The Successor to AC'97: Intel High Definition Audio 428
An anonymous reader writes "A few days back Intel announced the name to
its previously dubbed 'Azalia' next-generation audio specification due out by midyear, under royalty-free license terms. The
Intel High Definition Audio solution will have increased bandwidth that allows for 192 kHz, 32-bit, multi-channel audio and uses
Dolby Pro Logic IIx technology 'which delivers the most natural, seamless and immersing 7.1 surround listening experience from any native 2-channel source'. The architecture is designed on the same cost-sensitive principles as
AC'97 and will allow for improved audio usage and stability."
Re:OSS drivers? (Score:5, Informative)
Re:It is still onboard sound (Score:2, Informative)
use a DAC out of your case
just use digital out to a good A/V receiver
I can't tell if you're joking or not (Score:5, Informative)
Not true discrete channels? (Score:5, Informative)
This sounds like it could be more smoke and mirrors, though there really isn't enough information to be sure.
ProLogic IIx will "synthesize" multiple channels from a stereo or 5.1 source. I sincerely hope Intel isn't thinking "we can do the same old thing (stereo) and marketing folks can call it 7.1 multichannel because we put this Dolby fake surround processing in the chip!"
Despite how much ProLogic has advanced, it still doesn't hold a candle to true, *discrete* 6+ channel sound (like DD/AC3 or DTS).
Re:It is still onboard sound (Score:5, Informative)
br> there is
digital out is common on today pc (either optical or coax) and any good A/V receiver with integrated decoder is able to convert the signal from digital to analog
Re:It is still onboard sound (Score:1, Informative)
Or you can just use USB; the EMagic 2|6 and Edirol UA-5 work well under ALSA and are prosumer studio quality for $200-$300. If you're into softsynth the USB stuff also tends to have low latency, and still works nicely on a laptop.
Re:It is still onboard sound (Score:1, Informative)
Uh... the computer I'm typing this on, with an Asus P4P800 [asus.com] motherboard has a built-in digital coaxial S/PDIF audio output (yes, the motherboard's built-in sound chip - and it does also have analog outputs). I can plug it into my Sony receiver which finds and decodes the digital signal just fine. I've not tried to get surround sound going because I don't have any surround sources, unfortunately.
Of course, that doesn't mean that the shovelware manufacturers (Dell, HP) are going to have digital outputs any time soon - but the moral is "build it yourself" to get the good stuff.
Re:That's audio ? (Score:2, Informative)
Even if frequencies that high cannot be heard, using such a sampling rate will decrease the noise added by analog->digital conversion.
Re:That's audio ? (Score:1, Informative)
Re:That's audio ? (Score:3, Informative)
Re:Isn't this just a bit much? (Score:1, Informative)
"How much of that clarity was due to the excellent sound engineers they probably hired?"
Most of it. The remasters come from the same source as the original: very high quality analog tape.
On the other hand, I can tell a big difference in dynamics, recording piano at 24/96 versus 16/44. Say what you want about "inaudible this" and "overkill that." Better headroom is higher fidelity. It's just not as important for playback as it is for recording. And, sadly for the future of our human rights, more poeple choose to playback than to record.
Re:Isn't this just a bit much? (Score:3, Informative)
Clean, no scratches and if I recall, the Japan import 1984 cd was worth a mint
Anyhow... we played each one and came to the result that the 2 channel 30th anniversary remaster was far superior, even on a great system, and the surround mix was simply amazing to hear
Re:That's audio ? (Score:2, Informative)
Re:That's audio ? (Score:5, Informative)
for example, a 22050hz sine tone (if you can hear that high!) sampled at 44100hz is only sampled twice per cycle, and would effectively be recorded as a square wave (although, admittedly at that frequency you'd need to be a dog to tell the difference!)
This is completely and utterly wrong. I hear this very often though.
At 44100Hz sampling, a 22050Hz signal will be reconstructed as a 22050Hz SINE WAVE. The reconstruction of sampled signals is not as simple as you think it is. This is covered in any elementary DSP book.
With IDEAL equipment sampling at frequency N allows perfect reconstruction of all frequencies N/2 in all cases. The rather = comes about because of the potential of sampling the frequency N/2 at the zero crossings. However, if only two nonzero points are sampled of the N/2 component, it can be reconstructed perfectly.
Using a higher sampling rate has to do more with counteracting clock jitter and the error introduced by non ideal equipment.
Re:I can't tell if you're joking or not (Score:1, Informative)
-greg
Definitely some fishy Marketing going on here (Score:5, Informative)
Re:7.1? (Score:5, Informative)
4.1: Front Left, Right; Mid Left, Right
5.1: Front Left, Right, Center; Mid Left, Right
6.1: Front Left, Right, Center; Mid Left, Right; Back Center
7.1: Front Left, Right, Center; Mid Left, Right; Back Left, Right
I always thought the mids ended up being farther back than shown in the picture though.
Re:Initial reaction (Score:2, Informative)
Unfortunately SPDIF is not bit-perfect by no means, you need ASIO for that. An easy way to tell is to play a Dolby Digital or DTS
Re:7.1? (Score:3, Informative)
It's basically : Left, Center, Right; SurroundX(left,rear left, rear right, right). Total overkill IMHO, 5.1 is good enough for me.
Re:Initial reaction (Score:1, Informative)
Re:7.1? (Score:3, Informative)
In the movie world, a 7.1 audio mix usually means a 5.1 surround mix plus a conventional 2 channel stereo mix. You can synthesis a conventional stereo mix from a 5.1 surround mix, but the result may vary. That is why some movies are mixed in 7.1, which really is both a 5.1 and a stereo mix.
When the movie is distributed on DVD or used in cinemas they use the 5.1. When the movie is sent on TV (eg. PAL with NICAM), you get the stereo mix.
Re:double-blind, controlled test, please? (Score:5, Informative)
Yep, you're denfinately a physics teacher, not an EE.
44 KHz sampling rate only lets you record frequencies up to 22KHz if you had a PERFECT d/a convertor and a PERFECT filter. It is provably impossible to implement a perfect filter. (One with a perfect cutoff and a perfectly flat passband.) Sampling at 44 KHz allows someone to design a decent recording setup with compenents that actually exist. Sampling at 96KHz gives the engineer even more breathing room when designing the filter in front of the A/D convertor. Instead of going from H(jw)=1 to H(jw)=0 in the space of 2KHz, he now can do it in 20. This means he can use a filter design with a flatter pass band. This means there is less distortion of all those frequencies that you can actually hear.
Even if there was a hypothetical human who could hear 30 kHz, there would be many other things preventing it from being useful musically. For instance, your tweeters most likely can't respond well to those frequencies. Furthermore, the music might sound worse to such a person if the 30 kHz stuff was left in.
Actually, it's much easier to build a tweeter than can handle 30KHz, than it is to build a subwoofer that can handle 20Hz. There are plenty of tweeters on the market right now which claim to work at 30KHz.
Second, your statement about the 30KHz stuff making the music sound worse doesn't make any sense. The goal of an audiophile-quality setup is to reproduce the original audio exactly. We're not talking about adding in some strange 30KHz waveform, we're talking about preserving the signals that were there in the first place.
People who really want to hear good stereo sound should spend their effort on the two things that will make a lot of difference: (1) getting good speakers, and (2) working on the acoustics of the room, the placement of the speakers in the room, and the placement of their own head in the room. Note that all the stuff under #2 is free or cheap.
Actually, they should buy a good pair of headphones. For $300 they can buy a pair of headphones that would be tough to beat with speakers at 10X the price.
Re:I prefer OSS (Score:5, Informative)
The distros that have shipped ALSA as default, like SuSE, have had pretty good dummy-proof setup of ALSA for a while. Probably every major distro will be using ALSA in 2.6, which means that the remaining OSS/Free holdouts, like Red Hat, will be doing up easy-to-use UIs for ALSA.
I also stopped using ALSA a while ago -- it was just a pain in the ass to recompile the alsa-driver package each time I upgraded the kernel, and all the software I use also supports an OSS interface (and *most* was using ALSA through the OSS compatibility interface). I expect I'll be using it again in 2.6.
Re:I can't tell if you're joking or not (Score:1, Informative)
Protocol vs. controller (Score:5, Informative)
Don't get me wrong, AC97 is cheap, but it really dragged on the CPUs of the timeframe it came out.
Well, that's not really AC97's fault.
AC97 is really nothing more than a 5 wire signal specification. It has more to do with voltages and waveforms on wires. And a register set in the codec that the wires are talking to.
But that's the idea of AC97 - you don't need to know who made the codec, only that it's AC97. Then it's a drop in replacement, pretty much.
But controllers - everybody and their brother has a different idea how to talk to an AC97 codec. And it's the controller that determines the performance. Are you bit banging your codec? Then performance will suck. Are you using interrupts? Performance will improve. Using DMA? Performance will improve again. Does your DMA engine suck? Performance will drop.
If you're having a drag on your cpu due to audio, it isn't AC97 that's at fault. It's someone's lousy idea for a controller. AC97 is a spec, not a gadget.
Weaselmancer
The labels will botch it (Score:3, Informative)
The main reason you need more than 16 bits is because, during soft passages, most of the high bits are zero and you may effectively have only six or four bit audio. Classical recordings that aren't compressed really do suffer from this problem.
But really, the number of people who buy classical piano recordings is small.
If the industry can agree that the reference level for popular audio is somewhere well below 100%, this could work out. But that won't happen.
Re:double-blind, controlled test, please? (Score:4, Informative)
Yes, 96/192 kHz sampling is a good thing for recording studios for the reason that you explain. Moreover, >=24 bit recording means that you don't get aliasing problems if the signals are amplified or attenuated during the mixing process.
However, this is all on the recording side. After sampling at >=96 kHz, you can apply a digital filter with a perfectly flat passband up to 20 kHz and stopband above 22.05 kHz, and then downsample to 44.1 kHz. In any CD player, the opposite process is performed (the famous "oversampling"): it is hard to filter the noise above 20 kHz in the raw 44.1-kHz signal. Therefore, the DAC converts the signal digitally to a 4 to 16 times higher sampling rate and with a slightly higher bitresolution (e.g. 18 or 20 bits). Then, the DAC digitally filters out everything above 22 kHz while leaving everything below 20 kHz.
The (still digital) signal is now a "smooth line" through the supplied data points at 44 kHz. This signal is converted to a voltage by the true (non-signal-processing part of the) DAC. The part of the spectrum below 20 kHz will be exactly the same independent on whether the original input to the DAC was 44, 96, or 192 kHz. (Note: 1-bit DA convertors use a slightly different approach, but with the same result).
As far as the bit resolution is concerned: in the final signal, 16 bits is enough for a dynamic range of 92 dB. If the hearing treshold is at 0 dB, that means that for peak levels of less than 92 dB, the resolution is fully sufficient to encode even the softest audible sounds. Note that 92 dB is quite loud: about 4 W power to a typical 87 dB/W loudspeaker at 1 m distance. It is defendable to use a bitresolution higher than 16, if you want to hear a ticking watch in the background while the music is playing at the pain treshold of 120 dB. For that, you need 5 more bits: 21 bits. On the consumer end, 24 or 32 bits is a waste of storage space.
Re:It is still onboard sound (Score:3, Informative)
A much better idea is to run a digital link to an outboard DAC that has its own power supply and is outside the computer. That would actually give you extremely high quality audio, assuming the DAC box is properly designed.
Re:It is still onboard sound (Score:5, Informative)
That said, I actually think 32bit audio may be at least 8 bits overkill. I'm all for 192Khz, because we can actually hear a difference in the resolution of the wave. 16bit audio allowed for 64K levels that were smoothed between. Most audio is pretty smooth sounding, and I doubt you can hear any difference between 16 and 32 bit unless you crank the volumn up to a level that could damage your hearing.
Also, 32bit DACs are practically impossible to buy last time I checked. A full 16bit DAC is pretty expensive relatively and it's exponentially more complicated with each bit to build a proper DAC. I'm expecting a lot of shortcuts. A 32bit ADC for recording is prohibitively expensive, so I gaurantee you won't be doing any 32 bit recording any time soon on a PC.
Basically, the 32bit idea is dead in the water. The machine will be long gone before any audio is distributed that takes advantage of it. You probably can't use it for mixing because you probably won't be able to record at 32bit. It's also going to be more expensive in components. Speakers aren't going to be accurate enough to 32bits of resolution. They may shoot for 24 bit, because you can get an OK DAC and ADC for working with 24 bits, but it'll still cost.
The 192Khz thing is awesome. Right now, you can get 48Khz out of some consumer cards, but 192 would be excellent. Maybe we'll get digital audio up to proffesional quality some day. Right now if you go get a recording from a studio, you get tape (unless you can't afford it). All professional audio equipment is not only analog end-to-end, it's also usually tube based. The average transistor is pure sewage, and even MOSFETs are lacking. There's gotta be a lot more R&D into just transistors before we have professional grade audio going anywhere near digital. This is still going to be helpful to the end user that likes music, but we are still a long way off from having no audible differences. Amazingly enough, I think speaker technology has advanced more over the last decade than digital audio.
Re:double-blind, controlled test, please? (Score:4, Informative)
Quantization errors occur in the less signifigant bits, a high quality ADC will have an uncerainty of about + or - 4 bits. Think of a 10khz signal on the edge of human hearing like a nice china boy cymbal -- a cycle of a 10khz audio signal will be represented by about 4.41 samples :) I know the nyqist limit/shannons theorom says thats enough, but out here in the real world where there's noise and quantization errors its not enough, which leads me to my next point **the nyquist limit is valid only for situations where there is no noise** in other words: THERE IS NO SITUATION FOR WHICH THE NYQUIST LIMIT IS VALID. The Nyquist limit is at best, a guideline.
So now the reason you need higher resolution/bigger samples is because that alters the noise floor. + or - 4 bits in a 24 bit recording is alot less signifigant then + or - 4 bits in a 16 bit recording. Also, imagine at 192khz your 10khz signal is now represented by 19.2 samples -- error and noise is MUCH less destructive with more samples.
I deal with these issues every day in my studio, and the rule with audio is pretty much always, more is better. However, There is a point of diminishing returns -- and IMHO I think that point is 24bit/96khz. It is very difficult to distinguish a 96khz signal from a 192khz signal.
Re:I can't tell if you're joking or not (Score:2, Informative)
Look at any of the commercial DVD-audio or SACD players available- none of them support digital output at 192khz/24bit. If one was available, anyone could bypass the huge amounts of DRM/watermarking on those new discs, and make bit-perfect copies by simply plugging it into your soundcard/dat recorder.
Anyways, the S/PDIF standard doesn't support bit rates high enough for 192/24 audio, so an entirely new format would have to be made, and somehow i doubt the RIAA will allow one to be made.
Of course, digital output at 44.1/16, (well AC'97 resamples it to 48khz) will still be available, and that's more than good enough for 99.999% of the market.
So basically, this high-res stuff is nothing more than a marketing ploy, there is no way you can achieve 24-bit performance on a noisy switching powesupply while blasted by EMI, reproduced by a 99 cent DAC/opamp chip. (well there's no way to achive 24 bit performance period at room temperature, since the johnston resistor noise of any system is greater than the resolution of a 24bit system, but that's another matter altogether)
Re:It is still onboard sound (Score:1, Informative)
Also, the highest sample rate available is 192khz not 192bps.
Re:It is still onboard sound (Score:1, Informative)
Re:Isn't this just a bit much? (Score:3, Informative)
Standard CDs are sampled at 44.1Khz, so the highest frequency they could possibly store is a sound at 22.05kHz. However, this doesn't meant that they will reproduce anything less than that with perfect accuracy. Firstly, the sound needs to be filtered to prevent anything over 22.05kHz hitting the convertors (as they'd cause very nasty artefacts); this filtering has a lower cut-off (usually around 20kHz) for safety, and although the filter has a steep response, it's not infinite; it'll reduce some lower frequencies too, and it'll also cause phase changes at lower frequencies. (I gather current filters are much better than those used for early CDs, which were responsible for much of the early complaints.) Filters are also needed in the player, which also affect the sound.
Greater sample rates would allow much gentler filters to be used, which would have less (or no) effect on audible frequencies, even those above 20kHz.
Secondly, it's claimed that although we can't hear sound at those higher frequencies, we can detect phase changes and timing changes occurring faster than CD can store; the additional timing resolution would help with that.
And thirdly, in the studio (and wherever sound is processed) the tiny changes caused by filtering and slight timing shifts can add up, to the point (it's claimed) where they can have a very audible result. The extra frequency and time resolution, just like the extra sample resolution of 24 or 32 bits, allows mixing and other processing to be done with less loss.
So there are reasons why 96kHz or 192kHz and 24- or 32-bit sound might provide real benefits. I'm unlikely to hear them myself -- I'm a musician, not an audiophile or sound engineer -- but as technology gets more powerful, faster, and cheaper, I'm sure sound quality will only improve.
(If the RIAA doesn't stop it... Oops, a little bit of politics there, yes indeed.)
Re:It is still onboard sound (Score:4, Informative)
You mean the top 16 bits at 65536x the voltage, and the other handling the low 16. Else you've just produced a 17 bit DAC.
Re:Can you recommend some computer speakers? (Score:3, Informative)
That said, I settled on the Logitech speakers for my computer after a lot of listening--they're the only ones I found that sounded like music. I will admit that I didn't listen to the Klipsches, because they were out of my price range. I expect that they'd be quite good, as they make non-computer speakers which are very nice indeed. (Mind you, Altec-Lansing makes stereo speakers too, and their computer offerings are without exception, unmitigated shite!)
If you have a passable amp, then unpowered stereo speakers are likely to be the best choice. A few years ago, it would have been the only choice, but a few computer speakers are at least considering.
But ignore the 200watt rating. Even if it were valid (it's not), it's completely meaningless and irrelevant to your shopping.
Re:It is still onboard sound (Score:1, Informative)
As for the tubes comment... I am rolling my eyes. The transistor is well understood. Mosefets operate like the way tubes *should* have worked. Tubes are flawed, get over it. You may like the sound, but think of it as a sort of postprocessing, it is not the true signal.
Re:double-blind, controlled test, please? (Score:3, Informative)
Wrong. 180 degrees will indeed sample at the valleys instead of the peaks but the magnitude is the same, only the sign is different. Perhaps you meant 90 degrees.
Nope. Still wrong.
For your own education, here is Nyquist's Theorem.
Notice the language "without error". There is no error. It's hard to grok, and impossible to believe without doing the maths, but it is 100% true.
Though as I said before, the real world is more fun because sampling is never exact. Errors in the times when samples are taken and errors in the magnitudes of the samples will screw you.
Re:double-blind, controlled test, please? (Score:1, Informative)
A number of us, myself included, could hear up to 25 KHz on it. Either his system wasn't "in tune" or human hearing is alot better than researchers give it credit for. The professor didn't have anything to say during the class when we protested we could hear frequencies that high.
Anyway, 44.1 KHz sampling does not mean you can get 22 KHz audio out of a CD. In practice there is a transition band right around 17-18 KHz. So in reality you must filter out frequencies around that point to avoid aliasing. So 44 KHz audio definitely does not cover human hearing up to 20 KHz accurately. It also isn't perfect near the bass end of the spectrum.
48 KHz comes closer, but 96 KHz looks alot better when you analyze the frequency response on a scope. And if I remember my DSP class correctly an even higher sampling rate lets you get away with simpler DACs.
Conclusion? Bring on the high sampling rate!
Re:It is still onboard sound (Score:3, Informative)
Here's one [markertek.com]
This box allows you to use spdif with your existing analog stereo.
Specs here [rdlnet.com]
Re:It is still onboard sound (Score:5, Informative)
I would rather take the one that can be fixed with an equalizer.
Where transistors really rule though is low power usage. A class A tube amp will keep you warm at night without even actually making noise. We need better transistors, but I'm not saying we need tubes everywhere.
What you are saying about needing twice the sampling rate is complete BS. Between that remark and the tube vs MOSFET remark, I can tell you care very little about the wave form.
If I tried to make a 20khz sound with 40khz sampling, I wouldn't hear a thing. Take two samples of one wave equidistant, and you get the middle and the beginning of the wave, which are always the same value. At 4X the sampling, you get a saw wave (or worse, a muffled trapazoid if you phase shift 45 deg). So, if you do 20Khz at 80Khz, you're still screwed. How many points do you need on a wave to make it smooth? I would say at least 8 at high frequencies (and that has a chance of only getting you about 66% of the power). That's about 160Khz for 20Khz sound.
I hate that someone actually thought the whole 2X the audible range was good enough to begin with. You may not hear a difference, but I can. If you don't, check the frequency response of your speakers.
If you want to calculate how many samples it takes to get >90% power, you should calculate the distance on the wave that is 90% power, then divide the length of the full wave by that distance, then multiply by 20Khz. So, here's a trusty sine table for you already measured in percentage of wave:
1/.14 = 7.14
20KHz*7.14=143KHz
This isn't RMS calibrated, but so what.
192KHz:
2pi/9.6=.65 radians
sin(pi/2-pi/9.6)= 94.69% power output.
Basically, it's the sine of half the distance away from the peak to the furthest point out. Why is is it not the average? If you take the average, then you are forgetting that you will miss the right side of the wave completely. The only case better is sometimes when you hit the exact peek.
Also, you have to consider that this is going to create distortion too. Consider that the resonance of 20KHz with the actual output level, since it varies around the 94% mark.