Yamaha DSP-3000 (1988)
Twenty-one pounds of black box that promised to replace your listening room with a concert hall — for $1,899, roughly twice what the DSP-1 cost.
Overview
The DSP-3000 was Yamaha's second-generation digital sound field processor, and it arrived with a price tag that startled people who had watched the DSP-1 debut in June 1987. That first unit was already considered sophisticated; the assumption, as Audio (1988, p.85) put it, was that the follow-up would be simpler and cheaper. It wasn't. The DSP-3000 was more sophisticated in several respects and roughly twice the price — $1,899 at introduction. It carried 35 preset processing programs and 20 user-set positions, a jump from the DSP-1's repertoire that added 17 new environments, including concert halls sampled in several countries, two new presence modes, a new surround program, and four movie-theater simulations labeled "Adventure," "Classic," "Musical," and "Standard."
The architecture underneath was serious. Yamaha used three of its own YM-3818 VLSI chips — each containing a high-speed multiplier plus an adder and subtractor — to calculate dozens of discrete early reflections in real time. The company claimed the unit could produce up to 88 discrete reflections, 22 per each of the four effects channels. A proprietary Hi-bit, floating 18-bit scheme combined with a dual-converter configuration to improve signal-to-noise ratio and dynamic range, and digital processing ran four-times oversampling filters for time-base resolution, phase coherence, and transient response. The main channels used 18-bit D/A conversion; the effects channels ran 16-bit. A/D conversion was 16-bit linear quantization at 48 kHz with internal dither, and the unit auto-selected among 32, 44.1, and 48 kHz sampling frequencies.
The connectivity was unusually complete for the period. Two analog inputs and a direct digital input let the DSP-3000 process CD or DAT signals without an extra A/D stage — eliminating a conversion that the DSP-1 required. There were tape recorder in/out jacks with a monitor switch, a video input/output loop to superimpose program parameters on a TV monitor, and 24 gold-plated phono jacks on the back panel. A switched AC outlet handled up to 300 watts, which Audio (1988, p.85) noted was "quite high, and much better than on many other units." Power consumption was 45 watts.
The measurements were strong. Audio (1988, p.85) measured harmonic distortion at 0.002% on the main outputs with analog input, 0.005% on effects outputs, and 0.003%/0.005% with digital input. Frequency response for the main channels ran 10 Hz to 100 kHz with analog input; with digital input, both main and effects held 20 Hz to 20 kHz within ±0.5 dB. S/N ratio was 110 dBA on main outputs with analog input, dropping to 94 dBA on effects; with digital input, main stayed at 110 dBA and effects improved to 105 dBA. Channel separation measured 80 dB at 1 kHz with analog input and 90 dB with digital. The reviewer's own bench work found the main channels down just 0.05 dB at 20 Hz and 0.4 dB at 20 kHz, with -3 dB points at 1.7 Hz and roughly 80 kHz.
The volume control tracking was exceptional. Audio (1988, p.85) reported the two main channels tracked within ±0.1 dB over the range from 0 down to at least 65 dB of attenuation — described as the best the reviewer had ever seen. Front-channel volumes tracked within 1 dB across the whole range; rear channels within 1 dB for about 50 dB; effects channels tracked main within 1 dB for about 45 dB. The output impedance measured 966 ohms, and channel separation at 1 kHz came in at 79 dB.
The remote was the primary interface — 43 buttons against just nine on the faceplate. Audio (1988, p.85) found the remote worked reliably from more than 27 feet on-axis, and at 10 to 15 feet it could be off-axis up to 80° horizontally and at least 30° vertically. The LCD display was legible at 15 feet and up to 45° off-axis. The "Bit Monitor" display showed bit depth in real time: 13 bits at 0.146 V input, 14 at 0.295 V, 15 at 0.594 V, 16 at 1.214 V, and "Full" at 2.440 V. Clipping appeared at 7.34 V on the main output. The reviewer noted the bit display was a genuine aid — being able to set levels exactly to the point of full 16-bit processing was part of why the DSP-3000 sounded quieter than the DSP-1 without ever being driven into distortion.
The flaws were real and worth stating plainly. The DSP-3000 dropped the main mono center and subwoofer outputs that the DSP-1 had carried. Four "Output/Mono" jacks on the back panel ("Front," "Right," "Left," "Rear") handled low-frequency reinforcement with individual level controls and low-pass filters switchable between 80 Hz, 150 Hz, and 5 kHz, but the reviewer found the front output level too low to drive a center speaker convincingly, and the mono output levels generally ran about 7 dB below the summed channels — potentially too low for some amplifier/subwoofer combinations. The workaround was to Y-connect the main outputs to a stereo amp in mono mode, which the reviewer did with good results, though Yamaha's manual never clearly explained what the four mono signals actually consisted of. The DSP-3000 also omitted the DSP-1's Sound Effector programs, a loss the reviewer considered negligible for music and irrelevant for movies.
The manual ran 64 pages and was generally clear, though the reviewer wished for more detail on room acoustics, speaker placement, and program parameter effects. The internal pink-noise generator — technically optimized for speaker balancing rather than true flat pink — was a genuine convenience, and the effects level and balance displays each used 10 vertical bars per 10%, with 1% step adjustments available. The transformer ran warm to the touch after hours of use but was heavily encased and shielded, and the chassis stayed rigid with the top covers off.
For listening, the Dolby Surround/20 mode proved a strong choice for Dolby-encoded movies, with excellent dialog centering and voices that stayed embodied. The reviewer's summary judgment was that the DSP-3000 was expensive but the premier means of enhancing the listening experience at the time — with the caveat that buyers would need to budget for the effect-channel amplification as well. Collapsing the sound field to plain stereo, the reviewer noted, was simply not pleasurable once you'd heard what the processor could do.
eBay Listings
As an eBay Partner, we earn from qualifying purchases. This helps support our independent vintage technology research.