Exactly so. I was hoping the ES9822 would be closer to the ES9038PRO architecture, unfortunately it doesn't seem to. Or at least we are not entitled to know about, according to ESS.
Interesting. ES9102 had the ASRC, it was a selling point for pro audio and recording where several ADC boxes can be synced to external clock. It is setup in registers 4 and 5.
An ASRC does not make much sense in an ADC. If not in host mode a clock cleaner may be needed but to sample rate convert as you are sampling seems kind of dumb.
I agree, if we have a good enough clock to feed ADC, which is designed to utilize low-phase noise from external oscillators, what kind of "jitter eliminator" could we need? Just turn the ADC to the master mode and enjoy.
I think it’s a misunderstanding, by ASRC he means an asynchronous mode, similar to the ES9038 DAC.
IVX, the ES982x in master mode may fit your bill, but certainly doesn’t fit mine. The more I am looking at the XMOS code, the more troubles I am having considering the integration of an ADC in master mode. Certainly possible, but hard to justify, unless there is a performance penalty in using the slave mode.
I using CT7601 Comtrue UAC2 bridge, because of cheaper, smaller, 3x times less current draws vs Xmos. Opened the source code makes me easy to fit all functions needed + free Windows driver(for ASIO, UAC2 already works with Windows 10). But.. the jitter of that bridge for 44.1 and derivatives is awful. It is no problem with ASRC like 9038 has but with 9822 of course I have to go to make the bridge slave.
Well, you are a pro designer/manufacturer and work in the hot bed of electronics. For me, not a chance in hell to get anything beyond the XMOS (a horrible architecture power hog, with the known licensing issues and no source code for the critical parts) or the now outdated/obsolete CMEDIA 6635, everything else is either not available, or no source code.
I haven’t even heard of Comtrue products, other than from yourself. Not to mention I don’t have any resources beyond myself, manually soldering SMD boards and hacking code in the spare time from my day job. I guess I’m too small for such a big war.
I haven’t even heard of Comtrue products, other than from yourself. Not to mention I don’t have any resources beyond myself, manually soldering SMD boards and hacking code in the spare time from my day job. I guess I’m too small for such a big war.
Last edited:
IMO it's the same - asynchronous sample rate conversion - resampling between input clock and the output clock, with the conversion ratio adaptively adjusting to the momentary ratio.
In DACs it allows to have a single master clock for the DAC of any frequency (e.g. the 100MHz clock in ESS) and feed it with any independent I2S signal of any samplerate. This feature simplifies the DAC integration a lot as no dividers of bitclock/frame clock from the masterclock are required. On the other hand, the ESS DACs, unlike most others, have fixed 128xFs masterclock for any samplerate when in the synchronous mode, simplifying the clock layer (frameclock = MC/128, bitclock for 32bit 2ch I2S = MC/2).
Last edited:
Comtrue would likely work with anyone interested in their products. I worked with them to test their ASRC and DSD converter chip. The USB Bridge docs are at: ComTrue Inc. - CT7601 ...Don't know about small volume retail sales though. One would have to inquire with them.
Last edited:
I watched that thread and concluded there's nothing beyond what I already mentioned. Any idea where to buy the ComTrue bridge in small quantities, download a decent data sheet, SDK, source code, application examples, find a support forum, etc...?
Unfortunately my life is too short for this kind of detective work, I've already had enough with ESS in the last 6 months. Also, I have a limited budget to afford much of a trial and error, with no commercial activity in sight. If I would have the time, I'd rather start from scratch with an UAC3 bridge project on a STM32 F series or, why not, on a TI OMAP DSP chip, with an USB PHY chip on the side.
Back in the day I implemented a UAC2 device on a LPC4330 microcontroller. It was a bit of a nightmare, but in the end it worked very well on Linux, there were no drivers on Windows then. However the SGPIO peripheral on that device which was marketed as capable of outputting multichannel I2S experienced rare and random glitches where it would shift by one bit mid-stream, with the results one would expect.
XMOS' huge power consumption was a big turn-off because all that power is eventually turned into noise. Also at that time the dev environment was unusable.
XMOS' huge power consumption was a big turn-off because all that power is eventually turned into noise. Also at that time the dev environment was unusable.
about 1 year CT7601 is available on Taobao for 1-3pcs. You can check Aliexpress as well. Actually, I know the Taobao shop owner and may recommend them to sell CT7601 by Aliexpress. As I remember, I told about that Comtrue guys but they are weird and didn't reply. BTW, sometimes, I feel the Comtrue guys are not really the same who designed the chip.
CT7601SR??SA9023A SA9023 TE7022L 24BIT/96KHZ??-???
1PCS~5PCS/LOT CT7601SR QFN48 New original|Air Conditioner Parts| - AliExpress
LOL, $41 )) Actually, I found last week 24c64 was offered for 46x price as well -$$ilicon!
CT7601SR??SA9023A SA9023 TE7022L 24BIT/96KHZ??-???
1PCS~5PCS/LOT CT7601SR QFN48 New original|Air Conditioner Parts| - AliExpress
LOL, $41 )) Actually, I found last week 24c64 was offered for 46x price as well -$$ilicon!
Last edited:
Thanks, but assuming I'd buy a few CT chips disregarding price, then what? I see a windows driver binary and a few schematics are available for download, and that's about it. Not even 20% of what would be needed to start any serious development.
Last edited:
I am not trying to persuade you to use this IC, I just explained why I used that. Markw4 asked how to get the IC, I answered him.