Hi ... First I would like to say that I have not read all the last posts in detail but it appears that you might have some issues with a proper signal for the RTZ DAC, right? I have for some time now been experimenting with HQPlayer in a trial version and it appears to work very reliably in terms of generating a signal for my DSD DAC "explorations" (although the distortion level appears to vary from time to time ...). Anyway, just mentioning this because the trial version allows for 30 min use at a time (without any other termination).
To my memory I have also used HQPlayer with the Amanero card (original). Here I downloaded a firmware from the Amanero homepage that was optimized specifically for DSD use (but please don't ask me which one
- I don't remember and the card is no longer in use).
Just FYI.
Cheers, Jesper
P.S.: it seems there have been more posts while I was writing ... Just hope you find a solution ;-)
To my memory I have also used HQPlayer with the Amanero card (original). Here I downloaded a firmware from the Amanero homepage that was optimized specifically for DSD use (but please don't ask me which one
- I don't remember and the card is no longer in use).Just FYI.
Cheers, Jesper
P.S.: it seems there have been more posts while I was writing ... Just hope you find a solution ;-)
I have both, makes no difference in sound.
The second one is the "Amanero Technologies USB Driver 1.0.67 [Wasapi]".
Hans
Hans,
ASIO or WASAPI Exclusive should sound the same for DSD. They may or may not sound exactly the same for PCM, depending. The main difference between the two drivers for DSD is that for some USB boards it is possible to play higher sample rate DSD if using ASIO.
Here is the the thing, for PCM in particular you want to avoid the part of the Windows sound system labeled as "Windows Audio Service" in the diagram below (don't know what it might do to DoP):
The problem is that plain WASAPI, which is shared, goes right through Windows Audio Service which will then resample your audio without telling you if it thinks it needs to. In many cases it will probably think it needs to. If that happens then SQ will suffer somewhat. Not everyone notices it though. They just live with inferior sound without knowing it.
Also, the diagram above is from the FlexASIO documentation. The ASIO driver shown in the diagram is for FlexASIO or ASIO4All, not for full-blown ASIO.
ASIO or WASAPI Exclusive should sound the same for DSD. They may or may not sound exactly the same for PCM, depending. The main difference between the two drivers for DSD is that for some USB boards it is possible to play higher sample rate DSD if using ASIO.
Here is the the thing, for PCM in particular you want to avoid the part of the Windows sound system labeled as "Windows Audio Service" in the diagram below (don't know what it might do to DoP):
The problem is that plain WASAPI, which is shared, goes right through Windows Audio Service which will then resample your audio without telling you if it thinks it needs to. In many cases it will probably think it needs to. If that happens then SQ will suffer somewhat. Not everyone notices it though. They just live with inferior sound without knowing it.
Also, the diagram above is from the FlexASIO documentation. The ASIO driver shown in the diagram is for FlexASIO or ASIO4All, not for full-blown ASIO.
Last edited:
So I tried a very low voltage current limited start up of each board seperately. Hmm something is shorting on both boards, but only after I move from 1V to 2V.
Turns out I have my opamps soldered in backwards
All follow this so it will be a little job. I suspect that there's no damage (certainly nothing appears to be bad electrically other than the wrong way around). So that's the next task when I have some time and I'm not too tired. I could kick myself as the long line obvious points to pin 1 now I look at it.
Turns out I have my opamps soldered in backwards
All follow this so it will be a little job. I suspect that there's no damage (certainly nothing appears to be bad electrically other than the wrong way around). So that's the next task when I have some time and I'm not too tired. I could kick myself as the long line obvious points to pin 1 now I look at it.Oops, good thing you had a current limiter...
When you swap the positive and negative supply pins of just about any IC in a junction isolated process, you forward bias lots of substrate diodes all over the chip. It's usually destructive without current limiter and harmless when the current limit is set low enough not to overheat anything.
When you swap the positive and negative supply pins of just about any IC in a junction isolated process, you forward bias lots of substrate diodes all over the chip. It's usually destructive without current limiter and harmless when the current limit is set low enough not to overheat anything.
During discussions with Hans Polak, it was found that it can sometimes be handy to have a DSD file with repetitive 101010... patterns. For a FIRDAC with a notch at half the sample rate, it eliminates ultrasonic quantization noise, folding effects due to ultrasonic quantization noise and non-linearities or crosstalk issues, and it largely suppresses the effect of far-off clock jitter, leaving only the plain old analogue circuit noise. Attached is a zipped dsf file with just over one minute of 101010... in DSD64. Mind you, the file becomes more than 1000 times larger when you unzip it.
Fingers crossed that it’s all ok. Someone has to be “that guy”. I may start the process tonight.
I was searching on AliExpress for an Antminer unit for use with a ppy DSD'it cape and stumbled across these
https://www.aliexpress.com/item/1005002325408576.html?spm=a2g0o.cart.0.0.6c6438daGr7Pgg&mp=1
https://www.aliexpress.com/item/4001318367175.html?spm=a2g0o.cart.0.0.6c6438daGr7Pgg&mp=1
I've only had a quick look and the descriptions aren't great - I need to read again and look carefully to be sure of what they are but the text does refer to direct decoding of DSD. Big price difference between them too.
Just posting for info - I now have an Antminer on order.
https://www.aliexpress.com/item/1005002325408576.html?spm=a2g0o.cart.0.0.6c6438daGr7Pgg&mp=1
https://www.aliexpress.com/item/4001318367175.html?spm=a2g0o.cart.0.0.6c6438daGr7Pgg&mp=1
I've only had a quick look and the descriptions aren't great - I need to read again and look carefully to be sure of what they are but the text does refer to direct decoding of DSD. Big price difference between them too.
Just posting for info - I now have an Antminer on order.
The stuff on Aliexpress in generally cheap junk. I have to explain that regularly to people expecting a bargain. The components are often fake, caps that should be X7R are Z5U or worse, resistors that should be metal thin film are thick film, etc. PCBs are generally 2-layer when they should be at least 4-layer. Its the only way those guys can sell dac boards that cheap and still make any money.
Usually there is a reason for genuine bargains.
I.e. I just received a few Zynq chips, at about 5 USD each, including VAT and shipping. I suspect that due the high cost of energy, the mining activity has been going down, so this is why there are many boards (Zynq based) and new chips available on Ali. I don't know, this is only my opinion, maybe I am wrong.
I never received a counterfait FPGA - Spartan6 or Cyclone-IV, even if their prices were ridculously small: SLX9 (for simple DSD modulator) at about 5 USD before semiconductors shortage to about 12 USD now, SLX16/25 at about 10-15 USD each. All of them work great.
It is hard to counterfait an FPGA and I supose that they should sell tens of thousands to be economicaly feasable. I.e. in case of opams is a different story, so I never buy opamps from Ali.
I.e. I just received a few Zynq chips, at about 5 USD each, including VAT and shipping. I suspect that due the high cost of energy, the mining activity has been going down, so this is why there are many boards (Zynq based) and new chips available on Ali. I don't know, this is only my opinion, maybe I am wrong.
I never received a counterfait FPGA - Spartan6 or Cyclone-IV, even if their prices were ridculously small: SLX9 (for simple DSD modulator) at about 5 USD before semiconductors shortage to about 12 USD now, SLX16/25 at about 10-15 USD each. All of them work great.
It is hard to counterfait an FPGA and I supose that they should sell tens of thousands to be economicaly feasable. I.e. in case of opams is a different story, so I never buy opamps from Ali.
I agree, being sensible and selective is the way to deal with buying from China - both member Cestrian and I have bought 'cheap' Spartan FPGAs (for PCM2DSD builds) from an AliExpress seller and they work just fine but, like you, I would be extremely cautious about components such as op-amps because there is a demonstrable track record of issues.
Also, I've bought some DAC and amplifier kits from DIYAudio member Abraxalito, who supplies Chinese sourced parts (and makes suggestions for complementary modules), and the only issues I've had have been of my own making.
I do though have other reservations about buying from China for other reasons, but here's not the place for that conversation.
Also, I've bought some DAC and amplifier kits from DIYAudio member Abraxalito, who supplies Chinese sourced parts (and makes suggestions for complementary modules), and the only issues I've had have been of my own making.
I do though have other reservations about buying from China for other reasons, but here's not the place for that conversation.
Related to the received Zynq chips, they are not the best candidate for DSM builds, but at least I can move some peripheral (I2C master/slave, push-buttons, encoders, random number generator) from logic part (PL) to the MCU (PS) part and I will also benefit from the extended logisc of Artix7 compared to Spartan6. Anyway, at first it will be a funny project, for start. My intention is to build later a full development board based on Zynq chip.
First - the correcting the filter board opamps. The board is pulling 37mA on both +15 and -15V rails. After painstaking operation.. seems ok for now.
Next - the DAC itself..
So I've completed correcting the orientation of the ICs in the DAC, the 15 supply draw seems ok at the moment, currently running through the logic section and that seems ok running the clock and a channel with a siglent sig gen. Seems ok so far but I have to go through the other channel too.