Rated Power Output, All Channels Driven: Stereo mode, 140 watts per channel into 8 ohms, 20 Hz to 20 kHz, at 0.05% THD or 170 watts per channel into 8 ohms at 1 kHz and 0.7% THD; surround modes, 140 watts x 5 into 8 ohms at 1 kHz and 0.7% THD.
Dimensions: 17 3/8 in. W x 7 1/2 in. H x 19 1/3 in. D (43.4 cm x 18.1 cm x 48.6 cm).
Weight: 54 lbs. (24.5 kg).
Price: $2,800.
Company Address: 222 New Rd., Parsippany, N.J. 07054; 201/575-7810.
Denon's AVR-5600 comes as close to fulfilling my A/V receiver wish list as anything I've seen. First of all, it's a powerful beast, rated at 140 watts per channel into 8 ohms with all five of its channels driven. As with most A/V receivers, the five-channel rating is specified at 1 kHz (with 0.7% distortion), but the stereo rating (also 140 watts but from 20 Hz to 20 kHz with 0.05% THD) hints that real-world performance is likely to be better than claimed. Second, the AVR-5600 is equipped with a Dolby Digital Surround (AC-3) decoder and is Home THX-certified in both Dolby Pro Logic and Dolby Digital Surround, i.e., for both matrixed and discrete surround soundtracks. Its Dolby Digital section can accept an RF-modulated AC-3 signal from today's laserdisc players without an external box, as well as direct AC-3 bitstreams from DVD players (and, in a year or two, HDTV receivers). What's more, the AVR-5600 can accept the Dolby Digital bitstream from any of its four digital inputs. (Of these, one is a coaxial connection; the other three are optical.) In a novel twist that provides the AVR 5600 with unusual versatility, the receiver's setup menus enable you to allocate any dig ital input to any of seven program sources. However, only one digital input, "Optical-3," is accompanied by an output (also optical) to feed a digital recorder. Three sources ("Phono," "Tape-2," and the built-in tuner) cannot be assigned to digital inputs. When you select a source, the AVR-5600 uses the digital input, if one has been allocated; otherwise, it uses the analog input whose name corresponds to the source. For digital in-puts, the Denon analyzes the input bit-stream to determine whether it's standard PCM or needs to be fed through the AC-3 decoder. Thus, the AVR-5600 handles all current (or soon-to-exist) audio program sources-stereo or matrixed analog, stereo or matrixed digital, and discrete 5.1-channel digital (including RF-conveyed AC-3 from laserdiscs). Like most A/V receivers these days, this one can display menus, set tings, and even a speaker-setup diagram on your TV screen.
The AVR-5600 has provisions for five audio/video and five audio-only sources (counting its tuner), including two VCRs and two audio recorders. The tuner features 40 presets in five banks of eight and automatic memorization of local stations (with manual override). It also incorporates the Radio Data System (RDS), which Denon has persistently championed for car and home FM. As more and more FM stations start broadcasting with RDS, this will en able you to search among the RDS stations in your area to find a particular type of pro gram, get traffic reports, display transmit ted text (such as song and album titles), and receive emergency alerts. (Editor's Note: About 12 RDS stations can be received in suburban New Jersey, where I live. All of them list their program types, but none transmits traffic reports or useful text yet. –I.B.) The AVR-5600 has recording outputs to complement its audio tape and VCR inputs and, of course, an output for your TV monitor. Denon also provides stereo connections for an audio system and an additional TV in a second room. All connections are on the back, and all video signals (except that for the second TV) are carried in S-video and composite-video forms. All audio and composite-video RCA jacks are gold-plated, as is the front-panel head phone jack. There are preamp outputs for each of the five main channels and for a powered subwoofer but no input connections for the internal power amps. Since the AVR-5600's five power amps are identical, the omission is justified; there's little purpose in swapping identical power amps.
The amplifier outputs are multiway binding posts that accept single or double banana plugs. Parallel-connected pairs of posts are provided for the three front speakers, to facilitate bi-wiring. Three switched convenience outlets, with a total power rating of 120 watts, are provided. The FM antenna connects to a 75-ohm F connector; the AM loop connects to a pair of wire clamps. A grounding terminal is near the phono input, for easy connection to a tonearm or a turntable frame.
The AVR-5600 may not win any beauty contests, but its front panel is inoffensive and, when the hinged door at the bottom is closed, rather simple. The "Power" button is on the left, "Master Volume" is on the right, and below the display are 11 small se lectors: "AC-3 RF," "Phono," "CD," "Tuner," "VDP/DVD," "TV/DBS," "VCR-1," "VCR-2," "V.AUX," "DAT/Tape-1," and "Tape-2/Monitor" (with a nearby LED to indicate when that monitor is in use). I prefer individual program selectors like these to cyclic selectors, because they're faster and easier to use and eliminate an on-screen menu. (I also approve of Denon's use of re lays rather than solid-state switches to route signals, because relays are less likely to inject noise and distortion.) Behind the hinged door are 16 buttons that control surround modes, input adjustments, tuner presets, and recording/multiroom program selection. These are flanked by a "Phones" jack on the left and "Bass" and "Treble" knobs on the right.
Except for manipulating the tone controls, you can select any thing from the remote that you can from the front panel, albeit some times in a different fashion. For ex ample, selecting sound modes via the remote involves cycling through the modes, whereas the panel offers direct access to most of them.
These sound modes are "Direct" (which bypasses the tone controls when you're listening in stereo), "Stereo" (which activates the bass and treble controls), and six surround options. The surround modes are "Dolby Surround" (which automatically chooses Dolby Digital over Pro Logic if the receiver is fed an AC-3 program), "Home THX Cinema" (which adds Home THX enhancement to what ever Dolby Surround mode is used), "Mono," "Wide Screen" (DSP-enhanced Dolby Surround), "5CH Stereo" (which redistributes a stereo source among all speakers), and "DSP Simulation." The last chooses, in cyclic fashion, among five simulated surround modes ("Super Stadium," "Rock Arena," "Jazz Club," "Classic Con cert," and "Matrix"). Both "Home THX Cinema" enhancement and the various simulation modes are implemented in a Motorola 56004 DSP chip.
Three buttons behind the hinged panel control the audio input settings for each source and, once adjusted, remain associated with that source unless you reset it. The first two raise or lower input level, and the third toggles "Analog/Digital" input connections. (In retrospect, it strikes me that this third switch probably lets you use four more program sources than appear on the front panel: Just connect whatever sources you wish to the analog inputs that correspond to the four source positions you allocated digital inputs to, and then use the toggle to select either input set.) Next come three tuning buttons ("Shift," "Down," and "Up"). With the last two but tons on the bottom row, you set the program source to be fed to a second room and to the recording outputs. The "REC/Multi Source" button cycles through the 10 possible sources or, in its 11th position ("Source"), uses whatever source is set by the main program selector. The other but ton, "REC/Multi Mode," assigns the selected source to the recording outputs or to the multiroom jacks.
The AVR-5600's display is in three sections. On the right is a two-digit master volume LED that's always active. On the left are 10 small LEDs. Six of them indicate the presence of a Dolby Digital AC-3 signal and how many channels it contains. (AC-3 is just a compression system and can be used to carry stereo or mono audio as well as 5.1-channel surround.) The other LEDs indicate when a digital (rather than analog) in put is in use, when Dolby Surround decoding is on, whether there are signals in the low-frequency effects (LFE) channel, and when circuits (presumably the inputs) overload. Unfortunately, the LEDs' labels are essentially unreadable, which makes the array virtually useless except as decoration.
The display's center section is a fluorescent panel whose brightness can be set to three different levels or turned off by the remote's "Dimmer" but ton. If you turn the display off, it comes back to life briefly when you change the program source or other settings.
The remote can control three audio and three audio/video sources. It's a universal remote, preprogrammed with the control codes of equipment from several major companies and capable of learning the codes of other makers' equipment. The remote has 59 but tons and two switches. The buttons don't illuminate, but they are color-coded, clustered in groups, and shaped differently to give a tactile indication of use. Twenty-one "in frequently used" buttons are covered by a hinged panel, which keeps them out of the way. Strangely, the major tuner controls ("Tuning" up/down, "Band," "Mode," "Memory," "RDS," and the buttons that activate RDS program search and text display) are among those that are hidden.
The setup buttons are also be hind the remote's hinged panel.
"Enter" brings up the main on screen menu ("Surround Parameters," "System Setup," and "Menu Off"). The on-screen display indicates the active selection by a pointing finger, with upward- and downward-pointing arrows marking the selections above and below.
The indicator is moved by up/down "Cursor" buttons; "Enter" activates your selection.
The main menu's "Surround Parameters" option is used for toggling the Denon receiver's "Cinema EQ" circuit on and off and for adjusting "Room Size" and "Effect" levels when you are using DSP-based enhancement.
The "System Setup Menu" offers eight submenus. These are used for matching the AVR-5600's individual-channel frequency responses, delays, and levels to your speaker setup; controlling a peak limiter for the subwoofer; allocating the digital inputs; toggling the AC-3 decoder's dialog normalization on and off; setting tuner presets automatically; and controlling the on-screen display. Several of the submenus are self-explanatory, but four of them deserve a bit of explanation.
In the "Speaker Configuration" sub menu, you inform the AVR-5600 whether you have center or surround speakers and whether each speaker, except a subwoofer, is small or large. (More and more micro processor-controlled surround decoders use this setup method, which is clearer and more to the point than choosing "Center, Wide" for a large center speaker and "Center, Normal" for a small one.) If you indicate you have small main speakers, you're offered a choice of only "Small" or 'None" for the center and surrounds, and the AVR 5600 redirects bass energy accordingly.
To set "Delay Time," you enter the distances between the listening position and each speaker, in feet or meters, rather than having to calculate and enter delay times per se. The left and right front speakers are assumed to be equidistant from the viewing position, so one entry serves both.
For "Channel Level" adjustments, you can press buttons to move test tones from speaker to speaker or have these tones cycle around the array automatically. Once you've activated "Test Tone Start," an array of speaker icons and five channel-level adjustment bars appear on-screen to guide you through the procedure.
The " Subwoofer Peak Limit LEV." Sub-menu enables you to activate or deactivate a peak limiter on the subwoofer channel. It also helps you determine what peak limiter level will give you maximum bass from your sub without excessive distortion.
========
MEASURED DATA
FM TUNER SECTION
Sensitivity: IHF usable, 15.6 dBf in mono; 50-dB quieting, 18.6 dBf in mono and 41.1 dBf in stereo.
S/N Ratio, 65-dBf Signal Input: Mono, 78.1 dB; stereo, 71.7 dB.
Frequency Response, Stereo: 20 Hz to 15 kHz, +0.32,-0.44 dB.
Channel Balance: ±0.00 dB.
Channel Separation, 100 Hz to 10 kHz: Greater than 42.2 dB.
THD + N at 65 dBf, 100% Modulation: Mono, 0.42% at 100 Hz, 0.36% at 1 kHz, and 0.46% at 6 kHz; stereo, 0.38% at 100 Hz, 0.33% at 1 kHz, and 0.74% at 6 kHz.
Capture Ratio at 45 dBf: 1.4 dB.
Selectivity: Adjacent-channel, 4.6 dB; al ternate-channel, 59.5 dB.
Image Rejection: 53 dB.
AM Rejection: 62 dB.
Stereo Pilot Rejection: 34 dB.
Stereo Subcarrier Rejection: 50.2 dB.
AMP SECTION, STEREO MODE
Output Power at Clipping (1% THD at 1 kHz): 8-ohm loads, 185 watts/channel (22.7 dBW); 4-ohm loads, 290 watts/channel (24.6 dBW).
Dynamic Output Power: 8-ohm loads, 195 watts/channel (22.9 dBW); 4-ohm loads, 310 watts/channel (24.9 dBW); 2-ohm loads, 500 watts/channel (27 dBW).
Dynamic Headroom re 8-Ohm Rating: 1.4 dB.
THD + N, 20 Hz to 20 kHz: 8-ohm loads, less than 0.045% at rated output and less than 0.025% at 10 watts/channel; 4 ohm loads, less than 0.103% at 200 watts/channel, less than 0.066% at 145 watts/channel, and less than 0.041% at 10 watts/channel.
Damping Factor re 8-Ohm Loads: 215.
Output Impedance: 39 milliohms at 1 kHz, 63 milliohms at 5 kHz, 107 milliohms at 10 kHz, and 165 milliohms at 20 kHz.
Frequency Response: "Direct" mode, 20 Hz to 20 kHz, +0,-0.32 dB (-3 dB be low 10 Hz and at 117.6 kHz); "Stereo" mode, 20 Hz to 20 kHz, +0.54,-0.12 dB (-3 dB below 10 Hz and at 70 kHz).
Tone-Control Range: Bass, +9.4, -7.6 dB at 100 Hz; treble, +9.3, -8.7 dB at 10 kHz.
Subwoofer Crossover: High-pass, -3 dB at 78 Hz and -6 dB at 60 Hz, 12 dB/octave; low-pass, -3 dB at 45 Hz and -6 dB at 68 Hz, 24 dB/octave.
RIAA Equalization Error, 20 Hz to 20 kHz: "Direct" mode, +0.34, -0.57 dB; "Stereo" mode, ±0.5 dB.
Sensitivity for 0-dBW (1-Watt) Output: CD input, 9.8 to 77.4 mV (15.6 mV with input attenuator at 0 dB); MM phono input, 0.127 to 1.13 mV (0.246 mV with input attenuator at 0 dB).
A-Weighted Noise: CD input, -83.5 dBW in "Direct" mode and -80.9 dBW in "Stereo" mode; MM phono input, -78.3 dBW in "Direct" mode and -77.7 dBW in "Stereo" mode.
Input Impedance: CD input, 54 kilohms; MM phono input, 46 kilohms + 200 pF.
Input Overload (1% THD at 1 kHz): CD input, 3.2 to more than 10 V (6.4 V with input attenuator at 0 dB); MM phono input, 54 to 168 mV (105 mV with in put attenuator at 0 dB).
Channel Separation, 100 Hz to 10 kHz: Greater than 61.4 dB.
Channel Balance: "Direct" mode, ±0.015 dB; "Stereo" mode, ±0.05 dB.
Recording Output Level: 470 mV for 500-mV signal at CD input; 270 mV for 5-mV, 1-kHz signal at phono input; 650 mV for 100%-modulated FM signal at 1 kHz.
Recording Output Impedance: 700 ohms.
AMP SECTION, DOLBY PRO LOGIC MODE
Output Power at Clipping, 8-Ohm Loads: Main channels, 185 watts/channel (22.7 dBW); center channel, 200 watts (23.0 dBW); surround channels, 180 watts/channel (22.6 dBW).
THD + N at Rated Output, 8-Ohm Loads: Main, less than 0.038%, 65 Hz to 20 kHz; center, less than 0.041%, 100 Hz to 20 kHz; surround, less than 0.168%, 100 Hz to 10 kHz.
Frequency Response: Main, 20 Hz to 20 kHz, +0.57,-0.29 dB (-3 dB below 10 Hz and at 23.3 kHz); "Center, Wide" mode, 20 Hz to 20 kHz, +0.56,-2.64 dB (-3 dB at 19 Hz and 23.3 kHz); "Center, Normal" mode, 76.2 Hz to 23.3 kHz, +0.37, -3 dB; surround, 80.5 Hz to 7 kHz, +0,-3 dB.
A-Weighted Noise: Main,-80.4 dBW; "Center, Wide" mode, -75.8 dBW; surround,-75.8 dBW.
Channel Separation at 1 kHz: 57.9 dB or greater.
DOLBY DIGITAL (AC-3) MODE
Channel Balance: 1.03 dB or better.
Frequency Response: Main channels, 20 Hz to 19.1 kHz, +0.57,-0.09 dB; center channel, 20 Hz to 19 kHz, +0.57,-0.12 dB; surround channels, 20 Hz to 19.1 kHz, +0.03,-0.35 dB.
THD + N at 1 kHz for 0-dBFS Signal: Main, 0.011%; center, 0.056%; surround, 0.058%.
Channel Separation at 1 kHz: 45.3 dB or greater (see text).
D/A CONVERTER SECTION
Frequency Response: 20 Hz to 20 kHz, +0.01,-0.32 dB.
THD + N at 0 dBFS: Less than 0.089%, 20 Hz to 20 kHz.
THD + N at 1 kHz: Below-77.9 dBFS from 0 to-90 dBFS and below-83.3 dBFS from-30 to-90 dBFS.
Maximum Linearity Error: Undithered signal, 4.73 dB to -90 dBFS; dithered signal, 5.2 dB to -100 dBFS.
A-Weighted S/N: 93.6 dB re 0 dBFS for infinity-zero signal.
Quantization Noise:-85.5 dBFS.
Dynamic Range: Unweighted, 89.1 dB; A-weighted, 91.9 dB.
Channel Separation: Greater than 44.5 dB, 125 Hz to 16 kHz.
=========
Measurements
Fig. 1--Quieting characteristics and separation, FM tuner section.
Fig. 2-THD + N vs. frequency, FM tuner.
Fig. 3--THD + N vs. power output.
Fig. 4-Amplifier THD + N vs. frequency, 8-ohm load (A) and 4-ohm load (B).
Fig. 5-Frequency response from CD analog input.
Fig. 6-Tone-control range and response of subwoofer crossover.
Fig. 7-RIAA phono equalization accuracy.
Fig. 8-Frequency response, Dolby Pro Logic mode.
Fig. 9-THD + N vs. frequency, Dolby Pro Logic mode.
Fig. 10-Frequency response, Dolby Digital mode.
The AVR-5600 has a number of things going for it. Its FM tuner is a good bit more competent than those I've seen in other recent receivers, and its power amps are clean and potent. The analog preamp portion is fine too; so, in most respects, are the surround decoders. The major weakness I find lies in its D/A converters, which is odd, as Denon has established a rather good reputation in this area. But let's start with the AVR-5600's strengths.
As you can see in Fig. 1, the FM tuner snaps to attention quickly. The benchmark 50-dB-quieting point is reached with an RF input of 18.6 dBf in mono and 41.1 dBf in stereo, both very respectable. Band-limited (but unweighted) signal-to-noise ratio is greater than 78 dB in mono and almost 72 dB in stereo at 65 dBf, improving by another 2 dB or more with stronger stereo signals-impressive! Channel balance was perfect. Channel separation was remarkably uniform over the full audio band and was, at 42 dB or better, above average over the most important audio region.
Evaluating a receiver this elaborate requires a huge number of tests, with far more graphs than there's room for in this review.
Mostly, I've omitted those that are easy to describe. For example, FM frequency response was almost ruler flat over most of the band, drooping slightly in the bass (-0.44 dB at 20 Hz) and rising a trifle in the treble (+0.32 dB at 15 kHz).
I measured the FM tuner section's total harmonic distortion plus noise (THD + N) using an RF input signal of 65 dBf. Usually, I present the data taken at 100% modulation because that's what the standard calls for; in Fig. 2, I've added curves made at 50% modulation, which is probably more typical of average broadcasts. (The results in "Measured Data" are for only the worse case, 100% modulation.) At 50% modulation, THD + N remains less than 0.25% to 4 kHz in mono (2.7 kHz in stereo) and less than 0.3% across the entire band in mono and out to 4 kHz in stereo. This is pretty good performance, especially in light of the tuner section's reasonably high adjacent-channel selectivity of 4.6 dB and its alternate-channel selectivity of nearly 60 dB.
Image rejection didn't set any record, but very few people live in places where this should be of any concern. For the rest of us, the AVR-5600's combination of relatively low capture ratio and high AM rejection bodes well for handling multipath. Stereo pilot and subcarrier rejections weren't bad but could stand improving; if you intend to record cassettes with Dolby noise reduction, be sure to use the deck's MPX filter.
Figures 3 and 4 show the distortion characteristics of the AVR-5600's power amplifiers when driving 8- and 4-ohm loads. The data was taken by driving the analog CD in put and using the "Direct" mode in order to bypass the tone controls and all processing circuits. As the curves in Fig. 3 indicate, the AVR-5600 can deliver 185 watts per channel continuously into 8-ohm loads (290 watts into 4-ohm loads) at clipping (1% THD) with a 1-kHz signal. A trifle less power is available at 20 Hz and 20 kHz.
To assess THD + N versus frequency into 8 ohms (Fig. 4A), I used output levels from 1 watt/channel up to rated output (140 watts/channel). However, distortion was so low at 1 watt/channel that the curve showed only noise, so I omitted it. Denon doesn't specify power for 4-ohm loads, so in Fig. 4B I added a curve at 200 watts/channel, safely below the maximum output power capability indicated by Fig. 3. (I also assigned that output level as my 4-ohm power rating for "Measured Data.") As you can see from Figs. 4A and 4B, distortion is pretty constant and remains quite low from 100 Hz to 3 kHz or above. It turns up ever so slightly in the bass and more sharply in the treble. The curves show performance of the right channel, the poorer of the two on my sample. But because that channel cleared Denon's 0.05% 8-ohm full-power specification, there's no reason for complaint.
The AVR-5600 delivered just a trifle more power into 8- and 4-ohm loads with the IHF tone-burst signal than with a continuous signal. This suggests that the clipping point is determined primarily by the power supply's voltage rather than by its available current and that performance is unlikely to degrade substantially when more than two channels are driven simultaneously. With 2-ohm loads, the AVR-5600 delivered a whop ping 500 watts/channel on tone bursts. The damping factor clocked in at more than 200, but output impedance slightly more than quadrupled when the signal frequency increased from 1 kHz (39 milliohms) to 20 kHz (165 milliohms).
I measured frequency response (Fig. 5) from the AVR-5600's analog CD input to its speaker outputs, using 8-ohm loads. (Only the left channel's response is plotted; the right channel's was essentially identical.) Response is flatter in the "Direct" mode, which bypasses the tone controls. In "Stereo" mode, the tone controls are in the circuit; even though I set them to their center detents, the curve for that mode is some what swaybacked. The response in "Direct" mode is admirable, but the "Stereo" curve is more the one to study because, as we will see, its behavior carries through across the board-oddly, even in Dolby Digital Surround! The-3 dB points in "Stereo" mode are okay, but I'm less sanguine about the broad rise in bass response (+0.54 dB at 150 Hz) and the smaller rise in treble response (+0.2 dB around 5 kHz). I presume that these anomalies came from the tone-control circuitry, but I can't be certain.
Tone-control characteristics are shown in Fig. 6. Both controls' curves shelve, the bass below 100 Hz and the treble above 10 kHz. As indicated in "Measured Data," each control provides a bit more boost than cut at the benchmark test frequencies of 100 Hz and 10 kHz.
Figure 6 also shows the subwoofer cross over's characteristics. The high-pass section's performance is typical for Home THX:-3 dB at just about 80 Hz and-6 dB at 60 Hz, with a 12-dB/octave slope. The low-pass section is peculiar in two ways. Al though its slope ultimately reaches a nominal 24 dB/octave, it starts off rather gradually. And its response is never really flat except between 15 and 20 Hz (not seen in the graph).
Phono response (Fig. 7) has a sway backed shape in "Stereo" mode, similar to what I measured from the analog CD input.
In "Direct," it's flatter, though with a relatively minor bass rolloff.
For "Measured Data," I tested input sensitivity and input overload over the full range of input attenuator settings; all other tests were made with the input attenuator at 0 dB. (The input attenuator does not function in "Direct" mode.) With the attenuator at 0, the sensitivity data was typical of receivers in general, and the levels and source impedance of the recording outputs seemed fine to me. It's possible to overload the analog CD input on the test bench, but it's highly unlikely to occur in practice, especially if you use the input attenuator. The phono input's overload point should be ad equate for moving-magnet (MM) cartridges, though here, too, you might need to use the input attenuator. Phono input impedance was classic, in that it could be modeled by a parallel combination of a resistor and a capacitor, and the results were right on the mark; analog CD input impedance was also fine.
Channel separation and channel balance from the line (CD) input were good. How ever, as is increasingly common, the AVR 5600 doesn't have a conventional balance control; if a source is unbalanced you must correct matters with the speaker calibration controls.
On my sample, A-weighted noise was 0.6 to 0.7 dB higher in the right channel than in the left, so the right channel's results are reported in "Measured Data." The "Direct" mode was several decibels quieter from the line-level (CD) input, as you might expect. But on the phono input, preamp noise pre dominated, leaving little difference between "Direct" and "Stereo." Noise spectrum analyses revealed small but discernible power-line-related components at 60 Hz and at each harmonic through the fifth (300 Hz).
I evaluated Dolby Pro Logic operation by driving an analog input with simulated Pro Logic test signals and monitoring the speaker outputs of each channel. To save space, Fig. 8 presents frequency response only for standard Dolby Pro Logic (although I checked the Home THX enhancements and Denon's Cinema EQ system with the same test signals). The front channels have the same mid-bass response peak that appears in the "Stereo" curves in Figs. 5 and 7. On the other hand, surround-channel response is dead flat until it exhibits the "normal" bass rolloff. I admit to being somewhat surprised by this rolloff, since I had set the receiver up for large main and surround speakers. (The curve marked "Center, Wide" was taken using the setting for a large center speaker; the "Center, Nor mal" curve was taken with the setting for a small center speaker. Both were taken with the AVR-5600 programmed to assume that a subwoofer was in the system.) In each front channel, treble response was limited to 23.3 kHz by the anti-aliasing filter at the input of the A/D converter that feeds the processing circuitry from analog signals.
As usual, the surround channel rolls off sharply above 7 kHz, as called for by the Dolby Pro Logic standard.
The curves I took with Home THX essentially reflected the Dolby Pro Logic responses, modified by THX re-equalization and timbre matching.
Response was down 3 dB at 8.5 kHz (-6 dB at about 15 kHz) in the three front channels (THX re-equalization); surround-channel response showed signs of comb filtering and timbre matching.
The Cinema EQ curves were rather different. With this mode engaged, bass response in all three front channels rolled off sharply below 73 Hz (the-3 dB point), whether the AVR 5600 was set up for large or small center and main speakers. There was some additional mid-bass boost (about 0.76 dB at 210 Hz) and a gradual treble rolloff (-1 dB at 2.2 kHz,-3 dB at 5.8 kHz, and-4.7 dB at 20 kHz). Surround-channel response was the same as in Dolby Pro Logic mode.
Figure 9 shows THD + N versus frequency at rated output (140 watts into 8 ohms) in all five channels, using Dolby Pro Logic decoding. I used a 22-kHz low-pass filter here, to remove the ultrasonic garbage generated in the digital converters. Distortion in the three front channels remains at or be low 0.04% from 100 Hz to 20 kHz. Surround-channel distortion is higher but still very good.
Fig. 11-THD + N vs. frequency, D/A converter section.
Fig. 12-THD + N vs. signal level, D/A converter section.
Fig. 13-Linearity error, D/A converter section.
Fig. 14-Fade-to-noise test, D/A converter section.
In Dolby Pro Logic, the 1-kHz clipping point for each channel occurred at pretty much the same output level as in stereo.
Steady-state channel separation at 1 kHz ranged from a low of 57.9 dB (surround to right front) to a high of almost 105 dB (surround to left front). On average, separation was well into the 60-dB range, which is very good. The A-weighted noise was good in the main channels and fairly good in the others.
Strangely, channel separation in Dolby Pro Logic was better than in Dolby Digital. This can be some what misleading, however, as Pro Logic will achieve such separation only for strongly steered signals, whereas Dolby Digital will maintain it under all conditions. The worst separation in Dolby Digital was between the left front channel and the center, between the left and right front, and between the left front and right surround; each was just over 45 dB. Most of the other results were safely over 70 dB, al though a few were in the 50s.
Figure 10 shows frequency response of the various channels with a Dolby Digital test signal.
Again, we see the 0.57-dB mid-bass response hump in all three front channels but essentially flat response in the surrounds. Distortion at 1 kHz (0 dBFS) was a low 0.011% in the main channels and about 0.056% in the center and surrounds. Channel balance was within ±0.75 dB.
Lately I've taken to exploring behavior of the D/A converters in products like this, since I've un earthed surprising and disquieting results from some of them. The frequency response of the AVR-5600's converter was unexceptional in the good sense; except for a slight rolloff at both ends and a treble ripple so small as to be hardly noticeable, it was flat. The curves for THD + N versus frequency (Fig. 11) are relatively unremarkable, too. Except for a peak at 16 kHz (which better converters would not produce these days), distortion is less than 0.02% on the right channel, the poorer of the two.
More worrisome characteristics appear in the curves that relate to D/A converter linearity. Those plots are for THD + N versus level at 1 kHz, linearity error versus level, and the fade-to-noise test.
The AVR-5600's THD +N at signal levels from 0 to-6 dBFS was higher than I usually see, so I had to adjust the vertical scale of Fig. 12 to accommodate it. Sometimes an audio component has elevated THD + N near 0 dBFS because of problems in its analog circuitry, but that doesn't seem to be the case here. The right-channel curve flattens between-3 and-6 dB, suggesting that converter nonlinearity, not output amplifier compression, is the cause of this behavior. I can accept the left channel's performance, more or less, but note the kick-up at -30 dBFS on the right channel. These converters are simply not as linear as I would have hoped.
In Fig. 13, I again had to adjust the vertical scale to accommodate the level of error that I found. And the poorer channel's fade-to-noise curve (Fig. 14) occupies al most ±10 dB, which is unusual. Neither channel's performance is comparable to that of today's better converters. If you check "Measured Data," you'll find that the AVR-5600's A-weighted S/N ratio, quantization noise, and dynamic range also are less than stellar; channel separation was actually worse with the digital input than with the analog CD input.
Use and Listening Tests
So what's going on? I've not seen a really bad DAC in a CD player in some time, and I had assumed I could count on basically decent DACs in processors and A/V receivers.
I have felt that my time would be better spent exploring output power, high-level distortion, surround sound characteristics, and the like. But it takes six DACs to implement Dolby Digital, and I guess that's put pressure on the manufacturers to cut corners. Denon is not the only one to have done so, but I'm still surprised to find it here, especially since the company claims to use 18-bit DACs in the AVR-5600.
Could the right-channel DAC be a fluke of my particular sample? Perhaps, but the left-channel converter, while decent, isn't outstanding either. There are four more DACs in this receiver; unfortunately, they're not easily accessible for measurement. I have to assume that what I can measure is representative of what exists.
Could I hear the problems I measured? Can't say I did on Dolby Pro Logic or Dolby Digital movie soundtracks, but I could hear something amiss on well-recorded stereo. I preferred the sound of the DAC in my Sony CDP-XA7ES player to the sound of the DAC in the Denon. Okay, so the Sony's converter is extraordinary, and the player alone costs a bit more than the AVR-5600. But I also preferred the MASH converter in my old Sansui CD-X711 to the Denon's DAC.
Movies fared much better than CDs; maybe I become less critical with so much going on. Yet I didn't feel that the AVR-5600 provided quite the same bass impact as my reference audio system when I listened through the same speakers and subwoofer. I can't make quick and direct A/B comparisons in my home theater (nor can one match levels quite so precisely when using digitally adjusted attenuators), so I'm less sure of my conclusions-but that's how I felt.
As I said at the beginning of this review, the Denon AVR-5600 fulfills my wish list of features better than other receivers I've seen. A more robust and generally capable A/V receiver would be hard to come by. It could use a better manual (Denon tells me one's in the works), and it could also use a few modifications to improve the converters and remove a few operational oddities (limited bass response in some channels under some conditions, swaybacked response, etc). Some of these may require only software modifications (so much is done in software these days!); others may require component changes. I hope Denon will make them, because the AVR-5600, which is already a very good receiver, has the potential to be a really great one.
- EDWARD J. FOSTER
(adapted from: Audio magazine, March 1997)
Also see:
Denon DRA-400 AM/FM Receiver (Feb. 1983)
Denon AVR-3300 A/V Receiver (Equip. Profile, Sept. 1999)
Denon 20-bit CD technology (ad, Jan. 1989)
Denon DR-330 Cassette Deck (Mar. 1982)