O2 AMP + ODAC
May 8, 2014 at 5:05 PM Post #2,836 of 5,671
Hi miceblue,

the Vmoda is 32 ohms and that's most probably way too low for the macbook. The usual rule of thumb is that the headphone impedance should be 8-10 times or higher than the output impedance of the amp. That would mean the output should have 3-4 ohms or less and apple seems to state less than 24 ohms...

http://apple.stackexchange.com/questions/57186/what-is-the-impedance-of-the-line-headphone-jack-on-macbook-pro-retina

So the sound of the Vmoda is probably a bit colored when directly connected to the macbook and will sound different when when the macbook just drives the O2 with >10k ohms input impedance.

Joachim

Yeah that may very well be the case. Is there a way to measure the output impedance with things lying round the house? That stackexchange link didn't give a specific value. I'm inclined to think Apple updated their specs since the Retina Display line since there's a pretty substantial difference in sound between that and my 2008 MacBook.

The O2's output impedance is about 0.5 Ω according to the blog.



I cant even belive this is happening......OMG...

This is really well.....you know.....

Have fun....

A.

:deadhorse:

Thank you for the constructive feedback and explanations, I feel edified now.
 
May 8, 2014 at 7:01 PM Post #2,837 of 5,671
If you want some good advice get an amp that will play your tunes at the volumes you desire instead of double amping.....
 
Save your self time and effort...
 
But if you persist you can easily measure impedance if you have an oscilloscope or ac voltmeter, a signal generator, a variable resistance load or decade box. You need to understand the difference between DC resistance and impedance....so if you have these items "lying around the house" have at it....
 
A.
 
May 8, 2014 at 7:37 PM Post #2,838 of 5,671
I wasn't using maximum volume from the MacBook to the O2, just volume 1/16 for example, so the VRMS value going into the O2 and the volume going to the headphone shouldn't be a problem at all with the O2 at unity gain.
1/16 volume from MacBook -> 1.0x gain/no gain from O2 at maximum volume = 1/16 volume to ears assuming the O2 is just a "wire with gain"
 

 
The signal itself may be low but the amplification further down the chain will only amplify the noise and raise the noise floor.  Its very dangerous to do that as any noise introduced into the signal after the macbook will play out your headphones unattenuated. For example have you ever heard a pop or rustling sound when plugging/unplugging an interconnect?   Don't ever try that with the O2 on full volume if you care about your hearing or headphones.  I also would'nt trust the macs s/w volume control not to revert to full volume and pop something.
 
Quote:
As for someone actually doing this in reality, it's definitely a legitimate situation. I like the sound out of ___ system, but the volume is too quiet and I would like to make it louder. A "wire with gain" would provide exactly this, no? And say it's a computer soundcard (MacBook in this example), you can't actually separate the DAC and amp components.

 
No easy way around that other than using a DAC with a toslink or USB connection from the Macbook.
 
May 8, 2014 at 9:16 PM Post #2,839 of 5,671
  I had a question about the ODAC. How does the ES9023 chip objectively perform compared to the ES9018 chip? I know the latter is much more expensive and is used in high end DACs. Any thoughts?

It really depends on the implementation. ODAC sounds as good as NFB11.32 with ES9018.
 
May 9, 2014 at 6:33 AM Post #2,840 of 5,671
  Whitecitadel...are your sure u need 3x gain? What cans are you going to drive?
 
Alex

 
So I have low=1x, high=3x, I am using Sennheiser Momentums which on paper are super-easy to drive. The 1x is for the Momentums and the odac as the source, the 3x was just some future proofing and for using 0.5v LOD from an iPod etc.
 
I don't anticipate using 3x really, it was just a decision on what to build it with really...
 
May 9, 2014 at 6:43 AM Post #2,841 of 5,671
So I have low=1x, high=3x, I am using Sennheiser Momentums which on paper are super-easy to drive. The 1x is for the Momentums and the odac as the source, the 3x was just some future proofing and for using 0.5v LOD from an iPod etc.

I don't anticipate using 3x really, it was just a decision on what to build it with really...


That's the same spec as I use and 1x is enough to drive my HD650's with most albums.
 
May 9, 2014 at 9:38 AM Post #2,842 of 5,671
   
So I have low=1x, high=3x, I am using Sennheiser Momentums which on paper are super-easy to drive. The 1x is for the Momentums and the odac as the source, the 3x was just some future proofing and for using 0.5v LOD from an iPod etc.
 
I don't anticipate using 3x really, it was just a decision on what to build it with really...

 
If your building it yourself buy the resistors for the various gains (cheap) and a terminal strip to chop up and use as sockets.
 

 
May 9, 2014 at 9:41 AM Post #2,843 of 5,671


Quote: "Ultimately, the ES9023 properly implemented, is an audibly transparent DAC. The window glass is completely clean. There are no perceptible flaws. It's not possible to get the glass perceptibly cleaner. While the ES9018 or ES9012 might even better measurements, I honestly don't believe they can improve on the sound quality when listening blind."
 


I can't say I have heard both, but if it helps the ODAC designer wrote on his blog in the comments for the post "odac-may-update":
 
HTH

Ok let me get this straight, He Who Must Not be Named  "believes" the ES9018 or ES9012 can't improve the sound quality even though they measure better! Surely that statement goes against everything I thought he stands for.
 
May 9, 2014 at 9:49 AM Post #2,844 of 5,671
Ok let me get this straight, He Who Must Not be Named  "believes" the ES9018 or ES9012 can't improve the sound quality even though they measure better! Surely that statement goes against everything I thought he stands for.


Once you get past the point of transparency it could measure a million times better but our ears wouldn't know or care. That's the point. He believed in doing enough to bet the job done .
 
May 9, 2014 at 9:50 AM Post #2,845 of 5,671
Actually it doesn't. We're at a level where the limiting factor is no longer the abstract chip specs but implementation, even when well executed becomes the bottleneck and more relevant than chipset spec differences.
 
May 9, 2014 at 9:52 AM Post #2,846 of 5,671
  Ok let me get this straight, He Who Must Not be Named  "believes" the ES9018 or ES9012 can't improve the sound quality even though they measure better! Surely that statement goes against everything I thought he stands for.

 
Yeah that's right, the flag ship chip is no better than the bargain basement entry level chip
normal_smile .gif

 
May 9, 2014 at 10:02 AM Post #2,848 of 5,671
  The head'n'hifi kit comes with all the values and the SIP sockets, decided against the sockets and just soldered in the 3x and left out the others for 1x for now. 
 
Nice neat job by the way :D

 
Oh didn't know that, i only bought the PCB from them.
 
May 9, 2014 at 10:08 AM Post #2,850 of 5,671
  Ok let me get this straight, He Who Must Not be Named  "believes" the ES9018 or ES9012 can't improve the sound quality even though they measure better! Surely that statement goes against everything I thought he stands for.

 
According to manual my wife's car is good for about 130MPH and my car 150MPH, this makes my car the "better" car, as it measures with higher performance.
 
The performance of either in the real world is limited to the traffic flow on the local motorway (freeway) of about 70MPH, so using the one that measures better in day to day use makes no discernable difference from the one that measures in excess of the required performance.
 
If you were go to a racetrack and measure performance, a bad implementation (aka driver in my  analogy) could make the one with lower specs perform better...
 

Users who are viewing this thread

Back
Top