[Shike]

Quote:

You still don't see it, do you? You are suggesting putting a volume control in front of the gain stage, to make up for the fact that the amp does not have a volume control in front of the gain stage.
 
This is flat out dumb, because it completely negates the intended effect, and actually makes things worse than if there was a single volume pot where most other DIY other amps put it.

Then don't buy a "flat out dumb" source that can't follow redbook due to poor design.  It's really that simple isn't it?  It's not the amp designer that needs to make sure your source is behaving itself.

 

[nikongod]

Quote:

Yes definitely.  Obviously it's not flat as noise (which is really low) dominates on the left side and performance degrades on the right side.
 
Click on the first article and search for:
THD+N vs OUTPUT & MAX POWER ON AC
and also
THD+N vs OUTPUT & MAX POWER ON BATTERY
 
It's purportedly well below 0.01% into any load tested, even on battery, just before distortion really ramps up on the right side near clipping.  It's kind of inconvenient not to be able to link relevant information, but sure...
 

Thanks for pointing it out. 
 
Are they available on a logarithmic voltage scale? I own a few headphones that have never seen more than 100mv, and I cant read that section of the graph. 

 

[The Monkey]

Quote:

 
No, I mean that the dissonance is imagined, as the 'portable' (battery) O2 measures slightly worse than the 'desktop' (AC) O2.
F.ex 1% THD+N into 33 ohm is reached at 613mW by AC, and at 547mW by battery.
There is no reason why Head-Fi'ers heads should start exploding; their dogma that stationary (in general) beats portable (in general) still holds true.

Would those differences be audible?  And, regardless, comparing the O2 in its "desktop" configuration to other desktops would be appropriate, correct?
 
FWIW, HeadFi'ers heads regularly explode for no good reason.

 

[Beefy]

Quote:

Then don't buy a "flat out dumb" source that can't follow redbook due to poor design.  It's really that simple isn't it?  It's not the amp designer that needs to make sure your source is behaving itself.

Nice try to deflect away from the fact that your passive volume control was a stupid idea, but I will play along, and go through this very simply......
 
Using a volume pot to control the volume is good. Misuse does not inherently cause clipping issues in most amps.
 
Using a gain switch as a backup to account for soft vs loud recordings, while still maintaining good volume pot travel for both, is good. Misuse does not inherently cause clipping issues in most amps.
 
Being forced to change the gain with different sources and recording levels to prevent clipping and maintain adequate volume is bad.
 
This is a considerable usability issue. No matter what the builder does, these amps will end up in the hands of people who aren't informed enough to understand why the amp sounds bad at high gain settings.

 

[Steve Eddy]

Quote:

 

FWIW, HeadFi'ers heads regularly explode for no good reason.

"I don' know wuh happun'. Alla sudden, he haid, it jes' 'sploded!" --Eyewitness interview
 
se
 
 
 

 

[Steve Eddy]

Quote:

This is a considerable usability issue. No matter what the builder does, these amps will end up in the hands of people who aren't informed enough to understand why the amp sounds bad at high gain settings.

 
So what? Why must a designer be shackled by the ill-informed?
 
se
 

 
 

 

[The Monkey]

Is this group defining anything greater than 2 Vrms (is that the right way to state it?) as being a "hot" source?

 

[Head Injury]

Quote:

Is this group defining anything greater than 2 Vrms (is that the right way to state it?) as being a "hot" source?

I believe we're defining it at about that for portable sources, and since 2.8Vrms is apparently clipping level at the default 2.5x gain on AC, that would be hot for desktop sources.
 
I'm no good at the fancy math. At 2.5x gain, what input would cause clipping on the 9V batteries?

 

[nikongod]

 
Quote:

So what? Why must a designer be shackled by the ill-informed?
 
se
 

 
I laughed. 
 
You roxor!

 

[The Monkey]

Quote:

I believe we're defining it at about that for portable sources, and since 2.8Vrms is apparently clipping level at the default 2.5x gain on AC, that would be hot for desktop sources.
 
I'm no good at the fancy math. At 2.5x gain, what input would cause clipping on the 9V batteries?

It's weird.  Taking a quick look at the specs of the various desktop DACs I've owned, they seem to be all over the place from 2.0 - 2.7.  Then again, half the time I don't know what the hell I'm reading.

 

[Head Injury]

Quote:

It's weird.  Taking a quick look at the specs of the various desktop DACs I've owned, they seem to be all over the place from 2.0 - 2.7.  Then again, half the time I don't know what the hell I'm reading.

Yes. In other words, with 2.5x gain there isn't a problem with clipping. At least on AC power.
 
You could probably say that 2.8Vrms is "too hot", and 2.3Vrms+ (which I think was the clipping level at 3.1x gain?) is "hot".

 

[Shike]

Quote:

Nice try to deflect away from the fact that your passive volume control was a stupid idea, but I will play along, and go through this very simply......

Nice try, but it isn't a stupid idea if it works.  You're going to need to provide evidence that the impacts are clearly audible first.  You can say "but this wouldn't be an issue", it still doesn't matter.  It wouldn't be an issue if the source makers developed their gear properly.  If a source shoved a crapload of DC voltage on the output and the amplifier outputted it whose fault is it?
 
 
Quote:

Using a volume pot to control the volume is good. Misuse does not inherently cause clipping issues in most amps.  
Using a gain switch as a backup to account for soft vs loud recordings, while still maintaining good volume pot travel for both, is good. Misuse does not inherently cause clipping issues in most amps.
 
Being forced to change the gain with different sources and recording levels to prevent clipping and maintain adequate volume is bad.
 
This is a considerable usability issue. No matter what the builder does, these amps will end up in the hands of people who aren't informed enough to understand why the amp sounds bad at high gain settings.

 
People misusing a product now means we should limit the design?  That's stupid.  Let's all use plastic knives because someone may cut themselves otherwise.
 
An apt comparison is a really nice sharp knife being used with a bamboo board losing its edge.  You can't say the knife is the problem when it's clearly not meant to be used with bamboo.  Sure enough someone is going to do it and complain about the knife when it's their own fault.  Most chefs would say the person using the board was incompetent and move on - the knife (amp) isn't the problem here.  The board (source) is.  The best thing is a disclaimer can be made, which it has, and move on.

 
Quote:

Is this group defining anything greater than 2 Vrms (is that the right way to state it?) as being a "hot" source?

 
Pretty much.  It's just that some hot sources are closer to normal like the HRT streamer.  2.7V like the Audio-GD is no where near redbook standard.  IMO the smallest gain should be 2.5x, that way if the voltage is 2.5 it's safe.  Anything above that is closer to 3V anyway and is clearly disregarding standards being fringe cases.
 
 
@Head Injury
 
1.4V would cause clipping on low batteries - so a 2.5x gain is definitely safe for most portables - which tend to have volume control anyway so . . .

 

[Beefy]

 
Quote:

Why must a designer be shackled by the ill-informed?

 
Because that is how products gain mass-market acceptance. Esoteric products that require OCD attention are why most of the world think audiophiles are pathetic.
 
Following this, it really comes down to a question of why and when a gain switch should be used. In my opinion, it should only ever be need to be increased when you need more volume pot travel, or decreased when you have very little volume pot travel. Forcing users to switch gain depending on what source they have - in addition to the typical use case - is a defect of usability.

 

[Beefy]

 
Quote:

Nice try, but it isn't a stupid idea if it works.  You're going to need to provide evidence that the impacts are clearly audible first.

The designer put the volume pot after the gain stage to improve SNR, right?
 
So putting another volume pot before the gain stage would decreases SNR, right?
 
How can you not see the gaping holes in your own logic?

 

[Head Injury]

Quote:

The best thing is a disclaimer can be made, which it has, and move on.

Yep. Like I said, the problem would be solved with a warning at the top of the schematic which reads something like "DON'T SET GAIN TO X IF SOURCE OUTPUTS Y!" Then if someone does set the gain too high, it's nothing but their fault. They didn't read the instructions. It would be like if someone glued some tubes in where opamps are supposed to be. The design uses opamps, so it's your fault if you use anything else. The design uses a certain gain, so it's your fault if you set it too high.
 
Quote:

Because that is how products gain mass-market acceptance. Esoteric products that require OCD attention are why most of the world think audiophiles are pathetic.

It requires OCD attention to not mess with the specified gain?
 
 

The designer put the volume pot after the gain stage to improve SNR, right?
 
So putting another volume pot before the gain stage would decreases SNR, right?
 
How can you not see the gaping holes in your own logic?

Weren't you the one to suggest moving the volume pot to eliminate the issue, reducing SNR in the process? Shike's idea is the same in concept, but not permanent. If your source isn't too hot, you don't need the extra volume control. Your idea would reduce SNR for all sources, not just the troublesome ones.
 
Here's a wild thought, but can we discuss some other aspect of the design now? Or is user error the only actual flaw?