I can't imagine anything surpassing Quad Arc Prediction... apart from 'Octo' or 'Double Octo' AP of course! But this can't be the upgrade, because none of us would be able to use it right now.

Have you found a way of 'undoing' the distortions produced by cr*ppy ADCs maybe?

Mani.

I'm not sure if this is relevant, but Ritter talked about the 'upcoming' Model Two back in 1999. At that point, only the Model One was actually available. However, my particular Model Two was one of the first to be built... and that was in 2003!

Perhaps they were waiting for commercial 24/192 ADC chips to become available for the Model Two? However, Ritter talks about 'custom, discrete, full-ladder converters'...

Unfortunately, I'm not willing to take my unit apart to have a peek inside those PMADC-1 modules and see what chips they're using. Sorry!

Mani.

I haven't, but Romy has! Well his Model One at least:

http://goodsoundclub.com/Forums/ShowPost.aspx?postID=6566#6566

Not sure if you can make the chips out though...

Mani.

Yes, my experience too. But I think you need a 24/176.4 NOS DAC to appreciate this...

This weekend, I'll try to find some time to make some recordings from vinyl for you - 16/44.1, 16/44.1 with peak extension, 24/176.4 and 24/192. You won't like the music much (it'll be classical or jazz - that's all I've got on vinyl), but hopefully it might help you understand what's going on a bit better - or at least eliminate Keith Johnson's recording techniques from the equation.

Cheers,
Mani.

Hi Peter,

Here's what Michael Ritter (of Pacific Microsonics) wrote in 'Mix' magazine in 1999, before the Model Two was released (the highlights in bold are mine):

"The actual A/D converter in the Model One runs at 24 bits and 176.4 kHz currently; the Model Two will also convert at 192 kHz. We improve the linearity of our conversion with a high-amplitude broadband dither signal that we mix in with the program in the analog domain. The dither appears to be random, but the system knows at any given instant precisely what the amplitude of that dither signal is. And because we use our own custom, discrete, full-ladder converter with excellent amplitude and phase accuracy, we are able to apply an 'anti-dither' signal, exactly out-of-phase and matched in time, in the digital domain after conversion. That nulls the dither noise out of the signal.

If it's going to be a 176.4 or 192kHz DVD-Audio release, then we will not decimate that signal; we use a proprietary filter [non-oversampled] optimized to that sample rate. If it's going to be 88.2/96 kHz, we use 2:1 decimation, and once again we use a filter optimized to that frequency. But in both high-resolution settings, the Nyquist frequency is high enough that we don't use the 'dynamic decimation' process that becomes necessary when we go down to 44.1 or 48 kHz."

Note that there may be some confusion with regard to the editing by Mix magazine. Ritter mentions a 'proprietary filter' and the editor has included '[non-oversampled]'. I'm not sure how accurate this is...

HTH.

Mani.

Nice explanation Peter.


I think people will be shocked when they hear the NOS1 DAC with Quad (and greater) Arc Prediction. The benefits of this means of filtering is not subtle. It's totally obvious. BUT...

... it really requires a 24bit NOS filterless DAC to appreciate it. I have an oversampling DAC in my office system and prefer not using Arc Prediction at all with it. In my main listening room, I have a 24bit ADC/DAC that I believe is non-oversampling at >176.4 KHz, and I much prefer using QAP to anything else.

But this raises an interesting issue (interesting to me, at least):

Most recordings nowadays are made with delta-sigma ADC chips. What affect, if any, will this have? I mean, is the recording 'doomed' from the outset?

Any thoughts?

Mani.

PietPara,

You are going to have to increase your buffer size for Double and Quad. Repeat the method described in the tooltip for each of these separately.

As an example, with my RME, with NO Double or Quad, I can set the buffer to 32 in Adaptive mode and I get no buffer errors or pops 'n' clicks. Switching to Quad (with Arc Prediction), I have to set the buffer to 512 to get the same result.

Special mode with Quad Arc Prediction remains difficult for my setup - I have to set the buffer to ~200 to get a clean signal. This is substantially higher than Peter's 22 samples... or Russ's 1 sample! (or is it 2 samples nowadays?).

Mani.

Synology say, "The hard drives are hot-swappable to ensure continual service uptime should the drives need replacement." So, 'need replacement' doesn't mean if I want to upgrade from a 1TB to 2TB drive, say, but rather only when one of the drives fails and needs to be replaced???

Well, I'm about as angry as you are now Peter. (I tend to shoot the messanger though...)

I have my NAS drive placed down in my cellar (no noise in living space above) and am connecting it to 3 PCs via a switch and three CAT6 cables of between 5-15m length. Even when hot-swappable/pluggable eSata disks do arrive, I doubt I'll be able to have this configuration... and hence my decision to go for a NAS right now.

Mani.

Sounds like it's simply an issue with the folder being a 'shared' folder... In any event, this isn't a big deal, but it'd be nice to have it solved if it's a quick fix. Especially seeing as many people will start going the NAS root, I suspect.

Mani.

I've used dBpoweramp (I really like this SW, BTW) to rip an HDCD album to 24/44.1. I will compare this to the same album played with QAP and also with my built-in HDCD decoder later this evening.

Mani.

Hi Peter,

I'm in the process of transferring my whole music library to a NAS drive with RAID5 (better safe than sorry). But I'm getting the following error message, even though 'Music' definitely is a [EDIT: shared] folder.

Any thoughts?

Cheers,
Mani.

Yes, and hence why only 100-150 Model Twos were ever made before Microsoft pulled the plug. I'm not sure how many are still in the field...

But you know, I wouldn't be too bothered about the HDCD encoding. For the sake of argument, let's say just say that it's equivalent to 20/44.1 (probably optimistic). Better than 16/44.1? Well yes, but hardly anything to write home about.

But what really is of interest is that fact that all HDCD files have been digitized on a really good AD converter... it's a kind of guarantee of quality, if you like.

I've got the lab test results for my particular Model Two (THD, DIG-IN/OUT, etc, created on the 9/4/03) and the performance is truly remarkable.

Mani.

I just followed Peter's instructions... sort of! I set Q1=1 and started with the lowest buffer=32. With the 'log activity' box checked, I continued increasing the buffer size until there were no buffer errors in the X3PB log file. Now of course, I could reduce the buffer size from 1024 and increase Q1 from 1 and probably find a number of other values that work. But buffer=1024 and Q1=1 sounds good.

Incidentally, scheme 3 works fine for me on my Atom PC.

Mani.

Yes, this is exactly what I'm doing.

I haven't tried too many BNC wordclock cables. The one I use at the moment is a very reasonably priced one made by a UK company called Trichord (trichordresearch.com). As you said, it's a simple shielded 75ohm cable with BNC connectors - nothing more, nothing less.

Mani.

Of course, if you're going to use a Hiface, the spdif cable is going to be critical.


I'm currently using this cable between the Hiface and the X24 DAC. I'm not sure if it's down to the Hiface or the cable, but I've never heard this DAC sound so smooth... It could well be the cable, as this is exactly what other people have described when talking about the effect this cable has on the sound. I have a pretty expensive Cardas spdif cable knocking around somewhere and will do a comparison at some point.

Mani.