Ripping SACDs

Yes, I seem to be responding to my own message, but I did a little investigating.


A couple of pictures are illustrative...

The first, two dynamic envelopes of the entirety of "Money".


First, I took the Money .dsf file (a DSD file format for those that don't know) and converted it to 176.4k/24 PCM via Korg AudioGate 2, with no filtering and no gain normalization of the conversion. That's the bottom stereo pair. Next I ran the same AudioGate conversion, but this time with normalization...which works to not clip the maximum level. That's the top stereo pair. However, following the conversion I normalized the lower pair to the RMS level of the top pair--so the two renderings should play with approximately equal volume. When AudioGate normalized, it prevented clipping the peaks of the basic DSD signal. When I didn't allow AudioGate to normalize, the peaks were clipped. Incidentally, I also ripped the CD layer. The CD layer was clipped, and I'd have to say the clipping made the CD layer look exactly like the clipped stereo pair above (except for the RMS gain I induced for the picture).


Picture 2, the comparative spectrum curves of the two files above.



The two curves overlay exactly. So in this case, and looking at the entirety of Money, there is no difference in the overall spectrums of the clipped and unclipped renditions.

 

If the spectrum curves are over the entire duration of the song, then it is not surprising that the differences in the clipped areas are being masked. You could try comparing a small segment from one of the clipped areas, see if that turns up anything

I seem to recall that the CD layer of the hybrid was not liked at the time of release. Not sure what happened there as the Floyd team are usually so picky about sound quality, but then EMI in Europe has been releasing the wrong masters for years!! When I play this disc I only ever play the SACD layer.

 

The CD layer of the Dark Side Of The Moon SACD is known to be different in mastering than the DSD layer. Here's an analysis of the CD layer and DSD layer that Stereophile did: Dark Side of the Disc. The CD layer is louder and has more limiting than the DSD layer.

 

Thank you. That was good to read.

From the article, some key conclusions on the CD layer:

"Now, with the evidence from JA's graphs, I wonder what the point of limiting the CD's dynamics on a special release like this could possibly be. I've speculated that EMI may have wanted to give the CD layer more 'punch' since it is likely the one to be played on the radio. Or perhaps, as JA notes, EMI and Sony have conspired to place DSD in a more audiophile light with this manipulation—which is troubling when you start to ponder which other hybrids might have been altered in this manner.

"But, like JA, I'll guess the answer is actually more of a mundane 'business-as-usual' attitude at the CD mastering house. The paranoid audiophile in me suspects that the major labels now make it standard practice to push the audio level on all of their rock CDs to give them a more in-your-face sound. The evidence I've read in Mix magazine and that JA and others have gathered would support this contention. I can almost see the young EMI exec jabbing his finger at the mastering engineer and shouting, 'The audiophiles have got their prissy SACD layer, now make the other one ROCK!'"


Allow me to have a slightly different take on the situation. Possibly it's more a matter of sloppiness than something overt (for the radio crowd). Some pictures then, drawn from my pictures, my take.

Picture 1:


Two stereo pairs, both the entirety of "Money" (as displayed in Sony Sound Forge Pro 11, to keep things reasonably informative). The top pair is the result of using Korg AudioGate 2 on the Money DSD to render to 44.1k/16 PCM with no normalization. That is, the PCM was rendered directly from the DSD with no gain modification applied. The bottom pair is the result of ripping the 44.1/16 CD layer to PCM. Again, with no gain modification applied. The similarity between the two stereo pairs might be notable.


Picture 2:


The above is the result of running iZotope RX4 Advanced "Waveform Statistics" on the 44.1k/16 PCM rendered from the DSD by AudioGate 2. The statistics correspond to the top stereo pair in "Picture 1" above.


Picture 3:


The above is the result of running RX4 Advanced "Waveform Statistics" on the 44.1k/16 CD layer rip. The statistics correspond to the lower stereo pair in "Picture 1".


"Picture 2" and "Picture 3" have different values tabulated, though not greatly different. In particular note the LRA values which are the final element in each box. Yes, the two "Loudness range" values differ by only 0.4 dB. It's also worth looking at the Left versus Right "Total RMS level" in each box.

So here is my take... The DSD was mastered a bit loud, though not really a matter of concern while remaining within DSD-land. Based on the pictures above, however, the CD layer looks to have been the result of mapping the DSD layer directly to the CD layer...with no gain correction of the DSD beforehand. That's sloppiness, more care should have been taken. Of course the DSD could also have been mastered at a lower level, that could have worked, but the DSD layer doesn't seem broken (unlike the, say, U.S. released "Thriller" DSD layer). Actually, based on comparative channel Total RMS levels, the CD guys might have done a slight bit of work to bring the PCM rendered DSD channel levels into better balance.

Better attention should certainly have been paid in mastering the CD layer of DSotM, but it's possible there was nothing exactly diabolical about the CD mastering. Perhaps people were just not maintaining proper due diligence.

 

Edit - I posted this before seeing or reading the post immediately above.

Hi darkmass - I agree with you and Black Elk, that this idea of a comparison is probably doomed from the outset. Maybe there's a much simpler approach. You've done a lot of de-embedding and a lot of ISO ripping. Can I simply ask you, without a lot of charts and stuff, just to say in your experience whether there's a difference between the two methods in terms of accuracy? Is the one as close to bit-perfect as the other?

The reason I didn't suggest this before is because it unfairly places the burden on you and leaves me with nothing to do! But I trust you completely as to your findings.

In answer to your point about nulling, I was referring to the null test, where someone compares two supposedly identical tracks to see if they really are. One simply inverts the one and mix-pastes it upon the other and if they are identical, they will exactly cancel each other out and you will be left with nothing. But they really do have to be identical for this to work - same sample rate and everything - and the really tricky bit is exactly aligning them because if they are even one sample out of alignment, it won't work.

DSOTM would have been a very bad choice anyway, it seems. I don't think I've ever played mine!

 

And I may have played my DSotM only a single time. But on to other matters...

First, as I have previously noted, I got my KanexPro de-embedding working after the PS3 setup. Subsequent to that, there was at most a de-embedding test or two, till roughly last week solely for the purposes of this thread. I have not done a lot of de-embedding, I just know how to do it.

Don't worry, however, about placing the burden on me, because my thoughts place the burden entirely on you! (As an aside I am about to enter into a quite involving previous commitment. Starting early next week my posting will be close enough to "radio silence" for almost two weeks...and now I am entering prep.)

First thing, DSotM turns out to be near ideal for your own testing. You have seen my related posts and pictures. Run an analog drop of the DSD layer of Money, run a de-embedding of Money. If either appears clipped at the loudest points, that method is limited in the SACDs you can use it on. The native DSD is not limited. Obviously, that is your work.

You already have one DSD set that corresponds to an SACD you own. You can acquire more DSD download / SACD pairs should you desire. That can be done to the extent your funding permits. Then, the only thing you might be lacking is DSD -> PCM conversion software. But Tascam Hi-Res Editor is free...my understanding is that there are additional free conversion possibilities out there. Korg AudioGate 2 or Korg AudioGate 3 are free downloads, but using Korg software freely with no more than a Twitter account seems to be no longer a possibility. But if you are interested in using one of the AudioGates, eBaying the cheapest Korg DSD recorder you can find would give you the key to using AudioGate as frequently as needed. (You may even find yourself doing DSD needle drops!) When you install AudioGate, the Korg DSD recorder needs to be connected by USB...after that the connection is not required for usage. (I have a small Korg DSD recorder I bought a few years back, I am freely able to use AudioGate 2 as I desire.) Anyway, DSD download/SACD/software would allow you to fully carry out whatever comparisons held meaning for you. Everything could be done entirely in your own domain.

On the nulling... If you read my procedure, it amounts to what you describe. Of course careful and full alignment is necessary. But that is all something I carried out in my investigation of "Thriller" repair, as an example. I was not looking for zero differences ("null") in my investigations, however. I was looking at what the differences were between the results of one parameter choice versus another. I even ran differences of difference sets, to further allow a close look at what parameter changes could mean.

Okay, as I've indicated you have work ahead of you. Have fun!

 

Yes, it's true, I can compare my SACD de-embeds to online purchases of either DSD downloads or DSD conversions to FLAC, and that's what I'll do. I'm sure it will keep me happy! Ideally I would like to have brought ISO rips into the comparison, too, just to see if there was any difference - and if so, to ask if one was discernibly more accurate than the other. I'm beginning to think the de-embedding process is probably just as accurate as the ISO method, based on the results I've seen so far. Anyone else have a view on this, one way or the other?

 

Just a note that I seem to have confused myself in my previous post. I was thinking you had DSD files proper, but DSD files would have been of no use to you at the time. Of course you had the flac conversion of those files. But now you can buy actual DSD files, if you're so inclined, and make your own conversions to flac...given an acquisition of DSD/PCM conversion software. You could even use your Lynx Hilo to play DSD files and see what you thought of that.

Now that your de-embedding chain is working as it should, you really have a whole new world of options to explore. And listen to...and measure.

In a certain "purist" sense, making and working with ISOs could seem the only reasonable way to go, and there are certainly arguments for that. But it also depends on the endpoint you desire in your own DSD work. If you are ultimately going to flacs, any ISOs may well be superfluous, and with no actual advantage. If you are going to listen to DSD directly, that may be a stronger argument for ISO work...though with conversion software you might even determine that listening to DSD converted from 44.1k/16 CD rips, or converted from some higher rez, offers listening worth.

As one more aside, the DSD world can have some interesting surprises. For example, the Mobile Fidelity SACD Hybrid discs of Patricia Barber's early and most notable work (here is one example at somewhat reasonable current pricing) all have DSD layers that are DSD'd 44.1k. Peter Gabriel's "So" Hybrid SACD? DSD layer on that too is a conversion from 44.1k. Yeah, you probably know how I know.

But the thing is, you clearly care about your state of knowledge. An extremely admirable trait. You are building up your toolset, and that can mean in time doors you didn't even know the existence of will open up before you. A good toolset is not such a bad thing to have.

 

Here's a question for the SACD conversion gurus here. I recently posted to this thread about wishing I could rip the multichannel streams from the SACD's I own with surround sound. Shortly afterwards, I got a private message from a user (who shall remain nameless) who claims he recently converted the E.T. The Extra-Terrestrial SACD to a 5.1 Blu ray Audio track on his CD and claimed it was not difficult with the right software. He offered to tell me how he did it. I replied that I was certainly interested, but he never answered. I wrote him again, but still no answer. I've checked the forum and he's on here several times since I wrote, so I'm wondering if it was all a load of BS. Given that none of you are talking about it, I assume so. Just thought I'd bring it up since it seems clear the guy isn't going to divulge his secret!

 

It is indeed very simple to convert multi-channel DST64 area of SACD ISO to 24/88 (or, say, 24/96) multi-channel LPCM (or FLAC) in fb2k with, say, foo_input_sacd plug-in. I did several stereo downmixes of multi-channel SACDs, including this one (it takes ~15 min per SACD image on my 6-core AMD Phenom PC).

 

Yes but I assume this is only after one has ripped the iso using an old PS3, right?

 

Correct.

 

The ISO method is clearly preferable, since it gives you access to the DSD master on the disc. From that you can go in one of any number of directions: stick to DSD64, upconvert to DSD128/DSD256/DSD512, decimate to any form of PCM with any of a number of DSD to PCM converters (refer again to Archimago's links earlier in this thread). The PCM de-embed method makes the DSD to PCM conversion choice for you (and if you could configure a device to handshake at various PCM rates, you would be able to get alternatives to your current 176.4 kHz rips). You could also choose different playback software (foobar2000, JRiver, etc.) to see whether it has any bearing on sonics (playing either DSD as DSD or some form of PCM, or your DSD-to-PCM conversions).

If you want to determine how much/little you are losing via your analog and/or digital rips, I would be inclined to seek out a title from a company like Channel Classics, who are known for their very high recording quality, and who also sell DSD downloads of their titles (though they are rather pricey, so find one title you know you will love!). You will have to assume that the DSD download = the DSD ISO rip (I cannot think why they should not be the same, especially in the case of classical music which does not suffer from the loudness wars). You can then do all the various comparisons:

1. DSD (Sony) vs. DSD (Lynx)
2. PCM (Sony) vs. other DSD-to-PCM converters
3. DSD (Lynx) vs. PCM (Lynx)
4. analog rip (DSD or PCM) vs. DSD or PCM
etc.

It is impossible to say what, if any, differences you may hear, as some converters perform better with one format over another, or you may be susceptible to certain characteristics that are more/less emphasized by one technique, etc. However, comparisons of this type should satisfy your curiosity, and may save you from buying a ton of unnecessary DSD downloads.


Stefan,

With regard to your question, in principle, there is no difference between ripping the 2-ch and M-ch zones of an SACD, and it would all come down to the functionality of the de-embedder. It does strike me as odd that the devices mentioned in this thread are able to de-embed the M-ch PCM and convert it to analog, but do not make the 6-channel audio available digitally via ADAT or USB. Maybe there is no market for it?

I'm sure you do not want to use the lousy DACs in a cheap de-embedder to feed a nice M-ch ADC for archiving (that makes no sense!).

 

Edit: posted before seeing Black Elk's post above.

Yes, darkmass, I see what you mean. If DSD is your objective, then ISO makes more sense (because it's not even possible to de-embed DSD). But if FLAC is what you want, there's probably nothing to choose between them.

I've just suffered another small setback BTW, although thankfully not to do with the de-embedding process itself. In the course of my experiments I've discovered that there's a fault on the optical input of the Lynx Hilo - it's supposed to support all resolutions up to 192/24 but there's a temperature-related fault that shuts down the higher resolutions once the Hilo has reached normal operating temperature! Bizarre, I know, but I reported it to Lynx and they've been able to reproduce it in their lab so they've looking into it.

Life as a de-embedder is so much simpler with an Oppo. I may yet take that road - or just get a PS3.

 

On the idea of de-embedding 5.1 channels (or even more), I think the reason why it's not possible is not because of the limitations of the de-embedder but because of the limitations of S/PDIF. The burden of supporting six audio channels simultaneously at 24/192 would go way beyond what the S/PDIF standard supports, most especially via S/PDIF optical which struggles to achieve high-res even with just two channels. The S/PDIF standard only supports multi-channel in compressed formats, AFAIK.

 

Many thanks, Black Elk. Clearly you're right, that ISO ripping gives you access to the DSD master and also gives you far more options. If I wasn't intending to go down the DSD road, however, and only wanted FLAC at 88.2/24, I think de-embedding could make more sense so long as it didn't result in any loss of quality. I'd be interested to hear any argument as to why de-embedding would be any worse than converting from an ISO rip but not having heard one yet, I think what I need to do is just take your advice and listen for myself. That website you pointed to sounds very useful for that purpose so thank you very much for that and if I make any startling discoveries, I'll be sure to let you know!

 

I thought that the experiments in this thread answered that question, no? There has been hissy noise and failure after failure has there not?

I guess maybe I did not follow it all, as it has been a long winded read.

 

There are other digital interfaces which do support the data rate though.

It's not so much that it would be any worse, just that the choice is made for you, whereas in the ISO case you could change converter and/or noise-shaper/filter settings for every project, should you so desire.

 

If you had read all the posts, you would know that that problem has long since been fixed!

 

Speaking for myself, I wouldn't mind hearing about mundane discoveries as well. We are all here to learn, no?

I am not one to reflexively recommend the acquisition of a proper PS3. That said, I have no regrets that I have gone down that particular route. And...just think of the further options you'd have!

But whether you eventually go down the somewhat angular PS3 road or not, some advance data could always be useful. "ted_b" (also a forum member for Steve's site, if seldom seen here) is the one clear focal point of most things concerning the PS3 method (and before that, he was a de-embedding focal point). This is where he can most readily be found. Note the "guide" link in the signature block for him. While the guide is noted as being out of date at this point, it is ted_b's guide and there is certainly no better description of the overall process. Ted monitors the thread I have linked, and has been observed to respond to that site's PM system almost without fail.

For now, I'd recommend doing no more than reading about what the PS3 method entails. Then you'll be able to make a more informed decision about stepping on the PS3 road if that urge ever reaches a compelling state for you.

 

That's funny you should mention a guide because I just wrote one summing up what I've learned about de-embedding. Ted's guide on the PS3 method looks awesome and I'll read it with great interest - thanks very much for pointing it out.

Here's mine. I only wrote it for fun - I'm not planning to do anything with it except leave it here in case it helps anyone. If anyone spots any errors, do please point them out.

Ripping SACDs with a de-embedder

It should be easy to rip SACDs. In theory, you could just record the digital stream coming out of your SACD player through its HDMI output. The only thing that stops you doing this is the fact that virtually no PC sound card or DAC has an HDMI input. So you need an adapter that will take the digital signal out of the HDMI cable and push it into an S/PDIF coaxial or optical connection. That adapter is known as an HDMI de-embedder, a fairly simple little device with an HDMI input, an HDMI output, and an S/PDIF output on the side.

Depending on your SACD player, you may need something else, too. Usually, before an HDMI device will start outputting over an HDMI cable, it needs to have obtained a digital handshake from a receiving HDMI device using an anti-piracy protocol called HDCP. It also needs to have obtained the receiving device’s EDID, a digital ID that reveals what kind of formats the receiver supports.

If you have an Oppo, you don’t have to worry about the HDCP handshake because apparently, all Oppo models will send out HDMI without one (which is possibly a bit naughty, but let’s not go there). Nor do you have to worry about the EDID because you can control the Oppo’s output manually and tell it exactly what kind of signal to send out regardless of what’s at the other end of the HDMI cable.

It’s the complete opposite with the Sony SCD-XA5400ES - and probably most other players. The Sony definitely needs an HDCP handshake before it will send out any signal at all over HDMI. It also needs an EDID because there’s no way of controlling the Sony’s output manually from the player itself so you can’t tell it what sort of signal to emit. (I’ll explain why you need to do this in a moment.)

So the Oppo is ahead on ease of use. But the Sony will output at a resolution of 176.4/24 whereas the Oppo’s output is 88.2/24. The Oppo’s resolution is surely high enough but if you already have a Sony or you have some other player that isn't an Oppo, you will have to address its need for an HDCP handshake and EDID.

First, let’s take the Oppo. I don’t own one but as far as I know, setting up the de-embedding is simplicity itself. All you do is buy a KanexPro (HAECOAX) de-embedder, a tiny little box about the size of the palm of your hand. Using an HDMI cable, you connect the Oppo to the KanexPro. Then, using an S/PDIF coax cable, you connect the KanexPro to either your PC’s internal sound card or your DAC. Leave the KanexPro’s channel selector button in its default position. Tell the Oppo to send out PCM at maximum resolution. And away you go.

Turning to the Sony SCD XA-5400ES, things become more complicated. The KanexPro alone will not work with the Sony because the KanexPro doesn’t offer an HDCP handshake nor does it emit any kind of EDID. You could attach, say, an HDMI TV to the KanexPro’s HDMI output and that would provide an HDCP handshake, so the Sony would start sending a signal. But the TV’s EDID would reveal that it could only support low-res audio so that’s all the Sony would send it – there’d be no 176.4/24.

One way around this would be to attach an A/V receiver to the KanexPro’s HDMI output. That would provide the HDCD handshake and an EDID. But would it be the right EDID? The trouble is, the Sony’s default output is DSD, which these de-embedders don’t support. If your A/V receiver is DSD-capable, then DSD is what the Sony will send out, which is no use to you at all. You might think you can get around this by buying a cheap A/V receiver that doesn’t support DSD - just regular PCM. But the problem then is that, if the A/V receiver is that cheap, it probably doesn’t support a PCM resolution as high as 176.4/24 either! It’s surprisingly difficult to find an A/V receiver that’s good enough to receive PCM up to 192/24 yet not so good that it also supports DSD. But that’s what you’d need if you wanted to use a KanexPro with the Sony.

There is another solution, and that’s to use a slightly more expensive de-embedder that will not only supply its own HDCP handshake but also supply the correct EDID. This de-embedder is the Cyp AU-11SA, as it’s known in the UK. De-embedders looking exactly like the Cyp appear under other brand names, most famously the Atlona AT-HD570 in the US. I don’t know whether these are identical or what, and not having tried them, I can’t be certain whether they have exactly the same specification as the Cyp.

The point about the Cyp is that, because it supplies its own EDID and HDCP handshake, it will tell the Sony that it supports PCM but not DSD, and that will force the Sony to send PCM instead of DSD. The Cyp will also tell the Sony that it can support PCM in resolutions up to 192/24, so the Sony will default to its optimal PCM resolution which is 176.4/24.

However, there is one, big, huge problem with the Cyp: it uses an S/PDIF optical connection which, unlike an S/PDIF coax connection, cannot be relied upon to work at all resolutions up to 192/24. I can vouch for the fact that the Cyp’s optical output will definitely support an output of up to 192/24 but the problem lies further down the line. I’ve found that some cheaper optical cables won’t support resolutions up to 192/24 so you would need to get a high quality cable. But a far bigger problem is that few DACs or PC sound cards have an optical input that will support resolutions up to 192/24. For example, even Benchmark’s top-of-the-range DAC2 HGC, costing $2,548.50, only supports resolutions up to 96/24 on its optical input, and I have just discovered – and had it confirmed by Lynx – that my costly, semi-pro Lynx Hilo, although claiming to support up to 192/24 on its optical input, has a design fault that stops it going beyond 96/24 reliably.

If you can clear this hurdle, then using the Cyp with the Sony is as easy as using the KanexPro with the Oppo – it’s plug and play. Using an HDMI cable, connect the Sony to the Cyp. Using an S/PDIF optical cable, connect the optical output of the Cyp to the optical input of your PC’s sound card or DAC. Make sure that the switch on the side of the Cyp is in the 2-channel position, and away you go.

If you can’t clear the optical hurdle, you’ll either have to give up and get an Oppo or find some way of making the KanexPro work with the Sony. I can think of three possibilities. You could find a suitable A/V receiver to connect to the KanexPro, as explained earlier; you could buy an HDMI emulator, program it to imitate a suitable EDID (as well as a handshake) and hook that up to the KanexPro’s HDMI out; or you could even hook up the Cyp to the KanexPro’s HDMI out. (I’ve tried that and it works.) Any of these methods would allow you to use the KanexPro’s S/PDIF coax connector, which is trouble-free all the way up to 192/24.

I think that about covers the basics. I did have one other problem – my first few rips suffered from noise because I hadn’t realised that the clock sync settings on my Lynx Hilo had to be altered when using the S/PDIF digital inputs. But I think very few consumer DACs have these advanced settings so the vast majority of people wouldn’t need to worry about this.​

 

Very, very nicely done!

Here's how that menu item works on an Oppo BDP-93...


From the BDP-93 manual, page62, under Audio Format Setup...

2. HDMI Audio: To select digital audio output format from the HDMI port. The options are:

• Auto (default) – The player checks with the device connecting to HDMI output to
automatically determine which audio format to use.

• LPCM – Digital audio output via HDMI will be multi-channel Linear PCM format. When this
option is selected, compressed audio bit streams will be decoded by the player. This
setting is recommended when connecting the HDMI output directly to a TV or to a receiver
without advanced audio decoding capabilities.

• Bitstream – Digital audio output via HDMI will be in bitstream format. This option is
recommended when connecting the HDMI output to an A/V receiver or processor that
supports advanced audio decoding, such as Dolby TrueHD and DTS-HD Master Audio.

• Off – No digital audio output via HDMI.

What initially got in the way of my using the KanexPro with my Oppo was that the default of "Auto" was selected. Often enough Auto is a reasonable setting for any option, but here it meant "check with the KanexPro to see what it wants before doing anything". The KanexPro wanted LPCM but has no ability to convey that, so as a result nothing to de-embed went out over the HDMI. Once I finally figured out I needed to set the option to "LPCM" things went much better. But before I figured out "LPCM" was necessary I had already secured a PS3 and got it ISOing SACDs.

The subtle Oppo specific "feature" of the above options is that "LPCM" and "Bitstream" further mean to send data out over HDMI...even if there is no form of HDMI handshaking present.

 

Fantastic writeup @back2vinyl , and thanks for the Oppo info @darkmass !

For the first time, I'm actually thinking of ripping my CDs in PCM format, because I have an Oppo player (the BDP 105), I'm content with PCM 24/88.1 resolution for my rips, and I see the KanexPro unit is only $64 at Amazon.

One question, though: My ripping computer is an Apple unit, which only has the 3.5mm port option. I've seen plenty of little adapters to go from optical to 3.5mm - but I'm wondering if there are any that go from coax do 3.5mm.

I know you can probably buy a cable that has the coax-style RCA connector on one end and a 3.5mm connector on the other - but I believe most (all?) of those are designed for analogue conversion and so might not have the correct wiring or sufficient quality to pass along a digital signal.

I'm not saying an optical-to-3.5mm wouldn't work; just curious if there's an affordable, dependable digital coax-to-Apple-3.5mm solution out there.

Thanks!

 

Many thanks, darkmass. That's an extremely helpful supplement for Oppo owners.

Tmtomh, thank you too and I'm only sorry I can't answer your cable question! I hope someone else can.

 

I am not an Apple person, but as far as I can tell from my research, if your 3.5mm input is indeed digital (or possibly a digital/analog port that Apple seems to offer), the digital S/PDIF mechanism on the Apple end is optical, not coax.

If that is the case, if a KanexPro coax output -> Apple 3.5mm cable existed some included "cable" electronics package would be translating the coax to optical for the Apple port use. I'm not saying such a cable doesn't exist (though a little adroit googling didn't turn one up), but it seems to me that such a cable wouldn't serve much purpose for getting de-embedded output from a KanexPro to a Mac since it would still enter the Mac as optical. Might as well take the optical S/PDIF from the KanexPro--which ought to support the Oppo's 88.2k/24 PCM sampling of an SACD--directly into the Mac via a Toslink to 3.5mm optical adapter.