The other ESS DACS - ES9006 (Premier) and ES9016 (Ultra)

[regal]

interesting journey mate!, end game ESS rigs eh? products like the i2s fifo buffer/reclocker (no resampling) by Ian of this board have changed the landscape of diy clocking forever IMO, not new ideas, but new access to the diyer. I do wonder why you are still futsing about with spdif? seems you are going to at least as much trouble to end up NOT going the way that is so much trouble?

the fifo is asyc wrt the input i2s or spdif, but can be setup syncronous to the dac mclk in that they are one and the same, no division, the clock used to drive the flipflops for the other lines is also what drives the DSP and dac. the output clock bares no relation, or connection to the input clock, so the only remaining jitter is the jitter on that clock and its buffer, for a total of less than 1ps. DPLL bandwidth is set to lowest and the sound is magic!!

so this allows easy comparison of these modes, switched by the MCU, OSF bypass or not, sharp/slow rolloff, jitter rejection or not and the use of 64bit apodising OSF in software can be toggled in puremusic while switching the OSF on/off in the dac in a terminal window from an ipad in the listening position.

applying all this to the digital crossover is the next step, but I have to move away from the fifo for an intermediate sidestep, as its only 2 channel for now, till Ian gets around to coding the 8 channel later in the year. my usb->i2s has a fifo and can output 2-8 channels, so i'll toy with clocks and software for a bit to optimize the speaker setup.

all good fun, I dont think about the money or time, i'm putting together a system not available any other way and if time was taken into account it would far outweigh the cash outlay, so I dont see the cost as being all that relevant from my perspective. as it turns out the amps that have become available now also present 140+db and less than -110+N dB THD driving real loads

the akm4399 is interesting, as was the failed R2R release by ARDA and I just bought a couple parts of the AKM, but theres a long road traveled towards ESS that I would have to catch up with and tbh i'm pretty wrapped with the way things are falling into place

This is a great solution to the oversampling algorithm built in the ess, I sure hope PC foobar users can get a real time 64bit apodozing filter VST like is available for the mac.

 

[twest820]

If we assume the measurement jitter floor is negligible, the MCLK jitter distribution is Gaussian, and apply 66-95-99.7 it's probably reasonable to consider 600ps as an upper bound for the Cirrus codecs and 150ps for the Wolfson receivers (RMS jitter is one standard deviation).

Scratch the Wolfson part of that as baseband jitter is attenuated in measurements of SPDIF period jitter. Makes constructing a meaningful estimate of the baseband jitter rather difficult.

 

[twest820]

(moving DAC antialiasing filter programming discussion from the rePhase thread)

Twest, how are you getting the functions into the DAC? I think I have a few ES9016 implementations kicking around here and would like to try it.

I don't own an ESS DAC, so I've nothing to put the filters into. (I've other DACs I'm happy with so the primary reason for me to get a 9016 or 9018 is the control over the antialiasing filters. But I haven't got round to setting up minimum or constrained phase synthesis, which in turn means I don't need hardware as I've nothing to program into it.) Register and programming information is in the datasheets and ESS has an app note with additional details. I got my copy of the note from ESS directly but Shaw should be able to send it to you as well.

I'm not aware of any DIY targeted tooliing which allows programming the filters, though I assume ESS's eval board software supports it. Presumably this feature would fit naturally in the ES9018 I2C project but I've not been following that thread.