Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: USB transient noise with external DAC (Read 26685 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: USB transient noise with external DAC

Reply #25
I don't understand why people are suggesting workarounds for what is obviously broken hardware. 

When it comes to power on - power off transients the world is full of a lot of what some call defective gear.

Audio professionals are trained how to work out power sequences that avoid causing problems, It is possible to manage these things regardless of how the gear performs.  For example, one puts the power amp on a power source that adds a turn-on delay.

Such as: http://www.ebay.com/itm/AC-110V-H3Y-2-Power-On-Time-Relay-Delay-Timer-0-60-Second-DPDT-Base-Socket-NEW-/141186532701?hash=item20df5f855d:g:2pgAAOxyXDhSg0Jl

Quote
The device is generating a dangerous result during normal operation of a PC.  This is not his problem, just RMA it. 


 PC's are tools and system components, not one stop final solutions.  The problem of power amps that pass voltage surges from downstream equipment is as old as the hills, and so are the solutions.

Re: USB transient noise with external DAC

Reply #26
I sent it back yesterday.  At a minimum I will lose $25 for two way shipping.  There is a restocking fee of $22.50, but it will be waived if the manufacturer agrees the unit was defective, but they are already trying to game that.

Sounds like a good company to buy products from.

Even without an external DAC my NUC makes a few barely audible thuds when rebooting.  That is when plugged directly into a 200 w power amp with no input attenuation.  Using USB with the Modi turned elevated the noise to speaker blowing levels.  The tech at Schlitt ventured the lack of an input control was the problem, but when I put one in line and set it for my normal listening level (-15 db) the transient noises were still very loud.  It was something line bouncing a basketball on the floor of an indoor court.  That can't be normal, and if it was everyone would be complaining about it, but I can't find the complaints, so it must not be normal.

When you set a clock on a DAC you usually mute the line out amplifier so that all the glitching while the circuit synchronizes is blocked.  It sounds like that isn't happened for whatever reason. 

I hope the reason was a defective component and not some excuse like I should have used a volume control. 

Re: USB transient noise with external DAC

Reply #27
I don't understand why people are suggesting workarounds for what is obviously broken hardware. 

When it comes to power on - power off transients the world is full of a lot of what some call defective gear.

Perhaps, but in this instance the device is clearly broken.  Either a component has failed, or it has been incompetently designed.  In either case, the correct solution is to return it and ask for a working device. 

Re: USB transient noise with external DAC

Reply #28
I don't understand why people are suggesting workarounds for what is obviously broken hardware. 

When it comes to power on - power off transients the world is full of a lot of what some call defective gear.

Perhaps, but in this instance the device is clearly broken.

Based on what independent  authority?

Here, we have seen a solution that costs about $10 and addresses the problem of turn-on transients from all present and any future source.  Just delay turning on the power amp until the other components have settled down.  My system happens to have an receiver that does this right out of the box, but with less well-desgined and legacy equipment, the $10 solution works.

Re: USB transient noise with external DAC

Reply #29
Sarartoga, right now I don't know if I will get cash or if they will offer to replace the unit.  Remember the tech said they don't like plugging into a power amp directly.  I am listening to some 24/48 tracks right now via the ALC283 and it sounds good.  I did a some speaker placement  adjustments and updated the Realtek drivers.  Let's say there is no urgent reason to do more, but it just depends on the opportunity.

 

Re: USB transient noise with external DAC

Reply #30
I don't understand why people are suggesting workarounds for what is obviously broken hardware. 

When it comes to power on - power off transients the world is full of a lot of what some call defective gear.

Perhaps, but in this instance the device is clearly broken.

Based on what independent  authority?

Let's please stay on topic.  This side-conversation isn't helping the OP.


Re: USB transient noise with external DAC

Reply #32
192k isn't a major issue, but toslink would not do 88.2 because it depends on the Realtek chip which does not do 88.2.  The Realtek's implementation of 44.1 is  odd too.  It does a hardware resample to 48 but outputs 3900 blank frames per second.   I would have liked to use the Toslink input for a TV also.

Bottom line is I want something that works, not workarounds on a brand new piece of consumer audio gear.

Probably, what I need is something with variable output line out jacks.  Some that come to mind:
Audioengine D1
Origin Micca (reputedly has driver problems with win 10.)
IFI Nano (has a battery which can go bad.)

That's all I know of in the sub $200 price range.  Did I miss anything?

I don't expect to completely get rid of transient noises, but I expect to get it down to a dull thud.

Re: USB transient noise with external DAC

Reply #33
192k isn't a major issue, but toslink would not do 88.2 because it depends on the Realtek chip which does not do 88.2.  The Realtek's implementation of 44.1 is  odd too.  It does a hardware resample to 48 but outputs 3900 blank frames per second.  I would have liked to use the Toslink input for a TV also.
Did you actually hear that 3900 blank frames or did you just misinterpret Realtek's documentation? Did you use another audio interface to record the toslink output of your Realtek and verified it (bit-perfectness at 44.1k)? I replied to your similar claim several months ago.
 
https://hydrogenaud.io/index.php/topic,111980.msg922902.html#msg922902


Re: USB transient noise with external DAC

Reply #34
192k isn't a major issue, but toslink would not do 88.2 because it depends on the Realtek chip which does not do 88.2.  The Realtek's implementation of 44.1 is  odd too.  It does a hardware resample to 48 but outputs 3900 blank frames per second.  I would have liked to use the Toslink input for a TV also.
Did you actually hear that 3900 blank frames or did you just misinterpret Realtek's documentation? Did you use another audio interface to record the toslink output of your Realtek and verified it (bit-perfectness at 44.1k)? I replied to your similar claim several months ago.
 
https://hydrogenaud.io/index.php/topic,111980.msg922902.html#msg922902



Rather than speculating on whether I understand Realtek's documentation why don't you do a search for: realtek 147
  Then you read the documentation and report back here instead of jumping all over me.

You will not find documentation for the alc283, but doeumentation for similar chips explain the system.  I don't have the facilities to do the test you suggest.  However, Realtek sound chips are in most PC's.  It shouldn't take you more than 10 minutes to find one and do the test you described.  That would be more useful than jumping all over me.

You seem to have missed the part about a consumer buying something and expecting it to work as advertised. Like Arnold, you have jumped off into a different world.  If I want a Toslink input for another source then it doesn't matter if Toslink is just as good as USB.  If losing 88.2 doesn't matter (the much acclaimed ODAC does not do 88.2 native) then I could get a $30 UCA202 and run everything at 48k.  Perhaps the whole thing doesn't matter and external DAC's are no better than internal Realtek sound unless the motherboard was designed badly.

Does anyone around here have recommendations for a different USB DAC, preferably with variable line outputs, that has been successfully used without wildly loud transient noise when powering up the USB, or am I doomed to face a series of arcane technical arguments?

Re: USB transient noise with external DAC

Reply #35
Since my test is based on ALC892 so ALC283 may work differently, but if you see my posts and that null test video carefully, my ALC892 does not perform any resampling at 44.1k in exclusive mode. Null test is the most effective way to verify bit-perfectness in digital transfer.

So my standpoint is: Do ALL Realtek HD audio chips resample at 44.1k? No. How about ALC283? I don't know.

If you are still so eager to find another standalone DAC, you may read the thread below and see the possible risks of purchasing another USB DAC if it is not competently designed or built.

https://hydrogenaud.io/index.php/topic,111411.msg918545.html#msg918545

Re: USB transient noise with external DAC

Reply #36
Rather than speculating on whether I understand Realtek's documentation why don't you do a search for: realtek 147

Been there, done that. Got approximately 1.5 million hits, Looked at the first few 100. Nothing seemed to fit.  Ever hear of URL's? ;-)


Re: USB transient noise with external DAC

Reply #37
Rather than speculating on whether I understand Realtek's documentation why don't you do a search for: realtek 147

Been there, done that. Got approximately 1.5 million hits, Looked at the first few 100. Nothing seemed to fit.  Ever hear of URL's? ;-)


Read from section 7.2.5. I read that documentation several months ago since his last reply and that's why I said he misinterpreted it.

Re: USB transient noise with external DAC

Reply #38
Since my test is based on ALC892 so ALC283 may work differently, but if you see my posts and that null test video carefully, my ALC892 does not perform any resampling at 44.1k in exclusive mode. Null test is the most effective way to verify bit-perfectness in digital transfer.

So my standpoint is: Do ALL Realtek HD audio chips resample at 44.1k? No. How about ALC283? I don't know.

If you are still so eager to find another standalone DAC, you may read the thread below and see the possible risks of purchasing another USB DAC if it is not competently designed or built.

https://hydrogenaud.io/index.php/topic,111411.msg918545.html#msg918545

I expect the 892 and 283 work the same way because I have looked at the documentation for several realtek chips and all do it the same way.

Perhaps the word "resample" is not the right one.  The Realtek chip does not create 3900 new samples per second, it skips 3900 frames per second.  In a second 44100 samples come out as sound.  I don't actually know what happens to the 3900 skipped frames, but the time they take up has to be accounted for.  What I suspect is the 44100 samples are broadcast as analogue with pauses every 11 or 12 samples.  I wish someone would explain what is really happening.  Since no new samples are created a bit perfect recording appears possible, but what happens to the time?  Does it become jitter?

Believe me, I am aware of USB related problems.  Aside from transient noise problem I am having there is often a power issue on USB powered DAC's.  DAC's utilizing USB Class 2 run into driver problems on Windows.  If anything, I am cooling my jets on the entire concept.

Re: USB transient noise with external DAC

Reply #39
Does it become jitter?
Jitter can be recognized as raised noise floor and increased sidebands from test tone in analog output. I also measured the analog signal in 44.1 and 48k mode in the post below and attached the result. ALC887 is typo, it is ALC892. Did you recognize any major differences in my test result in 44.1k vs 48k?

https://hydrogenaud.io/index.php/topic,111980.msg922878.html#msg922878

Re: USB transient noise with external DAC

Reply #40
Rather than speculating on whether I understand Realtek's documentation why don't you do a search for: realtek 147

Been there, done that. Got approximately 1.5 million hits, Looked at the first few 100. Nothing seemed to fit.  Ever hear of URL's? ;-)


Read from section 7.2.5. I read that documentation several months ago since his last reply and that's why I said he misinterpreted it.

I just read section 7.2.4.  It is difficult for me to understand.  It appears to deal with multiple streams, not 44.1.  The next section clearly makes reference to empty frames in the stream.  I am not qualified to give an expert opinion about what happens to the empty frames or if there is any possibility of sound quality degradation.  However, processing 44.1 khz signals at 44.1 khz makes a lot more sense to my simple mind.

Re: USB transient noise with external DAC

Reply #41
Does it become jitter?
Jitter can be recognized as raised noise floor and increased sidebands from test tone in analog output. I also measured the analog signal in 44.1 and 48k mode in the post below and attached the result. ALC887 is typo, it is ALC892. Did you recognize any major differences in my test result in 44.1k vs 48k?

https://hydrogenaud.io/index.php/topic,111980.msg922878.html#msg922878


I remember that thread.  Sorry, but it is difficult for me to comprehend your testing.

Oftentimes I think the glowing reviews folks give external DAC's is because the vast majority of those writing the reviews listen to certain headphones which the on board headphone amps have trouble driving.

One of the things I have been experimenting with is using Sox mod 2 to resample 44.1 to 48.  I can't say if there is a real improvement here.  Fortunately, it cost's nothing to try.  I might try uninstalling the Realtek driver and see what happens.

Re: USB transient noise with external DAC

Reply #42
Quote from: bennetng

Read from section 7.2.5. I read that documentation several months ago since his last reply and that's why I said he misinterpreted it.

The PDF linked won't open with the current Adobe reader, but I found a similar document at:  http://realtek.info/pdf/alc882.pdf.

This cheap little Intel box seems kinda interesting and I'm in the process of negotiating to have one at my disposal to study in a more detailed way.

My interpretation is that audio data is communicated in 500 sample blocks with a bit clock rate of 24 MHz. A standard data block based on a 48 KHz audio channel occupies one block. If the channel's sampling rate is > 48 KHz then it occupies more blocks as required, and if its sampling rate greater than 48 KHz but not a perfect multiple of 48 KHz additional blocks  used,  padded with zeros as required.  This is also done with the one required block for data with sample rates < 48KHz.

This is one of several common and widely used standard practices for communicating digital data, all of which are  routinely executed in numeroous products without any audible degradation of the signal.

The most important consideration is that the logic on the chip is capable of operating at speeds well in excess of those actually required, so that work involved with ignoring the zero padding causes no unmanageable delays. 

Blocking and padding data are very common and required unless we are locked into a very short list of audio sample rates.

If this padding was causing problems the most likely symptom would likely be excess jitter or FM distortion of the audio signals that the chip processes. I call this framing jitter, and its not uncommon in digital audio gear, albeit generally at innocuous (inaudible) levels.

It is possible that with sufficiently sensitive measurements traces of this the effects of this operation will be found. I often find it when I measure digital audio gear. If at inaudible levels, it is just a trivial part of the neighborhood.

I have done detailed technical measurements of other similar  Realtek chips, and studied the measurements of others, and am unaware of any audible problems that can be traced to this form of operation. 

I concur with the relevant information presented by bennetng at https://hydrogenaud.io/index.php/topic,111980.msg922878.htmll#msg922878

Re: USB transient noise with external DAC

Reply #43
Arnold, it must work because the Realtek series of chips is found in millions of PC's.  Whether it works better than just running at a 44.1 clock the way most external dac's do (they have both 44.1 and 48 khz clocks) I have no way of knowing because if there is a sonic difference it is not gross enough for me to detect.

Just likely I will try and forget about the whole external DAC thing and keep using the internal DAC.  My NUC is a NUC5PPYH.  It cost $150 plus $20 for a 4 gig memory card plus sales tax.  The rest of the stuff came out of the recycle pile.  It will play back 1080p video all day long is silent and takes up very little space.  Before that I was using an old Macbook Pro but there was a lightening strike and it got fried. 

Re: USB transient noise with external DAC

Reply #44
Quote
I sent it back yesterday.  At a minimum I will lose $25 for two way shipping.  There is a restocking fee of $22.50, but it will be waived if the manufacturer agrees the unit was defective, but they are already trying to game that.
Bummer, but probably the best decision unless you wanted to keep it around for some other use.

If it's a design flaw and it's a normal characteristic for that particular product, they may not consider it a "defect", especially if it meets all of it's published specs.      I'm not saying that's right, but that may be the way they choose to look at it.

...I've had analog preamps with turn-on or turn-off transients so you had to turn on the preamp before turning-on the power amp and turn the power amp off before turning off the preamp.    The preamp wasn't defective it was just a "schitty" design.

Re: USB transient noise with external DAC

Reply #45
I used to use an M-Audio Revolution 7.1 sound card. Not only would it crackle-POP loudly every time it was initialized by something, but it also had a tendency to cause the whole machine to bluescreen if a static electric shock happened to cross wires with my headphones. Another example of a product that didn't behave properly out of the box. Unfortunately, I don't think it was ever corrected, either.

Your issue with your particular sound hardware may be fixable with a firmware update for the USB part of the device, assuming it's software upgradeable.

Re: USB transient noise with external DAC

Reply #46
DVDdoug, I suspect they will charge me the restocking fee, another $22.50.  That's just two martinis, LOL.  But, if all the Schiit DAC's are like mine, why isn't anyone complaining?

kode54, I am leaning towards not getting another external dac, at least for the time being.   The Realtek isn't evil; I was just hoping for a modest improvement.  Nobody needs BSOD's or loud noises on initialization.  Removing the Realtek driver seemed to help.  If my amp dies (it's 21 years old) I might buy an integrated with a built in DAC as many of them come that way.

Re: USB transient noise with external DAC

Reply #47
why isn't anyone complaining?
Probably because they accept them the way they are.  Turning the power amp off first and on last is a good habit to make.

Re: USB transient noise with external DAC

Reply #48

Jitter can be recognized as raised noise floor and increased sidebands from test tone in analog output. I also measured the analog signal in 44.1 and 48k mode in the post below and attached the result. ALC887 is typo, it is ALC892. Did you recognize any major differences in my test result in 44.1k vs 48k?

https://hydrogenaud.io/index.php/topic,111980.msg922878.html#msg922878


A very relevant part of that thread is IMO the archive of Audio Rightmark tests of the ALC892: https://hydrogenaud.io/index.php?action=dlattach;topic=111980.0;attach=9987

In those tests, the tests  involving a 1 KHz test tone show no artifacts that are less than 100 dB below the amplitude of the test tone.

This and the other tests suggest that this audio interface has high enough quality to not compromise the audio that they recover from the digital domain.



Re: USB transient noise with external DAC

Reply #49

I remember that thread.  Sorry, but it is difficult for me to comprehend your testing.

The tests are in a common format that is well-understood by people who are qualified to judge this kind of data.

Perhaps this short article might help:  http://www.edn.com/electronics-blogs/bakers-best/4313869/Understanding-FFT-plots

Quote
Oftentimes I think the glowing reviews folks give external DAC's is because the vast majority of those writing the reviews listen to certain headphones which the on board headphone amps have trouble driving.

IME this is very common and a good insight. One solution is to add an external headphone amp that can effectively drive the headphones at hand. The Topping NX1  or M-Audio Bass Traveler are common inexpensive high performance tools that I often use for this purpose.  To ensure that they are ready to use at all times I connect each to a dedicated USB power supply to keep their batteries charged at all times and avoid grounding prolbems.