Izotope SRC

[barrows]

Hey Jud, yeah, I cannot post here from home, only when I am on broadband at a coffee shop or friend's home. A lot of good sounding filters still allow for a lot of alias leakage, check out John Atkinson's DAC measurements...

Certainly more processing power can allow for better filter algorithms, but some DACs can do this really well if they use a separate DSP to do the filtering, agreed that in the limited silicone of a DAC chip the filters are necessarily compromised a bit, but an FPGA doing the first 8x can handle it really well. I am using a USB interface which incorporates a FPGA and runs an 8x OSF there which is much better than the stock filter in the ESS chip.

Folks like Chord, Ayre, Berkeley (I presume), and Auralic are all running proprietary filters in powerful DSP chips, so they can have no compromise in the number of taps to create the "best" filter design they can.

[Superdad]

...agreed that in the limited silicone of a DAC chip the filters are necessarily compromised a bit, ...

 

Barrows, you are a master of understatement!

[SiCHIPS]

Great! BUT can you please detail how to set these parameters using what program???

I have Pure Music, Audirvana and Fidelia...

I also HEAR the differences in my system with an NAD C390DD to Aerial Acoustics 10Ts and a Room EQ corrected room.

 

I would greatly appreciate the details on how you are setting these parameters!!!

 

-Laurier St.Onge ( [email protected] )

 

 

These are my settings from the other thread:

 

I've been going between these 2 sets of settings based on how the high-frequencies are attenuated. I feel that the edge might go to the 0.90 settings since symbols etc. sound a little more crisp and present. Adjust to taste based on your taste and your system's frequency response. For me I tend to lean toward the 0.90 filter because the highs, and air might bit just a touch sharper. On very bright systems/recording, one could go with a 0.88 cut-off filter with a steepness of 23. More rolled-off but sounds very analog.

Still need to play with filter length, and to a much lesser extent, pre-ringing. 0.6 +/- 0.1 sounds about right. Only a very slight amount of pre-ringing with these settings.

 

Power of 2 Upsampling

Steepness: 26

Filter Max Length: 1,800,000 (still testing)

Cut-Off: 0.90

Anti-Aliasing: 200 (max)

Pre-Ringing: 0.60

-0.1: 18,570 Hz

-1: 19,140

-3: 19,530

-6: 19,850

-113: Nyquist

 

Steepness: 24

Filter Max Length: 1,800,000 (still testing)

Cut-Off Freq.: 0.89

Anti-Aliasing: 200 (max)

Pre-Ringing: 0.60

-0.1: 18250 Hz

-1: 18820

-3: 19270

-6: 19600

-114: Nyquist

You have so much more quality, if you are willing to accept frequency response linearity about 19k Hz. A relatively shallow "brick wall" filter, with it's lower pre and post-ringing distortion, lower phase shifting, and good attenuation at Nyquist which helps reduce a lot of the digital hash not only from the playback filters, but possibly from downsampling during mastering.

[SpecMode]

Great! BUT can you please detail how to set these parameters using what program???

I have Pure Music, Audirvana and Fidelia...

I also HEAR the differences in my system with an NAD C390DD to Aerial Acoustics 10Ts and a Room EQ corrected room.

 

I would greatly appreciate the details on how you are setting these parameters!!!

 

-Laurier St.Onge ( [email protected] )

 

In Audirvana Plus preferences, select 'Audio Filters', and make sure 'Converter' is set to 'Izotope 64-bit SRC'. Expand 'Advanced Parameters' below that, and change the values to match those in your post (or to whatever you want, really). Note that they'll only have any effect if you're actually changing the sample rate of the tracks you're playing (typically upsampling, which is controlled by the next setting down on that page).

 

[ikkei]

Just a few remarks and suggestions about pre-ringing and steepness from my personal subjective experience, after having read this thread and fiddled with iZ settings on A2+ (and compared that with SoX on foobar to get some kind of base reference).

 

 

Pre-ringing

 

I personally use Pre-Ringing ~.36 (give or take .08). I think this setting is fairly hardware-dependent, and essentially changes what I'd call 'time-cohesion' (an umbrella concept that, for a subjective listener, pertains to phase shifting, and/or attacks/transients 'sync' and 'precision', and/or whatever makes music sound 'on time' or off of it…)

 

To critically listen to this particular resampler setting, to tune it with your personal audio system, I'd suggest

Human Nature (Michael Jackson, Thriller, make sure your source is a non-brickwalled master, and digital only such as SACD/download/CD with a good pressing). I got this idea from a Pono ad video, I really agree with the guy who suggested it.

 

This particular track indeed is a time-killer for most systems I've heard. It never fails to impress in state-of-the-art pure analog, and often manages to bring expensive digital systems to their knees, especially too-busy-CA and flawed operations in the digital domain (typically resampling). In my opinion, A2+ handles Human Nature flawlessly without resampling, so I'd expect a well-tuned resampling (ad hoc for one's audio hardware) should be flawless as well.

 

To check pre-ringing settings, I'd also suggest any track with extremely well recorded intimate voices (think jazz and the best pop productions, anything after 2000 should do it provided your source is, again, a very dynamic master). Listen to the breaths and saliva clicks and just about every noise that a mouth does, that's the kind of subtlety I find quite varying with resampling quality and accuracy. Look for 'broken' sounds (when your DSD feels like MP3 for a split second of artefact…), seek the most natural sound your brain believes.

 

Soundtracks (especially sound effects made with real-world objects, recorded with great dynamics preserved for theatre) also do a pretty awesome job for critical listening a particular sound processing: the point is that whatever you hear must sound natural; and we all have well-trained ears and well-refined brain filters for real-world sounds. A 'digitally flawed' resampling may make music somewhat off, but it simply destroys real sounds that our brain knows to recognise so well —of course, a trained musician / live enthusiast may achieve the same level of listening and discerning criticality with musical instruments, but for most of us, at the notable exception of intimate voices, we've mostly been listening to music through the eye of a particular audio system which, as far as I know, has never been perfect nor 'pure', lest you're a billionaire with a live band and orchestra at home, in an anechoic chamber to let them play for you.

 

I typically find that bad resampling breaks natural echo: metal, stone, reverberations and the likes sound weird, like a digital delay on a guitar amp, with a tad of flanging or phase shifting, discernible when the echo is long and pronounced. Just as compression flatly kills the illusion that a sound is real.

 

So, .36 does the trick for me. 1 sounds totally digital. 0 is very present, too much so, the sound is broken like there's too much gain/presence and everything seems lower in resolution. Anything from .8 down to .08 sounds at least realistic, and how much definition I can get varies depending on the genre (classical likes .44, compressed digital pop/electro sounds more 'club' or 'live' around .28). That's my personal experience.

 

The Black Triangle CP35-017 JP master of The Dark Side of The Moon tends to prove that, when compared to more "in your face" later digital masters even sourced from the original vinyl stereo master like that Blu-ray from late 2000's, unfortunately that particular recording wasn't precise enough, in my humble opinion, to tweak SRC pre-ringing.

 

I wish I could set different iZ settings depending on the genre and DR of the track.

 

I should have started with this, but it also seems to me that setting pre-ringing correctly (using the default settings for everything else) is the step to take first. I take it as a "align this resampling with my hardware" sort of step. Then I fiddle with everything else (especially steepness, the rest I don't find much room for tweaking).

 

 

Steepness

 

Steepness, I find on my system, is ultimately limited in 'acceptable range' by one's loudspeakers dynamics (how fast the mid and low speakers can reproduce a sound) and preamp/amp stage (same idea).

I find that the lower settings (anything below 96dB) gives a somewhat analog feel that I interpret as added noise and/or artefacts, thus it's very lenient with noisy analog setups such as tubes etc. (probably best suited for these systems, and to ears used to hearing the very characteristic tube noise-shaping and vinyl frequency variations).

 

On the contrary, I find that settings above 154dB are harsh, very digital, often too 'crude', too 'raw' in a bad meaning. I've loosely calculated a 'sweet spot' at 132dB on my main system, and indeed at 136-140dB it does a great job for these recent state-of-the-art 21st century fantastic classical and jazz recordings—try Opus3 Records, Chesky Records, B&W, Linn, Naim, etc. However for everything else, particularly compressed material, I tend to lean between 144dB (better suited to slow or minimalist material) 152-153dB (punchy! good for low-DR stuff and complex multi-track (i mean in studio) material that may require a bit of surgery not to sound like a giant pile of mud).

 

Indeed, as far as stereo is concerned, I find that higher steepness tends to do a better job at channel separation and mid-bass centred sounds definition (these ~11kHz harmonics that give (contra)bass its distinct timbre and grain depending on the instrument/amp used by the musician). You lose some of that to gain more overall cohesion and more natural sounds as you lower steepness, at least that's my perception.

 

Steepness seems like a compromise to me: lower (not too low though) sound more 'natural', higher more 'surgical'. I find that the more clinical the record, the lower I can get away with (which is always nice); conversely low-DR and bad recordings need that extra bit (pun intended) of surgery.

 

 

Ok well that's long enough, so to sum it up:

 

Steepness: 138-153dB depending on the material

Pre-ringing: 0.36 or .37 on my system

From memory (don't have my A2+ near to check), I keep aliasing between 190 and 200; cutoff at 1.08 Nyquist (I don't care for anything above 16kHz but it definitely sounds better on my system at 1.08-1.12); samples as much as I can get away with (I think I went with 1,411,200 with 8GB Ram, of which 4 are dedicated to A2+).

 

My 2 cts after reading this discussion for a long time.

[Jud]

I'll offer here four suites of iZotope SRC settings for Audirvana+, along with a brief explanation of what the settings do (to the level of my knowledge, which isn't much - some of what I say here may be plain wrong, and I welcome corrections). Two of these are done by ear by amateurs and are thus possibly "broken" or "pathological," or however you'd like to describe it; the other two are from pros who know what they're doing. I won't tell you which are which.

 

The effects of various filter settings are quite variable between systems. In particular, since most DACs are restricted to 192kHz input, but upsample internally to 352.8/384kHz, you will be hearing some combination of your own filter settings and the internal filtering in the DAC chip.

 

First suite:

 

Steepness - 5 db

 

This number is approximately one-fourth of the filter "order." A first order filter provides a slope (cut) of 6db/octave, second order 12db/octave, and so on. Thus steepness of 5db = ~20th order = ~120db/octave cut. The steeper the filter is, the more "ringing" it will generate; the less steep the filter is, the more aliases will be allowed to fold back into the audible range.

 

Filter max length - 10,000 samples

 

This is the number of samples over which the filter operates, and is again related to the filter order and thus the amount of slope or cut imparted to the signal. More samples = more cut.

 

Cutoff frequency - 1.00 x Nyquist

 

I believe this is the (sq rt of 2)/2 point, i.e., where the filtered signal is .707 of the unfiltered value.

 

Anti-aliasing - 200.0

 

Alias suppression in the filter's stop-band.

 

Pre-ringing - 1.00

 

This does not adjust the total amount of ringing. Steepness does that. It expresses the proportion of pre-ringing over post-ringing as a decimal. Therefore as the amount of total ringing energy that occurs as pre-ringing goes lower in proportion to the amount that occurs as post-ringing, the number goes down.

 

1.00 means the amount of pre-ringing and post-ringing energy are equal. This occurs when the filter characteristic is what is known as linear phase. When all the ringing energy is moved to post-ringing, then the pre-ringing proportion will be 0.00, and the filter characteristic will be what is known as minimum phase. A minimum phase filter is a "dispersive" filter. This means time to get through the filter is frequency-dependent. So different frequencies from the same instrument played at the same time (guitar or piano chords, for example) come out of the speakers at very slightly different times. (For more, see http://resonessencelabs.com/digital-filters/ .) For many people, the fact that minimum phase filters eliminate pre-ringing is significant to sound quality. For me, I feel there is something in the dispersive nature of a minimum phase filter (also sometimes described in terms of "group delay") that is very bothersome; I think it plays a part in messing up soundstage and localization of instruments and vocals within the soundstage. Some very smart audio engineers disagree over this: Keith Johnson, the designer of Spectral electronics and co-founder of Pacific Microsonics, feels it is important to eliminate group delay, while Charles Hansen, designer of Ayre electronics and the Pono player, uses minimum phase filters in his designs.

 

Forced Oversampling - Power of 2 only

 

Some people feel integer (power of 2) oversamplng values are best, others that this is a holdover from when filters and computers were less capable, and maximum sample rate can be used at all times.

 

So to summarize the first suite of settings:

 

Steepness - 5

Filter max length - 10,000

Cutoff frequency - 1.00 x Nyquist

Anti-aliasing - 200

Pre-ringing - 1.00

Forced Oversampling - Power of 2 only

 

Second suite:

 

Steepness - 7

Filter max length - 1,300,000

Cutoff frequency - 1.02 x Nyquist

Anti-aliasing - 200

Pre-ringing - 0.86

Forced Oversampling - ?

Third suite:

 

Steepness - 150

Filter max length - 500,000

Cutoff frequency - 1.00 x Nyquist

Anti-aliasing - 200

Pre-ringing - 0.75(?)

Forced Oversampling - ?

Fourth suite:

 

Steepness - 200

Filter max length - 10,000

Cutoff frequency - 1.03 x Nyquist

Anti-aliasing - 100

Pre-ringing - 1.00

Forced Oversampling - Maximum sample rate upsampling

[skatbelt]

According to information from Izotope on the various variables there is a direct relationship between steepness and pre-ringing:

 

"Low-pass filters are characterized by the amount of ringing they introduce into their output. Higher Steepness settings produce increased ringing. A setting of 100% produces a linear phase filter with equal pre and post ringing. A setting of 0% produces a minimum phase filter that offers no preringing but has nonlinear phase distortion. Intermediate settings allow a tradeoff between preringing and postringing and allows you to linearize phase in the pass-band."

 

This is exactly what I am experiencing and it makes tuning more complicated than just finding the system (mean) optimum for the 5 variables. I don't want to go as far as defining different 'presets' depending on material or genre.

[Jud]

According to information from Izotope on the various variables there is a direct relationship between steepness and pre-ringing:

 

"Low-pass filters are characterized by the amount of ringing they introduce into their output. Higher Steepness settings produce increased ringing. A setting of 100% produces a linear phase filter with equal pre and post ringing. A setting of 0% produces a minimum phase filter that offers no preringing but has nonlinear phase distortion. Intermediate settings allow a tradeoff between preringing and postringing and allows you to linearize phase in the pass-band."

 

This is exactly what I am experiencing and it makes tuning more complicated than just finding the system (mean) optimum for the 5 variables. I don't want to go as far as defining different 'presets' depending on material or genre.

 

There are at least two factors to be balanced for each of the settings you mentioned, steepness and pre-ringing. A steeper filter allows less aliasing distortion but causes more ringing. Less pre-ringing results in phase or group delay distortion. Up to you what sounds best, informed by the information in this thread and elsewhere on the Web.

[livelistenlearn]

I'll offer here four suites of iZotope SRC settings for Audirvana+, along with a brief explanation of what the settings do (to the level of my knowledge, which isn't much - some of what I say here may be plain wrong, and I welcome corrections). Two of these are done by ear by amateurs and are thus possibly "broken" or "pathological," or however you'd like to describe it; the other two are from pros who know what they're doing. I won't tell you which are which.

 

haven't slept for 24 hours or something. coffee tasting very good.

apologies if rambling ensues but just have to post this:

 

suite 1 (with 2 variances) on ver 2.0.7 has delivered something akin to a system upgrade for me.

 

improvements across the board >> airiness, sound-staging, layering, detail, presence, etc.

 

most importantly, music is no longer “played”, it just flows.

 

(present) settings for an early 2011 mbp 2.2GHz/i7 >> hugo:

 

upsampling: power of 2

steepness: 5

filter max length: 500,000 (somehow cannot reduce, defaults to this)

anti-aliasing: 200

pre-ringing: 0.8 (smoother sound + more transparency)

 

have not heard this level/magnitude of improvement since speaker and DAC upgrades.

 

for 16/44.1, as well as 24/88, 24/96 and 24/192 tracks. not sure about DSD though.

 

thank you so much, Jud, for pointing the way here and informative posts.

 

BIG thanks too to levandier for starting this thread, as well as junker, superdad, alexey lukin, miska, barrows, bdiament for leading the way. + everybody else who helped to pave the way.

 

still grasping at some concepts + tech points raised/discussed. no point pretending i can understand most of it.

 

but, holy guacamole. it works so good. can’t stop listening. and grinning like the (computer audiophile tech) idiot i am.

 

cheers to good music + good coffee.

 

btw, suite 3 seems to be the default settings.

[Jud]

haven't slept for 24 hours or something. coffee tasting very good.

apologies if rambling ensues but just have to post this:

 

suite 1 (with 2 variances) on ver 2.0.7 has delivered something akin to a system upgrade for me.

 

improvements across the board >> airiness, sound-staging, layering, detail, presence, etc.

 

most importantly, music is no longer “played”, it just flows.

 

(present) settings for an early 2011 mbp 2.2GHz/i7 >> hugo:

 

upsampling: power of 2

steepness: 5

filter max length: 500,000 (somehow cannot reduce, defaults to this)

anti-aliasing: 200

pre-ringing: 0.8 (smoother sound + more transparency)

 

have not heard this level/magnitude of improvement since speaker and DAC upgrades.

 

for 16/44.1, as well as 24/88, 24/96 and 24/192 tracks. not sure about DSD though.

 

thank you so much, Jud, for pointing the way here and informative posts.

 

BIG thanks too to levandier for starting this thread, as well as junker, superdad, alexey lukin, miska, barrows, bdiament for leading the way. + everybody else who helped to pave the way.

 

still grasping at some concepts + tech points raised/discussed. no point pretending i can understand most of it.

 

but, holy guacamole. it works so good. can’t stop listening. and grinning like the (computer audiophile tech) idiot i am.

 

cheers to good music + good coffee.

 

btw, suite 3 seems to be the default settings.

 

Ain't it grand?

 

Yep, Suite 3 is the default, so that's one of the two done by people who know what they're doing.

 

For max length, just highlight the number in the field, then type in the value you want. 10,000 is the minimum the field will accept.

[pl_svn]

Ain't it grand?

 

Yep, Suite 3 is the default, so that's one of the two done by people who know what they're doing.

 

For max length, just highlight the number in the field, then type in the value you want. 10,000 is the minimum the field will accept.

 

and, with my NOS DAC, I've been using the other one made by "someone who knows what he's doing" for quite a long time

I believe he uses x2 upsampling, btw

 

I only tweaked a veeery tiny little bit "filter max length" and "pre-ringing")

[Jud]

and, with my NOS DAC, I've been using the other one made by "someone who knows what he's doing" for quite a long time

I believe he uses x2 upsampling, btw

 

I only tweaked a veeery tiny little bit "filter max length" and "pre-ringing")

 

Well, I'd hoped not to give the mystery away so fast, but I know for certain that the other suite done by a pro uses max sample rate upsampling.

 

So that will tell you the two done by amateurs that you should watch out for!

[pl_svn]

Well, I'd hoped not to give the mystery away so fast, but I know for certain that the other suite done by a pro uses max sample rate upsampling.

 

So that will tell you the two done by amateurs that you should watch out for!

 

... nothing in my post to guess which one I was talking about

[skatbelt]

haven't slept for 24 hours or something. coffee tasting very good.

apologies if rambling ensues but just have to post this:

 

suite 1 (with 2 variances) on ver 2.0.7 has delivered something akin to a system upgrade for me.

 

improvements across the board >> airiness, sound-staging, layering, detail, presence, etc.

 

most importantly, music is no longer “played”, it just flows.

 

(present) settings for an early 2011 mbp 2.2GHz/i7 >> hugo:

 

upsampling: power of 2

steepness: 5

filter max length: 500,000 (somehow cannot reduce, defaults to this)

anti-aliasing: 200

pre-ringing: 0.8 (smoother sound + more transparency)

 

have not heard this level/magnitude of improvement since speaker and DAC upgrades.

 

for 16/44.1, as well as 24/88, 24/96 and 24/192 tracks. not sure about DSD though.

 

thank you so much, Jud, for pointing the way here and informative posts.

 

BIG thanks too to levandier for starting this thread, as well as junker, superdad, alexey lukin, miska, barrows, bdiament for leading the way. + everybody else who helped to pave the way.

 

still grasping at some concepts + tech points raised/discussed. no point pretending i can understand most of it.

 

but, holy guacamole. it works so good. can’t stop listening. and grinning like the (computer audiophile tech) idiot i am.

 

cheers to good music + good coffee.

 

btw, suite 3 seems to be the default settings.

 

Well, out of curiosity I tried your settings. It gives me a deep and well ordered soundstage but the presentation of instruments and voices are flat. My own settings (in my setup) are much more life like, I can hear more details (tape hiss, chair movement etc.) but - more important - instruments and voices feel like volumes in space. They do have a front and a back, are more 3D so to speak. As I mentioned above, steepness and pre-ringing are strong related so it took a while to find my optimum. My settings are:

 

Steepness: 83

Filter max length: 800,000

Cutoff freq.: 1,08

Anti-aliasing: 200.0

Pre-ringing: 0.30

Forced Upsampling: Power of 2

[juliocat]

Well, out of curiosity I tried your settings. It gives me a deep and well ordered soundstage but the presentation of instruments and voices are flat. My own settings (in my setup) are much more life like, I can hear more details (tape hiss, chair movement etc.) but - more important - instruments and voices feel like volumes in space. They do have a front and a back, are more 3D so to speak. As I mentioned above, steepness and pre-ringing are strong related so it took a while to find my optimum. My settings are:

 

Steepness: 83

Filter max length: 800,000

Cutoff freq.: 1,08

Anti-aliasing: 200.0

Pre-ringing: 0.30

Forced Upsampling: Power of 2

 

Your settings are the Bomb!!!!! Great Sound

[YashN]

Haaa, how many thanks can I send you for this, Jud?

 

Many many, many ones indeed for sharing.

 

 

I'll offer here four suites of iZotope SRC settings for Audirvana+, along with a brief explanation of what the settings do

[skatbelt]

Your settings are the Bomb!!!!! Great Sound

 

Looks like it is not so system dependent after all...

[sdolezalek]

Is there any way to "save" the settings for each "Suite" so that they can be easily compared? Maybe this should be added to the wish list for A+ upgrades, along with the ability to name them, like Jud 1, Jud 2, etc...

[Jud]

Looks like it is not so system dependent after all...

 

I think it likely is, and that your systems may work similarly.

[Jud]

Is there any way to "save" the settings for each "Suite" so that they can be easily compared? Maybe this should be added to the wish list for A+ upgrades, along with the ability to name them, like Jud 1, Jud 2, etc...

 

No way yet. It would be a nice extra touch.

[YashN]

This should be a feature of Aurdirvana, really...

 

Is there any way to "save" the settings for each "Suite" so that they can be easily compared? Maybe this should be added to the wish list for A+ upgrades, along with the ability to name them, like Jud 1, Jud 2, etc...

[dabears999]

Well, out of curiosity I tried your settings. It gives me a deep and well ordered soundstage but the presentation of instruments and voices are flat. My own settings (in my setup) are much more life like, I can hear more details (tape hiss, chair movement etc.) but - more important - instruments and voices feel like volumes in space. They do have a front and a back, are more 3D so to speak. As I mentioned above, steepness and pre-ringing are strong related so it took a while to find my optimum. My settings are:

 

Steepness: 83

Filter max length: 800,000

Cutoff freq.: 1,08

Anti-aliasing: 200.0

Pre-ringing: 0.30

Forced Upsampling: Power of 2

 

Interesting Result: Applied these settings and have observed a noticeable difference in A+. In my system, which has some similarity to yours, perhaps improved depth & width of soundstage, excellent detail, improved space between instruments. That said, I have always preferred A+ when playing DSD files as compared to Amarra, yet always preferred Amarra when playing 44.1khz and all HiRez files up to 192khz. Something about Amarra has just sounded right - maybe a warmer, less analytically result than A+.

 

I look forward to listening to Redbook CD & HiRez music this weekend to determine if A+ has closed that gap or exceeded Amarra here. No doubt these settings have made an improvement. Thanks!

[skatbelt]

Interesting Result: Applied these settings and have observed a noticeable difference in A+. In my system, which has some similarity to yours, perhaps improved depth & width of soundstage, excellent detail, improved space between instruments. That said, I have always preferred A+ when playing DSD files as compared to Amarra, yet always preferred Amarra when playing 44.1khz and all HiRez files up to 192khz. Something about Amarra has just sounded right - maybe a warmer, less analytically result than A+.

 

I look forward to listening to Redbook CD & HiRez music this weekend to determine if A+ has closed that gap or exceeded Amarra here. No doubt these settings have made an improvement. Thanks!

 

I was an Amarra fan myself from early 2010 until a month ago or so. I tried everything else in trial but always came back to Amarra for sound quality. Even despite of their bad test and quality assurance management and follow ups on known errors. When I tried version 2.0.6 of Audirvana plus I all-most didn't believe my ears. After trying out different 'Audio System' settings that is. My settings now are: exclusive access mode active, direct mode active, and integer mode 1 active. Combined with the SRC settings as described it gives me superior results over Amarra, especially with 44/16 up to 96/24 files. I use Audirvana stand alone (I sync with the iTunes library now and then and manage my files there). I used Amarra in playlist mode most of the time.

[Mystic]

These settings work quite nice in my setup (Oppo BDP-105). Being mainly a vinyl junkie I like more natural/warmer sound I get with these settings. I would still like to get the detail and liveliness with CA that I get out of my Vecteur L4.2 CDP, but I'm sure that's achieved by stepping it up in the DAC department.

Well, out of curiosity I tried your settings. It gives me a deep and well ordered soundstage but the presentation of instruments and voices are flat. My own settings (in my setup) are much more life like, I can hear more details (tape hiss, chair movement etc.) but - more important - instruments and voices feel like volumes in space. They do have a front and a back, are more 3D so to speak. As I mentioned above, steepness and pre-ringing are strong related so it took a while to find my optimum. My settings are:

 

Steepness: 83

Filter max length: 800,000

Cutoff freq.: 1,08

Anti-aliasing: 200.0

Pre-ringing: 0.30

Forced Upsampling: Power of 2

[Barcabale]

Well, out of curiosity I tried your settings. It gives me a deep and well ordered soundstage but the presentation of instruments and voices are flat. My own settings (in my setup) are much more life like, I can hear more details (tape hiss, chair movement etc.) but - more important - instruments and voices feel like volumes in space. They do have a front and a back, are more 3D so to speak. As I mentioned above, steepness and pre-ringing are strong related so it took a while to find my optimum. My settings are:

 

Steepness: 83

Filter max length: 800,000

Cutoff freq.: 1,08

Anti-aliasing: 200.0

Pre-ringing: 0.30

Forced Upsampling: Power of 2

 

Your settings works extremely well on my system Thanks so much its make a really huge difference than before !