Ticks going from one track to next

[mnauta]

Just joined the site after reading it for a while, finally decided to join in.

 

I downloaded form Channel Classics NETHERLANDSBACHSOCIETY_30809_24BIT_96KHZ_2CH_FLAC_COMPLETE. However, there is a tick sound every time it goes to the next track. I have never had this before on any download? Does this happen more often, anyone experienced this?

 

 

 

[flatmap]

Welcome, mnauta.

 

Since this doesn't seem to happen with your other files, I think I would inspect these files with an audio editor. A good (and free) choice is Audacity which is open source and cross-platform software for recording and editing. You can load in a file and look to see if there is a short burst at the end where you are hearing the tick.

 

And if the files do have some kind of artifact, you can also use Audacity to trim off the offensive part. Also you could email the vendor and tell them what you hear and see and ask them to fix it.

[mnauta]

Thanks for the Audacity software suggestion, I'll give that a try. Also waiting on Channel Classics support response.

 

 

Sent from my iPhone using Computer Audiophile mobile app

[mnauta]

Welcome, mnauta.

 

Since this doesn't seem to happen with your other files, I think I would inspect these files with an audio editor. A good (and free) choice is Audacity which is open source and cross-platform software for recording and editing. You can load in a file and look to see if there is a short burst at the end where you are hearing the tick.

 

And if the files do have some kind of artifact, you can also use Audacity to trim off the offensive part. Also you could email the vendor and tell them what you hear and see and ask them to fix it.

 

Ran the tracks trough Audacity and found the clicks on all tracks:

 

[Axiom05]

Just joined the site after reading it for a while, finally decided to join in.

 

I downloaded form Channel Classics NETHERLANDSBACHSOCIETY_30809_24BIT_96KHZ_2CH_FLAC_COMPLETE. However, there is a tick sound every time it goes to the next track. I have never had this before on any download? Does this happen more often, anyone experienced this?

 

 

[ATTACH=CONFIG]23021[/ATTACH]

 

Yes, I have experienced this on several Channel Classics PCM conversions. My understanding is that it is an artifact from converting the DSD master to PCM format. This is the reason that I will no longer buy PCM files from Channel Classics. If you can play DSD music files, you are better off using that format, otherwise buy from a different label.

[Axiom05]

Did you get any resolution to this?

[mnauta]

Did you get any resolution to this?

 

I got email back form Jared Sacks verily quickly. I mailed him the results of the Audacity analysis, now just waiting on the response.

[jazz01]

I've recognized similar 'ticks'/or clicking noise how I've descript them between tracks of some DSD alben - it was really annoying on Gershwin's Rapsody in Blue from 2xHD downloaded from Prostudiomasters so I wrote to Prostudiomasters; this was their response:

 

'The clicking between tracks usually goes away when files are played through music software such as J. River, iTunes or equivalent. The clicking occurs where the individual tracks are indexed but are intended to be played continuously without interruption'

 

I later bought a FLAC 96kHz version of the same Album which originally was also recorded in 96kHz and this version has not the clicking noise - so I am more convinced with the explanation that it might come from converting by whoever has converted them ...

[mnauta]

I thought I would give it some time to allow for a response, but no luck. This non-response and blame the hardware excuse was the first and last thing I heard from Channel Classics

 

On 12/30/15, 3:13 PM, "[email protected]" <[email protected]> wrote:

 

Dear Manuel,

This click is dependent of the DAC being used. From the hundreds of downloads we have only had a handful of feedback about clicks between movements. I do not know whether a Flac or a Wave file makes the difference in this. It is the file going to digital ‘0’. Most DAC’s solve this problem by making a kind of crossfade when there is a stop to the data.

Since you are not having this on other files, maybe are files are different which I would like to look into.

 

Greetings

Jared Sacks

 

 

I send Jared the Audacity screen shots, but no response. I then downloaded a free DSD sample track from Native DSD Music (ironically they had the exact same Bach album track as a free sample) and no end of track tick. Turns out Jared Sacks also runs Native DSD site. Got a marketing email from Jared via Native DSD asking how I liked the service. I replied and asked how the status of my question was .... LOL! then never heard back.

 

It is most likely a PCM to DSD issue, all that is needed is an acknowledgement. Blaming hardware might not be the best business practice.

[audiventory]

Clicks for DSD to FLAC conversion can depend not only by conversion software but can depend on DSD stuff.

 

It is main reason of development of clickless technologies kit.

 

Details here (as for ISO as for files album) How Convert ISO to DSF, WAV, etc. without Clicks (Mac OS X and Windows)

[Axiom05]

Clicks for DSD to FLAC conversion can depend not only by conversion software but can depend on DSD stuff.

 

It is main reason of development of clickless technologies kit.

 

Details here (as for ISO as for files album) How Convert ISO to DSF, WAV, etc. without Clicks (Mac OS X and Windows)

 

Is there any way of removing "ticks" from the flac file?

[audiventory]

It depend on where placed clicks. If clicks placed in time distance (from begin and end) less than several milliseconds by begin and end - AuI ConverteR can supress it via accurate fade in/out (Smooth DC function). Smooth DC turn ON/OFF in Settings window > General tab

 

 

In FREE DEMO possibly try process all FLAC files of album sequentially by one (batch converting not available).

 

If time distance more need manual editing of the FLAC files.

[mnauta]

I downloaded my second album form Channel Classics - Antonio Vivaldi, L'Estro Armonico with Rachel Podger because, a) I really wanted this highly praised album and b) I wanted to know if other downloads from CC had ticks in track changes.

 

The good news is that there are no ticks between any of the 32 tracks and it's a great album in every aspect.

 

(I'm still bummed though that the Bach Oratorio download has ticks. )

[Axiom05]

I downloaded my second album form Channel Classics - Antonio Vivaldi, L'Estro Armonico with Rachel Podger because, a) I really wanted this highly praised album and b) I wanted to know if other downloads from CC had ticks in track changes.

 

The good news is that there are no ticks between any of the 32 tracks and it's a great album in every aspect.

 

(I'm still bummed though that the Bach Oratorio download has ticks. )

 

Yes, I found that the Podger download was tick-free as well. This was my first Channel Classics PCM download and I was quite happy with it. This also shows that the DSD to PCM conversion can be done without ticks between tracks, certainly other labels have been doing this conversion successfully for years. I share you disappointment in my other Channel Classics PCM downloads.

[tailspn]

All Channel Classics recordings from 1999 onward were recorded in DSD. All DSD bit streams contain a DC offset component, specified for SACD production as being no greater than -50dBfs. That's about several millivolts in consumer 2 Volt full scale system usage. Many (depending when the conversion took place) Channel Classics DSD edited masters were converted to PCM (FLAC) for downloads using the PQ markers contained in the DSD edited master file. That's the text information that drives the SACD player display when played as an SACD, and also allows player entry at a track start. The timing/positioning of these PQ markers has no relevance to the instantaneous DC offset level, only to a convenient time to change the SACD player track information.

 

Channel Classics recordings sold on nativedsd.com originate from those same DSD edited masters, but each track slicing point was manually chosen to be about one second prior to music, and at a zero DC offset crossing (minimum tick point). The combination of the software player, and the DAC usually mutes any tick present in a DSD bit stream, and may, or may not work so effectively on FLAC files.

 

Mnauta, if you will PM me with your mail address, I'll create a FLAC at the sample rate you specify of the nativedsd DSD edited master with much reduced, or eliminated inter track ticks, and send it to you via Dropbox.

[Axiom05]

All Channel Classics recordings from 1999 onward were recorded in DSD. All DSD bit streams contain a DC offset component, specified for SACD production as being no greater than -50dBfs. That's about several millivolts in consumer 2 Volt full scale system usage. Many (depending when the conversion took place) Channel Classics DSD edited masters were converted to PCM (FLAC) for downloads using the PQ markers contained in the DSD edited master file. That's the text information that drives the SACD player display when played as an SACD, and also allows player entry at a track start. The timing/positioning of these PQ markers has no relevance to the instantaneous DC offset level, only to a convenient time to change the SACD player track information.

 

Channel Classics recordings sold on nativedsd.com originate from those same DSD edited masters, but each track slicing point was manually chosen to be about one second prior to music, and at a zero DC offset crossing (minimum tick point). The combination of the software player, and the DAC usually mutes any tick present in a DSD bit stream, and may, or may not work so effectively on FLAC files.

 

Mnauta, if you will PM me with your mail address, I'll create a FLAC at the sample rate you specify of the nativedsd DSD edited master with much reduced, or eliminated inter track ticks, and send it to you via Dropbox.

 

Tom, thanks for this explanation (I wish I fully understood it all). Are you saying that the ticks are actually in the original DSD file and that the nativedsd files have been prepared to minimize or reduce their audibility? If so, preparing the PCM conversions from the nativedsd files would provide a better result than what is used on the Channel Classics site? Will the CD layer on the SACD also have these ticks?

[tailspn]

Hi Axion,

 

Essentially yes to your questions about the tick origin, and the nativedsd file prep.

 

What's in any DSD file is a varying bit width pulse stream who's pulse width is proportional to the digitized input analog signal level, along with a resulting DC offset from the modulation process, plus any DC offset in the A/D converter.

 

A DSD edited master, from which the SACD DST cutting master originates, is a continuous (~2.8GB stereo, 5.8GB surround) file. When played as a SACD, there are no interruptions in the data flow between tracks. There's just the player's display displaying the metadata including the Table Of Contents etc. The inherent DC offset in the DSD bit stream is never perceived within its specified -50dB limit.

 

For selling as a download, the DSD file needs to be broken into tracks, to emulate how it's played out in a SACD player. Now the DC offset is perceived as a tick, both at the end of the ending track, and the beginning of the next, with the resulting analog signal level abruptly going form the DC offset value to zero, then back to the DC offset value. This happens unless the file is sliced with no eliminated bits (controlled by the editor used), the independent track files are played consecutively, and the combination of software player and DAC either does not interrupt the bit flow from track to track (through a buffered lookahead), or bridges the the gap through muting. Obviously there are many variables.

 

As discussed in these pages, there are several methods available for track slicing of DSD files for DSD delivery. For DSD file preparation of the Channel Classics catalog, I chose employing a manual non invasive method. This involved auditioning every track slice point, searching for the time point where the varying DC offset level (tick) is at a minimum, or non existent. This is different than the PCM file prep process, which used the embedded PQ marker's time point used by the SACD player to drive its display.

 

Will the Channel Classics PCM FLAC files have greater tick levels than the same DSD file from nativedsd? It depends, for the DSD version has had much less data conversion applied, and is more susceptible to having the ticks heard. It also depends on the amount of offset originally in the recording, determined by the A/D converters used at the time of recording. In general though, using the same time points for the slice, the PCM version typically has less of a perceived tick problem. The CD layer however is authored in a completely different process, with much filtering that seems to eliminate the issue.

 

I prepared a 96KHz FLAC stereo file of the mentioned in this thread Channel Classics Christmas Oratorio for Mnauta using the nativedsd file, and he promised me a review here.

[jazz01]

Hi tailspn - thanks a lot for your explanation - makes it a much clearer to me. I understood that there are numerous reasons at the end why we can hear the clicking noise between DSD tracks on same files, but question: would it be fair to say that the label or the person whoever remastered the files could have been done better/did a second best job?

[mnauta]

After some communication with Channel Classics, they tried very diligently to address my concern with regard to ticks between track transitions on the Bach Christmas Oratorio download. I commend their effort in trying to provide the best possible product within the constrains of software capabilities. As I mentioned, I recently downloaded my second Channel Classics download, the Podger / Vivadi, and it was perfect musically and production wise.

[tailspn]

...question: would it be fair to say that the label or the person whoever remastered the files could have been done better/did a second best job?

 

Hi Jazz,

 

Fair question. If by that you are asking if the DSD edited master could have been sliced and converted to PCM better using a different tool set and process, then we'd have to know which tools and processes were available at the time the DSD files were sliced and converted. But "better" is somewhat subjective. The Weiss Saracon converter used by Channel Classics at the time, and even today, is arguably the best sounding, most transparent format converter available, even bettering the converter within Pyramix according to Channel. Its drawback is it has no associated capability to seamlessly slice a DSD edited master file, as does the Pyramix (provided an acceptable to Pyramix .lbm text file is associated). Saracon is a format conversion tool, Pyramix is a complete Digital Audio Workstation.

 

So as a mastering person, which trade-off do you choose; the better sounding format converter without DC offset suppression, or the more automated subjectively lesser sounding process with tick suppression process capability? Or do you as a producer rely on the consumer's software player to suppress the ticks? So many options. Recognize please also this entire issue is with a very small number of DSD releases recorded early in the life of DSD, where A/D converters were less concerned with DC offset nulling adjustment.

[Axiom05]

Hi Jazz,

 

Fair question. If by that you are asking if the DSD edited master could have been sliced and converted to PCM better using a different tool set and process, then we'd have to know which tools and processes were available at the time the DSD files were sliced and converted. But "better" is somewhat subjective. The Weiss Saracon converter used by Channel Classics at the time, and even today, is arguably the best sounding, most transparent format converter available, even bettering the converter within Pyramix according to Channel. Its drawback is it has no associated capability to seamlessly slice a DSD edited master file, as does the Pyramix (provided an acceptable to Pyramix .lbm text file is associated). Saracon is a format conversion tool, Pyramix is a complete Digital Audio Workstation.

 

So as a mastering person, which trade-off do you choose; the better sounding format converter without DC offset suppression, or the more automated subjectively lesser sounding process with tick suppression process capability? Or do you as a producer rely on the consumer's software player to suppress the ticks? So many options. Recognize please also this entire issue is with a very small number of DSD releases recorded early in the life of DSD, where A/D converters were less concerned with DC offset nulling adjustment.

 

Tom, thank you for your input to this discussion, I for one appreciate that you take the time to help educate us. Also, please don't think that this is an attack on Channel Classics as this is certainly not my intention in my postings on this issue.

 

Regarding your comment about early DSD recordings, I have the Fischer Mahler 9th (192/24) which is a pretty recent recording. Lovely recording but it does have the pops/ticks between some of the tracks, as you and I have previously exchanged emails about, so your comment confuses me a bit. I would still love to have a pop-free PCM version of this.

[tailspn]

Regarding your comment about early DSD recordings, I have the Fischer Mahler 9th (192/24) which is a pretty recent recording. Lovely recording but it does have the pops/ticks between some of the tracks, as you and I have previously exchanged emails about, so your comment confuses me a bit. I would still love to have a pop-free PCM version of this.

 

Well, the choices from the same original DSD edited master are slice points chosen through the PQ marker positions, then converted with Weiss Saracon to PCM FLAC, or slice points chosen manually by listening, then PCM converted with Pyramix Album Publishing. If you'd like to try the latter, send me a PM of your mail address, and I'll make the sliced file for you. I didn't do the Channel PCM conversion, so for other than saying how it was accomplished, I can't comment further than I have in previous posts.

[jazz01]

Hello tailspn - thanks again so much for your explanation. For me, it is just a hobby, I enjoy much listening to highres music and getting a little more insight about the complexity behind is really very interesting. I was wondering why some files have clicking noise and other not, but now I know :-)

[tailspn]

I was wondering why some files have clicking noise and other not, but now I know :-)

 

Hi Jazz

 

Frankly, I wish I knew all the factors that cause one recording to be easy to slice clickless, and others to be almost impossible. As mentioned earlier, virtually all DSD recorded projects were/are intended for SACD delivery. The files, be they stereo or multi-channel are the DSD edited master, converted to DST format for any necessary lossless compression required to fit the roughly 8GB of data (stereo plus multi-channel content) into the 4.7GB available space of the SACD layer. Also, the DSD modulation process has a DC component which is both the DC offset of the A/D converter, as well as a program amplitude generated offset that goes both above and below zero level.

 

The DSD edited master (the SACD player drive converts the DST encoded SACD back into the DSD edited master losslessly) is played without interruption as one continuous file, and/or is muted by the player if started or stopped within the file.

 

However, none of that applies if played as a downloaded DSD edited master file, for there are no track separation indicators within the DSD bit stream itself. So the continuous DSD file has to be sliced into separate individual track files that are then assembled when played by the software player and DAC. The problem arises when transitioning from the end of one file (track) to the beginning of the next. They're independent separate DSD files, so the player transitions from whatever instantaneous DC offset exists at the track end abruptly to zero level, then with whatever gap avoidance routine existing in the player's amount of time, from zero level to the DC offset level of the new track.

 

If the software player is truly gapless, that is the next track beginning is already sitting in a buffer before the end of the previous track, and the union occurs on the very next clock/bit time, then a recording slice that did not drop bits when edited at the slice time would stream from the same DC offset level at the slice point, producing no click. Lots of if's!

 

Today, the majority of recordings I receive for distribution through nativedsd.com come already sliced by whatever Digital Audio Workstation the label employed, which is primarily Pyramix. There are a number of ways of separating DSD edited masters into tracks both in Pyramix, and stand alone slicing apps, each with their tradeoffs and associated support file requirements. But by in large, the slicing process has now been resolved such that few people notice any track transition noises today at normal listening levels.

[Axiom05]

Tom, thank you for your input to this discussion, I for one appreciate that you take the time to help educate us. Also, please don't think that this is an attack on Channel Classics as this is certainly not my intention in my postings on this issue.

 

Regarding your comment about early DSD recordings, I have the Fischer Mahler 9th (192/24) which is a pretty recent recording. Lovely recording but it does have the pops/ticks between some of the tracks, as you and I have previously exchanged emails about, so your comment confuses me a bit. I would still love to have a pop-free PCM version of this.

 

I want to thank Tom (tailspn) for helping me with this issue. Tom has supplied me with a pop-free version of the Mahler 9th. I appreciate the time that he spent on making this possible. I must say that I really enjoy this performance and the recorded sound is excellent. Thanks Tom!