[supercurio]

Quote:

@supercurio: which output method are you using with foobar? If it is DS you can't be sure that the sampling rate reaching the NFB-12 is what you set in foobar player. Say you have 44.1KHz from foobar goes to Xonar set at 44.1KHz but inside Windows Playback Devices Control panel you set your Xonar output to something different then Windows will resample the output. Unless you're using WASAPI you have to set sampling rate at 3 places (foobar, Xonar, Windows playback devices control panel) to be the same.

 
Hi, I did counter-measurements with my AVR SPDIF input (Denon AVR-1911) , and also tested NFB-12 with Linux & Windows both plugged on SPDIF and USB.
Also of course i triple checked the various sampling rate parameters, first thing.
 
No sound issue you can hear or measure with the AVR (clean sound & graphs) - results as described in the measurements graphs with NFB-12 @44k.

Sorry, I packed the NFB-12, I won't do new measurments on it (48k or else)

 

[dw1narso]

Supercurio,
 
I found something interresting...
 
I download the file that you mention: http://dl.project-voodoo.org/killer-samples/udial.wav
 
try it on my notebook, which use Conexant 20561 soundcard/chip. (which can accpet 24bit/96KHz)
 
I play it on foobar. I always get distorted sound when I use 44.1KHz rate. I did try different output format 16/24/32bit does not help.
and the sound become clean and undistorted when I upsample to 48,88, and so on above.
 
The result is the same whether I upsample using SoX or PPHS. The sound become clean when I upsample.
 
I wonder now... could it be that this is the issue on the WM8741 in general?
 
Guys, especially Madwolf,
 
please download the file and try it by yourself with NFB12, with all condition whenever possible. Madwolf, please try play the file in 44.1KHz with different OSR on the chip. As well upsample the file and with different OSR.
 

 

[PPkiller]

something in the sound chain is messing up the 44khz input. i do know some sound chip / audio codec has issue with 44khz. they do not natively support it and instead they resample it to 48khz(lousy resampling). a simple way to overcome this issue is for user to resample it to 48khz with software and then send it to the audio chain.

 

[Elanzer]

Can you describe the distortion you're getting? Using no resampler, or using the SoX plugin to upsample to 4x or 192khz sounds identical, but I'm also not using the NFB-12's amp output. I do hear something in the background of the file, but it doesn't change no matter what I resample it to. It might be just a part of the source file though that I'm hearing, since it doesn't sound like distortion.

 

[dw1narso]

if you listen on the 44.1KHz, there's like a pulsing rattle sound in the background...
 
this ratlling sound simply gone and the main sound become much cleaner when I upsample (actually downsample also clean this background "noise", but this could be because the downsample make the frequency response been cut?) and you can hear then that the 19KHz that supercurio mention is actually a pulsing wave...
 
I found this "noise" on my pc default soundcard Conexant which got the stream directly from Directsound, while Supercurio mention he found it on WM8741 on NFB12, fed by SPDI/F, coming from his PC...
 
this is strange... why the noise only come up in 44.1Khz rate?
Could this be Windows codec issue?
 
Elanzer, are you using Directsound or ASIO to connect to NFB-12. Can you try both DS and ASIO?
 
EDIT: I use Windows XP. And for now, I cannot use other way to test. I'm no longer holding my Audinst.

 

[holland]

Quote:

if you listen on the 44.1KHz, there's like a pulsing rattle sound in the background...
 
this ratlling sound simply gone and the main sound become much cleaner when I upsample (actually downsample also clean this background "noise", but this could be because the downsample make the frequency response been cut?) and you can hear then that the 19KHz that supercurio mention is actually a pulsing wave...
 
I found this "noise" on my pc default soundcard Conexant which got the stream directly from Directsound, while Supercurio mention he found it on WM8741 on NFB12, fed by SPDI/F, coming from his PC...
 
this is strange... why the noise only come up in 44.1Khz rate?
Could this be Windows codec issue?
 
Elanzer, are you using Directsound or ASIO to connect to NFB-12. Can you try both DS and ASIO?
 
EDIT: I use Windows XP. And for now, I cannot use other way to test. I'm no longer holding my Audinst.

Can you describe the noise better and where?  There are artifacts, but usually in the last 4 digits of the phone number.  It's a high frequency whirring which doesn't change from DAC to DAC.

 

[Moving Air]

Hi Supercurio,
 
I have been reading your posts on evaluating the NFB-12 and the professional Measurements done.  I'm very impressed with your determination to make things work right.  I'm MovingAir from Singapore and previously worked in Panasonic and later Philips Electronics as an Audio/Acoustics Design Engineer...but currently, I'm lecturing sound technology in a Technical college.  
 
Initially wanted to buy the NFB-11,but waited too long...so now I switched my order to NFB-12...I'm not quite please to see that the high freq. roll-off coming down too fast...plus not to mention the sampling selection issue.  
 
Are you returning your set?  Aren't you exploring using 3 way toggle switch to correct the sampling selection problem?  If that can be easily done, you can still keep it....btw, I'm curious to hear from your comments on the sound quality aspects assuming all problems have been solved?  
 
Your comments would be appreciated.....thank you very much.

 

[vrln]

Moving Air: The NFB-11 is available once again (non-USB version). So if you are looking for something more neutral (NFB-12 is intended to sound smooth and relaxed), you can change your order back.

 

[FeiJai]

 

[Elanzer]

Quote:

if you listen on the 44.1KHz, there's like a pulsing rattle sound in the background...
 
this ratlling sound simply gone and the main sound become much cleaner when I upsample (actually downsample also clean this background "noise", but this could be because the downsample make the frequency response been cut?) and you can hear then that the 19KHz that supercurio mention is actually a pulsing wave...
 
I found this "noise" on my pc default soundcard Conexant which got the stream directly from Directsound, while Supercurio mention he found it on WM8741 on NFB12, fed by SPDI/F, coming from his PC...
 
this is strange... why the noise only come up in 44.1Khz rate?
Could this be Windows codec issue?
 
Elanzer, are you using Directsound or ASIO to connect to NFB-12. Can you try both DS and ASIO?
 
EDIT: I use Windows XP. And for now, I cannot use other way to test. I'm no longer holding my Audinst.

I have tried a few things. The following is all done with windows set to 192/24 playback device output.
 
Foobar2000:
 
NFB-12 DS no resample - background noise, but sounds like a part of the recording, almost like an echo / analog feedback from the device that's dialing
NFB-12 DS SoX resample 2x = same as above
NFB-12 DS SoX resample 4x = same as above
NFB-12 DS SoX resample 96KHz = same as above
NFB-12 DS SoX resample 192KHz = same as above
 
NFB-12 WASAPI 44KHz = Very present whiring noise
NFB-12 WASAPI SoX 48KHz = Very present whiring noise, even more than 44KHz
NFB-12 WASAPI SoX 96KHz = Background noise, same as all the DS samples
NFB-12 WASAPI SoX 192KHz = Background noise, same as all the DS samples
 
Onboard Realtek ALC892 DS 44KHz = Background noise, same as all the DS samples
Onboard Realtek ALC892 DS SoX 48KHz = Background noise, same as all the DS samples
Onboard Realtek ALC892 DS SoX 96KHz = Background noise, same as all the DS samples
Onboard Realtek ALC892 DS SoX 192KHz = Background noise, same as all the DS samples
 
Onboard Realtek ALC892 WASAPI 44KHz = Present whiring noise, but not as present as NFB-12
Onboard Realtek ALC892 WASAPI SoX 48KHz = Background noise, same as all the DS samples
Onboard Realtek ALC892 WASAPI SoX 96KHz = Background noise, same as all the DS samples
Onboard Realtek ALC892 WASAPI SoX 192KHz = Background noise, same as all the DS samples
 
Tried on both LD MKIII amp out, and NFB-12 amp out = no difference
Tried on ATH-M50 instead of T1 = no difference
 
Cowon D2 = Very present whiring noise.
Nokia N900 smartphone = Present whiring noise.
 
Setting NFB-12's output in Windows to 44KHz/16BIT output, no resample DS = Whiring noise.
 
If the whiring noise is intended, then resampling actually seems to make things worse. All my other devices put forward the whiring noise just like the NFB-12 is doing at 44KHz without any resampling, so I don't see how 44KHz playback is broken.

 

[Currawong]

Here's a question: If normal music is tested with the NFB-12, does it have artifacts? The reason I ask is, I tried playing that tone track through my system and had very weird results if I up-sampled it with cracking distortion coming through.  Checking it with a music program, it seems to have a lot of sound above 20 kHz, which is very unlike anything we'd be playing back normally.

 

[movi]

this is all confusing. can someone please clarify for me weather there's anything wrong with the NFB-12 that i'm waiting for?

 

[holland]

Quote:

If the whiring noise is intended, then resampling actually seems to make things worse. All my other devices put forward the whiring noise just like the NFB-12 is doing at 44KHz without any resampling, so I don't see how 44KHz playback is broken.

LOL.  I actually think it improves a bit, the whirring becomes more focused.  However, not tested on NFB-12 as I don't have it yet (may go for NFB-11), but tested on all my other DACs.
 
I don't think it's broken.  I think it may be some aliasing artifacts, but it's hard to determine without an oscilloscope and a FFT running close to the nyquist frequency (19kHz, 20kHz, 21kHz, etc.).
 
The NFB-12 does nothing special.  It utilizes the WM8741 filters, but choses a low oversampling rate (192kHz selection).   I don't think it's optimal, YMMV.  After the WM8741, it goes to a discrete buffer stage.  It's nothing special, it's not a high end CDP which would have a DSP or FPGA before the DAC and disabling all the DAC filters.  It's a very basic off the shelf DAC to a buffer, but at a good price.
 
FWIW, I configured my DIY dual WM8741 DAC just like the NFB-12, except it uses opamp line drivers with no headamp.  My DAC has 3-state switches for oversampling and filter, so I can choose and listen to every HW filter configuration at whim.  So, while my opinions about udial are not on the NFB-12, I have tested it in a similar configuration and on various other DACs, (Alien DAC, Buffalo DAC, Porta Corda, PCM1794 DAC, all DIY).

 

[holland]

Quote:

this is all confusing. can someone please clarify for me weather there's anything wrong with the NFB-12 that i'm waiting for?

Wrong?  In what way?  There are opinions on the filters.  My opinion is that it is configured too wide, but my preference is for high oversampling, with a narrower slow roll-off (much like some higher end CDPs).
 
It's easy to change in any way you want, as the NFB-12 is a simple design, and rightly so.  I would buy it, and mod it if you are capable of doing such.  YMMV.  You can also upsample it, as most people will use it with a computer and not a CD transport.
 
FWIW, I can hear differences in filters with pretty much any setup.  The frequency roll-off is noticeable, but subtle as some sounds push to the background.  Others may notice it too.  I do agree the NFB-12 is valued well.  It's a pretty basic design, and really hard to get wrong.  If there were switches for oversampling rate and filter selection, then there really wouldn't be much to argue about.  Select what you want.  I would disagree it is nitpicking off of measurements.  The WM8741 is warm even without the rolloff, IMO.  The Sabre chips, those are colder.
 
If your choice is between the NFB-12 and some soundcards or onboard audio, NFB-12 all the way.  You can change the "flavor" by upsampling.  If that's not in your ability then one should consider the frequency graph.  If you like the NOS TDA1543 with passive I/V, you'll probably like the NFB-12 in it's standard configuration without upsampling.

 

[Elanzer]

Quote:

this is all confusing. can someone please clarify for me weather there's anything wrong with the NFB-12 that i'm waiting for?

I wouldn't worry about it. It's all nit-picking based off measurements (which we already know, aren't the end-all). I personally can't tell a difference between resampling and not-resampling at all even with these very revealing headphones, as the frequency that's rolled off and experiencing "artifacts" is inaudible (atleast, to me). For $200, the NFB-12 is still a very solid performer for the price, you can't expect to get a DAC/Amp that's perfect in every single way for $200. It's very good for the price and sounds awesome, I haven't heard any artifacts or anything abnormal while using it. If for whatever reason you don't like it, just sell it on the FS forum.
 
It's definitely worlds better than my onboard and xonar DX, and a bit better than my old Sparrow as well (mostly preference though, NFB-12 seems warmer than my Sparrow and I prefer it that way.