Hi All:
I need help with an issue using the Cronus/Hermes/Amanero with a Soekris DAM1021 DAC when playing music from YouTube.
The source is a MacBook Pro. I either play music through Audirvana or use a Chrome browser to play videos from YouTube on this Mac .
The DAM1021 runs on a +/- 9.3VDC PSU and the Cronus/Hermes runs on its own 6.5VDC PS.
When I play music with Audirvana, the whole setup runs perfectly fine. Solid lock to the i2s signal on all resolutions and beautiful music.
The problem comes when I play music from YouTube. The i2s lock is not stable with many irregular drop-outs lasting fraction of a second. Interestingly, the drop-out frequency varies from video to video, some videos consistently do worse than the others.
The odd thing is that if I play those same videos on YouTube using the same USB output on the Mac and through a USB-SPDIF converter and feed the SPDIF into the DAM1021, I have solid lock and no drop-out at all.
I can also play videos on YouTube through my dual-mono Buffalo II via another Cronus/Hermes/Amanero (a newer version than the one I am using with the DAM1021) with solid lock and no drop-out.
I have been able to consistently recreate all the above scenarios. There must be a logical explanation to this seemingly mystery. Can someone shed lights on what's going on and how to resolve this issue?
Thanks!
EDIT: forgot to mention that the same setup plays YouTube videos fine with solid lock using Amanero (same one I am using with Cronus/Hermes now) alone.
I need help with an issue using the Cronus/Hermes/Amanero with a Soekris DAM1021 DAC when playing music from YouTube.
The source is a MacBook Pro. I either play music through Audirvana or use a Chrome browser to play videos from YouTube on this Mac .
The DAM1021 runs on a +/- 9.3VDC PSU and the Cronus/Hermes runs on its own 6.5VDC PS.
When I play music with Audirvana, the whole setup runs perfectly fine. Solid lock to the i2s signal on all resolutions and beautiful music.
The problem comes when I play music from YouTube. The i2s lock is not stable with many irregular drop-outs lasting fraction of a second. Interestingly, the drop-out frequency varies from video to video, some videos consistently do worse than the others.
The odd thing is that if I play those same videos on YouTube using the same USB output on the Mac and through a USB-SPDIF converter and feed the SPDIF into the DAM1021, I have solid lock and no drop-out at all.
I can also play videos on YouTube through my dual-mono Buffalo II via another Cronus/Hermes/Amanero (a newer version than the one I am using with the DAM1021) with solid lock and no drop-out.
I have been able to consistently recreate all the above scenarios. There must be a logical explanation to this seemingly mystery. Can someone shed lights on what's going on and how to resolve this issue?
Thanks!
EDIT: forgot to mention that the same setup plays YouTube videos fine with solid lock using Amanero (same one I am using with Cronus/Hermes now) alone.
Last edited:
Hi Miero:
The DAM1021 uses a Si514 for the clock and a FIFO reclocker. The i2s is also isolated.
I know the Cronus/Hermes is probably an overkill in this case but I have an extra one from my Buffalo DAC project two years ago anyway.
What puzzles me is that it works fine with Audirvana and iTune, the drop-out only occures when playing YouTube. I tried Amazon streaming and experience similar drop-out but to a much lesser degree.
The DAM1021 uses a Si514 for the clock and a FIFO reclocker. The i2s is also isolated.
I know the Cronus/Hermes is probably an overkill in this case but I have an extra one from my Buffalo DAC project two years ago anyway.
What puzzles me is that it works fine with Audirvana and iTune, the drop-out only occures when playing YouTube. I tried Amazon streaming and experience similar drop-out but to a much lesser degree.
OK. Try to remove 48fs clock from Cronus if the sound will be playing from Youtube. It is possible that incorrect clock is being used for 44.1fs. If Youtube will not play then this might be the case.
Notice: I'm not sure if this would work, if it does not then try to swap clocks on Cronus with each other.
Notice: I'm not sure if this would work, if it does not then try to swap clocks on Cronus with each other.
Last edited:
Thanks Miero.
Unfortunately it’s not easy for me to remove or swap the clocks on my Cronus because I mistakenly soldered them directly on the Cronus board.
I plan to swap the Cronus/Hermes in my Buffalo DAC to use with the DAM1021 to see if that works and report back. This Cronus is the newer version with the buffered clock output.
I should mention I have many content in iTune that’s 44.1KHz frequency family and the Cronus has no issue choosing the correct clock.
Anything else do you recommend me trying?
Unfortunately it’s not easy for me to remove or swap the clocks on my Cronus because I mistakenly soldered them directly on the Cronus board.
I plan to swap the Cronus/Hermes in my Buffalo DAC to use with the DAM1021 to see if that works and report back. This Cronus is the newer version with the buffered clock output.
I should mention I have many content in iTune that’s 44.1KHz frequency family and the Cronus has no issue choosing the correct clock.
Anything else do you recommend me trying?
Last edited:
I experimented using the Cronus/Hermes (one with the buffered clock output) from my Buffalo DAC on the DAM1021. I also dropped the PS of the Cronus to 5.2VDC. Playing videos from YouTube, it improves significantly, eliminating about 80% of the drop-out but not completely.
I guess my only remaining option is to experiment with firmwares on the Amanero which I'll do next.
At this point I am prepared to accept that the Cronus/Hermes/Amanero may not be compatible with the DAM1021. It still puzzles me very much why the combo works fine when playing music through Audirvana and iTune but not YouTube. Can someone venture a logical hypothesis? Could it be that the DAM1021's own i2s isolator and re-clocker be the reason?
Thanks in advance for any thoughts you can share!
I guess my only remaining option is to experiment with firmwares on the Amanero which I'll do next.
At this point I am prepared to accept that the Cronus/Hermes/Amanero may not be compatible with the DAM1021. It still puzzles me very much why the combo works fine when playing music through Audirvana and iTune but not YouTube. Can someone venture a logical hypothesis? Could it be that the DAM1021's own i2s isolator and re-clocker be the reason?
Thanks in advance for any thoughts you can share!
The symptoms you are seeing could be timing related (because of the extra isolators at your DAC which will introduce a phase shift), but the reason I say to check driver/firmware is because it only happens during certain playback conditions - those that are not exclusive access (like your audivana player likely is) are the only ones that are problematic. This leads me to believe some experimenting on that end may yield results.
@ Miero: I started this project using Amanero alone with the DAM1021 and it works fine with both Audirvana and YouTube.
@ Russ: I am going to experiment with both Amanero CPLD and CPU firmwares as you suggested. I was doing some research last night and not able to find much information of the different versionns of the firmware. Maybe I did not look at the right places. There are a few versions of the CPLD firmware for the slave mode and many versions of the CPU firmwares most of them seem to be addressing DSD playback. Any suggestion of which ones to start with would be much appreciated.
Thanks both for your help!
@ Russ: I am going to experiment with both Amanero CPLD and CPU firmwares as you suggested. I was doing some research last night and not able to find much information of the different versionns of the firmware. Maybe I did not look at the right places. There are a few versions of the CPLD firmware for the slave mode and many versions of the CPU firmwares most of them seem to be addressing DSD playback. Any suggestion of which ones to start with would be much appreciated.
Thanks both for your help!
Last edited:
I took a short look at the way your DAC works - and I suspect the issue might be around the way that DAC generates it's clock - the skew between the cronus and the DAC may be too great. (phase not aligning as expected) - but I am still baffled why that would only occur in certain playback scenarios 
Can't say I have ever seen anything like that.
Can't say I have ever seen anything like that.[QUOTE ...I am still baffled why that would only occur in certain playback scenarios Can't say I have ever seen anything like that.[/QUOTE]
That’s the part confuses me the most. Here are few possible clues:
1. seems like the signal streamed from the internet (YouTube, Amazon, etc.) are the ones that exihibit the drop-out issue.
2. On YouTube, some videos have noticeable more drop-outs than others do and it’s consistent every time I tried playing them.
3. My Cronus with the buffered clock output has far less drop-out than the one without the buffered clock output playing the same material. The difference is very significant and noticeable.
4. Some drop-outs are very short, lasting fraction of a second, barely noticeable. Some are bad enough the DAM1021 would temporarily lose the lock for maybe a second or so but always regain the lock quickly.
Can't say I have ever seen anything like that.[/QUOTE]That’s the part confuses me the most. Here are few possible clues:
1. seems like the signal streamed from the internet (YouTube, Amazon, etc.) are the ones that exihibit the drop-out issue.
2. On YouTube, some videos have noticeable more drop-outs than others do and it’s consistent every time I tried playing them.
3. My Cronus with the buffered clock output has far less drop-out than the one without the buffered clock output playing the same material. The difference is very significant and noticeable.
4. Some drop-outs are very short, lasting fraction of a second, barely noticeable. Some are bad enough the DAM1021 would temporarily lose the lock for maybe a second or so but always regain the lock quickly.
Thanks Russ and Miero again for your help.
I am happy to report that experimenting with firmwares paid off. I found a combination that works for both Audirvana and YouTube playback: CPLD = "Slave_RJLR" CPU = "firmware_1096d_RJ".
I started out by trying every slave CPLD firmware available in the OEM1.17 tool from Amanero and paired them all with CPU firmware "firmware_2002b". CPLD firmware Slave_1095_div2 and Slave_RJLR2 could not obtain lock at all. Slave12288 gave distorted sound on both Audirvana and YouTube playback. Slave2224 plays Audirvana fine but gives scattered cracking noise playing YouTube. Slave_for_1080 and Slave_RJLR plays Audirvana fine but have drop-outs playing YouTube. I also paired Slave_RJLR2 with CPU firmware_1095LR as indicated by the description of Slave_RJLR2 but could not get lock at all.
The description of the CPLD firmware Slave_RJLR states "Slave Mode Input MCLK PIN6 22.572MHz and 24.576MHz Left = DATA_PIN Right = DSDOE RJ aligned - CPU firmware_1096LR". I could not find firmware_1096LR so I used firmware_1096d_RJ instead and it worked.
I should mention that the first few seconds of the first YouTube video I played with this working combination were all cracking noises, the second video had a few clicking noise but no drop-out. The third video played fine with no noise or drop-out I could detect. I then went back and played the first two videos and they played fine - no noise, no drop-outs (knock on wood). It's as if the Cronus/Hermes/Amanero and the DAM1021 had a "learning mode" where they learned to work together.
I could not find any information on Slave_RJLR and firmware_1096d_RJ on the web and would appreciate any information about these firmwares that anyone can share.
I am happy to report that experimenting with firmwares paid off. I found a combination that works for both Audirvana and YouTube playback: CPLD = "Slave_RJLR" CPU = "firmware_1096d_RJ".
I started out by trying every slave CPLD firmware available in the OEM1.17 tool from Amanero and paired them all with CPU firmware "firmware_2002b". CPLD firmware Slave_1095_div2 and Slave_RJLR2 could not obtain lock at all. Slave12288 gave distorted sound on both Audirvana and YouTube playback. Slave2224 plays Audirvana fine but gives scattered cracking noise playing YouTube. Slave_for_1080 and Slave_RJLR plays Audirvana fine but have drop-outs playing YouTube. I also paired Slave_RJLR2 with CPU firmware_1095LR as indicated by the description of Slave_RJLR2 but could not get lock at all.
The description of the CPLD firmware Slave_RJLR states "Slave Mode Input MCLK PIN6 22.572MHz and 24.576MHz Left = DATA_PIN Right = DSDOE RJ aligned - CPU firmware_1096LR". I could not find firmware_1096LR so I used firmware_1096d_RJ instead and it worked.
I should mention that the first few seconds of the first YouTube video I played with this working combination were all cracking noises, the second video had a few clicking noise but no drop-out. The third video played fine with no noise or drop-out I could detect. I then went back and played the first two videos and they played fine - no noise, no drop-outs (knock on wood). It's as if the Cronus/Hermes/Amanero and the DAM1021 had a "learning mode" where they learned to work together.
I could not find any information on Slave_RJLR and firmware_1096d_RJ on the web and would appreciate any information about these firmwares that anyone can share.
Found another firmware combination that works with DAM1021: CPLD = Salve _For _1080" and CPU = "firmware_1096d_RJ". Looks like the CPU firmware may be the magic here. There are likely other combinations that will work.
I did go back and use Amanero alone with the DAM1021. Even though the DAM1021 has on board i2s isolator and re-clocking, I feel the Cronus/Hermes/Amanero combo still sounds better than using Amanero alone. This, of course, is subjective impression.
I did go back and use Amanero alone with the DAM1021. Even though the DAM1021 has on board i2s isolator and re-clocking, I feel the Cronus/Hermes/Amanero combo still sounds better than using Amanero alone. This, of course, is subjective impression.
Last edited:
I2S to my knowledge. Isn't I2S the interface protocol and the PCM the data format?
I cannot find any information on RJ-PCM. I assume RJ is "Right Justified"? I see that in the Amanero's firmware tool, under the "stream format" the choices are "I2S", "Left justified" and "Right justified". Right now I use "I2S" and the whole combo works. I wonder if I should try "Right justified"?
I cannot find any information on RJ-PCM. I assume RJ is "Right Justified"? I see that in the Amanero's firmware tool, under the "stream format" the choices are "I2S", "Left justified" and "Right justified". Right now I use "I2S" and the whole combo works. I wonder if I should try "Right justified"?
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.