Weiss Saracon

Bruce B

WBF Founding Member, Pro Audio Production Member
Apr 26, 2010
6,738
127
63
Seattle, WA
www.pugetsoundstudios.com
#1
Weiss Saracon is a software program for sample rate conversion. The program handles every known sample rate all the way to 384kHz in PCM and DSD128fs. What we found out over the weekend was an eye-opener.

When converting DSD->PCM, Saracon puts a very steep filter starting around 22k and dropping over 50dB IN ONE OCTAVE. By the time it gets to 40k, there is NOTHING! You have no option for different filter settings like other programs.

This means that when converting DSD files to PCM in Saracon, THERE IS NO SUCH THING AS ANYTHING GREATER THAN 24/96!

Below are 2 files at 88.2 and 176.4 THEY ARE EXACTLY THE SAME!
 

Attachments

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,431
21
38
Seattle, WA
www.genesisloudspeakers.com
#5
I guess that the question I really wanted to ask was - DSD -> PCM isn't transparent?

I wonder how much music was originally recorded in DSD and converted to PCM for release (either as redbook, DVD-A, or high-rez downloads) and we are hearing the conversion.....
 

RBFC

WBF Founding Member & Super Moderator
Apr 20, 2010
5,127
7
38
Albuquerque, NM
www.fightingconcepts.com
#6
I guess that the question I really wanted to ask was - DSD -> PCM isn't transparent?

I wonder how much music was originally recorded in DSD and converted to PCM for release (either as redbook, DVD-A, or high-rez downloads) and we are hearing the conversion.....
This question mirrors my inquiry into the Korg Audiogate software for DSD playback. With an intermediate conversion stage, I don't see how it can be as transparent as a direct DSD>analog conversion.

Lee
 

Bruce B

WBF Founding Member, Pro Audio Production Member
Apr 26, 2010
6,738
127
63
Seattle, WA
www.pugetsoundstudios.com
#7
I guess that the question I really wanted to ask was - DSD -> PCM isn't transparent?

I wonder how much music was originally recorded in DSD and converted to PCM for release (either as redbook, DVD-A, or high-rez downloads) and we are hearing the conversion.....
The conversion from DSD to PCM is transparent using the correct methods, sample rates and bit depths.

The problem is now everyone is archiving tape/vinyl in DSD. If they are using Saracon to make hi-rez PCM files, they're not gaining anything.
 

weiss2496

New Member
Jun 25, 2011
4
0
0
#8
Here are the same 2 files using a different program.
There are various "philosophies" when designing a sampling rate converter. Our Saracon makes sure that there is no aliasing occurring and also that the very high level of RF noise present in a DSD signal is not present in the PCM signal. One has to be aware that the noise floor in a DSD signal is heavily shaped, which means that the noise energy above about 22 kHz rises steeply. While audio signals above 22kHz probably are fairly weak... i.e. mostly swamped by noise.

The "different program" mentioned above obviously allows for aliasing.

Daniel
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,591
34
48
Monument, CO
#10
Hmmm... Interesting, but I would appreciate less OOB (out-of-band) energy; saves my tweeters and I can't hear it (I know there are all sorts of arguments on the audibility of ultrasonics). Not too mention (so I shall ;) ) less risk of exciting an unstable amplifier into ringing (or worse). All in life, and in filters, is a compromise...

I still get tripped up by "hi-res" = higher resolution and higher sampling rate; I am used to those being independent (or rather inversely dependent, as it is harder to achieve higher resolution at higher sampling rates, but as specifications they are independent). I understand folk consider higher sampling rates as providing higher time resolution, just not the way I was weaned. :)

Bruce, are there any negative audible effects from the Saracon filters? Or, any benefit from those programs that do not filter the ultrasonics?

Curious (as always) - Don
 
Apr 3, 2010
15,820
1
0
Seattle, WA
#12
Late comer to the thread but I have to go with Daniel's view here. We can see the climbing power in higher frequencies which tells us that it is due to noise shaping. SACD process creates incredible amount noise in the audible spectrum. The encoding then, shifts that to higher frequencies above 20 Khz that is supposed to be inaudible. Since we can then no longer differentiate between what is signal and what is bunched up quantization noise in the ultrasonic range, I don't see an issue with their decision to apply a gentle roll off.

Put another way, one-bit DSD is a one-way encoding. Backing out its effect and getting back to PCM is impossible.

But sure, having that programmable and letting the operator pick how much of that noise+signal should be preserved is a good feature request.
 

Vincent Kars

WBF Technical Expert: Computer Audio
Jul 1, 2010
860
0
0
#13

RBFC

WBF Founding Member & Super Moderator
Apr 20, 2010
5,127
7
38
Albuquerque, NM
www.fightingconcepts.com
#15

Orb

New Member
Sep 8, 2010
3,010
0
0
#16
A company like Channel Classics records in DSD.
They use the Weiss software to convert to FLAC



If Bruce is right, buying the 24/192 is a waste of money.
http://www.channelclassics.com/high-resolution-audio-downloads/dijkstra-27108.html
Not disagreeing but this is strange as any detailed editing (used to anyway) would require them to convert from DSD to PCM and then back to DSD again.
From what I understand this was one of the reasons Linn moved away from their native DSD recording studio setup ages ago and to hirez PCM.

That said as I mentioned in some other threads it seems there may be a bit of luck from the consumer end to whether the hirez track they purchased is done correctly and has high bandwidth information or just noise.
Keith Howard developed some software to enable accurate analysing of what is recorded in an easy to read way; shows both highest amplitude and average over the whole frequency so any issues with the filter or any noise show up.
From his experience around 40% of those he purchased (I admit it was not many albums and he states that) were not true hirez tracks (this was not Linn, although he also gave example how even they had to remove a whole batch of tracks that they sold for another studio that were not true hirez).
So, if someone is into hirez tracks then TBH I would look to use some software to validate it and never rely on the company selling as even they can be caught out by other 3rd party studios whose albums-tracks they also sell - several hirez companies been caught out this way but one would hope they now have safeguards in place for their own validation.

Cheers
Orb
 
Last edited:

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,591
34
48
Monument, CO
#17
Higher sampling rates do not help with time resolution. Wordlength (i.e. SNR) is important for time resolution.

Daniel
Uh, I think we said the same thing? Though I am not sure I completely agree, as it depends somewhat on the source and the system... We want higher rates for wider bandwidth in RADAR, LIDAR, and other pulse systems, as well as ELINT systems etc., and in those cases a faster sampling ADC can provide greater time resolution (in terms of being able to accurately resolve smaller time differences). In those systems, the edge matters, and a 5 GS/s 8-bit is better than a 100 MS/s 16-bit ADC for precisely placing a pulse edge in time.

Wordlength and resolution are only quasi-related in my world, due primarily to two factors:
1. The system "wordlength" is a term usually used by the DSP guys and may not correlate with what the ADC/DAC can do; and,
2. Actual resolution, defined by the IEEE as ENOB although we use SNR. SINAD, and SFDR as well (among other things), is not always tightly coupled to the number of bits (specified resolution) of the converter. In the audio world (and beyond) there are plenty of cases of lower-resolution converters outperforming higher-resolution devices.

Note I define things from an engineering background dealing with wideband RF/mW/mmW systems, not (always) as an audiophile...

In any event, this is far afield of the original topic, and I've had enough debating for one week, so I'll back out. - Don
 

Bruce B

WBF Founding Member, Pro Audio Production Member
Apr 26, 2010
6,738
127
63
Seattle, WA
www.pugetsoundstudios.com
#18
If Bruce is right, buying the 24/192 is a waste of money.
Yes, I feel if a company that is using Saracon to convert DSD->PCM, then anything over 24/88.2 is a waste of money, even if it IS just noise.

The "different program" mentioned above obviously allows for aliasing.
Daniel
The "different program" is a Pyramix apodizing filter.
 
Last edited:

Bruce B

WBF Founding Member, Pro Audio Production Member
Apr 26, 2010
6,738
127
63
Seattle, WA
www.pugetsoundstudios.com
#19
Late comer to the thread but I have to go with Daniel's view here. We can see the climbing power in higher frequencies which tells us that it is due to noise shaping. SACD process creates incredible amount noise in the audible spectrum. The encoding then, shifts that to higher frequencies above 20 Khz that is supposed to be inaudible. Since we can then no longer differentiate between what is signal and what is bunched up quantization noise in the ultrasonic range, I don't see an issue with their decision to apply a gentle roll off.
Amir... 50-60dB/octave is not a gentle roll off. We have done extensive listening and feel it clouds the music/image. We have more "gentle" filters we use that sound more like the original source.
 

Attachments

Jul 1, 2010
8,678
1
0
#20
Jaded old geezer that I am, I yawn in the face of this stuff. I can't hear much above 12khz (and neither can most of the rest of you geezers), and what's there is noisy as hell, so if I could hear it, I wouldn't want to. I should thank Daniel for filtering it out. Thanks, Daniel.

Tim

PS: I can still hear a tube mic preamp and an old Gibson Jumbo on a decent recording, though, so I'm good. :)
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. A place where audiophiles and audio companies discuss existing and new audio products, music servers, music streamers and computer audio, digital to audio convertors (DACS), turntables, phono stages, cartridges, reel to reel, speakers, headphones, tube amplifiers and solid state amplification. Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing