VOGONS


OPL3LPT : OPL3 FM Synthesis on parallel port

Topic actions

Reply 21 of 264, by Dr.Yak

User metadata
Rank Newbie
Rank
Newbie
scroeffie wrote:

i have no idea what this is but it looks cool 😀

It's the cool OPL3 sound chip (Yamaha YMF262) that was featured on / emulated by nearly all sound cards in the DOS era, but dreamblaster has found a cool way to successfully cram it into a printer port.
So now you can have the authentic old school music, even on old laptop that have no sound cards and no free slots, and on modern 64bits machines which don't have ISA 8bit slots anymore.

noop wrote:

You may emulate the whole FM chip instead, it is fully digital anyway. 😀 {...}

It was more as an alternative to those people who do analog-loop recording for Youtube videos (Can get noisy).
And for the purists, as an alternative to using S/PDIF output of chips which aren't perfect replicates of YMF269.
And for the Lulz!, mostly.

noop wrote:

{...} In order to be able to stream data to it via usual typical USB LPT adapters one would probably need to change the interface to only use strobe/busy pins, besides data, again, with the help of microcontroller, something like ATtiny13 in sop package, executing very trivial program. This would also make the device compatible with Atari ST.

That's a fun project !
Though you need to change the protocol tiny a bit:

No actual distinction between register number and value.
You would need to send data in pairs, and automatically re-sync whenever you see a "0" : There are no "00" or "FF" registers number, so it's a sure way to check if you're still in sync with the pairs.
Also the drivers on the Atari ST/PC/Laptop could periodically send "0:0" pairs to make sure to sync the ATtiny13

(That's guaranteed to work on USBLPT too. Even printer port over network could work 😜 for the Lulz!)

Reply 22 of 264, by noop

User metadata
Rank Member
Rank
Member
Dr.Yak wrote:

(That's guaranteed to work on USBLPT too. Even printer port over network could work 😜 for the Lulz!)

Yeah, exactly. Too bad I'm already advancing too slowly with my current OPL LPT project (DIY LPT interface with swappable chips and my own version of the software in size-optimized assembly, which is easier for me to do experiments with, and possibly change the LPT interface as proposed above)

Reply 24 of 264, by jaZz_KCS

User metadata
Rank Oldbie
Rank
Oldbie
Fanatik wrote:

I will buy one of these when available. It would be cool to have soundblaster on my Tandy 1000hx!

Sorry to burst yer bubble. But OPL3 doesn't mean SoundBlaster compatibility. This is purely FM synthesis.

Reply 25 of 264, by Dr.Yak

User metadata
Rank Newbie
Rank
Newbie
jaZz_KCS wrote:

But OPL3 doesn't mean SoundBlaster compatibility. This is purely FM synthesis.

Fanatik should have said "soundblaster music compatibility", it would have been more correct.

Sierra adventure games could use the Sound Blaser driver for music, id software's Keen Commander 4-6 uses FM also, ...

Fanatik wrote:

It would be cool to have soundblaster on my Tandy 1000hx!

The music part (FM) of the Soundblaster : yes. Even in glorious 4 operator OPL3 mode (Sound Blaster 2.0 / Sound Blaster 16 / AdLib Gold).

For the playing of digital sound, you'd need DMA (You don' only need to twiddle settings on a chip, like the OPL3. You need the soundblaster so-called "DSP" to then start autonomously read sample from the memory and play), which old school (pre-ECP) parallel port don't provide.

So you'd need instead to go for some PLUS-to-ISA adapter, so you could try to plug an actual sound-blaster compatible ISA card into the PLUS expansion port of your tandy and get the digitized samples in all their DMA glory.

Or could try plugging a Covox-like simple DAC on the parallel port, and try some TSR driver to manually send the samples from RAM to the parallel port. (Dunno if the 286 CPU in your Tandy can handle the load).

Reply 27 of 264, by Dr.Yak

User metadata
Rank Newbie
Rank
Newbie
Fanatik wrote:

It's only an 8088 cpu. I guess I'm stuck with the opl2lpt.

A yeah.
So you're stuck with music-only on the parallel port using OPL2LPT or future OPL3LPT and patched software/drivers.

You should definitely try to find some PLUS-to -ISA adapter so you could plug in an ISA Soundblaster if you want digital audio.

Reply 28 of 264, by dreamblaster

User metadata
Rank Oldbie
Rank
Oldbie

Hi all,
I need your help :
Can you listen and compare the 3 tracks in this playlist ? : A,B and C
https://soundcloud.com/dreamblaster/sets/opl3 … pt-blind-test-1
This is a blind test, so I cannot tell you more about the details\differences of recording, if any.

Which track sounds best, and why ?
Do you hear clear differences ?

Visit http://www.serdashop.com for retro sound cards, video converters, ...
DreamBlaster X2, S2, S2P, HDD Clicker, ... many projects !
New X2GS SE & X16GS sound card : https://www.serdashop.com/X2GS-SE ,
Thanks for your support !

Reply 29 of 264, by keropi

User metadata
Rank l33t++
Rank
l33t++

I can't decide over A or B , they seem to be "fuller" sounding

🎵 🎧 PCMIDI MPU , OrpheusII , Action Rewind , Megacard and 🎶GoldLib soundcard website

Reply 30 of 264, by chartreuse

User metadata
Rank Newbie
Rank
Newbie

They're really close, they all sound quite nice. If I had to pick just one I'd go with C, the filtering of the highs helps to balance out the sound. I love the bass punchiness on A, but it almost drowns on the mids (at least on headphones). B is quite nice, but some of the high frequency sounds are bordering on grating with headphones. (Don't get me wrong I love the harsh sound of OPL3 but I don't think it's the best fit for some of these game musics)

If I had to guess what the differences were, I'd think that A is unfiltered, with a slight bass boost equalization. B is unfiltered, but no boost mostly a flat equalization. While C has some high frequency filtering on it.

Reply 31 of 264, by Utafuinki

User metadata
Rank Newbie
Rank
Newbie

> Hi all,

Hi!

> Can you listen and compare the 3 tracks in this playlist ?

Affirmative.

> Which track sounds best, and why ?

The differences are quite subtle. I'd say B sounds like OPL2 instead of OPL3, and A and C sound very similar to each other. B also looks different from the others, while also visually resembling the OPL2 more than the others.

C seems to have more 'impact' at some parts, and sounds almost 'louder' or 'brighter' than the others, B definitely sounds more 'thin' (if it was a Pulse, it would sound like it has a smaller width value than the others).

B's visuals are sometimes almost "reversed" (in Y-axis) compared to A and C, which look identical, but B has sometimes slightly higher 'peaks', visually speaking.

ABCO_Comparison.png

I can't hear the differences that well (could be because I am using earplugs because of my noisy neighbours - it sort of tunes out the higher frequencies as well), and if I had to compare between A and C purely aurally, I couldn't necessarily always hear which is which, they're so similar to my ear (though C seems to have a bit more 'impact').

But I would easily hear the difference in B.

Using Leisure Suit Larry III: Passionate Patti in the Pursuit of the Pulsating Pectorals might not be the optimal choice to test OPL3 or compare it to OPL2, because it sounds very good with OPL2, and I am not sure how well Sierra On-Line optimized their musics, instrus or sound effects specifically for the OPL3.

Perhaps something that utilizes the OPL3 in a more specific and optimized way might be a better candidate for tests like these (though I have no idea what that could be - perhaps Madbrain's Adlib Tracker II tunes? I am sure someone here can extend better knowledge about this topic).

> Do you hear clear differences ?

Depends on the definition of 'clear'. The only reason I was able to hear any differences, was because I focused on listening for exactly that. If it had been just casual playing, I wouldn't have heard the differences whatsoever.

For what it's worth; every one of them sounds good enough in my opinion - can't go wrong with any of them.

If I had to choose, then C would be my favorite.

Reply 32 of 264, by dreamblaster

User metadata
Rank Oldbie
Rank
Oldbie

thanks for the deep analysis,
don't look at the waveforms, purely judge by your ears.

my conclusion is, it all sounds very close
so what you vote, go for sound C ? or sound A ?

Visit http://www.serdashop.com for retro sound cards, video converters, ...
DreamBlaster X2, S2, S2P, HDD Clicker, ... many projects !
New X2GS SE & X16GS sound card : https://www.serdashop.com/X2GS-SE ,
Thanks for your support !

Reply 33 of 264, by Utafuinki

User metadata
Rank Newbie
Rank
Newbie
dreamblaster wrote:

thanks for the deep analysis,

You welcome.

don't look at the waveforms, purely judge by your ears.

Well, I am a visual (as well as aural) individual, I need to always see everything in detail. So I'll always be looking to get more information.

I had to record these sounds to wavs to be able to listen to them more conveniently (maybe you could've uploaded them as a zip file somewhere), and I use a wav editor to play wav files in the most pure and authentic way (some people say that certain music players distorts the sound), so it's only natural that they're visually in front of me.

Don't worry, though, from a 'distance', they all look the same to me anyway - I only went into a deeper visual mode after already listening to all of them multiple times and trying to compare. In fact, these visuals are only result of me trying to hear it better, by playing only a very tiny section of the song. Zooming in on a small sound sample from the whole and playing that over and over again makes it easier to hear the difference.

The side-effect of that is that I need to visually 'match' the waveform, so that it plays the exact same portion at the exact same place, so I need to edit the selection to be as perfectly identical as possible.

As a side-effect of -that- side-effect, I start noticing the differences in the waveforms themselves.

So it's not that I try to assess them purely visually, it's just that the visual happens as a natural side-effect of the flow that's necessary to use in order to try to hear the difference.

Don't worry, though, I did assess them purely by ear, it's just that the visuals of them are quite interesting to me regardless. I did select "C" as my favorite because of how it sounds, not by how it looks.

However, what's wrong with analyzing it visually anyway? If you want to know the differences between A, B and C the most accurately and in detail, the visuals help. The visuals can tell things that my low-quality ears can't, so it can help further the analysis to a level that wouldn't be reached otherwise.

If you want to really find out the differences and figure out which wav file has what properties and attributes, why deny the visual side completely? I realize it's audio we're talking about, but it can also add detail that the purely-aural-side can't reach. I'd think ANY way of analyzing these three things would be beneficial if we really want to scrutinize the differences between them and reach the final truth as to which wav file is what and in what way.

In any case, as I mentioned previously, B is the only one that sounds 'distinctly different' (OPL2?), and A and C sound very similar, but C has more 'impact', so it's my favorite. But I wouldn't be complaining if the song sounded like A or B, as they all sound very good.

The thing about the visuals is, after I've judged 'purely by ears', I want to know more (I am curious), and I want to go where my ears can't reach, I want to go the distance, I want to find out the most detailed and deep reality of these wavs that I can.

Of course judging sound purely visually would be completely wrong, but I don't see why visuals can't help after the fact.

I don't see how it detracts from the judging or invalidates anything if I know this wav looks like this, and that wav looks like that, when I am listening. All it tells me is that I am now listening to A, and now I am listening to B. It's my ears that do the hearing regardless of what I am looking at. I could be looking at a beautiful scenery or an oldskool computer, it wouldn't make any difference to how I hear it. It's not like one wav looks so pleasant that it makes it sound better, or one wav looks so unpleasant that it makes it sound worse.

The sound doesn't change from the visual (but I DID do a lot of 'closing my eyes while listening' when trying to hear the differences, and it didn't have an effect - everything sounded the same eyes closed and eyes open).

Besides, I figured the 'blind' part of the test simply means that we're not given any INFORMATION about what the wavs are, to retain impartiality (some might love a certain setting, chip or revision, and would be biased towards that if they knew all the info, and thus give a biased viewpoint - it's like in music competitions, where they won't show the name of the author until after voting, but there wouldn't be any harm in showing the song's waveform or structure in a tracker or whatnot), not a LITERAL 'blindness from all visuals'.. (:

I don't think you have anything to worry about; no one is going to judge these wavs purely by sight, the visual is just an interesting detail that helps the inquisitive mind to get some satisfaction. The judging and voting will always be based purely on how it sounds. If it was pretty, but sounded like a frog flatulence, I wouldn't favorite it. If it looked ugly, but sounded like heavely choir of angelic interdimensional energy entities, I would definitely favor it.

Also, it's always fun to bring in some visual to these posts.

so what you vote, go for sound C ? or sound A ?

As I said, my favorite is C. But everything sounds good.

There's not a -huge- difference between them, though - it's not like you can definitely say "ah, the chord sounds really bad with this wav, and really good with this other one", or "oh, the bass sounds so plain in this one, but so juicy in this one".

I hope this helped. I registered here and wrote these posts purely from the motivation of wanting to help.

As a sidenote, is there going to be a possibility for customizing where the customer wants to have the volume knob in OPL3LPT? I'd rather have it on the side and not back for pragmatic reasons.

Finally, I want to say that it's _VERY_ exciting that You are doing this type of wonderful products that are sorely needed, and that fill a deep void in the life of a musician that has too modern PCs for ISA cards, but still wants to use an authentic OPL(2 or 3) chip to create music, sound effects and experiments.

Heck, I just recently realized that some of my favorite arcade games used the OPL2 chip, so it's possible to play (and capture) the VGM-versions of their songs with the OPL2LPT. That's amazing to me - playing songs that weren't even made for the PC, as authentically as possible (Truxton is a good example). Mind-blowing! (and delicious)

Lots of Kudos and respect to You, Dreamblaster!

Reply 34 of 264, by dreamblaster

User metadata
Rank Oldbie
Rank
Oldbie
Utafuinki wrote:

I hope this helped.

This helps a lot !!
I have a suggestion : do a real 'blind' test : have someone else click A B or C, and you guess if it is A B or C
Would C still have the most impact ?

My initial idea was indeed a sidewards pointing potmeter, like on opl2lpt
but I got feedback from a 2 people, that the another cable obstructed the pot, and it would better be pointing backwards.

in the latest iteration, I found a good compromise :
The potmeter can be operated from the back and from the side !
(there is a cutout on the pcb under the pcb, you can easily operate it from the side

Clipboard01.jpg
Filename
Clipboard01.jpg
File size
173.35 KiB
Views
3277 views
File license
Fair use/fair dealing exception

Visit http://www.serdashop.com for retro sound cards, video converters, ...
DreamBlaster X2, S2, S2P, HDD Clicker, ... many projects !
New X2GS SE & X16GS sound card : https://www.serdashop.com/X2GS-SE ,
Thanks for your support !

Reply 35 of 264, by dreamblaster

User metadata
Rank Oldbie
Rank
Oldbie

okay so what should I choose ?
A or C ?

the majority says C ?

in my own blind test, there was a slight advantage for C,
but not really decisive

Visit http://www.serdashop.com for retro sound cards, video converters, ...
DreamBlaster X2, S2, S2P, HDD Clicker, ... many projects !
New X2GS SE & X16GS sound card : https://www.serdashop.com/X2GS-SE ,
Thanks for your support !

Reply 39 of 264, by dreamblaster

User metadata
Rank Oldbie
Rank
Oldbie
Jonas-fr wrote:

Very nice to see that the project is going well. Is the usb still a mini port or you're cswitching to the more common micro variant?

it is usb mini port (same as opl3lpt)
I find it more robust than micro.

Visit http://www.serdashop.com for retro sound cards, video converters, ...
DreamBlaster X2, S2, S2P, HDD Clicker, ... many projects !
New X2GS SE & X16GS sound card : https://www.serdashop.com/X2GS-SE ,
Thanks for your support !