Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › NuForce uDAC-2 Drama (detailed measurements)
New Posts  All Forums:Forum Nav:

NuForce uDAC-2 Drama (detailed measurements) - Page 14

post #196 of 208

Thanks for the replies, guys. I'm completely new to the world of DACs and this thing was my first purchase based on the rave reviews it got everywhere on the web (I didn't know anything about the "drama" review before I bought it, and I found out by accident while doing searches for additional information/reviews).

 

I do love the sound of this thing a lot. I'm not using it for headphones, I have it hooked up to my desktop to my receiver so I can listen to my Wharfedales.

 

I tend not to listen to my music loudly so you're saying what that really means is I'm just not getting the full dynamic range, the actual resolution is still the same. That's what I was confused about, then. It still makes me mad even knowing that but I guess I can't have everything. I was thinking about upgrading this to the SE but I'm unsure now. Does anyone know if it has the same issues as the udac-2 normal one? (I assume it does)

 

I've heard good things about the Music Streamer IIs except I've heard they're louder so if you, I guess, turn down the volume of the computer....would that in effect, do the same thing as this thing with the lowering of the dynamic range? Sorry, but I'm a technical ignoramus when it comes to technical knowledge of these things. (I know what sounds good, though)

 

Would you guys say the UCA202 or HRT MSIIs are a significant upgrade in sound quality than this thing? What about the V-DAC which I've also read good things about?

 

 

post #197 of 208
Quote:
Originally Posted by maverickronin View Post

If you get a 24 bit DAC like that HRT you can use the digital volume with out losing any resolution on 16 bit resolution files.


Isn't the udac-2 supposed to be 24 bit? Or is it basically because they screwed up the design it loses the 16 bit resolution without being 100% (and the HRT doesn't)? Sorry, like I said, I don't know too much about the technical details.

 

post #198 of 208

THe uDAC-2 clips with 0 dBFS digital input, meaning that if you listen to pop music which is oftne mastered loudly with a lot of content near 0 dBFS, it will continuously clip the sound unless you use digital control volume to attenuate the sound.

post #199 of 208
Quote:
Originally Posted by nbakid2000 View Post

Would you guys say the UCA202 or HRT MSIIs are a significant upgrade in sound quality than this thing? What about the V-DAC which I've also read good things about?


I don't find any DAC upgrades to be significant, except perhaps onboard sound to something if your computer's noisy. If you're able to get a return on the uDAC, I say get the UCA202. As a standalone DAC, it outperforms the uDAC (whether that's audible or not is up in the air) and is significantly cheaper. Might as well save some money wink.gif

 

Quote:
Originally Posted by nbakid2000 View Post

Isn't the udac-2 supposed to be 24 bit? Or is it basically because they screwed up the design it loses the 16 bit resolution without being 100% (and the HRT doesn't)? Sorry, like I said, I don't know too much about the technical details.

 

One thing to keep in mind is that bit depth only determines SNR and dynamic range. Every bit is 6 dB of SNR. 16 bits is 96 dB of SNR, 24 bit is 144 dB. That's all it really does when you consider playback (it's more useful for mastering because it allows extra headroom to play with effects before clipping). So there's a difference between supporting 24 bits (being able to convert 24 bit files) and actually playing at 24 bit resolution. The uDAC has a measured noise level of around -94 dB. That means noise is 94 dB below the signal. That means its SNR is 94 dB. That's not even enough for full 16 bit resolution. Forget 144 dB! In fact, I don't think there's any DAC on the market that supports a full 144 dB SNR, even though there's some DAC chips which can convert 32 bit files for whatever reason. It's not physically possible to get that sort of resolution.

 

There is an advantage to supporting 24 bits in that you can digitally attenuate the signal without losing SNR. The quantization noise exists in the least significant bits (the highest bits) as far as I know, and those are the first to go with digital attenuation. But it's really not a big deal, especially if you don't digitally attenuate much or at all.

post #200 of 208
Quote:
Originally Posted by nbakid2000 View Post

Isn't the udac-2 supposed to be 24 bit? Or is it basically because they screwed up the design it loses the 16 bit resolution without being 100% (and the HRT doesn't)? Sorry, like I said, I don't know too much about the technical details.


I was under the impression that while its DAC chip could accept 24 bit data the USB receiver chip didn't actually support transfering 24 bit audio.  I could be wrong though.

 

Open up the device properties on your computer and see what bit depths and sampling rates you can set it to.

post #201 of 208


OK, so basically, you're saying that even if I lower the volume on the computer from 100% to say like (well, you said to find a volume slightly lower than normal, so I found like...90% to sound good, I guess) it shouldn't really matter as far as my sound quality *that* much. Is that right?

 

As I said, I really really like what I'm hearing on the sound side of things. I just don't want to get clipping but I don't want to degrade my audio quality by avoiding that either.

 

Sorry for the questions regarding this stuff. redface.gif

 

Regarding audio properties on the PC:

 

I did check my settings on my computer and I have everything set to highest resolution for audio. Also: should I install WASAPI or ASIO4ALL on my computer? Would that help with the sound quality?

Quote:
Originally Posted by Head Injury View Post


I don't find any DAC upgrades to be significant, except perhaps onboard sound to something if your computer's noisy. If you're able to get a return on the uDAC, I say get the UCA202. As a standalone DAC, it outperforms the uDAC (whether that's audible or not is up in the air) and is significantly cheaper. Might as well save some money wink.gif

 

 

One thing to keep in mind is that bit depth only determines SNR and dynamic range. Every bit is 6 dB of SNR. 16 bits is 96 dB of SNR, 24 bit is 144 dB. That's all it really does when you consider playback (it's more useful for mastering because it allows extra headroom to play with effects before clipping). So there's a difference between supporting 24 bits (being able to convert 24 bit files) and actually playing at 24 bit resolution. The uDAC has a measured noise level of around -94 dB. That means noise is 94 dB below the signal. That means its SNR is 94 dB. That's not even enough for full 16 bit resolution. Forget 144 dB! In fact, I don't think there's any DAC on the market that supports a full 144 dB SNR, even though there's some DAC chips which can convert 32 bit files for whatever reason. It's not physically possible to get that sort of resolution.

 

There is an advantage to supporting 24 bits in that you can digitally attenuate the signal without losing SNR. The quantization noise exists in the least significant bits (the highest bits) as far as I know, and those are the first to go with digital attenuation. But it's really not a big deal, especially if you don't digitally attenuate much or at all.



 


Edited by nbakid2000 - 1/12/12 at 8:20pm
post #202 of 208


 

Quote:
Originally Posted by khaos974 View Post

THe uDAC-2 clips with 0 dBFS digital input, meaning that if you listen to pop music which is oftne mastered loudly with a lot of content near 0 dBFS, it will continuously clip the sound unless you use digital control volume to attenuate the sound.


With all the dis-information here it's nice to see something posted that is correct, and not mere speculation or guesses.  Thanks!

 

My understanding is that when it clips with a 0dBFS input, like with brick-walled pop music, the distortion climbs to approx 1% (from what I've been told).  I find that most brick-walled pop music sounds like crap out of any DAC, and the uDAC-2 is no different.  However, with all the properly recorded music that I listen to this is never a problem.  And I daresay that most people would NOT recognize the 1% distortion when it rears it's head in cases of poorly recorded music, otherwise they wouldn't listen to that music at all in the first place.  With brick-walled music, dropping the system volume just enough to keep the signal below 0 dBFS is not going to hurt the sound quality of the poor recordings, but has a chance to make them better.

 

For the others who posted recently - the uDAC-2 supports 24/96 USB Audio, in the receiver chip and the ESS DAC.  It sounds significantly better than my old UCA202 via headphone out, and still also a bit better than the UCA202 line-out (which I have owned twice).  The UCA202 headphone out had a large amount of hiss with IEM, poor power levels, and didn't sound very transparent vs even the original uDAC (which was more warmly colored than the uDAC-2).  The UCA202 line out seemed to lack micro-detail and dynamics in comparison to other DACs, and my Headstage DAC cable with PCM2906 had better sound quality (although not as good as a PCM2902E Japan).  If someone is looking for a cheap USB DAC/amp combo I'd be more likely to recommend the HiFiMan HM-101 that I haven't even heard yet, than to recommend the UCA202!  It uses the same DAC as the Meier 3MOVE and XM-5, which I have enjoyed despite the very slight graininess of their DACs vs those with WM8740 DACs.

 

Although the uDAC-2 headphone amp lacks that last bit of micro-detail, resolution, and soundstage that my DACport or upgraded iBasso D4 offer, I still think it's a nice step up in balance from the original uDAC and it bridges the gap much better.  I also find the uDAC-2 to sound very good via line-out into my Woo WA6, EF5 and ALO Amphora.  Plus, side by side with my Pico DAC and iBasso D4 line-out the uDAC-2 is almost impossible to distinguish between them when comparing them blind.

 

 

 

post #203 of 208
Quote:
Originally Posted by HeadphoneAddict View Post

With all the dis-information here it's nice to see something posted that is correct, and not mere speculation or guesses.  Thanks!

 

My understanding is that when it clips with a 0dBFS input, like with brick-walled pop music, the distortion climbs to approx 1% (from what I've been told).  I find that most brick-walled pop music sounds like crap out of any DAC, and the uDAC-2 is no different.  However, with all the properly recorded music that I listen to this is never a problem.  And I daresay that most people would NOT recognize the 1% distortion when it rears it's head in cases of poorly recorded music, otherwise they wouldn't listen to that music at all in the first place.  With brick-walled music, dropping the system volume just enough to keep the signal below 0 dBFS is not going to hurt the sound quality of the poor recordings, but has a chance to make them better.


Only 0.67% actually. But a lot of it odd order harmonics (the third order hamonic is at -48 dB). These are the harmonics that every audiophile always complains about with solid state equipment, even when it's magnitudes lower (of course, without actual proof that they hear it).

 

What speculations and guesses, by the way?

post #204 of 208

Audiable disortion is around 1% at 10khz with music , http://www.axiomaudio.com/distortion.html , if you listen to sine waves audiable disortion is around 0.2%.

post #205 of 208
Quote:
Originally Posted by JRG1990 View Post

Audiable disortion is around 1% at 10khz with music , http://www.axiomaudio.com/distortion.html , if you listen to sine waves audiable disortion is around 0.2%.


One possible flaw with the test: It appears they were told to raise their hand if they heard something that didn't sound right. But without being able to compare to an undistorted signal, it's hard to say what does or doesn't belong in the music. It may have been a better test if they did ABX tests between a distorted and undistorted signal at each frequency, rather than raise hands when it's loud enough to be obvious.

 

On the other hand they didn't use harmonics, just pure signals of different octaves. I'd assume that harmonic distortion would be harder to detect than a pure uncorrelated signal, so that might mean distortion in amps is even harder to detect.

post #206 of 208
Quote:
Originally Posted by Head Injury View Post

 

What speculations and guesses, by the way?


Things like not accepting 24 bit, or losing resolution, or that lowering the system volume (on the digital side) will decrease sound quality, among other things not at the top of my mind right now.  I thought your previous point about the SN ratio not being good enough to hear the improvements offered in 24 bit music was indeed valid, if the specs you quote are correct (I don't know).  

 

Where I see the improvements with 24/96 over 16/44.1 music is that you can digitally attenuate the music without much compromise, and the frequency response is more extended and airy.  The 96Khz sampling rate to me is more important for sound quality than the 24 bits which is mostly involved in Dynamic Range.  With most music, even with headphones which aren't soundproof, the ambient noise around us is going to reduce the dynamic range that we can appreciate to about 60 dB or so (an educated guess based on my heating system volume levels and the 120dB max I might expect to encounter).  So, to me an SNR of -94 is perfectly fine for most listening (budget gear or not).  I have the uDAC-2 set up on my Kitchen iMac, and there is no way I can appreciate a full -144dB of SNR, or even -80 or -90.

 

I leave my uDAC-2 and other DACs set for 24/96 all the time on my Mac (in Audio Midi Setup), so that nothing has to be down-sampled by my computer.  This way if I'm not playing 24/96 files iTunes still does a very respectable job up-sampling, rather than letting Apple Core Audio up-sample. If I leave the Mac set at 16/44.1 then I lose all the benefits of the higher bit-rate.  By having it at 24/96 all the time, I also have the extra bits to play with for digital attenuation if needed.  Usually that's only needed with extreme low volume listening with sensitive IEM, but at that point the volumes are so low I still would not use more than -60 dB of SNR.

 

 

post #207 of 208

Hi guys! I know that an old topic but i want to do a question biggrin.gif

 

For you, with a library of only 44.1/16 music, it's better to set 94/24 in the midi setting anyway?

 

Thanks wink.gif

post #208 of 208

When will you come back ?

If you read this , please come back :)

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › NuForce uDAC-2 Drama (detailed measurements)