|
Cloudschatze
Guest
|
 |
« Reply #20 on: October 28, 2005, 04:48:23 PM » |
|
Might Quest For Glory II's MT-32 driver perform this reversal? I am at work right now, else I would try using it with SQ1.
|
|
|
|
|
Logged
|
|
|
|
NewRisingSUn
Senior Member
Offline
Posts: 650
|
 |
« Reply #21 on: October 28, 2005, 04:57:57 PM » |
|
I read is as he composed the music, arranged it for the different modules, then simply made the MT-32 driver itself reverse the pan on the fly as it was played back. Well, IF they did that, they certainly didn't include that pan-reversing driver with the final product. Looking at the INSTALL.HLP file, there seems to have been a NOLOAD.DRV for internal use, maybe that one did it. I also find it funny that they would bother to make sure that the music pans right on MMA-conforming MIDI devices; I think at the height of the MT-32's era, the MT-32 was the only sound option that gave you stereo music anyway, wasn't it? Sierra certainly only provided a dedicated OPL3 option when they started supporting General MIDI, which made the MT-32 obsolete anyway... Might Quest For Glory II's MT-32 driver perform this reversal? No, no MT32.DRV that I've seen does that.
|
|
|
|
|
Logged
|
|
|
|
|
Cloudschatze
Guest
|
 |
« Reply #22 on: October 28, 2005, 05:17:00 PM » |
|
I think NOLOAD.DRV keeps the patch-bank from being sent to the MT-32. I have a copy of this, if you want to take a look at it. Well, drivers notwithstanding, one thing seems to have been verified... ...the MT-32 certainly should be considered the most f'd up synth ever produced. 
|
|
|
|
|
Logged
|
|
|
|
NewRisingSUn
Senior Member
Offline
Posts: 650
|
 |
« Reply #23 on: October 28, 2005, 05:30:54 PM » |
|
I think NOLOAD.DRV keeps the patch-bank from being sent to the MT-32. I know, but maybe it also performed that pan-reversal Seibert talked about. I have a copy of this, if you want to take a look at it. Yes, please. Where did you get it? If you got it from previous versions of my SQ4CD update, then it's not the original NOLOAD.DRV, but my substitute I wrote for my own use which I accidently included with the update distribution. Well, drivers notwithstanding, one thing seems to have been verified.. Not so fast, young man.  Titles like that must be earned, so what's the deal with that D-70 you mentioned?
|
|
|
|
|
Logged
|
|
|
|
|
Cloudschatze
Guest
|
 |
« Reply #24 on: October 28, 2005, 05:51:42 PM » |
|
I have a copy of this, if you want to take a look at it. Yes, please. Check your e-mail. Well, drivers notwithstanding, one thing seems to have been verified.. Not so fast, young man.  Titles like that must be earned, so what's the deal with that D-70 you mentioned? I nearly purchased one, but from what I've read, the initial release of D-70 keyboards came plagued with a variety of MIDI implementation bugs, some of which have yet to be fixed. Not to mention the fact that the LCD backlight will burn out and need to be replaced every so often.
|
|
|
|
|
Logged
|
|
|
|
|
Cloudschatze
Guest
|
 |
« Reply #25 on: October 31, 2005, 06:40:21 PM » |
|
I had never considered the following to be a flaw of the MT-32. Opinions? There is a design flaw... The D/A converters can be overdriven resulting not in a nice distorted guitar effect, but a loud, rude, equipment-damaging, ear-shattering crackle. If you have the oppurtunity to play a MT32, the easiest way to hear this, is to play the Honkytonk sound, or the orchestra smash at high velocities. It happens much more often when sequencing, due to the fact that you can change the velocities manually. I brought my unit back for service after hearing this, and they charged me 30 minutes labor, before I found out that ALL MT32's do this. Roland's explanation? When they admit the problem exists, they say it is not designed for professional use.
|
|
|
|
|
Logged
|
|
|
|
|
Alistair
|
 |
« Reply #26 on: November 01, 2005, 01:39:41 AM » |
|
No, it certainly is a flaw, but it won't necessarily happen to you. As the guy says, you must have high volumes and velocities set. Tom and I have our MT-32's set to 80 or 75% volume, which fixes that. Still in rare cases, like King's Quest 4, problems still occur..
- Alistair
|
|
|
|
|
Logged
|
|
|
|
NewRisingSUn
Senior Member
Offline
Posts: 650
|
 |
« Reply #27 on: November 01, 2005, 08:19:19 AM » |
|
I had never considered the following to be a flaw of the MT-32. So what had you considered it to be? This distortion is your typical "digital" distortion --- if the output signal gets too high, it wraps from +32767 to -32768, resulting in a cracking sound at maximum amplitude.
|
|
|
|
|
Logged
|
|
|
|
Laust
Senior Member
Offline
Posts: 722
|
 |
« Reply #28 on: November 01, 2005, 12:18:59 PM » |
|
This is one flaw I cannot understand that Roland didn't fix. It's even more annoying on the CM-modules, since the unit's volume control does not change the master volume level (as on the MT-32) but merely controls the analog volume level. This means, that if a game decides to set a high master volume level and it crackles as a result, there's not much you can do about it (on the MT-32 you can usually lower the master volume level from the control panel on the unit itself).
|
|
|
|
|
Logged
|
|
|
|
|
Cloudschatze
Guest
|
 |
« Reply #29 on: November 01, 2005, 01:25:44 PM » |
|
I assumed it to be normal, or at least common to Roland products. Certainly the JD-990 was designed for professional use, and yet, it is affected by the same "flaw".
|
|
|
|
|
Logged
|
|
|
|
NewRisingSUn
Senior Member
Offline
Posts: 650
|
 |
« Reply #30 on: November 01, 2005, 03:10:33 PM » |
|
This means, that if a game decides to set a high master volume level and it crackles as a result, Any game in mind in particular? The only game I know that forcibly sets the volume to 100 is 4D Sports Driving (aka Stunts), and while it crackles on an MT-32, it doesn't on a CM-32L/CM-500.
|
|
|
|
|
Logged
|
|
|
|
Laust
Senior Member
Offline
Posts: 722
|
 |
« Reply #31 on: November 01, 2005, 03:18:12 PM » |
|
Speedball 2 comes to mind. And Budokan crackles on when you visit the Sensei.
|
|
|
|
|
Logged
|
|
|
|
|
Alistair
|
 |
« Reply #32 on: November 01, 2005, 11:02:41 PM » |
|
Early Sierra games set it to 100, LSL3 and KQ4 spring to mind, and I've had 'crackles' with both. I assumed it to be normal, or at least common to Roland products. Certainly the JD-990 was designed for professional use, and yet, it is affected by the same "flaw". Well, I assure you I've never had the same problem with a Sound Canvas. - Alistair
|
|
|
|
|
Logged
|
|
|
|
NewRisingSUn
Senior Member
Offline
Posts: 650
|
 |
« Reply #33 on: November 02, 2005, 01:50:41 PM » |
|
And Budokan crackles on when you visit the Sensei. But Budokan doesn't set the main volume. At least my version doesn't... Early Sierra games set it to 100, LSL3 and KQ4 spring to mind, and I've had 'crackles' with both. Since when do you have a CM module? LSL3 cracks on the MT-32, but not on the CM-500. KQ4 sets the main volume very low at the beginning, at least version 1.000.111; no cracks there either. Well, I assure you I've never had the same problem with a Sound Canvas. The Sound Canvas doesn't crack like the MT-32; it just has a normal "clipping" if you drive it too loud (Pad 1: Phantasia at full volume with chorus=127, for example, or bank 0's "Whistle" on some models).
|
|
|
|
|
Logged
|
|
|
|
|
Cloudschatze
Guest
|
 |
« Reply #34 on: November 02, 2005, 06:30:40 PM » |
|
The Sound Canvas doesn't crack like the MT-32; it just has a normal "clipping" if you drive it too loud (Pad 1: Phantasia at full volume with chorus=127, for example, or bank 0's "Whistle" on some models). This is most likely what I've experienced with the JD-990; clipping, rather than cracking. Are they not the same thing though, just analog and digital varieties? Would reducing the voltage going into the DAC resolve the digital clipping found in the MT-32? I realize that turning the volume knob will accomplish the same result, but for the CM-500, a resistor could surely be installed. I now understand why you are interested in using an outboard DAC, NewRisingSun.
|
|
|
|
|
Logged
|
|
|
|
|
Zemus
|
 |
« Reply #35 on: November 02, 2005, 06:37:55 PM » |
|
A DAC accepts a digital input which it turns into an analogue signal. Now a digital input runs on two levels which is typically 0V for 0 and 5V for 1. If you reduce the maximum voltage it won't be able to read the 1s and you'll end up with silence.
Clipping means that the soundwave gets flattened at the top or the bottom if it goes above the maximum/minimum value a soundwave can have digitally. Think of a sine wave where the tops and bottoms are flat instead of round.
Crackling occurs if whatever makes the digital signal jumps from the max value a soundwave can have back to its minimum value. +32767 to -32768 for example (as NewRisingSUn pointed out earlier in the thread).
Both of these things happen at the mixing or amplification stage before it gets sent to the DAC. The DAC simply converts.
A sound module should, in a perfect world, have soundwaves and effects that would never reach the max value, but it doesn't look like Roland thought of that.
|
|
|
|
|
Logged
|
|
|
|
|
Cloudschatze
Guest
|
 |
« Reply #36 on: November 02, 2005, 08:30:58 PM » |
|
Thanks for the clarification, Zemus!
|
|
|
|
|
Logged
|
|
|
|
Laust
Senior Member
Offline
Posts: 722
|
 |
« Reply #37 on: November 02, 2005, 10:37:38 PM » |
|
And Budokan crackles on when you visit the Sensei. But Budokan doesn't set the main volume. At least my version doesn't... Maybe it crackles at the default volume then? I must admit, I don't recall under what circumstances it happened (it does happen at a certain volume and it does go away when you lower it. I'm too lazy to check the specifics right now, however), but since I got the CM-32L before I got my MT-32, I assumed it was with that module. Budokan would have been one of the first games I checked out - music by Commodore 64 legend Rob Hubbard  Speedball 2 crackles quite clearly on the CM-32L, at least.
|
|
|
|
|
Logged
|
|
|
|
|