Slashdot Mirror
The Successor to AC'97: Intel High Definition Audio
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."
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).
If only there were some way to have a digital output from the computer, and do the D/A conversion in a dedicated box.
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
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.
First off, 32 bit, 192 Khz, wants to appeal to those very serious about audio. 32 bit cards can have a dynamic range ratio of 144 db. That's beyond what normal humans can dfifferentiate, which is 120 db if we're lucky. Not only that, but professional 24 bit cards far exceed the needs and capabilities of most , if not every, user, with aaround 110 db of dynamic range. And they're going to put this mega high tech onboard? Hmm. 2ndly, the inclusion of Dolby. This is to appeal to the movie guys, but the real serious audio guys know that Dolby encoded audio is like an MP3, lossy compression. Serious audio guys will frown on that aspect. Incorporating these 2 aspects seems somewhat contradictory, which marketers always tend to do when trying to appeal to everyone. I, for one, remain highly skeptical. CD
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
Weaselmancer
rediculous.
Tubes generally have a flatter curve when comparing amps out to signal voltage, but MOSFETs have a flatter RMS Watts out compared to RMS signal level. Basically, MOSFETs screw up the wave form more than tubes, but manage to preserve the loudness at various frequencies better.
:)
.15 : .81 .18 : .90 .20 : .95 .25 : 1 .30 : .95 .32 : .90 .35 : .81 .32-.18=.14
/20KHz=9.6
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.
Karma Clown