Thanks for the update 640k!enough !!!

Thanks a lot 640k!Enough ! 😃 With your workaround I can continue my game of Simon The Sorcerer ][ enjoying it's awesome music on the sc88 without any ambient sound FX shredding my ears to bits. I'll keep my attention on a mixer that would allow me to adjust on the fly any sound source I have and this poor man mixing solution will be history.

I have not conduct much test but I was curious to compare CrystalFM to OPL3 volume for a same setting and it seem the OPL3 is louder than the CrystalFM, so maybe for game that use the OPL3 the mixing between the SB-PCMVol and OPL3 would be allright.

I notice the orphinit 53b (while not 52a and 53b-beta) output this string, I'm not sure what to make out of it :
WARNING: Plug and Play read data port changed from 3E3 to 3E7.
Board successfully initialised.

ElBrunzy wrote on 2021-04-27, 01:22:

I notice the orphinit 53b (while not 52a and 53b-beta) output this string, I'm not sure what to make out of it :
WARNING: Plug and Play read data port changed from 3E3 to 3E7.
Board successfully initialised.

I think I know why that is happening, but I am not absolutely sure. Please run the following command and post the resulting file:
ORPHINIT /vd > output.txt

I would like to have a more detailed look to see if there is something that can be further refined for the next verison.

There you go sir, I have no clue what you have in mind, but if you want anything else I'll be just glad to have the chance to help.

ElBrunzy wrote on 2021-04-27, 15:18:

There you go sir, I have no clue what you have in mind, but if you want anything else I'll be just glad to have the chance to help.

There seems to be something rather strange about the way your BIOS handles PnP support; I'm not sure if this points to another bug I have to fix, or if there isn't much to be done here. If you can tolerate one more request, can you attach the report, as I asked for in this post?

EDIT: I should have mentioned that this will be most useful if you create the report right after a cold boot, without first running ORPHINIT, UNISOUND or any other such tools.

Thanks for the EDIT, I was about to run the dump utility after init the card, silly me...

This i845 motherboard was nothing but trouble since I got it lately. Now things start to settle down and I can finally start to enjoy it. There is thing I have never seen before in this Phenixs AWARD bios, for instance I need to set DMA 1 in legacy ISA mode otherwise everything seem to work fine but I got no digital audio in some dos programs (like warcraft 1/2). At least I have a working Winbond W83628F W73629D PCI-to-ISA bridge. I was never able to identify this Chinese motherboard. That's why I was relieved someone with a more appropriate motherboard could confirm the SB-VoicePCM problem. And why I was selfish and say that this mixer issue was not really problematic, I was comparing to the other problems I faced with that motherboard (and I save you the hunt for a working AGP card or ram compatibility problem...). Anyway, who am I to judge of the severity of an issue ? 😅

I hope this dump is what you expect and I wish it help you with your project. Once again, if you need anything else, just ask!

ElBrunzy wrote on 2021-04-28, 02:41:

I hope this dump is what you expect and I wish it help you with your project. Once again, if you need anything else, just ask!

Thanks for the dump details. It provided a few hints about your system, and helped to confirm that ORPHINIT was not retrieving incorrect information, but the way that board handles PnP ISA devices is somewhat strange. I'm not sure if there is another obscure setting that would correct the behaviour, but it's probably not worth taking the time to look for it if you're not having trouble with other software or devices.

This whole investigation did lead me to make a few more small improvements to ORPHINIT, some of which may give better results on your system. Please be sure to update when the next release is ready.

Got everything up an running, but I have one issue: in SB mode at least, PCM is way louder than everything else, even with PCMVol at 1, FMVol at 7 and LineVol at 7. Am I doing something wrong, or is this just how sound blaster mixers work?

FortMax wrote on 2021-07-26, 23:49:

Got everything up an running, but I have one issue: in SB mode at least, PCM is way louder than everything else, even with PCMVol at 1, FMVol at 7 and LineVol at 7. Am I doing something wrong, or is this just how sound blaster mixers work?

In all likelihood, you are not doing anything wrong. This is a known issue that appears to occur when the OPL3 synthesiser is active; it appears to be a bug in the Crystal chip, while the older, genuine Creative cards behave correctly in this regard.

For software that doesn't actually use the FM synthesiser, you can either use FMMode=off or FMMode=CrystalFM to regain volume control for SB-PCMVol. If you need FM and don't want to settle for Crystal FM, there is currently no work-around. However, I am still experimenting, and there may be a way to abuse one of the modes of the chip to approximate FM volume-level control in Sound Blaster mode. If it is workable, this will not likely be in the upcoming version (which I expect to be ready in a few weeks), but maybe the one after that.

EDIT: I remember now that this was discussed at length a short while ago. I stand corrected — this cannot be fixed in the way that I was thinking; that was for another, mostly unrelated, problem. This would require a good bit of hardware-level experimentation, and it was decided that we didn't want to go down that road. So, for this, the above-mentioned work-arounds are the only options currently available. If there is another way to fix it, I have yet to discover it.

Hey 640K!enough, I just got around to installing my Orpheus and attempted to load Orphinit. I received the error..."unidentified chip found; quitting" error. I ran the program with the defaults. I did not edit the .ini and I have not used any other versions or different drivers.

I've searched and couldn't find this issue previously discussed.

S_crumb wrote on 2021-08-28, 19:47:

I received the error..."unidentified chip found; quitting" error. I ran the program with the defaults. I did not edit the .ini and I have not used any other versions or different drivers.

This could happen for a number of reasons, but we'll start by assuming that it will be one of the simpler possibilities: a resource conflict. Firstly, are there any other sound devices in the system? Even if not a separate card, some systems used to ship with on-board audio hardware that could be causing a conflict.

For now, this part is done by accessing the control registers on the card, so the first thing to try is changing CtrlBase. Try replacing the line in ORPHEUS.INI with CtrlBase=120, save your changes, run ORPHINIT again and see if there is any difference. If not, we'll have to look into some of the other possibilities.

Can you describe your configuration in more detail (processor, speed, motherboard or brand/model of machine, etc.)?

Thanks for taking a look.

The machine is a Gateway 2000. 486 DX/2 66 mhz. 16 mb RAM. Phoenix Bios 4.05.03. Running a CF adapter in lieu of a hdd and has a Gotek. A 1mb vga card. No other sound cards.

I made the change you suggested and got the same result.

edit: I decided to give Unisound a try and it doesn't work either. I set my variables in my autoexec.bat.....the initialization goes smoothly. Unisound detects the card successfully but no software can see the card. For example Duke Nukem 3d's sound config allows me to select the card and configure it. The Sound Fx test hangs then fails and the Music test says that it cannot detect an FM chip. So if Orphinit and Unisound don't work then there must be another issue....probably operator error.

I got serious today and stripped the machine down. I swapped my cards around and put the orpheus into a different ISA slot. Magically it works....to a point. Now with Orphinit and Unisound theres a crazy echo. I'm sure this is a known issue that I haven't gotten a chance to research.

S_crumb wrote on 2021-08-30, 18:34:

I got serious today and stripped the machine down. I swapped my cards around and put the orpheus into a different ISA slot. Magically it works....to a point. Now with Orphinit and Unisound theres a crazy echo. I'm sure this is a known issue that I haven't gotten a chance to research.

hmmm
no this is not a known issue
I would suggest you clean your isa slots and the edge connectors of your cards , if moving them made the card work then it could mean there are contact issues , doesn't hurt to try!

there was 1 incident of such (similar?) behavior - when using some PCI IDE controller : apparently there was some addresses conflicting and sound was distorted heavily
once the controller was replaced with another model all was fine

The issue seems to be surrounding the configuration of Sound FX. The sound fx echo uncontrollably. No matter what type of Sound Blaster that I choose within the configuration, I get one of two results...either a failed sound fx test or uncontrollable echos.

Music seems to play fine.

I chose a game that would attempt to autodetect my hardware and it identified the card as an Adlib Gold. That was odd.

This occurs under Orphinit/Unisound/Crystal drivers. So I guess technically these posts no longer belong here.

edit: This motherboard is slightly odd as it uses an ISA riser card. Placing the card in different slots results in different errors. I'm going to either reflow the connectors or buy a new rise card.

I just put together a simple test that should help in diagnosing the problems that some of you have been reporting lately. For the moment, it is very simple and inflexible, so please pay attention to the details below:

1. The WSS base address must be 534H.
2. This test uses simple programmed I/O to play back a simple waveform, so no DMA or interrupts are needed. If we can communicate with the card at all on those ports, it should work.
3. Start with low volume, using cheap amplified speakers, to avoid damage to your speakers or hearing, should something go wrong.
4. Press q to silence the playback and return to DOS.

If you wish, you can give this a try. It will play a simple sine wave at approximately 440 Hz, using WSS mode. This will tell us if we can get any PCM playback working, using the most basic method possible.

For anyone who also has an oscilloscope, feel free to have a look at the generated waveform, measuring its frequency. If everything looks (or sounds) good, there should be no reason we can't successfully play back digital audio with the card.

Thank you 640k!enough. I'll give this a try for thoroughness' sake.

Though this is looking more like a hardware issue. I thought I might have issues with my riser card, but I've ruled that out. I swapped in a Vibra16 card and it plays flawlessly. The Orpheus plays music perfectly but the sound effects echo.

but did you clean the isa slots?
assuming the card is bad it does not explain moving it in other slot and getting it half-working
if you did then you can return it for checking to replace or refund , I cannot think of anything else to help you

If I had to guess about the problem with this Gateway machine, and the suspicion is that it isn't a configuration or init problem, it could certainly be a problem of worn or loose contacts within the slot. This is supported by the fact that moving/re-seating the card gave different behaviour. The Creative card working doesn't disprove this hypothesis, since it may be fractionally thicker, thereby making more solid contact between card edge and slot contacts.

It's not something that most people have sitting around at home, but a precision caliper could lend additional support to the idea, by measuring the thickness of both cards, including ISA "fingers".

EDIT: Thinking about this some more, another possibility is that the Orpheus is not able to make proper contact in the slot due to the way the metal bracket was screwed on. Maybe try re-installing it, making absolutely sure that it is installed in such a way as to allow the card to sit as far into the slot as possible, while being as close as possible to perfectly perpendicular to the riser card. I have seen cases like this before.

I've cleaned the contacts and tried it in multiple slots. I agree with you 640k...I was thinking along the same lines...but I was thinking since the two cards use a different configuration of contacts the pins the Orpheus needs to engage don't need to be engaged on the Vibra.....you get where I'm going.

To cover all of my bases, I ordered a new riser card, albeit a slightly different revision.

The easiest solution for me would just be to solder in brand new connectors but I can't find a source for new ISA edge connectors.

If anyone knows where I can buy some, let me know!