A few hours with a Unitiserve, Snait, nDAC and my PC

Posted by: Occean on 28 September 2010

Currently I have a Supernait which is fed by my PC (Win 7 64bit/3.0ghz/ssd/nas/foobar/ESI Juli@) into my Spendor S5e’s + REL Stampede. I know the obvious upgrade for this is a Hicap2, but as it happens I currently have £2k ish I can spend. I really like the idea of the Unitiserve – yes my PC works just fine but a separate solution seems like a good idea to me. Also with all the hype of the nDAC and it being £2k, I thought I had better have a listen too….

So I organised a demo at O&L in Clapham, called silenttim from the forums to join me.

Unitiserve + Snait
Unitiserve + DAC + Snait
My PC + SNait
My PC + DAC + Snait

Speakers were PMC GB1 which I have heard many times before.

So I brought a couple CD’s (Martina Topley Bird, Antonio, Audio Bullies, Waldeck) ripped them with the Unitiserve, nice – must say that’s easy! So first up was the ‘serve, snait and with a DC1. Oh my it sounded so different to my home setup, the top end was really edgy in a negative way, I guess I was not as familiar with the PMC’s as I thought. Listened to a few tracks, all had the edgy topend, overall I wasn’t happy – I check with the guys that everything was warm and run in, they confirmed.

So we dropped in the DAC – everything started to sound lovely. The top end was smooth again, more detail and all the good things you expect but it wasn’t a million miles from my home setup. Not sure about the PMC’s at all. Repeated all the tracks

Next step was to setup my PC, optically into the Snait with the Serve connected via a DC1. The difference was astounding, we could both hear quite a big difference…my PC was better! The ‘serve had the same edgy sound to it as before, but my PC had all the detail and resolution (and reading FLAC’s) but the topend sounded so much better to both of us, even a quick blind test, everytime we could tell the difference. I was shocked and I must say a little disappointed, I loved the idea of the serve. But my confidence was restored in the PMC’s.

So finally – PC to DAC to Snait as well as the ‘serve – the DAC made a nice improvement to the PC, soundstage was widen and everything was just a little sharper but I was not convinced at £2k better…I know a Hicap’s improvement is a lot more than this.

So what did I decide, well I am booking the DAC for a home demo. The serve, well if it sounded as good as my pc (I was expecting that at the least) I would have ordered there and then, it didn’t – this has made me realise that my PC does a damn good job, it has always outperformed my Mac and sonos. So I may build the *perfect* PC instead.

As for the DAC, it was good, but not £2k better than the Snait bare, I need to thing long and hard about my upgrade path, if I decide to go 282/250 the DAC is a must, but I am not sure I want to go that far and unbalance my AV system so much (it sounds friggin awesome right now).
Posted on: 01 October 2010 by Occean
quote:


IMHO, if you start out with such an upscale case, then you may run the risk of pushing your total build cost pretty close to manufacturer-delivered turn-key solutions.



I believe better results can be achieved by building your own, a turn key solution being a compromise. That's why many of us *build* our own Hifi rather than an off the shelf setup (B&O...)

yes that is expensive...I wonder how much a naim case would cost in comparasion?
Posted on: 01 October 2010 by Hook
quote:
Originally posted by Occean:
Personally I think the CAPS build on the CA site gives away a little too much for ability to be portable and it also a little out of date now in terms of hardwear.

A single core atom was chosen as the duals were not passive at the time, now the next gen atom chips are in an the duals are even faster, more ram, less power consuming and passive.

The psu is a smps unit, a liner unit should be superior. Though not portable.

The case is cramped and the pci card is not actually mounted so the outputs are available on the outside, rather cables need to be routed though gaps. Something like atechs passive case would be much more suitable as a home unit...


Hi Occean -

You are 100% correct. It is amazing how quickly the PC marketplace moves forward. Those were likely the best choices last February, but there are better choices today.

I will say that the CAPS build does work well for what it is. The dual core Atoms available 8 months ago ran too hot for the smaller cases. In my experience, the single core Atom has plenty of horsepower to run a dedicated music server. But if you are building a system that will perform other functions as well, then dual core and a bigger case may be necessary.

Good luck on your build! Looking forward to seeing how it turns out.

Hook
Posted on: 01 October 2010 by pcstockton
quote:
the single core Atom has plenty of horsepower to run a dedicated music server.

Exactly. That list may be somewhat obsolete at this point, which should make everything even less expensive.

I had a Pentium 4 3ghz single core with 1 gb of RAM. It had NO PROBLEMS doing anything I wanted. Hi-res music (24/192), ripping and encoding to FLAC (very fast compared to my laptop with twice as much RAM), hidef video, etc... I even used to run a few pieces of work related software that require at least 1 gb RAM to themselves. I never noticed a drop in performance from my office set-up.

Now it has been tweaked slightly for quiet operation, and fully dedicated to music. Other than rebuilding it into a smaller case, I cant think of replacing it at this point. And if it is simply going to play music it will last for years to come.

I dont think I even use 10% of its capabilities when playing music.

The point of any dedicated solution isn't to be able to run 7 different intensive programs at once. There is a BIG difference between building a media computer and one for performance/gaming.

-Patrick
Posted on: 01 October 2010 by Hook
quote:
Originally posted by Occean:
...I wonder how much a naim case would cost in comparasion?


Good question. I would say that the metal work is relatively affordable, but it is that darned logo covering the IR port that costs the big bucks!

Of course, the logo does bring quality, warranty, service, support and high resale value, so I guess it is kind of worth it. Winker

Hook
Posted on: 02 October 2010 by js
More fun today. Our 3rd partner Mark just wanted an unbiased opinion and gave me 3 files to listen to on his Ipod touch. I didn't know what I was comparing and was to just rank them as they were simply numbered 1, 2, and 3. So I dutifully plugged in my $30 Brainwaves M1 in ears that I'd been tweeking a bit earlier and gave a listen. A couple minutes later I ranked them 3=>1 and then 2. 3 was a bit more full but 1 cleared out quite nicely and both were very tuneful. 2 was clearly below as though the notes weren't as complete, seemed to not be as flowing and the note a bit chopped though it was still overall not bad. Any description will sound more severe that it probably should under these conditions. So afterwards I looked at his notes and found all were the same cd ripped differently. 3 was Wavelab pro 6 via a great outboard transport and at a slow speed. 1 was the HDX and 2 was Itunes via a pc. Your results may vary.
Posted on: 02 October 2010 by Joe Bibb
You truly are a Golden Eared God js, you ought to take part in some of the file tests on the other well known UK forum, people have been unable to pick different interconnects, never mind lossless files on there. Winker The question we are left with is - would the results have seen the light of day had they not accorded? I think we should be told.
Posted on: 03 October 2010 by Guido Fawkes
I have to confess that I burned two CDs one using AIFF and the other using WAV. The AIFF was superior in every way and made the WAV file sound broken, almost unlistenable. I can honestly say the WAV file was quite dreadful and I fail to see how any sane person could prefer it to the AIFF.

The AIFF was created with iTunes no error correction and the WAV with EAC and matched against a database of accurate rips.

So why was AIFF so audibly superior?

A friend suggested that the AIFF burned CD being Sandy by the late great Sandy Denny and the WAV burned CD being Barry Manilow's Greatest B-sides had something to do with it.

On the same content, I can't tell an AIFF from WAV, EAC rip from iTunes and can hear no advantage from things like Pure Music; I can hear an advantage of using the BNC connection on the Naim DAC and, if giving it some stick, the USB port at the back. I don't like the optical port on the Mac and think the hiFace is worth the money. My experience with UPnP is not good so I've given up on it.

The main thing is that now I have something that works and is easy to use; I've stopped trying to tweak it. This is usually a good sign for me.

If you are using a Mac then my advice is run the minimum on it while playing music - if you can avoid networking then do so: if not and I don't always, at least disable network junk like directory services. Avoid third party plug-ins, do not run virus scanners and disable things like all those useless widgets that appear when you hit F12. Use lots of RAM and put all your music on a separate hard disk (go for a quiet one rather than worry about speed) in your Apple music server; if it'll only take one disk then use an external firewire drive (not a USB one - save that for exclusive use of the hiFace). If you can use SSD then go for it: when it becomes cheap enough then I'll swap all my HDDs for SSDs, but it is still too costly. Use BNC connectors whenever you can (unless you have a DAC with an APT-X BT interface, which unfortunately is not present on the Naim).

If you follow this you'll end up with a system like mine. No idea if you'll like it, but I do.

Two years ago I don't think you could achieve this level of replay from a computer and I put that down to the lack of a good Mac to DAC interface and a good DAC. The hiFace and the latest generation of DACs from Naim and Chord change this: this could be the buffering algorithms used by these two companies or it could just be magic.
Posted on: 03 October 2010 by Simon-in-Suffolk
quote:
Originally posted by ROTF:
A friend suggested that the AIFF burned CD being Sandy by the late great Sandy Denny and the WAV burned CD being Barry Manilow's Greatest B-sides had something to do with it.

Big Grin
Posted on: 03 October 2010 by js
quote:
Originally posted by Joe Bibb:
You truly are a Golden Eared God js, you ought to take part in some of the file tests on the other well known UK forum, people have been unable to pick different interconnects, never mind lossless files on there. Winker The question we are left with is - would the results have seen the light of day had they not accorded? I think we should be told.
It was possible that the setup wouldn't have been revealing enough to hear the type of diffs with the cheap phones and dap and in that case no, it wouldn't have seen daylight as a nonresult without the top kit in play doesn't say as much. It would have had to be redone on better kit. I don't trust nonresults without a bit more scrutiny.

Mark felt the same about the files but he knew them and is why he just handed me the touch and asked for the compare. All were made on a PC or HDX. All were wav. As I said, your results may vary but those were mine.

You'll also notice that I didn't think the HDX beat an optimized PC setup with an outboard plextor cd drive and unique ripping program at low speed so I am also not indicating that a HDX is the only viable avenue of attack which would be the case if I was just trying to pull some wool. That was at the limit of what was meaningful in that config so no clear winner. By the way, I was using a more standard version of Wavelab for rips long before the HDX appeared as it sounded better than any of the standard media programs years back before I ever heard of EAC or accurip. I don't think Wavelab calculates drive offset either but I suspect their solution gets around it by how it handles the files. Still the best software player I've heard also but not at all practical for most.