What is purpose of a cd transport?

CD data is encoded using Reed-Solomon error correction. The job of the transport is to read the bits (1's and 0's pits/no pits) off the disk including the additional data for error correction and place it all into a buffer for software error correction and for eventual feed to a DAC.

After error correction the result is two channels of 65536 bit word data streams at 44.1 KHz or a data stream rate of 1411200 bits/sec.

There is not much rocket science to this and, as it is all digital, the data will either be correct or incorrect. Incorrect words will cause skipping of the sound as their remains a gap or blank in the data stream.

The relative difference in transports therefore will be related to how easily it skips and how robustly it reads all the words from say a scratched or damaged disk.

The output bitstream from one transport to another should therefore be the same provided there are no skips.

It is downstream of the transport itself where differences can occur. The DAC conversion requires an accurate clock speed to clock out the buffer and very slight (often inaudible) differences can occur due to different DAC designs/specifications/quality, such as immunity to clock jitter. Things like filters, over sampling, one bit sampling, and multi-bit sampling are all methods to achieve the most accurate conversion and all have advantages/disadvantages and can cause slight differences in the ANALOG output. These differences occur in the DAC after the buffer data from the transport.

The quality of the transport itself should not normally make a difference in the quality of the output provided it is capable of properly reading the bits from the CD. It would require a very bad transport design or a very badly damaged disk in order for the system to fail to get all the correct words off a CD disk under normal conditions.

Yes all CD player/transports can read the data, but the problem is timing and distortion. Imagine what happens to the sound if the speed of the data flow is off and/or it fluxuates because of vibration, irregular speed of the platter, poor cabling, and/or fluxuations from an inadequate power supply (and a little of "all of the above" at the same time).

The data correction in the DAC can only do so much, and of course the quality of the DAC affects this also.

Speed of data flow is related to the accuracy of the clock on the data buffer being clocked out to the DAC....so it is not related to the transport itself as in a conventional turntable...as the buffer makes a disconnect in the flow of bits. Think of the buffer as a hot water storage tank in your home - the boiler/heaters job is to keep it with enough hot water for your needs and then the speed of flow is controlled by your faucets. The transport job is to read the bits accurately from the CD and at speed that ensures the buffer does not empty.

Thanks for the info Shardorne. I've often had trouble finding something that makes logical sense in the audioworld... ;-) I've heard the argument that a transport reads a 1 or a 0 and the distance between them make the note (or is subject to error) - like an analogue record player. This is a digital medium, not an etched vinyl record. Wow and flutter shouldn't affect the digital signal to the DAC when it has correction data, right? It would be great to be located somewhere with enough equipment that I could just sit down and blind compare a discman (assuming it had digital out) and a krell transport to the same DAC.

Crudo20,

You are correct - transport wow and flutter do not matter. It can run at any speed as long as it is fast enough to ensure the digital buffer does not run out. Just as you will not run out of hot water if your boiler/heater tops up the hot water quickly enough.

As for a discman digital out....yes it should be as good as the best...after all it is digital and that is the beauty of digital...digital preserves the original data perfectly.

A caveat, this is provided the DAC conversion is using its own digital ref clock to clock out the buffered bits and not slave to the clock signal from the discman....the discman clock might be less robust than a high quality DAC. A variable clock signal gives jitter and therefore distortion in some rare cases (badly designed equipment)

See this link for a good overview

http://www.tc.umn.edu/~erick205/Papers/paper.html#jitter