[Shahrose]

Quote:

 
I'll be interested in reading your review slim.

  Where will you post it?
 

 
x2.
 

 

[haloxt]

Loss of bit-perfect is pretty common in computers because of software quirks or hidden sound processing. I'd appreciate if people described what is necessary to get the tenor usb chip to be bit-perfect, not because I think it's essential, but just to find out if the tenor chip has any compatibility issues with different bit-perfect software.

 

[erin]

 
Quote:

Loss of bit-perfect is pretty common in computers because of software quirks or hidden sound processing.

I'm under the impression that if you use ASIO or KS or Direct out. that you will get bit perfect audio. ( however when I try these different output types - they all sound slightly different) So, this must mean that the interaction between the plugin and the player software must create different processing spikes and current draw, I guess resulting in RF spikes from the PS or Mobo, transferred down the USB cable to the converter. thus increasing  or decreasing  jitter. Resulting in the different sound.
 
I guess  this is why people are reporting positive results with the USB isolators.
 
From my own experience, reducing the number of background processes running on the computer can have a large impact on the sound quality. I use CMP + Cplay and find that this software is the best that I have heard. Others disagree, or don't want to use it because of its relative inconvenience. I now believe that all parts of your computer contribute to the overall sound (power supply, mobo, processor, attached graphics or sound cards etc.) , and so the subjective results on one computer with one converter, may be different on another computer with the same converter - which would explain why people report different results. Oh yea, and then there is the different DAC's that people use when reviewing.  It makes it very challenging  to know  whats best for your system. Only your own ears can tell you. Unfortunately its always X hundred dollars to try the newest toy. Hopefully the pockets are deep....
 
There is a lot of information around on various forums about getting the best sound out of your computer.
 
Cant comment on the tenor chip specifically because I don't own one.
 
Looking forward to some reviews of this Audio Gd converter big time

 

[regal]

Slim,  A DSP (Digital Signal Processing) chip in a transport is a red flag!  If we don't show bit-perfectness,  hell that thing could be applying EQ or anything else.  You cannot ever review a transport without first proving they are bitperfect (at 44.1k) this gives a level playing field.
I think you, as a budding reviewer should embrace this philosophy if you want to maintain credibility.
 
(It also helps those of us who need bitperfect for HDCD and DTS pass thru.)
 
Quote:

Hi upstateguy,
 
I will probably post my review in a separate thread. My "usb to spdif shoot out thread" has been closed for an unknown reason so I might have to start another one.

 

Hi regal,
 
In my personal experience, I have found that bit perfectness (in 44.1 and 48 frequencies) is not an issue with 24 bit capable transports. So far, all the 24 bit capable transports I have tried were able to pass HDCD data unharmed to the DAC.
 
The DSP-3 can do upsampling internally, so in that case it will definitely not be bit perfect.
I don't know about the dsp-3, but the dsp-1 works in 32 bits internally if my memory is correct.
In that case, any processing/rounding errors shoud fall under the less than 20 bit real world resolution of most DACs. Would that be audible? Hard to say without listening.
 
My main issue with the audio-gd digital interface and similar usb to spdif converters that use the Tenor chip (Teralink X2, Firestone Bravo...) is that they can't pass 88.2 without resampling to 96. (Please correct me if the audio-gd digital interface can pass 88.2)
I can understand why some people would want to resample from 44.1 to 96 (or 192). But doing 88.2 to 96 conversion (by the drivers or by software) is just a waste of processing power and a additional step where data could be harmed.

 

[slim.a]

Quote:

Would it be possible to repost your reviews from the "usb to spdif shoot out thread" in the beginning of the new thread, so that information isn't lost? 
 
USG

Hi USG,
 
I just started another "usb to spdif shoot out thread" here: http://www.head-fi.org/forum/thread/503472/the-usb-to-spdif-converters-shoot-out-thread
I reposted what I could find from the old thread (I didn't keep a copy of everything).

I will post my review of the audio-gd digital on that thread whenever it is sent to me.

 

[slim.a]

Quote:

Slim,  A DSP (Digital Signal Processing) chip in a transport is a red flag!  If we don't show bit-perfectness,  hell that thing could be applying EQ or anything else.  You cannot ever review a transport without first proving they are bitperfect (at 44.1k) this gives a level playing field.
I think you, as a budding reviewer should embrace this philosophy if you want to maintain credibility.
 
(It also helps those of us who need bitperfect for HDCD and DTS pass thru.)
 

 

 
I understand that is important to test bit perfectness especially for those who need HDCD or DTS pass thru. I don't have a HDCD capable DAC anymore but I will try to test bit perfectness on a friend's HT receiver.
 
You are right, the DSP is powerful enough that it could be doing a number of things. However, one could ask: if it doesn't raise the voulume (or compress the sound) and if any processing improves the sound, wouldn't it be a good thing for non HDCD data? Aren't all non-NOS DACs messing with the data anyway?

My main concern (besides the 88.2 capability) is the jitter reduction capability of the DSP. Recently, I found that, in my system, the sound improved a lot by bypassing the PLL feature of the DSP-1. My guess is that the "clean" clock provided by the battery powered Hiface I am using as a transport was compromised by the PLL of the DSP-1. This is of course only speculation to try to explain what I heard. I might be totally wrong about the explanation but I heard an improvement by bypassing the PLL feature on the DSP-1 (Jumper 1).

From that experience, I doubt that the reclocked digital stream by the DSP-3 can better a battery powered hiface. But since there are many other things at play (output stage, impedance,...), only direct listening experience will tell.
 

 

[haloxt]

Erin, I agree with your explanation, but still I've read people say they fail bit-perfect test either due to hidden DSP and resampling, or because they didn't choose the right settings. I'd like to know if DI uses any hidden EQ, but I just want people to make sure they try all the settings and a few different software on the computer if they can't manage to pass bit-perfect tests, this rules out the computer and lets us know if bit-perfect is lost (or retained) in the DI.
 
If this device actually uses software EQ "to improve sound quality" that'd be pretty surprising. But it wouldn't be the first, two popular ones for audiophiles are xxhighend and izotope ozone.

 

[Currawong]

Quote:

 
I understand that is important to test bit perfectness especially for those who need HDCD or DTS pass thru. I don't have a HDCD capable DAC anymore but I will try to test bit perfectness on a friend's HT receiver.
 
You are right, the DSP is powerful enough that it could be doing a number of things. However, one could ask: if it doesn't raise the voulume (or compress the sound) and if any processing improves the sound, wouldn't it be a good thing for non HDCD data? Aren't all non-NOS DACs messing with the data anyway?

My main concern (besides the 88.2 capability) is the jitter reduction capability of the DSP. Recently, I found that, in my system, the sound improved a lot by bypassing the PLL feature of the DSP-1. My guess is that the "clean" clock provided by the battery powered Hiface I am using as a transport was compromised by the PLL of the DSP-1. This is of course only speculation to try to explain what I heard. I might be totally wrong about the explanation but I heard an improvement by bypassing the PLL feature on the DSP-1 (Jumper 1).

From that experience, I doubt that the reclocked digital stream by the DSP-3 can better a battery powered hiface. But since there are many other things at play (output stage, impedance,...), only direct listening experience will tell.
 

Non-NOS DACs use oversampling, so there will always be some kind of pre-processing of the digital data before it reaches the actual DAC.
 
Your PLL comment is interesting.  I might try switching off the PLL of the DSP-1 in my Ref 1 since I'm using the Ref 3, to see if it makes any difference.  

 

[regal]

The Danger with hidden EQ is it can give you a wow affect and sound great on some music but poor on others,  you never want eq applied that you don't know of,  you want that to be done under your control or as most prefer just the mastering engineer himself (when the CD is made.)
 
Oversampling DAC's just oversample and dither thats it,  they don't change the data which is called mastering.
 
 
A transport that isn't bit perfect at 16/44.1  isn't a transport its a dumbed down DAW (Digital Audio Workstation.)
 
Quote:

Erin, I agree with your explanation, but still I've read people say they fail bit-perfect test either due to hidden DSP and resampling

 

[leeperry]

 

Non-NOS DACs use oversampling, so there will always be some kind of pre-processing of the digital data before it reaches the actual DAC.

Oversampling is done within the DAC chip at the conversion stage, anything else is called upsampling...and should be avoided in many ppl's opinions(including mine): http://www.audioholics.com/education/audio-formats-technology/upsampling-vs-oversampling-for-digital-audio/upsampling-vs-oversampling-for-digital-audio-page-2

The effects of upsampling are no doubt overstated. By carefully designing the sampler, ADC, digital processing path, and oversampling DAC, the upsampling and asynchronous rate transfer can, in my opinion, be avoided.

 
The only thing upsampling does is feed interpolated bogus data to the DAC oversampling algorithm..and yes, we can measure the THD/THD+N increase very easily. Some ppl might still like it, but there is no question that it's coloring the sound.

 

[ninjikiran]

for me I want it to be as transparent as possible, as in being what a transport is getting something from point A to B with little to no interference.  Obviously nothing in this world is "perfect", the only place any noticeable changes(unless your golden eared) should be when it comes in contact with any kind of DAC circuitry where its supposed to get converted and changed.
 
Of course without some kind of direct to driver mode(WASAPI/Real ASIO) you are reliant on Windows to mess up your sound in any way possible via software.  As for the players I have never heard a difference between say Foobar and Cplay, one player I did try seemed very very very slightly louder.  Enough to color your opinion if not careful... 
 
Everything else that people feel mess with audio, timing, jitter ect ect list goes on is not something you can really prevent(regardless of whether it effects you or not) besides buying overly expensive items.  The DSP chip in this unit in this device scares me as well, actually most high end audio products scare me because they all seem to want to paint a picture for things that should for all intents and purposes be completely transparent.  I mean if your adding warmth to 1's and 0's theres an obvious flaw or some kind of purposeful gimmick built into the device, or placebo(most likely the case).
 
This sound card I am using is a perfect transport but it is a sound card made for other goals.  Having a simple USB solution would save time and free the card up to be used for its real purpose

 

[Trapper32]

Hmmmm  Doesn't everything ( caps, wire, cables, psu, etc etc ) in the signal path affect EQ???   Isn't that why some products can be varied to sound warm, thin, detailed,  etc etc.  Shouldn't it be about the Music???  Subjective??  Already ppl are making judgements about how it will sound based  on pure speculation on what they think it will sound like.  What about just letting our ears be the judge.  If  product A is technically superior and passes a  bit perfect signal but  doesn't sound as good as B to your ears, which one will you listen to??  I think some will actually listen to A because its technically superior, even if its sonic capability is inferior.  Count me on the side of the best sounding unit, whichever technology is used.  

 

[slim.a]

Quote:

Non-NOS DACs use oversampling, so there will always be some kind of pre-processing of the digital data before it reaches the actual DAC.
 
Your PLL comment is interesting.  I might try switching off the PLL of the DSP-1 in my Ref 1 since I'm using the Ref 3, to see if it makes any difference.  

I am intersted in having your feedback on the PLL setting.

 
Quote:

The Danger with hidden EQ is it can give you a wow affect and sound great on some music but poor on others,  you never want eq applied that you don't know of,  you want that to be done under your control or as most prefer just the mastering engineer himself (when the CD is made.)
 
Oversampling DAC's just oversample and dither thats it,  they don't change the data which is called mastering.
 
 
A transport that isn't bit perfect at 16/44.1  isn't a transport its a dumbed down DAW (Digital Audio Workstation.)
 

I agree that a hidden EQ would be bad. It is also something that might show on a RMAA test I believe.
 
Regarding oversampling, I don't believe it is done the same way by all the algorithms. If you take Ayre's Minimum phase alogrithm for example (see here: http://www.ayre.com/pdf/Ayre_MP_White_Paper.pdf) you will see that they alter in purpose the impulse response in order to avoid the pre-ringing (I am simplifying things).
In that case, would the minimal phase processing be called mastering or is it still "regular" oversampling?
 
What I am trying to say is that I am usually a pragmatic person. While I also think that in most cases "zero processing" at the source/transport is the best way to go, I am not against processing if it can improve ALL my CDs.
For example, I don't always use upsampling but if Kingwa developped some smart and very effective upsampling algorithm that improves the sound quality of all my files, I would rather have the upsampling than having a "bit perfect" path with a lesser sound quality.
 
In other words, it is good to know the technical stuff to avoid buying the wrong components. However, at the end of the day, what matter most (to me at least) is how good it sounds. This is of course just my personal opinion.
 
 
Quote:

 

Oversampling is done within the DAC chip at the conversion stage, anything else is called upsampling...and should be avoided in many ppl's opinions(including mine): http://www.audioholics.com/education/audio-formats-technology/upsampling-vs-oversampling-for-digital-audio/upsampling-vs-oversampling-for-digital-audio-page-2

In the case of R2R DACs such the PCM1704, there is no built in digital filter. The digital filter has to be external and we still use the term oversampling: you can have a look at the DF1704 digital filter datasheet (see here: http://focus.ti.com/lit/ds/symlink/df1704.pdf). If you think that TI should use the word upsampling instead of oversampling, you might want to send an email...I wonder what they would respond

 

[slim.a]

Quote:

Hmmmm  Doesn't everything ( caps, wire, cables, psu, etc etc ) in the signal path affect EQ???   Isn't that why some products can be varied to sound warm, thin, detailed,  etc etc.  Shouldn't it be about the Music???  Subjective??  Already ppl are making judgements about how it will sound based  on pure speculation on what they think it will sound like.  What about just letting our ears be the judge.  If  product A is technically superior and passes a  bit perfect signal but  doesn't sound as good as B to your ears, which one will you listen to??  I think some will actually listen to A because its technically superior, even if its sonic capability is inferior.  Count me on the side of the best sounding unit, whichever technology is used.  

 
Trapper32,
 
This is exactly what I was trying to express: if the DSP-3 can help us get a better subjective listening exprience, even if it means that it will mess with the bit-perfectness, why not?
 

 

[leeperry]

 

In the case of R2R DACs such the PCM1704, there is no built-in digital filter. The digital filter has to be external and we still use the term oversampling: you can have a look at the DF1704 digital filter datasheet (see here: http://focus.ti.com/lit/ds/symlink/df1704.pdf). If you think that TI should use the word upsampling instead of oversampling, you might want to send an email...I wonder what they would respond

Aren't we talking about obsolete parts here?

The best audiopath is the shortest, and delta-sigma DAC's such as the AK4396 and its newer revisions do it all at once in the same chip, providing "the performance and linearity of a delta-sigma device with the noise performance of an R-2R part"...you seem to care about jitter, yet you're OK to use an external oversampling chip? I sense contradiction in your faith, Jedi.
 
Just as a reminder: 

 
Richard Kulavik of AKM Semiconductors explained it this way : "This DAC is a large departure from other delta-sigma DACs designed by us and others like BurrBrown, Analog Devices and Cirrus Logic. The AK4396 is an entirely new modulator, pioneered and patented by AKM. It achieves something unique. In the past, many of the old Phillips and BurrBrown parts were R-2R* based products. These older products were looked upon as some of the best. One of the reasons was high frequency noise. In older R-2R parts, HF noise was not present. In all delta-sigma parts prior to the AK4396, everyone has fought HF noise caused from the delta-sigma modulator with the insertion of large filters and other parts to attempt to solve a problem created by the delta-sigma design. The AK4396 today effectively does not suffer any modulator-induced HF noise and is over 60dB better than the nearest Cirrus and BB devices. All of this HF noise can cause many audible artifacts downstream. That is the 'miracle' we believe is making the difference today. This part gives you the performance and linearity of a delta-sigma device with the noise performance of an R-2R part, something that was never previously available."

 
But if you like to have a 2ft long audiopath full of resistors/caps/transistors and several clock references not being synced together, your choice