[castleofargh]

This is amazing - this is exactly what I was looking for. Thank you sir! (Obviously I need to work on my internet search skills, lol)

I’m not going to advertise the results here for fear of sparking an unnecessary debate, but, suffice to say, they’ve resolved my concerns quite nicely.

Thanks again - I really appreciate that a community like this exists in a world that’s gone altogether mad about advertising!

It's only a white paper and done with few subjects on one format. So you probably shouldn't draw definitive conclusions from just that. Plus, no matter how many similar trials we present that fail to disprove the null hypothesis under typical listening conditions, it never proves that nobody could. Or that we couldn't create conditions where it's easy to tell the formats apart. So, technically you cannot hope for science to give the definitive answer that high bitrate lossy codecs sound exactly like higher resolutions.

With that said, in this highly debated question, we tend to have mostly people performing blind tests on one side and mostly people doing sighted listening on the other. So while that's also not conclusive about the possibility to perceive a difference by ear, it's at least strongly suggestive that the vast majority of people claiming to hear a difference doesn't know how to do a listening test.

 

[skhan007]

Great thread and the white paper is a good indicator of statistical trend. With a large enough sample size, a follow up study might reach statistical significance. I’m curious to do my own single subject design and listen to the same song in 16/44.1, 24/96, etc. in a blinded manner. Most of what I’m learning on this Sound Science forum is telling me that my ears (likely) can’t perceive any appreciable superiority above my ALAC 16/44.1 music, but I’m very keen on seeing if I can acquire some hi-res files to hear for myself. At the end of the day, I agree it’s incredibly subjective.

 

[OhmAlone]

It's only a white paper and done with few subjects on one format. So you probably shouldn't draw definitive conclusions from just that. Plus, no matter how many similar trials we present that fail to disprove the null hypothesis under typical listening conditions, it never proves that nobody could. Or that we couldn't create conditions where it's easy to tell the formats apart. So, technically you cannot hope for science to give the definitive answer that high bitrate lossy codecs sound exactly like higher resolutions.

With that said, in this highly debated question, we tend to have mostly people performing blind tests on one side and mostly people doing sighted listening on the other. So while that's also not conclusive about the possibility to perceive a difference by ear, it's at least strongly suggestive that the vast majority of people claiming to hear a difference doesn't know how to do a listening test.

I understand. Thanks for clarifying. I’m really just trying to make the best financial decision for myself personally (and to stop worrying so much about bitrates when listening!) by keeping some scientific data ready at hand as a guide light for those times when I inevitably descend back into the marketing rabbit hole.

 

[bigshot]

When I first started building my media server, I spent a couple of weeks testing codecs and comparing them. I ripped and encoded a bunch of different kinds of music at a range of rates using several different codecs. Then I did careful listening tests to determine the threshold of transparency. My goal was to see if lossy could sound *exactly* like lossless. AAC turned out to be the most efficient codec- AAC at 256 was transparent, while MP3 LAME was transparent at 320. I decided to encode my entire library in AAC 256 VBR. I've encoded hundreds and hundreds of CDs and have never encountered any artifacting. Whatever you decide, always use VBR. It can only help. It can't hurt.

If you are interested in 24/96 and how it compares, see the article in my sig file called CD Sound Is All You Need

 

[skhan007]

When I first started building my media server, I spent a couple of weeks testing codecs and comparing them. I ripped and encoded a bunch of different kinds of music at a range of rates using several different codecs. Then I did careful listening tests to determine the threshold of transparency. My goal was to see if lossy could sound *exactly* like lossless. AAC turned out to be the most efficient codec- AAC at 256 was transparent, while MP3 LAME was transparent at 320. I decided to encode my entire library in AAC 256 VBR. I've encoded hundreds and hundreds of CDs and have never encountered any artifacting. Whatever you decide, always use VBR. It can only help. It can't hurt.

If you are interested in 24/96 and how it compares, see the article in my sig file called CD Sound Is All You Need

I actually read the articles in your sig. I found this info (and a lot of other discussions in the Sound Science forum) to be compelling, fascinating, and eye-opening. On a different thread, I learned that the remastering process yields (potentially) the big audible improvements that people attribute to hi-res.

 

[bigshot]

Yes, the quality of mixing and mastering is the most important factor in sound quality. Garbage in- garbage out. CD quality sound is designed to cover the entire range of human hearing. It actually has quite a bit of overkill baked into it. Lossy at a decent data rate is perfectly capable of sounding just as good as lossless or "HD" audio to human ears. The only reason you would ever need more is if you plan to process the audio in the future.

 

[old tech]

I actually read the articles in your sig. I found this info (and a lot of other discussions in the Sound Science forum) to be compelling, fascinating, and eye-opening. On a different thread, I learned that the remastering process yields (potentially) the big audible improvements that people attribute to hi-res.

You can also fine some good articles (such as the one below) and listening tests on Mark Waldrep's website. He makes a living from recording/producing/selling hi res music so it is quite refreshing that he doesn't gild the lily.

https://www.realhd-audio.com/?p=7241

 

[bigshot]

The pinned post in this forum, Testing Audio Myths has lots of interesting stuff too.

 

[skhan007]

Quick follow up: If I'm listening to an ALAC file as 16/44.1 and I'm switching the internal clock on my DAC from 44.1 to 192 to let's say 768 kHz, will the DAC simply be sampling the stream of information in the file that many more times, but no more information is there, so it will sound the same? I ask because as I'm playing with sampling rate, I swear I hear more treble frequencies with higher rates. Not necessarily better resolution or quality, but an ever so slight tonal pronunciation in the highs.

 

[bigshot]

There won't be more information. It should sound the same, but I don't know how good the upsampling is in your DAC. There's really no need to do that, so they might not have made it a design priority. Using my Mac, sample rate changes are transparent.

The only way to know if there actually is a slight change in the highs is to do a blind test. Expectation bias could produce the same impression.

 

[skhan007]

Makes sense. I imagine that upsampling is similar to what technology does with video- upsampling to 4K resolution for example. My DAC is the RME ADI-2 and I love it so far (only received it a few days ago). Perhaps the switching of the sample rate is also producing a slight boost in volume and I'm perceiving that and/or change in treble frequencies.

 

[bigshot]

It could be a slight volume boost. They might have designed it that way so you could hear some sort of difference. I had an SACD player that had a placebo button. It was labelled Direct/Clean and when you pushed it, it lit up a beautiful color of purple. I thought it might be making some kid of difference, then I did a blind listening test and realized it was just a purple light. They should have had it boost the volume a hair. That would make for more effective placebo.

 

[BrianovichIV]

It could be a slight volume boost. They might have designed it that way so you could hear some sort of difference. I had an SACD player that had a placebo button. It was labelled Direct/Clean and when you pushed it, it lit up a beautiful color of purple. I thought it might be making some kid of difference, then I did a blind listening test and realized it was just a purple light. They should have had it boost the volume a hair. That would make for more effective placebo.

My receiver has a similar button, but it actually does do stuff! It shuts off all the lights and also turns off any EQ, phono gain tweaks, or other settings. So it actually makes my music sound worse because I've tuned it a little to fit my living room.

 

[neil74]

Yes, the quality of mixing and mastering is the most important factor in sound quality.

100% this ^^

When comparing say 320 kbps streaming vs lossless the service used is really important I think. Spotify is my favourite service but there are a LOT of turds in their library that have nothing to do with the codec that they use. Tidal and Apple music for example I think have the better masters and there is an argument with Tidal that 320 kbps AAC is about as good as you need unles you have real high-end gear and very good ears!

Re Spotify I read somewhere before a rumour that when they moved from 192 to 320 that they upscaled a lot of files which could explain a lot but that is probably a whole other thread in itself.