A "bit" of information

Besides word clock jitter, which some DACs are capable of almost eliminating (Benchmark DAC-1), what other things can make one transport sound different from another? Aren't they just machines that spit out a stream of ones and zeros? Is it all just cosmetics for more money?

I'm no engineer, but just from a simplistic standpoint:

The transport has to spin a plastic disc around at high rpm and simultaneously track a laser across the radius of the disc while picking up the information contained therein, translate that information to the zeros and ones you mentioned, and then spit it out in a continuous stream. Just the mechanics of accomplishing that task should give some idea of where there is room for making one better than another. To make those mechanisms stable and accurate would seem to be the name of the game. I have no idea what it takes to do that, but certainly some tranports seem to do better than others. Like all the other connections, the link between transport and DAC can also play a roll, again, just speaking from experience and not from knowledge of how it all actually works. I've used a cheap Tohiba DVD player as a transport on my Muse DAC hooked up by S/PDIF. Comparing it to the dedicated Model 5 transport and propietary connector it is not a subtle difference. The Toshiba was not very engaging, it seemed to suck out all the PRAT..the toes stopped tapping. How's that for vomiting forth some Audiophile rhetoric, eh? Seriously, it does make a difference to my ears.

Sure, if it was perfect. But it isn't. We have CDDA and SPDIF to contend with. And our methods of dealing with these appear to be flawed. If it was WAV files read over something like USB or firewire, we wouldn't have a lot of these problems. But we do. And it sucks.

C L E A N U P ...Aisle Five!
The party line from Musical Fidelity (per a dac review in stereophile) is that all transports are the same if your DAC reclocks the bitstream.
My belief is that all digital cables also sound the same if your DAC reclocks (and very likely even if it does not).
Without reclocking there may be merit to Jax2's perspective.
I think audiophiles are sometimes encouraged to lose sight of the fact that bit rate via redbook cd is ridiculously slow compared to modern digital devices, and potential problems inherent to rapid data transfer over long distances are not an issue with a painfully slow cd and a 2 meter digital cable.
The power supply design, common-mode noise, EMI/RF, pulse aberration,quailty of the parts/design of the digital output circut, vibration control and the power cord all can effect on the sound quality?
my marantz 5 disc cd changer sounds VERY different(like crap to be exact) from my dvd player when both used as transports with my Bel Canto DAC2, with Philips $65.00 dvd player being a clear winner. And WireWorld Super Nova 3+ glass toslink sounds way better than the $25 Monster coax when fed into the dac from same dvd player, and I am not the only one to confirm this in my own system, just for those who beleive all digital cables sound the same. go figure.....
Welcome to the real world Neo...
Pondering your question got me thinking about optics, like the ones in cameras. Precision optics render sharp images. Dull optics produce errors, particularly at the edges. While maybe irrelevant to CD transports, the precision concept by itself may not be. The transport conveys electrical or optical representations of the ones and zeros, not the numbers themselves. Because the transport performs a translation of the pits and lands embedded in a CD, how well it achieves this feat becomes important. From the responses you can see that not all transports are created equal.

But more to your point, and processing considerations aside, in any piece of audio gear the quality of the parts used, their stability over time, the quality of assembly and the rejection of outside influences like vibration and radio interference, along with cosmetics, play a role in decision making too.
Koestner, in my opinion, although some dacs claim superb jitter rejection, their ability to deal with certain amount of junk that comes in from the transport is important. I personally don't think that Benchmark DAC-1 or my Bel Canto DAC2 are so jitter immune. They probably are better in this regard than some other dacs, but again it all depends on how much of that junk that comes in they can deal with. Recklocking the signal could be affecting other aspects of the dac's performance, making overall impact on the sound. I think that the cleaner the signal from the transport, the better the dac's overall performance. Also, this is where a good digital cable comes in. I seem to agree with Rockvirgo
here, it's just like a lens on the camera. Better transmission of the signal, that's all. Just speaking from my experience, this is how I picture it.
I am an engineer, with 30 years digital design experience, so I'll be happy to give you the technical explanations:

1) The S/PDIF output from a transport is specified at 75 ohms characteristic impedance. This means that the output impedance of the transport should be 75 ohms and the input impedance of the DAC should be 75 ohms, as well as the connecting cable and connectors. Mismatches in this impedance can cause reflections that cause jitter, and these mismatches are quite common in even the best transports. I have measured and fixed plenty of them.

2) The slew-rate or rise-time of the S/PDIF digital signal has a large effect on the jitter when the signal is detected at the DAC. All transports have quite slow risetimes by design in order to pass FCC emissions testing with the poorest S/PDIF cable attached and mismatched impedance. Therefore, they cause unneccessary jitter. Some Transports will have slower risetimes than others, but unless they are modded, they are all slower than desired.

This white-paper I wrote details the analysis of why the cable length, impedance matching and risetime are all critical to minimizing jitter:

Steve N.
Empirical Audio
Rockvirgo makes an important point. This is why computer audio is intrinsically better than CD playback. The pits are jittery as written on the disk. The Transport also adds jitter when it reads these pits, because each read head has different accuracy of reading the pits. When a pit is actually detected on an inside track and on an outside track probably varies from one transport to the next.

Steve N.
Empirical Audio
Thank you all for your responses. I was trying to convince myself that I didn't need a new transport. I've been told my Sony DVP S900es is not good as a transport.
"This is why computer audio is intrinsically better than CD playback. The pits are jittery as written on the disk. The Transport also adds jitter when it reads these pits, because each read head has different accuracy of reading the pits."

And the cheap CD drives in PCs don't add jitter when they
read the data on to the HD, and the noisy PC PSU doesn't effect the output signal?
"And the cheap CD drives in PCs don't add jitter when they
read the data on to the HD, and the noisy PC PSU doesn't effect the output signal?"

No. I'm talking about ripping the CD and playing it back from RAM on the computer. This is computer-driven audio. The CDROM drive is not involved in the payback. As for the noisy power supply and noisy ground, there are converters that go from USB to S/PDIF that have their own power supply or even a battery supply and they even isolate the computer and audio system grounds, so no noise of any kind gets injected from the PC. The USB interface does not affect the sound quality, no matter what the length.

This truly is the highest quality method for playback of CD images.

Steve N.
Empirical Audio