Best sound quality - run 96khz samplerate, do not oversample.

[Fritze]

not to get OT, but could some one recommend a (very) good SRC converter for downsampling because i find the one in Logic to suck ass pretty bad...

Voxengo r8brain

What puzzles me is that according to http://src.infinitewave.ca the free version is perfect while the pro version is "only good"?

[AudioGuy720]

not to get OT, but could some one recommend a (very) good SRC converter for downsampling because i find the one in Logic to suck ass pretty bad...

Voxengo r8brain

What puzzles me is that according to http://src.infinitewave.ca the free version is perfect while the pro version is "only good"?

You have to look at the other drop down menus, the free version only has a perfect frequency sweep but the other items not so much. Not bad though, like I said before charts/graphs may show something but may sound great.

Another thing you need to look out for which the charts don't have is the amount of pre-echo a steep filter has. So it will do a nice clean frequency sweep but it will add other artifacts like a linear phase equalizer.

When looking at the infinitewave charts bear in mind what you're looking at. Anything under -60dB will be essentially inaudible. And anything above 20 kHz will also be inaudible to 99% of humans.

My favorite comparison is between the Weiss SFC2 and anything else. If the sofrware is as good as a $5,500 converter or better then I'll use it, LOL!

[Fritze]

not to get OT, but could some one recommend a (very) good SRC converter for downsampling because i find the one in Logic to suck ass pretty bad...

Voxengo r8brain

What puzzles me is that according to http://src.infinitewave.ca the free version is perfect while the pro version is "only good"?

You have to look at the other drop down menus, the free version only has a perfect frequency sweep but the other items no so much. Not bad though, like I said before charts/graphs may show something but may sound great.

Another thing you need to look out for which the charts don't have is the amount of pre-echo a steep filter has. So it will do a nice clean frequency sweep but it will add other artifacts like a linear phase equalizer.

When looking at the infinitewave charts bear in mind what you're looking at. Anything under -60dB will be essentially inaudible. And anything above 20 kHz will also be inaudible to 99% of humans.

My favorite comparison is between the Weiss SFC2 and anything else. If the sofrware is as good as a $5,500 converter or better then I'll use it, LOL!

Thanx, lotsa good info though I disagree on the anything below -60dB being inaudible.

And a big on the Weiss!

I made some measurements since my last post and found out that there's so much room for fake in order to get nice pictures...

[AudioGuy720]

not to get OT, but could some one recommend a (very) good SRC converter for downsampling because i find the one in Logic to suck ass pretty bad...

Voxengo r8brain

What puzzles me is that according to http://src.infinitewave.ca the free version is perfect while the pro version is "only good"?

You have to look at the other drop down menus, the free version only has a perfect frequency sweep but the other items no so much. Not bad though, like I said before charts/graphs may show something but may sound great.

Another thing you need to look out for which the charts don't have is the amount of pre-echo a steep filter has. So it will do a nice clean frequency sweep but it will add other artifacts like a linear phase equalizer.

When looking at the infinitewave charts bear in mind what you're looking at. Anything under -60dB will be essentially inaudible. And anything above 20 kHz will also be inaudible to 99% of humans.

My favorite comparison is between the Weiss SFC2 and anything else. If the sofrware is as good as a $5,500 converter or better then I'll use it, LOL!

Thanx, lotsa good info though I disagree on the anything below -60dB being inaudible.

And a big on the Weiss!

I made some measurements since my last post and found out that there's so much room for fake in order to get nice pictures...

-60dB of something is audible during silent passages and such, yes. You have to read how their tests are done though to see why I said that. They're running a -6 dBFS sine wave sweep through the converters. A signal at -6 dBFS vs. one at -60dBFS is essentially inaudible. Mix a guitar -54dB below a vocal to see what I mean. A question for people smarter than me though is would running a quieter amplitude sweep result in quieter aliasing/distortion on the graph?

Also what is a little misleading is the phase chart. If you don't know about phase shift and how that works you may think the SRCs that show it are bad. Again, pre-echo (aka pre-ringing) is a side-effect of linear phase filters and can sound more unnatural than phase shift.

[mystran]

not to get OT, but could some one recommend a (very) good SRC converter for downsampling because i find the one in Logic to suck ass pretty bad...

Voxengo r8brain

What puzzles me is that according to http://src.infinitewave.ca the free version is perfect while the pro version is "only good"?

You have to look at the other drop down menus, the free version only has a perfect frequency sweep but the other items no so much. Not bad though, like I said before charts/graphs may show something but may sound great.

Another thing you need to look out for which the charts don't have is the amount of pre-echo a steep filter has. So it will do a nice clean frequency sweep but it will add other artifacts like a linear phase equalizer.

When looking at the infinitewave charts bear in mind what you're looking at. Anything under -60dB will be essentially inaudible. And anything above 20 kHz will also be inaudible to 99% of humans.

My favorite comparison is between the Weiss SFC2 and anything else. If the sofrware is as good as a $5,500 converter or better then I'll use it, LOL!

Thanx, lotsa good info though I disagree on the anything below -60dB being inaudible.

Well, I disagree as well if we can test with arbitrary signals. For actual musical content -60dB is often acceptable though. Even then, if you put require -100dB (better than 16-bit signal) or even -150dB (better than 24-bit signal) those filters are still essentially perfect. In fact I'd be willing to bet Aleksey's target was probably around -160dB or so; more than that and you should be using your computational resources for other things like transition bandwidth or less pass-band ripple.

[Fritze]

-60dB of something is audible during silent passages and such, yes. You have to read how their tests are done though to see why I said that. They're running a -6 dBFS sine wave sweep through the converters. A signal at -6 dBFS vs. one at -60dBFS is essentially inaudible. Mix a guitar -54dB below a vocal to see what I mean.

You have to take frequencies into concideration: The masking effect is heavily frequency dependant.

[Fritze]

Well, I disagree as well if we can test with arbitrary signals. For actual musical content -60dB is often acceptable though. Even then, if you put require -100dB (better than 16-bit signal) or even -150dB (better than 24-bit signal) those filters are still essentially perfect. In fact I'd be willing to bet Aleksey's target was probably around -160dB or so; more than that and you should be using your computational resources for other things like transition bandwidth or less pass-band ripple.

Absolutely agreed!

[JasonSpatola]

No it's crazy to accept the degradation of language.

Place a comma after an interjection.

I did mention it's true meaning,

"It's" is a contraction for "it is". You're looking for "its", a pronoun signifying possession. You used it correctly in your first sentence.

which of course it always third on the list.

The pronoun "it" used in this context contravenes proper sentence syntax. A better choice is "is", the singular present indicative of "be".

Here's a corrected version of your statement:

No, it's crazy to accept the degradation of language. I did mention its true meaning, which of course is always third on the list.

Just messing with you.

(Note that despite being an American, I prefer the British method for comma placement outside of quotes.)

Speaking on your actual point, you should recognize that many scientific terms wouldn't actually be scientific terms without those third/fourth/fifth/etc. entries.

Also, equating polysemy/semantic change with language degradation is an assailment of the nature of language as we've always known it. While I agree that sudden, radical changes to (or reinterpretations of) our lexicon could lead to a massive communication breakdown, I don't view this particular use of "virtue" as a threat.

[AudioGuy720]

-60dB of something is audible during silent passages and such, yes. You have to read how their tests are done though to see why I said that. They're running a -6 dBFS sine wave sweep through the converters. A signal at -6 dBFS vs. one at -60dBFS is essentially inaudible. Mix a guitar -54dB below a vocal to see what I mean.

You have to take frequencies into concideration: The masking effect is heavily frequency dependant.

Run a full mastered mix against a vocal -60 dB under its peak. Barely if at all perceptible IMHO. We'll have to agree to disagree then on this.

[Fritze]

-60dB of something is audible during silent passages and such, yes. You have to read how their tests are done though to see why I said that. They're running a -6 dBFS sine wave sweep through the converters. A signal at -6 dBFS vs. one at -60dBFS is essentially inaudible. Mix a guitar -54dB below a vocal to see what I mean.

You have to take frequencies into concideration: The masking effect is heavily frequency dependant.

Run a full mastered mix against a vocal -60 dB under its peak. Barely if at all perceptible IMHO. We'll have to agree to disagree then on this.

We wouldn't have to disagree if I would get my point across. Please check this:

http://en.wikipedia.org/wiki/Auditory_masking

[Compyfox]

My favorite comparison is between the Weiss SFC2 and anything else. If the sofrware is as good as a $5,500 converter or better then I'll use it, LOL!

Looking a the Wess SFC2 and the comparisions form that page, I must say that pretty much EVERYTHING is better.

I use Audiomove for quite a while now for both upsampling and downsampling (48kHz zo 44kHz). Personally I think it's the best tool in my equipment in that particular section. Never liked Wavelab for that purpose other than it's included Apogee Dither/Noiseshaper for 44/16 finalising - which however only takes care of the bitrate, not the SRC.


EDIT:
Stuck keys...

[The Chubu]

When i gave a look at this thread a few days ago, with no answers i thought "whoa, this ppl already grew up past all these sound quality endless discussions"

And here i am, 12 pages after lol

[Fritze]

When i gave a look at this thread a few days ago, with no answers i thought "whoa, this ppl already grew up past all these sound quality endless discussions"

And here i am, 12 pages after lol

Well, epic fail then. But thanx for letting me know.

[Alexey Lukin]

Another thing you need to look out for which the charts don't have is the amount of pre-echo a steep filter has.

It is shown in the "Impulse" chart.

Looking a the Wess SFC2 and the comparisions form that page, I must say that pretty much EVERYTHING is better.

I'd put it differently: Weiss SFC2 is quite good, just as most other tested converters. This is stated in the FAQ section of the site.

[Compyfox]

Actually, according to the FAQ they use the Izotope RX2 (Adv2) as reference. People even bragged on why it is among "the best" and why one of the founders works for Izotope (aka the bias comment).

To my understanding, the less "blue" is on the sweep plot, the better. And here, the Weiss SFC2 IMO fails. Also with the transition and the 1kHz tone comparision.

But then again - what do I know, I'm just a normal "audio guy".