erikistired
Diamond Member
- Sep 27, 2000
- 9,739
- 0
- 0
Rubycon
Madame President
- Aug 10, 2005
- 17,768
- 485
- 126
Loudness contour is a function where a gentle boost of bass frequencies is augmented at lower levels and gradually tapers down to flat as the volume is raised. It helps out to hear extended bass response at gentle levels.
The "crystalizer" function on the sound cards is a compander algorithm with a dynamic gate that modifies dynamic range and works on low and higher frequencies. On some movies the effect can be desirable under certain conditions.
Claiming it "remasters" compressed content to sound better than original is ludicrous, however.
I agree. We're going in the wrong direction.
Hey, why don't we record at 24kbps then upconvert to 320kbps? It's gotta sound better than a recording originally encoded at 192kbps. 320 is a higher number.
This is the "digital vibrance" of audio. UGH!
What? You mean it can't convert my rotting feces into beautiful music? That is pathetic, it's 2006, come the ****** on.
Not true, not true at all. As statistical understanding of signals and computing power improves its possible to predict (based on priors) the original signal. Will it be identical? Maybe, maybe not. Can a lossy signal be improved to be near lossless? Yep. My research is in image processing, but given that sound is only 1D this is definately possible in real time.
Again wrong. Look into the literature on higher order statistics (HOS) of signals. Applying a lossy compression algorithm on a signal with higher-order dependencies preserves these dependencies. These are far more important to signal quality than lower-order correlations (eg, 1st and 2nd).
erikistired
Diamond Member
- Sep 27, 2000
- 9,739
- 0
- 0
Once it's lost, it can never be found again, it can only be predicted. I guess you could loop through all the possibilities and compress them and find out what the compressed data ends up as but it'd be a billion years till we have the processing power to brute force it. And frankly I haven't found any upsampling (image or audio wise) or deblocking that impressive. Not that that's surprising. I do remember Intel had some project that took some huge supercomputer months to redo this JPEG image and it actually came out decent and revealed more detail. Good luck doing that in realtime though.
I'm only an aficionado when it comes to computers and acustics, but what you say seems to make a lot of sense.
As far as I know, lossy compression algorithms in sound work by using linear prediction patterns to reduce redundancy to the minimum, not true? Also, by taking advantage of the intrinsic limitations of the human ear, certain quick frecuency shifts are not coded at all, like a soft sound after a loud sound, or by not coding those +20KHz/-20Hz frequencies 99% of the population can't hear anyway.
Given that I can't imagine how the file might be restored, much less in real time.
- Mar 4, 2000
- 27,370
- 239
- 106
Algorithms, schmalgorithms. I daresay that there is prolly no user's sound system tied to a computer that would allow the user's ears to notice any difference (if it did work as advertised.)
Nearly all of my MP3 listening is done in the car - and that environment really doesn't allow for noticeable improvement to the ear.
My senses agree with MS Dawn.
Nearly all of my MP3 listening is done in the car - and that environment really doesn't allow for noticeable improvement to the ear.
My senses agree with MS Dawn.
More ridiculous BS from Creative; trying to fix a non-existent problem. Upconverting MP3s? How about just re-ripping them! Effing retards.
Oyeve
Lifer
- Oct 18, 1999
- 21,980
- 847
- 126
Originally posted by: corkyg
Algorithms, schmalgorithms. I daresay that there is prolly no user's sound system tied to a computer that would allow the user's ears to notice any difference (if it did work as advertised.)
Nearly all of my MP3 listening is done in the car - and that environment really doesn't allow for noticeable improvement to the ear.
My senses agree with MS Dawn.
A few of us have high-end systems hooked up to our computers. My receiver is probably worth more than my pc. I hear the difference between 128 and 192 mp3s. I heard a difference between my ZS2 and X-fi (although not worth the upgrade IMO). I even have a highend turntable hooked up to my PC. Yes, I am one of "those" people who beleive that vinyl (other than ticks an pops) sound better than CDs. So there are a few of us computer guys and gals out their who have highend hi-fi equipment connected to their computers.
Originally posted by: Oyeve
Originally posted by: corkyg
Algorithms, schmalgorithms. I daresay that there is prolly no user's sound system tied to a computer that would allow the user's ears to notice any difference (if it did work as advertised.)
Nearly all of my MP3 listening is done in the car - and that environment really doesn't allow for noticeable improvement to the ear.
My senses agree with MS Dawn.
A few of us have high-end systems hooked up to our computers. My receiver is probably worth more than my pc. I hear the difference between 128 and 192 mp3s. I heard a difference between my ZS2 and X-fi (although not worth the upgrade IMO). I even have a highend turntable hooked up to my PC. Yes, I am one of "those" people who beleive that vinyl (other than ticks an pops) sound better than CDs. So there are a few of us computer guys and gals out their who have highend hi-fi equipment connected to their computers.
Fix your rig link in your sig so we can see what you have.
What high-end equipment? Amplifiers etc? How does it all hook up to the PC and substitute for a sound card? I already have a great visual system on this PC, but I'd love to have a high-end audio solution (higher end than X-Fi and what not) as well.
Rubycon
Madame President
- Aug 10, 2005
- 17,768
- 485
- 126
Originally posted by: Oyeve
So there are a few of us computer guys and gals out their who have highend hi-fi equipment connected to their computers.
Does this count?
MS Dawn, how come all of your images come from that same root URL? Is that where you host images or something?
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 20K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 12K
-
Discussion Speculation: Zen 4 (EPYC 4 "Genoa", Ryzen 7000, etc.)- Started by Vattila
- Replies: 13K
-
Facebook
Twitter