Quote from Mathias (RME designer):
"Its most spectacular feature is the addition of multi-client mixing. You can now use any number of programs of any sort (WDM, ASIO) to play back on the same channels. Note that as long as you use only one program per playback channel the driver works as before, including bit-perfect playback."
"Its most spectacular feature is the addition of multi-client mixing. You can now use any number of programs of any sort (WDM, ASIO) to play back on the same channels. Note that as long as you use only one program per playback channel the driver works as before, including bit-perfect playback."
Hallelujah!
It might also be a good idea to spend some time into marketing/documentation to gain some more customers(me for instance). I do not understand what sets the exa apart in "bit-perfectness". Therefore I do not "buy" that argument. A white paper could perhaps convince me and bring me on board. Undocumented claims tends to do the opposite.
regards,
Øyvin
I thinks we might have some crossed wires here.
I presumed exa was discussing bit-perfect in regard to his USB to I2S converter and comparing that to RMEs spdif output.
I now see he may have been referring to his new DAC product and comparing that to RMEs DACs. This is a whole different kettle of fish
However, beyond 16bit AFAIK, there are no DACs that can truly be called bit-perfect. If his new product achieves this then it is quite an achievement indeed - I look forward to evidence of this being the case.
I presumed exa was discussing bit-perfect in regard to his USB to I2S converter and comparing that to RMEs spdif output.
I now see he may have been referring to his new DAC product and comparing that to RMEs DACs. This is a whole different kettle of fish
However, beyond 16bit AFAIK, there are no DACs that can truly be called bit-perfect. If his new product achieves this then it is quite an achievement indeed - I look forward to evidence of this being the case.As I understand it the rme and many other sound devices will output bit perfect audio streams . I believe the issue is that there can be significant jitter and/or noise. Many people feel that using a USB or FireWire interface to the ps can yield far less of both. Many people feel that using a i2s interface between the USB and the DAC is optimal for similar reasons. Many people feel the b-III is an excellent DAC. Exa's u2i interface card is the only way I know of to bring all of these things together in m-ch.
"Quote:
Originally Posted by exa065
...It was discussed in previous posts in this thread, you can browse back and read it.
This thread is 106 pages(!) It would be useful if you took the time to write a white paper or something to explain this issue in detail and what exaU21 does to overcome it. "...
I believe the discussion he was referring to began 1 page back and is the same discussion you have joined in on.
"It might also be a good idea to spend some time into marketing/documentation"...
Likely everyone would agree more documentation would make all of our lives easier. This statement applies to this card, the buffalo-III and in fact the whole DIY experience of which both are a part. But then again I guess that's why it's called DIY ;-)
oyvine ,
Here is one of many threads on a site you may find helpful: computeraudiophile.com
Music player: what is there beyond Bit Perfect'ness? | Computer Audiophile
Here is one of many threads on a site you may find helpful: computeraudiophile.com
Music player: what is there beyond Bit Perfect'ness? | Computer Audiophile
This discussion has been going on for a while. Periodically I see claims such as in this posting:
http://www.diyaudio.com/forums/exad...hronous-usb-i2s-interface-96.html#post2846190
"I may say as much as when a 16 bit audio file are played and the so called bit-perfect chain to the DAC are following the standards - then you may get bit-perfect performance ONLY if you use a pure 16 bit DAC.."
I have also seen a posting where the actual C-library that is used was mentioned. I am not able to find it at the moment though. According to RayCtech, the exa065 and its drivers converts 16-bit data to 24-bit data in a "bit-perfect" way. I would really like to get some understanding on what this means and also understand why and how "all others" fail to do this correctly.
regards,
Øyvin Eikeland
its simple - you add 8 zeros as 8 LSBs to 16-bit word and you get 24-bit word with 16 MSBs untouched.
When you use for example digital filter or ASRC with 24-bit output length the input word is recalculated and 16 MSBs of 24-bit output word are not the same as 16 MSBs of input word
Connecting to a Micro Controller
I am trying to figure out the best way to input the bit rate into a micro controller. I know I would have to remove the LED's on the exaI2S. I have a lot of experience with the micro controller and basic digital IO. I am having trouble figuring out the best way to do this interface. The Xilinx chip that controls the LED's is new to me and I am having trouble getting what I need from their documentation. One idea I had was to use optical isolators. But they can draw 60ma. Is this too much for the Xilinx? I want to be careful with my investment in the exa board.
I am trying to figure out the best way to input the bit rate into a micro controller. I know I would have to remove the LED's on the exaI2S. I have a lot of experience with the micro controller and basic digital IO. I am having trouble figuring out the best way to do this interface. The Xilinx chip that controls the LED's is new to me and I am having trouble getting what I need from their documentation. One idea I had was to use optical isolators. But they can draw 60ma. Is this too much for the Xilinx? I want to be careful with my investment in the exa board.
Hi John,
You can replace the LEDs with optical isolators. The FPGA outputs are connected via 390 Ohms resistors, so you can draw about 5-6 mA.
George
Sure, I will try to put together a white paper. Most of the information is on our websites.
George,
Have you thought about changing the LED for 352 to light up for 384 instead of or in addition to 352. Jriver supports 384. I suspect more people would use 384 then 352. It works with my setup.
By the way, you were right about being patient with the Exa / BIII setup. It does take a while to get used to the sound. I am convinced that it is the best thing I have ever heard in my listening room. This with everything from CD's to hi res downloads. Every once in a while I get some weird things out of it. This usually happens on recording with a lot of reverb from the 80's and 90's. I think that this might be due to the fact that the recording engineer never heard the song through a state of the art DAC. They over did the reverb to get the effect they wanted. Only within the last few years could we hear it the way it was recorded verses they way they wanted it to sound.
John
Have you thought about changing the LED for 352 to light up for 384 instead of or in addition to 352. Jriver supports 384. I suspect more people would use 384 then 352. It works with my setup.
By the way, you were right about being patient with the Exa / BIII setup. It does take a while to get used to the sound. I am convinced that it is the best thing I have ever heard in my listening room. This with everything from CD's to hi res downloads. Every once in a while I get some weird things out of it. This usually happens on recording with a lot of reverb from the 80's and 90's. I think that this might be due to the fact that the recording engineer never heard the song through a state of the art DAC. They over did the reverb to get the effect they wanted. Only within the last few years could we hear it the way it was recorded verses they way they wanted it to sound.
John
Hi John,
Thank you for posting back about your listening experience. All of us are so used to the sound of the mainstream audio equipment, that it takes a while to appreciate the sound of exaU2I - the flood of detail and the lack of makeup and extra spices. I remember the reaction of a friend of mine, when he listened for the first time to my system. I played a piece from "Brothers in Arms" - 24bit/96kHz 5.1, and his comment was: "This is not the voice of Mark Knopfler"
To answer your question - there is a unique LED combination for 384 kHz - The FIFO light is On, and all frequency indicator lights are Off.
-Best,
George
Thank you for posting back about your listening experience. All of us are so used to the sound of the mainstream audio equipment, that it takes a while to appreciate the sound of exaU2I - the flood of detail and the lack of makeup and extra spices. I remember the reaction of a friend of mine, when he listened for the first time to my system. I played a piece from "Brothers in Arms" - 24bit/96kHz 5.1, and his comment was: "This is not the voice of Mark Knopfler"
To answer your question - there is a unique LED combination for 384 kHz - The FIFO light is On, and all frequency indicator lights are Off.
-Best,
George
DSD256 and DSD512 play with TPA Buffalo II
I could play DSD256 and DSD512 audio sources with the combination of ExaU2I and TPA Buffalo II. It's amazing!
As for the DSD256 source, I converted a sample file of WSD format available on "1 bit consortium" web site. In the case of DSD512, I crafted it from 2L DXD sample file. It's not an originally recorded one.
In the both cases, Korg AudioGate program was useful for WSD->DFF and WAV->DFF conversion.
Bunpei
I could play DSD256 and DSD512 audio sources with the combination of ExaU2I and TPA Buffalo II. It's amazing!
As for the DSD256 source, I converted a sample file of WSD format available on "1 bit consortium" web site. In the case of DSD512, I crafted it from 2L DXD sample file. It's not an originally recorded one.
In the both cases, Korg AudioGate program was useful for WSD->DFF and WAV->DFF conversion.
Bunpei
Correct me if Im wrong, but the exaU2I dies not have either a clock or a power supply. Both are within the the B-II/B-III (the DAC). the exa card is the interface between the computer and the DAC.
Or, are you asking what you could do to upgrade the DAC?