Search options (Page 6 of 16)

The problems you might have with .tfi is not with the algorithm and operator's connection as this is just a typo error in the GenMDM doc but mainly from the Second decay rate (D2R) mapping error in GenMDM 102 I related a while ago. This parameters greatly influence the final sound.
You also need to take in account that velocity is assigned to TL levels in GenMDM and this also affect the final sound a lot in FM.

That is why you might often end up with a different patch than the one you should when D2R is involved.

Until a small fix is available (maybe this as been addressed since, but not publicly here by Seb) a workaround is to set Sustain level to max attenuation (this disables the second decay rate).
By default GenMDM seems to have Second decay rate set high with a long decay and only OP1 (M1 the one with feedback) accept control over it in a limited way (only goes from max to middle rate instead of min).

Hope that helps,
between this does not impact to much what you can do sound-wise

reminder for the EG :

Here is a small tutorial on how you can quickly create a simple drum loop pattern on only one mono FM channel within FMDrive vst...you can even be way more complicated by sending program changes and around 100 registers that can be automated! Here we use Cubase 5 as a DAW.
All the data can be transfered in real time to the Sega Genesis hardware via GenMDM Midi Interface or will be trans-coded into a VGM File (soon).
FMDrive vst is available at  http://www.alyjameslab.com

Yes I reported this here a few pages back >> http://chipmusic.org/forums/post/171808/#p171808
for example you shown algorithm 0 in your example :  1->3->2->4  where only OP 4 is the carrier.

It is a typical typo error in some documentations, OP 3 & 2 are often swapped.

Just to let you know that FMDrive vst got updated to 1.23!
-added extra features and corrections.
-accuracy improved again.
-GenMDM midi mapping already done on startup for those who have one.

Free for all registered user  head over to alyjameslab.com and drop me a mail
I have posted an update log at the beginning of the thread so you can check your version against
(GenMDM owners>> let me know, so I can send the pre mapped version)

UPDATE

what I usually do is making the song with FMDrive (and lately Super PSG beta).If I want the hardware output, I then hook up GenMDM and route the MIDI content to it and voila!

for now on it is damn close Smile

DIRECT AUDIO COMPARISON

REMINDER of my last little GenMDM bug/questions report

-PSG side: channels 1,2,3,4

Not mentioned in the CC list but:
CC#11 (expression)
can control the PSG volume too, witch is great but it only affect PSG channel 1 and 2.
PSG channel 3 CC#11 seems to affect volume on PSG channel 1
CC#85 do not work on either.
CC#42 on channel 1 : seems to change the TET tunning on channel 1 but also affect channel 02
CC#42 on channel 2 : seems to change the TET but only affect channel 03
CC#42 on channel 3 : seems to change the TET but only affect channel 01

-YM2612 side:
See previous posts.

VGM Logger:
CC#114 set above 65
GenMDM do not output any CC Message.

Anybody who can send informations on this is welcomed
Seb, PM me about this when you got some time, I would really loved the logging feature

Of course I will add that you can also write to the volume register at CPU speed without using Hblank timing.
Writing this sequence of data for a 16 data point sinewave,
.db 04,02,01,00,00,01,02,04,11,13,14,15,15,14,13,11
gave me something like 1978Hz without any wait...
this gives us roughly a 31,648 KHz sample rate, it should be the same thing for PWM or PCM sample reading...assuming there isn't any other code going round then the audio stuff.
So it is twice as fast as Hblank with near 32µs against 64µs
the simplest square wave could be .db 00,15 as high as 15824Hz
anyway at high rate things get screwed up easily

If it can help...I have never read a story about somebody being screwed up by Seb, I know it will not make your device delivered faster and the delivery/making/answering is reaaaaaly long but I am sure it will come at some point and Seb will provide a solution for the undelivered ones.