Jorge ,Hi
I present the project AYIA (in tribute to his "perpetrator", Smanz in Diyaudio forum)i. I have not made myself, just so enjoy it.
It is a dual mono dac cv181 tube.
Princiales components are:
- 2 Reference 9018 Sabre DACs (Twisted Pear Audio)
- USB EXAU2I
- Analog output with Psavane cv181 tubes, capacitors Mundorf MCAP gold silver oil, transformer Sowter
- Inputs 1 USB (exau2i), 2 Digital coaxial (75 Ohm WBT), 1 Toslink, 1 BNC.
- Outputs RCA analog outputs (WBT) Balanced (Furutech)
The sound is realistic, three-dimensional (in untreated room) and detailed. I've never heard anything like it, and is still running.
The output selector allows dsd for exadevices dsd player.
People who have heard talk sound much better than more expensive commercial dacs (and is not cheap)
Is everything perfect, I just need to get exadevices Audirvana direct mode
Some pictures:
Imageshack - img2078e.jpg
Imageshack - img2079n.jpg
Imageshack - p1010106il.jpg
Best, Jorge.
how do you switch digital input to BUFFALO between MUX and EXA ( SPDIF and USB )
Zoran
Is it not possible to make this software selectable via the driver CP?
Whatever the end product please count me in for at least one.
This is my wishlist:
-8-channel with pcm1704
-2MB+ fifo buffer (i think there is already in the current version but not sure)
-wsapi/KS driver support up to at least 384khz
USB High speed Line Isolation
I'd like to report two products, at up-steam and down-steam of exaU2I.
I hope exa065 will not delete this post as "Off-Topic" or "Opposing his idea".
1. Adnaco USB High-Speed optical isolator
In the early days of this exaU2I thread, RayCTech reported his attempt of isolating USB High-speed input and an improvement of SQ. However, his method seemed too proprietary and violating USB specification. No one might have reproduced it.
Besides, exa065's position was, according to my understanding, that the most effective isolation point was the I2S as he provided.
Recently, I could get a certain level SQ improvement by applying an optical isolation on USB High-speed 480 Mbps connection between PC and exaU2I with a USB spec compliant generic product.
The product is "Adnaco-S3B – USB 3.0 Over Fiber Optic Expansion System" .
I tried it with DSD256 and DSD512 stereo play and found no problems even on such a high speed data transfer.
2. ElectrArt's "DSD principle board" (DSD filter)
Based on my experiences, I think it's not easy to establish a stable play of DSD512 using such commercial DAC devices as ES9018.
I applied ElectrArt's board as a beta-tester and got a good result.
ElectrArt?Digital Audio??? ???????? 1bit delta sigma DAC (In Japanese)
His board simply filters DSD signals and output audible analog signals through FIR circuit and OP-amp circuit.
Here is a schematic.
I could play both DSD256 and DSD512 output from exaU2I without any harmful "unlock interruption" nor "digital burst noises specific to DSD".
In the uploaded picture,
A. at the upper-right, Adnaco USB extender board with a battery power
B. at the upper-center, exaU2I board with USB bus power from Adnaco board
C. at the upper-left, ElectrArt's "DSD Principle Board"
I hope this information is something useful to "DSD-orienting" exaU2I users.
I'd like to report two products, at up-steam and down-steam of exaU2I.
I hope exa065 will not delete this post as "Off-Topic" or "Opposing his idea".
1. Adnaco USB High-Speed optical isolator
In the early days of this exaU2I thread, RayCTech reported his attempt of isolating USB High-speed input and an improvement of SQ. However, his method seemed too proprietary and violating USB specification. No one might have reproduced it.
Besides, exa065's position was, according to my understanding, that the most effective isolation point was the I2S as he provided.
Recently, I could get a certain level SQ improvement by applying an optical isolation on USB High-speed 480 Mbps connection between PC and exaU2I with a USB spec compliant generic product.
The product is "Adnaco-S3B – USB 3.0 Over Fiber Optic Expansion System" .
I tried it with DSD256 and DSD512 stereo play and found no problems even on such a high speed data transfer.
2. ElectrArt's "DSD principle board" (DSD filter)
Based on my experiences, I think it's not easy to establish a stable play of DSD512 using such commercial DAC devices as ES9018.
I applied ElectrArt's board as a beta-tester and got a good result.
ElectrArt?Digital Audio??? ???????? 1bit delta sigma DAC (In Japanese)
His board simply filters DSD signals and output audible analog signals through FIR circuit and OP-amp circuit.
Here is a schematic.
I could play both DSD256 and DSD512 output from exaU2I without any harmful "unlock interruption" nor "digital burst noises specific to DSD".
In the uploaded picture,
A. at the upper-right, Adnaco USB extender board with a battery power
B. at the upper-center, exaU2I board with USB bus power from Adnaco board
C. at the upper-left, ElectrArt's "DSD Principle Board"
I hope this information is something useful to "DSD-orienting" exaU2I users.
Attachments
Last edited:
Hello Bunpei,
Thank you for posting about this project. The DSD512 board is exciting.
I see the benefits of the optical link when a longer cable is required. In terms of isolation it is redundant. There is no point to isolate twice, I am sure that that this will have no sonic effect.
Beside the cost, optical isolation is great. If you want to take this experiment further, you can remove the GMRs from the exaU2I. This will reduce the jitter. It will be still higher than the jitter of the new exaU2I V3.
Note to the exaU2I users with ES9018 DACs - don't bother removing the GMRs - in your case removing the GMRs won't make any difference. ES9018 takes care of the GMR jitter, and replacing the GMR isolation with optical isolation will not present a rational improvement.
Thank you for posting about this project. The DSD512 board is exciting.
I see the benefits of the optical link when a longer cable is required. In terms of isolation it is redundant. There is no point to isolate twice, I am sure that that this will have no sonic effect.
Beside the cost, optical isolation is great. If you want to take this experiment further, you can remove the GMRs from the exaU2I. This will reduce the jitter. It will be still higher than the jitter of the new exaU2I V3.
Note to the exaU2I users with ES9018 DACs - don't bother removing the GMRs - in your case removing the GMRs won't make any difference. ES9018 takes care of the GMR jitter, and replacing the GMR isolation with optical isolation will not present a rational improvement.
Last edited:
I can definitely commit to ordering the PCM1704 version (although the suggestion someone else posted to make this jumper-selectable makes sense too). Cost is not that much of an issue. I'm not sure whether I would need synchronous operation -- asynch clocks only seems fine, unless synchronous has noticeably lower delay. I would need multichannel -- this is a major attraction of the exa boards.
I hope the following improvements in the authorized DSD support of the new model of exaU2I.
- compatible with such players, HQPlayer and JRiver Media Center, through a genuine ASIO DSD framework
- free from any glitch or click noises at start play or stop play timings of DSD sources, and also protective for unintended DC output
( Not a player on PC but your board may need to insert DSD-zero signals at inter-tracks or paused or initial state.)
- compatible with such players, HQPlayer and JRiver Media Center, through a genuine ASIO DSD framework
- free from any glitch or click noises at start play or stop play timings of DSD sources, and also protective for unintended DC output
( Not a player on PC but your board may need to insert DSD-zero signals at inter-tracks or paused or initial state.)
New version
Hi, Should have answered you long ago. These are my preferences:
- 8 channel PCM.
- Specification and flexibility before cost.
- I2S output to external DAC's of own choice (PCM 1704, PCM17XX, ADxxxx, TDA1541 etc)
- Low CPU-load drivers (Yes they are today...)
What I did like from your original list with deleted items and some changes:
(quick and dity editing....)
Keep up the really good work you are doing.
(( Per
Hi, Should have answered you long ago. These are my preferences:
- 8 channel PCM.
- Specification and flexibility before cost.
- I2S output to external DAC's of own choice (PCM 1704, PCM17XX, ADxxxx, TDA1541 etc)
- Low CPU-load drivers (Yes they are today...)
What I did like from your original list with deleted items and some changes:
(quick and dity editing....)
- Just about 1ps jitter -> YES
- Two sets of clocks for synchronous and asynchronous operation with ES9018 -> YES (or other DAC)
- 8 channel DXD / 384 kHz PCM -> YES
- Individual LEDs for all PCM sampling rates
- Easy elimination of USB power -> YES
- Macro USB connector
- Header for connecting external USB connector
- Connectors for external power sources -> YES
- New ground isolation technology -> YES, if you know how to improve
- Micro BNC output connectors -> Would be nice
- No need to rewire the outputs for DSD and PCM (ES9018 only)
- Multichannel support for PCM1704. This option will rewire alternative firmware, and it may be a separate product.
- All clocks installed -> With option to supply my own clocks for synchronous operation.
- ES9018, PCM1704 or a third party DAC.
Keep up the really good work you are doing.
(( Per
For me THE key-priority: Linux-Suport like WaveIO gets automatically detected by MPDPUP. Once you have listened to a Linux-Setup, you find the Windows Stuff unnatural. And I tried a lot to make it work. Today I am a happy camper with Alix-Board with USB-SOTM-Card for USB. I even have your product, bought, but cant use it as it is currently not designed for Linux.
I am sorry, I don't follow you. If you use bit-perfect jitter free interface, you won't be able to tell apart Windows from Linux. When you use ASIO on Windows, you don't listen to Windows, you experience ASIO. There is no such thing as Windows Stuff. There are different implementations.
I understand that there are pros and cons to every platform, but I disagree with the notion that Windows is inferior to Mac or Linux when it comes to high-end audio.
Has anyone measured the noise floor of the USB-SOTM-Card?
Regards,
exa065
I understand that there are pros and cons to every platform, but I disagree with the notion that Windows is inferior to Mac or Linux when it comes to high-end audio.
Has anyone measured the noise floor of the USB-SOTM-Card?
Regards,
exa065
V3
Hi George,
Really excited about the new version!
Yes, would like to register my interest for purchasing one. Even the current one has become my preferred transport for the AckoDAC System
Mine is all 9018/12 DACs so hopefully you would maintain support for these.
The rest on the list is really good, native DSD especially. Possibly dedicated 2CH or 8CH versions if these make things more optimized?
OSX support?
Is the I2C a slave or master? Can the sample rates be read over I2C? I will be using the AKC12 Controller (master), so DAC Vol Cntrl is taken care of but sample rates from EXA2UI needed (for display purpose mainly).
Those U.FL coax, if you could leave it unpopulated would make it easier for my DAC that uses the smaller w.fl types. I could kludge a connection in place.
From the looks of it you have got re-clocking done after the isolation. If true this is great! For your current version (2.0) I am doing the re-clocking externally using a separate module after the isolators. So this could become redundant
Has the dimensions and mounting positions been finalised? Nice if you could keep it the same as the current one for easy upgrades. ETA? would like to start the integration process already.
Well done and thank you for the great efforts!
Hi George,
Really excited about the new version!
Yes, would like to register my interest for purchasing one. Even the current one has become my preferred transport for the AckoDAC System
Mine is all 9018/12 DACs so hopefully you would maintain support for these.
The rest on the list is really good, native DSD especially. Possibly dedicated 2CH or 8CH versions if these make things more optimized?
OSX support?
Is the I2C a slave or master? Can the sample rates be read over I2C? I will be using the AKC12 Controller (master), so DAC Vol Cntrl is taken care of but sample rates from EXA2UI needed (for display purpose mainly).
Those U.FL coax, if you could leave it unpopulated would make it easier for my DAC that uses the smaller w.fl types. I could kludge a connection in place.
From the looks of it you have got re-clocking done after the isolation. If true this is great! For your current version (2.0) I am doing the re-clocking externally using a separate module after the isolators. So this could become redundant
Has the dimensions and mounting positions been finalised? Nice if you could keep it the same as the current one for easy upgrades. ETA? would like to start the integration process already.
Well done and thank you for the great efforts!
Last edited:
I have used of course Asio. I used jriver, jplay, cmp, cplay etc. I tweak win7 and before vista. I build 5 different pcs with special components for best sound.
Bottom-line: Linux with Alix-Board and Sotm-usbcard with external linear psu. Most natural and transparent sound ever in my setup. Therefore, it would be great if the new version will support Linux...any chance ?
Regards