Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › O2 AMP + ODAC
New Posts  All Forums:Forum Nav:

O2 AMP + ODAC - Page 77

post #1141 of 3780

That's cheap multimeters.  Certainly people test signals outside of mains frequencies, when using multimeters in a lab or something more serious.  But if in doubt of your multimeter's capabilities, use 60 Hz if you can, because even the cheapest ones should be fine with that.  It's the most basic expectation and usage for a user.

 

If you're referencing O2 and Clip+ crosstalk measurements, note the load impedance, as it can well be different for different loads.  The crosstalk also may vary with frequency.  For that matter, -50 dB is undesirable but not a lot at all (would be considered super way overkill for most speakers playback, for sure), particularly if most of the sounds in a recording are amplitude panned so most sounds bleed L/R much much more than -50 dB and the extra amount from the hardware would be trivial.  Again, be careful of quoting certain specific measurements as the explanation of perceived differences.

 

You can make an ABX box if you really want.  There could be some that are sold out there too.

post #1142 of 3780
Quote:
Originally Posted by mikeaj View Post

That's cheap multimeters.  Certainly people test signals outside of mains frequencies, when using multimeters in a lab or something more serious.  But if in doubt of your multimeter's capabilities, use 60 Hz if you can, because even the cheapest ones should be fine with that.  It's the most basic expectation and usage for a user.

 

If you're referencing O2 and Clip+ crosstalk measurements, note the load impedance, as it can well be different for different loads.  The crosstalk also may vary with frequency.  For that matter, -50 dB is undesirable but not a lot at all (would be considered super way overkill for most speakers playback, for sure), particularly if most of the sounds in a recording are amplitude panned so most sounds bleed L/R much much more than -50 dB and the extra amount from the hardware would be trivial.  Again, be careful of quoting certain specific measurements as the explanation of perceived differences.

 

You can make an ABX box if you really want.  There could be some that are sold out there too.

I've asked this in another thread but the original poster hasn't had time to answer it. How does one read and interpret crosstalk graphs and measurements?

The C5 for example: Crosstalk @ 150 Ω -67 dB

^ I have no idea what that actually means, and various RMAA measurements have a graph to go with this measurement.

post #1143 of 3780

Quote:
Originally Posted by miceblue View Post

I've asked this in another thread but the original poster hasn't had time to answer it. How does one read and interpret crosstalk graphs and measurements?

The C5 for example: Crosstalk @ 150 Ω -67 dB

^ I have no idea what that actually means, and various RMAA measurements have a graph to go with this measurement.

 

More or less, play a tone into one channel, measure the output into that channel (A) and then also into the other one (B).  B is undesired; ideally it would be 0.  Check out the ratio of B/A.  20*log10(B/A) is the quoted result in dB, if A and B were voltage measurements.  It's negative because B is a lot less than A.  

 

A graph might show values for different frequencies.  Or I guess x axis could be something else.  If somebody gives a single value and doesn't specify frequency, it's probably 1 kHz because in audio land, that's what most people do by convention.  Just like THD measurements—usually at 1 kHz just because that's common practice, not that it particularly means anything special.

 

Result will depend on frequency tested, load impedance, and potentially even some other factors like position of volume control.  You have to think about the ways in which signals could be mixed or coupled inside an amp.

 

When driving 150 ohms, O2 gets -91 dB supposedly.  It's -65 dB for 15 ohms, -72 dB for 33 ohms, -91 dB for 150 ohms, -95 dB for 600 ohms.  Even if you hook up the leads correctly, RMAA gives some weird results for crosstalk sometimes.  I don't really know what's up.  The software is pretty buggy, or it also could be a hardware grounding issue or something else that causes the weird results.  I wouldn't take RMAA figures too seriously, anyway.

 

For that matter, practically I don't really think -91 dB is an improvement over -67 dB when you're listening to anything, particularly if it's not all just supposed to be in one ear.  Actual music in each ear masks any of these effects, even if you could actually discern a difference when testing with a tone in one channel and silence in the other.

 

 

So what I described at the top is the mechanism by which the measurement is taken (at least conceptually; I don't know the actual procedure for sure but that would produce the desired result... they could do sweeps or something else maybe).  The result should apply for any signal, including music.


Edited by mikeaj - 3/24/13 at 2:48pm
post #1144 of 3780
Quote:
Originally Posted by mikeaj View Post


Quote:
Originally Posted by miceblue View Post

I've asked this in another thread but the original poster hasn't had time to answer it. How does one read and interpret crosstalk graphs and measurements?

The C5 for example: Crosstalk @ 150 Ω -67 dB

^ I have no idea what that actually means, and various RMAA measurements have a graph to go with this measurement.

 

More or less, play a tone into one channel, measure the output into that channel (A) and then also into the other one (B).  B is undesired; ideally it would be 0.  Check out the ratio of B/A.  20*log10(B/A) is the quoted result in dB, if A and B were voltage measurements.  It's negative because B is a lot less than A.  

 

A graph might show values for different frequencies.  Or I guess x axis could be something else.  If somebody gives a single value and doesn't specify frequency, it's probably 1 kHz because in audio land, that's what most people do by convention.  Just like THD measurements—usually at 1 kHz just because that's common practice, not that it particularly means anything special.

 

Result will depend on frequency tested, load impedance, and potentially even some other factors like position of volume control.  You have to think about the ways in which signals could be mixed or coupled inside an amp.

 

When driving 150 ohms, O2 gets -91 dB supposedly.  It's -65 dB for 15 ohms, -72 dB for 33 ohms, -91 dB for 150 ohms, -95 dB for 600 ohms.  Even if you hook up the leads correctly, RMAA gives some weird results for crosstalk sometimes.  I don't really know what's up.  The software is pretty buggy, or it also could be a hardware grounding issue or something else that causes the weird results.  I wouldn't take RMAA figures too seriously, anyway.

 

For that matter, practically I don't really think -91 dB is an improvement over -67 dB when you're listening to anything, particularly if it's not all just supposed to be in one ear.  Actual music in each ear masks any of these effects, even if you could actually discern a difference when testing with a tone in one channel and silence in the other.

 

 

So what I described at the top is the mechanism by which the measurement is taken (at least conceptually; I don't know the actual procedure for sure but that would produce the desired result... they could do sweeps or something else maybe).  The result should apply for any signal, including music.

Well there goes that theory for the perceived soundstage difference I hear between the O2 and the C5. >_<

 

Anywho, thanks for explaining that. I still don't quite understand what the number physically means though. -91 dB meaning...the crosstalk is pretty much inaudible? What is considered a "bad" crosstalk value? Likewise for the signal-to-noise ratio measurement. 110 dB meaning.....the signal power is much larger than the noise power? What is considered a "bad" SNR value?

 

 

 

Quote:
Originally Posted by Lord Voldemort View Post

Haven't been on head-fi for a while... 960 posts since I left this thread. Time to get out my reading glasses.  

If Voldemort is mentioned in a post, it's most likely referring to the designer of the O2 (AKA He-Who-Must-Not-Be-Named), not you.


Edited by miceblue - 3/24/13 at 4:18pm
post #1145 of 3780

There's bad relative to what you can hear and bad relative to other devices.  It's going to depend on a number of things, same for SNR.

 

Also, when looking at SNR, keep in mind that most recordings have noise from the recording process, already in them.  There's also already acoustic noise in your listening environment.  So past a certain point it doesn't make much sense to obsess about noise from the electronics, if other noise is at a significantly higher level.

 

If you want, I can post a few tracks with slightly mixed L and R channels to simulate different levels of crosstalk (say, not today but maybe tomorrow; my upload speeds at home leave a lot to be desired).

 

 

Anyway, this is kind of OT, but I think important.  Don't get too caught up in reading numbers without a frame of reference for them, pretty much.

post #1146 of 3780
Quote:
Originally Posted by mikeaj View Post

There's bad relative to what you can hear and bad relative to other devices.  It's going to depend on a number of things, same for SNR.

 

Also, when looking at SNR, keep in mind that most recordings have noise from the recording process, already in them.  There's also already acoustic noise in your listening environment.  So past a certain point it doesn't make much sense to obsess about noise from the electronics, if other noise is at a significantly higher level.

 

If you want, I can post a few tracks with slightly mixed L and R channels to simulate different levels of crosstalk (say, not today but maybe tomorrow; my upload speeds at home leave a lot to be desired).

 

 

Anyway, this is kind of OT, but I think important.  Don't get too caught up in reading numbers without a frame of reference for them, pretty much.

That is definitely a good point. I didn't even think about that!

 

I don't think it's completely off-topic. A whole lot of thought and numbers went into the design of the O2. It's only natural to wonder if these numbers are actually meaningful for someone without an electrical background, and how it affects the end sound we hear, at least it is to me.

post #1147 of 3780
Quote:
Originally Posted by mikeaj View Post


If you're referencing O2 and Clip+ crosstalk measurements, note the load impedance, as it can well be different for different loads.  The crosstalk also may vary with frequency.  For that matter, -50 dB is undesirable but not a lot at all (would be considered super way overkill for most speakers playback, for sure), particularly if most of the sounds in a recording are amplitude panned so most sounds bleed L/R much much more than -50 dB and the extra amount from the hardware would be trivial.  Again, be careful of quoting certain specific measurements as the explanation of perceived differences.

Yes, sir!

In anycase, it's the only measurement that explains the significant difference in stereo that I'm hearing. I'm not saying that the clip+ has little stereo, so while -65db versus -50db difference may not be big, it may still explain the audible loss of stereo I'm experiencing.
post #1148 of 3780
Quote:
Originally Posted by miceblue View Post

If Voldemort is mentioned in a post, it's most likely referring to the designer of the O2 (AKA He-Who-Must-Not-Be-Named), not you.

Oh yeah definitely. The way I got this name in the first place was because the Nwavguy had a reputation for this name. I thought it'd be fun to take it and heck, it was available. I'm a big Harry Potter fan as well. And I did start this thread about a year ago and it's interesting what people have been discussing here. I've learned quite a bit. 

post #1149 of 3780

re: crosstalk

 

 

I created a new thread with a few audio files with different levels of simulated crosstalk.  For best effect, I guess you should use an amp with very low crosstalk of its own.  Details in the link:

 

http://www.head-fi.org/t/657280/simulated-crosstalk-listening-example-audio-files-inside

post #1150 of 3780

I am very happy with the ODAC and O2 (as two separates).

 

Should I buy a better quality AUDIOQUEST - FOREST USB MINI CABLE ($29)

 

 

or this http://www.walmart.com/ip/Monster-Digital-Life-High-Speed-Mini-USB-Cable-0.5/17490789  ???


Edited by JakeJack_2008 - 3/27/13 at 3:39pm
post #1151 of 3780

Forest.

post #1152 of 3780
Quote:
Originally Posted by JakeJack_2008 View Post

I am very happy with the ODAC and O2 (as two separates).

 

Should I buy a better quality AUDIOQUEST - FOREST USB MINI CABLE ($29)

 

 

or this http://www.walmart.com/ip/Monster-Digital-Life-High-Speed-Mini-USB-Cable-0.5/17490789  ???

 

Unless one cable or the other is defective, there will be zero audible difference. If one is defective, you'll hear genuine noise when using it.

post #1153 of 3780
Quote:
Originally Posted by HamilcarBarca View Post

Unless one cable or the other is defective, there will be zero audible difference. If one is defective, you'll hear genuine noise when using it.
This man speaks the truth. I'll give you analog cables, but with digital, you won't get any upgrade from the $5 Walmartspecial unless you are trying to run more than 15-20 ft
post #1154 of 3780

Dont waste your money.....what they said....the Walmart cable will do just fine unless its defective.....

 

Alex

post #1155 of 3780

Thanx to all of You. So, no Forest, Monster, ... USB cables.

 

However, I've a little problem. The USB cable which came with the ODAC -

it's also posted among cables on the JDS Labs site - is way too long.

So, this fact triggered my search for shorter (and perhaps better quality)

USB cables. Now, in Staples there's a Monster USB Mini and Micro cable of which length

is only 6 inches (and costs $20).

This is the same Monster cable which  I posted above (1.5 feet=45 cm long for $25).

 

Can you recommend any shorter (and cheaper) USB cables than the 6" Monster cable ($20) for the ODAC to connect to my laptop.

Also, is a 6" USB cable a good idea? While using the 6" cable, the ODAC and also the O2 (as two separates) must be very close to my laptop's EMI/RFI?  Does it matter?


Edited by JakeJack_2008 - 3/28/13 at 9:37am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Headphone Amps (full-size)
Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › O2 AMP + ODAC