[money4me247]

  Can you define "transparency or soundstaging" and ascertain what affects them? I just love these audiophile terms.

transparency is how true to the source the sound is. of course hard to say what exactly the "true to source" sound as our equipment for replicating sound introduces coloration and distortion. can use measurements to see if the distortions are below the threshold that the human ear can detect.
 
sound stage is the subjective perception of depth & space in music. primarily dependent on the spatial cues in the source file. also can be altered by the design of the headphone (open/closed), damping, and earcup size/design.
 
relating to how a dac would affect those factors... honestly, i personally have no idea as i have not read much about that specific field. probably relates to how accurately the dac reproduces the original sound file. additional distortion/noise can alter the transparency and the spatial cues from the original sound file.
 
well, there you go... a easy explanation that everyone can understand. lol

 

[StanD]

  transparency is how true to the source the sound is. of course hard to say what exactly the "true to source" sound as our equipment for replicating sound introduces coloration and distortion. can use measurements to see if the distortions are below the threshold that the human ear can detect.
 
sound stage is the subjective perception of depth & space in music. primarily dependent on the spatial cues in the source file. also can be altered by the design of the headphone (open/closed), damping, and earcup size/design.
 
relating to how a dac would affect those factors... honestly, i personally have no idea as i have not read much about that specific field. probably relates to how accurately the dac reproduces the original sound file. additional distortion/noise can alter the transparency and the spatial cues from the original sound file.
 
well, there you go... a easy explanation that everyone can understand. lol

Nice work.
The spacial cues are in the recording and may be affected by the phase shifts or the FR in a headphone, no such thing in an amp or DAC that will be a cause for concern,
To me a DAC or an Amp with no perceptible distortion and a flat FR is transparent.
If one wanted to get a DAC or Amp with noticeable distortion, I'm certain that it can be found from a dubious source on fleaBay, real cheap.

 

[Bill-P]

Sorry, couldn't resist jumping in...
 
Allow me to cite InnerFidelity's measurement of these headphones:
Sennheiser HD 650
Sennheiser HD 800
 
Notice their THD+Noise is in the ~1% range and the HD800 is very close to 0.1% across the board.
 
But every little bit of that above 0.1% is audible and even measurable as an effect on frequency response. This is quite obvious when doing free air measurement on the driver (no effect due to enclosure), and then with enclosure (now distortion and reflection is added due to enclosure).
 
Some amps would actually add its own distortion/noise on top of this, although the end result is that there may be little change to FR if an equal amount of distortion/noise is added, but it is definitely very audible.
 
And yet when measured, the measurement system insists it's 1% or less.
 
So it is very much in the math and how people would like to represent it.
 
And now I'll pass this back to you.
 

 

 

[StanD]

When it comes to headphones there's much to measurements vs. the acoustic coupling to each individual's head, some are denser than others 

. Sorry I couldn't help that.
Bill we're talking about DACs not headphones, keep it in context.

 

[Bill-P]

Hahaha, I know.
 
And yeah, hence why I apologized. 

 
But anyway, back to our regular DAC discussion.
 
I think that in some cases, there is interaction between the DAC and the amp that causes the signal to be... well, "perverted". For instance, a DAC with an output impedance of 1000 Ohm would typically be happier driving an amp with an input impedance of 10000 Ohm, because then the output impedance of the DAC won't cause a current limitation, which would drop the voltage level, causing the signal to be clipped. Some amps have input impedance up to 100K, but some are as low as 5K, and those would make a DAC with 100 Ohm and some current limit to cry and bawl.
 
On the other hand, there are also amps that are not happy with too much voltage coming from the DAC. For instance, the Objective 2 will not be happy with a DAC that outputs more than 2Vrms. So anything above that will cause the amp to clip. Some (if not most) big desktop DACs do output higher than that... as high as 4Vrms in some Audio-GD DACs, I think.
 
When there's clipping of either the amp or DAC, bad things happen. And I think that's one way some may hear a difference between different DACs.

 

[StanD]

DACs don't have 1K output impedances, something around 75 Ohms is more like it. I don't think the Studio line impedance of 600 Ohms is such a big thing anymore as output impedances narrow in to the zero Ohm goal and input impedances go higher, The only thing staying is the reference voltage levels. If you find a DAC or Amp that doesn't conform to level standards, I'd say, don't buy it. If you buy a Audio-GD DAC and you have such a problem, attenuate the output with an L-PAD. Where are you going with this? It does not compute.
Huh? "But some are as low as 5K, and those would make a DAC with 100 Ohm and some current limit to cry and bawl." Why would a 5K input impedance cause problems when attached to something with a 100 Ohm outut impedance? That's an easy load. What current limit? One more Huh for that one.
 
Which Audio-GD DAC has a 4V SE output level?

 

[Bill-P]

ODAC actually has an output impedance of around 400 Ohm, according to its creator.
 
And some other DACs typically also have equally high output impedance to suppress distortion due to capacitive/inductive load (bad for current-limited devices).
 
So no... output impedance hasn't gone to the zero ohm range. It only does so for amplifiers, while DACs can be as high as... well, however high their creators want them to be. Unless you're saying the ODAC is now not conforming to level standards, but... there was never any standard for DAC output impedance as far as I know. There has only been a recommended input impedance for amplifiers to be around 10K Ohm.
 
And a 5K input impedance certainly wouldn't cause any problem with a 100 Ohm output impedance, indeed, but with a 1K output impedance, I think it would, because VDR would say that the signal is clipped to about 5/6 its original value, or approx. 17% of the signal is attenuated by the output impedance. That's not good.
 
And DACs actually do have an output current limit. Or I guess I should put it differently: DACs should not be required to output a lot of current, because higher current output causes higher distortions/noise, and all sorts of bad things. Also, DAC chips typically have much lower current output limit than op amps, so they must be designed with a buffer in mind in order to mitigate this issue:
http://dkc1.digikey.com/us/en/tod/ADI/Five-Things-DACs-Part-4/Five-Things-DACs-Part-4.html
 
Here's the datasheet for the PCM1794, one of the more popular modern DAC chips, which cites its current output limit, and why it needs more buffering to be done:
http://www.ti.com/lit/ds/symlink/pcm1794.pdf
 
Notice TI does recommend low-noise op amps to go with that DAC, because everything else around the DAC affects it. A DAC is so much more than just the chip it uses in that way.
 
Edit: not all DAC chips need a buffer, but I have heard and seen some measurements of those that do not go with a buffer, and their typical performance gets crappy at lower frequencies because the DAC chips inside lose its quoted specs as soon as they are stressed to output high current.

 

[StanD]

According to JDS, "You may connect any two devices each with input impedance of at least 10k" That means a 5K load without crying.
I'm not sure that I'd want to buy a DAC where the DAC chip is directly connected to the RCA Jacks.
Attenuation does not necessarily mean clipping.
Regarding Audio-GD which you referred to, the NFB-1, NFB-1S, NFB-3 and NFB-7 all have an output impedance of 10 Ohms, 
Your reference to digikey is for a DAC chip, not a commercial Audio DAC product as we are discussing here. Try to stay relevant.

 

[Hi Rez]

 

 
Anyone with actual experience comparing dacs taking into account that they all have different output levels want to tell me their opinions?

 

Quote:

  Such hatred in this thread.

 
Seems to be a lot of debate in this thread when the OP was simply asking for opinions....
 

  Back on topic, my experience has been that with high-quality recordings, the differences between inexpensive and expensive (and good quality) DACs has been that the latter are more revealing of the fine details and the less expensive DACs have been less resolving. The important part is the bit in bold -- I know that an expensive DAC will do nothing for, say, my Van Halen albums but will make all the difference with the best classical, so I think all the arguments are moot unless the music is being discussed. 

 
Agree that recording quality will influence identification of differences. 
 
Personally, I have both an ODAC and a QB-9 DSD.  The QB-9 DSD clearly is more resolving and dynamic in the bass region.  The QB-9 DSD also gives a more natural presentation of the music, comparatively, the ODAC sounds flat and is missing the nuances that the QB-9 DSD resolves.  When I listen to the ODAC, it initially sounds pretty good, yet it doesn't take long before what's missing (or extra?) makes me want to go back to the Ayre.
 
If you can make the opportunity, see if you can get a loan of a better DAC and discover for yourself if there is enough difference to be meaningful in your system.
 
Happy listening!

 

[Bill-P]

  According to JDS, "You may connect any two devices each with input impedance of at least 10k" That means a 5K load without crying.
I'm not sure that I'd want to buy a DAC where the DAC chip is directly connected to the RCA Jacks.
Attenuation does not necessarily mean clipping.
Regarding Audio-GD which you referred to, the NFB-1, NFB-1S, NFB-3 and NFB-7 all have an output impedance of 10 Ohms, 
Your reference to digikey is for a DAC chip, not a commercial Audio DAC product as we are discussing here. Try to stay relevant.

Yeah, because as said, it's 400 Ohm... so it shouldn't have issues with a 5K load. But a DAC with more than 500 Ohm may actually have some issues.
 
And there is one such DAC where the DAC chip is connected to the RCA jacks here:
http://www.ebay.co.uk/itm/Non-Oversampling-DAC-TDA1543-DIR9001-96kHz-SPDIF-Coax-Optical-9V-NiMH-charger-/200858459050?rmvSB=true
 
(passive I/V) 

 
Attenuation is clipping out low-level signals. It's not like the amp is perfectly transparent and won't add its own noise... so when the signal that comes into the amp is as low as the noise floor, it'll just get lost. Attenuation will cause that to happen more often.
 
As for the Audio-GD devices, I meant they were not appropriate for the O2. The O2 would audibly clip when fed a signal greater than 2Vrms. Audio-GD DACs typically go up to about 2.5Vrms, I think, so that's more than the O2 can handle, and it'll clip. You may very well build an attenuator and mitigate the issue just fine, but then you're clipping out about 20% of the signal in that case (2V vs 2.5V). I don't think I'd choose that route.
 
Either way, I'm just trying to come to some explanation as to why some DACs may sound different, coming from an actual technical standpoint, and not just due to subjective impressions. Granted, this has no bearing on their actual pricing. I'm using a $150 DAC (Geek Out 450), and loving it. 

 

[billybob_jcv]

Seems to be a lot of debate in this thread when the OP was simply asking for opinions....


Agree that recording quality will influence identification of differences. 

Personally, I have both an ODAC and a QB-9 DSD.  The QB-9 DSD clearly is more resolving and dynamic in the bass region.  The QB-9 DSD also gives a more natural presentation of the music, comparatively, the ODAC sounds flat and is missing the nuances that the QB-9 DSD resolves.  When I listen to the ODAC, it initially sounds pretty good, yet it doesn't take long before what's missing (or extra?) makes me want to go back to the Ayre.

If you can make the opportunity, see if you can get a loan of a better DAC and discover for yourself if there is enough difference to be meaningful in your system.

Happy listening!

Considering that true DSD content is made from the best possible recordings available, and costs $25-$50/album for downloads, I certainly hope it sounds better...

 

[StanD]

 
Yeah, because as said, it's 400 Ohm... so it shouldn't have issues with a 5K load. But a DAC with more than 500 Ohm may actually have some issues.
 
And there is one such DAC where the DAC chip is connected to the RCA jacks here:
http://www.ebay.co.uk/itm/Non-Oversampling-DAC-TDA1543-DIR9001-96kHz-SPDIF-Coax-Optical-9V-NiMH-charger-/200858459050?rmvSB=true
 
(passive I/V) 

 
Attenuation is clipping out low-level signals. It's not like the amp is perfectly transparent and won't add its own noise... so when the signal that comes into the amp is as low as the noise floor, it'll just get lost. Attenuation will cause that to happen more often.
 
As for the Audio-GD devices, I meant they were not appropriate for the O2. The O2 would audibly clip when fed a signal greater than 2Vrms. Audio-GD DACs typically go up to about 2.5Vrms, I think, so that's more than the O2 can handle, and it'll clip. You may very well build an attenuator and mitigate the issue just fine, but then you're clipping out about 20% of the signal in that case (2V vs 2.5V). I don't think I'd choose that route.
 
Either way, I'm just trying to come to some explanation as to why some DACs may sound different, coming from an actual technical standpoint, and not just due to subjective impressions. Granted, this has no bearing on their actual pricing. I'm using a $150 DAC (Geek Out 450), and loving it. 

Bill, why are you looking for a lousy DAC on fleaBay that nobody should buy to try to sell a pointless point? Attenuation is reducing he signal level, clipping is hacking off the top of a waveform resulting in distortion that nobody here is interested in. Please stop looking for oddball fringe cases that are not relevant.
What are you going on about noise for? If one needs attenuation because of the oddball case you presented where the signal is too hot then after proper attenuation it will NOT be as low as the noise floor. Stop being silly.

 

[money4me247]

  Bill, why are you looking for a lousy DAC on fleaBay that nobody should buy to try to sell a pointless point? Attenuation is reducing he signal level, clipping is hacking off the top of a waveform resulting in distortion that nobody here is interested in. Please stop looking for oddball fringe cases that are not relevant.
What are you going on about noise for? If one needs attenuation because of the oddball case you presented where the signal is too hot then after proper attenuation it will NOT be as low as the noise floor. Stop being silly.

bill likes to diy and fiddle with his external components so he gets really into random options hahah.

 

[StanD]

  bill likes to diy and fiddle with his external components so he gets really into random options hahah.

This is more like silly stuff that I'd refrain from.

 

[Bill-P]

Sometimes I like to poke fun at stuffs. 

 
But yeah, I guess I'll stop since the rest of all that really belongs in Sound Science.
 
All I really wanted to say was that beyond the interaction of the headphone and the amplifier, there is interaction between the amplifier and the DAC. So it's good to know all the gears in the equation before proceeding.
 
Anyway, as for opinions, I currently have a Geek Out 450, and a Neko Audio D100 Mk.2 (this one is interesting, but I won't try to derail thread 

).
 
My subjective impressions are that... the 2 sound very audibly different.
 
The Neko DAC being... very smooth, silky, and very much devoid of noise or anything else like that. It truly sounds remarkably transparent but without being dry, grainy, textured, or... put another way, it's like music is a river that keeps flowing when I'm using it. Its lack of grain and textures is sometimes unrealistic as I have heard the same tracks on multiple other DACs and heard grain and texture. Not sure what is causing that to happen, but just to say... this DAC has the ability to essentially make any amp that is plugged into it sound like a smooth tube amp that's not warm, thick, or syrupy. Despite the overly smoothed sound, details are very apparent, and can sometimes be a bit unsettling because it's all so very smooth. Also it has a very unique way to present bass that truly brings out the "bass" in pretty much anything. Even a SR-009 sounded thumpy and weighty with this DAC.
 
In comparison, the Geek Out 450 is a fraction of the price, and gets to about 80-90% of the Neko, minus the silky smooth sound, that last bit of soundstage depth and detail extraction, and... "noise floor". I call it "noise floor", but I don't think it's just that, please go with it for now. What it really is is that... when I listen on the Geek Out 450, I get the "sound coming out of thin air" feeling every once in a while with the right track. Meanwhile, the Neko DAC does it a lot more often, and so individual sounds stand out in the space they are at. This is not the same as "separation", since both DACs can present distinct instruments with their unique tone quite well. I think detail extraction may be related to this, but I'm not sure. The GO450 is noticeably less detailed than the Neko. The GO450 is more "textured", or "grainy", sometimes, though, and I consider that to be more realistic.
 
But either way, close call. I made it sound like they are completely different, but I must admit... they are very very similar-sounding on a quick swap. The Neko is 10% better where it counts, but it really depends on the headphone I'm using. With the Sony MDR-Z7, I can barely hear that difference, if at all. With the Hifiman HE-560, I can hear the differences more consistently. And then with my Audeze LCD-2, I can hear the differences very plainly.
 
Anyway, I think the OP would be more interested in this last part:
Neko Audio D100 Mk.2 - $1400
LHLabs Geek Out 450 - $199
 
Sound-wise, very close, and one probably wouldn't notice in a noisy environment (at meets, etc...). It takes some hard concentration to notice.

So why did I say "very audibly different"? Because I like to poke fun at stuffs. 

 Okay... seriously, I'll stop.