• Yes. Not sure about the second part of your question, but If you are asking how the digital signals are converted to analog, then the answer is "usually sound card", unless you use the digital output of your soundcard.
• IDE ribbon cable, though in theory SPDIF should also be OK (never used this output in practice).
• The analog output has minimal impact on performance, it goes straight to the sound card. This could make some difference on an old computer. On the other hand, in an old OS you wouldn't be able to use the "digital" (over IDE) playback, not sure when it was implemented in Windows.
• In Windows 98, in Control Panel, Multimedia applet, CD Music tab you have to check "Enable digital audio playback for this CD-ROM device".
• Don't know about the WDM drivers.
• Usually it will offer the best quality.

• If using the 4-pin analog, then yes, the signals are processed by the optical drive's DAC. Not sure how the 2-pin signals are processed, though (maybe it's the same path as the 4-pin, just that the audio is not processed at the optical drive level but instead at the sound card level).
• The digital way (IDE or the 2-pin digital cable) should produce better sound quality than using 4-pin, though it's not really that noticeable in my opinion.
• The 4-pin works under all operating systems (DOS, Windows, etc.) and with the 4-pin, the audio playback is mostly handled by the optical drive itself, so there would be significantly less load on the optical drive, the CPU, and the IDE bus.
• For Windows 98, same as GL1zdA's answer (Control Panel -> Multimedia -> CD Music -> Enable digital audio playback). If this option is checked, the audio goes through the IDE cable; If unchecked, the audio goes through the 4-pin wires.
• For best compatibility, I think you do need WDM to make sure the Enable Digital Audio Playback option isn't greyed out.
• I sincerely don't suggest using the IDE cable mode, as this puts the drive (possibly the IDE Bus) and the CPU constantly under high load. While it certainly can provide better audio quality, the loud noise from the optical drive spinning at full speed alone could be considered a degradation, not to mention possible system lags and audio stutters with older CPUs when playing games with CD Audio as the BGM. Use the 4-pin or 2-pin whenever possible (unfortunately with later Windows and SATA optical drives this is no longer the case).

Wow. Thanks guys! That answered most of my questions.

However, your answers regarding my fourth question bring up another question. You see, in my build, which you can find here, I have no cables whatsoever (either 4-pin analog or 2-pin digital) going to either my Sound Blaster Live! or AWE64 Value soundcards, and yet, even though "Enable digital audio playback" is greyed out, whenever I play games that utilize a CD (such as The Need for Speed SE (which uses the latter soundcard), Need for Speed 2 SE and Diablo (which use the former)), I still get music and such. So, therefore, I have a "counter-question" to those answers, as you will: As that option only applies to my Sound Blaster Live! card, does that option allow for the transfer of digital audio over the 2-pin SPDIF cable, allow for the digital output port of the card to output digital sound straight from the CD/DVD drive, or allow for the transfer of digital audio over the IDE cable? I'm confused. 😀

As for the fifth question, I wanted to double-check that was the case because I tried a WDM driver from a Sound Blaster Live! CD hosted on the VOGONS driver archives, and, although it un-greyed out that "Enable digital audio playback" option, the result of using that driver that volume increased explosively exponentially when increasing the volume level in the taskbar, with after around 10 being pretty loud. Obviously, I did not want to use a "broken" volume system, so I uninstalled that driver and reinstalled the latest VXD (?) driver from Creative's website. Hopefully, when I get around to building my second Windows 98 SE/Windows XP retro gaming machine and I try to use WDM drivers with my Sound Blaster Audigy 2 ZS, the same thing does not happen.

Finally, for the sixth question, in conclusion, are you saying LSS10999 that I should use the 2-pin SPDIF instead of IDE?

Thanks again in advance!

the_ultra_code wrote:

Wow. Thanks guys! That answered most of my questions. […]
Show full quote

Wow. Thanks guys! That answered most of my questions.

However, your answers regarding my fourth question bring up another question. You see, in my build, which you can find here, I have no cables whatsoever (either 4-pin analog or 2-pin digital) going to either my Sound Blaster Live! or AWE64 Value soundcards, and yet, even though "Enable digital audio playback" is greyed out, whenever I play games that utilize a CD (such as The Need for Speed SE (which uses the latter soundcard), Need for Speed 2 SE and Diablo (which use the former)), I still get music and such. So, therefore, I have a "counter-question" to those answers, as you will: As that option only applies to my Sound Blaster Live! card, does that option allow for the transfer of digital audio over the 2-pin SPDIF cable, allow for the digital output port of the card to output digital sound straight from the CD/DVD drive, or allow for the transfer of digital audio over the IDE cable? I'm confused. 😀

As for the fifth question, I wanted to double-check that was the case because I tried a WDM driver from a Sound Blaster Live! CD hosted on the VOGONS driver archives, and, although it un-greyed out that "Enable digital audio playback" option, the result of using that driver that volume increased explosively exponentially when increasing the volume level in the taskbar, with after around 10 being pretty loud. Obviously, I did not want to use a "broken" volume system, so I uninstalled that driver and reinstalled the latest VXD (?) driver from Creative's website. Hopefully, when I get around to building my second Windows 98 SE/Windows XP retro gaming machine and I try to use WDM drivers with my Sound Blaster Audigy 2 ZS, the same thing does not happen.

Finally, for the sixth question, in conclusion, are you saying LSS10999 that I should use the 2-pin SPDIF instead of IDE?

Thanks again in advance!

I think those games you mentioned do not involve Redbook CD audio (which are separate tracks from the data track, and normally not visible to anything other than CD audio players). Their music data are stored as binary data (installed to HDD or loaded directly from the CD Drive), and has nothing to do with the 4-pin or 2-pin cables. Those data are always transmitted through IDE just like any other data, as they're technically files.

I'm not entirely sure about the 2-pin SPDIF as I haven't tested it myself. I mostly use 4-pin analog wires whenever possible for better compatibility and availability. Personally I could not hear any noticeable quality difference with the 4-pin analog output.

If you don't have games that plays Redbook CD audio as BGM then I think you don't need the 4-pin or 2-pin wires, since those are only applicable to Redbook CD audio playback and nothing else.

the_ultra_code wrote:

[*] What is the benefit of using the 4-pin analog wire over the digital means of transmitting audio?

This is mainly a legacy interface. When the first CD-ROM drives became available, most sound cards didn't have CD-quality DACs yet, let alone SPDIF support (we are talking SB Pro-era, older cards didn't even have a CD input at all). So direct IDE transfer or SPDIF transfer of digital audio simply wasn't an option. It would have to be downsampled first, which at the time would be too resource-intensive.
The analog cable was an easy and compatible way to connect the CD audio directly to the analog onboard mixer of any sound card.

the_ultra_code wrote:

[*] After plugging in the 4-pin wire into both the optical drive and a soundcard, are three things that need to be set to within an OS to have that 4-pin to work?

You'll need to make sure the CD input on the mixer is not muted, and the level is set so you can hear something.
You can test by muting the CD input whether it is being used or not.

Again, thank you guys for the answers!

Okay, two more things.

One, to phrase it differently, what does the "Enable digital audio playback" do exactly, in-depth?

Two (this one could be subjective), is an optical drive's DAC better than a soundcard's?

I cannot think of any other questions at the moment.

LSS10999 wrote:

I think those games you mentioned do not involve Redbook CD audio (which are separate tracks from the data track, and normally not visible to anything other than CD audio players). Their music data are stored as binary data (installed to HDD or loaded directly from the CD Drive), and has nothing to do with the 4-pin or 2-pin cables. Those data are always transmitted through IDE just like any other data, as they're technically files.

That would make sense.

Scali wrote:

You'll need to make sure the CD input on the mixer is not muted, and the level is set so you can hear something.
You can test by muting the CD input whether it is being used or not.

Got it! 😀

Again, thank you everyone for your help, and thanks Scali for the history lesson - I like to learn little factoids like that 😀.

the_ultra_code wrote:

One, to phrase it differently, what does the "Enable digital audio playback" do exactly, in-depth?

Technically, with this checked, the system will read the Redbook CD audio data directly through the IDE cable to the memory and let the CPU to resample and feed the digital data to the sound card, just like playing any other audio files (.wav for example) on the hard drive or optical drive. In this way, the 4-pin analog cable is not required, and all the processing work is done on the system rather than on the optical drive.

the_ultra_code wrote:

Two (this one could be subjective), is an optical drive's DAC better than a soundcard's?

Like Scali said, back then sound cards like SBPro can only do 22.05kHz stereo, which is below CD Quality (44.1kHz), so if you're comparing optical drive's DAC with the sound cards at that time, the optical drive's DAC would be relatively better, in terms of capabilities.

Recent onboard and discrete sound card's DACs, however, are most likely better than the one on the optical drives, with the capability to sample at 48kHz and possibly higher, and with SNRs often reaching over 100dB. This is, without taking the "forced resampling" (here, and here) from 44.1kHz to 48kHz into account. (Personally to what extent the issue can impact the audio quality is a matter of opinion, and it depends on the hardware design of the sound card.)

Also, with the modern CPU's processing power, there's virtually no need for the 4-pin analog cables if you're playing Redbook CD audio on supported operating systems (you'll still need those if you want to support legacy operating systems such as DOS, and the connectors are often provided even for onboard audio cards).

As the usefulness of the optical drive's own DAC diminishes, SATA optical drives no longer include onboard DAC (thus no more 4-pin connectors anymore).

the_ultra_code wrote:

Two (this one could be subjective), is an optical drive's DAC better than a soundcard's?

I think it's safe to assume that the DAC on a CDROM drive was mainly a 'checkbox feature'... They put it on there because it had to be there. They probably just picked some cheap generic DAC, and probably also didn't bother to put a lot of work into optimizing the analog output for good sound quality.
Then again, many early sound cards had appalling sound quality as well. But there were some decent high-quality ones around, like Turtle Beach, and Pro Audio Spectrum 16. Those will probably be better than your average CDROM.

I'm back from enjoying a Thanksgiving meal. It was deliciously filling 😀.

Anywho, back on topic.

Taking in what you guys have told me, and doing some light research, I have come to the conclusion that, at least with my build, I probably should continue to let my soundcards handle DAC-related tasks, given one they have DAC "frequencies" (is that the correct verbage?) above or equal to 44 kHz (SB AWE64 Value has around a 44 kHz I think, because it has the same frequency as the AWE32; SB Live! has a 48 kHz frequency), so the IDE optical drive would not do much to help in the sound quality department if it handled DAC-related tasks (I have not looked up the DAC inside of the IDE optical drive my build is using; I am assuming it is either on par or worse at the task than my soundcards are), and two, I have such a powerful processer (Pentium III 600E 600 MHz), that having my system + soundcards handle Redbook audio (which is a concept I am now just understanding), both in DOS and in Windows 98 SE, should not have too much of a performance impact in-game.

Is this a reasonable conclusion? I would very much appreciate your opinions.

It's funny. Every time I think I'm out of questions, more come to my mind, so I have yet another question (sorry). Here is my new question: Let's take a moment to think about how audio played from actual files stored on a CD/DVD works. If I'm correct, that non-Redbook audio-file "data" is sent over the IDE cable normally and is played by the system and/or soundcard. So, my question is, if I use a 2-pin SPDIF cable from my optical drive to my SB Live!, would my optical drive send that data over the SPDIF cable straight to my soundcard instead of over the IDE cable? And if that is the case, wouldn't that help performance and potentially sound quality (since that signal is not being sent next to 30+ other signals which could distort that signal)?

the_ultra_code wrote:

It's funny. Every time I think I'm out of questions, more come to my mind, so I have yet another question (sorry). Here is my new question: Let's take a moment to think about how audio played from actual files stored on a CD/DVD works. If I'm correct, that non-Redbook audio-file "data" is sent over the IDE cable normally and is played by the system and/or soundcard. So, my question is, if I use a 2-pin SPDIF cable from my optical drive to my SB Live!, would my optical drive send that data over the SPDIF cable straight to my soundcard instead of over the IDE cable? And if that is the case, wouldn't that help performance and potentially sound quality (since that signal is not being sent next to 30+ other signals which could distort that signal)?

I believe in theory there will be a performance difference yes. IDE data will be received by the CPU first, and then has to be sent back to the sound card.
SPDIF is a direct input on the card itself.

Sound quality will not be affected, since it is digital. It's all either 1s and 0s. If there was a problem with distortion in the IDE cable, then you would have problems accessing data from the CD as well. Just one bad bit can be enough to crash a program, or at least show heavy glitches.

the_ultra_code wrote:

... that having my system + soundcards handle Redbook audio (which is a concept I am now just understanding), both in DOS and in Windows 98 SE, should not have too much of a performance impact in-game.

Be noted that DOS does not have the capability to play Redbook audio through the IDE cable. Only Windows 9x and later can.
There will be a performance impact, depending on the slowest part of the workflow (Optical Drive -> IDE Bus -> Memory -> CPU -> Sound Card).

the_ultra_code wrote:

It's funny. Every time I think I'm out of questions, more come to my mind, so I have yet another question (sorry). Here is my new question: Let's take a moment to think about how audio played from actual files stored on a CD/DVD works. If I'm correct, that non-Redbook audio-file "data" is sent over the IDE cable normally and is played by the system and/or soundcard. So, my question is, if I use a 2-pin SPDIF cable from my optical drive to my SB Live!, would my optical drive send that data over the SPDIF cable straight to my soundcard instead of over the IDE cable? And if that is the case, wouldn't that help performance and potentially sound quality (since that signal is not being sent next to 30+ other signals which could distort that signal)?

After a bit searching, it seems the 2-pin cable most likely does the same job as that of the 4-pin, just without going through the onboard DAC. The decoding is done on the sound card instead, also without the system CPU's intervention. In such case, the audio quality is determined by the sound card's DAC.

So no, those cables (2-pin and 4-pin) are only applicable to Redbook audio and nothing else. Non-Redbook audio files can only go through the IDE cable, and possibly needed to be decoded by the CPU before it can be played (such as MP3).

I'm back. Sorry for taking so long.

Anywho, I want to make sure all that has been said here regarding the different means of CD audio playback is in a nice little "list", as you will:

• Non-Redbook Audio:
  • Using Over-IDE Playback:
    Path: Optical Drive -> IDE Cable -> Motherboard (includes CPU and Memory) -> Soundcard -> Soundcard's DAC -> Audio Out (i.e. the port where we hook up our headphones).
• Redbook Audio:
  • Using Over-IDE Playback (with "Enable Digital Audio Playback" Enabled in OS):
    Path: Optical Drive -> IDE Cable -> Motherboard -> Soundcard -> Soundcard's DAC -> Audio Out
  • Using Over-Analog (4-pin Analog Cable) Playback:
    Path: Optical Drive -> Optical Drive's DAC -> Soundcard -> Audio Out
  • Using Over-SPDIF (2-pin Digital Cable) Playback (Best Option):
    Path: Optical Drive -> Soundcard -> Soundcard's DAC -> Audio Out

*Update: this list has been updated per other users' comments down below.

Please let me know if anything in the above list is incorrect, it's missing parts and pieces, or there is another method I missed.

And another two questions. LSS10999, going off on what you said earlier::

the_ultra_code wrote:

One, to phrase it differently, what does the "Enable digital audio playback" do exactly, in-depth?

Technically, with this checked, the system will read the Redbook CD audio data directly through the IDE cable to the memory and let the CPU to resample and feed the digital data to the sound card, just like playing any other audio files (.wav for example) on the hard drive or optical drive. In this way, the 4-pin analog cable is not required, and all the processing work is done on the system rather than on the optical drive.

in terms of sound quality (assuming my listing above "Using Over-IDE Playback (with 'Enable Digital Audio Playback' Enabled in OS)" is correct), does this method produce better sound than without the feature on while still going over IDE, because it is using the CPU to do DAC than the optical drive? And does this method "perform better" than without "Enable Digital Audio Playback" enabled while still going over IDE?

Again, thanks in advance!

the_ultra_code wrote:

[*] Using Over-IDE Playback: Path: Optical Drive -> Optical Drive's DAC -> IDE Cable -> Motherboard -> Soundcard -> Audio […]
Show full quote

[*] Using Over-IDE Playback:
Path: Optical Drive -> Optical Drive's DAC -> IDE Cable -> Motherboard -> Soundcard -> Audio Out
[*] Using Over-IDE Playback (with "Enable Digital Audio Playback" Enabled in OS):
Path: Optical Drive -> IDE Cable -> Motherboard (with CPU Acting as DAC) -> Soundcard -> Audio Out

The CPU can not act as a DAC, and the optical drive can't send analog data from its DAC either. You can only send digital information over the bus.
The "Enable Digital Audio Playback" option simply enables over-IDE playback, as opposed to SPDIF or analog via the 4-pin cable.
So I believe there is only one variation here, and it looks like this:
Path: Optical Drive -> IDE Cable -> Motherboard -> Soundcard -> Soundcard's DAC -> Audio Out

Scali wrote:

The CPU can not act as a DAC, and the optical drive can't send analog data from its DAC either.

Correct. The CD-ROM performs DAE (digital audio extraction) on the redbook tracks yielding 44.1k/16b/2c PCM data. The CPU will perform any necessary format conversion before outputting via the sound card's PCM channels.

First of all, thanks for the correction! I have updated the list to reflect the new information.

However:

Scali wrote:

The "Enable Digital Audio Playback" option simply enables over-IDE playback, as opposed to SPDIF or analog via the 4-pin cable.

gdjacobs wrote:

Correct. The CD-ROM performs DAE (digital audio extraction) on the redbook tracks yielding 44.1k/16b/2c PCM data. The CPU will perform any necessary format conversion before outputting via the sound card's PCM channels.

I'm missing something here...

Okay, so if "Enable Digital Audio Playback" (I will now say "EDAP" for short 😀 ) is NOT enabled in the OS or if the OS does not support it, and NO audio cables whatsoever (either the 4-pin or 2-pin) are connected to the optical drive, and the only way to transfer Redbook audio is over IDE, would such a "path" look like this?

Path: Optical Drive -> Optical Drive's DAC -> IDE Cable -> Motherboard -> Soundcard -> Audio Out

the_ultra_code wrote:

Okay, so if "Enable Digital Audio Playback" (I will now say "EDAP" for short 😀 ) is NOT enabled in the OS or if the OS does not support it, and NO audio cables whatsoever (either the 4-pin or 2-pin) are connected to the optical drive, and the only way to transfer Redbook audio is over IDE, would such a "path" look like this?

Path: Optical Drive -> Optical Drive's DAC -> IDE Cable -> Motherboard -> Soundcard -> Audio Out

There is no such path.
The DAC is hardwired only to the 4-pin output and the headphone jack on the CD-ROM (if any). The IDE cable does not support any analog audio signals.
So if you get sound while not having a 4-pin analog cable or 2-pin SPDIF cable attached, then it MUST be doing DAE, if you can still hear redbook audio. That is the only way you can get audio via an IDE cable.

So then without "EDAP" enabled and no audio cables going from the optical drive to a soundcard, there would be no sound?

the_ultra_code wrote:

So then without "EDAP" enabled and no audio cables going from the optical drive to a soundcard, there would be no sound?

Indeed.

Got it!

Whoof. That was a lot of questions. 😀

I think all my questions regarding IDE optical drive audio playback have been answered, and anything regarding the technicalities about the different methods I should now be able to do research on by myself (I hope 😀 ). Thank you all guys very much!

Now, just a final note, not related to this topic's main, well, "topic": Were most of the games that used Redbook audio DOS CD games, or were there just about a relatively nice spread of games utilizing Redbook audio from the DOS-era to the XP-era? Just curious.