| View previous topic :: View next topic |
| Author |
Message |
tomaitheous
Very interested
Joined: 11 Sep 2007
Posts: 236
|
 Posted: Fri Apr 25, 2008 2:12 am Post subject: Yamaha internal waveform info |
 |
|
Hey guys,
Not sure if this is the correct forum (or place) for this, but I'm looking for the resolution and length of the internal sine wave used by the YM2612, OPL2, and some other Yamaha FM chips. I *think* I remember TmEE mentioning the 2612 had a 9bit resolution on the sine wave, but no mention of length.
Also looking for more info on phase distortion synthesis by Casio. This is for a synth emulation project on a non PC platform.
Thanks,
Tom |
|
| Back to top |
|
 |
TmEE co.(TM)
Very interested
Joined: 05 Dec 2006
Posts: 1944
Location: Estonia, Mahtra village
|
| Posted: Fri Apr 25, 2008 10:42 am Post subject: |
|
|
AFAIK, the sampling rate is around 52KHz (53203424(PAL) / something), accuracy 9 or 10 bits on discrete Yamaha YM2612 and about 12 bits on Sega internal YM2612. The bit count is an estimate, which I estimated by listening the amount of aliasing noise produced and comparing with WAV files degraded to some bit count.
OPL2/3 is 49KHz... I don't know about accuracy, it can't be less than 12 bits I think, my sound card sound very clean.
_________________
Mida sa loed ? Nagunii aru ei saa  |
|
| Back to top |
|
|
tomaitheous
Very interested
Joined: 11 Sep 2007
Posts: 236
|
| Posted: Fri Apr 25, 2008 2:11 pm Post subject: |
|
|
| TmEE co.(TM) wrote: | AFAIK, the sampling rate is around 52KHz (53203424(PAL) / something), accuracy 9 or 10 bits on discrete Yamaha YM2612 and about 12 bits on Sega internal YM2612. The bit count is an estimate, which I estimated by listening the amount of aliasing noise produced and comparing with WAV files degraded to some bit count.
OPL2/3 is 49KHz... I don't know about accuracy, it can't be less than 12 bits I think, my sound card sound very clean. |
Hmm but that doesn't tell me anything about the digital sine waveform stored on the internal rom. Couldn't you silence all but one channel and silence all but one operator to get a clean/unmodified carrier output? Setting it to some like 40-80hz and record that. One could figure out the length of one the sine wave and possibly the resolution of it (assuming the DAC is equal or higher res than the internal waveform). That or decap the chip. |
|
| Back to top |
|
|
TmEE co.(TM)
Very interested
Joined: 05 Dec 2006
Posts: 1944
Location: Estonia, Mahtra village
|
| Posted: Fri Apr 25, 2008 7:33 pm Post subject: |
|
|
OK, I'll run some tests when I've finished my exam tomorrow.
_________________
Mida sa loed ? Nagunii aru ei saa |
|
| Back to top |
|
|
tomaitheous
Very interested
Joined: 11 Sep 2007
Posts: 236
|
| Posted: Sat Apr 26, 2008 3:36 am Post subject: |
|
|
| TmEE co.(TM) wrote: | | OK, I'll run some tests when I've finished my exam tomorrow. |
Awesome  |
|
| Back to top |
|
|
TmEE co.(TM)
Very interested
Joined: 05 Dec 2006
Posts: 1944
Location: Estonia, Mahtra village
|
| Posted: Sat Apr 26, 2008 1:25 pm Post subject: |
|
|
I recorded form my MD2 and MD1, and MD2 looks way nicer... I'm waiting when my friend uploads it on his site... I'll give a link when its done.
_________________
Mida sa loed ? Nagunii aru ei saa |
|
| Back to top |
|
|
TmEE co.(TM)
Very interested
Joined: 05 Dec 2006
Posts: 1944
Location: Estonia, Mahtra village
|
| Posted: Sat Apr 26, 2008 3:09 pm Post subject: |
|
|
http://www.epicgaming.us/tiido/MD_Sine.rar
Kindly done by Epicenter 
I hope this sheds some light on YM2612's internals
Each WAV has 5 or 6 samples in it, First with Total Level at 00h, and each following has it incremented. 0, 1, 2... notice how the wave gets better with lower TLs... and they should be 27.5Hz...
_________________
Mida sa loed ? Nagunii aru ei saa |
|
| Back to top |
|
|
Sik
Very interested
Joined: 10 Apr 2008
Posts: 168
|
| Posted: Sat Apr 26, 2008 3:39 pm Post subject: |
|
|
Is it me or the MD1 fails to do the peaks properly? Also it seems to have a lower volume.
_________________
Sik is pronounced as "seek", not as "sick".
Mega Drive Research and Development Center: http://srb2town.sepwich.com/mdrdc
The Zen forums: http://zenforums.ipbfree.com/ |
|
| Back to top |
|
|
TmEE co.(TM)
Very interested
Joined: 05 Dec 2006
Posts: 1944
Location: Estonia, Mahtra village
|
| Posted: Sat Apr 26, 2008 3:57 pm Post subject: |
|
|
The graph of MD1 is bigger than MD2.... and I recorded the sound both having about same level. and MD1 does have bad peaks, on all TL values.
_________________
Mida sa loed ? Nagunii aru ei saa |
|
| Back to top |
|
|
Shiru
Very interested
Joined: 07 Apr 2007
Posts: 786
Location: Russia, Moscow
|
| Posted: Sat Apr 26, 2008 5:10 pm Post subject: |
|
|
| This recordings was taken directly from YM2612 output or from amp output? |
|
| Back to top |
|
|
TmEE co.(TM)
Very interested
Joined: 05 Dec 2006
Posts: 1944
Location: Estonia, Mahtra village
|
| Posted: Sat Apr 26, 2008 5:38 pm Post subject: |
|
|
directly
_________________
Mida sa loed ? Nagunii aru ei saa |
|
| Back to top |
|
|
HardWareMan
Very interested
Joined: 15 Dec 2007
Posts: 542
Location: Kazakhstan, Pavlodar
|
| Posted: Sat Apr 26, 2008 6:54 pm Post subject: |
|
|
| TmEE co.(TM) wrote: | | directly |
So, that's means it is not true signal. Because YM2612's output more complex - all 6 channels is separated in time with gaps between them. The channel change speed is about 120kHz, so one channel discretization is about 20kHz. If you got normal oscilloscope, please make a record YM2612 output at 120-200kHz samplerate.
I had made very deep analysis of analog output of YM2612 at one russian forum, but now this forum is dead. I can rebuild that job, if you want it. |
|
| Back to top |
|
|
TmEE co.(TM)
Very interested
Joined: 05 Dec 2006
Posts: 1944
Location: Estonia, Mahtra village
|
| Posted: Sat Apr 26, 2008 7:15 pm Post subject: |
|
|
Unfortunately I have no access to oscilloscope... actually my friend has one, 2ch 50MHz, but its non digital and I have no idea how would I record its output... my sound card can record at 500KHz(according to chip datasheet) but Win drivers don't allow it...
It would be nice if you rebuild your job. MD1 and MD2 sound different, MD2 seems to have more accuracy (less aliasing noise) and better sine output (at least according to the recordings above)... MD1 used is same what HardWareMan has, just less modded.
_________________
Mida sa loed ? Nagunii aru ei saa |
|
| Back to top |
|
|
Sik
Very interested
Joined: 10 Apr 2008
Posts: 168
|
| Posted: Sat Apr 26, 2008 7:57 pm Post subject: |
|
|
The different peaks are enough to make it sound more aliased. Those sudden changes make it sound like noisy, in fact sudden changes are the reason for clicks in sounds.
And I think you should take a photo of the oscilloscope output 
_________________
Sik is pronounced as "seek", not as "sick".
Mega Drive Research and Development Center: http://srb2town.sepwich.com/mdrdc
The Zen forums: http://zenforums.ipbfree.com/ |
|
| Back to top |
|
|
Shiru
Very interested
Joined: 07 Apr 2007
Posts: 786
Location: Russia, Moscow
|
| Posted: Sat Apr 26, 2008 8:17 pm Post subject: |
|
|
| Such small and same changes can only change timbre slightly, they can't make clicks or noise - they have same frequency like carrier tone. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
