Your objectives are laudable, Bruce. To achieve them, go first to the local Apple Store. PC's are for other stuff - as demonstrated by the many frustrations you're encountering. Best of luck.

Great post!

I guess I went a different route. Just bought a Transporter. IMHO it sounds better in my system than the lavryDAC10 or Benchmark and rivals the AYRE.
I doubt i"ll ever use a CDP again.
I use ALAC for the MAC convenience and run the TP into Rowland Watt/Puppies

I'm back in audiophile heaven

Bruce,

I agree, a fascinating post, I

Yes, I'm still wide awake! I read both posts with interest. Thanks a lot to both of you for sharing your experiences, as I am puzzling with this same subject, and seeking a "high-value" solution.

Okay Bruce, so three months in and I am still peeling layers of inpenetrable complexity and obfuscation from this digital onion. However, I feel like I have graduated out of novice class ("you rearning glasshopper!") and feel I am now able to take a crack at your unanswered questions. Hope you're still reading the forum and didn't just conclude we were a bunch of Johnny-come-latelys and give up on us!

Quote:
Is my soundcard truly passing bit-for-bit without any manipulation?

Knowing what I now know I would say there is a very good chance it isn't. I have a "Creative Sound Blaster Audigy ADVANCED MB" (came built-in to my Dell XPS) which refused to do that. The giveaway was that even when playing back files ripped at 44.1kHz using KS or ASIO to defeat kmixer my Grace m902 would show it was locking on at 48kHz. MUCH to-ing and fro-ing with Dell and Creative support only induced severe hypertension and an overwhelming desire to harm the cat. Eventually the helpful souls on the Creative user forum provided the answer. Creative cards, and I don't know if it's some or all of them but definitely mine, upsample everything to 48kHz in hardware, independent of kmixer. This is apparently impossible to disable. However, my card lacks the "bit-for-bit transfer mode" which you mention. (Btw, where does that setting show up in your s/w? Mine only has one setting for the S/PDIF output: ON/OFF!) I ended up giving up and buying an M-Audio Audiophile 2496 on ebay.

Quote:
Foobar has direct kernel streaming but why does it have a volume control that works?

I am using foobar2000 v0.9.4.2 and can say that in 'standard' guise (i.e no UI plug-ins) there is no volume control visible when playing files through either KS or ASIO output devices and the windows volume control has, as expected, no effect.

Quote:
I still don

I have found this thread to be most interesting. I'm thinking ahead to the day that my transport (10 year old Museatex CDD) gives up the ghost. So I'm wondering if a computer audio/music server is the way to go. A couple of my friends maintain that cds that are bit perfect rips to a hard drive sound better than the cd player. One of them is a designer and small volume manufacturer of high end equipment so I tend to give this great credence. I need to study this so I can make an informed decision about which way to go and the products in the market place. Can you suggest websites/magazines/reviews for the pc audio/music server neophyte looking towards the future?
Thanks
Mike

Mike,

One of my main learnings thus far is how poorly this subject matter is mapped out. I think Bruce describes quite vividly the feeling of being out there alone armed only with a strong sense of conviction and basic common sense, a sort of digital Columbus trying to get to India and ending up in The Bahamas. That's how I felt too.

I wish I had noted down the various sites and resources I stumbled across along the way that were really helpful. Sometimes the nuggets were very few-and-far between, such as one post in a long discussion thread whose author really "got it" and who managed to undress some impenetrable contradiction and hit the switch on the elusive bulb of enlightenment.

Here are a few that spring to mind, I will add others if I find them but this should get you started. Please feel free to share any you find here too!

Your first port of call in my view should be Gordon Rankin's Wavelength Audio site. Also, I found that Art Dudley's review of the Wavelength Brick contained useful additional insights and context. I am sure the products are first-rate as well but I have no first-hand experience of them myself.

I found this review over at sixmoons.com very helpful, apart from the product under review it mentions a good number of the other contendahs. I have not checked out each of their sites but wouldn't mind betting there is more good info to be had there.

Generally, I found Hydrogen Audio a good place to get basic definitions/explanations and links to more information as well as specific pieces of software. For instance searching for arbitrary plug-ins on the foobar2000 site can be pretty daunting and sometimes impossible, whereas the relevant HA pages provide links that take you straight there.

Other than that I have found various bulletin boards very helpful. A couple of Swedish ones are particularly good but might not help you, however the Slim Devices user forum is a goldmine of information , and I am sure there is a wealth of good stuff over at the asylum if you have time to sort the wheat from the chaff.

All-in-all this is not a journey for the faint-hearted. Good luck!

Struts,
Thanks for the links, I will visit the sites and try to learn something. My preliminary reads indicate the Olive products (i.e. Symphony and Musica) are very interesting. I did note a passing concern about their copy protection methods; not sure if the issue was audibility or philosophy or both. Anyhow, I've got a lot of reading to do to getup to speed. How's the weather in Sweden?
Mike

Mike,

My pleasure.

Elk, who frequents these boards has an Olive, I am sure he would be happy to share his experiences. From what I understand..

Quote:
I have an Olive as I did not want a separate computer on all the time. At most it consumes seven watts and is silent. I added a USB external drive for backup/additional storage. I am accessing it through a Nokia N800 Interest Tablet. It is easy to find what I have stored on it. It sounds as good as the DAC that I am using. I like being able to easily compare recorded performances such as a particular Goldberg variation played by different pianists. The ability to easily find high bit-rate Internet stations and stream them to a good DAC is fun. I like being able to easily stream music that I have personally recorded and am editing to my main system for final sound checks without burning a CD-R and transferring it via sneaker net.

I am still not completely enamored with the technology however. I like the physicality of CDs and LPs. As I have no problem finding CDs in my collection, even though I have thousands, the computer database of recordings isn't that much of an advantage. I dislike needing to catalogue recordings while I am ripping them - the automatic CD database catalogs are great for pop and rock but lousy for classical - for example, the "composer" field is almost always blank and the composer's name and/or performer is in the "artist" field.

..he is happy with the product, but nevertheless puts his finger very deftly on many of the ways in which the user experience is still insufficiently refined to support the hump of the adoption bell-curve. You'll find the full thread here if you haven't seen it already.

My own opinion, fwiw, is that Olive is really onto something. It's ironic that as much as I consider myself a hardcore computer geek, figuring out how to get really high quality sound out of my PC has been a humblingly challenging experience. If the Olive system, as I understand, sidesteps many of these man-traps then I think they have targetted a very interesting segment of market.

The Sonos system, which I use, is a paragon of simplicity from an install/configure/maintain perspective however (to a first approximation) it still assumes one already has one's music ripped, correctly tagged, cover art loaded in the appropriate compressed format etc. to get the full benefit of the user experience. Recently helping my 65-year-old mother get up-and-running with an iPod (by telephone I should add) really opened my eyes to the magnitude of prerequisite skills and knowledge that is required but which any gen-Y-er would take completely for granted.

The sound quality question is one I find particularly interesting. I have done a considerable amount of background reading on this subject of late, covering the full gamut from subjective opinions to engineering papers and I have to say that the jury is still out for me. Hard-disk- or even RAM-driven playback certainly has some significant theoretical advantages over silver disc spinning but also suffers a number of practical problems.

When it comes to subjective comparisons I have yet to see the one I personally consider most interesting, namely, streamer vs CD transport with both connected via digital outputs to a SOTA DAC with a separate wordclock interface and with the DAC providing the master clock for both. Afaik this can only be achieved today with a Slim Transporter connected to either a dCS or Esoteric DAC with matching transport. No other products I am aware of offer the right interfaces.

I will hopefully be conducting a comparison of my own in the next few weeks, although this will perforce be using S/PDIF embedded wordclock as my DAC and streamer both lack separate clock interfaces. I'll certainly report back on my findings.

If you know of anybody who has actually documented relevant comparisons (something none of the audiophile press, print or web, appears to have done) please point me in the right direction. Until I either have the opportunity to compare myself or read a really good report from someone who has done a proper job of it, my great audiophile posterior will remain balanced delicately on the fence.

Weather here is great. Spring has come early and today we're basking in 15

Starchild,

I have an Olive music server. There is no copy protection in its software or otherwise.

The only copy protection you will encounter using a server is that which already exists in your source download, if any. For example, my understanding is that iTunes downloads incorporate DRM. (I may be wrong on this, I don't use iTunes.)

Hi Elk,
Are you able to add external hard drives to the Olive to expand your storage capcity? I don't understand why they're asking $2/gig for a larger hard drive when you can buy external drives for $.50/gig all day long?
Mike

Quote:
I don't understand why they're asking $2/gig for a larger hard drive when you can buy external drives for $.50/gig all day long?

I smell margin recovery .

.. or rather I thought I did. This from the Olive website:

Quote:
You can use an external USB hard drive to backup your music or extend the capacity of your Olive music center.

Note: To use an external hard drive you first have to format it correctly. Please use one of the following file systems:

FAT32 (MS Windows)
EXT3 (Linux)
UDF (UNIX)
HFS (Mac)
ISO9660

Other file systems like NTFS are not supported! FAT32 supports hard drives up to 2 TB, but formatting hard drives with bigger capacities under Microsoft Windows (in the above mentioned formats) requires special software/utility packages. We recommend the Freeware H2Format. You can get more information here or contact Olive support for details.

Quote:
Hi Elk,
Are you able to add external hard drives to the Olive to expand your storage capcity? I don't understand why they're asking $2/gig for a larger hard drive when you can buy external drives for $.50/gig all day long?
Mike

I have the same question.

External USB drives work great, both as backup to the onboard drive and as an additional storage source. I have a 250GB USB drive plugged into my Olive which works wonderfully.

By the way, I am slowly going over to the dark side. I am beginning to really enjoy how my CDs are soooo easy to access on the music server's drive. I didn't orignally see this as much of an advantage. Maybe I am just getting lazy, but this is pretty cool.

Elk,
I've recently met some folks at our fledgling audio society who maintain that digital played from a hard drive sounds better then digital from a cd player. I just haven't had a chance to listen to a music server/hard drive based system long enough to to zoom in it. The fact that these guys have equipment that I have little experience with only makes harder to identify what's happening. In other words I have no reference. As I spend more time listening to those systems, I'll get a better handle on it. I don't want to just run a cable from lap top to my pre-amp because I doubt I'd get decent results.
Mike

Quote:
I've recently met some folks at our fledgling audio society who maintain that digital played from a hard drive sounds better then digital from a cd player.

I have heard this before, but have not experienced this. On my system, digital files played back from my server's hard drive sound the same as those played from the original CD. I know that there are potentially some advantages to a hard drive by way of less jitter, but I have not heard any difference in practice.

Quote:
On my system, digital files played back from my server's hard drive sound the same as those played from the original CD.

Mike,

Elk's observation correlates with JA's in his review of the Sonos Music System and his (in my view) legendary quote:

Quote:
I was hard-pressed to hear much of a difference between the Levinson driven by the ZP80 receiving Apple Lossless Compressed or AIF files, and the original CDs from which I had ripped the tracks, as played back by the Class

Quote:
Somewhat frustratingly he left this thought hanging tantalizingly in the air and without exploring the obvious corollary. If an early example of a mass-market audio streamer is virtually indistinguishable as a digital source from a reference-quality silver disc spinner, then surely it opens the possibility for streamed audio to significantly eclipse 'spun' audio in the not-too-distant future with a truly no-holds-barred design? Unfortunately WP offered no additional enlightenment in his IMHO very disappointing review of the Slim Transporter where he seemed to ignore its digital output completely. A conspiracy theorist would no doubt speculate that this was all part of some kind of cover-up to protect certain interests!

.....

However it is also true to say that most computers are horrendously noisy RF environments with chips and busses operating at radio freqeuncies and switch-mode power supplies chugging out RF pollutants like an old coal-burning power station. It might therefore be true to say that streamed audio has the potential to offer superior quality to 'plate spinning' at a given cost point but that it requires very careful design of the serving system and a lot of care applied to traditional audiophile preoccupations such as power, grounding, RF shielding etc. to fulfil that promise.

Personally I have pretty much gone over to the dark side and consequently have no high-end transport to compare with but I am hoping to borrow one from my dealer over the coming weeks to conduct a comparison of my own.

Based on everything I have read so far, I am expecting any differences, in whichever direction, to be very slight indeed.

Struts,

Once again your post is a fine example of clear and concise writing with each point carefully examined and very well taken. Truly a pleasure to read.

Regarding the sections I've quoted above I think that you missed the point on the first one which has led you to be slightly off the mark on the later point.

Let me explain, I think that Wes Phillips' review of the Transporter did not "ignore its digital output completely" but rather chose to concentrate on the analog outputs because a large part of the cost of the Transporter is dedicated to the built in DAC. If one was going to use just the digital output of a Transporter one would be better served by buying the much less expensive Squeezebox and using the Squeezebox's digital output. WP did go into that very set up, i.e. a Squeezebox coupled with an outboard DAC, in some detail and found the sound of the Transporter's analog outputs to be somewhat superior to the sound of a Squeezebox feeding an external DAC.

As to the second part of your post, I completely agree that a computer is a terrible environment for the reproduction of high quality audio but I must point out that your statement "very careful design of the serving system and a lot of care applied to traditional audiophile preoccupations such as power, grounding, RF shielding etc. to fulfil that promise" is exactly what Slim Devices has tried to address with the Transporter. Sure the Transporter is among the first of its kind (a high end music server) and as other manufacturers present their own versions of this type of device we can only hope to see (and hear) further improvements and refinements to the concept but make no mistake, the Transporter is most definitely an all out assault on the state of the art.

Since I now own a Transporter and have been doing and will continue to do careful listening tests and comparisons between the Transporter and my CD player, I can safely say that, as you so rightly stated, the "differences, in whichever direction, to be very slight indeed."

Quote:
Struts,

Once again your post is a fine example of clear and concise writing with each point carefully examined and very well taken. Truly a pleasure to read.

I second this. Strut's posts always delight.

Quote:
Since I now own a Transporter and have been doing and will continue to do careful listening tests and comparisons between the Transporter and my CD player, I can safely say that, as you so rightly stated, the "differences, in whichever direction, to be very slight indeed."

Fully agreed.

I did not want to like music servers; they seem so pedestrian. WiFi streaming devices held little appeal and conjured images of computer speaker sound. Instead of embracing these bits-o-bias I bought one and have found it to work exceedingly well.

But for already owning a couple of DAC's and wanting to buy prior to the advent of the Transporter I would have bought one - a great DAC and streaming WiFi receiver all in one. Neat unit.

Thanks for the kind words jazzfan (and Elk). It feels like I am very much still wrestling this whole subject matter to the ground so it's nice to hear that my thoughts appear so organized when they hit the page!

I think you put your finger on a common phenomenon on boards like this jazzfan, namely that everyone's views are coloured to a greater or lesser degree by their own situation. My wife works in market research and they have a great expression in that business "where you sit is where you stand" which sums it up quite nicely. Despite knowing exactly where this mantrap lies it appears I have gone and stuck my foot right in it!

I have a pretty decent DAC (actually a preamp/DAC) in my system so I am only really interested in a streamer as a digital source. Of course I then tend to assume that everyone else is in the same place which as you rightly point out has made an ass out of me and umption. However that said, I think I will stick to my view that Wes 'ignored' the TP's digital out. And this even though I take your point about a large part of the raison d'etre for the TP being its high quality DACs and corresponding potential as an analogue source. For one thing I think it is reasonable to appraise the digital output of any component offering one, and secondly, I feel the provision of an external wordclock interface sets an expectation that this is a digital output with real ambitions in life! Although maybe not of interest to the majority of Slim's customers I would expect it to be of greater interest to the subset who are Stereophile readers.

To your second point, I agree entriely. All your points are well taken and it was not the TP I had in mind at all with those comments. I very confusingly changed the subject without changing the tone of my voice! I was actually thinking about starchild's comments questions about computer audio in a broader sense where essentially there seem to be three broad approaches to getting the music off the disc and delivered to the DAC:
1. A PC soundcard as demonstrated by Bruce in his original post (and as I use myself in my study).
2. A music server "all-in-one" like the Olive.
3. A wi-fi streamer such as the Squeezebox, TP or Sonos.

Clearly in the latter case the streaming device does not need to be connected directly to the PC or NAS and so issues of PSU or other RF pollution are less acute than when, for instance, connecting a soundcard directly onto the motherboard of a PC. Despite this, I agree that the TP adresses them in an exemplary audiophile manner.

I am fascinated by the TP, although I realize that I would only ever use a third of its capability and so it doesn't seem ideally suited to my needs. I would love to see a stripped down 'hotrod' version engineered to the same standards but only supporting a fixed level digital output, although in reality I fear that such a product would only find a very small market. What I would really like to see is said product delivered by Sonos with all the benefits of their fantastic handheld controller. Although given their mainstream consumer electronics positioning I fear this will also have to remain a pipedream. I expect Sonos's focus to be on broadening their addressable market rather than narrowing it!

Hi struts,

I do believe the computer is POTENTIALLY the superior platform. However it is all about:
- execution, how well is it done
- availability, what solutions are available that can
actually do the theoretically best job

I have tried an ' el cheapo' usb/spdif converter as input for my Monarchy Audio DIP & Mark Levinson 36 DAC, playing music directly from iTunes.

This turned out to be inferior (very much so) to the same music played via my Sonos music player as input to the same MA DIP & ML 36.

The wavelength brick sounds better than the 'el cheapo' usb/spdif, but not as good as the sonos, and that is because it uses a lesser DAC than the ML36 (both filtered)

same for the transporter, which was:
- worse (using the internal dac),
- slightly worse (using it to feed the ML36) or
- slightly better (feeding the MA DIP), but it is a bit expensive to serve as just a music server

I really want the computer to be my source though, as it opens up the audio future to true high resolution audio (some downloads available) with it's tremendous amount of software available...

although an audiophile grade Sonos would be fine as well, but as you suggested, they appear to be targeting the mainstrema rather then the audiophilealready...)

I am very interested in what www.lessloss.com is doing, they stll hold that a tuned setup of a slaved cd transport and their DAC wil win from any computer based setup, including their own...

I will be using a CEC TL-51x CD-transport shortly, and am expecting an improvement over the sonos (as it was an improvement on the Krell 300cd used as a transport, so I sold my Krell)

See there is potential, but it may take a while before it is realised in real products that we can buy..

My latest setup:

(a CEC TL-51x CD-transport ordered but not yet received/tested)

sonos music player, network attached (hardwired/Ethernet), 250 GB NAS storage (to be expanded shortly) ->

1.0 m Chord signature se digital ic ->

Monarchy Audio DIP ->

0.5 m Brilliance Purity DCT AES/EBU digital ic ->

Mark Levinson 36 DAC ->

0.5 m Briliance Euphony stereo RCA ic ->

Musical Fidelity KW hybrid preamp ->

0.5 m Briliance Euphony stereo RCA ic ->

Musical Fidelity KW750 ->

Brilliance Prodigy 2.5m speaker cable ->

2x Sonus Faber Amati

as soon as a se version of the prodigy is available, that is what I will use, far superior to the chord, for slightly less money

matsbeem,

Thanks very much for keeping this thread alive and for posting these findings - interesting reading!

You are the second person I have seen reporting on positive results using the Monarchy box with a Sonos source in as many weeks - the other was here, over on the Sonos forums - although that was the 48/96 upsampling version.

My own project to assess the impact of introducing an elderly dCS Purcell between my ZP80 and Boulder 1012 DAC/Pre has unfortunately been significantly delayed by the late arrival of the evaluation unit. I have recently started a new job and in the month or so since I received it I have simply not found the time to sit down and have a good listen.

If I ever get round to it I will certainly report back!

Let me just add some thoughts:

1/ Drop the audigy (or any other creative product) (granted, they recently purchased the e-mu brand. Those are OK). The chances that you manage to get it to play bit-perfect are very near nil. Personally, Ive opted for the m-audio revolution 5.1, and it does indeed play bit-perfect (assuming you don't attenuate/equalise), AND it can output 192Khz/24bit on its SPDIF out.

2/ I use winamp, with hqsoftproc mod to upconvert from 16 to 24 bits (so I CAN in fact use digital volume control without much loss.) converting 16->24bit at 100% volume is still perfectly lossless!

3/ I use winamp kernel streaming (out_ks36.dll I believe) to output bit-perfect signals (at 100% volume), bypassing windows XP volume.

4/ I use remoteamp2, on a pocketpc, connected through bluetooth, to control my media PC. If I do use winamp volume (through remoteamp) this first digs into the 'extra 8 bits' (16->24bit) so as long as you dont overdo it this should not hurt audio quality much.

5/ My Dac died recently so Im now in the market for a new one. Looking at:
- the Benchmark DAC1 (cheapish, good anti-jitter, but max 96khz sources)
- the AQVOX USB 2 (same price, not as good anti-jitter but accepts 192Khz sources)
- the Bel Canto DAC3 (expensive, good anti-jitter, maybe 192Khz sources!)

As you can tell, my main concern is jitter reduction, secondary is I want to be ready for DVD-A audio resolution sources, i.e. 192K/24bit

puntloos,

Also consider the Grace M902 and the Ps Audio DAC III. Both meet your requirements and are excellent sounding DACs.

I bought a black slim devices Transporter from www.broadbandstuff.co.uk to stream my music from my wireless network. The sound quality is great it is really easy to use I can also stream thousands of internet radio stations I

It's good to hear that you are so pleased, despite the technological setbacks. I recently set up a server system that I'm very happy with. I have an Apple iMac G5, use iTunes for management, and send the tracks to an Apple Airport Express. The digital signal from the router runs into an Audio Alchemy DDE v1.2. Though it can't quite match the liquidity of my CD player, it sounds excellent and I use it more often than the disc spinner.

Thank you for sharing your knowledge and experiences here. You inspired me to put together my computer-based system.

- Audigy might not be a good choice for high quality audio playback. Creative's driver s/w is huge and not very good.

Gaming has been the main market for consumer soundcards. Pro audio souncards are more useful for high quality 2-channel audio playback.

- Consider getting some backup hard drives. USB external drives are cheap and effective for backup.

- If you are re-ripping lots of CDs, you might consider alternatives for ripping s/w. If you want to do secure rips where errors are caught and usually corrected, EAC has been a standard recommendation. dBpoweramp costs money but uses the AccurateRip database and C2 error correction to rip considerably faster than EAC.

I use J. River Media Center 12 for playback and its own secure ripper as well. It is no faster than EAC but I can tag classical music the way I want to as I rip it. (MC 12 isn't free but it has an iTunes-like interface with far more features. It handles large libraries very well.)

> 2

I just added on of these, and now have no use for my old McCormack UDP-1. I will never go back:) Combined with a Mac, I am over the moon. The sound stage is wonderful, and the convenience of playlist is a whole new world. I was hoping for equal quality, but am blown away that it is a nice bump better:) (Genesis 501/McCormack DNA-2) I am sure the Brick (cheaper) is also excellent. Computer based audio is here to stay, IMHO.

Bruce,

I use a D-Link DIR-635 which is the predecessor to the DIR-655 and essentially the same unit but without the Gigabit ethernet support. I bought it for the same reason of the advanced QoS support (which is licensed form a 3rd party btw, you can read more about it here if you have trouble getting to sleep), as well as needing the extra range of 802.11n in my long stretched-out turn-of-the-century appartment (18" walls!).

The Qos seems to work well as neither VoIP calls nor music streaming are noticably affected by other network traffic like file downloads. However the 802.11n claims concerning 600% or whatever it was range improvement are pure BS IMO. After buying the dedicated cards for two PCs and spending hours on the phone to D-Link support to get the magic parameters to maximize range (no, they're not the defaults - PM me if you want them) it is still not producing any better signal strength (according to netstumbler) at the two most distant client machines than a bog standard Linksys WRT54GS. In some rooms my machine can see over 30 APs all trying to share 11 channels so signal strength is kind of important to me!

Have you had any issues with wireless coverage?

Same here, the QoS worked fine straight out of the box. It is only the wireless parameters I have had to tweak.

Let me share with you guys my 2 cents.

I have a room with large bookshelf speakers and an integrated amplifier being fed from my PC. I have lots of lossless .ape and .flac files, which are separated track by track, and I listen to them through Winamp.

I went first from an onboard soundcard to an Audigy 2 ZS, then to an M-Audio 2496, and I currently have an ESI Julia. Each upgrade was noticebly heard and felt, each was well worth their prices!

As for software, I used Winamp's DirectSound until I tried the otachan's ASIO plug-in along with upsampling. That was an excellent free upgrade. Now that I have a brand new dual-core system, I run 176.4kHz upsampling at Ultra mode and the microdetails are gorgeous. Winamp's default dithering and 24-bit support also must be enabled: they really add to the soundstage. My tests favored Winamp over Foobar, and ASIO over kernel streaming, upsampling over none, ultra over the other modes.

My next upgrade will be a Benchmark DAC1 USB. But I'll not sell my Julia: I also play and record instruments and this card is excellent for the home-studio guy like me. But for audiophile listening, I'll be using the card's SPDIF or TOSLINK (to be tested) into the Benchmark DAC1. I expect that the soundstage will be a lot wider and deeper.

My advice is: get rid of Creative's cards ASAP. Studio cards sound WAY better, and Benchmark DAC1 (some say) better still.

Summing up my testing experience:

• use only professional, studio-grade soundcards or a dedicated external DAC
• Winamp (using ASIO 176.4kHz Ultra mode, dithering, 24 bit)
• always lossless compressions (APE, FLAC)

Quote:
I expect that the soundstage will be a lot wider and deeper.

I am sure you won't be disappointed. Welcome to the forum.

The Juli@ is quite good, but its DAC is its weakness. The Benchmark DAC1 will be a huge improvement.

Great post! I hope you hang around.

I like the white soldermask on the Juli@ card! Very distinctive. Better than the boring, everyday green.

Just had to give props even though I contributed nothing to the thread.

Bob

Bruce,

I understand that you prefer not to upsample. Even though my tests showed improvements in my system, there is still great controversy about upsampling among audiophiles. But let's leave it for another thread entirely.

I also do know what "bit identical" is and I completely understand that you don't want to do digital to analog conversion inside the computer. Me too! My current setup is indeed converting inside the computer while I'm using the Julia, but when I upgrade to the Benchmark DAC1, the conversion will be done externally, just like your Accuphase player.

But even using PCs as digital sources, some soundcards do change the digital sound, especially Audigys. The Audigy series have this reputation of resampling all incoming sampling rates to a default 48kHz, because it's DSP works that way. Try googling for "Audigy resample 48khz", this is also described in Wikipedia. So even if you use the Audigy's digital outputs, it will NOT be bit identical, afaik. If you intend to use a digital output from your computer, be it SPDIF or TOSLINK, you'll need to choose a card that is bit identical, such as this Julia of mine (Claus Riethmueller, Managing Director of ESI, answered me in this thread).

As for storing your files in WAV, I don't know how much you know about lossless compression, but these APE and FLAC formats are bit identical. That means that you can compress and decompress a file 5, 10, 1000 times in a row and the resulting WAV file will be identical to the original WAV. It is like a ZIP, but optimised for audio compression. There's no need to store all your songs in WAV if a lossless compression is identically the same and files are 50% smaller (half the storage costs).

As for jitter in digital outputs, that's where Benchmark DAC1 does it's magic. Page 18 of the manual states that the DAC1 reclocks the digital interface to use it's high precision inner clock. That means that even if a signal full of jitter (digital timing errors) arrives at the DAC1, it ressamples the digital input in a frequency thousands of hertz higher, buffers the stream and feeds it's DAC with high precision timing. Great, huh? That's why the Benchmark DAC1 is such a succesful product.

I have only about 2 years of audiophile pursuit, and that's pretty much all I know about PC as an audiophile source.

Quote:
So even if you use the Audigy's digital outputs, it will NOT be bit identical, afaik.

Matias,

It depends on the model of card. My Dell PC came with a motherboard-based Audigy card where your statement was true, however I recently helped my neighbour connect his PC to his AV receiver via its Sound Blaster X-Fi Audio card which, unlike mine has a bit-for-bit mode selectable from the setup screen. Both my neighbour's receiver and my Grace m902 (and I believe Bruce's Accuphases) indicate the sample rate they are locked on to which makes it a bit easier to know when you've reached that elusive goal.

So I still think Bruce is right, there is no reason to believe that Juli@ would be better than a 2496 (that's what I use btw) or an Audigy when judged solely on the quality of its digital output. Neither am I aware of any comparison of sound cards that took digital outputs and their jitter levels into account. Lastly, I am not aware of any sound card aimed at audiophiles (would output low-jitter 16/44.1, 24/96 etc. 'bit perfectly' on Coax and XLR, and would be able to accept an external wordclock on BNC) and lacking any recording/gaming/home theatre functionality. Let me know if you know differently on any of these points!

Like you, I am a believer in upsampling and use a software-based upsampler in my headphone system and hardware-based one in my big-rig (see further up this thread for details). In fact my DAC upandoversamples - go figure!) Yet the best sounding all-in-one CD players I have heard like the Nagra and the Accuphase don't upsample so there is certainly more than one road to Rome. At the end of the day it's the sound, not the math, that counts.

Thanks for posting the photos Bruce, no idea what it is that I find so fascinating about looking at other people's systems (and the clutter on their desks). However there is a gallery specifically for the purpose right under 'Forums' on the menu. I would just hate to see this great thread veer off topic after 42 really interesting posts!

Quote:
Sound Blaster X-Fi Audio card which, unlike mine has a bit-for-bit mode selectable from the setup screen.

So I still think Bruce is right, there is no reason to believe that Juli@ would be better than a 2496 or an Audigy when judged solely on the quality of its digital output.

struts,

That inner ressampling to 48kHz applies only to Audigy cards, not the newer X-Fi. Therefore, there is reason to believe the Audigy's digital output is not bit identical. I also did state that Julia is bit identical, but I never said the M-Audio 2496 is not: it may even be bit identical, I don't know. So I agree, the Julia and the 2496 may be soundwise identical in their digital outputs.

What I said earlier was about their analog outputs: Audigy < 2496 < Julia.

Matias,

Ah, thanks for clarifying, I was inadvertantly tarring all Creative Audio cards with the same brush.

I seem to recall Bruce once stating that his particular Audigy card claimed to be able to operate in a 'bit-perfect' mode, that there was such a setting in the driver, and even that his Accuphase reported locking on to a 44.1 kS/s signal, but also that he subsequently said it was unreliable and seemed to reset itself occasionally.

However, if you are stating that you know for a fact that all Audigy cards resample then I certainly can't contradict you based on my own experience. I am just a little bit sceptical of 'proof by Wikipedia', I've seen some real howlers in there along the way.

Anyway, apologies if I came across as argumentative. I am just as interested as everyone else to get to the bottom of this.

PS I can confirm that the 2496 does not perform sample rate conversion however I realize that this is insufficent to prove that it is indeed passing a bit-perfect stream. Proving that conclusively (even for one sample rate let alone for all) is actually deceptively difficult and is not something I have attempted.

Bruce,

Believe me, if I had the answer I would have told you long ago! Unfortunately, as you yourself have pointed out on many occasions, clear facts are in short supply here.

There was some discussion of this in a thread about the Benchmark DAC1 over on the head-fi board. Unfortunately the site is down at the moment so I can't give you the exact link but it was on page 48(!) of the thread when it comes back up.

One of the regulars, thomaspf, who posts frequently on many forums including the Audio Asylum, AVS, J. River etc. and appears to work as a consulting engineer, had tested the bit-transparency of some USB cards and was challenging the validity of some of Benchmark's claims:

Quote:
[The Benchmark DAC1] works bit transparently just fine with kernel streaming like most of the other USB adapters using the USBaudio.sys driver. Your device is basically not better or worse than anyone else. The one custom driver that can work bit perfect with wav and DirectSound on XP that I know of is the M-audio Transit.

He was also arguing, if I got the right end of the stick, that Benchmark's method of testing bit-transparency using the Audio Precision signal generator/analyzer is flawed specifically because it encodes the test data in a WAV file. He implies that WAVs are generated using DirectSound library functions and are consequently manipulated by KMixer during ripping, regardless of whether KMixer is bypassed on replay using KS/ASIO. I have not seen Benchmark or Audio Precision conclusively rebut these claims.

Quote:
The problem is that Kmixer in Windows is actually multiplying all values by a number smaller than 1.0 at max volume so there is actually no way this can be bit perfect.*

In other words, even if your replay chain is bit-perfect your carefully ripped programme may not be...

Hey, step back from the edge Bruce, it might not be true! I don't know enough about the internals of Windows XP to be able to be able to evaluate this (although it doesn't seem far fetched based on what I do know) and I might well have gotten the whole thing upside down. I have spent a couple of hours digging though the Microsoft Developer Network site and I'm none the wiser. Consider this another unproven hypothesis to add to your growing pile. thomaspf goes on:

Quote:
On Vista you can get the stack to pass 16 bits through unmangled at 44.1 Khz if you select 24bit / 44.1 Khz in the audio control panel and output the 16bit stream as 24bit data with the data residing in the 16 most significant bits.

So maybe ripping to WAV under Vista is the answer? (I am planning to upgrade anyway just-in-case). However at the very least it appears that the sound card, device driver (version) and operating system (version) variables all need to be bound before anyone can test/prove/warrant anything concerning bit-transparency. Welcome to the wonderful world of open systems!

If I stumble on any other nuggets of enlightenment along the way I'll let you know. In the meantime, Happy holidays!

* This assertion is contradicted by the information provided by Microsoft:

Quote:
In Windows XP and later, the operating system uses a different set of default volume settings to avoid this loss of audio quality. It sets the volume levels on the wave, CD audio, MIDI, and other audio sources to zero decibels of attenuation (pass-through mode). This translates to a full-volume slider setting for each of these sources in SndVol32. These settings improve the default sound quality by ensuring that KMixer does not degrade the original resolution of the source signal.

In Windows XP and later, the default master-volume setting remains unchanged at 6 decibels of attenuation. The master volume control represents the hardware volume node on the adapter pin to which KMixer's output pin is connected.

Although in my opinion this is insufficient to debunk thomaspf's claim completely and prove that ripping to WAV is indeed bit-perfect.

Have been learning from others who have gone foward with setting up a computer based audio system. Last week I downloaded iTunes for my HP laptop computer. I purchased a Sea gate 500 GB external HD and a Benchmark USB DAC.

Have recorded over 40 CD's so far. Very pleased with the sound quality through by tube McIntosh stereo components. Truly enjoy the ease in which I can record and playback my music.

There is still room for improvement in computer based audio systems. The new issue of Absolute sound has some interesting comments on what is expected for digital audio in the coming years.

Hopefully we will be able to download 24/96 bit digital to our computer and no longer be handicaped with the 16/44.1 format of current CD's.

Only time will tell.

Bill

Quote:
Hopefully we will be able to download 24/96 bit digital to our computer and no longer be handicaped with the 16/44.1 format of current CD's.

You already can, and believe-it-or-not some of them are actually free! Check out the appropriate music forum further down the menu and search on "lossless download" for lots of links. Depending on your taste in music this might be a good place to start.

Welcome to the forum!

Quote:
Depending on your taste in music this might be a good place to start.

Wow, this is what the future of music looks like: track by track, multi-resolution audio downloads. I believe Linn Records is ready for the 21st century.

Struts,

In my tests the M-Audio Audiophile 2496 does pass a perfect 16 bit 44.1kHz data stream using the Foobar ASIO driver or the DS driver. However the datastream is altered using the Foobar KS driver.

I use a Lynx L22 card in my music player PC and have an Audiophile 2496 in my main PC. I hooked the digital out of the 2496 to the digital in of the Lynx and used Foobar 0.9.4.5 to play a Wave file and Sony Sound Forge Audio Studio to record it via the Lynx. I used the resynchronising compare function of a hex editor to compare the original and recorded files.

I wasn't surprised that the ASIO driver produced a perfect copy but, based on previous experience, I expected the Directsound driver to alter the datastream very slightly and it didn't. I tried using Windows Media Player 9 to play the file and it also produced a perfect copy. Maybe something has changed recently in Directsound as my previous testing has shown that it altered the data.

I thought you might be interested.

Dave