AMR's research and development team's prior experience with the Philips TDA1541A in other compact disc players as well as designs of their own told them that the full performance of this chipset had yet to be extracted.

Some manufacturers advertise the use of Philips Double Crown TDA1541A chips, which were selected at the time of manufacturing for specific performance parameters by Philips. These can sound slightly better than the standard TDA1541A's when used in conventional circuits. However, Philips Double Crown TDA1541A chips are not longer widely available and the differences have often been overplayed.

AMR went to the drawing board and analysed the Philips TDA1541A at the silicon-die level in extreme detail. AMR discovered a number of areas where correctly designed external circuitry would significantly bridge the small performance gap between the Double Crown and standard TDA1541A chips.

From the power supply to the digital input signal conditioning, the CD-77 circuit has been implemented to a degree of optimisation extending beyond anything we are aware of. The implementation is far beyond the simple manufacturer's datasheet application schematic commonly found in CD-Players and DACs employing the TDA1541A. The resulting core technologies are the OptiSignal® conditioning circuit and OptiClockLock®.

Under testing, these circuits considerably improved the performance of standard TDA1541As to a level that was largely en par with Double Crown chips in measured and subjective performance evaluation. (Note AMR used verified genuine Double Crown chips for these tests, as the market has recently seen many standard TDA1541A which were surreptitiously remarked as Double Crown to increase their sales value.) We at AMR believe that the result of all this surpasses most if not all other TDA1541A designs, even those fitted with a genuine Double Crown chip.(please refer to the Advanced Design Features section for more details).

"SuperClock" is simply a marketing term. The CD-77 uses a military-grade crystal oscillator equal or superior in specification to any of the industrial-grade so-called "Super Clocks" offered as an aftermarket modification. AMR has combined this master clock with a dedicated power supply whose noise performance is at the limits of what is physically possible.

Going beyond even this, all other clocks within the CD-77 are synchronised to the superior precision timing signal of the master clock module to avoid increased jitter due to multiple oscillators creating "beat frequencies."

CD clock crystals are often promoted using a measure that specifies the difference of the actual frequency from the nominal frequency, usually specified as the deviation in ppm (parts per million). It should be noted that this specification has no bearing whatsoever on the jitter in CD players as it is a purely static measurement.

Phase noise in the clock is the primary cause of jitter. No reliable and standardised protocol exists to measure and specify jitter in a form that makes measurements comparable. AMR as a result, has chosen not to publish a specific number that would not be reliably comparable to the numbers published by others. Rest assured that the CD-77 offers one of the lowest levels of phase noise of any CD player produced.

In recent years, a technology known as "Upsampling" as well as one called "Zero-Oversampling®" or " Non-Oversampling" have gained popularity with CD player and DAC manufacturers. Having compared these approaches to traditional "Oversampling" we at AMR feel that in most cases, the omission of Oversampling or Upsampling and the use of a simple analogue circuit, is the most musically satisfying.

However, in a significant minority of recordings, we found that either traditional Oversampling or Upsampling was preferable. Therefore, we have implemented all of them and made them user-selectable (please refer to OptiSample® in the Advanced Design Features section for more details).

close

Philips Electronics ceased production of its own 'industrial standard' compact disc transports in 2001. Presently, new so-called Philips CD transports are far-eastern clones constructed to much lesser standards and are of questionable reliability and are the only remaining off the self dedicated CD transports.

Furthermore the exceptional Victor (JVC) transport the XL-Z900 was based upon has been discontinued and several other suppliers of high quality transport solutions have cancelled their CD lines in favour of Multiformat DVD designs.

At AMR we found that if CD replay is the desired application neither computer CD Drives nor the current "universal" DVD based solutions perform equal to the highest quality dedicated transport mechanisms. AMR therefore set out to create a proprietary CD transport system that would provide the same or better performance than the legendary transport solutions of yesteryear, be they the original Philips swing arm type transport or Victor's XL-Z900.

AMR combined genuine parts from a number of major manufacturers and mounted these to our machined platform. This platform is then suspended on dampeners to isolate the Transport further from any vibrations. The resultant OptiDrive® transport maximises the accuracy of the data retrieval. (Please refer to OptiDrive® in the Advanced Design Features section for more details).

Yes, the CD-77 has a 16-frame buffer (or 512K Bit of Memory). That said, nearly all CD players have a memory buffer of one form or other. This buffer is used among other things to allow for error correction if the data is read incorrectly.

The error correction, IF implemented correctly recovers the original data through the use of redundant data written to the CD in almost all cases. This process does NOT "interpolate" but actually recovers the exact data originally written on the CD.

The necessary calculation may however take a short time. Without a large enough buffer, the timing of the data would be effected negatively. Only in case of gross problems does the error correction fail and then ALL CD-Players, even computer-based ones or software such as Exact Audio Copy, will first attempt to interpolate samples that cannot be read correctly (in many cases the software will first attempt to re-read the samples several times) from samples before and after the affected area.

The data transmission from the drive to the DAC is timed by the clock supplied to the CD-Transport or generated in the transport itself. The spindle motor servo keeps the buffer normally 1/2 full and therefore is linked to that master clock indirectly.

If the master clock driving the transport is accurate and the buffer large enough (8 frames or more) the output from the memory buffer is "jitter free". Some chipsets designed for portable CD Players have a larger amount of memory, to provide "jog proofing". This means that even if the player is violently shaken and hence mistracks, several seconds of music are in the buffer and the player normally can re-acquire tracking before the buffer runs out.

Other than this (which is of no relevance to a CD Player used at home on a solid surface) there is no further advantage from excessively large memory buffers and such buffers usually slow down operation, by delaying the time from pressing play to hearing music by the time needed to fill the buffer, track skipping etc is equally delayed.

The situation changes dramatically if we use a separate Transport and DAC linked via S/P-DIF or AES/EBU cables. The problem when using a DAC with an S/P-DIF or AES/EBU input, which combine the data and clock is that the DAC itself must recover the clock from the datastream, which is not a very accurate process and made more problematic by the specific way the clock is modulated to carry the data.

To address this significant problem, additional memory buffers have been occasionally used in the highest "high-end" DACs, both pro-audio and consumer audio in order to allow a local clock which is NOT PLL synchronised (PLL = Phase Locked Loop - a process analogous to Negative Feedback, except in the time domain).

The clocks between CD Transport and DAC may now differ to a small but significant degree. The CD Specification allows +/- 1000ppm or 0.1% deviation, meaning that up to 1 in thousand samples would be "too much" or "missing" per second, so the memory buffer must hold many more frames than the one in the CD player.

In fact, it needs to be able hold around 44 frames per second playtime of CD. Thus the minimum memory buffer needs to handle around 210,000 32-Bit frames of under or overrun and needs at least 2MB in size to accommodate this.

Further, the buffer would need filling with 210,000 frames and would introduce around 5 second delays between data being read and received on the audio output. As no "off the shelf" solution exists to provide such a memory buffer, it would need to be designed from ground-up.

We hope this brief explanation has shed some light as to why DACs very rarely have memory buffers (no simple and readily available solutions) and CD-Players always have them (< 1KB required to work properly and built into most commonly available chipsets).

The key to high sound quality is not in the size of the memory buffer, but first in the implementation of the mechanical system to maximise the correct reading of data followed by the error-correction routines, which requires extended routines beyond the original "red book" specification to correctly recover the original data.

This correct and original data then needs to clock into the DAC with a precise, jitter-free clock, so following the mechanical implementation of the drive, equally important is the master clock for the CD player.

In summary, yes the CD-77 as a one-box processor, has a more than ample 16-frame memory buffer but so does almost every other compact disc player. However, DACs need much larger memory buffers but due to the lack of a readily available solution, very few actually have a buffer.

The only methods to transmit data between transport and DAC that are standardised are S/P-DIF and AES/EBU. As these combine clock and data signals, they produce higher levels of jitter than found in well-engineered single unit players.

While AMR could have implemented a non-standard, proprietary connection, such as an implementation of the I2S bus, none are suited to long transmission distances and no common standards exist.

By keeping all digital circuits within a very compact 4 inch square area, we negated the introduction of jitter arising from long connections. We kept the analogue stages very close to the digital to analogue converter, to avoid introducing noise and distortion picked up from the outside. We also include the power supply in the same case to ensure that clean power is generated where it is consumed rather than being sent over long cables as with a two-box setup, which would again, pickup noise and distortion.

close

The Digital to Analogue Converter section selected by AMR together with the thermionic electron valve-based analogue stages provide a performance far superior to that of most stand-alone DACs.

During the original conception of the CD-77, there was no standard to allow for the connection of a transport to a DAC without introducing substantial levels of jitter.

However, in the last two years, the technology for transports and DACs has matured. We have therefore developed a "jitterless" solution which we first implemented in the CD-777. We have now applied this solution to the CD-77 and from the beginning of 2010, there is an option for the CD-77 to ship with S-P/DIF input/output via RCA. In effect, this means the CD-77 can also serve as a dedicated transport or DAC via S-P/DIF as well.

close

In recent times, although many music formats (including WAV, AAC, WMA, MP3 etc.) have become computer-based, the sound system of most computers however, leaves much to be desired.

The USB input of the CD-77 allows its superior digital-to-analogue conversion and analogue stage to be used in the replay of computer-based music formats to the same level of quality provided by the CD-77 for compact discs. (please refer to OptiBus® in the Advanced Design Features section for more details)

CD is an optical format. The presence of varying external light conditions such as that admitted by a transparent lid was found to be detrimental to the sonic performance.

No. The CD-77 uses a completely single-ended signal path as we have found this to give a more realistic reproduction of music. However, in terms of sound quality, we found the XLR connector preferable to the RCA connector, even when used with un-balanced signals.

Consequently, we have offered the XLR output in a manner that is fully compatible with balanced equipment: takes full advantage of balanced connections yet retains the desirable single-ended signal circuit.