I recall this as well, although I expect representatives of ADI weren't addressed those of us they might consider as "fanatical radicals", running them open loop into a resistor connected from the Tz node to ground.
When reducing the Tz node resistance from 2M Ohm to 2K Ohm the bandwidth can increase by a factor of 1000. Alternatively, if the Tz node capacitance is increased 1000x from 5pF parasitic to 5,000 pF the bandwidth at the Tz node remains the same. Hence this permits simplistic filtering by selecting capacitance values to support either increased or decreased bandwidth. The performance doesn't necessarily suffer by adding even large capacitance values at the Tz node terminal, rather can dramatically improve performance by passively filtering out high frequency disturbances.
From an experimental perspective it becomes difficult even for beginners in the DIY community to accidentally turn this form of application into an oscillator.
There was a long exchange between Barrie Gilbert and Charles Hansen that was quite amusing, that I have lost. Mr. H tried the open-loop just has its own sound line on Barrie and he didn't quite know what to say or if to take it seriously.
Is, or was, Barrie Gilbert ever an audiophile? My background is one of constructing all manner of specialized electronic equipment for research purposes, with a primary hobby interest in excess of 20 years (actually 40 but i don't want to confess my age) in designing and modifying audio equipment.
The difficulty was never in creating variant circuit designs to achieve some object (since that is what I did for a living) rather most entirely in understanding the psycho-acoustic relationship to devices designed to address parameters believed relevant in the reproduction of audio. Barrie Gilbert can rightly be considered an expert in his field, and worthy of great respect, yet incompetent as such relates to identifying the relationship of innumerable known and perhaps unknown design parameters related to psycho-acoustics in audio reproduction. Any arrogance on his part is understandable yet would be considered misplaced without some knowledge of psycho-acoustics.
Barrie had a vast knowledge and love of music, he was a composer of original works and had an entire orchestra of synthesized instruments in his home. He told me he always wished he had saved a pair of Beverage electrostats. OTOH he confessed that he could not tell the difference between an early class D demo board and his Classé power amp.
So what is an audiophile? BTW Mr. Hansen's (RIP) ramblings were nonsense, calling criticism of them arrogance is fatuous.
OK... I had to look it up. An audiophile is "a person who is especially interested in high-fidelity sound reproduction".
Barrie seems an interesting and most respectable person... thanks for the heads-up.
Even if Barrie could hear the difference how would that help him to identify those design parameters most important to improve upon perhaps the Classé power amp, or to improve upon the AD844 for that matter? There seems a vast number in the DIY community that have a greater understanding of electronic design parameters as such relates to psycho-acoustic phenomenon. Unlike Barry we attempt to address what we perceive (as delusional or otherwise), to evolve mechanisms for higher fidelity.
It seems that we (at least those as still active members of the DIY) are audiophiles and Barrie isn't. Barrie is clearly interested in sound and its reproduction, yet seems mostly careless of fidelity, since he doesn't hear differences. For him there is no difference between what we consider high and low fidelity to cause him to engage any interest on his part. Hence there is more interest for him not to care. Someone without the ability to hear difference also has no feedback mechanism to determine if fidelity is being improved, or otherwise, in the making of changes.
For Barrie, and perhaps many others, low-fidelity doesn't exist of advancing significance to cause them to be, or remain, "particularly interested in high-fidelity". Once an audiophile isn't "always" an audiophile
I don't know if this is vanity or fantasy. Best let it lie, this has no where good to go.
Dialog respecting Barrie is agreed as a diversion from the substance of this thread. But then again so is voting with a “+1”.
Sorry abraxalito... meant to respond to you. "Best let it lie, this has no where good to go."
+1
A good argument is one whose conclusions follow from its premises; its conclusions are consequences of its premises. My replication as a "+ 1" contains no variant presentation to the original as to permit your presentation being strengthened by its replication. The "+1" in this context is simply to show my consideration and understanding of your submission. Thanks.
On a different topic I have a question for you. Although it seems clear that you and many others perceive parallel'ed DAC's to sound better, to what do you attribute the failure of singular DAC's and/or AD844's?
Curious - where did you gain the impression that I perceive paralleled DACs to sound better? I only have noticed that in one particular circumstance - that being where those paralleled DACs feed a step-up transformer. In all other cases I haven't noticed a difference or there's been a very slight deterioration in the sound.
Where is the 'failure of singular DACs/AD844s' evident?
My use of the word 'failure' is in relation to a reference point whereupon multiple DACs/AD844's are considered 'successful' as 'a priori' true by those who believe this is true. In that sense you have answered part of my question in that you have found this true under circumstances employing a transformer.
With some background... I looked up your profile and found the thread you started on the lingDAC - cost effective RBCD multi-bit DAC design, with post #460 showing pictures of the paralleling. In a limited review of the entirety of this thread your contributions appear analytically astute and prolific to the extent that my own limited intellect and experience can comprehend. My question was motivated upon a perception of the level of your expertise that at minimum recognizes the advantages of ferrite's, particularly on the input of the I/V.
In a bit of background I purchased the Chinese 4xTDA1543 DAC box (Muse Dac I believe) and wasn't impressed. Although I like the idea of singular power supply DAC's like the TDA1543 and others, the need to reference the output to ground creates a number of design challenges that in conjunction with other factors as poorly addressed in the Chinese device. Hence this does not mean that the concept of paralleling devices was condemned, rather that the implementation was not considered adequate to make that determination.
In reviewing some older threads I recently came across the thread 'TDA1387 continuous calibration dac' that included your posts #4 and #5, as seemingly the background for the work you are engaged currently. For the most part this answers my questions, yet if you have anything to add of current thoughts as addressing the mechanisms as to why this works so well (as evidenced by the complexity of its continued implementation several years later) I would be appreciative to hear them. Thanks
I believe ESS Sabre dac are voltage output, and therefore don't need I/V conversion.
Cheers George