Thank you mates,
If John and his brother want to gift us we must contribute at our simple level to make it time light as we saw no one ihere s really able to draw a sota TDA1541A board. Pedja's Rogic from Audial is a good illustration imho showing devil is in the details and only guys as him, John from ECDESIGNS, Thorsthen Losch from AMR can do it and share in tne same time with great returns. So a KISS for the New-Year could be cool imho, no discussion on the design to let egotic claims outside:
- Standalone board, just add the PS traffos and the stream via usb
- No gerber, no pcb draw, it stays EC DESIGNS New Year Kiss property, so produced by EC designs only. It means no draw and errors made by egotic members or partial designs with multi boards (So no discussions)
- No interaction with John but for congrats, just populate with the definitive given BOM (change it at your own risk if you want but keep it for you)
- A building thread written by members to help
- cms friendly and eventually add ufl I2S inputt option
- KISS means a single tda1541a chip and single end outputt.
- 1 dollar given to Diyaudio which supports this great thread from years
I add an idea : if the cost of the board stays friendly, combos it with EC Designs'S USB transport UTOS device (sorta win-win to advertise off shelves products outside the diya forum (so inputt of the dac will be toslink John's way for this board)
Max would you like to help and to be the op of the building thread or manage the GB of the board ? We could help by writting the BOM with suppliers reference as we read the thread and know what John is liking : Acrylic caps, Rhopoint in the analog signal path, sussumu or else for the digital side, etc ? Idea is to stay kiss to save John's time and stays on the sharing attitude.
What do you think John , it was your idea some times ago to make us a board when you gave up tda1541a
for discrete r2r todays line? Please
...as we must save the soul of our great Poet here known as Max Lorenz Neruda from Chile to allow him to listen to the TDA1541a .
If John and his brother want to gift us we must contribute at our simple level to make it time light as we saw no one ihere s really able to draw a sota TDA1541A board. Pedja's Rogic from Audial is a good illustration imho showing devil is in the details and only guys as him, John from ECDESIGNS, Thorsthen Losch from AMR can do it and share in tne same time with great returns. So a KISS for the New-Year could be cool imho, no discussion on the design to let egotic claims outside:
- Standalone board, just add the PS traffos and the stream via usb
- No gerber, no pcb draw, it stays EC DESIGNS New Year Kiss property, so produced by EC designs only. It means no draw and errors made by egotic members or partial designs with multi boards (So no discussions)
- No interaction with John but for congrats, just populate with the definitive given BOM (change it at your own risk if you want but keep it for you)
- A building thread written by members to help
- cms friendly and eventually add ufl I2S inputt option
- KISS means a single tda1541a chip and single end outputt.
- 1 dollar given to Diyaudio which supports this great thread from years
I add an idea : if the cost of the board stays friendly, combos it with EC Designs'S USB transport UTOS device (sorta win-win to advertise off shelves products outside the diya forum (so inputt of the dac will be toslink John's way for this board)
Max would you like to help and to be the op of the building thread or manage the GB of the board ? We could help by writting the BOM with suppliers reference as we read the thread and know what John is liking : Acrylic caps, Rhopoint in the analog signal path, sussumu or else for the digital side, etc ? Idea is to stay kiss to save John's time and stays on the sharing attitude.
What do you think John , it was your idea some times ago to make us a board when you gave up tda1541a
for discrete r2r todays line? Please
...as we must save the soul of our great Poet here known as Max Lorenz Neruda from Chile to allow him to listen to the TDA1541a .
Last edited:
Max would you like to help and to be the op of the building thread or manage the GB of the board ? We could help by writting the BOM with suppliers reference as we read the thread and know what John is liking : Acrylic caps, Rhopoint in the analog signal path, sussumu or else for the digital side, etc ? Idea is to stay kiss to save John's time and stays on the sharing attitude.
What do you think John , it was your idea some times ago to make us a board when you gave up tda1541a
for discrete r2r todays line? Please
Oh, dear!
I feel flattered beyond measure but, dear Diyiggy, seriously, I'm your worst pick.
I have no practical experience with the DAC and I am too bad at organizing: I am a consigliere. or "consisigliere"Not to mention that I am lazy as f***.
Best wishes.
M.
PS: of course I could help here and there...
PS2: I think I will give a try to the LED idea...
Please
Seems the noble cause ;-)
Were you thinking 1541A core PCB with only I2S/PCM via u.fl inputs, or USB> FPGA> ISO> RE-CLOCK> I2S to PCM> 1541A all on the same PCB?. First one surely easier and with less interruption/BS as eluded to, likely good enough.. despite modular, or no? ..
Either way, I admire the respect you show when asking, also the humanistic tone in your dialogue.
Kind regards,
HK
Last edited:
Thank you and so sorry to see what happens in Oz, hope Adelaid's area is not too much firered.
I don't know what John had in mind when he told us he will make a tda 1541 pcb for Diya people.
My point is, it should stay simple for him while having all John's mojo on the digital side and analog I/V on a standalone pcb as Rogic did. If he want to do it (halas his silence is deafening... as the pcbs tries here made before by others btw to stay sincere) I would say: free to him, he knows the tda1541 animal pretty well and few tame it as he did.
If I2S stays simple and his a good option for us it also means less John's mojo on the digital input. In my items wisch list: no fgpa (or buy a Chord or use others Fifo stuffs), "just" a cool usb or toslink input if people buy his of shelves usb to toslink transporter... so indeed sort of buffer/post passive reclocking as he shows in this thread by a local masterclock on the pcb as the good tda 1387 audiophonic dac hat for Rpi but here with usb or toslink instead Rpi I2S (btw I fed up with all this hip around Rpi factory multilayer skycreepers dacs). So no IanCanada or anyone board in my mind on the digital side to stay simple and manageable as a standalone pcb (just add raw PS traffos and pcm stream), good digital inputt management philosophy made by EC Designs. I think it could be a good thing, after all the diy Aya pcb of Audial gives it a lot of non diy customers for the off shelves final and more complex products whom listened the simplier diy sound signature of the brand at homes' friends.
Edit: maybe an intention member list to show if there is a need ?
I don't know what John had in mind when he told us he will make a tda 1541 pcb for Diya people.
My point is, it should stay simple for him while having all John's mojo on the digital side and analog I/V on a standalone pcb as Rogic did. If he want to do it (halas his silence is deafening... as the pcbs tries here made before by others btw to stay sincere) I would say: free to him, he knows the tda1541 animal pretty well and few tame it as he did
. If I2S stays simple and his a good option for us it also means less John's mojo on the digital input. In my items wisch list: no fgpa (or buy a Chord or use others Fifo stuffs), "just" a cool usb or toslink input if people buy his of shelves usb to toslink transporter... so indeed sort of buffer/post passive reclocking as he shows in this thread by a local masterclock on the pcb as the good tda 1387 audiophonic dac hat for Rpi but here with usb or toslink instead Rpi I2S (btw I fed up with all this hip around Rpi factory multilayer skycreepers dacs). So no IanCanada or anyone board in my mind on the digital side to stay simple and manageable as a standalone pcb (just add raw PS traffos and pcm stream), good digital inputt management philosophy made by EC Designs. I think it could be a good thing, after all the diy Aya pcb of Audial gives it a lot of non diy customers for the off shelves final and more complex products whom listened the simplier diy sound signature of the brand at homes' friends.
Edit: maybe an intention member list to show if there is a need ?
Last edited:
Cheers. ADL itself is okay, some rural properties affected >50km to CBD, impact is comparatively very small. Reported animal loss is 480 million, which includes 50% and 30% of Koala on Kangaroo Island and throughout NSW, respectively.
RE: John DAC PCB, hopefully it gets some traction because its a great idea!.
RE: John DAC PCB, hopefully it gets some traction because its a great idea!.
Attachments
Last edited:
Interesting thread.
Mathematically I see this as time super resolution. It is over sampling in that there is a higher time resolution grid applied. So mathematically it is over sampling. The difference is that rather than have a single sample faster stream, the conversion occurs in parallel at normal speed (just time shifted).
So the interpolation is linear with 8 points caused by the sub-division of time on the output. As such you should not have overshoot but your waveform may show a slower rise/fall time as the interpolation ignores the previous samples.
This is missing a trick.
What about the bit depth resolution - this could be made into higher resolution by also shifting the input voltage for each bit depth sub-diction and then provided to a dac - doubling the number of favs again!
I would also suggest an fpga - not for the butte path (although it could dejitter the i2s stream) but instead working at 300+MHz it could provide a better inter-sample synchronisation of the DAC clocks without using the inverse delay of each chip? Perhaps an added complication.
Mathematically I see this as time super resolution. It is over sampling in that there is a higher time resolution grid applied. So mathematically it is over sampling. The difference is that rather than have a single sample faster stream, the conversion occurs in parallel at normal speed (just time shifted).
So the interpolation is linear with 8 points caused by the sub-division of time on the output. As such you should not have overshoot but your waveform may show a slower rise/fall time as the interpolation ignores the previous samples.
This is missing a trick.
What about the bit depth resolution - this could be made into higher resolution by also shifting the input voltage for each bit depth sub-diction and then provided to a dac - doubling the number of favs again!
I would also suggest an fpga - not for the butte path (although it could dejitter the i2s stream) but instead working at 300+MHz it could provide a better inter-sample synchronisation of the DAC clocks without using the inverse delay of each chip? Perhaps an added complication.
In the softest way one can ask these things, are you not Rudolf Broertjies?.can't even layout a basic TDA1541A dac?
Last edited:
I see your point - super resolution is more ADC, inverse super resolution is more DAC.
Super resolution ('Drizzle') was made for Hubble - by taking multiple images with sub-pixel movements across the field of view it increases the pixel resolution once the final images are realigned against the space object. In this it's using the x,y being the spacial axis. This realignment can be done on an upscaled image by using 2D FFTs (assuming a linear 2D point spread function).
The key here is there's limited sampling against an infinitely detailed target.
Here's the original paper: Drizzle: A Method for the Linear Reconstruction of Undersampled Images on JSTOR
For my image processing I simply;
* upscale the time to 10x (this can be with or without interpolation)
* rotate the images by 45 degrees (this smooths the corners)
* then use 2D FFT phase correlation to align the images at 10x the resolution
* stack the images, essentially averaging the pixels.
The images I use are with a camera running at -20degC cooling the CCD - the sampling ADC is measured in electrons (the CDD pixel has a finite electron well depth). The resulting images are very noisy so image alignment and stacking is a standard thing to increase SNR.
As a DAC, it's doing the inverse - you think if the stream of bytes - x axis being sample time and y axis being bit depth which is finite sampled then the time alignment is then generating a infinitely detailed output.
Thus we can use the same technique in bit depth - shifting the voltage baseline of the intermediate DACs for higher bit depth sources and generate a higher detailed output.
I hope that explains my reasoning.
As apostles go you guys are pretty disappointing. I cannot believe after all this time you have learnt so little. How can you spread the word if can't even layout a basic TDA1541A dac?
As you know, deep in you Sacred Heart, dear rfbrw, I Love you unconditionally
and I AM sending you a Pink Ball of my Pure Divine Unconditional Love (PDUL), which you may feel (or not; it depends...) as a warm feeling in your chest. 
This, in order not to confound the business of the left hemisphere with the dealings of the Superior Mind-Sacred Heart complex...
Best wishes of Abundance of All Kind of Blessings,
M.
I imagine NickKUK is talking about the Direct Interpolation DAC from the beginning of the adventure and which was abandoned more than a lustrum ago. Indeed, it was interesting: as I remember, with the smaller and humbler 1543 chip it produced an ample soundstage, a very dynamic "party-like" presentation, but was, alas, wanting on the already compromised HF department for this DIY friendly chip...
(I still have the PCBs)
The experiments that followed sound better IMHO, including the "scrambler-interpolator DAC" and my beloved "Dual-Mono DAC".
John can give the technical details...
Cheers,
M.
I imagine NickKUK is talking about the Direct Interpolation DAC from the beginning of the adventure and which was abandoned more than a lustrum ago. Indeed, it was interesting: as I remember, with the smaller and humbler 1543 chip it produced an ample soundstage, a very dynamic "party-like" presentation, but was, alas, wanting on the already compromised HF department for this DIY friendly chip...
(I still have the PCBs)
The experiments that followed sound better IMHO, including the "scrambler-interpolator DAC" and my beloved "Dual-Mono DAC".
John can give the technical details...
Cheers,
M.
Lol - yep, over 700 pages makes it difficult to piece together what the state of the title is
I was wondering over the last couple of days how todo DAC without silicon in the end you end up making the same journey that DAC IC manufacturer's have made. However the romantic idea of a vacuum tube ladder dac would be stuff of legend!
As you know, deep in you Sacred Heart, dear rfbrw, I Love you unconditionally
This, in order not to confound the business of the left hemisphere with the dealings of the Superior Mind-Sacred Heart complex...
Best wishes of Abundance of All Kind of Blessings,
M.
