IPB

Welcome Guest ( Log In | Register )

32-bit support, Split from: "The Future of FLAC"
Brand
post Aug 31 2012, 22:21
Post #1





Group: Members
Posts: 318
Joined: 27-November 09
Member No.: 75355



QUOTE (yourlord @ Aug 31 2012, 21:26) *
32-bit support

QUOTE (sshd @ Aug 31 2012, 22:16) *
>2G support

Both of these are important.

I was glad to see that in audio editing/music making software FLAC gained a lot of support in the last couple of years. I'd say right now most support at least importing FLAC.
But 32bit float is used a lot in audio editing, so FLAC could definitely improve there.


EDIT:
Also, I remember reading once that FLAC is too loosely defined (compared to ALAC where the rules are more strict regarding number of channels etc.). Don't know much about that stuff myself, but if it's true it's worth looking into. Hardware/software manufacturers want a reliable standard to work with.
EDIT2: found it

This post has been edited by Brand: Aug 31 2012, 22:31
Go to the top of the page
+Quote Post
 
Start new topic
Replies
Brand
post Sep 1 2012, 22:56
Post #2





Group: Members
Posts: 318
Joined: 27-November 09
Member No.: 75355



I'm not sure I agree with what I think you're implying (that adding ridiculously high specs is some sort of deceiving). Also FLAC already supports 32bit fixed point and 655350Hz sample rates, both of which are unnecessary by certain HA standards.
Anyway, the 64bit part was mostly a joke, although again, I don't see the harm in adding it. I would disagree with promoting it with statements like "it improves the sound", but I wasn't suggesting doing that.

However, I do believe 32bit float support would be useful, even if it's only used in certain niche cases and not for distribution. (But I should stop repeating myself now.)

This post has been edited by Brand: Sep 1 2012, 22:59
Go to the top of the page
+Quote Post
db1989
post Sep 1 2012, 23:11
Post #3





Group: Super Moderator
Posts: 5275
Joined: 23-June 06
Member No.: 32180



QUOTE (Brand @ Sep 1 2012, 22:56) *
I'm not sure I agree with what I think you're implying (that adding ridiculously high specs is some sort of deceiving).
You think wrongly, for I’m not implying anything. I meant exactly what I said (quite clearly, I had thought): The ability to use such resolutions should not be added for the sake of attempting to increase the format’s appeal to people who hold mistaken beliefs about their utility.

QUOTE
I would disagree with promoting [high resolutions] with statements like "it improves the sound", but I wasn't suggesting doing that.
You might not have stated that yourself, and I said neither that you said it nor that you believe it, but you did say that higher resolutions might help to assuage and appeal to those who do believe such myths. So, the two might become effectively equivalent, regardless of your intentions. The spread of falsehoods about higher resolutions is not likely to be decreased, as Hydrogenaudio desires, by their inclusion “for marketing purposes”. Rather, I can imagine gurus of woo interpreting it precisely the opposite way.

If you didn’t meant that the next maintainer(s) of FLAC should add support for higher resolutions in order to pacify those who wrongly believe that “bigger is better” in this context, you shouldn’t have chosen a configuration of words that says just that.
Go to the top of the page
+Quote Post
Brand
post Sep 2 2012, 13:29
Post #4





Group: Members
Posts: 318
Joined: 27-November 09
Member No.: 75355



QUOTE (db1989 @ Sep 2 2012, 00:11) *
If you didn’t meant that the next maintainer(s) of FLAC should add support for higher resolutions in order to pacify those who wrongly believe that “bigger is better” in this context, you shouldn’t have chosen a configuration of words that says just that.

Right, I should have separated those first two sentences when talking about popularity (and "marketing").
But--at the risk of stating the obvious and repeating again--let me clarify: If FLAC can benefit from increased popularity, then it's something the developers should take into account. Of course, not by selling snake oil, but with technical improvements, which can (at least potentially) be useful and make some people happy at the same time.

If someone thinks those improvements are good for the wrong reasons ("it sounds so much better") it's unfortunate, but if this helps make FLAC more popular and survive in the long term, I see it as a net positive result.
There's only so much you can do to keep people informed... and FLAC (not) including 'high res' capabilities in its specs IMO doesn't make a difference in this regard. Those who want or need to learn about this stuff will always find out what's what.
And I would be very much in favor of a FAQ on the official FLAC page, pointing out that good AD/DA converters yield around 20bits of dynamic range.. that 24bit is already overkill and that 32bit float is only useful for certain applications. That should keep uninformed speculations at bay.


Unless we want to limit FLAC to a final-product-distribution format, but I don't (yet) see a good reason for that.

In that sense, even my 64bit suggestion was semi-serious, after all 64bit float is used in some DAWs.
(here's a thread about its potential/theoretical usefulness, including a quoted post about how to bring 32bit float to its limits--I haven't tried that myself, don't know how valid it is)
Go to the top of the page
+Quote Post
[JAZ]
post Sep 2 2012, 16:43
Post #5





Group: Members
Posts: 1787
Joined: 24-June 02
From: Catalunya(Spain)
Member No.: 2383



QUOTE (Brand @ Sep 2 2012, 14:29) *
(here's a thread about its potential/theoretical usefulness, including a quoted post about how to bring 32bit float to its limits--I haven't tried that myself, don't know how valid it is)


If one reads that well, it just says that there is a mathematical difference doing operations in 32bit float versus 64bit float, and that if one increases the *difference* of both signals, it can be seen/heard. Well... yes... and the same happens if you get 128bit float versus 80bit float versus 64bit float... That's completely irrelevant to *audio*.

64bit float *math* is required in several audio algorithms to prevent audible problems, but 64bit float audio pipeline is relatively unnecessary versus 32bit float, because although you might be accumulating error, you're accumulating it on the 24th bit of the mantissa. How many operations does one need for that to become audible?

In the 90's, the audio pipeline of audio applications was 16bit INTEGER. (x87 floating point operations were too slow)
Go to the top of the page
+Quote Post

Posts in this topic


Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 24th October 2014 - 17:20