Error?! [Actually, maybe not...]

[Superaintit]

Hi 6th.replicant,

I'm sorry to hear the rdac is not what you expected. Reading your posts re the comparison with the cd37 was interesting. I haven't directly compared the rdac with an arcam cd player, but I have heard the cd192, the rdac connected to the top of the range arcam cd-player, the rdac connected to my mac mini by optical spdif and usb and the Cyrus dac-x connected to a cyrus transport and my mac.

Having owned the rdac now for several weeks, the tonal character compared to a cyrus dac-x is:

-a natural way of presenation, without sheen or brightness.

-detailed, although not as detailed as the dac-x. The cyrus can unravel a recording like no other. For example the Massive Attack's unchained melody starts with a deep base that's more defined on the cyrus. The Rdac's base is deep but doesn't look as deep into the recording as the cyrus. To me that's a good trait, as I own many pop and rock albums that I now can enjoy, whereas they previously made me wince.

-base: big, bold, but nothing too serious if the rest of your equipment doesn't emphasise this characteristic.

In my experience the rdac works best with the following settings:

-connected over usb (more detailed then optical)

-Audio/Midi configuration: standard out: Arcam Dac, 88.2 , 2 channel 24bit

-and of course all 'sound enhancing' functions in Itunes off. Volume should be set on maximum too. Volume control by amp.

-accessing music files via front row (the remote) yields slightly better results to my ears then via regular itunes. Probably less video processing going on in the background.

-the rdac has improved with time -from memory it took about a week.

Summarizing: the rdac is all that I could want, maybe a smidgen more detail and control in the base would be nice, but listening through the a32 it never bothers me. However, your preference may differ and the cd37 may well be a better package overall for you. For me there is no turning back from computer audio: it's got me the tonal characteristics I like and the user interface is so much better than all those cds hanging around.

Edit: thought I should add that I listened to a system comprising of a arcam cd-player, rdac via optical, a38 and dynaudio contour, which I didn't like at all. The rdac in my experience needs some carefull matching and is far better via mac then cd-player. All IMO of course.

 

[SteveR750]

shooter69:SteveR750:
As I understand it re-clocking is a process that takes place in a remote DAC, because of the difficulty of synchronising a master clock over s/pdif or USB for that matter...The transport will have its own clock which determines its own sampling rate. The challenge is to get the remote DAC to sample at exactly the same time interval as the output data, but the output device clock data is also subject to transmission errors, hence the need to use Phase Locked Loops etc to establish it. Ideally the master clock is controlling the DAC sample rate and is also controlling the output rate of the transport, but this is not easy to achieve. In a integratd CDP the master clock runs both processes, ad in theory is inherently more accurate, but then electrical intereference, power supplies, vibration all start to have an effect. Like many things, the reality is often a compromise of the ideal science.

Check this out.EDITED BY MODS - House Rules Explains how master clocks and DAC's are implemented and the cheapest ways to do it.

I posted that link elsewhere.

This is a good site:

AGAIN, EDITED BY MODS - House Rules

 

[6th.replicant]

Thanks for numerous replies/advice - much appreciated

(Although I'm a wee bit baffled by the discussions re 'ones and zeros science'

)

As for rDAC reviews throwing up the "careful system-matching required" caveat, I assumed that I'd ticked that box - my amps are also made by Arcam

Perhaps the rDAC does need some run-in? Been out since late last night and during that time the rDAC's been running 17 hrs non-stop. Before I give it a listen, I'll give it another overnighter, then have a play tomorrow. Might also call Arcam to get its views re run-in.

In hindsight, it's very naive of me to expect the SQ of £300 box to match its £1300 sibling, even if both have Wolfson WM8741 tech.

I'm sure there are better DACs than the rDAC, but here's my conundrum: I'm very happy with my CDP, thus only want a DAC for computer-based music duties and, therefore, to complement my CDP, not replace. Ergo, I don't want my digital music to have two different 'house sounds' via two different manufacturers' DACs - an issue that Audiolab's neatly addressed with its 8200CD CDP/DAC combo.

And I'm not up for replacing the CD37 with a CDP/DAC combo just now, because IMHO the CD37's SA-CD playback is currently hard to beat, for the money. Hopefully, the CD37's replacement will be a CDP/DAC...

Re iTunes setup: I'll try cutting Sound Enhancer, Equalizer - found it! Ta, JD - and Soundcheck; iTunes volume is already set to max.

Here's hoping that the rDAC does need run-in...

Laters

 

[SteveR750]

I am still not convinced that iTunes / Mac is the best SQ solution into a DAC. There is a review of J River on the computer audiophile website that found as much:

" In January 2010 at CES I listened to a demo using JRMC 14 and was really pleased with the sound. JRMC was running on a Mac laptop with Boot Camp and Windows 7. I compared the sound to the OS X / iTunes partition on the same laptop and was surprised at how much better I like JRMC in that system. I was finally convinced I needed to take JRMC much more seriously. Shortly after arriving home from Las Vegas I arranged a meeting with the people at J river and started using the application exclusively."

 

[6th.replicant]

To paraphrase a tall, dignified alien, with a gold hieroglyph embossed on his bonce: "Undomesticated equines could not induce me to return to using Windows."

But hey, maybe OS X + JRMC is worth a look - or is JRMC Windows-only?

Ta

 

[Craig M.]

i've never used a windows pc to play music, but logic says that given an identical bit stream from each platform, then there is no reason for them to sound different. one could have more jitter present, but then you're into the audibility of jitter and how well the dac reduces it. i've tried the demo version of pure music player in place of itunes, initially i thought it really sounded different, but, on repeated comparison, i think i was fooled by the difference in volume between the two players. pure music was set at something like 75% volume, which i didn't notice, and i turned the amp up to compensate. comparing them both at 100% volume yielded no apparent difference to me. maybe because i'm sceptical, i am conditioned to miss obvious differences.

a further confirmation that all is well with the optical out of my macbook, came when comparing it to a cambridge audio 640cd used as a transport into my cousins new adm 9.1s(

), there was absolutely no difference - as you would expect with two bit perfect sources.

when i read accounts of two bit perfect media players sounding different then, assuming jitter is dealt with by the dac, i have to conclude that the listener is at fault in some way. what other explanation could there be?

 

[DavieCee]

6th.replicant:To paraphrase a tall, dignified alien, with a gold hieroglyph embossed on his bonce: "Undomesticated equines could not induce me to return to using Windows." But hey, maybe OS X + JRMC is worth a look - or is JRMC Windows-only? Ta

Unfortunately, when I looked last week it was Windows only so I am sticking to iTunes for now. Some one on another thread suggested that Front Row gave better results for some reason. I will be giving it a go tonight.

 

[6th.replicant]

Superaintit:In my experience the rdac works best with the following settings:
...Audio/Midi configuration: standard out: Arcam Dac, 88.2 , 2 channel 24bit...
Hi Superaintit

Interesting to hear your views re the rDAC.

Can you please explain, in layman's terms, what you mean by, "Audio/Midi configuration: standard out: Arcam Dac, 88.2 , 2 channel 24bit" - are these settings something to do with the rDAC's setup or ...?

Ta

 

[Craig M.]

on your mac -> applications -> utilities -> audio midi setup.

change the format from 44100 and 16 bit -> 88200 and 24 bit.

you may as well leave the bit rate at 24, it will not effect the sound of 16 bit files and means you don't have to change bit rate if you d/l some high res music. not sure what, if any, benefit you will get from doubling the sample rate from 44.100 to 88.200 khz. audio midi setup, is where you adjust the sample rate if you want to play high res files. for changes to audio midi to take effect, you need to exit itunes, make changes to audio midi, and then restart itunes - making sure to change it all back when you listen to normal cd files. pain in the backside.

 

[AlmaataKZ]

Craig M.:

i've never used a windows pc to play music, but logic says that given an identical bit stream from each platform, then there is no reason for them to sound different. one could have more jitter present, but then you're into the audibility of jitter and how well the dac reduces it. i've tried the demo version of pure music player in place of itunes, initially i thought it really sounded different, but, on repeated comparison, i think i was fooled by the difference in volume between the two players. pure music was set at something like 75% volume, which i didn't notice, and i turned the amp up to compensate. comparing them both at 100% volume yielded no apparent difference to me. maybe because i'm sceptical, i am conditioned to miss obvious differences.

a further confirmation that all is well with the optical out of my macbook, came when comparing it to a cambridge audio 640cd used as a transport into my cousins new adm 9.1s(

), there was absolutely no difference - as you would expect with two bit perfect sources.

when i read accounts of two bit perfect media players sounding different then, assuming jitter is dealt with by the dac, i have to conclude that the listener is at fault in some way. what other explanation could there be?

 

[SteveR750]

Craig M.:
i've never used a windows pc to play music, but logic says that given an identical bit stream from each platform, then there is no reason for them to sound different. one could have more jitter present, but then you're into the audibility of jitter and how well the dac reduces it. i've tried the demo version of pure music player in place of itunes, initially i thought it really sounded different, but, on repeated comparison, i think i was fooled by the difference in volume between the two players. pure music was set at something like 75% volume, which i didn't notice, and i turned the amp up to compensate. comparing them both at 100% volume yielded no apparent difference to me. maybe because i'm sceptical, i am conditioned to miss obvious differences.

a further confirmation that all is well with the optical out of my macbook, came when comparing it to a cambridge audio 640cd used as a transport into my cousins new adm 9.1s(

), there was absolutely no difference - as you would expect with two bit perfect sources.

when i read accounts of two bit perfect media players sounding different then, assuming jitter is dealt with by the dac, i have to conclude that the listener is at fault in some way. what other explanation could there be?

You might be assuming wrong. Jitter can be reduced but not eliminated, and not all DACs are the same. I'm also not sure that you can claim the output from a CA 640 being "bit perfect", as it wold be a pretty unique trsnport if it was...! I had one, and it failed to read two discs that the NAD 542 that it was quickly replaced with did. It did it with 2 different machines so didn't appear to be isolated...

And it's worth adding, I have no special preference for Windows, so I'm really not interested in an "OSX is better than windows" argument, in fact I am still not 100% certain that I made the rght choice going for a PC anyway... I just have an open mind about what is possible using any type of PC in any configuration, and how to get it to work optimally.

 

[Craig M.]

i also have not the slightest interest in saying a mac is better then pc, if you read my post you will have seen that is what i said. i talked about jitter reduction and not elimination, and also thought i mentioned the dacs ability to deal with jitter potentially upsetting things

i don't see why you assume that because a cdp wouldn't read a couple of discs, that it wouldn't put out a bit perfect signal with the ones it did read. and, even if it didn't read them as well as another cdp, the error correction should be robust enough to correct this. another indication for me is that my dac locks on and displays 44.1 khz when hooked up to it.

 

[shooter]

Do some searching and you'll find lots of information regarding jitter within PC from inadequately shielded A/D and D/A converter chips on your PCI sound card, Radio Frequency Interference (high frequency noise) generated by high-speed signal lines. PCI sound cards drawing power from the computer's main power supply and generating substantial noise. And worst of all, it it all ends up on the HD when you rip resulting in loss of detail, clarity, width and definition in your soundstage.

In the DAC, the best way to deal with it seems to be shielding from electric or magnetic fields, good grounding, good power-supply decoupling, and a good signal transmission between the clock generator and the actual D/A chip.

So how can you tell if your PC or DAC a good one or not, unfortunately you cant unless its specified.

But there is some good news, EMU sound cards 1010, 1212m and 1820/m have some good press for audiophiles regarding low jitter. These will eliminate as much jitter as possible at source.

Just out of interest has anyone got a firewire DAC up and running? If so which one.

This type of DAC could interest me as it doesn't seem to be inhibited by the length of cable between PC and DAC.

 

[Craig M.]

i know what you mean shooter, but i have heard several cdps as transports versus a mac mini and macbook and they sounded the same into each dac i've tried, bar a cyrus dacxp.

if the bits are the same and the jitter is reduced by the dac to below audible levels (if it's even a problem!), then i don't see the problem. i'm not being purely theoretical, my own experience backs it up.

the adm9.1s i listened to the other night are pretty revealing, not one iota of difference between transports to mine, or my cousins, ears.

edit: i've just read your post again, and don't agree at all with this bit "And worst of all, it it all ends up on the HD when you rip resulting in
loss of detail, clarity, width and definition in your soundstage.". if all that information has been lost, then why isn't it missing from the bit perfect copy? as long as it is ripped accurately, which is easily done, then the information on cd and hdd is the same. none added and none taken away.

 

[Anonymous]

Also using optical to the DAC removes any electrical interference.

Incidentally, does anyone know at what level jitter becomes audible? I was under the impression most DACs fall well under that level, whatever it is.

 

[6th.replicant]

Today I spoke with a very helpful, and patient, fella at Arcam, who confirmed that the rDAC needs a week/100-ish hours of run-in.

So, thanks to Dan Turner and Superaintit for the heads up re run-in - were it not for your advice, I'd have returned the rDAC by now. Speaking of which, the fella at Arcam said that if I'm still not happy after a week's run-in, then I should call him and he'll arrange for the dealer to accept a return/full refund. Blimey!

Have now given rDAC some non-stop overnight/daytime running - must have c. 50 hours by now - and the flabby/overbearing bass has tightened by a tidy amount, but it still bloats when frequencies go very low, although there is now some extra mid-range detail appearing.

Then tried the various suggestions re iTunes settings, which have added more clarity and detail. Thanks for that advice

However, with Soundcheck off, shuffle is a 'mare as volume levels soar and plummet from track to track

Compared to the CD37, IMO the rDAC is still a 'blunt instrument' overall, but maybe it needs the full 100 hours?

Ta, again, for advice

 

[SteveR750]

Craig M.:
i don't see why you assume that because a cdp wouldn't read a couple of discs, that it wouldn't put out a bit perfect signal with the ones it did read. and, even if it didn't read them as well as another cdp, the error correction should be robust enough to correct this. another indication for me is that my dac locks on and displays 44.1 khz when hooked up to it.

In which case I'm not sure I understand error correction!

I presumed that if the laser does not read the sector first time round then error correction "fills in the blanks". I can accept (though not yet understand) that to some point, this is mathematically correct (and thus and most importantly inaudible), but what happens when it fails to read that bit and has no time to re-read (from what I can deduce from CDP design off the net so far buffering in a integrated CD is not the usual approach as designers prefer to use a master clock close to the DAC) and the error correction is either audible (and how does that sound different from jitter) or as fatal as "clicking" or dropout?

I have assumed that the definition of bit perfect means that the output stream from the player is 100% the same as the data from the file. I guess error correction (like lossless compession?) can correct to a 100% accuracy, but at some point it fails, otherwise my ears are grossly deceiving me! I may be completetely misunderstanding you, but I think that what you are saying is that error correction fixes all errors, no matter how big so long as there are no "obvious" catastrophic failures, in which case my chocco wrapper transport would sound fabulous, only it doesn't!

What I learnt thus far from the small bits of info I have read is that inherent in any disc steaming system are reading + clocking errors when viewed at the input to the DAC. Jitter it seems to me appears to manifest itself audibly as roughness, coldness, clinical sound; which seems logical as jitter would cause less accuracy in reproduction at higher frequencies where you are approaching the resolution limits of the sampling frequency.

Ultimately, I can only go by what I have heard, and what I do know is that my PC system is as much an improvement over the CD6SE as the latter was over the NAD 542 when played through my old C352 and B&W speakers. I cannot understate how big it is, though it took a few days configuring it to get it to right. I would just like to understand why.

 

[Craig M.]

steve, you make some interesting points, if i can find where i got my information on how error correction in a cdp works (in real time) i'll try to post it. i think it had something to do with a mathemetical formula and redundant data, so if the laser missed the bit it wants, the exact bit is recoverable from the data around it. or something. i'll try to find it but i remember thinking that only a huge scratch that would create an audible skip, would defeat it.

to have a go at your last question though - maybe system synergy? maybe in a completely different system you would prefer the cd6se to the dm? maybe if you could feed the dm with the bit stream from the cd6se (ala 8se), you wouldn't hear a difference to your pc providing the bit stream?

i went through all this internet searching a while back

, and never found anything that provided a robust argument for bit perfect sources sounding different, jitter being audible at anything like the output from most modern dacs/cdps/dvdplayers, or error correction not working!

edit: just found this: "The CIRC
code used on audio CDs can correct burst errors of up to 3500 bits (2.4mm) and
can interpolate error bursts of up to 12,000 bits (8.5 mm). "

this means that upto 2.4 mm long scratches can be corrected, doesn't sound long but i would guess the odds of a scratch following the track of the cd exactly, is pretty slim. i'll keep looking for the original info i read.

p.s., i hope i'm not coming across like a git. a 5 week old baby keeps some of my responses a bit shorter then i would like, and i know it can be hard to determine meaning from a post.

 

[SteveR750]

Hey Craig, no offence taken don't worry! I did my late nights shifts a while back but know what they are like...

maybe in a completely different system you would prefer the cd6se to the dm? maybe if you could feed the dm with the bit stream from the cd6se (ala 8se), you wouldn't hear a difference to your pc providing the bit stream?

Unlikely - I reckon the Cyrus DAC is *probably* *Better* than the DM, plus there is no interefering s/pdif, windows, optical cable jitter etc etc between the 6s reader and dac.

never found anything that provided a robust argument for bit perfect sources sounding different, jitter being audible at anything like the output from most modern dacs/cdps/dvdplayers, or error correction not working!

Nor me, other than other opinions (as in what they heard) and my ears... Agree there is no concensus on what are the measurable -> audible levels, and begs the question if jitter is the only audible, or even critical source of distortion...given the difference in the filters on the DM probably not, and possibly not respectively. Unfortunately I have a problem, and that it is that if I don't know I have got the best set up, then I won't believe my ears unconditionally, but then, I don't think I am alone in this.

Sleep well dude!

 

[6th.replicant]

So, the rDAC's now had a week's/100-ish hours of run-in...

And Dan Turner, Superaintit and the fella I spoke with at Arcam are correct - the rDAC does need running-in.

To my ears, there's now more overall detail, especially mid-range and top-end, along with more snap and attack. And the all-consuming, Flubber-style low-end bass bloat has simply gone, replaced by a much tighter and more focussed punch, which is also slightly deeper.

However, I am bemused - what the hell happens during run-in that it should make such a fundamental difference??

It's tempting to think that I've now merely become accustomed to the rDAC's sound, but having done some A/Bs with the CD37, that's not the case.

Sure, the rDAC still can't quite match the CD37's detail, texture and imaging, but at least they both now sound as if they're 'from the same family'.

Result!

And thanks again to all for their advice.