Sign in to follow this  
Followers 0
Grunkk

Framerate limit

11 posts in this topic

Is there a reason why you don't impose a cap on the framerate permitted for your music? The human ear isn't capable of distinguishing sounds much higher than 128kbps (not to mention higher ones have considerably inflated filesizes). I have spent a great deal of time downloading some nice songs here, but every single one has had a greater framerate than justified. While I have gotten in the habit of editing these songs for personal use, removing undesirable elements, sometimes adding my own, and truncating framerate on the way, not all people are inclined or know to do this.As a matter of fact, inflated files are absolutely everywhere, like a sweeping tide of inefficiency called forth in the name of planned hardware obsolescence (though most of the time it's just not knowing better). I'm sure in today's age size consolidation does not matter so much, but back in the day it was a pleasant skill to practice, and was good form. If you wish to test this, feel free, but I am under the impression you knew this and have some other reason for not doing it. Would it really hurt the uploaders' feelings so much? Thanks in advance for your thoughts,

Share this post


Link to post
Share on other sites

First of all, frame rate only applies to videos. YouTube videos currently peak at 60 fps for HD and 30 fps for SD. :P

Second, people can hear differences between 128 kbps and 192+ kbps bit rates. Well, I can anyway, and I expect some of the younger people on the Judges Panel to be able to as well. A 128 kbps MP3 degrades audio above 16000 Hz, whereas I can hear pretty much all the way up to 20000 Hz.

128 kbps constant bit rate:

128 kbps.png

224 kbps constant bit rate:

VBR1.png

The difference is not obvious, but I hear it, and I can definitely see it. It's not the same as simply hard-low passing (high dB/octave slope) at 16000 Hz, but I do hear a slight degrading of the upper treble frequencies as a result of compression artifacts in 128 kbps relative to 192+ kbps.

djpretzel likes this

Share this post


Link to post
Share on other sites

I love this post.

I absolutely love this post.

It proves a certain point; for every individual convinced he/she has golden ears and can hear the difference between 1024Khz and 4096Khz and 128-bit audio vs. 2048-bit audio (intentionally ridiculous numbers here, folks!), there's someone questioning the use of bandwidth and (incorrectly) invoking something like Nyquist to argue that we've gone past the limits of human hearing...

@Grunkk, my brubba.... 128Kbps is a bitrate, not a sampling frequency/depth. You can have a 128Kbps "quality" stream of 16-bit, 22khz mono audio. You can have a 320Kbps stream of the same. You can also have a 64Kbps stream of 24-bit, 96Khz stereo...

I'm not gonna bore you with the details, here, but in simple terms: the MP3 bitrate dictates the lossiness of the compression, whereas the bit depth & sampling frequency dictate the fidelity of the source audio. The first is the lossiness of the digital compression being applied to the captured data, the second is the "lossiness" (more like "quantization") inherent in the conversion from the analog realm to the digital. Metaphor: MP3 bitrate is JPEG lossiness on a scale of 0-100%. where as audio bit depth & sampling frequency are the resolution and color depth of the lossless source image being compressed. You can have a JPEG 100% quality image that's 100x100 pixels, and you have have JPEG 20% quality image that's 1000x1000 pixels...

128Kbps can be noticeably lossy, even to the layperson, depending on the nature of the audio in question... hence our decision to increase to either 192Kbps or VBR1, with a Nyquist-friendly and industry-standard 16-bit + 44.1khz stereo source.

If all of this seems overwhelming, I guess you'll just have to trust us: we've landed on a standard that 99.9% of people couldn't reliably pass an A/B Coke/Pepsi taste test on when comparing to lossless, including Neil Young and Brendan Becker, yet which doesn't unnecessarily wander into the realm of 320Kbps, 24-bit, 96khz diminishing returns...

Garpocalypse, Slimy and timaeus222 like this

Share this post


Link to post
Share on other sites
7 hours ago, djpretzel said:

hat 99.9% of people couldn't reliably pass an A/B Coke/Pepsi taste test

Actually, it's quite simple to tell the difference - Coke has much stronger carbonation.

Share this post


Link to post
Share on other sites

While I found a satisfactory reply from Timaeus222 (thanks for clarifying minor differences in jargon between bit and frame rate), I am disappointed that the administrator djpretzel chose to be so flagrantly patronizing. And I won't stoop to his level by bolding that. Putting his ego aside, I'm sincerely sorry to hear that you think there is enough of a difference to justify doubling and quadrupling the size of your files. Naturally, 128kbps isn't the exact limit of the human ear; it varies somewhat so I can sort of see why 160kbps could contain some info (though how much value that has is negligible), but certainly not 192 or 320, no matter how much awesomesauce he puts in his coffee. Sure, you can see it on your software, and possibly hear something if really trying hard, but how meaningful is the difference? Is it twice the quality? I think not.

Share this post


Link to post
Share on other sites

In our day and age, we have access to some pretty hefty storage. I mean, look at this 128 GB flash drive for $5.59 (sure looks bleh, but man, that's almost as much storage as one of my external HDD's!) . I don't know how much that price got you in thumbdrive storage size in the times before I was born, but definitely much less than that! (Silly me, I still use an 8 GB thumbdrive for school.)

I really don't think storage is an issue, so sure, the particular MP3 for the example I showed you (1:33 long) was 2.48 MB at 224 kbps and 1.42 MB at 128 kbps, but the main benefit is that a 192+ kbps bitrate minimizes compression artifacts across most genres of music (whether it has a lot of crisp synthesized elements or poorly-recorded mono distorted electric guitar), and keeps a consistent bar of quality when it comes to the "age" of ReMix submissions.

I think it's a good thing that we at OCR favor higher bitrates over lower bitrates. :)

djpretzel likes this

Share this post


Link to post
Share on other sites
15 hours ago, Grunkk said:

I am disappointed that the administrator djpretzel chose to be so flagrantly patronizing.

You can take it like that if you want, but there was a lot of solid info in his post and you just seem unwilling to learn from it.

Share this post


Link to post
Share on other sites

oh man, I'm just getting into video editing to do my own music videos and this would totally explain why i'm having some synching issues. My camera records at 24 fps and my music is recorded starting at 01:06:00:05 on the Rendering Frame Timeline.  Divide that by the total average amount of measures in my music and i think i'm hitting somewhere between 32 and 55 fps.  So, hypothetically, i could fix this by halving the speed on my metronome and subdividing twice as fast...?  

Share this post


Link to post
Share on other sites
On 6/18/2016 at 7:05 AM, Grunkk said:

The human ear isn't capable of distinguishing sounds much higher than 128kbps (not to mention higher ones have considerably inflated file sizes).

Sorry if my response seemed flippant to you, but here you are stating as fact what is completely erroneous, and seem to be faulting us for it at the same time... if you had approached this topic as more of a point of inquiry rather than assuming your own understanding to be accurate & whole, and ours to be faulty, I think I would have responded differently...

19 hours ago, Grunkk said:

I am disappointed that the administrator djpretzel chose to be so flagrantly patronizing. And I won't stoop to his level by bolding that. Putting his ego aside, I'm sincerely sorry to hear that you think there is enough of a difference to justify doubling and quadrupling the size of your files. Naturally, 128kbps isn't the exact limit of the human ear; it varies somewhat so I can sort of see why 160kbps could contain some info (though how much value that has is negligible), but certainly not 192 or 320, no matter how much awesomesauce he puts in his coffee.

You're still not quite getting it, yet you project confidence and speak in certitudes ("negligible", "certainly not"); when talking about the limits of the human ear, sampling frequency and bit depth are the relevant metrics, since they dictate the fidelity of the conversion of audio from analog to digital. When talking about the lossiness of a bitrate, it's about how the audio in question is compromised to fit into that space... MP3 has a lot of tricks up its sleeve to increase quality at lower bitrates... things like cutting off low & high frequencies, joint stereo, etc... how perceptible this lossiness is depends on the actual audio in question... it also depends on whether the bitrate is constant or variable... it also depends on the MP3 encoder being used, because not all encoders are created equal...

Your initial post stated that the human ear couldn't hear sounds "higher" than 128Kbps, and your follow-up post still isn't understanding the difference between bit depth, sampling frequency, and the effective bitrate of a lossy compression algorithm like MP3.

Some decent additional info, with links to studies, at: http://skeptics.stackexchange.com/questions/3219/are-there-any-perceptible-differences-between-the-sound-quality-of-192-versus-32

Share this post


Link to post
Share on other sites
19 hours ago, Grunkk said:

(thanks for clarifying minor differences in jargon between bit and frame rate)

"Minor differences between bit and frame rate." Haha! You have no right to call anyone patronizing.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0