User profile for user: Joseph Covert
Joseph Covert Author
User level: Level 2
195 points

16/44.1k has more BASS than 24/88.2k ??? Why?

I made some sample tests over the weekend to try to figure out at which rate I want to sample with, and I'm a bit puzzled at my findings. Turntable -> Rane mixer -> M-Audio Delta 66 -> Peak ->Toast CD was the chain. All files were recorded in Peak. When I made the 24-bit files at different khz rates, I was careful to also use Peak to down-convert them to CD specs, because there was a difference between Peak & Toast's internal conversion. I really think I thought of everything with regards to putting the samples on equal ground.

Anyway, I found that...

- 16-bit, 44.1Khz - the sound was not quite as "wide" as the 24 bit files, but the kick drum, bass guitar, and even the snare sounded markedly more full and punchy. The bass line could be heard un-interrupted pulsing at the bottom.
- 24-bit, 44.1Khz - the sound was a bit "wider" as I said, and a good touch clearer in the high end. But the lack of "energy" was also noticable. The kick drum and bass diminished even though they were technically still there. The sound wasn't as "connected" feeling.
- 24-bit, 48k - my ears couldn't tell a diff....I really couldn't. It was too close to 24/44 so I sort of mentally eliminated it from the tests eventually.
- 24-bit, 88.2Khz - This is as good as my sound card will do, and I had to have a 1024 buffer to do it. The sound was noticably the "widest" and also a "deepness" started to become apparent. But still, when compared to the 16/44.1 file, it really didn't sound as nice in the bass area, even though it was nicer in the mids and highs, and it was def wider.

I've wondered why the Akai MPC's (except for the 4000) are all at 16-bit/44.1Khz, and these machines are much-loved in Hip-Hop of course, and 100 out of 100 people claim that drums sound better through them....is it something to do with the sample rates??

I was theorizing (scary I know) that maybe the big blocky pieces of 16-bit data actually add a synthetic component to bass waves that make them sound even punchier? I've got a 12-bit S-950 sampler, and the sample disk that came with it (Akai Soundset) has EXTREMELY punchy, in your gut, sort of sounds.

o.k. sorry for rambling here...but does anyone ever sample at 2 different rates and then layer them later? (for drums etc.)

thanks for listening.

G4 Dual 500, 1.75GB RAM, LaCie D2, Mac OS X (10.4.8), Logic Pro 7.2.3, Delta66, JV1080, S950, Serato Scratch, 2xTechnics, TTM56, Avalon737,Aphex 204, misc. guitars, mixers etc. etc.

Posted on Jan 2, 2007 1:52 PM

Reply
17 replies
User profile for user: Joseph Covert
Joseph Covert Author
User level: Level 2
195 points

Jan 2, 2007 2:10 PM in response to jord

What dithering algorithms did you use?

jord


Good question and thanks jord...I used the Pow-r dithering, I'm pretty sure it was Type-I since there was no choice for the other types in my version of Peak. Obviously, the 16/44.1 file didn't need anything.

(My earlier Toast conversions made me suspect that the type of dithering used could be a factor, so...I just picked Peak's conversion because it sounded closer to the original when A/B'ed against Toast's conversions)

...very interested to know your take on this Jord.
Reply
User profile for user: jnashguitar
jnashguitar
User level: Level 2
345 points

Jan 2, 2007 4:52 PM in response to Joseph Covert

jizzer,

Did you try listening to the source files themselves, too ? (i.e. without downsampling, dithering, and burning to CD ?)

Obviously, if your final product is CD, the only thing that matters is how the stuff sounds burned to CD, but I wonder what you thought about the sound quality when played back at higher sampling rates from inside Peak ? Did you still prefer the low-end of the 16-bit, 44.1k file, or did the 24-bit, 88.2k file have better "energy" when played back at its native resolution ? Maybe the downsampling and dithering is what killed things for you ?

James
james@nashvillain.com
Reply
User profile for user: Joseph Covert
Joseph Covert Author
User level: Level 2
195 points

Jan 3, 2007 7:57 PM in response to jnashguitar

Thanks for the replies all...

jnashguitar - No, I didn't specifically test the actual files in my studio. For one, my current monitors s*ck rocks and I can't hear bass at all, so I opted to do this test on my surround system w/sub downstairs in living room. I also thought it would be a poor comparison of the 16/44.1 file since my system seems to run at whatever clock rate I set it at...so in my soundcard settings, I can choose between 44.1/48/88.2/96...and I can *sort of* hear the differences (I can really tell when I set it lower..like 22khz or 11 khz etc.). So I figured if I played a 16/44.1 file on my system, there would be some conversion of some sort.
Reply
User profile for user: Joseph Covert
Joseph Covert Author
User level: Level 2
195 points

Jan 3, 2007 8:04 PM in response to David Robinson9

because there are less genuine high frequencies at
any given comparable level.


That makes sense, but are you suggesting that I simply percieved more bass in the 16/44.1 because of the lack of high-end? Because it would be hard to convince me there wasn't really more bass there (although I remain open-minded).

This may or may not be relevent, but I did the test from start to finish twice, and the second time I normalized all tracks. But even so, I didn't percieve a volume change between files. (although Peak said there was a diff between a few .1dbs)

and fwiw, I listened to the same tracks on repeat while web-surfing until I forgot which song file was playing....and it was only then that I started my tests because I wanted to be very sure I was ONLY using my ears for this.
Reply
User profile for user: Joseph Covert
Joseph Covert Author
User level: Level 2
195 points

Jan 3, 2007 8:10 PM in response to Rik-The-Mix

I'd be curious what difference if any shows up in frequency analysis. > On a metering plugin, does the 24 bit version show less bass


I didn't even check Mike. Plus, this was really a 'real world' test for me, and I didn't want to be influenced to believe this or that. There is enough disagreement amongst the experts that i felt a listening test was in order for myself.


If I was a betting man I would say this has more to
do with the quality, or lack of quality, of the
M-Audio Delta 66 than anything else! I'd like to see
the results of this test using a bit of pro kit.


You're probably right. I know a pro kit would sound better in every way, but I'm curious as to whether it would still exhibit this phenomena of tighter bass coming from a lower bit-rate?
Reply
User profile for user: lwilliam
lwilliam
User level: Level 2
363 points

Jan 4, 2007 12:51 AM in response to Joseph Covert

"- 16-bit, 44.1Khz - the sound was not quite as "wide" as the 24 bit files, but the kick drum, bass guitar, and even the snare sounded markedly more full and punchy. The bass line could be heard un-interrupted pulsing at the bottom."


The "wide" is perfectly understandable since our perception of depth and width relies a lot on low-level ambient cues that could easily be lost in a 16-bit system vs a 24-bit system.

The other part is harder...just another man's theory here, but if we were to take this to extremes...say a 3-bit system: off, soft and loud.

It would be true then that anything over a certain input level threshhold would be assigned as a "loud" bit. Could that make certain characteristics and sounds appear more "punchy"? Obviously, there are 2 to the 16th power levels of resolution in a 16-bit byte, but I'm just wondering if this concept could affect the perception more so than with a 24-bit system.

Another factor to consider is that 12-bit systems would generally have more distortion in the analog section than a high-end 24-bit converter. Distortion can definitely affect the perception of both loudness and 'punchiness'. Remember, the signal goes through analog circuits before it hits the converter chip. That analog portion is pretty critical to how the system sounds.

That's why two a/d converters can use the same converter chips and even the same clocks, but still sound different.
Reply
User profile for user: Jope
Jope
User level: Level 1
70 points

Jan 4, 2007 1:07 AM in response to Joseph Covert

One simple question: Did you use Space Designer for reverb? I remember when I went from 44.1kHz to 88.2kHz for better anti-aliasing of software synthesizers, the output level of Space Designer changed... Obviously an abnormality, but I didn't make much fuzz about it since all I had to do was adjusting the output control of Space Designer.
Reply
User profile for user: Jope
Jope
User level: Level 1
70 points

Jan 4, 2007 3:49 AM in response to Jope

Oh, sorry, I just realized Logic isn't used here... Well, maybe there is an analog filter in your Delta 66 whose cutoff frequency is switched to another value when the sample frequency is changed. Or maybe the Delta 66 always uses 88.2kHz internally but then does a digital sample rate conversion with an inadequate algorithm. I don't believe a higher sample rate will result in lower bass by itself.
Reply
User profile for user: Joseph Covert
Joseph Covert Author
User level: Level 2
195 points

Jan 4, 2007 1:12 PM in response to lwilliam

Thanks again for the replies guys...if I read between the lines of what most of you are saying, it's that something is fishy about my test...you guys do NOT expect a dimished sense of bass with 24b files.

I was really hoping to nail down a sampling resolution I could count on for smpling all my vinyl drums, but I guess it's back to 'play it by ear on each track'. Booo. 😉

Thanks again for the feedback all.
Reply

This thread has been closed by the system or the community team. You may vote for any posts you find helpful, or search the Community for additional answers.

16/44.1k has more BASS than 24/88.2k ??? Why?

Welcome to Apple Support Community
A forum where Apple customers help each other with their products. Get started with your Apple Account.
Learn more Sign up