Product Review - Sherwood Newcastle R-963T Audio/Video Receiver, Part 2 - April 2002
The Heart of the Matter
The crucial piece of any receiver or processor these days is the choice of digital components. For a product such as this, the 3 main digital items are the Digital Signal Processor (for algorithmic operations on digital audio data), Digital to Analog Converter (for digital inputs) and Analog to Digital Converter (for processing analog inputs digitally). These 3 pieces can define the inherent capabilities of a particular product.
In the case of the 963, the DSP is the Crystal Semiconductor (Cirrus Logic) CS-49326. This DSP has a host of functions included, among them are DD, DTS, DTS-ES, DPL-2 and MPEG decoding. In addition, the DSP performs time alignment and could perform Bass Management in the digital domain (this receiver has shifted Bass Management to the analog domain). The limiting capability is the number of MIPS (Millions of Instructions per Second, or if you're cynical, Meaningless Index of Processor Speed). DSPs can be ganged together to provide more horsepower (i.e., more intense calculations / operations) - as our systems utilize more and more digital functions, the MIPS will need to increase.) Fortunately, MIPS for DSPs seem to be following Moore's Law just as CPUs are. Since a DSP is a dedicated function CPU, this isn't surprising to me.
Digital to Analog Conversion is provided by Analog Devices' AD-1852, which is a stereo Delta-Sigma DAC, capable of converting PCM data up to 24bit/192 kHz to analog, with a 114 dB Signal to Noise Ratio, and 117 dB dynamic range. These indicate real-world resolution of just under 20-bits. Since this is a stereo DAC, and the R-963T has 8 channels, there are 4 discrete DACs for each pair of channels. Identical DACs means identical performance for all channels from a D/A conversion perspective. To make things interesting, an Analog Devices AD1896 Sample Rate Converter is employed for upconverting to 24bit/192kHz on the front left/right channels. Without going into a protracted argument about the merits of doing so, I will point out that the immediate positive benefit is the employment of a milder output filter at the DAC. No additional data is in the conversion, but gentler output filters can result in better measured performance. Interestingly enough, the AD-1896 is even engaged on the DD and DTS decoder outputs (the DD and DTS decoder output is PCM), so for any digital material, the capability to upconvert to 24bit/192kHz is there. This "remastering" as the R-963T refers to it, is user selectable, and a brief discussion of its effects will occur in the subjective listening portion of this review.
Analog to Digital Conversion is provided by the Asahi-Kasei Microsystems (AKM) AK-5380, which is a 24bit/96kHz A/D stereo converter. Incoming analog signals (if so chosen) will be sampled at 24bit/96kHz for processing as would occur with any other digital source. Signal / Noise Ratio is 105 dB, indicating slightly less than 18 bits accuracy for input - which is fairly good for a consumer grade A/D converter these days. While there are certainly better performing A/D converters, they are also substantially more expensive. Although the R-963T is by no means an entry level product, there are still some places where the law of diminishing returns does not justify further expenditures, and in Sherwood's estimation this is one of them.
This area is in my estimation not correctly implemented in the 963, and here's why.
Bass management is a well known process - it's comparable to the Low-Pass Filter in a two (or more) way speaker for the woofer section. In the case of the 963, it's 80 Hz for speakers defined as small, or is it large? My indecision stems from the flawed implementation. When using a digital input, the LPF (Low Pass Filter) is engaged, regardless of whether the speaker is defined as large or small. So even if your mains have a -3 dB point of 35 Hz, and you've defined them as large, bass between the crossover frequency (80 Hz) and the speakers' -3dB point will be doubled up. Why doubled up? Large speakers do not have the complimentary High-Pass Filter engaged, resulting in (potentially) bloated bass within the crossover region. In discussions with Sherwood, I was told that customers expect their subwoofers to be active at all times, but personally I disagree with this. Hopefully this might be changed in a future release of firmware.
The multi-channel inputs have a bit of unexpected behavior as well. I would not go so far as to say that they have bass management; rather I would call it speaker protection. For any speaker defined as small, a 2nd order (12dB/octave) HPF (High Pass Filter) is engaged, so that inappropriate frequencies are filtered out, preventing damage to a speaker. However, no bass redirection occurs, which means you are losing information below the crossover frequency of 80 Hz.
This is where it gets very tricky. Suppose you're aware of your speaker systems' inability to reproduce the last octave and a half, and have installed an Outlaw Audio ICBM. You have engaged crossover points of 60 Hz for mains and surrounds, 80 Hz for the CC (that's my settings in case you're wondering!). For the CC, you've implemented a double crossover, meaning you will be -6 dB from the CC at 80 Hz, otherwise, things will be fine - it's just a slight loss in the 80-160Hz octave. However, for the mains, you have dropped in a hole from 80 Hz - 60 Hz, which cannot be recovered.
Below are shown the results of the filter in action on a bass decade (11 tones from 200 Hz - 20 Hz, in 1/3 octave intervals), which was recorded at my listening position (values are not corrected for inaccuracies in the Radio Shack SPL meter). Highlighted frequencies indicate room modes, with primaries between 40 Hz and 60 Hz, where a large spike in output is to be expected. [Note: Bass traps are being worked on to minimize the modes.] The in-room results with small speaker correlates well with a 12dB/octave HPF being engaged.
Frequency (Hz) Large Speaker (dB) Small Speaker (dB) 200 70 70 160 71 70 125 72 71 100 79 77 80 81 78 63 74 71 50 70 60 40 72 59 31.5 72 57 25 65 51 20 59 Too low to read
Pull up a chair and listen with me for awhile . . . .
I started off my sessions using the R-963T as a receiver in a System B I had running for a period of time. Components utilized in System B consisted of a Toshiba SD-5700, the Sharp DX-SX1, and the Home Theater Direct Level 3Y speaker system. A tuner was utilized as an additional source. I also had the receiver in System A for use as a preamp/processor. Due to the number of available formats, I am not going to spend the amount of time I would normally on the nuances of each format. For the record, I barely have space for a 5.1 setup due to my penchant for large planar speakers, so I was not able to test 6.1 or 7.1 decoding with this receiver.
The meat and potatoes of any current product is its performance on good old fashioned Redbook CD (20 years later, and now it's good old Redbook?) While I am fond of both high-resolution sources (SACD and DVD-A), the fact is that there remains a dearth of material, with about 600 titles on SACD and 200 titles on DVD-Audio at this writing. So I pulled out my trusted CDs and did some listening - often switching back and forth between source direct (which invokes no processing) and digital remastering which performs sample rate conversion to 24bit/192k. My defined reference for CD was Carmen McRae's "Carmen Sings Monk", track 12, "Suddenly (In Walked Bud) (Live Version)". This is a 4:00 or so of groovalicious hard swinging. In System B, I was hard pressed to find a difference between the direct and remastered. Maybe a little more splash from the cymbals but that was the extent of it. When I listened to the same tracks in System A (I'm aware all other things aren't equal here) I was able to hear the differences clearly between the two - with Digital Remastering on, I experienced several subjective improvements - the biggest was in reproduction of the drum kit. Brushes are used for this selection - there seemed to be a clearer indication that the brush is made up of numerous steel bristles, whereas with source direct the effect was much less clear. In addition, the ride cymbal during the tenor sax and acoustic bass solos rang through with a greater clarity. Additionally, it seemed as though the depth of the soundstage increased, and I was aware of more hall presence from this highly reverberant recording. The acoustic bass had a slightly warmer sound, and the strings had more snap to them. Acoustic piano had a little more body, with a more distinct and clear tone. These are subtle but worthwhile improvements when placed into System A.
Super Audio CD
I had the opportunity to utilize SACD as well with the player, both with the Sharp DX-SX1 in System B, and the Marantz SA12-S1 - I did not have the opportunity to record my thoughts with the Sharp player in System B, as its return meant I only had about 4 days with it in System B. In System A, the Marantz acquitted itself quite well, and if you read my commentary within that review, you will find my thoughts through this product. Suffice it to say that I was highly pleased with the results. The R-963T performed very well indeed with Super Audio CD.
Dolby Digital and DTS
I bet a lot of eyebrows will be raised when they see these two formats grouped together in this review. That's because I used all of the usual suspects for testing DD and DTS decoding. I have yet to hear major differences between DD and DTS decoders, and as such I will only report that the 963 decoders worked nicely, and without any problems (audio dropouts).
Dolby Pro-Logic 2
I remember when I first got DD running in my home with five speakers. I thought it was the coolest thing ever, really! This was just like a cineplex theater, except for the sticky floors, overpriced popcorn, and the cell phones ringing. Dolby Pro Logic-2 (DPL-II) reminds me of that moment. Rather than rehash what DPL-II brings to the table, I will point you instead to our article on the technology.
Although there is the potential for a variety of customizable settings with DPL-II, the 963-T implements the choices in their most basic form. You can choose between Music mode and Movie mode. I found myself utilizing Music mode much more often than Movie mode. While both did an outstanding job of deriving surround from stereo and Dolby Stereo tracks, I preferred the more diffuse front presentation of Music mode to be more to my liking.
I tried DPL-II with TV broadcasts and I was always rewarded with a near Dolby Digital like experience - with a "realistic" surround presentation. I use the term realistic guardedly as one can't really say it's real, when it is a synthesized creation based on algorithmic manipulation.
So what did I hear? Mostly an almost DD-like separation that didn't collapse into the Center Channel as regular DPL so often does. I can't wait until I get some time with baseball broadcasts to see how it does on stereo broadcasts of baseball. DPL would always collapse to center.
Had I not heard DPL-II, I would have thought that DTS Neo:6 was the best new matrix surround process. Neo:6 has Cinema and Music modes, so it shares similarities with DPL-II. Adjustable parameters are optional with Neo:6 as they are with DPL-II. The biggest difference I heard was that Neo:6 produced a discontinuous soundfield most of the time. (As a reminder, both DPL-II and Neo:6 take a two-channel stereo signal and convert it to 5.1 surround, with stereo in the front as well as in the rear. This contrasts with DD and DTS digital 5.1, where each channel is discrete to begin with.)
When engaged, there seemed to be a conscious effort by the algorithm to point out that you have engaged a surround processing matrix extraction algorithm. What does that mean anyway? Simply put, the blend between surrounds and mains was not well executed, and I felt like there was a front to back "stereo" effect, with little to no blend between fronts and surrounds. The auditory effect to me was that I had two soundfields front and back, instead of a continuous soundfield enveloping the listening space.
Perhaps with access to the tunable parameters I could have created a more cohesive presentation, but as it was, I felt that Neo:6 was an improvement over original DPL, but not up to the level of DPL-II.
I have mixed feelings about the R-963T. When talking about the subjective aspects of performance, I was quite pleased with this receiver's performance, regardless of whether it was in System B as a receiver, or System A as a preamp/processor. 24/192K sample rate conversion brought about definite improvements in two-channel reproduction from PCM sources, although I did not hear them in a lower end system in my home. DPL-II is well implemented and is now turning up in many current market products. Its improvements over the first incarnation are well documented. DTS Neo:6 had its moments as well - and you might find its approach preferential to DPL-II, even though I did not.
The amplifier section was able to drive loudspeakers to near ear-drum shattering levels in my System B, and there was only the mildest hint of distortion at these levels. I did not test the amplifier into impedance loads below 3 Ohms.
The parts that I am less than enamored with are outlined above. The user interface needs to be improved to make the experience easier for consumers. Bass management (as presently implemented) is flawed in a way that I consider unacceptable. Until Bass Management implementation is corrected, I would suggest running all speakers as small. With that rule in place, the receiver will reward you with excellent sonics, and a workable (though cumbersome at times) user interface.
- John Kotches -
© Copyright 2002 Secrets of Home Theater & High Fidelity
Return to Table of Contents for this Issue.