Audio-GD NFB-12
Feb 4, 2011 at 11:43 PM Post #541 of 2,278
Not sure why it matters. I am at ~250 hours and It still doesn't sound good with the 580. As good as it is, my DIY amp and 13 year DAC are still a level above, especially in dynamics and extension. I wish that wasn't the case, but it is. The NFB-12 is not going to become more dynamic and extended. The mids are thick and haven't changed. This works great for some phones and not others. It's just the way it is. I believe in burn-in, but it's not magic. I have only found burn in to smooth things and tighten them. Not change the sound signature.
 
Quote:
Quote:
Chill out Tim. I posted initial impression which are done on this forum every day and then I updated after a good amount of burn in. And remember, these come from the factory with 100 hours on them already, so we are not talking about brand new gear. It's funny on this forum because burn in is always longer than what someone has on it. If I say I have 300 hours, someone always comes along and says it needs 400. If I have 400, it needs 500. People like to know what something is going to sound like when they get it. It helps with any initial disappointment and buyers remorse. 
wink_face.gif

 
Quote:
Isn't the HDP $450? Isn't the NFB12 $200? All/most comments are based on a handful of hours use? No wonder reviewers like Skylab are appreciated here, he lives with the gear for days/ weeks before rendering an opinion.


Views on burn in for Audio-GD gear are pretty consistent though. Have you owned any of their gear before or read any of the other threads for their other gear? Not being accusatory here, just curious.



 
Feb 4, 2011 at 11:55 PM Post #542 of 2,278
Thanks for the advice guys!
 
Feb 5, 2011 at 12:28 AM Post #544 of 2,278







A
It would be helpful to get other impressions of how the NFB-12 couples with the HD650. I personally think there is great synergy between the two - the sound is full bodied, dynamic, and very musically engaging. Its done a nice job of brining out the fun personality out of the HD650s which apparently is a latent/hidden characteristic of these phones. FYI, I have the HD650s with new drivers which may explain why I'm not getting the veil with them. 
 
I prefer the Moon hdam installed in my Audio-GD Fun as it adds warmth and body to the sound signature. It's not as resolving as the more analytical Earth hdam, but I feel Moon has wonderful synergy with the HD650. It's energetic and engaging. Fwiw, I also have an HD650 with the "new" version drivers, with no "veil" in the first place, other than the less refined sound when not running off a dedicated amp and good source.It's worth noting that people commenting on the NFB-12's unsuitability for the HD650 don't even have that headphone in their inventory.



 


Agree. Over a year ago I bought a HD650 and a audio-gd C-2 amp and DAC19 DSP1. I just listened to Paul Simon's Graceland from "You Can Call Me Al" to the end of the album for the first time on my rig. My rig is burnt in. Guess what? "You Can Call Me Al" was similar to what the previous writer described a few pages back-it was not bright and the bass was present and fairly tight but I was not wowed with the track at a low medium volume. I believe this is because the HD650s and the audio gd gear are neutral and revealing of the source recording. As I listened to other tracks I heard more treble where appropriate, a wide sound stage and great instrument separation and imagery-in other words other tracks wowed me like "diamonds". The DAC19 has the PMK1704 chip but I e-mailed Kingwa the other day to ask which chip is closest to the Wolfson (NFB-12) or the Sabre (NFB 11) was closest to the 1704 and the audio gd reference series DACs. Kingwa said the Wolfson. I now have on pre-order a NFB-1 WM and a NFB-12 both using the Wolfson. All of this is a long way of saying I think the HD650s sound great on audio-gd gear.

I love that Kingwa jumps on the thread during his holiday. Great customer service.

Slaughter hookem! I am a displaced horn and Austinite in S. Cal.
 
Feb 5, 2011 at 1:37 AM Post #545 of 2,278
I ever wrote a article and post at a Chinese audio magazine in 2010.I can't fully interpret the article to English base my English ability.
I think the 192KHz support not a real upgraded.
The software (the recorders ), there is a real improved than 44KHz.
But  the hardware, a lot newest D/A chips only support up to 96KHz or lower. The newest DA chips declare support up to 192KHz not real open a new field than the older PCM1704.
The WM8741, the ES9018, at 192KHz input, they want to work on the "high rate" filter mode.(Had declare in their PDF files.)
Read from the output waves of the WM8741 on oscillograph , I presumed the 192KHz input is working at 2X times oversample. At 96KHz is 4X (8X of some other chips ) oversampling.
If the digital filter setting to 2X oversampling, the PCM1704 can working at 384KHz ( PCM1704 can support highest 768KHz input ).
In theory  the 8X oversampling sound more detail and black backgroup than 4X and 2X modes , and I gree the throry with my experience.
So 192KHz files sound maybe no benefit than 96KHz files after the DAC.
But I gree the WM8741 is a excellent DA chips, its sound mellow and  analog taste, the soundstage is quite wide and deep than some other chips. So we design the WM8741 based DACs was follow its flavors.
Sorry again for my English.
PS: Edwin is not me, his English is much better.
 
Feb 5, 2011 at 1:59 AM Post #546 of 2,278


Quote:
I ever wrote a article and post at a Chinese audio magazine in 2010.I can't fully interpret the article to English base my English ability.
I think the 192KHz support not a real upgraded.
The software (the recorders ), there is a real improved than 44KHz.
But  the hardware, a lot newest D/A chips only support up to 96KHz or lower. The newest DA chips declare support up to 192KHz not real open a new field than the older PCM1704.
The WM8741, the ES9018, at 192KHz input, they want to work on the "high rate" filter mode.(Had declare in their PDF files.)
Read from the output waves of the WM8741 on oscillograph , I presumed the 192KHz input is working at 2X times oversample. At 96KHz is 4X (8X of some other chips ) oversampling.
If the digital filter setting to 2X oversampling, the PCM1704 can working at 384KHz ( PCM1704 can support highest 768KHz input ).
In theory  the 8X oversampling sound more detail and black backgroup than 4X and 2X modes , and I gree the throry with my experience.
So 192KHz files sound maybe no benefit than 96KHz files after the DAC.
But I gree the WM8741 is a excellent DA chips, its sound mellow and  analog taste, the soundstage is quite wide and deep than some other chips. So we design the WM8741 based DACs was follow its flavors.
Personal I like it's working at 96KHz or lower modes, but people always like the 192K support on paper.
Sorry again for my English.
PS: Edwin is not me, his English is much better.

 
haha well i love you both 
 
 
Feb 5, 2011 at 2:20 AM Post #547 of 2,278


Quote:
Isn't the HDP $450? Isn't the NFB12 $200? All/most comments are based on a handful of hours use? No wonder reviewers like Skylab are appreciated here, he lives with the gear for days/ weeks before rendering an opinion.


I'm glad there are people on here like skylab too.  I'm also glad that there are people who will post quick impressions on a new piece of gear instead of not contribute and just complain about it.  It's not like there isn't time in the future to write a more thorough review.  Plus if you're so wound up about it (you obviously are since you keep making posts about it) then go buy a NFB-12, listen to it for a few weeks, and then post your impressions.  At least in the end that will be exponentially more constructive than just complaining. 
 
Feb 5, 2011 at 7:15 AM Post #548 of 2,278
Hi,
first of all many thanks to supecurio with his measurements!
I have to say, although I'm engineer in communication electronics, I'm not familiar with the layout of hi-fi DAC design.
After some resaearch I found the audio-gd solutions quite attractive in function and build quality, maybe like some of you. My NFB-12 is ordered since end of decembre and in the pipeline...
 
Looking at the frequency responses I became some headache!
The usual approach of hifi (High Fidelity!) should imo mean to try to harm or violate a signal as less as possible. Also from "exotic" solutions I expect today, that the main technical data follow these rules. There are enough solutions showing that this is possible.
I know of course, filter design is alway a compromise beetween response os frequency, phase, group delay and pulse response in time domain. NFB-12 seems to me not very well balanced therefore. It can be very helpful to compensate e.g an agressive sounding headphone. But don't forget, this works only in this combination well. And don't forget, NFB-12 has also RCA outlets, they sound the same.
For sure each designer, constructor ore entrepeneur may create devices as he likes to do, but then the approach or philosophy should be clearly mentioned.
The drop-off of the NFB-12 was confirmed to me in PN by kingwa and now in his posts above.
 
For me that means they are providing a sound, or better different sounds (NFB-11...). But the NFB-12 is never ever "neutral" as mentioned. They specify 20Hz - 20KHz, but this absolutely useless without specfication of the limits, in this case -3.5 dB.
This is a lot and for sure hearable. This will never be compensated by burn-in and there will be no need to think about tuning with cables, capacitors or transistors. These influences are are by far smaller than cutting high frquencies by such a filter.
Additionally, if I got kingwa right regarding the analog filter, then the frequency response must be identically regardless the sample rate, but this is not the case.
 
BR
 
Michael
 
Feb 5, 2011 at 7:16 AM Post #549 of 2,278
Modding NFB-12 to 96kHz rate?
 
Hi King-wa and other DAC-engineers, how to mod NFB-12 to fixed 96kHz rate to make it 4X oversampling as King-wa said? I have read in an earlier post thet this is set by certain voltage on some leg of the WM chip.
 
But, if I convert my output to 192kHz in the PC before sendig it to dac, does the WM chip make a new sampling anyway?
 
 I am 51 years old and I can't hear anything over 15kHz anyway, so I'm not concerned of analog filter slope! But the benefit of more X multiplied oversampling should be hearable to all of us. I cannot solder the board but I can find a friend to do that. How about produding a 96kHZ (audiophile) version?
 
I have placed an order of NFB-12 from ampcity early in January 2011, and I'm waiting for delivery very anxiously. This is my first separate DAC and I'll hook it to my PC/Foobar-player and BD-player. I have a Vincent SV-129 2ch amp (to line-out) and a pair of AKG 601 and Sennheiser HD540 Reference (a classic from 1980's) and the legendary 414 to HP out
 
Greetings from snowbound Finland!
Juha
 
Feb 5, 2011 at 8:06 AM Post #551 of 2,278

 
Quote:
I ever wrote a article and post at a Chinese audio magazine in 2010.I can't fully interpret the article to English base my English ability.
I think the 192KHz support not a real upgraded.
The software (the recorders ), there is a real improved than 44KHz.
But  the hardware, a lot newest D/A chips only support up to 96KHz or lower. The newest DA chips declare support up to 192KHz not real open a new field than the older PCM1704.
The WM8741, the ES9018, at 192KHz input, they want to work on the "high rate" filter mode.(Had declare in their PDF files.)
Read from the output waves of the WM8741 on oscillograph , I presumed the 192KHz input is working at 2X times oversample. At 96KHz is 4X (8X of some other chips ) oversampling.
If the digital filter setting to 2X oversampling, the PCM1704 can working at 384KHz ( PCM1704 can support highest 768KHz input ).
In theory  the 8X oversampling sound more detail and black backgroup than 4X and 2X modes , and I gree the throry with my experience.
So 192KHz files sound maybe no benefit than 96KHz files after the DAC.
But I gree the WM8741 is a excellent DA chips, its sound mellow and  analog taste, the soundstage is quite wide and deep than some other chips. So we design the WM8741 based DACs was follow its flavors.
Personal I like it's working at 96KHz or lower modes, but people always like the 192K support on paper.
Sorry again for my English.
PS: Edwin is not me, his English is much better.

 
For people who think that NFB12 sound bad, I believe you are feeding the NFB12 using a 44/48khz input. 
I strongly urge you to feed the NFB12 with a 192khz source and hear the difference. 
Using a 96k source is only slightly better as it is just half way there. 
I know 192khz source is hard to find but try. 
 
After reverting to Pin 22 to 3.5V. the sound become darker and veil, The tonal balance is off at the high end. After some listening I know that this cannot be right.
Again I remove Pin 22 from the PCB and after I confirm with a few song the sound is great. At least the high end gives the correct tone.  
AB test with another system to confirm. 
 
The same amount of hissing is heard at both setting. 
 
Taken from Page 40 of Wolfson 8741 datasheet

We need to set the correct setting for this pin to the source sampling frequency if not the digital filter will be off. 
The NFB12 current fix this pin at one fixed setting. 
 
From supercurio graph the NFB12 is correctly configured if the input source is 192khz
There a very slight error if the input source is 96khz
But if the source is at 44khz the high frequency is rolled off prematurely.
 
It cannot be the 470pf or 100pf as pointed out by KingWa. because if it is this it will affect all the graph at different source frequency.   
 
 
高音的问题只有在用户使用44/48khz输入时才会出现。
如果输入是192khz高音是没有问题的。 
96khz输入时影响不大。 
 
你说的那个470pf capacitor,看来影响不会那么大。有影响的话,影响44/96/192的程度也会是相同的。 
可是我从supercurio的报告看到影响的程度是不一样的。
 
Wolfson8741的说明书第40叶,写到不同的输入是要用不同的设置的。 可是我量到这是锁定在3.5V的。 
 
 

 
 
Feb 5, 2011 at 8:34 AM Post #552 of 2,278
Thanks madwolf,
yes, the NFB-12 must sound better @192kHz, simply look at the f-response.
and yes, this cannot  be caused by the analog 6dB-lowpass after D/A conversion.
 
IMO, these filters are more important for the let me say 'sound-signature' of an DAC then the Converter itself.
 
@kingwa: Meantime I head a look in some theory. Logically such a filter is really needed with delta sigma converters too. 'Normal'  approach seems to be a 2 step filter SC/RC and not simple 6- dB one stage low pass. The example you are mentioning is still better compared to NFB-12 with -1dB @ app. 16-17 kHz...
 
BR
Michael
 
Feb 5, 2011 at 8:37 AM Post #553 of 2,278
I think it is fine for each product to have it own sound signature 
But the NFB12 sound differently with the same song with the input set to the different sampling frequency. 
 
一个产品有自己的风格 是个好事。 
可是如果用不同的输入带音,有不同的效果,是个很难理解的事。 
 
Feb 5, 2011 at 9:02 AM Post #554 of 2,278
I certainly can't write good English but I think I can read English not too bad, but thanks the Chinese explain.
 
Seach and read the test diagram in this link
http://www.stereophile.com/digitalprocessors/
A lot D/A gears (include the VERY hi-end brands) all have different frequency band at different sample rates.
 
How easy to design and build a gear with 20-20KHz less than 0.5DB ? An electronical university man can do well.
 
I just surprise there are some people still like to look at the specs on paper at first , then  decide how to listen the sound.
 
Feb 5, 2011 at 9:27 AM Post #555 of 2,278
thanks kingwa,
 
the link does not work for me..
For sure many devices have different f-response depending on the sample rate. But more above 20 kHz and this means not in the range people can recognize huge differences.
 
You know, people in europe are buying a lot from the net, but mostly inside their own country. Buying from China is for most including me a different story. I'm sure you are very supportive and helpful, but I simply ask for your understanding, that I want to get as much information as possible before the product is delivered.
In EU for instance you may by law return each product to supplier wthin 14 days without any explanation and without any costs. Also here I'm sure you would agree on that too, but it would be more complicated and expensive than inside Europe anyhow.
 
And, a graph or a number is - at least for me - much concreter information than words like 'neutral, dynamic, smooth , forgiving...'. Such is always a matter of taste.
BR Michael
 

Users who are viewing this thread

Back
Top