Pioneer VSX-72TXV receiver.

Manufacturer: Pioneer Electronics (USA) Inc., 2265 East 220th Street, Long Beach, CA 90810; 800-746-6337; www.pioneerelectronics.com

Price: $1,200

Source: Manufacturer Loan

* DTS-ES 

* DTS-96/24

* DTS NEO-6 Movie and Music (no parameter control)

* Dolby Digital EX

* Dolby Pro Logic IIx adjustable for Movie, Music + Parameter Controls

* THX Cinema (Dolby Pro Logic and Neo 6 overlay)

* THX Select 2 Cinema

* THX Surround EX

* THX music and game

* WMA9 Pro decoder

* 11 Surround effect modes with individually adjustable effect levels

* 10 Stereo analog inputs including Phono 

* 3 analog outputs

* 5 Optical and 2 Coax Digital Inputs

* 1 Optical digital outputs

* XM Radio Input and remote control

* Ipod input terminal and remote control

* Speaker output jacks will not accept double banana plugs

* Automatic Format Detection

* Automatic Speaker Set-up (gain, distance, phase check)

* Crossover frequency fixed at 80 Hz

* Analog stereo bypass mode

* 3-band automatic (3-measurement averaging) or user-controllable parametric EQ

* 9-band automatic or user controllable digital EQ (fixed frequency and Q)

* Variable gate time control for automatic adjustment of 9-band EQ (reverb characteristics for each channel and frequency band displayed on the front panel or a PC though the serial port)

* 6 memories for speaker parameter settings with data management

* Optional dB per octave EQ setting from -0.5 to -3dB from 2 kHz

* 8 Pre-Amp Outputs

* On Screen Display (available from all video outputs) for setup screens only

* 7 Composite video inputs

* 3 composite video outputs

* 5 S-video video inputs

* 3 S-video video outputs

* 3 component video inputs

* 1 component video output

* 3 HDMI 1.1 video and audio inputs (may not allow transmission of copyright protected material depending on the capability of the signal source. SACD not supported.)

* 1 HDMI output

* Second room output (2-channel analog) with fixed or variable volume control

* Video up and down converts from composite, S-Video or component video

* Video up converts to HDMI (no down conversion)

* Sound delay to 6 frames

* Watch video and audio sources separately

* Attenuator for analog signals with overload indicator

* 8-unit (including the AV receiver) learning remote control

Once again we preface a review of an AV receiver with a long list of features: you should read this carefully (better yet, go to the Pioneer web site and pull down the user manual). For example, this unit does not appear to support MPEG inputs at the optical port. Missing from this unit are Dolby virtual surround and Dolby headphone. The crossovers are fixed to 80 Hz, which I think has something to do with the fact this unit has THX decoding.

Perhaps most important, you can encode only 2 and not 6 channels of analog inputs to digital. That means all the nifty features I describe below will not be useable in 5.1 if your SACD/DVD-A player has only analog outputs. Instead, the receiver supports direct digital connection for DVD-A if you have a DVD-A player that supports HDMI 1.1 (the 1.0 version will not transmit audio). If you want SACD signals to be processed in multi-channel, you will have to move up to the VSX-74TXV. That unit has an ILink input that will transport digital SACD provided you have an ILink output on your SACD player. I think only Pioneer supports this at the present time. In the case of a DVD-A, you will find it also has standard DVD encoding in Dolby Digital or DTS. While it is possible these lossy coding systems may not be as transparent as 24-bit/96 kHz PCM, they will allow you to use the full features of the receiver in 5.1 from any DVD player using the standard SPDIF digital link. SACD fans are plan out of luck, because these discs have no DVD fallback for 5.1 playback. None of these issues apply if you are using SACD or DVD-A for stereo only reproduction since the unit does have a stereo ADC or you can just listen in CD mode through the SPDIF port (caution: some SACD/DVD-A players mute the SPDIF when they are playing an SACD or DVD-A.$50 Wall Mart DVD players will present no such problem).

This is an early look at this unit, which has been under test for only a couple of weeks. I have not yet found all the warts in its man/machine interface because I have not engaged all the modes associated with audio reproduction. Those of you who want to know about the quality of the HDMI up-conversion and other things video will have to look elsewhere. I can report, though, that everything I have tried so far is intuitive to use and glitch-free. (The exception being the onscreen display, which is dead except in the setup mode, and even in that mode it starts up strange. When you press setup it takes a while for the screen to display, but then it's only for a second and then the screen is black again until it is finally stable. With the on-screen display dead for normal operation, you have to look at the front panel to find out information such as what source you are on, what surround mode you are using, and what the volume level is. Pressing any function on the remote yields nothing visible on the screen. This is maddening because you need a TV in the room--I use a cheap 13-inch tube unit since I am only listening to music--to view DVD-A menus and run the setup functions. I have also not tested the HDMI audio input, since I do not have a DVD-A player with HDMI 1.1 outputs

The big news on this unit is the speaker and room correction software, which Pioneer calls Advanced MCACC--Multi-Channel Acoustic Calibration. Unlike most of these things, it is fully user-adjustable. Results from the auto calibration system appear accurate. For example, this is the first unit where distance settings are always correct. EQ includes a 9-band with fixed center frequency and Q (often called a "graphic equalizer"). Frequencies start at 63 Hz and end up at 16 kHz. The center frequency approximately doubles with each band. In addition, the unit has three bands of parametric EQ, designed to control standing waves.

The automatic setting system for the standing wave system allows 3-position averaging. From my tests, the system only works correctly if you do the 3-position averaging. Unfortunately, standing wave correction is limited to about 60 Hz. Stereo channels receive the same standing wave correction. I have not investigated what it does with center and back yet.

The automatic parameter setting for the 9-band equalization gives the user the option of setting the gate time for the measurements. You can delay the gate time up to 60 msec in 10-msec steps. Now, how do you know where to put the gate time? Pioneer supplies what they call the "reverb view." I do not have all the details, but it looks like an integration of an energy/time curve. The optional advanced instruction manual (you download it from the web) helps you to interpret what the curves are indicating. A curve is generated for all 9 frequency bands on all 7 channels. All curves can be viewed individually on the TV on-screen display. In addition, you can view them in a bunch if you download the results to a PC through a serial bus.

The software on the PC side is downloaded from the Internet. Unfortunately, many new computers no longer support a serial bus. I have yet to try this feature because I would have to drag an old desktop into the listening room to view it. For those you old enough to have played with a serial bus, you will recall that special cables are need depending on how the data are to be transmitted and received. The cable is not included, but enough detail is in the advanced instruction book to make sure you get the correct one from Radio Shack.

From the displays (a sample PC display is shown in Figure 1) you can see one channel's reverb profile across the frequency band. Figure 2 shows a different display of all the channels at one frequency. Here we see a problem with a surround channel. Fixing the problem may be as simple as moving a potted plant that somehow found its way near the left rear speaker when you were not looking. More complex issues require more time to resolve and you may need to try different gate delays to find the optimal sonic result. Since the unit has 6 memories you can store the results for different gate delays and compare them later. You can also copy the results from one memory to another. That way you can keep the original MCACC in one memory and make manual adjustments in the other.

[FIGURES 1-2 OMITTED]

It takes time to get it correct. Think of this as highly productive tweaking. The reverb display and the frequency domain information displayed presented in other setup windows can guide you to where to find better places to put your speakers and listening chair. Now, I have seen reviewers complain about why this receiver cannot just do the setup in a minute and then let them get on to listening. This is just plain wacky. A good deal of the fun of this hobby is taking the time to improve the sound quality of your system. After the initial investment in the VSX-74TX, it costs nothing but time to use the reverb curves, frequency data, and all the manual adjustment options to improve the sound of your system..

The computer display does not include the parametric EQ levels. The on-screen display does not connect the dots between the center frequency of the 9-band graphic EQ. To get a real feel of what MCACC has done, we need to measure the analog transfer function. Figure 3 shows a sample correction curve from the Pioneer taken from an analog input to the preamp out. I used the ETF system to make this plot to show how this low-cost software package can be used for these measurements. For this measurement all you need is the software and a good sound card. You will also want ETF around to check the room response after EQ to make sure the auto EQ has improved things and make manual adjustments as needed. For in-room measurements you need a good microphone in addition to the software and a good external sound card.

[FIGURE 3 OMITTED]

Returning to Figure 3 (note that the grids are at 1dB per division, magnifying the correction), it can be seen the 16 kHz EQ has been pushed a little too much as the Pioneer attempts to flatten out the response of a speaker with limited dispersion above 10 kHz. It is best to go in and reduce this manually. Ignoring the response above 10 kHz, the total correction window is within 3 dB. Note the correction above 250 Hz is at the fixed center frequencies of the 9-band EQ (500Hz, 1kHz, 2kHz etc). If the 9-band EQ were parametric, the correction would contain more granularity.

Figure 4 gives a better view of the EQ applied in the room-dominated response area below 300 Hz. Here, I manually turned off all EQ above 250 Hz. The three parametric equalizers are clearly seen pulling down standing waves, while the three fixed equalizers combine to reduce the total energy below 160 Hz and give a little lift at 250 Hz. Nothing much happens below 80Hz. Unfortunately, the limit for the equalizer is about 60 Hz. This 60-Hz limit appears to apply to most of the equalization systems being offered in AV receivers. This limit may be related to a coefficient resolution problem that occurs when designing low frequency digital filters with a sampling rate of 44 kHz or more.

[FIGURE 4 OMITTED]

Moreover, the correction is very sensitive to listening position, so things may be improved in your main listening chair but made worse elsewhere. Lots of folks are trying to solve this problem at the moment. It may not even be solvable with a single point source. The workaround is to use the multiple memory function and run MCACC for each listening position. Of course this does not solve the problem if multiple listeners are in multiple places in the room.

So, does it improve the sound when you sit in the sweet spot? I have not had enough time to say what it does in all cases, but it made significant improvements in the sound of the Phase Technology 3.3 II speakers when used in stereo (correction curve not shown). Not only was the uniformity of the bass improved, but it took out some energy between 800hHz and 4 kHz. The maximum reduction was 3 dB at 2 kHz. Measurements with the ETF acoustic analyzer with the microphone at the listening position confirmed that the overall response of the speaker was flatter in this room with the EQ applied. Now, I know the Phase Tech is flat from 800 Hz to 4 kHz in quasi-anechoic measurements, so the correction is related to room response.

It is strange that the resultant transfer curve the Pioneer produced looks like the depression found in some audiophile-voiced loudspeakers. What this means will have to wait for the future, but the key is that we started out with a flat loudspeaker in one specific room. It is not as easy to start with a voiced loudspeaker and ask the Pioneer to EQ it, especially because the off-axis radiation pattern response of some audiophile loudspeakers can be very irregular. With the full EQ switched in, the Phase Tech sounded like almost like a megaphone had been removed (again we are taking about the result in one 16 X13 X 8 room), and you will recall the Phase Tech 3.1 II was a great-sounding speaker to start with.

With the high end push (figure 3 shows an example for another speaker), things sounded a little bright. The Pioneer can fix that also. It features something called an "X curve adjustment." What it does is roll off the speaker starting at 2 kHz at fixed per-octave increments (you can only do this with DSP). Use 3dB/octave of the X curve and you can turn any flat speaker into an AR 3. I added in a 0.5dB of tilt per octave (this is not as small as you think with a 2dB cut at 16 kHz which is three octaves up) and the resulting sound was just amazing.

Again, I caution this is one speaker in one room. I have to do a lot more testing. I have done extensive listening only with the Phase Technology 3.1 II. Very preliminary tests of other speakers in bigger rooms have not resulted in such dramatic improvement and in one case I thought flat might be better. In general, the fixes below 250 Hz are usually positive unless the speaker has been voiced to sound correct with a broad push in level below 250 Hz (see the Energy RC-70 review in this issue).

A word of caution on doing AB comparisons between flat and EQ modes The Pioneer does not normalize the EQ correction. In most cases, the broadband level of the speaker will be significantly reduced with EQ on. You must correct for this before making judgments.

We are running out of space for this issue so let me quickly focus on build quality. The TI DSD1791 DAC used in this unit is a significant move up in performance (worst-case SNR at 18 bits and worst-case distortion at 16 bits) relative to what we have seen in lower-priced units, but it is not the same as the top-of-the-line DACs found in some of the $2,000 and up AV receivers (no pun intended, but you get a bit more) Analog amplification is done using dual opamps and not large-scale integrated circuits. This adds to the cost of the unit with so many opamps and the associated resistors and capacitors for each complete stage of analog signal processing. The result of the complexity is reduced distortion and noise in comparison to LSI solutions.

On the down side, analog power rails are reduced to [+ or -] 7 volts in comparison to the [+ or -] 15 volts we have seen on other units including some at lower price points. On the other hand, large scale LSI chips seen in lower priced receivers, that replace 10 or more opamps, electronic pots, and switches in a single package, sometimes have rails as small as [+ or -] 2.5 volts, although [+ or -] 5 volts is more typical. The unit does not have individual opamp buffers for each analog input as some more expensive units do. You thus must rely on the unit driving the Pioneer having a low output impedance to ensure the best performance from the analog switches. The opamps used are mostly the BA4560RF with the NEC 4570G2 used at the preamp outputs. While the NEC 4570G2 is roughly competitive with the more commonly used New Japan Radio NJM 2068, the 4060 is not. The 4060 has higher noise, a lower slew rate, and is generally considered a lower-end opamp.

The unit needs two DSPs to do all the digital processing. An Analog Devices ADSP-21266 S capable of doing 400 million 32-bit floating point multiply accumulations per second is first at bat. It does most of the Dolby digital, THX, and DTS processing. A Freescale (Motorola spin-off) DSP56371 24 bit fixed point is up next (48-bit double precision is also possible, but this slows the chip down). It does 180 million multiply accumulations per second. Its job in life is to do the EQ functions and bass management. Three microcontrollers and a lot of other digital chips are in the AVR 72 to keep all the digital systems working in harmony.

The maximum power draw for this unit is 7A with a UL safety mark. That indicates that it is not going to produce full power with 7 channels driven from 20Hz to 20kHz; the users manual does not claim it does (some of the promo literature does imply this--probably the fault of the copy editor) The 35-lb weight of the unit indicates the transformer and heat sinks are designed for about 7 amp maximum power supply draw. Getting all 7 channels to full power would require twice the iron. The only difference in the power amp between the VSX 72 and the more expensive VSX 74 is a change of the primary filter capacitor from 10,000 [micro]Fd to 15,000 [micro]Fd per unregulated rail. The output devices comprise a single Sanken SAP17N and SAP17P Darlington pairs. These are more than capable of driving a 4-ohm load continuously from 20 Hz to 20 kHz at more than a 200 watts. The amp is not going to be able to do that at 2 ohms, however. Pioneer rates the unit as capable of delivering continuous power with two channels driven of 120 watts from 20 Hz to 20 KHz at 0.05% distortion into an 8-ohm load. Into 6 ohms at clipping the continuous output for two channels is 160 watts. Again, this is very good, but some $2,000 and up receivers can come up with more power at lower distortion into 6 ohm loads and below. Beyond that you need to go with a separate power amp.

One really important feature of the Sanken SAP17 is the built-in temperature-compensating diodes used to bias the transistors. This is the first time I have seen these in a commercial unit. Distortion can be dramatically reduced, since fast thermal changes in the output device are communicated to the diodes that are on the same hunk of silicon. Bias stabilization time is also dramatically reduced. The front end of the power amp is not as fancy as the amps found in the $2,000 and up AV receivers, but it does have what it takes to produce the low 0.05% distortion figures 20 Hz to 20 kHz specified for this unit. With the most advanced topologies found in some $2,000 AV receivers, distortion may be specified as low as 0.02%. Very expensive separate stereo power amplifiers such as the Bryston 4B SST can do 0.005% at 300 watts into 8 (0.007% at 500 watts in 4 ohms continuous), but the Bryston costs $3,000 and do not even think about using more than one Bryston 4B in a 15-amp power line. If you want 7 channels of Bryston you will need multiple power lines and more than $10,000.

So what do I conclude after two weeks of evaluation? Build quality is good enough so the receiver should be transparent in its sound. Moving up in price will only give marginal improvements that will be measurable but I doubt they will be audible unless the power amp runs out of current or voltage. No question about it--the unit is worth its $1,200 price. With the electronics earning the seal of approval, we get to marvel about the power of the EQ system of the VSX-72. Even at this early stage, I recommend you try it even if you are only going for a two-channel system. Make sure you get a money-back guarantee if the EQ system does not improve the performance of your system, but make sure you take the time to adjust it correctly with manual corrections of some of the parameters as needed before you draw any conclusions.--DAR