@ThorstenL : Thanks for your feedback in #2964! ... Sort of what I had figured out in relation to the quite high distortion levels of R2R DACs, however, I did not know that Holo Audio uses a corrective DAC to achieve their low distortion levels. Interesting approach ...
And then, if I may, one question in relation to your #2910 post: You show a "1+5" inverter topology for a sine wave inverter/squarer. I have seen this before (e.g. Andrea Mori uses it) and have been wondering if using the same IC, i.e. just one package, would not cause significant inter-gate disturbances as well as a (much) higher ground bounce due to now 5+1 gates drawing current through the GND pin? Also, won't the use of 5 parallel output gates cause some glitches on the output that may disturb the clock?
Anyway, just have been puzzled about this approach ... Wouldn't just one NC7WZ04 - or maybe two NC7SZ04s - make for a less noisy solution while still keeping phase noise levels very low?
BTW your 74ALVC164245 does not appear to exist ...
Cheers, Jesper
And then, if I may, one question in relation to your #2910 post: You show a "1+5" inverter topology for a sine wave inverter/squarer. I have seen this before (e.g. Andrea Mori uses it) and have been wondering if using the same IC, i.e. just one package, would not cause significant inter-gate disturbances as well as a (much) higher ground bounce due to now 5+1 gates drawing current through the GND pin? Also, won't the use of 5 parallel output gates cause some glitches on the output that may disturb the clock?
Anyway, just have been puzzled about this approach ... Wouldn't just one NC7WZ04 - or maybe two NC7SZ04s - make for a less noisy solution while still keeping phase noise levels very low?
BTW your 74ALVC164245 does not appear to exist ...
Cheers, Jesper
It takes a sophisticated calibration system.
Essentially for any given binary input there needs to be specific compensation. Even with that, I'm hardly blown over by results:
Review and Measurements of Holo Audio Spring DAC
Compare that to TDA1541 - which surely is a total dinosaur:
Measurements of Hifidiy forum Anniversary TDA-1541 My DIY Replica
Hifi Forum TDA-1541A DAC Review
No, this is DRIVER. It means you take your clock (or for arguments sake AES-EBU signal) and pass it through this, to get a low impedance drive with more precisely defined impedance.
The "squarer / receiver" uses just 2 Inverters and is best done as pop-corn logic.
This system would (using 3.3V rails) have a clock with a very tightly controlled 50 Ohm impedance and 1.65V PP driver / receiver system with very low added jitter.
Specifically, it was meant for passing squared or clipped sine clocks, without resorting to LVDS, which I would argue is a better choice.
Indeed. Hence my suggestion of using multiple NC7 popcorn logic items instead.
However, these circuits are common and well worn and appear to have less phase-noise jitter than what is added by even very expensive crystals. So perhaps loosing too much speel here is a waste of time?
Probably not. We have separate decoupling and capacitive feedback into the clock outputis unlikely, as it would look commonly something like this:
The second inverter squarer usually gets closer to 50% duty cycle and isolates clock and rest of circuitry.
The problem is the very unequal H/L output impedance. If we want a solution that can drive "universally" long cables that are 50 Ohm back terminated, a single inverter as driver is just not that good?
BTW your 74ALVC164245 does not appear to exist ...
Really? Why? Mouser lists 1,000 in stock from 3 Vendors and I even found domestic chinese versions from multiple vendors.
https://th.mouser.com/c/?q=74ALVC164245
Thor
Last edited:
.. Following your link I can see that there are lots of these ICs - sorry, I missed that 
Jesper,
While you are at it, it's also worth looking up 74AHCT/LVC/AUC16374.
Peruse the datasheet, maybe you will "get" why I keep tubthumping and welkin ringing about these 16-Bit widebus parts.
"A squid eating dough in a Polyethylene bag
Is fast and bulbous, got me?" Captain Beefheart
Thor
While you are at it, it's also worth looking up 74AHCT/LVC/AUC16374.
Peruse the datasheet, maybe you will "get" why I keep tubthumping and welkin ringing about these 16-Bit widebus parts.
"A squid eating dough in a Polyethylene bag
Is fast and bulbous, got me?" Captain Beefheart
Thor
Ok something very odd with the new PCM2DSD FW , I have two boards here one with original FW and one with new. I plugged the new one in and thought this sounds very odd , scratching my head I've tried a few tracks I know well . For some reason the Right/Left is reversed , I plug the other in and it is normal . Tried it again reversed , is there anything I've done that can be causing this ? I just tried the L&R test with windows in sound settings to confirm and also plugged into my MacBook . I'm using a Douk U2 pro to a HDMI to i2s that plugs into the pcm2dsd to Marcels FIRDAC
There are two standards for DSD channel assignments on I2S bus. The new firmware has versions to support either standard. The old firmware only supported one. I just switch the output cables for now if the wrong standard is selected.
Also, there are various ways to know which channel should be L or R. For example, if there is a pan pot in your player app you can pan to L and see if the sound goes to the left speaker (say, maybe with Audacity).
Also, there are various ways to know which channel should be L or R. For example, if there is a pan pot in your player app you can pan to L and see if the sound goes to the left speaker (say, maybe with Audacity).
Ok thanks, I'm quite late to the party so not had the PCM2DSD and FIRDAC long , a friend kindly sent me the boards to wire up and listen to . He sent me another PCM2DSD with latest FW to try . I've been using the same dac for over 12 years and wanted something newer so I'm grateful having chance to try this out
I listen to a variation of music , today was Keb Mo , Natalie Merchant , Bob Moses , Leonard Cohen , Chantel Chamberland , Ghost Rider . Latest version sounds little less strident seems a little more reserved mainly with female and some instruments . It's not huge but I can definitely hear a difference especially with my HD800, its quick to swap them around . I need to spend more time with it
I've got two active filter boards , original with last stage bypassed and output opamps changed to OPA1612 . The other board is based on Bohroks 1632 , I'm mainly using SE output . Everything is supplied by separate shunt regs
I've got two active filter boards , original with last stage bypassed and output opamps changed to OPA1612 . The other board is based on Bohroks 1632 , I'm mainly using SE output . Everything is supplied by separate shunt regs
Do you have anything with natural percussion, say simple music with live drums, cymbals, shakers, conga, etc? Music where you can hear the resonance of the body/hollowness of a drum, where you can hear if cymbals sound like bursts of noise, or else if you can hear the ping of a wooden stick on each cymbal, and each cymbal sounds different? Also, what about the dynamics of vocalists, can you hear to what extent their vibrato is frequency only, or if includes volume level modulations (same for wind instruments too, hear the volume level puff of a note attack or soft building volume level, or else maybe volume level undulations of a vibrato, or is it only pitch variation)? Can you hear the reverberation/echo of the recording room, and from that guess/estimate its width and depth? If not those sorts of things, I can put something in dropbox to try...
Last edited:
I mainly use Amazon prime unlimited so there maybe something on there ? I have the free version of HQplayer but unfortunately it doesn’t work with Amazon , it’s no problem to download a file though and use HQplayer .
It’s a little late over here now (UK) so will have another listen again tomorrow after work .
It’s a little late over here now (UK) so will have another listen again tomorrow after work .
If you are listening streaming that's probably already too damaged to tell much because of the usual lossy encoding. I will put three files in dropbox and PM you a link if that's okay. Whenever you can get around to seeing if there is anything you would like to say about the sound of the files would be fine. No rush, no pressure. Just think its probably better if you start with real wav files. Worth a try?
Sure no problem, I’ve got both Mac and windows so can use either . Amazon HD is supposed to be lossless . I use it mainly for convenience but if there’s any albums I listen to I like I’ll just buy them . Probably should compare it against downloaded albums using alternative player I can’t imagine the streaming being quite as good