second,play, unload). In that case 1mb imho will be enough. + it will be easy for the os4/mos in that case, just use os4/mos versions of ptreplay.library while the api and your 68k binary are the same.

But imho that is not big problem all in all, dbm modules are nice too (why not). Hope to see more music disk in future from MDS :)

Last edited by kas1e on 27-Jan-2010 at 08:57 PM.

_________________
Join us to improve dopus5!
zerohero's mirror of os4/os3 crosscompiler suites

Installed the libs with the included installer and it started without problems...aside from some sound glitches/back-ground noise (using SamFlex built in audio).
Could be my failing speakers, but I'm not so sure...

_________________
SamFlex Complete 800Mhz System + AmigaOS 4.1 Update 4
Amiga 2000 DKB 2MB ChipRam GVP G-Force040 Picasso 2 OS3.9 BB2
AmigaCD 32

@DAX
Our DaFreak has tested the musicdisk allso on Sam Onboard Sound, he had the same problems with loud white noiseground. Can't remember what he did until it was perfect, but he got it right at the end. Check several AHI settings in the musicdisk prefs, one will fit :O)
And for the rest, it is 8bit Protracker format converted to 8bit dbm format, so no clean samples here, but I did my best while sampling to keep the soundquality as good as possible for Paula and above. But you know, the music was ment to be played on protracker replayer enviroment, the DBM format is a litte bit diferrent to MOD.
What you could do when nothing really works... load the dbm files into your Tunenet with dbm plugn and let them play ;)

@Psycreator

Quote:

As far as I know, all Module formats the engine can handle renders the audio to an AHI stream and then it goes to the soundcard at the end.

On new machines this is no problem but with real Amiga hardware a paula mode should be used. Infact the AHI music interrupt should be used to play the module as it was intended. Otherwise rendering the audio is asking for trouble.
Quote:

We are trying to get those engine better and better, perhabs in the future we find better sollutions for all tracker formats we use internaly.

It is possible to make a module replayer that uses AHI in a player interrupt that can work on all machines. You mask out the default mode and if it's paula then set DMA mode, otherwise it's on a soundcard. Then swap the left/right channels as in Blom's country the Amiga audio channels were reversed. I've modified players to use AHI and this works quite well.

Quote:

And for me it is a goodby to my old A4000TE and al the good times I had with that machine :O)

OH no! Still the only way to play real Amiga songs.

@Psycreator

Quote:

We tested it many times without success at all, something went wrong anytime our coder tried to fix problems.

Depending on the source I found a bug in a [fixed] PT1.1B player routine. The samples were all cleared at the start for the blank loops needed but sometimes random noise would be heard. This was due to using $0 and not the sample start as the blank loop address and also a length of 2 was used. Random data in $0 will be heard. Apart from that a length of 1 should be used. And using 1 with a blank sample start fixes up problems with funny samples. IIRC this also fixes up problems on 060 machines.

@dOc.K

Quote:

PtReplay Library 6.6 seems not working correctly.

It's not if it doesn't use an interrupt and play it in real time!

Quote:

Besides that the modules have to be loaded into chip ram which is a knock out if the sizes are above 1 meg.

Well that can't be helped sometimes. Bit hard for an interrupt to copy samples from fast to chip memory at will. But the module itself could be in fast, though it would take up less space I'd say.