Slashdot Mirror

I also spent 4 years studying an EE degree, and although it was not especially focused on signal processing, I now work for a large pro audio company.

Some of the issues pointed to in this and other posts regarding oversampling and AA filters are not really relevant to the subject at hand, given the technology currently in use. A statement like 'oversampling at 192 kHz' shows a lack of knowledge regarding the kinds of audio converters that have been in use for a good while now. A Delta Sigma ADC running with an Fs of 48 kHz might often be oversampling at 3.072 MHz or 6.144 MHz. Anti aliasing filters that many people have mentioned are implemented digitally inside the converter (no need for external analog filters, which may well exhibit many of the problems mentioned), and actually have extremely good pass band ripple.

Look at datasheets for converters from manufacturers such as TI (burr brown), cirrus [page 36 here has detailed plots of 48, 96, and 192 kHz pass pand characterisitcs for the device, highlighting the fact that increasing the sampling rate does not improve pass band ripple for this device (also note the scale is 0.02 dB/div)], AKM, Wolfson micro You will find pass band pass responses that are flat to within less than +/- 0.05 dB over the audible range, and stop band attenuation in excess of 100 dB, whether sampling at 48 kHz or 192 kHz. If you can find anything in actual converter datasheets that points to better converter performance from selecting a higher sampling rate, I would be interested to see it.

All in all, the basics of sampling theory don't really help people to understant the real world issues in designing a moden high end audio device. And in the end, surely the proof of the pudding is in the blind tests, that never seem to show that anybody can tell any difference when moving to higher rates? Even if there were a few people who could hear this difference in some perfect listening envirmonment, would it really make sense for everyone else to go out and buy 192 kHz equipment?

Ok, sick and tired of those who don't understand what is in a DAP:

iPods:
http://ipodlinux.org/wiki/Generations
Note they are all ARM processors with no hardware decoder.
Older ones are PortalPlayer
http://en.wikipedia.org/wiki/PortalPlayer
Newer ones are Apple labeled ARM chips
Also note on the wikipedia page how many DAPs use these chipsets.
Again, not hardware decoders.

Latest Sansas:
http://www.rockbox.org/twiki/bin/view/Main/SansaE200v2
http://www.anythingbutipod.com/archives/2008/03/sandisk-sansa-fuze-disassembly.php
Note AMS SoC
http://www.austriamicrosystems.com/eng/content/download/7921/128739/version/1/file/AS3525_PB_1v0.pdf

Older Sansas:
http://www.rockbox.org/twiki/bin/view/Main/SandiskE200HardwareComponents
Note PortalPlayer Soc
http://en.wikipedia.org/wiki/PortalPlayer

Note when you see the word "codec" when referring to a chip it is the D/A converter, not a MP3 decoding chip:
http://www.semiconductors.philips.com/acrobat/datasheets/UDA1380_4.pdf
http://www.wolfsonmicro.com/products/WM8987/

Philips:
http://www.rockbox.org/twiki/bin/view/Main/GoGearHDD6330

Cowon:
http://www.rockbox.org/twiki/bin/view/Main/CowonD2Info

Creative:
http://www.anythingbutipod.com/archives/2008/07/creative-zen-xfi-disassembled.php
look up the chip numbers - general purpose CPU, no hardware decoder.

Do you want me to link more?

Actually, I work with mixed-signal stuff where I work, including DACs/ADCs, filters, amps, and so on. I do some FPGA stuff here and there, but my main interest is analog. I am by no means an expert, but I work with analog experts on a day-to-day basis. And yes, I've read Jim's book, it is quite excellent.

As far as Sandisk: they are not a no-name brand, but they are mainly known as a manufacturer of solid-state storage (aka digital chips). Analog is not their expertise, and they are pushing the player on the basis of features, not sound quality. That tells me the sound quality is "good enough", rather than "excellent". Yeah, this is a heuristic, but a pretty damn good one.

As far as integrating the DAC with the headphone amp: it's an excellent way to improve sound quality while reducing costs. Modern DACs are made on a very high-performance process. A substantial part of the cost of the chip is the package. All other things being equal, a two-chip solution is going to be more expensive than a single-chip solution. To maintain low cost, lower-performance components will be chosen. If the headphone amp is integrated into the DAC, you can take advantage of the high-performance process for next to nothing. Performance isn't significantly reduced over a two-chip solution, since the DAC is largely an analog circuit. In any case, if you don't believe me, look at the specs of the chip Apple uses in the iPods: WM8971. This chip has amazing performance given its power consumption.

As I said, I am not an Apple fanboy. I just strongly believe in evaluating products based on their performance. It's hard to dispute the iPod's superiority over all of its competitors when it comes to the quality of the user interface, the software, and the hardware itself. In fact, I am tempted to plug it in to the Audio Precision system at work and see how well it really performs compared to my friend's iRiver player (sorry, don't know anyone with a sandisk player). Any serious discussion of sound quality is rather meaningless without at least a THD+N measurement.

Apple never intended for them to be used as audio recorders, and they have no control over the quality of third party dongles.

Your analogy is flawed or, rather, you are too passive. These are not "third party dongles", these are licensed and manufactured in partnership with Apple (that provides the firmware support and allows access to the iPod's innards). You don't get Apple's blessing, you don't get very far. Look at the incredibly slow progress the iPod Linux has made relative to, say, RockBox. This is because Apple actively works to lock out unauthorised development.

The iPod's hardware seems well capable of supporting high-fidelity recording, both analog and digital. The PortalPlayer PP5002B chipset (and derivatives on current models) used in all the big iPods since the early days is capable, according to PortalPlayer itself, of encoding MP3, WAV, AIFF, WMA, and ATRAC3 at up to 320Kbit/s.

A little over a year ago iPods switched to the Wolfson WM8731L ADC/DAC ($5 each in small lots!), which can sample at 44.1kHz, 48kHz or 96kHz. I haven't kept up with current iPod offerings because they are of little interest to me but I would assume Apple has not regressed on the ADC capabilities. It's hard these days to spend more than $3 on a signal chip and *not* get high-quality ADC. I note that most of the other players based on a similar PortalPlayer/Wolfson platform (eg Samsung, Philips, iRiver) offer high-fidelity recording.

So you see you are wrong. The iPod's lack of high-fidelity sound recording is not the fault of "third party dongles", it is not a limitation of iPod hardware, it is simply that Apple has chosen to intentionally limit the available quality of the recording function. As to why Apple would choose to cripple the iPod this way, many people probably have different opinions on that. personally, I feel that it's Apple's way of making nice with the RIAA.

iPod 3G has a 220uF capacitor at the headphone output stage

The later Wolfson WM8750 (PortalPlayer iPods use WM8731 and WM8721) offers "No DC blocking capacitors required (capless mode)"

SigmaTel's D-Majors have a <0.05% THD headphone driver, including anti-pop and short-circuit protection

Why would it need short-circuit protection? Hmmm....

iPod 3G has a 220uF capacitor at the headphone output stage

The later Wolfson WM8750 (PortalPlayer iPods use WM8731 and WM8721) offers "No DC blocking capacitors required (capless mode)"

SigmaTel's D-Majors have a <0.05% THD headphone driver, including anti-pop and short-circuit protection

Why would it need short-circuit protection? Hmmm....

Rebuffering...
and
Cannot find stream
and
Would you like to upgrade to Real 9?

This is the Audio Codec used by the iPod.

It has an internal clock generator for the DACs.

The data isn't "clocked" anywhere until the DAC is literally pulling the data out of the filter output buffer. Everything else is asyncronous.