Hello everyone,
I am looking at these two chips and cannot decide which is better in terms of real-world performance and reliability. Does anyone have any experience with both CM102S and PCM2706?
Also, I was under the impression that CM102S supported asynchronous transfer mode (a good friend of mine told me that), but the datasheet only mentions Isochroous 😱 transfer: "Isochroous transfer using adaptive synchronization with internal PLL". Could somebody please enlighten me whether CM102S supports async USB or not?
Thanks!
I am looking at these two chips and cannot decide which is better in terms of real-world performance and reliability. Does anyone have any experience with both CM102S and PCM2706?
Also, I was under the impression that CM102S supported asynchronous transfer mode (a good friend of mine told me that), but the datasheet only mentions Isochroous 😱 transfer: "Isochroous transfer using adaptive synchronization with internal PLL". Could somebody please enlighten me whether CM102S supports async USB or not?
Thanks!
Shinier of two turds... although for a hobbyist wanting to build something from the ground up they are the only choices available since the more advanced chips are volume purchase only with lots of NDA.
Async or adaptive synchronization refers to the clocking mode, or where the USB controller gets its clock from. The data transfer mode of audio is almost always isochronous, which suits audio best because audio can only be played after the computer software generates it. Unless Stephen Hawkins found a way to transfer audio data from the future.
That said, some oddball interfaces may use bulk mode for audio data. This mode, usually for external storage and etc, gets guaranteed data delivery through error detection and resending. But this is not a good mode for time-critical applications e.g. audio. That said, there are oddball users who buy external dacs which add a full 1 second delay to the sound while it streams and re clocks it through a buffer. (Stressing again, you can't get audio data from the future.)
Async or adaptive synchronization refers to the clocking mode, or where the USB controller gets its clock from. The data transfer mode of audio is almost always isochronous, which suits audio best because audio can only be played after the computer software generates it. Unless Stephen Hawkins found a way to transfer audio data from the future.
That said, some oddball interfaces may use bulk mode for audio data. This mode, usually for external storage and etc, gets guaranteed data delivery through error detection and resending. But this is not a good mode for time-critical applications e.g. audio. That said, there are oddball users who buy external dacs which add a full 1 second delay to the sound while it streams and re clocks it through a buffer. (Stressing again, you can't get audio data from the future.)
No experience with the cm102s but a bit with the pcm2707. It's quite reliable but it's still easy to kill it if you don't respect the timing on power connections, especially in self powered mode. There are quite a few threads about this on this forum, a quick search could be useful.
Yea come to think of it the PCM270x seem to be more prone to getting hangups by power surges (e.g. somebody turning on the fluorescent lamp). Meanwhile all the cheap USB sticks that are as big as a thumbdrive and powered by CM102 (it is a very low cost part) have no issue despite having nothing on the PCB.
Oh I forgot to mention, CM102 has very bad analogue performance. Nowadays I assume that people use these two chips only for USB to I2S or SPDIF, but if one chip is going to handle everything then PCM2707 is the way to go.
Oh I forgot to mention, CM102 has very bad analogue performance. Nowadays I assume that people use these two chips only for USB to I2S or SPDIF, but if one chip is going to handle everything then PCM2707 is the way to go.
- Status
- Not open for further replies.