[bimmer100]

I'm sure I can get ahold of Robert @cerious.
I've been talking quite a bit with him over the last week. I've ordered about 3 cables recently, and had another half dozen ordered from friends. Along with a group buy deal i'm putting together. What questions do you want to know?

 

[bimmer100]

  I found some commercial gimmicks in TaoBao today:
 
Q-1  Xmos/XU208 @RMB239.00 each,

 
Q-1S  Xmos/XU208 @RMB239.00 each,

 
X-1  Xmos/XU-208 @RMB369.00 each

 
F-1  Xmos/XU-208 @RMB999.00 each

Is there any more info on the F1?  as in...is it powered with USB only? or can it accept external PSU...any instructions on this? and what is the pin/out for this?
 
I would consider integrating this into my M11 as the form factor might be possible. But the PSU would be relevant.

 

[hugoboss]

@bimmer
you can use dual head usb cable with the 5volt going to linear psu.
iam gonna use this way

 

[Caper]

  Is there any more info on the F1?  as in...is it powered with USB only? or can it accept external PSU...any instructions on this? and what is the pin/out for this?
 
I would consider integrating this into my M11 as the form factor might be possible. But the PSU would be relevant.

I am going to use one Aqvox psu feeding Intona USB isolator and then to the F1.

 

[abartels]

Hi guys,
 
Just for your information, since half a year there's a new Crystek TCXO called the CCHD-950.
This is probably the best commercial TCXO (not talking about Pulsar clocks of $350......)
 

 

[hgpsemaj]

  Is there any more info on the F1?  as in...is it powered with USB only? or can it accept external PSU...any instructions on this? and what is the pin/out for this?
 
I would consider integrating this into my M11 as the form factor might be possible. But the PSU would be relevant.

 
 
Q1, X1 & F1 could be powered either (i) via USB only, or (ii) via an external 5V PS, but it would still consume 5V via USB simultaneously. And, you may need to remove some parts, if you adopt external 5V method.
I did asked for the images in respect of removing those parts. However, the supplier hasn't got any for this very moment. He also insists that this requires a certain ability to perform such task, and generally is not recommended.
 
 
有關 Q1,X1 和 F1, 我要幫 Head-Fi 論擅一眾確定一些電源問題.(i) Q1,X1 和 F1,都可單獨地從 USB 供電; 或 (ii) 拆掉隔离电源模块，然后单独使用5v直接供电给音频那边,但它同時需要從USB口輸入5V電.



對嗎?



xiu305470199

是的



xiu305470199

音频那边可以单独5V供电，同时USB这边用USB供电就可以了。



 

18​

18:47​

有 '拆掉隔离电源模块' 的圖片嗎?



xiu305470199

没有哦



ok,



xiu305470199

这个需要一定的动手能力的，一般不推荐改动

 

[hgpsemaj]

  Is there any more info on the F1?  as in...is it powered with USB only? or can it accept external PSU...any instructions on this? and what is the pin/out for this?
 
I would consider integrating this into my M11 as the form factor might be possible. But the PSU would be relevant.

 
F1 Xmos USB

 
1 , full isolation technology, 150Mbps full-chip isolation, interference can be completely isolated from the PC;
 
2, source synchronous shaping technology and CPLD technology, I2S signal isolation after the re-shaping; thus eliminating the isolation chip to bring added jitter;
 
3, the independent development of the system clock, using CRYSTEK customized high-performance crystal, ultra-low phase noise, low jitter.
F-1 digital interface is our team to develop a high-performance USB digital audio interface, using the XMOS latest xCORE-200 series chips. F-1 is a digital interface, ultra-low jitter clock system. F-1 digital input interface using a standard USB2.0 interface, while F-1 has a wealth of output interfaces, including coaxial RCA and I2S interfaces.
 
Features:
1, F-1 is the first onr,using the latest XMOS xCORE-200 series of high-end USB digital interface products. New generation XMOS chip, using more advanced process technology, performance than the older chip U8 doubled, reaching 1000MIPS processing capabilities.
2, it is known, the weight of the core is digital audio algorithms. Our team in the field of digital audio work for many years, and the strong support of XMOS FAE team, we conducted a number of technical improvements and optimization algorithms on the basis of the original firmware.
 
Specifications:
Each output interface supports sample rates:
PCM: 44.1KHz, 48KHz, 88.2KHz, 96KHz,
   176.4KHz, 192KHz, 352.8KHz, 384KHz
    [Wherein I2S out support full sample rate, S / PDIF supports up to 192KHz]
DSD: 2.8 MHz (DSD64) - DoP, native
    5.6 MHz (DSD128) - DoP, native
    11.2 MHz (DSD256) - DOP, native
     [Wherein all DSD format support I2S out, S / PDIF and AES / EBU support DSD64 DOP mode]
Bits wide: the highest 32 bit over I2S output
The highest 24 bit over S / PDIF
 
Screw position, I2S pin position is fully compatible with Italy Amanero module
The entire board with three crystal design, all are imported SMD crystal using active crystal. Wherein the audio oscillator is CRYSTEK CCHD series customized version of the crystal.
 
Each interface electrical standards:
1, USB input socket is a standard USB-B type female, USB power supply range 4.5V-5.1V;
2, RCA connector outputs standard S / PDIF signal level is 550mV (standard connection load), the output impedance of 75 ohms;
 
Design details:
1, carry out electricity as the mother of the acoustic design concept, pay attention to the power supply circuit design, high-performance, low noise and fast response LDO as the main power source. The entire board using a 2-way power supply independent LDO, wherein the clock section even using ADI's ultra low noise LDO, in order to ensure the ultimate output performance laid a solid foundation.
2, PCB board with 4-layer circuit board design, ensure that you have full power and ground plane layer, using the method of high-speed digital design, signal integrity and power integrity of the entire board. The clock signal for special treatment, using package design and precise impedance control, thereby ensuring the signal quality and increase the clock anti-jamming capability. Input and output signals follow a standard design, especially high-speed USB signals using a 90 ohm differential impedance control.
 
System Compatibility:
1, Windows XP, Windows 7, Windows 8; 32/64 bit, you need to install special drivers
2, Native MacOS 10.6 and later, using the system comes with drivers
3, Native Linux with UAC2 compliant kernel, using the system comes with drivers
4, Android OS 4.2 and above, the need to support OTG function.
 
 
 
I2S signal output socket Description:
1, MUTE is high mute; the DSD data stream DSD_ON high level;
2. All electrical average LVCMOS, voltage of 3.3V;
3, VD3.3 output power, low output current 50mA;
4, MCLK clock output 22.5792Mhz or 24.576Mhz;
5, FS0, FS1, FS2 is sampling rate indicator level, but FS3 not;

 ​

6, LRCLK = DSDR, SDATA = DSDL, BCLK = DCLK.
 

 
 
 
WARM WARNING: diyinhk has launched Xmos XU216 2000MIPS recently.

 

[hgpsemaj]

  Hi guys,
 
Just for your information, since half a year there's a new Crystek TCXO called the CCHD-950.
This is probably the best commercial TCXO (not talking about Pulsar clocks of $350......)
 

 
Looks great, what's the RRP? You sure it could match NDK NZ2520SD?
 
 
Regards,

 

[abartels]

It sells for about $20
 
Below 1000 Hz NDK NZ2520SD still is better, but I think this is a very good performer.
 
It seems it's not only the low Phase Noise that's important for audio.
There are OCXO's who sound better than NDK but have Phase Noise specs which are worse, so I'm not sure.
 
Crystal, for sure, is a company which designed, developed and produced audio xo's for a very long time.
The first ones I bought from them was at the beginning of the nineties, when I developed and produced a commercial dac with Crystal Semiconductor (nowadays Cirrus Logic) CS4328 and CS8412.
Crystal even advised in their whitepapers to use the Crystek TCXO's back then.
 
Until now I didn't compare NDK with Crystek soundwise, probably I will when I have Ian's FIFO II - Isolator - Dual Clockboard II working with my RPI.
The clockboard has sockets so very simple to change the XO's. I have 45.xxx - 49.xxx NDK's at hand, will use them first, and order the Crystek 950's to compare.
 
Later on I will try OCXO's, as you advised me
 
 
Regards,
 
Alex

 

[Caper]

Oh wow, didnt even take two hours since I mentioned in the other usb/ddc thread that the next 'usb wonder' is coming soon

Good luck with your new thingie but dont count one people like me .. nowadays I'm trying to get rid of all the usb nonsense I bought last year.

Other than that I got some hopes for alex's no-usb solution .. an rpi with clean i2s conn inside my dac sounds just perfect. No extra boxes, no cable clutter, no worries about the poor bits ... just music .. controlled from my phone on the couch

 
I havent followed all threads, what is " alex's no-usb solution" ??

 

[prot]

See abartels' posts in this very thread
Basically you put an rpi inside the dac. Music bytes go from rpi to an i2s filter and then directly to the dac via i2s. Like mentioned, no usb, no extra boxes, cables, etc. Sounds like a winner to me. And his first experiments seem to be quite encouraging in the SQ department.

 

[Superdad]

  Question for Superdad: Have you considered designing a state of the art DDC with your upcoming uber-Regen built-in?

Oh wow.  Well the answer(s) to that question would be rather complicated and involve my revealing a lot of forward-looking things--and new interfaces--that John Swenson and I (well he's the engineer, I'm just 'Jobs') are cooking up.  
 
A big part of the answer is no. And it is not because John doesn't know how to design world-class S/PDIF inputs and outputs (for which there is far more to than what I see discussed here).  Take a look at the Sonore Signature Rendu, an Ethernet>S/PDIF (and also I2S over LVDS) renderer.  Yes, it is pricy at $3K, but that is largely due to the cost of the Ethernet renderer module that Sonore OEMs from elsewhere.  The S/PDIF output stage (and the LVDS I2S output) is a Swenson-designed board and considered among the greatest S/PDIF outputs ever.  And then, on the other side, for S/PDIF in, the input of the BottleHead DAC he designed is chock full of innovation.
 
But the truth is, the whole clock-embedded S/PDIF interface is a flawed dead end (sacrilege!).  And the S/PDIF inputs of DACs have wild variability as well.  So sure, USB is not perfect--especially due to the many stumbles we see in USB input stages and USB>I2S board designs--but that's what the computer OSs support best at this point in time, and until there are broadly compatible, open-source "virtual sound card" Ethernet drivers, USB is here to stay if one wants commercial (read compatible) viability.
 
And all a great USB>S/PDIF interface says to me is that the DAC's USB interface should have been done better.  Believe me, I am tempted every week to consider working with John on an awesome USB>S/PDIF (and I2S) converter.  But neither of our hearts would be in it, and while the REGEN is modestly priced enough, I don't think people would be too pleased with what we would have to sell a SOTA DDC for.
 
Yet there are other ways to approach USB interfaces.  Ways that promise to overcome most all the cable, ports, power supply, and computer issues--as in making none of that stuff matter any more.  It involves splitting the whole thing into two ends, with a really accurate and robust interface in between.  And completely different tech and electrical signaling than is used anywhere else in audio.  Can't say more.
Trouble is, the question then becomes what is at the "DAC end" of this new interface?  Not another USB or S/PDIF output--yuck--defeats the purpose.  I2S over LVDS?  Sure, but that limits popularity with DACs--and still the master clocking is not where you want it, which is close to the DAC chip.  What one wants is a DAC with this interface built in.  Which leads us to either enticing OEMs to embrace and license a new solution, and/or do our own DAC to demonstrate the tech.
 
But we have two other products to get into production first--and I have already said way too much. 

 
 
Have a great weekend everyone.
 
--Alex C.
 
P.S.  Took my wife and kids to see Paul McCartney at the Fresno Savemart Center (16K sold out seats) last night.  As our 14 year old son put it, "the sound was like ass", but our seats were great and Sir Paul puts on a great show!

 

[robertsong]

   
 
But we have two other products to get into production first--and I have already said way too much. 

 
 
Have a great weekend everyone.
 
--Alex C.
 
P.S.  Took my wife and kids to see Paul McCartney at the Fresno Savemart Center (16K sold out seats) last night.  As our 14 year old son put it, "the sound was like ass", but our seats were great and Sir Paul puts on a great show!

Haha. Thank you very much for that info. Greatly appreciated. I'll keep an eye out for the uber.

 

[rb2013]

 
Oh wow.  Well the answer(s) to that question would be rather complicated and involve my revealing a lot of forward-looking things--and new interfaces--that John Swenson and I (well he's the engineer, I'm just 'Jobs') are cooking up.  
 
A big part of the answer is no. And it is not because John doesn't know how to design world-class S/PDIF inputs and outputs (for which there is far more to than what I see discussed here).  Take a look at the Sonore Signature Rendu, an Ethernet>S/PDIF (and also I2S over LVDS) renderer.  Yes, it is pricy at $3K, but that is largely due to the cost of the Ethernet renderer module that Sonore OEMs from elsewhere.  The S/PDIF output stage (and the LVDS I2S output) is a Swenson-designed board and considered among the greatest S/PDIF outputs ever.  And then, on the other side, for S/PDIF in, the input of the BottleHead DAC he designed is chock full of innovation.
 
But the truth is, the whole clock-embedded S/PDIF interface is a flawed dead end (sacrilege!).  And the S/PDIF inputs of DACs have wild variability as well.  So sure, USB is not perfect--especially due to the many stumbles we see in USB input stages and USB>I2S board designs--but that's what the computer OSs support best at this point in time, and until there are broadly compatible, open-source "virtual sound card" Ethernet drivers, USB is here to stay if one wants commercial (read compatible) viability.
 
And all a great USB>S/PDIF interface says to me is that the DAC's USB interface should have been done better.  Believe me, I am tempted every week to consider working with John on an awesome USB>S/PDIF (and I2S) converter.  But neither of our hearts would be in it, and while the REGEN is modestly priced enough, I don't think people would be too pleased with what we would have to sell a SOTA DDC for.
 
Yet there are other ways to approach USB interfaces.  Ways that promise to overcome most all the cable, ports, power supply, and computer issues--as in making none of that stuff matter any more.  It involves splitting the whole thing into two ends, with a really accurate and robust interface in between.  And completely different tech and electrical signaling than is used anywhere else in audio.  Can't say more.
Trouble is, the question then becomes what is at the "DAC end" of this new interface?  Not another USB or S/PDIF output--yuck--defeats the purpose.  I2S over LVDS?  Sure, but that limits popularity with DACs--and still the master clocking is not where you want it, which is close to the DAC chip.  What one wants is a DAC with this interface built in.  Which leads us to either enticing OEMs to embrace and license a new solution, and/or do our own DAC to demonstrate the tech.
 
But we have two other products to get into production first--and I have already said way too much. 

 
 
Have a great weekend everyone.
 
--Alex C.
 
P.S.  Took my wife and kids to see Paul McCartney at the Fresno Savemart Center (16K sold out seats) last night.  As our 14 year old son put it, "the sound was like ass", but our seats were great and Sir Paul puts on a great show!

Great post - I laugh when I look at my EMU 1616M from 11 years ago (still a great ADC for digitalizing LP's), it used it's own PCIe card linking to the main box by CAT5 ethernet!  Maybe we should have stayed down that road. 

 
Cheers!
http://www.soundonsound.com/sos/oct05/articles/emu1616m.htm

 

[abartels]

 
Comparison "Monster Dac" with USB - isolated XMOS - I2S interface VS Low Cost AK4495SEQ Dac with RPI3 - DAC+Pro - I2S interface:
 
Setup:
 
1- HP Probook i7 - 16Gb RAM - Server 2012 - 2x 250 GB SSD - Audiophile Optimizer - JRiver as player - connected with USB to "Monster Dac"
 
2- HP Probook i7 - 16Gb RAM - Server 2012 - 2x 250 GB SSD - Audiophile Optimizer - LMS server (Logitech Music Server)
 
Both JRiver and LMS server running simultaneous on HP Probook.
 
Both "Monster Dac" and Low cost AK4495SEQ Dac use LME49720 as output stage.
 
 
To begin with:
 
I bought the low cost Chinese AK4495SEQ last year, just to check out the performance and sound characteristics of AKM AK4495SEQ chip.
Since I wanted to build a new dac I could decide whether this chip had a sound character I liked and if I would base my new dac design on it.
 
This little baby performed very well, but in it's stock version it was not as airy as I had in mind, so I started to mod it with lots of Nichicon FG caps,
Cree diodes, different opamp as output stage and more. At the end it sounded fabulous, especially when I connected Diyinhk Isolated Xmos to it.
The design of this dac incorporates 1x psu for output stage, and 1x psu (with several regulators) for Digital section.
 
Since it's fabulous performance I decided to build my new dac, based on this AK4495SEQ chip.
 
I started designing and building and eventually finished my "Monster Dac"
 
First listening sessions I was amazed about it's airyness, bass control, big deep wide soundstage, lots of blackness around voices and instruments.
It was a huge step forward from the little baby.
 
So, where are we now, today?
 
I connected both dac's and played 2 albums on them, running simultaneous from HP Probook.
If I wanted to switch between dac's, I only had to switch between them on my pre-amp.
 
To begin with, they sound almost identical!!!
 
- Monster Dac has a little better bass control (and definition in it's lower frequencies), it's a tiny bit more airy, and it seems (not 100% sure) if it's more quiet, more black around voices and instruments.
 
- RPI has better vocal body and seems to have more 3D artifacts. It's bass output is more evident.
 
The two are VERY close to each other in this setup.
 
Conclusion:
 
Since low cost AK4495SEQ performed incredibly different when using isolated XMOS, it couldn't walk in Monster Dac's shadow, I have to conclude that:
 
RPI3 + DAC+ Pro > I2S output > based on PiCorePlayer and LMS server, is running circles around isolated xmos solution.
It's a VERY good performing solution!
 
AND
 
Keep in mind, RPI3 + DAC+ Pro together cost around $100. Software (LMS server and PiCorePlayer) are freeware.
 
Isolated XMOS interface cost about the same, but it needs:
 
1- Very stable OS and drivers
2- Optimized OS for Audio only
3- Powerful CPU
4- Needs to pay for Player software if one wants to remote control it with Graphical Interface on iPad or Android device
5- Needs optimized USB port on PC
6- Needs USB optimizing equipment like Intona / Regen / iFi purifiers and others (which will let you break the bank.......)
7- Needs high quality USB cable (which will let you break the bank.......)
 
Keep in mind, I also installed LMS server on my Daily work laptop, which I am typing this review on.
I connected PiCorePlayer to this LMS server (it finds it automatically in the network) and played music from it.
There was NO sonic difference between Setup-2 and this setup!!! So, NO need for expensive hardware to run LMS server!!!
 
Second thought:
 
You can use RPI3 + DAC+ Pro in every room in your house, if you have network connection. You can use same LMS server, and, even Analog output sound quality of RPI3 + DAC+ Pro is rather good!!!
I suppose that there are many expensive dac's out there which perform much less than this little baby! (if connected separate psu's the way I did).
 
Final thoughts:
 
I will have to modify my Monster Dac to replace it's I2S interface with RPI3 + DAC+ Pro combo, more on that next week or so 

 
Last word:
 
Since this RPI3 + DAC+ Pro combination is performing that well, it asks for the best of the best, which is Ian Canada's FIFO II solution with Isolation board and Dual XO II clock board.
I ordered them today, so, let's have some fun the next weeks 

 
 
 
Wish you all a prosperous week 

 
Alex

 
Update 14-04-2016:
 
Last night had a short listening session with RPI3 - DAC+ Pro > I2S > low cost AK4495SEQ.
 
At this time low cost AK4495SEQ with RPI3 + DAC+ Pro has BETTER performance than my "Monster Dac". It gained in every aspect!
Even SACD's which sounded some kind of thin before are sounding great now! Every recording has details in them I never heard before, and, this is with LOW COST dac.
Can't imagine how it will sound with my "Monster Dac". This weekend I will connect and review it!
 
In the meantime Ian's stuff arrived 

 

 

 

 

 

 
The Dual Clockboard II has two low cost TCXO's on it, they will be replaced with NDK's at first, and later on they will be replaced with OCXO's.
 
 
Let the fun begin