Taperssection.com
Gear / Technical Help => Playback Forum => Topic started by: Ryan Sims on January 04, 2008, 02:41:55 PM
-
Alright, so I got my first 24-bit capable system for Christmas:
Foobar 2K>Audiophile 2496 (coax)>Zhalou DAC/Headphone Amp (http://www.obadimports.com/catalog/item/4377355/4344428.htm)>Audio Technica ATH-m50 phones
I just cannot hear a discernable difference in files recorded at 24 from their dithered counterparts. Note that this is all on files from the Archive, as I don't own a DVDA player.
Foobar and the 2496 both display output at 96khz or what have you, but I just don't hear it. What might be going on?
TIA
Ryan
-
Just a guess but does your DAC upsample? If it does then the 16/44.1 info you are sending it should (theoretically) sound reasonably close to the original 24/xx recording.
-
is the "output data format" area in Foobar set to 24 bit output?(file>preferences)
I have never been able to hear an enormous difference ..at least immediately. The advantages will become more apparent over time. (Greater dynamic contrasts, depth, seperation). The source material is a major factor.... I find it very hard to discern the differences between high res and redbook "PA" recordings of mine, but with some of my symphonic recordings it is much easier. Just give it time, and dont expect it to be some major "eureka!" type of moment.
Alright, so I got my first 24-bit capable system for Christmas:
Foobar 2K>Audiophile 2496 (coax)>Zhalou DAC/Headphone Amp (http://www.obadimports.com/catalog/item/4377355/4344428.htm)>Audio Technica ATH-m50 phones
I just cannot hear a discernable difference in files recorded at 24 from their dithered counterparts. Note that this is all on files from the Archive, as I don't own a DVDA player.
Foobar and the 2496 both display output at 96khz or what have you, but I just don't hear it. What might be going on?
TIA
Ryan
-
As far as I know, when using scientifically valid double blind tests, nobody has reliably heard a difference between good 16/44 and 24/96, or between analog line original and 16/44. So no reason to worry (exept the possibly wasted money).
-
Well I didn't waste any money, considering I still love the sound improvement of my 16-bit stuff with a decent playback system.
-
a) dont expect a big difference
b) if youre not using kernel streaming, you almost certainly arent putting out a 24-bit data stream, thanks to microsoft's kmixer
-
b) if youre not using kernel streaming, you almost certainly arent putting out a 24-bit data stream, thanks to microsoft's kmixer
can you elaborate?
i play my audio through the VX Pocket 440. should i assume that i'm hearing 24 bit audio?
-
I have had folks do a double blind for me on my home playback system and I have have picked the 24bit every time. The difference is in the dynamic range and in both the smoothness (less "electronic sounding") in the upper end and the tonal richness in the mid-bass/bottom end. It is not a night and day thing and it takes bot a tuned ear and a playback system capable of displaying the difference but it is certainly there. Keep playing around with your software as recommended above and try doing your won dithering from the 24bit sources you DL from the archive. If you have a bud with a kick-ass playback system see if they will let you do a few tests of your own on their system. I sold audio gear for the last few years and had plenty of time to do comparisons in our high end room and I can safely say that, once you get used to the difference, there is no going back.
-
if youre using the asio or directsound drivers with the vx, you should be ok. digigram cards are a lot better than the m-audio when it comes to 24-bit
I'd test it (record the bitstream, look at the resulting wav with the bit-depth meter in wavelab or invert it over the original file using soundforge)
b) if youre not using kernel streaming, you almost certainly arent putting out a 24-bit data stream, thanks to microsoft's kmixer
can you elaborate?
i play my audio through the VX Pocket 440. should i assume that i'm hearing 24 bit audio?
-
Ryan, any one piece of audio gear might sound different from another for many reasons. But when you ask about the sonic difference between good 16-bit and good 24-bit, that difference in itself is so slight that it wouldn't normally be noticeable unless the playback volume was set very, VERY high.
When the CD format was still in its design stage, Philips seriously proposed that it be limited to 14-bit resolution, because they had a huge stockpile of 14-bit DAC chips and for home listening, 84 dB dynamic range would already be a great improvement over the best LPs or FM broadcasts. It's considerably wider than the dynamic range of 15 ips Dolby-A master tapes. 16 bits gives another 12 dB range, which puts its noise floor far below that of any home listening environment other than a distant farmhouse in the middle of the night. Even then, if an ant scratches its groin in the pantry, you lose it.
So maybe you didn't do anything wrong at all--maybe the two things really do sound the same under the listening conditions that you have there. On the other hand, maybe that's because as some here seem to be suggesting, your setup is feeding the same 16-bit (i.e. truncated) signal to both playback systems. You should definitely check that out. A truncated 16-bit rendition of a 24-bit recording isn't valid for making any comparisons other than to test the effects of truncation itself--sooner or later you'll hear why it's bad.
The type of sonic differences described by Stagger in this thread are clearly being caused by some other, extraneous variable. It could be truncation but more likely it's something else (perhaps in addition). The only inherent difference between 16-bit and 24-bit systems is their dynamic range; there are no differences of timbre, imaging, or anything that is particular to any one part of the audio frequency spectrum. When you hear any of that stuff, it's time to take a break and figure out where your experimental design went to hell.
There is an astonishing amount of persistent nonsense floating around on this topic ... and not just on this forum by any means. When the CD was introduced in the early 1980s, the audiophile magazines went all Lou Dobbs on the infidel invader, and the simple technical facts of the situation have nearly been banished to living in caves.
--best regards
-
I would say the differences on headphones are less exaggerated. With the 24bit playback on my system the real noticeable difference is on the size and clarity of the soundstage. It's certainly more apparent on the speaker system. On headphones what you'll notice is slightly better definition of the trailing and leading edges of notes. How they emanate and fade away, decay trails are cleaner and a bit more natural. The music is less "confused" especially during peaks. Definitely make sure you're using Kernel Streaming. One easy way of knowing if it's happening is that you're volume slider in the software will have no impact on the sound.
-
Jammin72, the comparison you describe isn't valid unless you've done the following first:
[1] Take the 24-bit sound file and make a copy of it; keep the original in a safe place.
[2] Bring the 24-bit copy into some software such as Adobe Audition that lets you dither it to a particular bit depth.
[3] Dither that copy to 16 bits and store the result.
Now compare the original 24-bit file with the dithered 16-bit copy, using the exact same hardware and volume settings for both. There should be no audible difference other than (possibly) a very slightly higher noise floor in the 16-bit version.
--best regards
-
Jammin72, the comparison you describe isn't valid unless you've done the following first:
[1] Take the 24-bit sound file and make a copy of it; keep the original in a safe place.
[2] Bring the 24-bit copy into some software such as Adobe Audition that lets you dither it to a particular bit depth.
[3] Dither that copy to 16 bits and store the result.
Now compare the original 24-bit file with the dithered 16-bit copy, using the exact same hardware and volume settings for both. There should be no audible difference other than (possibly) a very slightly higher noise floor in the 16-bit version.
--best regards
I have done so with my own recordings. Not using Adobe but with both Reaper and Sequoia. However it may be the re sampling that it's causing the difference I don't usually dither without in another step re-sampling to 44.1 from the original 48. It may also be that the number crunchers in my receiver (STR-DA5000ES) prefer the 24bit data on it's way to becoming a one bit stream. Many possibilities. All I know is there IS a discernible difference between the 24/48 and the 16/44.1 files with no variance in volume played back through the same system. I often spend time flipping between the two since it's simple enough in MediaMonkey to do so. I've read so many technical papers and seen so many arguments that I'm not really sure what is fact other than on my system with my ears and my own little brain the 24bit files just sound more open and real.
Thanks for the information, it's always nice to know some of the technical underpinnings of the theoretical debates. I'll spend some more time with the dithered files only. +T