[umvue]

According to this paper, they sound the same:
 
https://secure.aes.org/forum/pubs/journal/?ID=2
 
Does anyone know any replication studies that support or refute their conclusion???

 

[bigshot]

It makes sense if you look at the difference between 16 and 24 bit sound. At peak level, 24 and 16 are identical. 24 bit has more resolution at very low volume levels. So in order to hear a difference with music, you would have to turn up the volume to ear splitting levels. 24 bit is great for mixing where they need to bring up quiet stuff in the mix, but for normal listening, it makes no difference.

 

[nick_charles]

 
Quote:

According to this paper, they sound the same:
 
https://secure.aes.org/forum/pubs/journal/?ID=2
 
Does anyone know any replication studies that support or refute their conclusion???

Meyer and Moran has appeared in this subforum many times. It has generated much heated debate. The short answer is no there have not been any replication studies at all. No serious researchers or multi-national audio companies have shown any interest in the last 6 years of picking this one up any further. I'll summarize the critiques of the paper to save time.
 
Their kit was not good enough, it is not enough to have a high res player it must be a "good" high res player, and the recordings (some of them anyway) were not good enough, it is not enough to have a DVD-A or SACD it must also be a good one high res born.
 
The supporters say - not one person and not one recording ever managed a stratistically significant level of detection of difference and the 16/44.1 stage was a separate AD stage which is much more degrading, not an internal conversion .
 
I would personally like M and M to do a replication study, but it is not going to happen.
 
You pays yer money ($20 for non AES members) and you takes yer choice...

 

[goodvibes]

Done proper transfer from analog to both formats and there's no question in my mind that 24/96 is superior. Don't find much advantage above. With that said, it's not the most important aspect of a transfer and significance can get lost in both transfer and playback, especially via a PC so I understand both sides. Personally, I use 24/96 whenever possible as it does get further in for me.

 

[digger945]

Quote:

Done proper transfer from analog to both formats and there's no question in my mind that 24/96 is superior. Don't find much advantage above. With that said, it's not the most important aspect of a transfer and significance can get lost in both transfer and playback, especially via a PC so I understand both sides. Personally, I use 24/96 whenever possible as it does get further in for me.

Hello goodvibes!
 
What dac or integrated amp/dac are you listening to?
 

 

[Head Injury]

Quote:

Done proper transfer from analog to both formats and there's no question in my mind that 24/96 is superior. Don't find much advantage above. With that said, it's not the most important aspect of a transfer and significance can get lost in both transfer and playback, especially via a PC so I understand both sides. Personally, I use 24/96 whenever possible as it does get further in for me.

 
Quote:

Hello goodvibes!
 
What dac or integrated amp/dac are you listening to?

And what were the parameters of your double blind test?

 

[bigshot]

And what kind of a difference are you hearing?

 

[goodvibes]

Naim Dac w/ps via Naim streamer. Regardless of what I do, I can't get the dig out of a PC or Mac to sound as good. Weiss interface and Amarra or Wavelab still doesn't perform as well, for me. Use both the 57s and 2805s quads with Naim 500 series. Nagra 4s analog dubbed to a Nagra VI. All that said I would still be able hear these things dubbing vinyl with a Nagra LB and top table using a stand alone UnitiQute for playback. Key is to have a good enough source and transcription.
 
 Sounds just extend and flesh out, better harmonics texture and timing, the way ambiance extends until something louder masks it instead of just stopping. A bit more going on while sounding more relaxed or less busy/noisy at the same time. Blacker. Don't think going beyond 24/96 gets you much if anything so I'm not of the more is always better contingent. These may not be big or important differences for most, even negligible in many cases. A lot of people here grew up on 16 bit. I didn't (had quads and LP12 with moving coils over 30 years back) and never fully got used to it though it's clearly better than it's ever been in the past due to some of the latest DAC topologies. You would be amazed at how easy it is to hear different dither profiles. So many variables that no other point of view will surprise me. Leave it to my opinion as I'm out and don't care what others do. These discussions never go well but I thought I'd give a sincere answer before I split.

 

[bigshot]

Well, it sounds very good, but I have no idea what you're talking about. Can you describe what you're hearing using terms that describe sound rather than visual terms like "blacker" and emotional terms like "relaxed" or "busy"? Dynamics? FR? Distortion? Technically, the big difference should be dynamics, and if you can hear that, you would have to be compressing pretty hard or playing back at extraordinary volumes.

I totally agree with you about dithering. There is always one dither that works perfectly and a bunch that don't. You can't just blindly use any old dither.

At work, I used a ProTools work station that we used to record dialogue and music. We would always record and mix 24 bit to allow room for boosting levels. But once we finished the mix, the bump to 16 always sounded identical. If it doesn't for you, I'd suggest trying other dithers or see if there is something wrong with your software or hardware. It really should sound identical. My bet is your problem is in the transcoding. If you captured at 16/44.1 and avoided the bump down, it would probably sound fine.

As for vinyl transcription, going to 24 bit should only give added resolution a mile below the noise floor. It might help define transients for impulse noise reduction filters, but I always found that the minute (if any) improvement in noise reduction wasn't worth the crawl that my ProTools work station slowed down to. This was a few years back. Comps may have gotten faster and filters better designed for 24 bit.