Talk:Intel High Definition Audio

Codename
Why is IHD called AZALIA ? Any special reason ??
 * Just a codename. For example, Intel uses geographical names for their CPUs and chipsets (look here). Why not use flower names for audio codecs? 217.172.21.161 (talk) 06:51, 9 January 2008 (UTC)

Throughput
How could it be that 192 ks/s times 32 bit times 2 channels (12.288 Mbit/s), when downsampled to 96 ks/s, provides 8 channels (24.576 Mbit/s)? Logically, «low-end» mode should carry 16 bits only, i. e. “192/32/2 vs 96/16/8”. 217.172.21.161 (talk) 06:51, 9 January 2008 (UTC)

the HDA in nforce 8200 with vista x64 totally sucks, it just doesnt work. It works in linux

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Intel High Definition Audio. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added tag to http://www.opensolaris.org/os/project/opensound
 * Added archive https://web.archive.org/web/20130819015410/http://www.conexant.com/Product/Audio/pchdaudio/Pages/default.aspx to http://www.conexant.com/Product/Audio/pchdaudio/Pages/default.aspx

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 15:17, 14 November 2017 (UTC)

Incorrect note about AC'97 dongle triggering headphone or microphone detection.
The page states "The different signal assignments can cause trouble when AC'97 front-panel dongles are used with HDA motherboards and vice versa. An AC'97 dongle returns audio on pins 6 and 10 rather than digital plug sensing signals. Consequently, a loud audio passage may cause a HDA motherboard with a AC'97 dongle believe headphones and microphones are being plugged and unplugged hundreds of times per second."

This isn't how the HDA header is designed as per the Intel High Definition Audio spec, in fact it seems specifically designed to avoid this issue. This passage lacks citation, it seems like it was made up based on an incorrect assumption.

Per the spec, the SENSE1_RETURN and SENSE2_RETURN lines are simply bridged to ground in the motherboard, via a reference resistor. There is no logic from the CODEC IC to either of these pins. Instead, SENSE_SEND measures the resistance between itself and ground, and then determines what jacks are connected based on the measured resistance. SENSE1_RETURN and SENSE2_RETURN simply act as convenient references for the most common HD audio dongles which bridge SENSE_SEND to SENSE1_RETURN or SENSE2_RETURN, but SENSE1_RETURN and SENSE2_RETURN are not strictly required. A dongle can bridge SENSE_SEND to ground via the same resistor values internally. The HD Audio Interface seems to have been designed specifically with this flexibility in mind.

This means that it's impossible for an AC'97 dongle to cause a HDA motherboard to detect a headphone or microphone connection when a "a loud audio passage" is playing looped back into the SENSE1_RETURN and SENSE2_RETURN pins, since they don't actually do any active detection by themselves, they are just high impedance routes to ground. Crozone (talk) 11:57, 4 September 2023 (UTC)