Jump to content
IGNORED

AuI ConverteR settings


mfaoro

Recommended Posts

Hello All,

 

I have a couple terabytes of DFF files that I need to convert to FLAC (my Linn DSM does not support DSD).

 

After a good bit of research, I have settled on performing the conversions using AuI ConverteR. I am interested in the best sound quality. If you know of a better tool re: sound quality then I am all ears. (edit: I use a mac).

 

I am not super-technical when it comes to DSD or these conversions. I have read a lot of threads here about these conversions but most them evolved into technical details that I did not understand.

 

So, I was wondering: given the settings available in AuI ConverteR, what should I change and what should I leave alone?

 

I don't care about things like file size and I am more than happy to let a computer churn at these files for days on end.

 

Thank in advance for any advice you can share.

 

Michael

FRONT END: Analog: Radikal Linn LP12 > Linn Urika 2 phono stage. Sound: Linn Klimax Organik DSM > Linn Duo amp >Maggie 3.7i  Wires + Power: Transparent: Reference Speaker, XL Power Conditioner + XL Power Cords. Furutech NFC Rhodium outlet on 10 gauge dedicated circuit with isolated ground Isolation: HRS SXR stand, M3X2 Bases. 
 

Connected to back end by: Transparent Ethernet 

 

BACK END: Digital: Internet > OpticalModule > EtherREGEN < AD Queen Squarewave Clock < Roon Nucleus + (internal 7TB SSD music library) Isolation: Salamander Archetype rack, HRS M3X2 base the under Nucleus, ER,Stillpoints under all others Power: Paul Hynes SR7T > Clock, Nucleus. SR7T > ER & OpticalModule, SR4 > Switch. Furutech NFC Rhodium outlet on 10 gauge dedicated circuit with isolated ground 

 

Link to comment

I have a license for AUIConverter and use it now and again.

 

For batch processing, the version I have falls over and there's no progress bars, and that's just a couple of DSD albums converting DFF to DSF....

When creating Redbook files (for burning to CD) from Hires FLAC, the SRC misses out the last 2kHz to 20KHz, wheras Izotope does not (using Sony Sound Forge). For these reasons Use the Trial version for AUI and check it out thoroughly before committing.

I have a lot of trouble understanding the author, some Google Russian-English translation difficulty perhaps, certainly not transparent.

Unfortunately there is a lot of technobabble which is over my head, also leading to frustration in getting AUI working. The GUI is wanting a polish!

 

I don't see a computer in your system, a NAS which hasn't a hope in hell of performing DSD to PCM on the fly. Online conversion perhaps with a hardware device may be more of the way to go, say a Microrendu and a DDC that outputs 176.4 to S/PDIF Coax to the DSM.

 

dbpoweramp has an encoder for DSD, but no decoder...and it's only for Windows.

 

Other than that I can't think of anything given the streaming setup that's there now. Don't complicate life with a computer that does the processing, I gather you've been there and done that.

AS Profile Equipment List        Say NO to MQA

Link to comment

Thanks 1 1/2.

 

I have, as you said, been there and done that. Highly-modified Mac Mini running a skeleton OS on an SSD etc.. serving music to an ARC DAC8. Got sick of the lack of stability and all the fiddling. Really turned me off to computer audio for a while. The Linn DSMs have renewed my interest somewhat. That said, I would still rather listen than fiddle.

 

I am not a big fan of 'on the fly' conversion. I can't imagine the kind of computer one would need to do these conversions without sacrificing sound quality. I know I don't want it in my listening room!

 

Re: the source misses out the last 2khz to 20 kHz.

How do I tell if AUIConverter still has this issue? Look at the spectrum? Here is what I see in Adobe Audition for a FLAC that I converted using AUI:

 

Screen Shot 2016-12-16 at 7.56.51 PM.jpg

 

These spectrum graphs don't tell me much. Yet.

 

The free version of AUIConverter that I have been testing has a progress indicator and has handled converting two albums at a time without falling over (at least as far as I can tell.

 

Michael

FRONT END: Analog: Radikal Linn LP12 > Linn Urika 2 phono stage. Sound: Linn Klimax Organik DSM > Linn Duo amp >Maggie 3.7i  Wires + Power: Transparent: Reference Speaker, XL Power Conditioner + XL Power Cords. Furutech NFC Rhodium outlet on 10 gauge dedicated circuit with isolated ground Isolation: HRS SXR stand, M3X2 Bases. 
 

Connected to back end by: Transparent Ethernet 

 

BACK END: Digital: Internet > OpticalModule > EtherREGEN < AD Queen Squarewave Clock < Roon Nucleus + (internal 7TB SSD music library) Isolation: Salamander Archetype rack, HRS M3X2 base the under Nucleus, ER,Stillpoints under all others Power: Paul Hynes SR7T > Clock, Nucleus. SR7T > ER & OpticalModule, SR4 > Switch. Furutech NFC Rhodium outlet on 10 gauge dedicated circuit with isolated ground 

 

Link to comment
Thanks 1 1/2.

 

I have, as you said, been there and done that. Highly-modified Mac Mini running a skeleton OS on an SSD etc.. serving music to an ARC DAC8. Got sick of the lack of stability and all the fiddling. Really turned me off to computer audio for a while. The Linn DSMs have renewed my interest somewhat. That said, I would still rather listen than fiddle.

 

I am not a big fan of 'on the fly' conversion. I can't imagine the kind of computer one would need to do these conversions without sacrificing sound quality. I know I don't want it in my listening room!

 

Re: the source misses out the last 2khz to 20 kHz.

How do I tell if AUIConverter still has this issue? Look at the spectrum? Here is what I see in Adobe Audition for a FLAC that I converted using AUI:

 

[ATTACH=CONFIG]31603[/ATTACH]

 

These spectrum graphs don't tell me much. Yet.

 

The free version of AUIConverter that I have been testing has a progress indicator and has handled converting two albums at a time without falling over (at least as far as I can tell.

 

Michael

 

Here's a MusicScope report of a 96 file SRC to Redbook, for CD burning using AUI

 

01 Jo.wav_AUI version_report.png

 

The same file, converted to Redbook using Sony Sound Forge with Izotope SRC, note the above zero for 20kHz content.

 

01 Jo.wav_report.png.png

 

Fully appreciate the reasoning for streaming and more of an appliance. I did look at the Lumin T1 and Aurender A10 and decided the DSD ability rested with a computer better than the aforementioned,to a disappointment. That workhorse is in the garage, well away from the listening area, so not a real problem.

 

The DSM looks like a great machine, the DSD no go zone is problematic in more ways than one. Maybe Yuri will chime in and give me a roasting :)

AS Profile Equipment List        Say NO to MQA

Link to comment

Hi Michael,

Hi One and a half,

 

Michael, thank you for interest to our converter software.

One and a half, thank you for your reply.

I prefer to get any type of feedback, bug reports, etc., because this way light issues that may be fixed.

 

For batch processing, the version I have falls over and there's no progress bars, and that's just a couple of DSD albums converting DFF to DSF....

The GUI is wanting a polish!

The free version of AUIConverter that I have been testing has a progress indicator and has handled converting two albums at a time without falling over (at least as far as I can tell.

 

I suppose, in latest versions no same issues. In 2016 we fixed several interface issues for hanging avoiding.

Periodically new issues appear. We also work under it.

 

When creating Redbook files (for burning to CD) from Hires FLAC, the SRC misses out the last 2kHz to 20KHz, wheras Izotope does not (using Sony Sound Forge).

 

There are alternatives:

1. Slope filter with keeping 20…22 kHz, but with artefacts.

2. Sharper filter in lossing smallest frequency range about 22 kHz, but more ringing.

3. Сompromise: Deep suppressing in 20 … 22 kHz (due hearing border is 16 …20 kHz). There are artefacts about quantization noise (about -170 dB), ringing lesser comparing sharper filter.

 

For sample rates upper 44 kHz may be applied non-optimised filter.

There is output band cut like traditional PCM resampling: close to value of minimal sample rate (input and output) divided 2 times.

For converting DSF/DFF/ISO to PCM cut from 24 to 27 kHz (depend on values of input and output sample rates).

 

There are possible several directions for better sound achieving:

1. Sample rate

2. Bit depth

3. Linear phase/minimum phase filter

4. Optimised/Non-optimised mode

 

Minimum phase filter is compromise decision, because ringing energy same to linear phase, but after front of signal.

 

Converter can’t improve quality of audio file as itself.

But hardware can provide different levels of distortions for different modes (sample rate + bit-depth).

 

Thus we can found «best sounding» mode (combination of points 1.-4.) and use it.

 

As default AuI use the best mode (Optimised + Linear filter) in my opinion. But it is not constant principle for anybody.

AuI ConverteR 48x44 - HD audio converter/optimizer for DAC of high resolution files

ISO, DSF, DFF (1-bit/D64/128/256/512/1024), wav, flac, aiff, alac,  safe CD ripper to PCM/DSF,

Seamless Album Conversion, AIFF, WAV, FLAC, DSF metadata editor, Mac & Windows
Offline conversion save energy and nature

Link to comment
Hello All,

 

I have a couple terabytes of DFF files that I need to convert to FLAC (my Linn DSM does not support DSD).

 

After a good bit of research, I have settled on performing the conversions using AuI ConverteR. I am interested in the best sound quality. If you know of a better tool re: sound quality then I am all ears. (edit: I use a mac).

 

I am not super-technical when it comes to DSD or these conversions. I have read a lot of threads here about these conversions but most them evolved into technical details that I did not understand.

 

So, I was wondering: given the settings available in AuI ConverteR, what should I change and what should I leave alone?

 

I don't care about things like file size and I am more than happy to let a computer churn at these files for days on end.

 

Thank in advance for any advice you can share.

Michael

I gather you are only interested in converting DFF to FLAC, so I will limit my comments to that particularly process, on a mac. I use AUIConverter regularly (current version), but I have never tried to batch terabytes of DFF files at once - only a few albums at a time. It is very stable when batching a few albums at a time. It does have progress indicators.

 

There are not many settings that need to be changed. When downsampling, the developer believes very strongly in limiting the maximum frequency to 20kHz, irrespective of the sample rate you want to convert to. For me, I don't agree with that. Fortunately, there is an option to chop it at around 26 kHz instead, by setting Filter Mode to "non-optimized resampling filter". That one is important for me.

 

Under the metadata tab in settings, check "create subdirectories as performer/year" etc. That way each album will be given its own output subdirectory instead of having a mass of files in one output directory.

 

There are many options for reducing or eliminating ticks or pops between songs, that sometimes happen with some converters. I have never found the need to use them, I would suggest not touching them, and listening to a few converted albums. If you find that there is a problem (particularly on a gapless album), go back and experiment with the relevant settings.

 

One of the most important things when batching DFF or SACD ISO to FLAC is peak normalization. This should be done on a "per album basis". You can set up AUIconverter to do this by going to the "Levels" tab. For "Levels ajusting", select "to max peak" and "by directory". Then make sure each album is in its own directory. Each album will then have the level optimized individually to a peak album level of 0dB.

 

If I am converting from DFF to 24bit flac, I switch off dithering, If I am converting to 16 bit, I switch on dithering.

 

That's about it. I don't touch any other settings. I will admit that the english (both in the app and in the manuals/guides) is not easy to follow. But I have not found it to be a real issue for me. The purchase options are very confusing.

 

In terms of other software for macs, I like DSDMaster. It is simple and the sound quality is very good (equal to AUIconverter). But there is a disadvantage for batching multiple albums. Peak normalization can either be done per song, or per batch of files being converted. So if you drag multiple albums into the app for conversion, they will be normalized as one large album. AUIconverter can handle each album separately.

 

Paul

Link to comment

Hi Paul,

 

Thank you for so detailed review.

 

I will admit that the english (both in the app and in the manuals/guides) is not easy to follow. But I have not found it to be a real issue for me.

 

I permanently work under it. It is really very important.

Currently manuals don’t cover all questions what I get.

First thing now is covering of wider frequently questions area and systematisation of knowledgebase.

 

 

Fortunately, there is an option to chop it at around 26 kHz instead, by setting Filter Mode to "non-optimized resampling filter". That one is important for me.

 

Especially for DSD64 and 128 after 20 kHz noise floor rapidly growing with frequency growing (depend of DSD modulator).

1-2 years ago, when I developed non-optimised mode, planned point of limitation was 30 kHz.

But when I implemented it, I was disappointed in my hypothesis.

Noise level at higher frequencies was close to CD’s one.

 

Each excessive ultrasound noise can cause noise in audible range due intermodulation distortions.

 

On of similar cases see on video

 

But, may be, will applied "super wide" option in the future.

 

The purchase options are very confusing.

 

Absolutely. Saving of customer’s money is single goal of existing severals editions and sale-per-module.

Though maximal PROduce-RD is not most expensive software among pro conversion ones.

 

One of my permanent minds is how achieve balance «lower price - easy choice».

But, unfortunately, though there nothing doing so very successfully.

I hope, I will getting insight, how to right implement these things.

 

In terms of other software for macs, I like DSDMaster. It is simple and the sound quality is very good (equal to AUIconverter).

 

Here comparison of DSD decoders Archimago's Musings: ANALYSIS: DSD-to-PCM 2015 - foobar SACD Plug-In, AuI ConverteR, noise & impulse response...

 

Unfortunately, there aren't DSDMaster and some other DSD decoders.

Also it would be interesting to read about comparison DSD encoders.

 

Best regards,

Yuri

AuI ConverteR 48x44 - HD audio converter/optimizer for DAC of high resolution files

ISO, DSF, DFF (1-bit/D64/128/256/512/1024), wav, flac, aiff, alac,  safe CD ripper to PCM/DSF,

Seamless Album Conversion, AIFF, WAV, FLAC, DSF metadata editor, Mac & Windows
Offline conversion save energy and nature

Link to comment

Yuri -

A little while ago there was discussion of a manual that would be written by a native english writer with the appropriate technical knowledge. Did anything ever come of that?

 

All of my conversions are flac to dsd and the default settings have yielded satisfactory results. No complaints at all. But I must admit that the recent writings on clicks have gone unread because of the translations.

 

Again, this is not a complaint. I am a very satisfied customer.

 

"The function of music is to release us from the tyranny of conscious thought", Sir Thomas Beecham. 

 

 

Link to comment

Hi Augustine,

 

Thank you for your reply about my software here.

 

A little while ago there was discussion of a manual that would be written by a native english writer with the appropriate technical knowledge. Did anything ever come of that?

 

We began this work. There text was corrected (thanks for Sdolezalek and SwissBear), but when I began read the manual I found many things that is obsolete in current versions. So I temporary stop correcting of the manual. There need fix obsolete information, remove parts that duplicated at site and sent paper for correction again.

 

At the AuI ConverteR 48x44 - HiEnd Audio Converter ISO DFF DSF WAV FLAC AIFF currently placed many separate stuff about audio quality.

But I have a few time to systematise it.

 

Of course, I understand that the user manual is very important part of the software.

And I will correct the manual. But I won’t promise exact term due I don’t like empty promises.

 

 

But I must admit that the recent writings on clicks have gone unread because of the translations.

 

I suppose, in my case main reason of translation lost is different language patterns (Russian and English).

 

For last translations I use online English-Russian phrase context dictionary rather than Google.

 

I thought, that my language in the article more clearer than before. But looks like that need work there more an more.

 

Thank you for reply about the article. I suppose, next 4th part will final (if I don’t get new information).

 

Now re-reading of the article is planned to correct it for better understanding.

 

The article is part of the user manual too.

 

I suppose, in the manual need more pictures less words.

 

Best regards,

Yuri

AuI ConverteR 48x44 - HD audio converter/optimizer for DAC of high resolution files

ISO, DSF, DFF (1-bit/D64/128/256/512/1024), wav, flac, aiff, alac,  safe CD ripper to PCM/DSF,

Seamless Album Conversion, AIFF, WAV, FLAC, DSF metadata editor, Mac & Windows
Offline conversion save energy and nature

Link to comment

Especially for DSD64 and 128 after 20 kHz noise floor rapidly growing with frequency growing (depend of DSD modulator).

1-2 years ago, when I developed non-optimised mode, planned point of limitation was 30 kHz.

But when I implemented it, I was disappointed in my hypothesis.

Noise level at higher frequencies was close to CD’s one. Each excessive ultrasound noise can cause noise in audible range due intermodulation distortions.

 

On of similar cases see on video

 

But, may be, will applied "super wide" option in the future.

 

My limited understanding of the disadvantages of CD/Low-Res audio is that there are audible compromises due to resampling. CDs need to be resampled to 22 kHz. AuIConverter's cutoff is even lower at 20 kHz. If there are any artifacts/distortions due to resampling, it seems to me that they might occur in the audible frequency range and perhaps affect the sound quality.

 

My main use of AuIConverter has been to convert DSD to PCM, but I use it for other things as well. My personal, unscientific feeling is that your "non-optimized" cut-off (around 25-26 kHz) is about right, for converting DSD to PCM. Anything higher than that (super-wide option) has doubtful benefits. Anything lower than that, and the distortions caused by resampling are getting too close to the audible frequency range... in my opinion. That's why I choose to use the "non-optimized" filter option. I am not really comfortable resampling hi-res DSD files down to 20 kHz when converting to PCM.

 

As with rodrigaj, I am a very satisfied customer. It's my "swiss army knife" for audio conversions. Just expressing my view of the two available filter options.

 

Regards

Paul

Link to comment

Hi Paul,

 

AuIConverter's cutoff is even lower at 20 kHz.

 

In range 0 … 20 kHz filer provide flatness significantly lesser 1 dB (filter calculated for 1e-6 dB).

After 20 kHz filter go to -200 dB (calculated value) at 22050 kHz or other (depend on sample rates).

See pictures at SRC Comparisons

For estimations sweep sines used in 32-bit float point bit-depth.

 

If there are any artifacts/distortions due to resampling, it seems to me that they might occur in the audible frequency range and perhaps affect the sound quality.

 

1. Resampling artefacts are distributed across full spectrum of output file. The artefacts must suppressed via filter that cut above 1/2 sample rate (minimal between input and output ones).

 

2. Ultrasound noise (even original, non distorted) don’t cause distortions for playback on ideal (without non-linear distortions) hardware.

Because we can’t hear ultrasound by current theory and experiments. May be in the future we will have other knowledges.

 

But we have hardware with non-linear distortions. When signal pass through the hardware there appear combinations of signal’s components.

I.e. new harmonics. It is distortions. Combinations of ultrasound components (intermodulation distortions) may appear in audible range.

 

Especially dangerous significant level sines into ultrasound. One my customer suggested that it is products of recording apparatus.

 

Optimised filter used for intermodulation distortions only.

 

For test of apparatus (ultrasound intermodulations) should be used pair of ultrasound sines.

 

As example, we have 33 and 35 kHz test tones.

We can get products at 2 kHz = 35 - 33 kHz. Or 4 kHz=35*2-33*2. Etc.

 

Amplitude of the products depend on amplitudes of test tones.

 

As with rodrigaj, I am a very satisfied customer. It's my "swiss army knife" for audio conversions. Just expressing my view of the two available filter options.

 

Thank you for kind words, Paul.

 

Best regards,

Yuri

AuI ConverteR 48x44 - HD audio converter/optimizer for DAC of high resolution files

ISO, DSF, DFF (1-bit/D64/128/256/512/1024), wav, flac, aiff, alac,  safe CD ripper to PCM/DSF,

Seamless Album Conversion, AIFF, WAV, FLAC, DSF metadata editor, Mac & Windows
Offline conversion save energy and nature

Link to comment

Thank you so much. This is very useful information to me.

 

One question for you and Yuri: Can FLAC compression be turned off completely? Is that what setting Flac Compression=Faster does?

 

I get radically different file sizes when converting the same file with AuI and DSDMaster (have set Flac Compression=Faster in AuI and Compression=none in DSDMaster). DSDMaster creates a 25% bigger file. Perhaps thats because it has data above 20k - will have to use your settings and see what happens.

 

<edit> I used your settings and the file is still ~25% smaller than the same file from DSDMaster.

 

Another question: Do you turn off dithering by setting Dithering Mask Depth...=Light and not hitting the button in the main Ui?

 

I gather you are only interested in converting DFF to FLAC, so I will limit my comments to that particularly process, on a mac. I use AUIConverter regularly (current version), but I have never tried to batch terabytes of DFF files at once - only a few albums at a time. It is very stable when batching a few albums at a time. It does have progress indicators.

 

There are not many settings that need to be changed. When downsampling, the developer believes very strongly in limiting the maximum frequency to 20kHz, irrespective of the sample rate you want to convert to. For me, I don't agree with that. Fortunately, there is an option to chop it at around 26 kHz instead, by setting Filter Mode to "non-optimized resampling filter". That one is important for me.

 

Under the metadata tab in settings, check "create subdirectories as performer/year" etc. That way each album will be given its own output subdirectory instead of having a mass of files in one output directory.

 

There are many options for reducing or eliminating ticks or pops between songs, that sometimes happen with some converters. I have never found the need to use them, I would suggest not touching them, and listening to a few converted albums. If you find that there is a problem (particularly on a gapless album), go back and experiment with the relevant settings.

 

One of the most important things when batching DFF or SACD ISO to FLAC is peak normalization. This should be done on a "per album basis". You can set up AUIconverter to do this by going to the "Levels" tab. For "Levels ajusting", select "to max peak" and "by directory". Then make sure each album is in its own directory. Each album will then have the level optimized individually to a peak album level of 0dB.

 

If I am converting from DFF to 24bit flac, I switch off dithering, If I am converting to 16 bit, I switch on dithering.

 

That's about it. I don't touch any other settings. I will admit that the english (both in the app and in the manuals/guides) is not easy to follow. But I have not found it to be a real issue for me. The purchase options are very confusing.

 

In terms of other software for macs, I like DSDMaster. It is simple and the sound quality is very good (equal to AUIconverter). But there is a disadvantage for batching multiple albums. Peak normalization can either be done per song, or per batch of files being converted. So if you drag multiple albums into the app for conversion, they will be normalized as one large album. AUIconverter can handle each album separately.

 

Paul

FRONT END: Analog: Radikal Linn LP12 > Linn Urika 2 phono stage. Sound: Linn Klimax Organik DSM > Linn Duo amp >Maggie 3.7i  Wires + Power: Transparent: Reference Speaker, XL Power Conditioner + XL Power Cords. Furutech NFC Rhodium outlet on 10 gauge dedicated circuit with isolated ground Isolation: HRS SXR stand, M3X2 Bases. 
 

Connected to back end by: Transparent Ethernet 

 

BACK END: Digital: Internet > OpticalModule > EtherREGEN < AD Queen Squarewave Clock < Roon Nucleus + (internal 7TB SSD music library) Isolation: Salamander Archetype rack, HRS M3X2 base the under Nucleus, ER,Stillpoints under all others Power: Paul Hynes SR7T > Clock, Nucleus. SR7T > ER & OpticalModule, SR4 > Switch. Furutech NFC Rhodium outlet on 10 gauge dedicated circuit with isolated ground 

 

Link to comment

Hi Michael,

 

Can FLAC compression be turned off completely? Is that what setting Flac Compression=Faster does?

 

I get radically different file sizes when converting the same file with AuI and DSDMaster (have set Flac Compression=Faster in AuI and Compression=none in DSDMaster). DSDMaster creates a 25% bigger file. Perhaps thats because it has data above 20k - will have to use your settings and see what happens.

 

AuI applied only compressed FLACs.

It may it with several degree of compression.

From «with minimal size» to «maximal speed of conversion to FLAC».

It is standard features of FLAC library.

At my computer I don’t see big difference in size of compressed files.

In speed of processing too :-)

 

We don’t applied uncompressed FLACs.

Because I have not seen objective proofs (measured figures) of quality losses due compression yet.

Even as increasing of noise at output DAC due higher CPU load spent to FLAC un-compression.

Binary content FLAC and WAV was compared experimentally

 

Even 2 FLACs compressed in one degree, there may be difference in size:

1. Paddind block (free space before audio data for writing metadata without re-writing full file).

2. Size of metadata, additional records there (what software compress as example).

3. Size of artwork, recompressing and/or resizing artwork.

 

As example, AuI may apply resizing artwork (adjustable). It need for some portable audio players, as example.

 

 

Another question: Do you turn off dithering by setting Dithering Mask Depth...=Light and not hitting the button in the main Ui?

 

Dithering may be used for:

1. Conversion to 16-bit resolution.

2. Conversion from 16 bit resolution.

 

«Dithering mask depth…» it is level of dithering noise.

 

For case #1 I recommend to use «Light» (minimal level) position always. It is standard situation.

 

Case #2 is not standard. Result depend on dithering applying to source 16-bit stuff during production.

Dithering noise mask (due higher level) non-linear distortions that present into such source.

In a musical signal we can’t exactly detect distortions visually.

So there may applied control by ears primarily.

 

Anyway I recommend to check case #2 just for effect.

 

If Dither lamp in main window is light - dither is applied.

 

If both input and output resolutions above 16 bit (1-bit considered here as above 16 bit too) dither auto OFF despite lighted lamp.

It need for batch conversion of files with mixed resolution.

 

More information about dithering What is dithering audio?

 

 

Best regards,

Yuri

AuI ConverteR 48x44 - HD audio converter/optimizer for DAC of high resolution files

ISO, DSF, DFF (1-bit/D64/128/256/512/1024), wav, flac, aiff, alac,  safe CD ripper to PCM/DSF,

Seamless Album Conversion, AIFF, WAV, FLAC, DSF metadata editor, Mac & Windows
Offline conversion save energy and nature

Link to comment
Thank you so much. This is very useful information to me.

I get radically different file sizes when converting the same file with AuI and DSDMaster (have set Flac Compression=Faster in AuI and Compression=none in DSDMaster). DSDMaster creates a 25% bigger file. Perhaps thats because it has data above 20k - will have to use your settings and see what happens.

 

Another question: Do you turn off dithering by setting Dithering Mask Depth...=Light and not hitting the button in the main Ui?

 

It might be that the FLAC file is smaller, if there is no signal above 20 kHz. I am not really sure. A way to test it would be to do a conversion in AuIConverter with the filter set to optimized, then the same AuIConverter conversion with filter set to non-optimized. If they are the same size, then the difference between DSDMaster and AuIConverter would be just the FLAC compression settings. I would not worry about it too much.

 

The dither button is just under the output settings (format and bit depth). As Yuri said, if the light is on, then dither will be applied. Except, apparently, if the input and output are both above 16 bit (I did not know that).

 

Paul

Link to comment
It might be that the FLAC file is smaller, if there is no signal above 20 kHz.

 

I'm not expert in compression algorithms and don't know how higher frequencies impact to size.

Theoretically, it is possibly.

 

AuI has settings for purer experiment.

For bit-perfect compression (similar input and output sample rates and bit depth without optimization and dithering)

 

in Settings > General > need check «Don’t make DSP …» switch.

 

It work for same sample rates. Optimization is not applied. Input and output bit depths may be different.

 

If input and output bit depths is same (sample rates too), signal passed by without changes in this mode.

 

Such settings was used in the above video (WAV vs. FLAC).

 

Best regards,

Yuri

AuI ConverteR 48x44 - HD audio converter/optimizer for DAC of high resolution files

ISO, DSF, DFF (1-bit/D64/128/256/512/1024), wav, flac, aiff, alac,  safe CD ripper to PCM/DSF,

Seamless Album Conversion, AIFF, WAV, FLAC, DSF metadata editor, Mac & Windows
Offline conversion save energy and nature

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...