[samsquanch]

Quote:

That line of thought suggests you're going to own the same set of headphones for the rest of your life, and considering in this hypothetical situation, you've just favoured the sound of an amp that significantly changed their signature - it's pretty unlikely that is going to be the case.
 
If you buy amp A now because it rolls of the bass of headphone C and you like the sound better - it's also going to roll off the bass of headphone X - which WOULD have been your perfect headphone had you not bought a rubbish amp that was now masking its potential.

this would imply that people on this site AREN'T stock piling amps and headphones like they were food before a nuclear holocaust.

 

[bcg27]

Quote:

this would imply that people on this site AREN'T stock piling amps and headphones like they were food before a nuclear holocaust.

 Lol, very true

 

[Mr.Sneis]

I think I'm going to try to build this.

 

[Hennyo]

Quote:

this would imply that people on this site AREN'T stock piling amps and headphones like they were food before a nuclear holocaust.

Lol, x2. But stockpiling food here too. =d

 

[LizardKing1]

Quote:

That line of thought suggests you're going to own the same set of headphones for the rest of your life, and considering in this hypothetical situation, you've just favoured the sound of an amp that significantly changed their signature - it's pretty unlikely that is going to be the case.
 
If you buy amp A now because it rolls of the bass of headphone C and you like the sound better - it's also going to roll off the bass of headphone X - which WOULD have been your perfect headphone had you not bought a rubbish amp that was now masking its potential.

Well I do prefer clear amps, and believe you should only use the last link in your chain to change the sound signature, i.e. the headphones. If you're not up to having a different headphone for every sound signature, which is perfectly rational, a decent EQ is also good. However, the fact that I think this way doesn't mean that someone else will too and that necessarily they're wrong. We all here share the opinion that the O2 is a really good amp, but if a subjectivist tries it and it doesn't go along with his taste for bass or something, will he be wrong to prefer something warmer? It just didn't appeal to him. I think someone would have to be very influenciable to buy something they know they don't enjoy. Of course if he later tries that same amp, and now the warm gets bloated, sucks to be him, and that's why wires with gain work better than synergy ^^ but then I guess the O2 is only good for people who think ahead, right?

 

[mikeaj]

Quote:

There are published specs for the B22 here: http://www.amb.org/audio/beta22/, under the Specifications tab.

 
An issue is that many people have called into question those RMAA specs.  Obviously it's RMAA and limited by the input/output interfaces as well as the software itself.  But, for example, the crosstalk results don't make sense (not that crosstalk is that important IMHO, just an example).  Multiple people have gotten crosstalk results 10s of dB worse than what is shown, in some AMB designs. 
 
To me the simplest explanation is that the results were done with impedances much higher than 33 ohms and 330 ohms as claimed.  i.e. something wrong in the test setup...partially because the 3 channel topology is an issue and something was not done right?  It's hard to say.  If it were true that impedances higher than 33 ohms and 330 ohms were effectively used, then that would impact all the other results too.  If higher impedances are used, we would expect better measured performance than in reality.
 
Hence that's why some independent tests would be nice.  I'd like independent tests for almost everything though--no need to single out AMB except that we were discussing the Beta22 as a potential reference amp.

 

[EddieE]

I'm not suggesting anyone should either, I'm just saying that you shouldn't buy an amp based on how it makes one set of headphones sound. An amp that does its job will not impart any sound of the headphones - if you find you don't like the sound of a set of headphones on such an amp - you don't like the headphones, not the amp.

 

[Pars]

Quote:

 
An issue is that many people have called into question those RMAA specs.  Obviously it's RMAA and limited by the input/output interfaces as well as the software itself.  But, for example, the crosstalk results don't make sense (not that crosstalk is that important IMHO, just an example).  Multiple people have gotten crosstalk results 10s of dB worse than what is shown, in some AMB designs. 
 
To me the simplest explanation is that the results were done with impedances much higher than 33 ohms and 330 ohms as claimed.  i.e. something wrong in the test setup...partially because the 3 channel topology is an issue and something was not done right?  It's hard to say.  If it were true that impedances higher than 33 ohms and 330 ohms were effectively used, then that would impact all the other results too.  If higher impedances are used, we would expect better measured performance than in reality.
 
Hence that's why some independent tests would be nice.  I'd like independent tests for almost everything though--no need to single out AMB except that we were discussing the Beta22 as a potential reference amp.

I realize that RMAA can give somewhat optimistic results, but it is not wholly useless as a test tool. The RMAA results that Amb provided are clearly labeled with the load that they were run at, so I have no reason to doubt that... not sure where you are coming up with questioning that.
 
I'm not a fan of 3-channel topologies, but since the B22 does not use a rail splitter (unlike the mini-3 and M^3), I'm not sure why you would find an issue with that? It's just a buffered ground for the audio signal.
 
 

 

[EddieE]

Quote:

Well I do prefer clear amps, and believe you should only use the last link in your chain to change the sound signature, i.e. the headphones. If you're not up to having a different headphone for every sound signature, which is perfectly rational, a decent EQ is also good. However, the fact that I think this way doesn't mean that someone else will too and that necessarily they're wrong. We all here share the opinion that the O2 is a really good amp, but if a subjectivist tries it and it doesn't go along with his taste for bass or something, will he be wrong to prefer something warmer? It just didn't appeal to him. I think someone would have to be very influenciable to buy something they know they don't enjoy. Of course if he later tries that same amp, and now the warm gets bloated, sucks to be him, and that's why wires with gain work better than synergy ^^ but then I guess the O2 is only good for people who think ahead, right?

Sure each to their own and all, if someone wants to buy a messed up amp rather than use an EQ or get new headphones, it doesn't affect me.
 
I just think the whole "synergy" mind set and the belief that components like amplifiers and DACs and even cables should have "characters" leads people to spend an inordinate amount of money over the years chasing a perfection, being sold by a glowing subjective poetic-language review of every new fad.
 
If we collectively got in the mind set of demanding measurements from amp and DAC manufacturers, or only taking reviews that included them seriously, then it would be good for absolutely everyone who deserved it to be good - those making quality products and us buying them. 
 
We can still be as poetic and wax lyrical as much as we like about headphones and music - but the other components are about science enabling beauty not creating it.

 

[LizardKing1]

Quote:

Sure each to their own and all, if someone wants to buy a messed up amp rather than use an EQ or get new headphones, it doesn't affect me.
 
I just think the whole "synergy" mind set and the belief that components like amplifiers and DACs and even cables should have "characters" leads people to spend an inordinate amount of money over the years chasing a perfection, being sold by a glowing subjective poetic-language review of every new fad.
 
If we collectively got in the mind set of demanding measurements from amp and DAC manufacturers, or only taking reviews that included them seriously, then it would be good for absolutely everyone who deserved it to be good - those making quality products and us buying them. 
 
We can still be as poetic and wax lyrical as much as we like about headphones and music - but the other components are about science enabling beauty not creating it.

 
I truly agree. When I think of synergy I think of people buying 2,000$ cables to get a slightly warmer sound.

 

[Head Injury]

When I think of synergy I think of people buying $2,000 cables to get the same sound 

 

[MisterX]

 
Quote:

An issue is that many people have called into question those RMAA specs. Obviously it's RMAA and limited by the input/output interfaces as well as the software itself. But, for example, the crosstalk results don't make sense (not that crosstalk is that important IMHO, just an example). Multiple people have gotten crosstalk results 10s of dB worse than what is shown, in some AMB designs.

 
 
Are these people short circuiting the ground channel when running the test?
 
 
Here are summary results I got from the other day for a High performance Mini3 configured with the gain set to +5:
 
 

 
Here are summary of the results of the same amplifier with the ground channel short circuited:
 
 

(click to expand the images)
 
Idle current of the amplifier before running the test with the shorted ground channel was 24.3mA, 26.024mA after.
Average current draw when playing sweeping test tones into my HD580s at twice the normal listening level was 28mA before the test, 36mA after.
DC offset of the ground channel was 0 volts before the test, +6mV after the test.
A quick A/B listening test against another similar amplifier and it was obvious...
 Not only does shorting out the ground channel predictably effect the RMAA results it damages the ground op-amp.
 
Meh....I had more for you but FF crashed when I clicked preview and I don't feel like retyping it so all you get is what was in my clipboard.
For now at least 
 
 
 
 
 
 

 

[mikeaj]

edit: didn't see above when posted, will read
 
edit2: Thanks for the post.  I'm not other people, so maybe they shorted the ground channel when testing.  I'm in no position to ask people for anything, but do you have results for a lower impedance, closer to 33 ohms?  What exactly does your setup look like?  How exactly was the amp connected to the load (all three channels) and where were the input leads connected?  I didn't search everywhere, but I did not find satisfactory answers to these before, so I'd be glad if you could fill me in.
 
edit3:  So I think the thing to clarify is:  how much current was going out L and R channels and how much current into the ground channel, while testing?

 

[estreeter]

Quote:

I think I'm going to try to build this.

Wild about your optimism, Mr Sneis - lets hear something along the lines of 'I *will* build this amp, or die trying !' 

 

[Malvin]

 
Quote:

 
An issue is that many people have called into question those RMAA specs.  Obviously it's RMAA and limited by the input/output interfaces as well as the software itself.  But, for example, the crosstalk results don't make sense (not that crosstalk is that important IMHO, just an example).  Multiple people have gotten crosstalk results 10s of dB worse than what is shown, in some AMB designs. 
 
To me the simplest explanation is that the results were done with impedances much higher than 33 ohms and 330 ohms as claimed.  i.e. something wrong in the test setup...partially because the 3 channel topology is an issue and something was not done right?  It's hard to say.  If it were true that impedances higher than 33 ohms and 330 ohms were effectively used, then that would impact all the other results too.  If higher impedances are used, we would expect better measured performance than in reality.
 
Hence that's why some independent tests would be nice.  I'd like independent tests for almost everything though--no need to single out AMB except that we were discussing the Beta22 as a potential reference amp.

 
 
Ok, let me share mine..
 
Just not to get into this 2ch vs 3ch debate let's take a passive-ground amp: CK2III (I have one in my headamp collection). Test setup should be straightforward in this case.
 
My CKK is a pretty standard (and economical) build, with a power transformer in a separate enclosure. All parts are as per AMB's site, the only exception is output transistors which are Toshiba.
 
Measurement tool: Sony Vaio laptop + EMU 0404 USB (unmodified) + RMAA 6.2.3, dummy load 33/330 ohm
 
Here is what I have:


 
 
if you doubt whether 33 ohm load was there, look at the distortion numbers.. they are higher for 33 ohm load vs 330 ohm, however crosstalk is still above what is referred to as theorethically possible.
 
(if anyone insists I can even post pictures of my setup with a multimeter connected to the dummy load..)
 
In any case.. I see that AMB is criticized a lot based on the fact that specifications he provides are beyond theorethical limits. This argument is used further to question credability of his designs in general.
 
IMHO it is quite possible that some specialized and expensive equipment would produce totally different results, and crosstalk numbers would be much worse.
 
But we need to compare apples to apples, right?
 
Based on what I posted above I don't see that anything was faked or made up by AMB - my RMAA results are in line with the specs posted on his site.