Pioneer DV-868AVi (aka DV-59AVi) Review (Multichannel DVD-V/A, SACD & CD Player)

[Scytales]

Hello to everyone.

This is a review and test of the Pioneer DV-868AVi DVD-V/A, SACD and CD player. This is an European model. Its US equivalent is known as the Elite DV-59AVi. Both model actually share the same service manual. There is also a seemingly similar Asian model known as the DV-S969AVi.

I recently acquired an old Audio Precision System One + DSP SYS222A for an investigation project I may publish on ASR. In order to familiarize myself with this complex test device, I have decided to measure one of the DV-868AVi that I have bought to play back multichannel music discs.

The DV-868AVi makes an excellent « guinea pig », not only because it is a wide spread model across the world, but also because of its numerous amount of features. It can play back CD Audio, DVD-Video and DVD-Audio as well as Super Audio CD, in stereo and multichannel, on analogue outputs. It is also able to decode Dolby and DTS encoded discs (be them CD or DVD) up to 24 bits. Moreover, it is capable of channel trim level and limited bass management for DVD as well as SACD. For PCM contents, it gets a re-quantization feature to expand 16 bits or 20 bits to 24 bits and five different digital interpolation filters. Yes, I am going to test all these features. So, I apologize in advance for the length of this review.

I am not able to conduct tests on digital outputs, so only the analogue outs will be measured. But as you can see on the photo below, the DV-868AVi is well equipped in digital outputs (for its time): coaxial and optical SPDIF, 2x Firewire (i-Link) and HDMI. Only the Firewire ports are able to transmit DSD signals read on SACD as well as any other type of digital signal. The HDMI output is able to transmit any kind of perceptually coded digital bitstream and PCM (save from copy-protected DVD-A: only sample rate restricted digital signals are put out).

As it is a first to me, this review shall be in some way a work in progress. Feel free to criticize the methodologies, relevance of tests and text and graphs formatting for future use in case I wish to repeat the experience.

To end this forewords, I have to say that besides the obvious steep learning curve to master the Audio Precision, the test process has required a huge amount of work, discipline and many checks. This has made me more humble and helps me better appreciate the considerable efforts that NTTY, pma, pkane, … and above all Amirm have to put in to provide the content for this website. They deserve many thanks from all consumers for these efforts.

But enough talk, let's get to the test!

To keep the workload of publishing the measurements manageable and help readers to navigate through many information without being overwhelmed at once, I have decided to divide the review per episode in several messages:

- in this message: CD measurements (PCM 16bits @ 44.1 ksps).
- in message #2: stereo DVD-V measurements (PCM 16 bits @ 48 ksps and 24 bits @ 96 ksps) - Published on 11/15/2025
- in message #3: perceptually coded DVD-V measurements (multichannel Dolby Digital and DTS) - Published on 11/22/2025
- in message #4: stereo and multichannel DVD-A measurements (PCM 24 bits @ 48 ksps and 24 bits @ 192 ksps) - Published on 11/29/2025
- in message #5: stereo and multichannel Super-Audio CD measurements (DSD) - Published on 12/1/2025

In each episode, I will do my best to adhere to the following general framework, when it is possible with the various test discs available to me:

1. Dashboard @ about 1 kHz, 0 dBFS
2. Frequency response related measurements
3. Noise and distortion related measurements
4. Linearity tests
5. Special tests
6. Disc readability tests

Part I: Measurements of the Pioneer DV-868AVi as a CD Player

Unless otherwise noted, the tests have been conducted with NTTY’s Test CD Version 7.2. The Audio Precision has always been given the 30 minutes preconditioning period mandated in its calibration procedure and the device under test the 5 minutes mandated by the AES-17 standard.

1. Dashboards

As said earlier, the Pioneer DVD-868AVi gets five interpolation filters to choose from: a no name one and four filters dubbed “Legato Pro”, which are distributed as “Standard” and “Effect” 1 to 3. The player also has a re-quantization feature dubbed “Hi-Bit”. The first dashboards were made with all these features disable, i.e. with the no name interpolation filter and no re-quantization.

Please also note that in order to navigate through the configuration menu of the player to conduct the tests, one of the SCART video output of the Pioneer player was plugged in an old cathode ray tube monitor with a cable of several meters. Some tests conducted without the monitor have shown that it has no effect whatsoever with CD replay, but very faint noises at the mains frequency and its harmonic are detectable with 24 bits source materials due to the lower noise floor. Hence, I have systematically engaged the “Pure Audio” feature of the Pioneer: that switches the video output off unless it is called back with the remote control unit and that kills the noise from the monitor.

Keep in mind that the old Audio Precision System One has no dual analogue to digital converter per analyzer channel, hence cannot display a recombined representation of both the test signal and the distortion residual. Contrary to Amirm's reviews, you will only see the distortion residual with a “notch” on each side of the leftover of the test signal: this is where the analyzer band-reject filter has removed the test tone prior to the digitization and measurement of the residual signal.

As the tests has revealed a significant difference between the left and the right channels, you get 2 dashboards for the cost of one. First, the left channel, which is good but somewhat 2 dB worse than the other:

The right channel reveals exceptionally good performance for a CD player, although the output level is about 0.1 dB lower than the left channel, which is fair:

You may notice the complete absence of visible power-supply related noise (mains frequency in Europe is at 50 Hz). Further investigations will confirm this fact.

To comply with the practice set by NTTY in his exceptionally thorough CD player reviews, here are dashboards of the two channels configured the same but at -6 dBFS:

At -6 dBFS, the respective THD+N (inverse of SINAD) of the two channels are almost the same: this indicates that they are dominated by noise rather than distortion. The “Hi Bit” feature does change nothing, but the four optional digital interpolation filters show some effects. I will save you from publishing measurements of all the available filters (I can publish them later on request), because they show only minimal worsening of the harmonic distortion, except for the "Effect 2" filter, which shows clear signs of clipping at 0 dBFS, albeit at a very low level (right channel only to let the Pioneer shines):

Fortunately, the clipping vanishes when the signal level is just 1 dB below 0 dBFS:

All in all, there is not much to see with the optional interpolation filters, except with filter #2, which raises an eyebrow at the maximum digital level. At least, none of the four optional filters significantly alters the output level. The analogue analyzer reading meter usually shows lower THD+N with the optional filters, but as will be seen shortly, this is just the by-product of a premature filtering of higher frequencies by those filters. That lowers the N component of the THD+N, but there is slightly more distortion nevertheless. Thus, to me the non-Legato Pro filter seems the best. Further tests may determine if this remains true.


2. Frequency response

Let's see how the five available interpolation filters of the Pioneer act on the player’s frequency response (only one channel shown). The two following measurements has been taken with the glide tone at -15 dBFS track of the Denon Audio Technical CD:

Obviously, the main effect of the choice of filter is to attenuate more or less the treble frequencies in the audio band. The good point is that all 5 filters produce almost the same output level at 1 kHz.

Here is a high-resolution frequency response and inter-channel phase deviation plot of the player with the Legato Pro feature disable:

Apart from the somewhat 0.1 dB level difference already seen between left and right channels, the frequency response in the audio band is quite perfectly flat and there is no deviation of phase response between the two channels. All the 4 other digital filters produce negligible inter-channel phase deviation above 15 kHz or so.

Another way to look at the effect of the digital filters is to trace the wideband response with white noise up to ultrasonic frequencies (in this case 85 kHz). To make the reading more comfortable, I have split the analysis in two graphs.

First, the no name filter and the Legato Pro Effect 2:

Then, the Legato Pro Standard, Effect 1 and 3:

The most obvious effect of all the optional filters is to alter the high frequency response in the audible band, a somewhat crude tone control, and let more out-of band images bleed in the ultrasonic range. Because of the attenuation of high frequencies brought by these filters, THD+N measurement restricted in the audio band can show a slight improvement over a more flat response, but that is of course misleading.

Decoding of emphasized data has been assessed with tracks from the HiFi-News and Record Review Test CD II (HFN 015). The DV-868AVi decodes emphasized tones at -10 dBFS at almost correct levels:

Further checks have shown that the decoding of pre-emphasized data at -60 dBFS from the same test disc is also correct.

Crosstalk evaluated with spot tones from the Denon Audio Technical CD is good:

3. Noise and distortion measurement

Let's start with some Left/Right single point measurements about noise and distortion before proceeding with more graphs. All measurements are done with standard Audio Precision tests:

The huge signal to noise ratio indicates that when it is fed with a digital signal containing only “infinity zero”, the DAC of the DV-868AVi mutes its output and/or that the muting circuit of the analogue output stage is engaged (which would have the effect to ground the Audio Precision analyzer inputs). We will see further below in the review confirmation of that fact (see section about Dolby Digital and DTS measurements). Therefore, dynamic range and quantization noise give a more realistic picture of the performance of the player with 16 bits contents.

We also see confirmation that the left channel is less good than the right thanks to the intermodulation tests. It seems the left channel does not like signals at 0 dBFS very much.

The default digital filter has no headroom to reproduce inter-sample overs, as do the 4 other available filters.

To go further in the analysis, Audio Precision provides some standard tests to assess the noise performance of a CD player.

First test is to look at the FFT spectrum of an “Infinity Zero” signal from the Denon Audio Technical CD up to 80 kHz:

The other tests use the analogue analyzer to plot 1/3 octave curves of the noise with the same "infinity zero" signal. Although this method has less frequency resolution than the FFT analysis (the higher the frequency, the wider the noise bandwidth that each point of the curve represents contrary to an FFT analysis where each “bin” has equal bandwidth; that is why the curve obtained with the analogue analyzer regularly increases towards higher frequencies), it takes advantage of the greater dynamic range and the wider frequency response of the Audio Precision System One’s analogue analyzer over its ADCs.

A correlation can be seen on both the FFT and the wideband analogue noise spectrum, where two spikes appear slightly above 30 and 70 kHz. The curve obtained with the analogue method gives a hint of the presence of two more noise spikes at about 95 and 140 kHz. The overall levels of the noise spikes remain very low, indeed, especially in the left channel. As the DV-868AVi is powered by a switch-mode supply, those tones are perhaps the product of the switching frequency of this power supply.

Restricting the analysis to the audio bandwidth shows greater details, with some spikes of unknown origin on the left channels:

Lastly, I add a test inspired by NTTY's practice to show an FFT spectrum up to 1 kHz when the player is reproducing a 1 kHz tone at 0 dBFS in search of power supply spurious noise. For this test, I set up an FFT at 8 ksps in order to improve bin frequency resolution to look after the output of the notch filter. Here I have to make a choice about which signal to use. I have chosen the 999.91 Hz sine with noise-shaped dither, which produces the lowest noise floor and is actually the only way to see what seems to be harmonics of the main 50 Hz frequency at 100, 200 and 250 Hz (only one channel shown, the other being essentially identical):

As the Pioneer DV-868AVi uses a switch-mode power supply, this spurious might have been introduced by the interconnection with the Audio Precision.

THD+N vs frequency has been assessed with spot tones at 0 dBFS without dither from the Pierre Vérany Digital Test CD. First within a 20 kHz bandwidth (i.e. the audio band):

In the bass, the THD+N is higher at 20 Hz then decreases rapidly. Cross checking with spot tones from the Denon Audio Technical CD has shown that at 40 Hz, the THD+N already rejoins the level at 50 Hz. The downward slope of the curve above 10 kHz is due to the fact that the second harmonic at 10 kHz and above falls outside the pass-band of the test, hence there is no longer any harmonic to measure, only noise. The upward curve from about 15 kHz most probably indicates an increase in the noise (including quantization noise) or distortion due to imaging above the 22.05 kHz Nyquist frequency of CD folding back into the audio band when the player has to reproduce high frequencies at high level.

Second, here is the same test, but with a wider measurement bandwidth up to 80 kHz:

THD+N in function of levels has also been assessed with the help the 999.91 kHz from NTTY CD test disc. First with dithered tones (which gives a typical performance curve, because CD production is almost universally done with dither):

The flat curve up to a level of about -9 dB indicates that the measurement is dominated by the dither noise and that excess distortion begins to appears at signal levels above -9 dBFS. It is possible to see a truer picture of the level of performance of the player with the same measurement but this time with the sine tones without dither:

Here, the two curves stay below -98 dB THD+N up to -6 dBFS level or so. That means that the Pioneer DV-868AVi is capable of 16 bits accuracy over almost a 70 dB wide dynamic range, except above -6 dBFS, from where little excess noise or distortion begin to appear.

Audio Precision provides a standard test to evaluate more qualitatively the SMPTE intermodulation distortion by looking at the digitization of the output of the analogue analyzer filter that removes the 60 Hz and 7 kHz tones to observe the actual distortion products. First with the Legato Pro feature disable:

And here is the same test with Legato Pro Effect 3, which shows a somewhat worsened distortion pattern (Effect 1 is almost the same):

The old 16 bits Burr Brown PCM78 ADCs of the Audio Precision System One probably limits the usefulness of FFTs of high level CCIF twin tone signals, because of the high dynamic of this signal relative to the noise floor. Here is a first FFT of a -10 dBFS twin tones (from the HiFi-News & Records Review CD Test II) in a 23 kHz bandwidth:

Obviously, the test signal was not created with dither, which explained the "grass" made of many low-level spikes above the noise floor at about -130 dBr. Nevertheless, pairs of odd order distortion products (at 18 and 21 kHz, 17 and 22 kHz and 16 and 23 kHz) are visible, the highest pair being at about 85 dB under the level of each twin tones. The even order distortion at 1 kHz is some 100 dB under the level of each twin tones. To compare, here is the same measurement with the same type of signal from NTTY's Test CD, but at a higher level of -3.02 dBFS:

We see an elevated noise floor that now masks the "grass" (the same with a similar test signal at 0 dBFS from the HiFi-News test disc, not shown), but also higher distortion products that are easier to see. Here is an analysis of the very same signal with a bandwidth extended to 85 kHz:

And here is the same signal analyzed in the same 85 kHz bandwidth with the Legato Pro Standard filter to compare to the above:

For the same reason already explained about the dynamic range of the old Audio Precision ADCs, I do not think an FFT of the multitone signal from NTTY's test CD can give much detailed information. Moreover, the maximum FFT length that the System One is capable of restricts the bin frequency resolution in the bass. Anyway, here is the FFT spectrum of such signal, but from 100 Hz only:

Even considering the limitation of the System One ADCs, this curves will be a useful reference for further tests of the same kind that will show some detrimental effects. More of that in a further part of the review. I should add that it is probably possible to overcome the limited resolution of the analysis in the bass by lowering the sample rate of the FFT to 8 ksps and restricting the analysis to a lower 4 kHz bandwidth in order to observe only the lower part of the multitone signal, but I have not try.

To end this chapter about distortion, I have something very special to measure.

The Pierre Vérany Digital Test CD is unique in that it gets tracks to perform a peculiar test called "intermodulation by crosstalk". The theory of operation of this test is as follows: the channel under test plays a 5 kHz tone and the other channel simultaneously plays a 1 kHz tone, both at 0 dBFS. If there is a significant crosstalk between channels, the 1 kHz tone can theoretically intermodulate with the 5 kHz tone to produce on the channel under test a difference frequency of 4 kHz. This test has been designed at a time when most CD players have only a single digital to analogue converter for two channels for cost-savings reason. These CD players had to demultiplexe the left channel from the right by routing the output of the single DAC alternatively to each one of the two channels at twice the CD sampling frequency. I do not know if this test has still relevance nowadays, but here it is (left channel only, because it shows the biggest effect and I soon will hit the limit of possible number of images to upload!):

Please do not pay attention to the wide skirt at the base of the 5 kHz tone: it is due to the Flat Top window I have use to get the most accurate level of each tone visible on the FFT spectrum. You may have noticed that a very faint 4 kHz tone has appeared at less than -105 dBr compared to the situation where only a 5 kHz tone at 0 dBFS is played in each channel. Indeed, it seems that the phenomenon explained by the author of the Pierre Verany Digital Test CD might exist. The right channel (not shown) is almost not affected by an increased of the level of a 4 kHz tone, but very low level tones at 1 and 2 kHz at -110 dBr do appear when this test is performed. You can also see on the analysis of the left channel in inclusion in the above image that some tones are also noticeable at 1, 2 and 3 kHz in the no intermodulation situation.

The relevance ot this measurement as an intermodulation distortion assessment is debatable, but a case can perhaps be made that it is in some way a more stringent crosstalk test than the one I performed and have shown the result above. In an usual crosstalk measurement, one channel only plays back some signal whereas the analyzed channel remains silent. In this case, many D/A converters may mute the channel under scrutiny. When the test signals designed for the Pierre Verany Digital Test CD are played back, on the contrary, both channels are very well alive because they have both a tone at maximum level to reproduce.


4. Linearity Tests

I wish to use the spot tones from 0 to -110 dBFS on NTTY's Test CD to plot the deviation from linearity of the player's DAC. Audio Precision provided a standard test to measure linearity with the Audio Precision System One. Unfortunately, the manufacturer warned that the System One's analogue band-pass filter and analyzer are unable to measure under -90 dBFS due to the presence of the dither noise necessary to encode tones at those levels and the intrinsic performance of the analogue analyzer. Attempts to use the standard test have shown that that is indeed the case. But I have not given up on plotting linearity. Having thoroughly studied the many System One manuals prior to its purchase, I have created a new linearity test using another tool of the System One: a software-implemented selective voltmeter. It works great in conjunction with the noise-shaped dithered spot tones created by NTTY for his test CD. So, low and behold the deviation of linearity plot of the DAC of the Pioneer DV-868AVi down to -110 dBFS:

Please take notice that I have used a much more dilated vertical scale than the one Amirm uses for its own tests. The yellow curve (right channel) stops at -100 dB because the right channel of the Audio Precision is consistently unable to read any signal under. I have not find why yet. Nevertheless, we get an almost perfect linearity down to -90 dBFS on both channels and the left channel shows a deviation from perfect linearity that remains well under a 0.2 dB positive error down to -110 dBFS. To me, this is an outstandingly good result from CD data.

An FFT spectrum of the -110 dBFS sine shows that even the right channel is able to play the test tone at the correct level and that the Audio Precision selective voltmeter is not cheating:

For this graph, I have referenced the 0 dB to the full output level of the right channel, which is about 0.1 dB lower than the left channel, remember. This difference is most probably the reason why the positive error on the left channel appears to be slightly higher than what the selective voltmeter has measured.

5. Special Test

Here is a capture of the waveform of a 997 Hz sine at -90.31 dBFS without dither from NTTY Test CD:

The waveform is a bit noisy, but the three discrete voltage steps are clearly visible.

Same signal, but with the "Hi Bit" 24 bits re-quantization feature enable:

This re-quantization feature works after all, even if it seems that it makes no difference on other tests.


6. Disc Readability

To end the first part of this review dedicated to CD replay, I have used the famous Pierre Vérany Digital Test CD2 to check the ability of the Pioneer DV-868AVi to play damaged or non-compliant disc pressings to the Red Book (RB) specifications. As NTTY obviously uses the same disc (or a very similar one) to perform its own tests, I will borrow his table:

Test Type	Variable parameter(s)	Results
Variation of linear cutting velocity	1.4 then 1.2 to 1.4 m/s in 0.05 m/s steps	All tests passed
Combined variations of track pitch and velocity	From 1.2 m/s to 1.40 m/s combined with 1.5 µm & 1.7 µm pitch	All tests passed
Variation of track pitch	1.5 µm to 1.7 µm in 0.05 µm steps	All tests passed
HF detection level	Variation of the pits/lands ratio from +2 to +18%	All tests passed
Drop-outs tests	0.05 to 0.2 mm (RB spec.) and 0.3 to 4 mm (non-spec.)	All RB spec. tests passed - Pass to .75 mm Fails from 1 mm
Combined drop-out size variation & minimum track pitch	1.5 µm + 1 to 2.4 mm	not performed (failure to previous test)
Successive drop-outs	2x0.1 mm to 2x3 mm	Pass to 2x1.5 mm Fails from 2x2.4 mm

I hope you have enjoyed the first part of this review and found it interesting!

Now, it's time to move on to DVD-V replay in the next message.

 

[Scytales]

Part II: Measurements of the Pioneer DV-868AVi as a DVD-V Player (Linear PCM)

I was fortunate enough to get a complete DVD test set from the German test and broadcast equipment manufacturers Rohde&Schwarz and Burosch. All the following measurements have been done with one of the DVD of this set. This part of the review will be shorter, because the old Audio Precision System One + DSP is sadly not suitable to use all the test signals on the DVDs, which have obviously been designed to be used with a digital control loop with a suitable analyzer, such as an Audio Precision Dual Domain or, of course, comparable Rohde&Schwarz audio analyzers.

As will be seen shortly, the optional digital filters also work with high sample rate signals. Unless otherwise noted, all measurements were done with the Legato Pro as well as the “Hi-Bit" features disable.

We will look at 16 bits @ 48 ksps material first, then 24 bits @ 96 ksps (a so called "hires" format).

A. 16 bits @ 48 ksps Stereo Tests:

1. Dashboards

Unsurprisingly, the output voltage levels are the same as with CD, with the same 0.1 dB difference between the two channels. The overall THD+N of both channels are mostly noise dominated, presumably by the dither used to produce the 997 Hz test signal (the use of dither for this specific test tone frequency is mandated by the AES-17 standard). No effect of the slightly higher sample rate compared to CD is visible. This dashboards should be viewed as typical of what the player would produce with dithered test tones in CD mode.

2. Frequency response

The frequency response obtained from a glide tone signal at 0 dBFS is a bit more extended than with CD, as expected:

Here are the responses of the five digital filters with white noise. Their responses are just scaled to higher frequencies compared to what they are with 44.1 ksps signal:

3. Noise and distortion measurements

I have much less to measure with the DVD test set. Interesting observations to make are:
- as is the case with CD data, the DAC obviously mutes its outputs with infinity zero signal, making the signal to noise ratio unrealistically high;
- the dynamic range data are obtained from a dithered - 60dBFS signal (again a requirement of the AES-17 standard) that explained the slightly less good numbers compared with the CD measurement above, but they are more typical of what can be expected from commercial CDs;
- the difference between left and right channels (not unexpectedly) remains.

The effect of the dithering applied to a 997 test tone can also be seen in the slightly elevated noise floor on the power supply noise test compared to the above measurement with CD data:

FFT Spectrum analysis of the residual of the SMPTE intermodulation test and wideband analysis of CCIF intermodulation signal:

THD+N vs level has been measured with a 332 Hz tone from 0 dBFS to -18 dBFS:

The effect of the dither shows up again in the elevated overall THD+N, but that is more typical of what can be obtained with CD replay at 16 bits depth, not only DVD replay. That does not prevent the left channel to perform slightly less than the right channel. Please also take note that the test signal is in Prologic II, but apart from the stated phase shift in the right channel, do not ask me what that does as I am not an hometheater guy at all!

The Rohde&Schwarz test DVD has two kind of multitone signals: 20 and 100 sines in the audio band. Here is the former:

4. Special tests

Before I move on to 24 bits test signals, let me take the opportunity to show you how the bass management and channel trim level features of the Pioneer perform.

Here is the response of the two main channels of the player to the 100 sines signal, one channel being trimmed to maximum (+6 dB), the other to minimum (-6 dB), and both beeing declared "small" in the player configuration menu:

The -3 dB point of the high-pass filter is set at about 80 Hz. When measured with a single sine tone, the output stays at the nominal level of 2 V on the left channel, but is 12 dB below on the right channel. That means that the processor of the Pioneer normalizes the levels of the highest trimmed channels to the nominal output level and attenuates the other channels accordingly, which is the smartest thing to do to avoid clipping on the positively trimmed channels and preserves as much resolution as possible. That is because when the channel trim feature is enable, a fixed 6 dB attenuation is applied to all channels, i.e. the output level is reduced to about 1 V RMS with PCM data and 880 mV RMS with DSD data. This fixed attenuation gives room to trim all channels up to +6 dB (see also Part IV).

As can be seen on the multitone response, the right channel seems not to suffer much from the attenuation. Performing a "dashboard" on the trimmed left channel provides confirmation that it is identical to the above published graph and that the trimmed right channel remains consistent:

B. 24 bits @ 96 ksps Stereo Tests:

Now it's time te begin the exploration of so-called high resolution, or "hires", test signal source materials!

The DVD-V format has been used under the "HDAD" acronyme by some music publishers that have released reissues in 24 bits/96 kHz PCM of some stereo albums before the advent of DVD-A (or SACD). So, it makes sense to test the performance of the Pioneer on that format.

1. Dashboards

Sadly, the left channel under-performs still, but the right channel shows an almost 6 dB improvement of its performance compared to 16 bits signal, mostly thanks to a lower noise level. The level difference between the two channels remains the same as with CD.

2. Frequency response

The frequency response and interchannel phase deviation obtained from a glide tone signal at 0 dBFS show nothing particular, except of course the higher bandwidth due to the higher sample rate:

It's interresting to compare this to the frequency response of one of the 4 optionnal digital filters, in this case the Legato Pro Effect 3 (please, pay attention that the vertical scale is different):

Compared to 44.1 ksps signal, the -1 dB point of the filter has moved from about 8.5 kHz to 20 kHz, and the -3 dB point from 16 kHz to about 30 kHz or so, leaving the audio band much less affected, but still.

Here are the responses of the five digital filters with white noise. Their responses are just scaled to higher frequencies compared to what they are with 44.1 ksps signal:

The extended frequency response is obvious, even when the effects of the filters is taken into account.

3. Noise and distortion measurements

As is the case with CD data, the DAC obviously mutes its outputs with infinity zero signal, making the signal to noise ratio unrealistically high. But the dynamic range measured with the converter alive and working thanks to the low level -60 dBFS tone is markedly improved compared to the 16 bits depth signal.

The power supply noise test shows a bit more discrete tones harmonically related to 50 Hz (European mains frequency) than with CD data, maybe due to run to run variation and induced by the display monitor that was used to navigate through the player's configuration menu:

THD+N vs. level shows a definitive superiority of 24 bits depth coding over 16 bits depth (pay attention to the fact that the vertical scale goes lower than with 16 bits signal!):

The higher definition of the player let us see that there is definitely something that limits the performance of the left channel. As the unit is 20 years old, it is perhaps the consequence of an aging, overheated or damaged component somewhere or it is by design, I can't say yet but we must keep an eye on this measurement to compare it with other channels when multichannel replay will be tackled further down the review. Anyway, at lower levels, the player shows pretty good performances.

As seen in the above table, there is also a significant improvement on the SMPTE intermodulation test compared to 16 bits signal. The FFT of this test residual shows a lower noise floor that unmasks more of the distortion products, but they remains more or less the same compared to 16 bits data:

The CCIF intermodulation spectrum:

4. Special tests

To end this part, I show how a -80 dBFS sine at 24 bits depth looks like:

And finally at -100 dBFS:

In the next message, I will go to perceptually coded digital formats before having a look at the DVD-A format.

 

[Scytales]

Part II bis: Measurements of the Pioneer DV-868AVi as a DVD-V Player (Dolby Digital and DTS coding)

All measurements in this part has been made with the Rohde&Schwarz and Burosch DVD-V.

The optional digital filters also work as well with perceptual CODEC source materials. Unless otherwise noted, the Legato Pro as well as the “Hi-Bit" features were disable.

We will look at Dolby Digital first, then DTS. This two CODECs are multichannel capable. Thus, some measurements will be shown on all six channels of the 5.1 output of the Pioneer.

A. Dolby Digital Multichannel Tests:

1. Dashboards

The Dolby Digital coding does not alter the output level but introduces some artefacts from about 2 to 13 kHz and an elevated noise floor. The spray of spurious tones at 2 kHz and above remains the same with a lower 332 Hz frequency test tone and at lower levels. Nevertheless, the end results remains solid on both channel, with a mean SINAD of 90.3 dB.

2. Frequency response

The frequency response obtained from a glide tone signal at 0 dBFS shows a very peculiar behavior:

There obviously is a heavy phase shift on both channels under about 70 Hz and above 15 kHz. The usable frequency response is thus more or less severely restricted in the highs. The low frequency response is much less affected, but some side-effects remain observable.

The responses of the five digital filters with white noise show more clearly what happens in the high frequency with this coding scheme:

There is a fair bit of chunk of high frequencies in the audio band that is filtered out. Presumably, this is where Dolby Digital information is encoded.

3. Noise and distortion measurements

Contrary to what has been observed until now with linear PCM signals, the DAC doesn't mute with an infinity zero signal encoded in Dolby Digital. But the signal to noise ratio is not more consistent with the dynamic range calculated from a -60 dBFS sine signal either. The reason seems to be that the noise floor is modulated respective to the signal:

Some modulation can also be observed in the somewhat elevated noise floor obtained from an infinity zero signal, proof that the DAC is alive with this kind of signal:

The higher noise level can also be seen along with more spurious tones with the test of power supply noise with a high level 0 dBFS signal, even if no modulation of the noise is apparent at this signal level:

Apart from the noise behavior, results with Dolby Digital encoded signals are slightly less good than with 16 bits linear PCM data. This can also be seen on the FFT spectrum analysis of the residual of the SMPTE intermodulation test:

In-band analysis of the CCIF intermodulation test tones reveals nothing particular, though:

Dolby Digital being a multichannel capable CODEC, wee have the opportunity to check the THD+N vs level on the main left and right channels as well as the center and surround left and right channels:

Interestingly, the two surround channels consistently show a sharp increase in THD+N level at 0 dBFS. I don't know if this is particular to the test disc (doubtful), the Pioneer Dolby decoder or the coding scheme itself. Future measurements performed on other players may give a clue. Overall, the lower performance of Dolby Digital against even 16 bits data is obvious, notwithstanding the discrepancy between the surround channels and the others.

The spectrum of the Multisine 20 signal from the Rohde&Schwarz test DVD indeed shows some ill-effects despite the limited resolution of the old Audio Precision analyzer on that test:

But, curiously, the frequency response above 15 kHz seems not to suffer from the limitation that has been observed with the glide tone signal used to trace the frequency and interchannel phase response.


B. DTS Multichannel Tests:

Having seen how Dolby Digital under-performs in some ways compared to linear PCM data, how will DTS do?

1. Dashboards

Actually, the dashboards look promising. Yes, the left channel under-performs compared to the right still, and the level difference between the two channels remains, but at 102 dB, the mean SINAD is slighty better than with 24 bits @ 96 ksps PCM data. Perhaps this is just due to the output levels being a hair lower than what they are with 24 bits data. The distortion spectrum is less clean and a unsual 500 Hz tone makes an appearance, though.

2. Frequency response

The frequency response and interchannel phase deviation obtained from a glide tone signal at 0 dBFS does not show the effects we have just seen above with Dolby Digital encoding:

The explanation is apparent when we look at the response of the five digital filters with white noise:

As is the case with Dolby Digital, DTS obviously uses part of its available signal bandwidth to code hidden informations and removes them from the frequency response with filtering, but the bandwidth used for that purpose is situated outside the audio band.

3. Noise and distortion measurements

We are back to the usual behavior: the DAC obviously mutes its outputs with infinity zero signal, making the signal to noise ratio unrealistically high. The dynamic range is actually on par with 24 bits signals already measured above, which is a very good result. An FFT of the -60 dBFS signal shows that DTS does not suffer from the noise modulation that plagued Dolby encoding (confirmed by a check with a -81.3 dBFS sine signal, not shown):

The power supply noise test is very clean, save for the 500 Hz or so tone already seen above on the dashboards:

THD+N vs. level on all channels save LFE shows somewhat inconsistent results across channels, but the THD+N remains pretty good up to 0 dBFS in any case:

The SMPTE intermodulation test shows a result in the same ballpark as with 24 bits data:

The CCIF intermodulation spectrum:

DTS has performed well on all distortion tests. Only a multitone signal does show some ill-effets of the coding scheme, but they are less intrusive than with Dolby Digital:

But take notice of something also visible on the wide-band FFT spectrum of white noise: the noise floor on this complex signals is higher with DTS than with Dolby Digital.

To conclude this part, I have to say that I was pleasantly surprised by the way DTS performed. It is clearly a better CODEC than Dolby Digital and it approaches the level of performance of plain linear PCM at 24 bits depth! How will the DVD-A format up to 24 bits @ 192 ksps perform?

 

[Scytales]

Part III: Measurements of the Pioneer DV-868AVi as a DVD-A Player (24 bits @ 192 ksps Stereo and 24 bits @ 48 ksps Multichannel)

All measurements in this part has been made with the Rohde&Schwarz and Burosch DVD-A.

Unless otherwise noted, the Legato Pro as well as the “Hi-Bit" features were disable.

We will look at stereo first, then multichannel.

A. 24 bits @ 192 ksps Stereo Tests:

1. Dashboards

No change in output levels, no change of the behavior of both channels (there is still a 0.1 dB difference in level between channels and the left remains a lesser performer than the right). The mean SINAD is 98.3 dB.

As you can see, the performances are slightly degraded compared to 24 bits signals at 96 ksps (see Part II). It has nothing to do with the DVD-A disc formatting opposed to DVD-V formatting, as will be obvious further down when we will deal with 48 ksps. It is an illustration that more of a thing (in this case, the sample rate) is not always better!

2. Frequency response

The frequency response obtained from a glide tone signal at 0 dBFS shows a flat response:

As the System One seems to have some trouble to follow the glide tone on the DVD-A (which goes up to 96 kHz), I also show an FFT of white noise to illustrate the wide frequency response:

The extended frequency response at 192 ksps is visible in all its glory. There is no need to test the Legato Pro filter as any effect, if any, would fall outside the bandwidth of the digital analyzer.

3. Noise and distortion measurements

We see the usual behavior we have been accustomed to see: the DAC mutes its outputs with infinity zero signal, making the signal to noise ratio unrealistically high.

The test of power supply noise with a high level 0 dBFS signal shows a very low broadband noise floor that unmasked some whispers of 50 Hz related spurious tones:

The SMPTE intermodulation test shows a low noise floor as well:

Th in-band analysis of the CCIF intermodulation test tones reveals nothing particular, even if it must be pointed out that the single point measurement shown above has revealed a particularly good result on the right channel, so perhaps the high sample rates helps for that kind of signals all else being equal?

THD+N vs level has been measured with a 332 Hz tone from 0 dBFS to -18 dBFS:

The results are behind those obtained with 24 bits @ 96 ksps.

To conclude, going up to 192 ksps does not bring something decisive over lower sample rates, at least with this player design and DAC. The bit depth matters more. The following measurement will make it even clearer.

B. 24 bits @ 48 ksps Multichannel Tests:

Please note that the DVD-A format does not alter the gain and bandwidth of the sixth channel that can be dedicated to LFE. The test signal on all six channels are thus identical, contrary to the Dolby Digital and DTS formats. Unless otherwise noted, the measurements have been performed on the two front channels.

1. Dashboards

At 101 dB, the mean SINAD is close to the best performance that was obtained with DTS data. Actually, this is the best result we can get from various linear PCM formats
with the Pioneer DV-868AVi. So, 48 ksps seems close to the sweet spot within the capability of the player's DAC.

2. Frequency response

No surprise, and the response of the five digital filters with white noise shows a now familiar pattern:

3. Noise and distortion measurements

No surprise either: the DAC obviously mutes its outputs with infinity zero signal, making the signal to noise ratio unrealistically high. The very good results of the right channel with the CCIF test signal is worth mentioning. A pity the two front channels does not behave the same.

To show that all six of the multichannel output behave the same noise-wise as the front channels, here are the FFT noise spectrum of an "Infinity Zero" signal taken from the Center and LFE outputs (take notice that the vertical scale is dilated compared to previous graphs of the same kind):

It is worth pointing out that the overall noise floor is identical than with a similar signal from a 16 bits source material (see Part I). This is of course because the DAC is muted, hence only the base noise floor of the analog stages of the player is left to measure, without any quantization errors or other source of spurious signals produced by the DAC to increase the level of noise.

The power supply noise test is very clean:

The THD+N vs. level on all channels shows somewhat inconsistent results across channels, but the THD+N remains pretty good up to 0 dBFS in any case:

The SMPTE intermodulation test:

The CCIF intermodulation spectrum:

Conclusion about the behavior of the Pioneer DV-868AVi as DVD-A player:

The player behaves very well. The fact that the LFE output is identical to the five others is advantageous, as it opens the possibility to use it as a sixth full range channel. Increasing the sample rate does not necessarily lead to better performances in the audio band. Maybe there are other gains to hope for from higher sample rates, but they are not obvious to me, at least with this player. CD sample rate (44.1 ksps) or close to (48 ksps) appears to be the sweet spot for the player's DACs.

This part ends the analysis of the DVD and PCM originated formats. In the last part, we will have a look on the Pioneer DV-868AVi as a stereo and multichannel SACD player as much as is possible. Interesting discoveries could emerge.

 

[Scytales]

Part IV: Measurements of the Pioneer DV-868AVi as an SACD Player (Stereo and Multichannel DSD)

Measuring an SACD disc player presents two challenges.

The first is the lack of availability of proper test disc. Only two test SACDs designed to assess performance of disc players have ever been made by Sony and Philips and they have been out of print for a very long time. The only available SACDs with some more or less useful test signals that are procurable are designed to set up stereo or multichannel Hi-Fi systems, not to measure the performance of the DACs of disc players. Nevertheless, I will try to make good use of the test SACDs that are available.

The second is the high level of out-of-band noise generated by the sigma-delta modulation. This out of band noise may upset wide-band analogue audio analyzer. The Audio Precision System One does not escape this problem. Thus, the measurements I am able to perform are limited for now, but I will proceed to work on the challenge.

For this part, I will use two test discs that have been designed mostly to set up multichannel audio systems : the Denon Audio Check SACD and the Stay in Tune with Pentatone SACD (formerly the Philips Super Audio CD - Surround Sound Reference Disc).


1. Frequency response

None of the two above discs has a 1 kHz or so sine tone at the nominal 0 dB SACD level, but the Denon SACD gets several sine signals at -16 dB SACD that are precisely aligned with the level of PCM sine signals at -16 dBFS. It is thus possible to calculate the nominal output level of the Pioneer by measuring the signals at -16 dB SACD and adding 16 dB to the measured level: 1.75 V RMS, which is about 1.4 dB lower than the output level in CD or DVD mode. This level difference will favor PCM source material during comparison unless it is compensated for.

The frequency response has been checked with a glide tone signal from 5 Hz to 30 kHz at -20 dB SACD from the Denon disc:

You can see that the response is tailored, with about 0.25 dB attenuation at 20 kHz and 0.6 dB at 30 kHz. As the analogue output low-pass filter is the same as for PCM source materials, this roll-off must be the effect of the built-in digital low-pass filter for DSD signal of the DAC chips.

It is possible to explore the response much higher in frequency with the pink noise tracks of the Denon SACD. It is well known that a 1/3 octave analysis of pink noise should show a flat response. Let’s check that with the Audio Precision analogue band-pass filter :

The responses of all six channels are identical and as expected. The usable frequency response extends to somewhat 50 kHz with a gentle roll-off. From 50 kHz, the response tilts upward. This is where the noise is no longer pink in nature: it is the shaped quantization noise typical of DSD that overcomes the pink noise. From about 70 kHz, the combined digital and analogue low-pass filters of the DAC and the player output stage begin to roll the shaped noise off. This graph may be a useful reference to compare to the performance of the low-pass filtering of other SACD players.

Another way to view this pink noise, though in a more restricted bandwidth, is to make an FFT at the maximum sampling frequency the Audio Precision System One is capable of. Moreover, as the Denon Audio Check SACD pink noise tracks get signal on one channel only at a time, we are going to see what is actually on the muted channels:

As you can see, the pink noise effectively extends to about 55 kHz before it is overcame by the shaped quantization noise. The tone at 31.5 kHz on the left channel can also be seen on the FFT spectrum of an "Infinity Zero" PCM signal: most probably it is related to the switching power supply. But the more interesting thing is that the quantization noise is also visible on the muted channel. That means that the silence is actually an analogue mute signal and that the DAC of the player stays alive, contrary to what happens with an "Infinity Zero" signal. This is good news, because we will be able to measure the true signal to noise ratio of the player thanks to this DSD analogue mute signal.

2. Noise measurements

To compute the signal to noise ratio, one must measure the noise floor in the audio band when playing an analogue mute track and refer the result to 0 dB SACD. Bear in mind though, that 0 dB SACD is not the maximum signal level that SACD is capable of: as I have explained on another thread, the SACD audio specification allows short term peak excursions at +3.1 dB SACD.

Here, I encountered a problem : the System One gives inconsistent results with the analogue mute signals of the Denon Audio Check SACD. Presumably, the shaped quantization noise level upsets the analyzer. Fortunately, the Pentatone SACD also have analogue mute tracks that all give consistent results. So here you are :

Taking into account the 1.4 dB difference in nominal level in detriment of SACD playback, the signal to noise ratio is consistent with the dynamic range measured with 24 bits PCM signals. So, the noise performance of the Pioneer in the audio band seems to be about the same with high resolution PCM signals and SACD. But a word of caution: there are many sigma-delta modulator designs that all have different characteristics, both in the audio band and outside. As a consequence, it is difficult to draw a definite picture of player performances without knowing the specifics of any given modulator used to produce DSD signals.

We can have a look at the noise floor in the audio band with an FFT, first with an analogue mute signal from the Denon SACD :

It can be seen that the two front channels are noisier than the two others with DSD. But the results do change when the same measurement is performed with analogue mute tracks from the Pentatone SACD (only the two front channels shown):

Now, the two front channels show a noise floor (referred to 1.75 V at 0 dB!) that is very close to the noise floor (referred to 2 V at 0 dB!) with a 24 bits PCM signal. Here is an FFT of a -100 dBFS 997 Hz sine signal in 24 bits to compare (please ignore the wide skirt at the base of the tone and its apparent erroneous level ; they are due to the absence of windowing):

Lastly, it is possible to check the lower part of the frequency spectrum in search of mains interference:

3. Special Tests

To end this part dedicated to SACD measurements, let’s have a look at how signal processing is performed on DSD by the Pioneer.

The DV-868AVi is based on a multichannel SACD decoder from Sony, the CXD2753. This chip allows to declare all main channels large or small or non-existent. It is possible to check the high-pass filter applied to a channel declared small by accumulating the FFT spectra of multiple low frequency test tones from the Pentatone SACD :

The actual -3 dB corner frequency is set at about 100 Hz.

The channel trim level feature of the Pioneer DV-868AVi also works on DSD data. Here are three successive FFT views which give the opportunity to see the pink noise and silence tracks from the Pentatone SACD to compare to those of the Denon shown above (beware the bandwidth of the analysis has been restricted to 85 kHz instead of 96 kHz!):

I have to correct what I have initially wrote above in Part II, a : the processor of the Pioneer does not normalize the highest trimmed channel to 0 dB and attenuate the others accordingly. When the channel trim feature is enable, a fixed 6 dB attenuation is applied to all channels, i.e. the output level is reduced to about 1V RMS with PCM data and 880 mV RMS with DSD data. This fixed attenuation gives room to trim all channels up to +6 dB.

You can see that the overall shape of the out of band noise is different than with the Denon disc. That means that a different delta-sigma modulator has been used to produce the Pentatone disc.

A concern is often expressed about signal processing on DSD data: does a player convert DSD to PCM to perform the processing? As we have just seen, the level control does not make the typical shaped quantization noise of DSD disappear, a clear sign that there is no decimation of DSD to low sample rate PCM. Let’s check all the signal processings, including the high-pass filter :

Although hard to see, the curves of the right channel declared "small" or "large" almost perfectly overlap in the audio band and out of band, except a slight difference in the noise level above 70 kHz.

The most interesting part are the three curves when the channel is level trimmed. As was already visible above on the three FFTs of pink noise and silence, the tone level (extreme left), the noise floor in the audio band up to 20 kHz or so and the quantization noise all have the same shape and are simply shifted by 6 dB in level.

To me, this strongly points to the fact that the level control is analogue in nature, not digital!

If I am not mistaken, where and how is the level controlled performed? I would venture a guess: in the player’s DACs. I have the feeling that the built-in volume control of the Burr Brown PCM1738 DAC chip used in this player may not be implemented purely in the digital domain, by computing the input data, but by controlling an analogue value made proportional to a digital code at the DAC output. That seems to be compliant with the rather vague statement in this DAC datasheet (page 23) that says: “Each DAC output has a digital attenuator associated with it”. It is possible that the value of the main current source supplying the 67 equally weighted current switches constituting the output of the PCM1738 is variable and digitally controlled by an 8-bits code, which would be sufficient to produce the 240 0.5 dB steps advertised for the output attenuator. This type of volume control is not unheard of: Sony had designed a similar volume control in an old homecinema processor, the SDP-EP9ES, which uses Sony’s own 1 bit DAC followed by a voltage to current converter where the volume control is implemented. Maybe the PCM1738 uses a similar technique in a more integrated fashion. It is worth pointing out that the PCM1738 is highly likely the product of a joint Sony/Burr Brown effort. It is indeed very well established that the PCM1738 also existed under a Sony part number, the CXD9657, that had been used in some Sony SACD players months before the PCM1738 was available on the open market. Apparently, it is common practice in this kind of joint venture that the senior partner gets a short-lived exclusive right to exploit a new chip under its own brand name.

Anyway, this potentially unprecedented discovery ends this review.

I hope you find it interesting.

 

[VintageFlanker]

Also missed that one. Again, thanks for this tremendous amount of work.

I was pretty sure I bought one of these in my early audiophile days... But mine was more likely an earlier model, without HDMI, which I found on LeBonCoin for about a hundred Euros.

 

[Scytales]

A Pioneer DV-868AVi without HDMI? It must have been a DV-757Ai.

 

[VintageFlanker]

A Pioneer DV-868AVi without HDMI? It must have been a DV-757Ai.

You nailed it ! Absolutely this one.

 

[Billy Budapest]

I believe this platform also serves as the basis for the McCormack UDP-1, albeit with a custom power supply and analog output circuitry. I’ve had a UDP-1 Conrad Johnson Edition for 20 years (or so) and never had any complaints about the sound. Nor would I expect it to sound any different than any other CD or SACD player.

 

[Billy Budapest]

Actually, it can’t be the same exact player because it lacks HDMI. I know it is based on a Pioneer Elite player from around the same timeframe. Just found out it’s the DV-45A that it’s based on.

 

[Scytales]

The DV-45A is the same player as the DV-656A, except for the lack of DSD-capable DACs and SACD/DVD-A decoder in the latter. The DV-757Ai is technically very close to the DV-45A, both players use lesser capable DACs and have different output stages for the surround left/right, center and LFE channels than for the main front left/right channels, whereas in the DV-59AVi/DV-868AVi/DV-S969AVi, the DACs and output stages of all six channels are identical. That's the key difference between all these players.

 

[Sokel]

The DV-45A is the same player as the DV-656A, except for the lack of DSD-capable DACs and SACD/DVD-A decoder in the latter. The DV-757Ai is technically very close to the DV-45A, both players use lesser capable DACs and have different output stages for the surround left/right, center and LFE channels than for the main front left/right channels, whereas in the DV-59AVi/DV-868AVi/DV-S969AVi, the DACs and output stages of all six channels are identical. That's the key difference between all these players.

I have the 656A right here and it can play SACD, that's correct.
Should I give it a go and measure it at first chance?

Thanks for the amazing work!

 

[Scytales]

I have the 656A right here and it can play SACD, that's correct.
Should I give it a go and measure it at first chance?

Thanks for the amazing work!

Thanks.

A DV-656A than can read SACD?! Ah, yeah! Now that you said it, I understand after a quick search. There are actually two different Pioneer players bearing the DV-656A designation: a 120 VAC (Region 1) version which cannot play back DVD-A or SACD and and a 230 VAC (Region 2) version (with the suffix letter "S" or "K" in its designation) than can play back SACD!

 

[Sokel]

Thanks.

A DV-656A than can read SACD?! Ah, yeah! Now that you said it, I understand after a quick search. There are actually two different Pioneer players bearing the DV-656A designation: a 120 VAC (Region 1) version which cannot play back DVD-A or SACD and and a 230 VAC (Region 2) version (with the suffix letter "S" or "K" in its designation) than can play back SACD!

Yes, the K version says so in the front too, under the FF button (picture of my very old one) :

..and that's the back with the K:

 

[NTTY]

What a review!!!
I missed it before. I will need 3 days to absorb it and I love that. Thank you!