• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

Cronus - It's about time.

Post 721 Russ mentioned something about it with newer version boards. That is all I know.

http://www.diyaudio.com/forums/twisted-pear/272007-cronus-time-73.html#post5468230

"Just a note - we have recently updated Cronus to version 1.4 (available now). It now includes and optional inverting of the buffered clock that goes out to the Hermes modules. This give the builder some flexibility on timing - inverting that clock can make edges align better at the FF with certain sources - like the Hermes-RPI - which is currently in testing.

So in 1.4 you will observe one more jumper that simply selects inverted on non-inverted Hermes output clock. Of course at the final output (same clock that feeds the FF) the signal is not inverted - and so all of the signals are aligned to the actual master clock.

Cheers!
Russ "
 
Hi there

Nice to hear a RPI Hermes is on its way:)

I have an older setup: RPI 3B -> OTTO -> Metronome -> COD DAC.

Would the new Hermes-RPI work fine as an isolator for the RPI and also work with the Metronome?

Does the Hermes comes with U.FL connectors? (If used as a single board without the Cronus)

Regards Mikkel
 
Member
Joined 2015
Paid Member
Need some assistance debugging a Cronus installation.
I followed this post: How to assemble Cronus-Hermes-Amanero to BIIISE

tl;dr:
45/49 rhea pair
J1:2
CPLD Firmware = "Slave_For_1080"
CPU Firmware = "firmware_1096c"
Config bits per the image there.

I'm using Foobar2000 on Windows.
On playback, most of the time there isn't any lock and the player freezes.
After a few tries, there could be a lock and it plays fine. Other times, it may lock but sounds too fast. All files are 44.1khz.

Any ideas?
 
Member
Joined 2015
Paid Member
Need some assistance debugging a Cronus installation.
I followed this post: How to assemble Cronus-Hermes-Amanero to BIIISE

tl;dr:
45/49 rhea pair
J1:2
CPLD Firmware = "Slave_For_1080"
CPU Firmware = "firmware_1096c"
Config bits per the image there.

I'm using Foobar2000 on Windows.
On playback, most of the time there isn't any lock and the player freezes.
After a few tries, there could be a lock and it plays fine. Other times, it may lock but sounds too fast. All files are 44.1khz.

Any ideas?

Not sure what I did differently but I reflashed everything and seems to be fine now :)
 
Need some assistance debugging a Cronus installation.
I followed this post: How to assemble Cronus-Hermes-Amanero to BIIISE

tl;dr:
45/49 rhea pair
J1:2
CPLD Firmware = "Slave_For_1080"
CPU Firmware = "firmware_1096c"
Config bits per the image there.

I'm using Foobar2000 on Windows.
On playback, most of the time there isn't any lock and the player freezes.
After a few tries, there could be a lock and it plays fine. Other times, it may lock but sounds too fast. All files are 44.1khz.

Any ideas?


I had similar issues with Amanero and i initially fixed them by moving J_INV jumper on MCK- on Cronus, as the issues were present with it on MCK+ ...
Strangely enough, after one week it started to work ok with jumper on MCK+ too.
There could be player/amanero problems too.
 
Last edited:
Member
Joined 2015
Paid Member
Not sure what I did differently but I reflashed everything and seems to be fine now :)

Rethinking about this for a moment, when I reflashed I had only flashed the CPLD and CPU firmwares.
I did not modify the config bits as in the image. (MCLK slave mode and pin 11 selector)

Could there be a possibility that music is playing but Cronus is not actually reclocking the input from amanero?
 
Member
Joined 2015
Paid Member
I had similar issues with Amanero and i initially fixed them by moving J_INV jumper on MCK- on Cronus, as the issues were present with it on MCK+ ...
Strangely enough, after one week it started to work ok with jumper on MCK+ too.
There could be player/amanero problems too.

Hmm, I had just realized that I totally skipped over this (I don't recall having 2 jumpers in the kit).
I hadn't soldered any headers on the J_INV... I am assuming that is problematic? :)
 
Hmm, I had just realized that I totally skipped over this (I don't recall having 2 jumpers in the kit).
I hadn't soldered any headers on the J_INV... I am assuming that is problematic? :)

What i understand is that it could help when having problems for a device to lock on cronus mck.
But if everything locks correctly without it, it is not strictly needed (it was not present until newer boards) - anyway if you have J_INV oboard, you have to bridge either MCK+ or MCK-, i think...
 
Last edited:
Member
Joined 2015
Paid Member
What i understand is that it could help when having problems for a device to lock on cronus mck.
But if everything locks correctly without it, it is not strictly needed (it was not present until newer boards)

I just soldered the J_INV header and put a jumper on MCK+ and it looks like its working now.
Is there a way to know the current firmware on the Amanero? I did flash it to Slave but just want to recheck it as a sanity check.

Previously I did not have a lock on the B3sePro
 
Hmm, I had just realized that I totally skipped over this (I don't recall having 2 jumpers in the kit).
I hadn't soldered any headers on the J_INV... I am assuming that is problematic? :)

Good catch - I will add this info to the Cronus overview and the first page.

The jumper allows you to send master clock to Hermes modules either inverted - or non-inverted. In certain (rare) cases sending the master clock inverted can help with phase alignment at the FF. I added the header to give that option. It should work with current Hermes modules with either position jumped.

Cheers!
Russ
 
Member
Joined 2007
Paid Member
Background: Back in March we had a short conversation that started by asking what newer SoC solutions might enable a new level of PCM digital processing. We just dropped it at the time, as the Hermes-Pi was still in development. At this time I don't see a Hermes-RPi among the current TPA offerings.

Yeah, I was under the impression that the Pi only had 2 i2s data channels, but then saw an 8-channel dac hat that somebody put together. Maybe they re-purposed some of the GPIO pins.

Now, 8 months later, several new SoC boards have been introduced. Beagleboard.org now offers a new 'mini' option - the PocketBeagle - and a WiFi version of BBB. Raspberry introduced what looks like the killer of the bunch - the RPi 3A+.

YouTube

The price of the RPi 3A+? ...are you sitting down? It's $25.00.

Idle speculation: A person could do some decent FIR filtering with 4 cores at 1.4GHz. For crossovers why not slave an additional RPi (per stereo channel) to the the master player board and run the filters full time on an RPi slave? Three-way multi-amped system (like mine)? Three output boards. This would eliminate the sample-rate paralysis that otherwise afflicts Linux running BruteFIR. What can the 1.4GHz chip do? Looks like 8192 taps @ 44.1kHz is reliable...
[Guide] High quality room correction with Raspberry Pi & Roon - Raspberry Pi - Roon Labs Community
dsp/brutefir/raspberry-pi-sample-config - audiophile

I believe 2048 taps could give adequate resolution in the bass-midrange interface. If so, the RPi 3A+ would have a good chance of running effective FIR crossover filters at 176kHz and 192kHz. So, integer-based up-sampling to either 176 or 192kHz (in ALSA?) would mean only two filter sets are needed, selectable by the master board. Room correction filters? ...Probably limited to 44.1 and 48kHz due to the need for 8k or 16k taps.

Questions: 1. What is the status of Hermes-RPi? 2. What is the current state of the Hermes-RPi driver? tpa-hermes-rpi-dev/README.md at master * miero/tpa-hermes-rpi-dev * GitHub. or maybe 3. Is the TPA skunkworks working on something new for this kind of PCM mutillation? :p
 
I'm just curious why have to use Hermes at all. A short ribbon cable can connect the BBB (or RPI) and the Cronus well, and:

1. Isolators often negatively effects sound quality. According to my personal experiencies always.
2. Excessive path length also can degrade sound, as clock sign is easily distorted.

If sound quality is not the most important aspect, there are also some benefits using such a card.

Is there any documentation of direct wireing BBB and Cronus?
 
Abra, Hermes-BBB also protects sensitive pins of BBB. There should not be any voltage on them if BBB is powered down.

Isolators often negatively effects sound quality. According to my personal experiencies always.
2. Excessive path length also can degrade sound, as clock sign is easily distorted.
Sound quality and signal quality is a different thing. If you are using Cronus then a possible negative effect of used isolator is compensated.

But if you want to experiment, you can check pin names of Hermes-BBB connectors in the schematics and make direct wiring according to them:
- Hermes-BBB/hermes-bbb-schematic.pdf at master * twistedpearaudio/Hermes-BBB * GitHub
 
Abra, Hermes-BBB also protects sensitive pins of BBB. There should not be any voltage on them if BBB is powered down.


Sound quality and signal quality is a different thing. If you are using Cronus then a possible negative effect of used isolator is compensated.

But if you want to experiment, you can check pin names of Hermes-BBB connectors in the schematics and make direct wiring according to them:
- Hermes-BBB/hermes-bbb-schematic.pdf at master * twistedpearaudio/Hermes-BBB * GitHub

Just wanting to try where the boundaries of synchronous mode are. My device is used in sync mode currently. Actually I like it's sound with the Pulsar clocks. But I found async mode also have sonical advantages in some respects, even if the DAC has a less refined NDK NZ2520SD clock.

I don't understand why to "compensate" parts, which probably unnecessary in my device, and can be removed instead. Anyhow it's worth a try. Thank you for your help, it makes easier to try the direct wireing.


"Abra, Hermes-BBB also protects sensitive pins of BBB. There should not be any voltage on them if BBB is powered down."
All the SBC-s I tried previously previously well worked without isolators. Different versions of RPI, Odroid C1 & C2, Cubie Truck and a few others.
Is the BBB a factory documented exception?