The new MSB Pro USB interface

Comments

Aug 10, 2015
43
12
8
#2
This is a very unique solution, but cannot understand how usb signal can be split into is2 or so called proISL.

It is usb signal deconstructed and reconstructed to besent through leds or laser.

The clocking signal would have been imbedded in the usb or as asyncronous usb , then there would not be a clock transmission involved here.

Interesting...
 

Taiko Audio

Active Member
Feb 10, 2017
434
220
43
The Netherlands
taikoaudio.com
#3
This is a very unique solution, but cannot understand how usb signal can be split into is2 or so called proISL.

It is usb signal deconstructed and reconstructed to besent through leds or laser.

The clocking signal would have been imbedded in the usb or as asyncronous usb , then there would not be a clock transmission involved here.

Interesting...
It seems the usb to i2s (or whatever proISL is) conversion takes place on the server side, the DAC seems to function as the masterclock for i2s.

This is the first real interface technology advance I’ve seen in years, and a very promising one.
 
Aug 10, 2015
43
12
8
#4
Thats why we dont know if ProISL, as with usb doesntnhave any clock signal transfer between server and dac.

Is this correct?

It would be much alike to ethernet transmission with a fibre converter in between.

In the proworld this could allow for usb connection between 2 devices for miles and miles.
 

Bodhi

Active Member
Apr 20, 2014
688
197
43
Melbourne, Australia
#5
Can someone explain to me what the difference is and associated benefits between this USB to ProISL interface vs a server or NAS streaming music over Cat 7/8 ethernet cable to a renderer/dac?
 

Taiko Audio

Active Member
Feb 10, 2017
434
220
43
The Netherlands
taikoaudio.com
#6
Can anyone explain to me what the difference is and associated benefits between this USB to ProISL interface vs a server or NAS streaming music over ethernet to a renderer/dac?
That is a very big difference, as ethernet, of all options, is the most processing intensive on the DAC side. This is the total opposite.
 

Bodhi

Active Member
Apr 20, 2014
688
197
43
Melbourne, Australia
#7
That is a very big difference, as ethernet, of all options, is the most processing intensive on the DAC side. This is the total opposite.
You'll have to bare with me as i'm not into computer audio or servers. So you're basically saying that Boulder's approach with its DLNA/UPnP compliant network player/dac (2120) is inferior to MSB's solution?
 

asiufy

Member Sponsor
Jul 8, 2011
3,157
247
63
San Diego, CA
almaaudio.com
#8
I've ordered some.
The idea is to fully isolate the server from the DAC, so you'd be able to use a normal PC, and the box would isolate it from the DAC, by converting and slaving the signal to the DAC's clock.
There's been many claims of devices that "isolate USB", regens, etc., so I'm a touch sceptic if this would pose a benefit over the Renderer. But we shall soon find out :)
 

Taiko Audio

Active Member
Feb 10, 2017
434
220
43
The Netherlands
taikoaudio.com
#9
You'll have to bare with me as i'm not into computer audio or servers. So you're basically saying that Boulder's approach with its DLNA/UPnP compliant network player/dac (2120) is inferior to MSB's solution?
MSB has something similar as Boulder is using called “The Renderer”.

They are different approaches.

There is a certain amount of processing necessary to convert and transport a digital file to something a DAC can convert to an analogue signal (i2s).

These different interfaces distribute processing loads differently between source and destination.

A few examples:
1) Direct I2S, all processing at source, virtually none at destination
2) AES/EBU, adds a little bit of processing at both source and destination relative to 1)
3) USB, adds a little bit more of processing at both sides relative to 2)
4) Ethernet, moves a significant part of the processing from source to destination

Now why we dont just simply use option 1) is it’s very sensitive to distances (as in each inch matters). So we use interfaces which are more robust to travel realistically useable interconnect lengths.

So this new MSB Pro USB interface is more like 1) and the already existing MSB Renderer is more like 4).
 

Taiko Audio

Active Member
Feb 10, 2017
434
220
43
The Netherlands
taikoaudio.com
#10
I've ordered some.
The idea is to fully isolate the server from the DAC, so you'd be able to use a normal PC, and the box would isolate it from the DAC, by converting and slaving the signal to the DAC's clock.
There's been many claims of devices that "isolate USB", regens, etc., so I'm a touch sceptic if this would pose a benefit over the Renderer. But we shall soon find out :)
Great, looking forward to your experiences!
 

Bodhi

Active Member
Apr 20, 2014
688
197
43
Melbourne, Australia
#11
Thanks for the explanation @Taiko Audio. From what you're describing, it sounds like the Boulder has sufficient processing power to handle the processing required to act as a render in an ethernet, without the need for an intermediary USB to ProISL interface.
 

Taiko Audio

Active Member
Feb 10, 2017
434
220
43
The Netherlands
taikoaudio.com
#12
Thanks for the explanation @Taiko Audio. From what you're describing, it sounds like the Boulder has sufficient processing power to handle the processing required to act as a render in an ethernet, without the need for an intermediary USB to ProISL interface.
Yes, so does the MSB renderer.

It’s not about “sufficient” processing power though. It’s about noise sensitivity, creation and isolation.
 

asiufy

Member Sponsor
Jul 8, 2011
3,157
247
63
San Diego, CA
almaaudio.com
#13
Thanks for the explanation @Taiko Audio. From what you're describing, it sounds like the Boulder has sufficient processing power to handle the processing required to act as a render in an ethernet, without the need for an intermediary USB to ProISL interface.
The Boulder does LESS than the MSB Renderer does, as it is not Roon compatible. It is in effect similar to a 5 year old Renderer v1.
Frankly, UPNP is old, archaic technology, that should be put to rest. Even Linn, that helped improve UPNP with their OpenHome protocol, has moved on to support Roon.
I'm sure Boulder is working on a new model, with Roon compatibility. But as it stands, it lags severely behind the MSB, and you can't say they both do the same thing.
The new MSB module has NOTHING to do with Ethernet, network, etc. It is a USB>ProISL converter. Remember way back, when no DAC had USB, and you had to buy those USB>SPDIF converters? This is like that, only it converts to a superior (to MSB DACs of course) format, one that carries signal and clock separately, and actually slaves the transport to the DAC's clock.
I only mentioned comparing it to an Ethernet/Renderer based solution as that's my current preference for sound quality, over any other solution, USB, AES, etc. There's a good chance this box might bring USB back to relevance, at least with MSB DACs.
 
Last edited:
Likes: CKKeung

Armsan

New Member
Jan 28, 2016
19
0
1
Portugal
#14
Just ordered another Pro ISL module to my Select and try this new development from MSB, the Pro USB.
It would be great, since the output of the server is USB, that it worked with the fantastic JPlay Femto.
Cheers.
 
May 30, 2010
15,086
477
83
Portugal
#15
MSB has something similar as Boulder is using called “The Renderer”.

They are different approaches.

There is a certain amount of processing necessary to convert and transport a digital file to something a DAC can convert to an analogue signal (i2s).

These different interfaces distribute processing loads differently between source and destination.

A few examples:
1) Direct I2S, all processing at source, virtually none at destination
2) AES/EBU, adds a little bit of processing at both source and destination relative to 1)
3) USB, adds a little bit more of processing at both sides relative to 2)
4) Ethernet, moves a significant part of the processing from source to destination

Now why we dont just simply use option 1) is it’s very sensitive to distances (as in each inch matters). So we use interfaces which are more robust to travel realistically useable interconnect lengths.

So this new MSB Pro USB interface is more like 1) and the already existing MSB Renderer is more like 4).
I think that there is lot more than distributing processes in this approaches - mainly 1 and 2 are synchronous ( 2 can use a master clock) , 3 and 4 are asynchronous.

Curiously this unit goes towards my view of the digital stream in audio - black boxes with input and output placed along the signal flow "purifying" the digital signal. :)
 

DMSB

New Member
Aug 10, 2017
5
22
3
#16
I’m from MSB Technology and I would like to share a few of the advantages the Pro USB has over other digital audio computer connections.

The Pro USB works as follows. The DACs clock is feed to the Pro USB through one fiber which controls the packet rate from the server, then the data is scrambled and streamed through a second fiber with added redundancy and error detection codes. The data is synchronized with the DACs clock to minimize noise and jitter. The received data is error corrected then checked for corruption by low noise single purpose decoding hardware and sent to be converted.

The Pro USB is different from other computer interfaces in subtle but important ways. Instead of offering partial isolation from the noise of the computer server like S/PDIF, copper Ethernet or USB the Pro ISLs single mode glass fiber cabling guarantees total, perfect isolation. Instead of only metering data packets using the DACs clock like an Ethernet streamer or USB interface the Pro USB streams the data synchronously with the DACs clock. Instead of assuming perfect data reception the Pro ISL first corrects any errors using redundant data coding and then checks for corruption using strong CRC error detection. Instead of settling for a high or medium amount of reception jitter the Pro ISL guarantees extremely low jitter. Instead of trying to isolate the DACs circuitry from the decoding noise of a USB receiver or Ethernet renderer the Pro ISL minimizes the amount of noise generated by data reception in the first place. Instead of having large data blocks that generate electrical in the audible band the Pro ISL streams and scrambles data to ensure that none of its electrical noise is in the audible range. The Pro ISL allows extremely long cables (more than 1km) without degradation. The Pro ISL was designed with future audio formats in mind so it has the ability to transfer 32bit/3MHz 2 channel PCM data or 12 channels of 32bit/768KHz PCM or 8 channels of 16x native DSD. Pro ISL cables do not generate or receive electrical interference.

The Pro USB allows most computing hardware (servers, renderers, etc...) to become first class transports. Downsides are few but include the need to correctly configure your software to ensure the Pro USB receives bit perfect data from the server. The power consumption of the Pro USB is 2W meaning it could be a significant power draw on battery powered equipment. The Pro ISL input is only available on MSB DACs currently. Compared with the MSB renderer it does not allow the computer to control the DACs volume.

As a side note, one really neat trick that can be done with the Pro ISL interface is optically splitting the data stream to multiple DACs. Only one DAC can be the clock master so the other DACs must be clock slaves, but all the DACs are completely isolated from each other. This can be really useful for multichannel data sources.
 

Taiko Audio

Active Member
Feb 10, 2017
434
220
43
The Netherlands
taikoaudio.com
#17
I’m from MSB Technology and I would like to share a few of the advantages the Pro USB has over other digital audio computer connections.
Thank you so much for sharing.

The Pro USB works as follows. The DACs clock is feed to the Pro USB through one fiber which controls the packet rate from the server, then the data is scrambled and streamed through a second fiber with added redundancy and error detection codes. The data is synchronized with the DACs clock to minimize noise and jitter. The received data is error corrected then checked for corruption by low noise single purpose decoding hardware and sent to be converted..
Am I right in assuming you're writing your own algorithms for this?

The Pro ISL input is only available on MSB DACs currently.
Is it possible to expand on what "currently" means? Do you have plans making this technology available to other DAC manufacturers on an OEM basis?
 
Likes: DMSB
May 30, 2010
15,086
477
83
Portugal
#18
(...) The Pro USB works as follows. The DACs clock is feed to the Pro USB through one fiber which controls the packet rate from the server, then the data is scrambled and streamed through a second fiber with added redundancy and error detection codes. The data is synchronized with the DACs clock to minimize noise and jitter. The received data is error corrected then checked for corruption by low noise single purpose decoding hardware and sent to be converted. (...)
Really nice to know some thechnical details of the ProUSB. Scrambling data in serial links seems a good move - it was done long ago by Ben Duncan (Audio Synthesis) using a proprietary SPDIF link - the Transcend N-Code. However I would not expect that such a link would need redundancy and detection codes - bit rates are tens of Mbit/s maximum - unless very long lengths are considered. Can I ask what is the typical corrected error rate?
 
Likes: Taiko Audio

DMSB

New Member
Aug 10, 2017
5
22
3
#19
Really nice to know some thechnical details of the ProUSB. Scrambling data in serial links seems a good move - it was done long ago by Ben Duncan (Audio Synthesis) using a proprietary SPDIF link - the Transcend N-Code. However I would not expect that such a link would need redundancy and detection codes - bit rates are tens of Mbit/s maximum - unless very long lengths are considered. Can I ask what is the typical corrected error rate?
Yes all hardware, software and algorithms are designed in house specifically for the Pro USB or Pro ISL.

Typical uncorrected error rate is virtually unmeasurable (a 24 hour test to count FEC events had zero events) but with a faulty cable, partially plugged in cable, dirt in the optical receiver, a 26db optical attenuator or failing optical module a raw uncorrected BER of up to about 10^-6 can be observed before the optical module reports loss of lock.The corrected BER is unmeasurable until a raw BER of about 10^-4 leads to some uncorrectable FEC events.
 

Taiko Audio

Active Member
Feb 10, 2017
434
220
43
The Netherlands
taikoaudio.com
#20
Yes all hardware, software and algorithms are designed in house specifically for the Pro USB or Pro ISL.

Typical uncorrected error rate is virtually unmeasurable (a 24 hour test to count FEC events had zero events) but with a faulty cable, partially plugged in cable, dirt in the optical receiver, a 26db optical attenuator or failing optical module a raw uncorrected BER of up to about 10^-6 can be observed before the optical module reports loss of lock.The corrected BER is unmeasurable until a raw BER of about 10^-4 leads to some uncorrectable FEC events.
Thank you, those bit error rates seem very good for a deliberately handicapped setup, better then normal ~50% utilized ethernet links if my memory serves me correct. I do not own a MSB DAC but this interface technology is actually making me consider buying one!
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. A place where audiophiles and audio companies discuss existing and new audio products, music servers, music streamers and computer audio, digital to audio convertors (DACS), turntables, phono stages, cartridges, reel to reel, speakers, headphones, tube amplifiers and solid state amplification. Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing