Don't forget Antares was missing from the night sky; I cling to my theory that it going supernova damaged to ozone layer sufficiently to precipitate an ice-age that dropped the ocean levels, closing Gibraltar.
When MPEG LA first announced the VP8 pool formation, a rush of companies applied to be in the pool, partly because everyone wanted to see what everyone else had. That gave way to some amount of disappointment. And by 'some amount' I mean 'rather a lot really, more than the MPEG-LA would care to admit.'
Eventually, things whittled down to a few holdouts. Those '11 patent holders' do not assert they have patents that cover the spec. They said '_may_ cover'. The press release itself repeats this. Then these patent holders said 'and we're willing to make that vague threat go away for a little cash'. Google paid the cash. This is what lawyers do.
That's why it's a huge newsworthy deal when companies like NewEgg actually take the more expensive out and litigate a patent. It is always more expensive than settling, even if you'd win the case, and very few companies are willing or able to do it. Google was probably able, but not willing.
We deal with this in the IETF all the time. Someone files a draft and a slew of companies file IPR statements that claim they have patents that 'may' read on the draft. Unlike other SDOs though, the IETF requires them to actually list the patent numbers so we can analyze and refute. And despite unequivocal third-party analyses stating 'there is no possibility patent X applies', these companies still present their discredited IPR statements to 'customers' and mention that these customers may be sued if they don't license. This is not the exception; this is standard operating procedure in the industry. These licensing tactics, for example, account for more than half of Qualcomm's total corporate income.
It's this last threat that Google paid a nominal sum to make go away. It's the best anyone can hope for in a broken system. If those 11 patent holders had a strong claim, it is exceedingly unlikely they would have agreed to a perpetual, transferable, royalty free license.
"After a decade of the MPEG LA saying they were coming to destroy the FOSS codec movement, with none other than the late Steve Jobs himself chiming in, today the Licensing Authority announced what we already knew.
They got nothing. There will be no Theora patent pool. There will be no VP8 patent pool. There will be no VPnext patent pool.
We knew that of course, we always did. It's just that I never, in a million years, expected them to put it in writing and walk away. The wording suggests Google paid some money to grease this along, and the agreement wording is interesting [and instructive] but make no mistake: Google won. Full stop.
This is not an unconditional win for FOSS, of course, the LA narrowed the scope of the agreement as much as they could in return for agreeing to stop being a pissy, anti-competetive brat. But this is still huge. We can work with this.
For at least the immediate future, I shall have to think some uncharacteristically nice things about the MPEG LA.*
>If you insist on recording in stereo though, you might do as they did, and record with a Mid-Side array and use a matrix to decode back to L-R, so you can control the stereo spread in post-production.
That would not have controlled the reverb; the space this was recorded in was a concrete floor with concrete walls and no acoustic treatment. Like I said, it was a tradeoff, and one that was successful if not perfect.
Oh! And 'linear' was completely wrong. I don't know how I braino-ed that in there, the [at very least perceived] distortion/noise tradeoff is not linear.
Right, and this is why dither is only applied to 'last-mile' audio intended to be consumed. Dither 'screws' you in other ways if you intend to use that audio in production, such as losing all the property of removing the distortion, yet still having the additive noise. But we're still talking about changes 100+dB down.
>Counter nitpick: Monty, as a professional motion picture sound designer, I cannot tell you how distracting it is to hear your voice constantly changing its pan across the stereo field:)
The audio was recorded with a stereo pair. It wasn't panned artificially:-) Look down a few comments for more about this, you weren't the only person to complain.
As a nitpick, you get dithering losses _or_ quantization distortion, or a linear tradeoff between the two. You don't get the worst case of both on top of each other unless you screw up.
Without dither, worst case, all your 16 bit quant distortion products will be under -100dB regardless of input amplitude. I actually display the worst case in the video to make it easy to see. Quantization distortion aliases, and I chose an integer sample period so the aliased distortion would always land in the same bins after folding. If I hadn't, it would have spread out more and been even lower. If I had chosen a relatively prime frequency, the quantization distortion would have spread out across all bins equally.
> Stop using cdparanoia - it isn't very good, at all. It tests poorly, we're sad to say.
Really! As the author, I'd love to hear hard specifics. or maybe a bug report.
> You want to use Secure Mode with NO C2, accurate stream, disable cache.
You can't disable the cache on a SATA/PATA ATAPI drive. The whole point of cdparanoia's extensive cache analysis is to figure out a way to defeat the cache because it can't be turned off. There is no FUA bit for optical drives in ATA or MMC.
The 'accurate stream' bit is similarly useless (every manufacturer interprets it differently) and C2 information is similarly untrustworthy.
Plextors are not recommended for error free or fast ripping. They try to implement their own paranoia-like retry algorithm in firmware and do a rather bad job about it. They also lie about error correcting information (you do not get raw data, you get what the drive thinks it has successfully reconstructed). Plextors often look OK on pristine disks, but if you hit a bit error (like on just about any burned disk), you don't know what it's going to do. Plextors are, overall, among the more troublesome drives _unless_ you're using a ripper that does no retry checking (ie, NOT cdparanoia and NOT EAC). If you use iTunes, you want a Plextor. Otherwise, avoid them.
That was true ~15 years ago. Since then, Plextor's firmware gets along very badly with the rippers that try to be frame accurate, because Plextor tries to implement a much lighter-weight more error prone version of the same algo on the drive. The drive still doesn't do a realiable job, and it seriously mucks up the ripper.
Regarding hte first point, that 120dB broadband noise figure is giving you at least 140dB of SFDR, probably much better, and the depth of any critical band is going to be even better yet. Even 16 bit data with a decent noise shaper is going to be 120dB deep in the 2-4kHz critical bands. (all of which doesn't disagree with anything you said of course)
Oh! I remember this one:-) I'll be honest here-- this particular debate is outside my core expertise. I have enough background to say "this is all plausible" (I'm an electrical engineer after all), and I've discussed it in person with an author of a few papers on the same subject, but I'm only a dilletante when it comes to building amps.
>I guess my point is that it's too easy to make an error when seeing an "interesting idea and no data" and dismissing it.
I agree with you completely. Interesting ideas should be published; no paper is born in the state of being independently verified. I object to those who take these papers as evidence to support a position when no such validation has taken place. Thinking aloud is useful, but thinking aloud != hard data.
I'm actually giggling that 20 bits of real dept is 'disappointing' in any sense of the word. It's bloody fantastic!:-) (I hadn't meant to trash the ESS chip in any sense. I've not used it, but it looks impressive)
Well, for the record, I've not been rejected, but I've only published within AES once.
It's not an attack, it's more a statement of truth. The AES publishes all sorts of things. Papers with interesting ideas and no data (eg, the J. Dunn 'equiripple filters cause preecho' paper, which presents a fascinating insight, even if it doesn't work out in practice), papers with data that are effectively WTFLOL (the famous Oohashi MRI paper) and papers that are more careful controlled studies. It runs the whole gamut on both sides, just as I said.
Do you deny that a substantial portion of the membership, including many elders of the group, are not 'bigger numbers are always audibly better' audiophiles? It was Andy Moorer himself who, with no hard data, kicked off the insane sampling rate race that now has some hardcore audiophiles wondering if 192kHz is enough.. they're holding out for 384kHz!
Is the AES a worthless cesspool? Oh heck no. Never said that. But treating its publications as more than a good industry rag (where it's sometimes hard to tell the research from the advertisements).. or perhaps an advanced debating club... is probably not a very good idea. Treating any one AES paper as gospel is just insane.
Parent post is talking about broadband SNR, not narrowband depth or SFDR. Very few ADCs/DACs reach 20 bits SNR. Almost all exceed 20 bits SFDR. Noise shaping has nothing to do with it; it actually penalizes SNR even though it is generally a [great] benefit to perceived fidelity.
Assuming from the press release that this is an apodizing filter that 'removes' Gibbs effect preringning... how many peer reviewed studies can we compile here below my post that indicate anyone can hear these 'artifacts'?
Ready, on your marks, go!
Be careful of publications from the businesses that are pushing these filters (eg, Meridian audio / J. Robert Stuart). AES papers count only if the results have been independently reproduced (the AES is in the business of publishing 'interesting ideas', and the papers run the gamut of careful science to raving lunacy).
"I listened and _totally_ heard a rounder, fuller sound with better staging" is not data. It's what audiophiles have said of every nonsense breakthrough of the past 40 years. The one true breakthrough, digital audio, they generally still roundly pan and feel the need to 'fix' by sprinkling fertilizer all over it. It's like holy penguin pee, only it smells bad.
'Proofs' and explanations are nice, I indulge in them myself, but the blind listening data is the final authority.
We had about six years where Linux wasn't the unsupported red-headed stepchild of the Web. It was nice while it lasted.
Firefox is losing market share, so it's understandable that they want to avoid the following: "We notice you have an unsupported browser. Please download Safari or IE to view this content". That wouldn't really serve anyone's interests.
Unfortunately, this just pushes the problem onto Linux as a whole: "We notice you have an unsupported OS. Please use Mac OS or Windows to legally view this content."
If you turn the samples up until you can hear the noise floor, you can easily hear the difference. Of course, at those levels, a full range signal would launch your speaker cones out of the cabinets. So is that a fair comparison of 16 vs 24?
There are any number of ways to cheat an ABX test to your own satisfaction. If the goal is to delude yourself, you'll probably succeed.
That copypasta hasn't been funny for at least five years if ever.
If you wanna troll, let's go... I'll take your side, you take mine and no one under the age of thirty will have any freaking clue what just happened. >>>/g/
Indeed. One of the overlooked but highly important issues with sampling rates is that although you can represent up to Nyquist in a periodically sampled signal, that is the limit for infinite length recordings. For finite-length recordings, it isn't all or nothing, represented perfectly or not at all -- instead the uncertainty (read: representation error) increases as you approach Nyquist.
Too bad Shannon and Nyquist are dead. It seems they've completely misunderstood the math. How embarrassing they passed on before you could correct their mistake. Now they'll never know.
Slashdot Mirror
"It is a damn poor mind indeed which can't think of at least two ways to spell any word."
--Andrew Jackson
Don't forget Antares was missing from the night sky; I cling to my theory that it going supernova damaged to ozone layer sufficiently to precipitate an ice-age that dropped the ocean levels, closing Gibraltar.
When MPEG LA first announced the VP8 pool formation, a rush of companies applied to be in the pool, partly because everyone wanted to see what everyone else had. That gave way to some amount of disappointment. And by 'some amount' I mean 'rather a lot really, more than the MPEG-LA would care to admit.'
Eventually, things whittled down to a few holdouts. Those '11 patent holders' do not assert they have patents that cover the spec. They said '_may_ cover'. The press release itself repeats this. Then these patent holders said 'and we're willing to make that vague threat go away for a little cash'. Google paid the cash. This is what lawyers do.
That's why it's a huge newsworthy deal when companies like NewEgg actually take the more expensive out and litigate a patent. It is always more expensive than settling, even if you'd win the case, and very few companies are willing or able to do it. Google was probably able, but not willing.
We deal with this in the IETF all the time. Someone files a draft and a slew of companies file IPR statements that claim they have patents that 'may' read on the draft. Unlike other SDOs though, the IETF requires them to actually list the patent numbers so we can analyze and refute. And despite unequivocal third-party analyses stating 'there is no possibility patent X applies', these companies still present their discredited IPR statements to 'customers' and mention that these customers may be sued if they don't license. This is not the exception; this is standard operating procedure in the industry. These licensing tactics, for example, account for more than half of Qualcomm's total corporate income.
It's this last threat that Google paid a nominal sum to make go away. It's the best anyone can hope for in a broken system. If those 11 patent holders had a strong claim, it is exceedingly unlikely they would have agreed to a perpetual, transferable, royalty free license.
I'll add my own thoughts here, also posted at http://xiphmont.livejournal.com/59893.html
"After a decade of the MPEG LA saying they were coming to destroy the FOSS codec movement, with none other than the late Steve Jobs himself chiming in, today the Licensing Authority announced what we already knew.
They got nothing. There will be no Theora patent pool. There will be no VP8 patent pool. There will be no VPnext patent pool.
We knew that of course, we always did. It's just that I never, in a million years, expected them to put it in writing and walk away. The wording suggests Google paid some money to grease this along, and the agreement wording is interesting [and instructive] but make no mistake: Google won. Full stop.
This is not an unconditional win for FOSS, of course, the LA narrowed the scope of the agreement as much as they could in return for agreeing to stop being a pissy, anti-competetive brat. But this is still huge. We can work with this.
For at least the immediate future, I shall have to think some uncharacteristically nice things about the MPEG LA.*
*Apologies to Rep. Barney Frank"
>If you insist on recording in stereo though, you might do as they did, and record with a Mid-Side array and use a matrix to decode back to L-R, so you can control the stereo spread in post-production.
That would not have controlled the reverb; the space this was recorded in was a concrete floor with concrete walls and no acoustic treatment. Like I said, it was a tradeoff, and one that was successful if not perfect.
Oh! And 'linear' was completely wrong. I don't know how I braino-ed that in there, the [at very least perceived] distortion/noise tradeoff is not linear.
Right, and this is why dither is only applied to 'last-mile' audio intended to be consumed. Dither 'screws' you in other ways if you intend to use that audio in production, such as losing all the property of removing the distortion, yet still having the additive noise. But we're still talking about changes 100+dB down.
>Counter nitpick: Monty, as a professional motion picture sound designer, I cannot tell you how distracting it is to hear your voice constantly changing its pan across the stereo field :)
The audio was recorded with a stereo pair. It wasn't panned artificially :-) Look down a few comments for more about this, you weren't the only person to complain.
I think it's more "I want people to know why I do the stupid things I do." Latent fear of being committed.
As a nitpick, you get dithering losses _or_ quantization distortion, or a linear tradeoff between the two. You don't get the worst case of both on top of each other unless you screw up.
Without dither, worst case, all your 16 bit quant distortion products will be under -100dB regardless of input amplitude. I actually display the worst case in the video to make it easy to see. Quantization distortion aliases, and I chose an integer sample period so the aliased distortion would always land in the same bins after folding. If I hadn't, it would have spread out more and been even lower. If I had chosen a relatively prime frequency, the quantization distortion would have spread out across all bins equally.
> Stop using cdparanoia - it isn't very good, at all. It tests poorly, we're sad to say.
Really! As the author, I'd love to hear hard specifics. or maybe a bug report.
> You want to use Secure Mode with NO C2, accurate stream, disable cache.
You can't disable the cache on a SATA/PATA ATAPI drive. The whole point of cdparanoia's extensive cache analysis is to figure out a way to defeat the cache because it can't be turned off. There is no FUA bit for optical drives in ATA or MMC.
The 'accurate stream' bit is similarly useless (every manufacturer interprets it differently) and C2 information is similarly untrustworthy.
Plextors are not recommended for error free or fast ripping. They try to implement their own paranoia-like retry algorithm in firmware and do a rather bad job about it. They also lie about error correcting information (you do not get raw data, you get what the drive thinks it has successfully reconstructed). Plextors often look OK on pristine disks, but if you hit a bit error (like on just about any burned disk), you don't know what it's going to do. Plextors are, overall, among the more troublesome drives _unless_ you're using a ripper that does no retry checking (ie, NOT cdparanoia and NOT EAC). If you use iTunes, you want a Plextor. Otherwise, avoid them.
That was true ~15 years ago. Since then, Plextor's firmware gets along very badly with the rippers that try to be frame accurate, because Plextor tries to implement a much lighter-weight more error prone version of the same algo on the drive. The drive still doesn't do a realiable job, and it seriously mucks up the ripper.
Regarding hte first point, that 120dB broadband noise figure is giving you at least 140dB of SFDR, probably much better, and the depth of any critical band is going to be even better yet. Even 16 bit data with a decent noise shaper is going to be 120dB deep in the 2-4kHz critical bands. (all of which doesn't disagree with anything you said of course)
Oh! I remember this one :-) I'll be honest here-- this particular debate is outside my core expertise. I have enough background to say "this is all plausible" (I'm an electrical engineer after all), and I've discussed it in person with an author of a few papers on the same subject, but I'm only a dilletante when it comes to building amps.
>I guess my point is that it's too easy to make an error when seeing an "interesting idea and no data" and dismissing it.
I agree with you completely. Interesting ideas should be published; no paper is born in the state of being independently verified. I object to those who take these papers as evidence to support a position when no such validation has taken place. Thinking aloud is useful, but thinking aloud != hard data.
I'm actually giggling that 20 bits of real dept is 'disappointing' in any sense of the word. It's bloody fantastic! :-)
(I hadn't meant to trash the ESS chip in any sense. I've not used it, but it looks impressive)
Well, for the record, I've not been rejected, but I've only published within AES once.
It's not an attack, it's more a statement of truth. The AES publishes all sorts of things. Papers with interesting ideas and no data (eg, the J. Dunn 'equiripple filters cause preecho' paper, which presents a fascinating insight, even if it doesn't work out in practice), papers with data that are effectively WTFLOL (the famous Oohashi MRI paper) and papers that are more careful controlled studies. It runs the whole gamut on both sides, just as I said.
Do you deny that a substantial portion of the membership, including many elders of the group, are not 'bigger numbers are always audibly better' audiophiles? It was Andy Moorer himself who, with no hard data, kicked off the insane sampling rate race that now has some hardcore audiophiles wondering if 192kHz is enough.. they're holding out for 384kHz!
Is the AES a worthless cesspool? Oh heck no. Never said that. But treating its publications as more than a good industry rag (where it's sometimes hard to tell the research from the advertisements).. or perhaps an advanced debating club... is probably not a very good idea. Treating any one AES paper as gospel is just insane.
Oh, what do you know. In that case, go forth and unsubscribe :-)
As for me, it was 'membership'. Peer review means little when a substantial portion of the peers don't believe in the scientific method.
Dynamic range (not SNR) is 22.5 bits ("How quiet the chip can get when it's doing nothing/off"). SNR is 20 bits at best given the 120dB THD+N spec.
Surely you mean membership...?
ESS's own specs list it at THD+N of 20 bits, and SNR can't be higher than that.
This is also an 'ultimate' DAC, supposedly the Terminator of the current generation (in ESS's own marketing anyway)
Parent post is talking about broadband SNR, not narrowband depth or SFDR. Very few ADCs/DACs reach 20 bits SNR. Almost all exceed 20 bits SFDR. Noise shaping has nothing to do with it; it actually penalizes SNR even though it is generally a [great] benefit to perceived fidelity.
Assuming from the press release that this is an apodizing filter that 'removes' Gibbs effect preringning... how many peer reviewed studies can we compile here below my post that indicate anyone can hear these 'artifacts'?
Ready, on your marks, go!
Be careful of publications from the businesses that are pushing these filters (eg, Meridian audio / J. Robert Stuart). AES papers count only if the results have been independently reproduced (the AES is in the business of publishing 'interesting ideas', and the papers run the gamut of careful science to raving lunacy).
"I listened and _totally_ heard a rounder, fuller sound with better staging" is not data. It's what audiophiles have said of every nonsense breakthrough of the past 40 years. The one true breakthrough, digital audio, they generally still roundly pan and feel the need to 'fix' by sprinkling fertilizer all over it. It's like holy penguin pee, only it smells bad.
'Proofs' and explanations are nice, I indulge in them myself, but the blind listening data is the final authority.
We had about six years where Linux wasn't the unsupported red-headed stepchild of the Web. It was nice while it lasted.
Firefox is losing market share, so it's understandable that they want to avoid the following: "We notice you have an unsupported browser. Please download Safari or IE to view this content". That wouldn't really serve anyone's interests.
Unfortunately, this just pushes the problem onto Linux as a whole: "We notice you have an unsupported OS. Please use Mac OS or Windows to legally view this content."
If you turn the samples up until you can hear the noise floor, you can easily hear the difference. Of course, at those levels, a full range signal would launch your speaker cones out of the cabinets. So is that a fair comparison of 16 vs 24?
There are any number of ways to cheat an ABX test to your own satisfaction. If the goal is to delude yourself, you'll probably succeed.
That copypasta hasn't been funny for at least five years if ever.
If you wanna troll, let's go... I'll take your side, you take mine and no one under the age of thirty will have any freaking clue what just happened. /g/
>>>
Indeed. One of the overlooked but highly important issues with sampling rates is that although you can represent up to Nyquist in a periodically sampled signal, that is the limit for infinite length recordings. For finite-length recordings, it isn't all or nothing, represented perfectly or not at all -- instead the uncertainty (read: representation error) increases as you approach Nyquist.
Too bad Shannon and Nyquist are dead. It seems they've completely misunderstood the math. How embarrassing they passed on before you could correct their mistake. Now they'll never know.