ackoDAC based on ES9018

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
PM sent, interested in the bare 9018 or perhaps a tweaker 9018 in 4 oz. this looks like the perfect build to play with LCDaudino for control/sensing. also, do you have any of the sabre chips?? are they on the board already, or do we have to source them ourselves?? I dont mind either way, but would be nice if it came with one. any idea on the pricing for a real dual mono version?? I see it mentioned on the layout and this interests me a great deal. perhaps would be cleaner to buy 2 x T version 9018 boards, but muxing would become a little tricky then.

and i'm with curly woods here, gotta give the guy something for the work, especially if you would like any support or development to continue. these are small runs of high quality PCBs here, not cheap to produce in such low numbers.
 
that PCB is OK... but I can't justify the 65Eur+shipping on the IC. Especially, because people use it for stereo, its just like the good old tda1541s2 "bidness" (and I tought that must be a fad) ... (...)
I recommended and co-design an ad1955 D/A right now by-the-way .

Who knows, some of the ESS chips might be EOL sooner than we think, once the marketing department pulls a string :headshot:
 
Last edited:
that PCB is OK... but I can't justify the 65Eur+shipping on the IC. Especially, because people use it for stereo, its just like the good old tda1541s2 "bidness" (and I tought that must be a fad) ... (...)
I recommended and co-design an ad1955 D/A right now by-the-way .

Who knows, some of the ESS chips might be EOL sooner than we think, once the marketing department pulls a string :headshot:

WOW that is pricey! Damn the taxes!!!!
 
Agreed, tritosine, about the IC, it's way over priced but again based on Curly's logic we're paying for their work in designing a new & unique oversampling methodology which hadn't been done before. Wow Curly, I'm surprised you didn't know this :confused:

This hobby is expensive enough as it is without over-paying for simple jobs within the capability of a fair number that frequent this forum. Remember, that you will have to solder the 0.05 inch pitch, 64 LQFP IC to this board - a job probably requiring more skill than doing the layout of the board :)

Thanks, Curly, I will take up your invitation to do this through Expresspcb & share it with 3 of my friends - as much because shipping charges from Oz to Irl don't make sense :)
 
I think you'll find that there is enough uniqueness in their design to apply for a patent back in 2006 United States Patent: 7330138

Thats about the DPLL part for the ASRC, every ASRC has to have one. :D
The oversampling/interpolation part is still a 'textbook' zero fitting implementation, and you can't really alter its 'textbook' approach by uploading new coefficients.
It also poses its requirments afaik, linear interpolation is greedy when it comes to stopband attenuation.
 
"Zero Stuffing not First Order Hold in up-sample
A first order hold in the over sampling filter results in a
narrower data path and hence less logic, but necessitates
sin(x)/x correction be at a lower rate. No sin(x)/x correction is
perfect and low rate sin(x)/x correction results in frequency
domain imperfections. Zero-stuffing is mathematically more
precise
and results in a sin(x)/x correction at a much higher
frequency and hence any errors are much further out-of-band"

thats what Im talkin about. The rest is debatable, like "low order noise shaping" (is 5th that low ? :D ) .
 
Last edited:
I'm not going to argue the pros or cons of their design approach - you consider it to be trivial text-book fare & I'm not Dustin, the designer, so I don't know enough to enter into this with you but I have seen it reported widely that they have implemented some clever tricks in their design that hadn't been done before & this is what you're paying for in the price!
 
OK, if you must get into this:
It is a multibit sigma-delta DAC so the input is highly oversampled using polyphase filters, then the synchronous resampling is done on the highly oversampled data which allows use of a very simple interpolation algorithm.

This is very clever (and patented).

It is a good example of lateral thought. Instead of solving a hard problem (asynchronous resampling at frequencies close to the Nyquist limit) it turns it into an easy problem (asynchronous resampling with a sample frequency way above the Nyquist limit).
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.