Toshiba SD-9000 (1999)
A fourth-generation DVD player that Toshiba billed as the pinnacle of the format - and that the bench tests greeted with a shrug.
Overview
Toshiba, the self-proclaimed prime developer of the DVD format, launched the SD-9000 as a fourth-generation player and the first DVD deck admitted to its exclusive Cinema Series. The marketing went into overdrive. Toshiba claimed the machine would redefine "the state-of-the-art in DVD-Video performance, quality and functionality" and deliver "the highest level of DVD video playback quality attainable." The price was $1,100, a serious sum in 1999 dollars for a device that, on paper at least, promised the moon.
The build suggests Toshiba meant it. The player weighs 15 pounds (6.8 kg) and measures 3¼ inches high, 17 inches wide, and 12¼ inches deep. Those dimensions reflect the company's stated effort to "dramatically reduce mechanically induced vibration and resonances." It has a resin-impregnated base, a shock-damped cover with a plate laminated inside over the transport mechanism, and a machined aluminum front panel. Owners report the heft is immediately apparent when lifting the unit. Audio (September 1999, p.54) confirmed the deck is heavier than many competitors, and attributed the mass to exactly those construction choices.
Toshiba equipped the SD-9000 with an advanced anti-aliasing filter and a 10-bit, 27-MHz video DAC said to provide resolution in excess of 500 lines. Audio (September 1999, p.54) found the player showed no sign of quitting at 440 lines, about as high as the wedge patterns on standard test discs go. The reviewer noted that no test disc existed to prove the 500-line claim, but the deck's resolution was clearly strong. On a projection monitor, picture quality matched the bench results. Even the tightest wedges of Snell and Wilcox test patterns were clearly delineated, and color rendition was pleasing. Some minor instability in midrange gray scale appeared on the bench but proved invisible in normal viewing.
The audio section is where the SD-9000 gets complicated. Toshiba fitted 96-kHz/24-bit "multi-port" DACs in what the company described as "a parallel configuration." The company claimed the converters filter the left and right audio signals three times to achieve ultra-low distortion, improved signal-to-noise ratio, and dynamic range greater than 112 dB. Audio (September 1999, p.58) tested those claims and found the filters less effective than those in many other upscale players. Frequency response stayed within ±0.07 dB across the audio band, but visible ripple appeared in the response curve. Distortion at 20 kHz in stereo hit 0.249%. That comes from a 24.1-kHz beat between the 44.1-kHz sampling carrier and the 20-kHz signal, a symptom of inadequate digital filtering. Below 20 kHz, converter distortion barely exceeded 0.03%, so the problem was confined to the top octave. The reviewer had hoped for better numbers given the price.
Total harmonic distortion plus noise was also mixed. Below -30 dBFS, noise and distortion were excellent at less than -96 dBFS. Distortion rose steeply at higher signal levels, reaching -78.6 dBFS at 0 dBFS, which Audio (September 1999, p.58) called nothing to write home about in a top-of-the-line player. The reviewer also noted that the 112-dB dynamic range claim is theoretically impossible with a 16-bit stimulus, suggesting Toshiba's figure was based on a 24-bit source. The output levels of the five channels were well matched within 0.04 dB, and separation measured 86 dB in stereo and 72 to 99 dB in Dolby Digital. The headphone amp clipped at a rather low level, though it was probably adequate for most consumer headsets.
For video processing, the SD-9000 has two-level 3D-DNR digital video noise reduction, which Toshiba claimed would reduce random video noise, particularly in background images. The system stores picture frames in a digital buffer and correlates the present frame with the previous one to identify and average out noise. Audio (September 1999, p.54) could find no evidence that the 3D-DNR did anything to prevent or obscure blocking in reduced-bit-rate test material. If anything, engaging it softened vertical detail, blurring ripples on moving water for instance, without reducing blocking on low-bit-rate DVDs. The reviewer also noted that blocking at any given bit rate seemed more apparent on the SD-9000 than on a reference Sony DVP-S7700.
The player passes DTS, Dolby Digital, and MPEG-2 audio bit streams via optical and coax connectors for downstream decoding. It also has a full 5.1-channel Dolby Digital decoder built in. Toshiba used Fujifilm's Vaddis II chip, a Zoran design licensed by Fuji, for this. The internal decoder is more accommodating than those of some other players. Bass-management options permit setup with or without a subwoofer and with "Large" or "Small" center and surround speakers, and users can choose to omit center or surround speakers entirely, with signals routed accordingly. Center and surround delay are adjustable relative to the main fronts via on-screen menus that Audio (September 1999, p.57) found clear and reasonably intuitive. There are limits. There is no option for "Small" main front speakers, no means to accommodate center and surround speakers of different sizes, and no choice of bass-management crossover frequency. The crossover point proved higher than usual at about 125 Hz (-3 dB), with gradual filter slopes of 6 dB/octave on the high pass and somewhere between 6 and 12 dB/octave on the low pass.
The SD-9000 also provides a two-channel mixdown feed for stereo TVs and for A/V amps and receivers that can manage Dolby Pro Logic but not Dolby Digital decoding. A headphone jack and volume control on the front allow private listening. For stereo enhancement, the player offers two Spatializer N-2-2 modes, one for loudspeakers and one for headphones. Audio (September 1999, p.57) described the virtual surround as better than nothing but no substitute for the real thing. All connectors save the optical socket and the S-video jack are plated in gold. The player has color-difference (component-video) outputs, which Toshiba branded ColorStream, as well as one S-video and one composite-video connection. The ColorStream outputs are fully compatible with any monitor that provides color-difference inputs, despite the proprietary name.
The remote control is long and rather heavy but relatively easy to grasp. It handles TVs and cable boxes from an extensive list of manufacturers, though it is not universally programmable. A slider on the side chooses between DVD, TV, and cable. The most frequently used keys are always available and illuminate at the touch of the "Light" button. Those are the transport controls, zoom, angle, subtitle, and the navigation arrows. The backlighting is none too bright, helpful in a dark room but invisible in a well-lit one. Audio (September 1999, p.57) found the deck required direct aiming when room lights were up, and a more powerful remote would have been appreciated. At the base of the remote, a sliding door reveals a numeric keypad and less frequently used controls, including a "3D DNR" button. Between the main bank and the keypad sits a jog/shuttle combination for slow and accelerated forward and reverse motion. The reviewer found it a pain to use. The deck seemed to take its own sweet time responding to commands, which encouraged over-twisting that shot past the intended point. Frame-advance and slow motion were admirably smooth forward but quite jumpy in reverse.
The SD-9000's video performance failed to live up to the hype about the 10-bit, 27-MHz video DACs, though Audio (September 1999, p.58) noted that the performance wasn't bad and there were no obvious problems on screen. The reviewer's complaint was that the player was measurably off the mark on luminance, chromaburst, and sync levels. That is a bothersome finding in a pricey, top-of-the-line model when so many other players were dead accurate. The SD-9000 also exhibited measurable chroma differential gain and chromaluma time displacement, which other players did not. Luminance channel response was admirable, as was chroma phase accuracy, and the deck responded well to rapid shifts in luminance level.
Audio performance on CDs, used rather than DVDs to ensure straight 16-bit PCM, was decent but by no means as good as it could and should have been. Audio (September 1999, p.59) found James Levine's piano a bit strident and Dawn Upshaw's voice rather shrieky on a Sony SBM recording, with the image and ambience distant. A Philips Bitstream recording of Mitsuko Uchida playing Schubert sonatas sounded hard, with confusing ambience and unnatural decay. The Rodrigo Concierto de Aranjuez on EMI Classics was nearly unrecognizable. The guitar was dull and characterless, the violins shrieky and edgy, and the orchestral transients tizzy. The reviewer's conclusion was blunt: "Not my first choice for a top-of-the-line DVD player."
The SD-9000's reception in the pages of Audio was not an outlier. A reader's letter published in the November 1999 issue (p.10) praised the magazine for not shying away from criticizing a higher-priced product with no advantages, and cited the Toshiba SD-9000 by name as the example. The player earned an "A" rating in that review, but the accompanying text made clear the grade came with reservations about value at the $1,100 asking price.
The SD-9000 is a substantial, well-built machine with a machined aluminum faceplate and genuine 5.1 decoding on board. That makes it interesting as a piece of late-1990s DVD hardware. It is also a first-generation-adjacent DVD player from an era when the format was evolving rapidly, and even modest later players have surpassed its video and audio performance. The 3D-DNR system, which was supposed to be a selling point, proved ineffective in independent testing. The remote's sluggish response and the jog/shuttle's imprecision are the kinds of quirks that owners either learn to live with or find endlessly irritating. Parts and service documentation are scarce, and the machine's complexity means repairs are not for the faint of heart.
eBay Listings
As an eBay Partner, we earn from qualifying purchases. This helps support our independent vintage technology research.