All you are saying is that doubling the frequency makes your phase error more impactful. That's a bit obvious a 1% error in a clock divided by 2 gives 0.5% error of the new frequency.
Anyway we are talking about a digital clock. There is no db. It literally does NOTHING by energise flip flops, shift registers and mosfet gates.
The maximal clock error for a 24.576Mhz clock, with reference to what it's being used for is 1/24.576MHz/2. Or 1/12.288Mhz or a period of 81 nano seconds and obviously a frequency component of 12.288Mhz.
You are not hearing that. 81nano seconds is orders of magnitude beyond what any of your audio hardware is capable of even seeing. Speakers do not respond to anything in 81 nano seconds. If you want empirical proof of that ... I don't think a high speed camera exists which can verify the speaker WILL NOT MOVE.
It's such a short amount of time that if you were sitting 8 feet away from your speakers it would take light longer to reach your eyes from the speakers than that phase error in the clock. If you think you can hear that you are nuts.
I don't want subjective, bias non-sense listening test from an audiophool I want you to show me this distortion, empirically on the audio signal. Switch clocks, clock rates, induce distortion on that clock line and show me the result on the audio trace when compared against a control.
Anyway we are talking about a digital clock. There is no db. It literally does NOTHING by energise flip flops, shift registers and mosfet gates.
The maximal clock error for a 24.576Mhz clock, with reference to what it's being used for is 1/24.576MHz/2. Or 1/12.288Mhz or a period of 81 nano seconds and obviously a frequency component of 12.288Mhz.
You are not hearing that. 81nano seconds is orders of magnitude beyond what any of your audio hardware is capable of even seeing. Speakers do not respond to anything in 81 nano seconds. If you want empirical proof of that ... I don't think a high speed camera exists which can verify the speaker WILL NOT MOVE.
It's such a short amount of time that if you were sitting 8 feet away from your speakers it would take light longer to reach your eyes from the speakers than that phase error in the clock. If you think you can hear that you are nuts.
I don't want subjective, bias non-sense listening test from an audiophool I want you to show me this distortion, empirically on the audio signal. Switch clocks, clock rates, induce distortion on that clock line and show me the result on the audio trace when compared against a control.
Good point Stefan, I did not do this recently (adding usb cable and blindly listen) - why should I ? Anyway, it is easy to do and as I am not sure anymore which clock I was using that day. I will do this little check in the coming days and report back if I still have the same experience….
Ok, just did some listening again between RPI usb WaveIO into fifopi and direct RPI to Fifopi.
with the andrea mori clocks, and for all the at 44.1 tracks, if I really, really listen “hard” I can indeed say that the Direct route has a very small amount of extra low level detail (texture and holographic reproduction) - but boy, is it small…. I would never be able to tell after a cold start without knowing what input is actually playing. Any way, there is no real need to add a usb-I2S converter in the digital fifopi chain. But…. I could absolutely recommend using an usb to I2S converter followed by the Fifopi if for whatever reason using a aRPI as “ I2S-generator” is not an option. For example if you run directly from a laptop or Apple device or a streamer with only usb output or if have a diretta streamer ?
with the andrea mori clocks, and for all the at 44.1 tracks, if I really, really listen “hard” I can indeed say that the Direct route has a very small amount of extra low level detail (texture and holographic reproduction) - but boy, is it small…. I would never be able to tell after a cold start without knowing what input is actually playing. Any way, there is no real need to add a usb-I2S converter in the digital fifopi chain. But…. I could absolutely recommend using an usb to I2S converter followed by the Fifopi if for whatever reason using a aRPI as “ I2S-generator” is not an option. For example if you run directly from a laptop or Apple device or a streamer with only usb output or if have a diretta streamer ?
Hi All - NEED HELP please!!
I am using Fifopi Q3 + Reclockpi combo. Till recently, I was using the Accusilicon AS318 45/49M clocks and signal was fed by WaveIO usb to i2s converter. The clocks shifted between 45 to 49 smoothly without any issues.
Now, I am using a BeagleBoneBlack to do the i2s conversion and have upgraded to Andre Mori Drixos and Doublers along with his TWTMC-FSDO Sine to square converter (switched output). I removed the DIL sockets U13 and U7 positions and instead installed 2 Ufl sockets at J12 and J13 positions on the Fifopi.
PROBLEM: Now, when i reboot the fifopi q3, for a brief second, the green LEDs X01 and X02 light up and then it stops at only the X02 (where the green LED stabilises). When I install the ufl cable carrying the clock signal from the Squarer to J12, Fifopi locks i2s signal and beautiful music plays. However, if i take the same ufl cable and install it on J13, the X01 does not get selected and i2s signal does not lock in and fifopi q3 is stuck at EMPTY with RED Light. How can I enable this second Xtal X01 also to accept clock signal and be selected?
I seek your help to solve this problem. Maybe others have also had it. I really hope to solve this problem as the investment made on DRIXOS, Doublers and the Squarer is significant for me.
I am using Fifopi Q3 + Reclockpi combo. Till recently, I was using the Accusilicon AS318 45/49M clocks and signal was fed by WaveIO usb to i2s converter. The clocks shifted between 45 to 49 smoothly without any issues.
Now, I am using a BeagleBoneBlack to do the i2s conversion and have upgraded to Andre Mori Drixos and Doublers along with his TWTMC-FSDO Sine to square converter (switched output). I removed the DIL sockets U13 and U7 positions and instead installed 2 Ufl sockets at J12 and J13 positions on the Fifopi.
PROBLEM: Now, when i reboot the fifopi q3, for a brief second, the green LEDs X01 and X02 light up and then it stops at only the X02 (where the green LED stabilises). When I install the ufl cable carrying the clock signal from the Squarer to J12, Fifopi locks i2s signal and beautiful music plays. However, if i take the same ufl cable and install it on J13, the X01 does not get selected and i2s signal does not lock in and fifopi q3 is stuck at EMPTY with RED Light. How can I enable this second Xtal X01 also to accept clock signal and be selected?
I seek your help to solve this problem. Maybe others have also had it. I really hope to solve this problem as the investment made on DRIXOS, Doublers and the Squarer is significant for me.
A friend of mine ask for audiophile RPi network streamer transport solutions. I have some recommendations for him.
Just to share the options here in case somebody want to know the selections clearly.
"I recommend three fully integrated audiophile RPi network streamer transport solutions for you:
A. Low cost but still great sound quality streamer transport solution
1. TransportPi Digi (#20B)
2. UcPi (#47A)
3. Optional MonitorPi (#49A)
4. Optional acrylic cover (#51A)
B. Fully function audiophile grade RPi I2S/DSD network streamer transport
1. TransportPi MKII (#20A)
2. FifoPi MA (#1C)
3. PurePi (#48A) with optional UcConditioner 3.3V (#26A)
4. Optional MonitorPi (#49A)
5. Optional acrylic cover (#51A)
C. Flagship RPi I2S/DSD network streamer transport
1. HdmiPi Pro (#17B) or TransportPi AES
2. FifoPi Q7 (#1D)
3. StationPi SMT (#35C)
4. LinearPi 5V(#32A)
5. UcConditioner 5V (#25A)
6. LifePO4 Mini 3.3V (#42A)
7. UcConditioner 3.3V (#26A)
8. MonitorPi (#49A)
"
D. Simplified flagship RPi I2S/DSD network streamer transport
1. HdmiPi Pro (#17B) or TransportPi AES
2. FifoPi Q7 (#1D)
3. PurePi (#48A)
4. UcConditioner 3.3V (#26A)
5. MonitorPi (#49A)
https://github.com/iancanada/DocumentDownload
UpgradeUcConditioner by Ian, on Flickr
Ian
Just to share the options here in case somebody want to know the selections clearly.
"I recommend three fully integrated audiophile RPi network streamer transport solutions for you:
A. Low cost but still great sound quality streamer transport solution
1. TransportPi Digi (#20B)
2. UcPi (#47A)
3. Optional MonitorPi (#49A)
4. Optional acrylic cover (#51A)
B. Fully function audiophile grade RPi I2S/DSD network streamer transport
1. TransportPi MKII (#20A)
2. FifoPi MA (#1C)
3. PurePi (#48A) with optional UcConditioner 3.3V (#26A)
4. Optional MonitorPi (#49A)
5. Optional acrylic cover (#51A)
C. Flagship RPi I2S/DSD network streamer transport
1. HdmiPi Pro (#17B) or TransportPi AES
2. FifoPi Q7 (#1D)
3. StationPi SMT (#35C)
4. LinearPi 5V(#32A)
5. UcConditioner 5V (#25A)
6. LifePO4 Mini 3.3V (#42A)
7. UcConditioner 3.3V (#26A)
8. MonitorPi (#49A)
"
D. Simplified flagship RPi I2S/DSD network streamer transport
1. HdmiPi Pro (#17B) or TransportPi AES
2. FifoPi Q7 (#1D)
3. PurePi (#48A)
4. UcConditioner 3.3V (#26A)
5. MonitorPi (#49A)
https://github.com/iancanada/DocumentDownload
UpgradeUcConditioner by Ian, on FlickrIan
Last edited:
@kartick
FifoPi can take single clock. That means you can use one clock either to U13 or U7.
Did you solder the J12 and J13 by yourself? It could be the connector or soldering issues if J12 works but J13 doesn't. You can measure the pad of U13 pin3 to see if there is correct clock signal in LVTTL levels.
Ian
FifoPi can take single clock. That means you can use one clock either to U13 or U7.
Did you solder the J12 and J13 by yourself? It could be the connector or soldering issues if J12 works but J13 doesn't. You can measure the pad of U13 pin3 to see if there is correct clock signal in LVTTL levels.
Ian
Last edited:
Very interesting Ian. I am so pleased with your FifoPi Q7. It is much better than my previous Q2 . If the clocks you are working on is at the same level, I cant’t wait to get them soon enough.
@smooth dancer
Your system looks awesome! What DAC are you using? How do you link the Mcintosh amplifiers?
Ian
Your system looks awesome! What DAC are you using? How do you link the Mcintosh amplifiers?
Ian
Thx, dddac it is, with four dac boards. Psu is from doede, with bigger trafos and chokes from Lundahl and Hammond.
Not sure what you mean with «link the mc amps»? It’s two monoblocks ( 1.2kw) and Mc52 preamp.
Source is Aurender N100h, and diy turntable with Jelco 850L Mk-II.
Dddac psu on pic.
Not sure what you mean with «link the mc amps»? It’s two monoblocks ( 1.2kw) and Mc52 preamp.
Source is Aurender N100h, and diy turntable with Jelco 850L Mk-II.
Dddac psu on pic.
Attachments
Love the big Mac’s. Have tried many amps, but theese are here to stay. They have full control over my jbl 4365’s, and I like to describe them as an iron fist in a cotton glove.
Looking forward to buy your new clocks, Ian. I bet they are better than the Accusilicon 338, which I am using for time being.
Looking forward to buy your new clocks, Ian. I bet they are better than the Accusilicon 338, which I am using for time being.
@smooth dancer
That's so great.
How about AS338. Didn't found from the Assusilicon official website. Is it an internal P/N? What frequencies are you using? How they compared with AS318B?
That's so great.
How about AS338. Didn't found from the Assusilicon official website. Is it an internal P/N? What frequencies are you using? How they compared with AS318B?
338, i bought them online in China. Part of May clock Review ….. Clock review dddac