I know that person... It's me! (hope you get the reference)
That means it needs to be made with an interruption so uC will snap out of whatever it's doing and slap the command on PCM1792.
Thats why I want to make a simple prototype first - to see if I can put it together in a workable way. Then I'll bother with great psu, i/u converter and stuff.
That means it needs to be made with an interruption so uC will snap out of whatever it's doing and slap the command on PCM1792.
Thats why I want to make a simple prototype first - to see if I can put it together in a workable way. Then I'll bother with great psu, i/u converter and stuff.
Hi SomekPoland & Markw4,
Just a couple of brief comments in relation to PCM17** DACs.
I currently use - and have used for 5-6 years now - a somewhat modified DDDAC based on the PCM1794, however, in a non-oversampling mode (NOS). Before completing the DDDAC version I also built a "normal" PCM1794 based DAC. And today I personally would not step back from the DDDAC approach - IMHO it is much smoother, speedier, less gritty - and to me most importantly - very musically sounding. I did not entirely find these qualities in the oversampled PCM1794 version.
I use the DDDAC with an amanero board where I have modified the power supply (as I believe many have done by "cutting" the USB supply). I have also inserted one of Acko's (diyaudio member) reclocking boards, and the clocks are NZ2520SD (the former version - I think the new one is callled NZ2520SDA ???). In my experience adding these clocks is one of the most significant changes I made to this DAC - it improved tonal integrity, tonal nuance, instrument definition, soundstage, precision, well practically all aspects of the sound to a degree that I also would not go back to the standard clocks on the Amanero. My two cents ...
That said I would also like to add that again IMHO it seems to me that the PCM179* DACs can be small gems of audible bliss
, however, they (and I assume "they" even if I have only experience with the PCM1794) really benefit from a careful PCB layout and good components. Also, in my measurements the PCM1794's output is quite HF noisy and my DAC clearly benefited from good HF filtering ....
Now, I don't know your background SomekPoland but I remember reading earlier in the thread that you were not that proficient in digital design (right?). If this is the case I might suggest that you take a look at digital PCB design theory & practice. Literature could be Howard Johnson's High Speed Digital Design & Henry Ott's Electromagnetic Compatibility Engineering. There are also many articles on this topic on the internet. These books & internet articles saved me a couple unpleasant surprises (and a few extra grey hairs) ...
I may also mention Saturn's PCB toolkit software. It is free to download and may help with deciding feasible track widths, PCB thicknesses, inductances, and capacitances when doing a PCB layout. Please note that this is not to deter you - but having my "digital newbie-hood" "behind me" (always more to learn though
) I remember being so very surprised that digital and analog are not really the same ... Well ...
About the DSD1792 I seem to remember reading somewhere that its DSD conversion is rather processed. Thus, depending on your sonic aim with this build I may mention that the AK4499 (native DSD conversion) or PPY's (diyaudio member - to my memory the DSC 2 thread) DSC 2.** DSD converters are considered very good.
Well, I think this is what just came up ... Good luck with your DAC adventure
Cheers,
Jesper
Just a couple of brief comments in relation to PCM17** DACs.
I currently use - and have used for 5-6 years now - a somewhat modified DDDAC based on the PCM1794, however, in a non-oversampling mode (NOS). Before completing the DDDAC version I also built a "normal" PCM1794 based DAC. And today I personally would not step back from the DDDAC approach - IMHO it is much smoother, speedier, less gritty - and to me most importantly - very musically sounding. I did not entirely find these qualities in the oversampled PCM1794 version.
I use the DDDAC with an amanero board where I have modified the power supply (as I believe many have done by "cutting" the USB supply). I have also inserted one of Acko's (diyaudio member) reclocking boards, and the clocks are NZ2520SD (the former version - I think the new one is callled NZ2520SDA ???). In my experience adding these clocks is one of the most significant changes I made to this DAC - it improved tonal integrity, tonal nuance, instrument definition, soundstage, precision, well practically all aspects of the sound to a degree that I also would not go back to the standard clocks on the Amanero. My two cents ...
That said I would also like to add that again IMHO it seems to me that the PCM179* DACs can be small gems of audible bliss
, however, they (and I assume "they" even if I have only experience with the PCM1794) really benefit from a careful PCB layout and good components. Also, in my measurements the PCM1794's output is quite HF noisy and my DAC clearly benefited from good HF filtering .... Now, I don't know your background SomekPoland but I remember reading earlier in the thread that you were not that proficient in digital design (right?). If this is the case I might suggest that you take a look at digital PCB design theory & practice. Literature could be Howard Johnson's High Speed Digital Design & Henry Ott's Electromagnetic Compatibility Engineering. There are also many articles on this topic on the internet. These books & internet articles saved me a couple unpleasant surprises (and a few extra grey hairs
) ...I may also mention Saturn's PCB toolkit software. It is free to download and may help with deciding feasible track widths, PCB thicknesses, inductances, and capacitances when doing a PCB layout. Please note that this is not to deter you - but having my "digital newbie-hood" "behind me" (always more to learn though
) I remember being so very surprised that digital and analog are not really the same ... Well ... About the DSD1792 I seem to remember reading somewhere that its DSD conversion is rather processed. Thus, depending on your sonic aim with this build I may mention that the AK4499 (native DSD conversion) or PPY's (diyaudio member - to my memory the DSC 2 thread) DSC 2.** DSD converters are considered very good.
Well, I think this is what just came up ... Good luck with your DAC adventure
Cheers,
Jesper
I think it is wise to consider advice put forward by those that have had
experience. I had previously recommended the JLsounds to Mark and he is
recommending it to you for good reason. It has superior performance due to
the galvanic isolation, reclocking and better clocks.
Many other people have made this same comparison with the same results.
WRT old R2R formats, the 179x requires right justified if you wish to run mono / external digital filter. This is also worth considering.
4 layer pcb should be sufficient for 179x DACs. Don't use 2.
Your I ref resistor should be 10k not 1k.
Good luck with the project.
TCD
WRT PCM1794 (hardware only mode) vs others, the former has some restrictions including a fixed modulator speed of 64 x FS.
PCM1792A (and others) have native DSD playback bypassing the internal
modulator, routing DSD directly to OP DACs CCS array to implement various
analog low pass filters.
TCD
If you didn't settled on digital interface, check out this review.
Measurements of GUSTARD U16 USB Digital Interface | Audio Science Review (ASR) Forum
The reviewed device is a Gustard U16 digital interface. There's a 60$ Chinese reference unit which outperforms much more expensive Gustard unit.
USBAS318B PCM1536 DSD1024AmaneroXMOSI2S-
USB digital interface AS318B PCM1536 DSD1024 compatible with Amanero Italy XMOS to I2S|Digital-to-Analog Converter| - AliExpress
Measurements of GUSTARD U16 USB Digital Interface | Audio Science Review (ASR) Forum
The reviewed device is a Gustard U16 digital interface. There's a 60$ Chinese reference unit which outperforms much more expensive Gustard unit.
USBAS318B PCM1536 DSD1024AmaneroXMOSI2S-
USB digital interface AS318B PCM1536 DSD1024 compatible with Amanero Italy XMOS to I2S|Digital-to-Analog Converter| - AliExpress
May I get a link of one these reviews please? But a review with measurements. I have both I2SOverUSB and Amanero and I don't think that it I2SOverUSB is a better unit according to my experiences. The magic word `islolator` doesn't work here because XMOS chip is getting highly jittered clock due to the placement of the osciilators (other side of the isolator).
There is a review in Greek. They measuresd V2 unit along with some others. It was worst despite isoltator + reclock stage. Amanero unit did best with around 100ps total jitter.
????????? ???? ?? USB Receivers
Last edited:
@Terry Demol: Hi ... I remember you are quite knowledgeable about the PCM/DSD179* series of DACs so pleased to see you chime in ;-)
As I sort of hinted at in my previous post I am not particularly familiar with the DSD1792 yet had read that the DSD was processed. However, with your comment in mind I have taken a quick look at its datasheet this morning. And I do notice that there are various FIR filters but I cannot from a quick glance see if those filters can be disabled? About feeding the DSD signal directly to the output I cannot see this information in the datasheet (again from a quick glance) but I reckon you have checked this out - so good information, thanks!
@terranigma: Very, very interesting with the USB adapter you linked to - I've been waiting for someone to develop a 1.536 MHz interface for a long time, and now it finally appears to be here ... I did try to download the user manual but I ended up in a dialog box with non-translated Chinese so I couldn't get any further. You wouldn't by any chance have a user manual that you can share (e.g. by email)? ... A P.S.: Came to think of if you know that it actually works - drivers etc.?
Cheers,
Jesper
As I sort of hinted at in my previous post I am not particularly familiar with the DSD1792 yet had read that the DSD was processed. However, with your comment in mind I have taken a quick look at its datasheet this morning. And I do notice that there are various FIR filters but I cannot from a quick glance see if those filters can be disabled? About feeding the DSD signal directly to the output I cannot see this information in the datasheet (again from a quick glance) but I reckon you have checked this out - so good information, thanks!
@terranigma: Very, very interesting with the USB adapter you linked to - I've been waiting for someone to develop a 1.536 MHz interface for a long time, and now it finally appears to be here ... I did try to download the user manual but I ended up in a dialog box with non-translated Chinese so I couldn't get any further. You wouldn't by any chance have a user manual that you can share (e.g. by email)? ... A P.S.: Came to think of if you know that it actually works - drivers etc.?
Cheers,
Jesper
Last edited:
I came across this unit accidentally. I don't have it yet but I will order it. For 60$, that jitter performance is ridiculous in my opinion. Seller on Aliexpress has sent me the manual. It is in Chinese but since it is a technical document, Google translate does it fairly. I'm attaching the document here.
Windows srivers are here. You can alter USB ID's of the device in order to use with other device's drivers.
Support – XING
Last edited:
Thanks Jesper, don't have much forum time these days but can add some info.
The FIR filter *is the OP stage. Similar to DSC1, there are various ways of
arranging the unity weighted OP CCS to form LPF.
cheers
TCD
Sorry, I don't have time to register with this Greek forum and so can't view the images. More importantly, how were the measurements made? Is there
any info WRT the test setup?
Thanks,
TCD
It is not possible for XMOS to decode PCM SR higher than 768k, since it only accepts 50MHz clock as maximum - see their datasheet.
Good point but device is not XMOS based.
Sorry, I don't have time to investigate and go in detail either. Please use Google Translate.
Yes I did (of course). However you have to be a member of the forum to view
any of the images. As far as I could see, there were no details given WRT exact
methodology. This is very important.
TCD
The Xing USB board with Accusilicon clocks might sound better than I2SoverUSB, we don't know yet. Also, we don't have good measurements of either USB board yet in terms of close in phase noise. Xing shows graphs from the Accusilicon data sheet, but that doesn't mean the MCLK output pin on the USB board is as good as the graphs might suggest. Graphs are supposedly of typical performance, and phase noise can be easily degraded by any subsequent circuitry.
For now, from lots of listening experience we know that I2SoverUSB sounds significantly better than any of the other commonly used USB boards including Amanero, at least in most cases and if using the standard onboard clocks.
It there were a question at this point it would only be between Xing and I2SoverUSB as to which one would be the better choice. That assumes of course that sound quality is the only issue, perhaps that and cost. If the primary issue is 'what everyone else is using around here' then maybe Amanero could be the best choice. Its the OP's project, he can decide what is important to him in a dac.
For now, from lots of listening experience we know that I2SoverUSB sounds significantly better than any of the other commonly used USB boards including Amanero, at least in most cases and if using the standard onboard clocks.
It there were a question at this point it would only be between Xing and I2SoverUSB as to which one would be the better choice. That assumes of course that sound quality is the only issue, perhaps that and cost. If the primary issue is 'what everyone else is using around here' then maybe Amanero could be the best choice. Its the OP's project, he can decide what is important to him in a dac.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.