Is this a reasonable Audiophile Mac-based Server?

The connection of using an Airport express to a DAC is not ideal IMO. A USB based DAC such as the Benchmark DAC1USB and Wavelength DACs give much better results.

I prefer having a Mac Mini connected to a LCD display, wireless keyboard, with quiet 2.5'' based HD directly connected in the same room. Simpler and works better than my experiments with the Airport Express.

You won't believe how much more convenient and better the sound is of the Benchmark, fed by the Mac Mini via USB rather than using a standard transport and SPDIF or Toslink interface.

Good luck!

I called Benchmark's technical support and spoke to them at length about your assertion that USB is superior to Toslink coming out of the Airport Express. They assured me repeatedly that Toslink would be "identical" to USB.

I know Benchmark says that it sounds identical. However, I can tell the difference between Airport Express (AE), Slimdevices transporter fed via toslink to the DAC1, or directly to USB. Direct via USB it better with a tighter sound, less glare in the highs, and more low lever resolution.

That allows me to save several hundred dollars by skipping the USB.

I would go with the USB version of the DAC1 in any case:
- More flexibility down the road as you can try USB direct.The excellent USB implementation in the DAC is one of the main reason to go with the DAC1USB in first place.
- Better driver stage in the USB version which is supposed to sound better than the last incarnation of the DAC1.
- You save yourself the $100 for the AE.

As for the interface, it's a matter of personal taste: Frontrow on a larger LCD screen, clicker software with a PDA, Laptop in you lap. Try them out if you can before settling on one.

Henry, looking at your plan B I still don't think that Toslink will give you the same performance as a USB connection. I had several people over today and we performed some more comparisons in my system:

1) Slimdevices Transporter direct using the internal DAC in the Transporter
2) Transporter feeding Benchmark DAC1 via toslink
3) Mac Mini feeding Benchmark DAC1 via USB

Out of the above 1) and 2) were very similar, despite the different DACs.
Switching to USB made the biggest difference, i.e. 3) was clearly better, less hf noise, tighter more extended bass.

As for remote contrallability: The Macbook as well as the Mac Mini come with a Infrared remote to control iTunes or Frontrow. Alternatively take a look at Salling Clicker, which allows you to use a PDA or cellphone as remote (works great too).

Again, just my own experience, but have been playing with several of the combinations you suggested in my own system.

Herman, the experiment you describe was the first one I did when I got the DAC1 a while ago (I started out with a Airport Express and a Macbook with optical out)...

A better comparison would be

1. Mac Mini feeding Benchmark DAC1 via USB
2. Mac Mini feeding Benchmark DAC1 via toslink

even though this isn't conclusive either since despite what Benchmark says the DAC may indeed work better with USB. This isn't saying USB beats toslink every time, only that the Benchmark may be better with USB.

The differences between 1) and 2) were similar to the more recent experiment but even more pronounced. Similarly for the Airport Express which I also have. In fact, I was expecting the transport to do much better since it is isolated from the PC, but I guess that is not necessarily the case.

Certainly the observations are true for the Benchmark, which is the DAC Henry was considering. In fact, your experiments with the Altman are more of a singular experience, since Altman put great care in the toslink interface ;)

Also just to ad some information to the above experiments: I used a Glass toslink with a mini to regular toslink adaptor which works better than the Monster plastic optical fiber cable (as determined in previous trials.

Restock, maybe I missed it but I don't see where you used anything but a jittery optical out compared to the USB.

I agree it looks like there is an excess amount of jitter in the toslink output of the Airport Express, Mac Mini and Transporter (at least compared to USB). But that is why I recommend to Henry not to got the Airport Express or Macbook to toslink route. By the way, I don't see why your source (Airport Express) has any less Jitter, just that you have a very good anti-jitter circuit in the Altman Jisco.

In any case, I would recommend (to Henry) to get the USB version of Benchmark due its upgraded circuitry and added USB output compared to the Standard DAC1, try the wireless setup with AE and compare it to the USB version. Best to make up your own mind about this and trust your own ears. The differences are subtle to say the least., but since the goal is to assemble the"best sounding Music Server regardless of price" it would be best to keep options open.

As far as the wireless 802 network vs. ethernet work debate goes, dropouts are going to be unlikely with an all Mac setup. you only might run in trouble down the road for playing back high rez 24/96 files. There I really don't see that much of a difference soundwise for 16/44.1 file playback.

S789f, no dropouts ever with my Mac Mini. What length USB cable are you using? USB may have problems if longer than 3m or 10ft.

I'm considering a Mac based server w/ outboard DAC, and just read in the current issue of Stereophile; Manufacturer's Comment from Benchmark Media Systems concerning "iTunes 7.5 bug" that sounds suspiciously like S789f's dropout problem....can anybody help with experience on the iTunes 7.5 (Mac updates automatically) so a heads up is worthwhile.

If you follow Benchmark's suggestions for setting the Audio Midi sampling frequency before opening itunes (check Benchmark website), then there should not be a problem. Also, Gordon Rankin of Wavelenth has been checking this by doing detailed measurements and could not verify a serious problem with itunes.

S789f, I guess there might be a problem with the Bel Canto then - do you know anyone in the area with other USB DACs that you could try out. Would a Benchmark have similar problems?