ChanServ changed the topic of ##yamahasynths to: Channel dedicated to questions and discussion of Yamaha FM Synthesizer internals and corresponding REing. Discussion of synthesis methods similar to the Yamaha line of chips, Sound Blasters + clones, PCM chips like RF5C68, and CD theory of operation are also on-topic. Channel logs: https://freenode.irclog.whitequark.org/~h~yamahasynths
cr1901_modern has quit [Quit: Leaving.]
natarii has quit [Quit: quit]
natarii has joined ##yamahasynths
<ej5>
i just got some AY-3-8910 and AY-3-8912 devices from UTSource
<ej5>
2 of the 8912s, 3 of the 8910s. only one of the 5 has not been sanded down and remarked: a single 8910.
<ej5>
it might be fun to curve trace the pins to see how closely they match. i don't like the idea of putting in a part like this in a circuit only to have it short vcc and gnd, for example.
<andlabs>
where are the chinese companies making clone chips?
<andlabs>
KitsuWhooa: lol that's not suspicious at all
<KitsuWhooa>
I never got around to testing them properly, but they produce a sine wave
<KitsuWhooa>
package doesn't match, so they are most likely clones
<andlabs>
how were you able to determine the sine wave part if the package doesn't match
<andlabs>
guessing which pin is output?
<andlabs>
*which pin is which
<KitsuWhooa>
I treated it as a YM2151, and the pinout matches
<KitsuWhooa>
they are claiming to be one, after all :p
<superctr_>
they are compatible with YM2151, but they are fakes
<superctr_>
used in bootleg arcade boards many years ago
<superctr_>
l_oliveira knows all about them
<l_oliveira>
hello guys
<superctr_>
so they aren't rebadged RAM chips like some other infamous fakes, but rather they are sound chips that are OPM-compatible, but internally implemented completely differently and produce inferior sound
<l_oliveira>
been absent for a while since I was out of town but here I am again
<andlabs>
looking for the exact part number on that giant gendev chip
<andlabs>
I shudder to think what their FM clone of the one super heavy 80s snare drum recreation sounds like
<l_oliveira>
the CY5001 sounds great on a System 16 board because SEGA did stick to the basics lol
<andlabs>
as for Seibu I don't think anybody noticed, not least because apparently according to vgmrips seibu's games are super soft for some reason
<l_oliveira>
then on a SF2 board you have funny guile music, I think probably the EG on the fake chip is fubar
<l_oliveira>
some fake chips do generate good sound though
<superctr_>
Seibu PCBs have a jumper to switch between OPM and OPL2
<superctr_>
either can be used on the same board
<l_oliveira>
yeah I copied that on my YM2151 custom board, I use it to run a OPL2 on my MSX sometimes
<andlabs>
Gallantry but in reality Timid
<superctr_>
but the sound program obviously needs to be replaced, and i don't think there were games actually used both
<andlabs>
*Timidity
<l_oliveira>
I have a software algorithm that detects whatever FM chip is on the I/O port, I use the same port for all the chips
<andlabs>
pretty sure there were some games that used either
<andlabs>
not by Seibu but by other companies
<andlabs>
Toaplan maybe?
<l_oliveira>
just have a daughter board on a 40pin IDC board which connects to a MSX cartridge
<l_oliveira>
40pin IDC flat cable I mean (reusing Parallel ATA cables is fun)
<andlabs>
okay but can you moonsound
<andlabs>
:U
<l_oliveira>
later guys gotta go out for a bit
<l_oliveira>
I can sure thing
<l_oliveira>
brb
<andlabs>
I still need to write the DX100 simulator for my SFG-05
<andlabs>
so we can finally figure out what those undocumented registers do
<andlabs>
just can't figure out where to start with MSX software development
<andlabs>
especially MSX-DOS software development in modern C with high memory usage (need at least 48K)
<superctr_>
i feel the same
<superctr_>
the thing with MSX: you have to ask for all documentation, because you won't find it on the internet
<andlabs>
oh no I can find all documentation easily
<andlabs>
what you won't find is a starting point
<andlabs>
and C makes it harder because there are really only two C SDKs that aren't from the 1980s
<superctr_>
I've used SDCC
<andlabs>
the accumulated-over-time one by many people including konamiman and two spanish guys that among other things does int main(char **argv, int argc)
<superctr_>
i didn't use a libc though
<superctr_>
i just did bare metal C with some custom written wrappers
<andlabs>
and Re:Action which is CC-BY-SA, which is not only a software license, but also viral
<andlabs>
*not only not a software license
<andlabs>
cool but I'd love to be able to allocate memory in MSX-DOS
<andlabs>
is that not possible
<andlabs>
I know for some dumb reason the MSX BIOS spec never provided a way to get all available RAM
<andlabs>
because hurr
<andlabs>
given that some Casio machines have this: 8K should be enough for everyone
<andlabs>
I wonder how the tiny slot checker determines RAM without a false positive
<superctr_>
that's the thing
<superctr_>
you need to ask for that
<superctr_>
example code counts as documentation
<superctr_>
and even if you find documentation, the example code is also scarce
<andlabs>
so how does anyone get anything done then
<cr1901_modern>
grauw sometimes shows up in here, but Idk if they have this channel on their auto-join
<andlabs>
I've already talked to them on twitter lol
<andlabs>
in fact it was their page on the YM2164 that I'm interested inc racking
<andlabs>
those undocumented registers? they're actually used by the DX100
<andlabs>
for something related to either portamento or breath control
<whitequark>
Sarayan: glasgow + yamaha opl is out of the box use case
<whitequark>
it doesn't do analog though, it captures the digital output
<whitequark>
DAC bitstream
<whitequark>
as for 5V, glasgow is designed to support true TTL IO
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
<l_oliveira>
I'm back
<l_oliveira>
MSX DOS2 has pretty decent memory management
<l_oliveira>
the VGM player for it can handle ludicrously large VGM files like 4MB+ no problemo
<andlabs>
I looked into that and it seems to just allocate contiguous memory?
<andlabs>
will have to check again
<andlabs>
alos grauw writes in assembly language (I shudder to think of writing BOTH a 6303 AND 44780 emulator entirely in assembly)
<andlabs>
and also uses a custom built assembler... written in java
<andlabs>
bah
<andlabs>
finally getting around to archiving micro-cap
<andlabs>
two files are screwed
<andlabs>
the micro-cap 6 executable is a permanent 403
<andlabs>
both microcap 5 v2.1.2 downloads are the same file
<andlabs>
fortunately micro-cap 6's CD image download works and the micro-cap 5 v2.1.2 download file that downloads is also existant
<andlabs>
I could possibly go back to see if micro-cap 5's issue here was just a mistake
<andlabs>
also huh this software dates all the way back to the Apple II (but nothing before v5 was available)
<andlabs>
who was it that linked the page here again?
<l_oliveira>
grauw's vgm player is a good example of how to allocate a ton of memory on MSX DOS2
<andlabs>
ah, Lord_Nightmare
<l_oliveira>
and a lot of other things like for example handle I/O, access the screen, do timer control, etc
<andlabs>
l_oliveira: so in this case I need to memory map the DX100 ROM (32KB), which is why I need so much RAM in the first place =P
<l_oliveira>
you don't need the whole rom do you?
<andlabs>
not at once
<andlabs>
most of the ROM is just the default voice bank
<l_oliveira>
you can split it in 16KB chunks
<andlabs>
192 voices lol
<andlabs>
but yeah no I'm not afraid of banking
<l_oliveira>
that's the minimum paging size of memory on MSX DOS2
<l_oliveira>
well on the whole slot subsystem too
<l_oliveira>
memory mapper is designed to "slot in" into the "original slots" concept
<l_oliveira>
and everything slots into something else in the MSX, that's why people dislike the platform when they first see it I suppose
<l_oliveira>
but heck load 6MB of VGM in a 30+ years old platform
<andlabs>
yeah I only have a fractal2000 512KB cartridge
<andlabs>
vgmplayer chokes easily
<andlabs>
if there are larger cartridges then cool
<andlabs>
also for some reason vgmplayer doesn't behave as advertised for things like OPN
<andlabs>
unless it demands some sort of PCM chip to even bother
<andlabs>
but yeah if you have the knowledge on how to write MSX programs and are bemoaning that it's not possible to get started doing so then you should make it possible
<andlabs>
and at least you can find shit for the MSX
<andlabs>
I still need to write my long amiga development question with an undertone of rantiness about how proprietary the fanbase is
<andlabs>
(unless you're pirating games)
<l_oliveira>
for example you can't properly play mega drive music on MSX without a Turbo R computer because of the 4khz timer and the stupid fast R800 CPU
<l_oliveira>
I mean mega drive VGMs
<l_oliveira>
also, vgmplay does not exactly support OPN on MSX without a physical OPN or a OPM connected