Review and Measurements of Topping A30 and Schiit Magni 3 Headphone Amps

[Hemicrusher]

These are used for DC servo.

And for those that did not know.
Most discrete designs are just op-amp circuits (inverting and non inverting inputs followed by gain stages and output stages) but made with discrete components.

Consider the fact that in opamps exotic components can be made (transistors with multiple emitters etc.) that perform better and are easier to 'match' than discrete components and do not exist in discrete components.

Makes sense......

Thanks!

 

[έχω δίκιο]

Fail.....

Yes, you failed at written English.

Saying I like amps without opamps in no way indicates that a prefer them.

What you wrote was "I actually like that the Magni 3 does not use an opamp." That indicates a preference for amps that do not use op-amps.

If that wasn't your intention, you could have written "I actually like that the Magni 3 does not use an op-amp but I would like it just as much if it did use an op-amp because I have no preference about the use of op-amps in amps -- which makes me wonder why I ever mentioned anything about op-amps at all because it makes no sense if I'm not trying to express a preference."

 

[Hemicrusher]

Yes, you failed at written English.

What you wrote was "I actually like that the Magni 3 does not use an opamp." That indicates a preference for amps that do not use op-amps.

If that wasn't your intention, you could have written "I actually like that the Magni 3 does not use an op-amp but I would like it just as much if it did use an op-amp because I have no preference about the use of op-amps in amps -- which makes me wonder why I ever mentioned anything about op-amps at all because it makes no sense if I'm not trying to express a preference."

You are adding intent to my statement. That is where you failed.

I even clarified my statement in a reply to sssn before you even started trolling me.....I said, " So, if I say I like vanilla ice cream, don't assume I hate chocolate. "

 

[έχω δίκιο]

You are adding intent to my statement. That is where you failed.

No, I am reading your statement as written.

I even clarified my statement in a reply to sssn before you even started trolling me.....

You are "adding intent" to my posts. Trolls are looking for a response, and I wanted none from you.

I said, " So, if I say I like vanilla ice cream, don't assume I hate chocolate. "

I said that you expressed a preference when you wrote "I actually like that the Magni 3 does not use an opamp." If you did not intend to express a preference, then that statement makes no sense at all.

 

[Thomas savage]

Yes, you failed at written English.

What you wrote was "I actually like that the Magni 3 does not use an opamp." That indicates a preference for amps that do not use op-amps.

If that wasn't your intention, you could have written "I actually like that the Magni 3 does not use an op-amp but I would like it just as much if it did use an op-amp because I have no preference about the use of op-amps in amps -- which makes me wonder why I ever mentioned anything about op-amps at all because it makes no sense if I'm not trying to express a preference."

He could just as easily be expressing simply that he likes the point of difference, in this case lack of opamp just because it’s a intresting implementation.

Let’s not get hung up on semantics, move on.

 

[Hemicrusher]

He could just as easily be expressing simply that he likes the point of difference, in this case lack of opamp just because it’s a intresting implementation.

Let’s not get hung up on semantics, move on.

Exactly......

I am done with this.

 

[mindbomb]

@amirm

Can you double check the magni 3 measurements? Its performance in the thd+n vs power 300 ohms and 33 ohms seems much better than I was expecting.

 

[amirm]

@amirm

Can you double check the magni 3 measurements? Its performance in the thd+n vs power 300 ohms and 33 ohms seems much better than I was expecting.

Gosh, i just tore all that down for the new review I did. And folks are screaming loud to review new gear. I will see if I can retest....

 

[Zilfallion]

Gosh, i just tore all that down for the new review I did. And folks are screaming loud to review new gear. I will see if I can retest....

Did you ever get around to testing the Modi 2's output level?

 

[Dana reed]

Thanks for the review, @amirm! By the way, the reason the Magni 3 and Modi 2 have different finishes is that the Magni 2 had the same finish as the Modi 2. With the Magni 3, they changed that finish and only updated the Modi 2 Uber and Multibit to match (the regular Modi 2 still has the old finish).

Yeah for a while they used painted steel on the cheaper models and anodized aluminum on the more expensive ones. Though I think at this point they are unifying to the aluminum for all of them. My original Mani phono preamp was painted steel, but when I gave that one to my daughter when she got a turntable, the replacement I ordered was aluminum.

 

[amirm]

@amirm

Can you double check the magni 3 measurements? Its performance in the thd+n vs power 300 ohms and 33 ohms seems much better than I was expecting.

I made a second run which is superimposed on the original measurement here:

So there is run to run variation which I have seen with some of these headphone amps. It tends to be related to the volume control. Tiniest bit of touch and the output changes some per above.

I dug up my previous measurement at the same 300 ohm:

While the general shape is similar (previous graph was scaled to be the same as this), the new measurement is definitely cleaner.

At first I couldn't figure out why but then it occured to me that the most improvement in the new Audio Precision APx555 is in its analog signal generator. If my memory serves me right, it is some 6o t 10 dB quieter. That may be the reason we are seeing better measurements now.

Most of my testing is with digital products like DACs where the above difference is not material as the AP is generating digital samples. But in this case, it matters.

That is the only hypothesis I have at this point.

P.S. I did find a mistake in the original measurements in the case of 33 ohm. The setting for watts measurement was wrong so the horizontal scale is different than it should have been.

 

[amirm]

One more thing actually . The test fixture I am using now was different than when I ran the original tests. That impacts amount of noise pick up which in turn changes THD+N.

In general the purpose of posting these THD+N graphs versus load is to show the power rating. There is much too much cabling and connectors involved for accurate noise/THD+N measurements this way. At near clipping where we measure power, noise is no longer important.

 

[rebbiputzmaker]

Why are you upset that I like amps without opamps?

Maybe because the Ken Doll is not anatomically correct and Barbi is very frustrated.

 

[Krunok]

Maybe because the Ken Doll is not anatomically correct and Barbi is very frustrated.

I would say Ken Doll is as anatomically correct as Barbi.

 

[Thomas savage]

Maybe because the Ken Doll is not anatomically correct and Barbi is very frustrated.

He must have a lot of free time on his hands.

 

[Wombat]

He must have a lot of free time on his hands.

He can spend it 'imagineering' on forums.

 

[Krunok]

LOOL

 

[Bhargu]

I might prefer a discrete component based amp over an opamp based amp if it measures as well. The reason is simple. It is an engineering challenge and I like well engineered devices.

I remember those old auto-reverse cassette decks. Some of them actually flipped over the cassettes instead of just flipping over the read-head. There were so many mechanisms, they were engineering miracles of their times. Though unwieldy, and prone to breakage, they belong in an engineering museum.

 

[Hemicrusher]

I might prefer a discrete component based amp over an opamp based amp if it measures as well. The reason is simple. It is an engineering challenge and I like well engineered devices.

I remember those old auto-reverse cassette decks. Some of them actually flipped over the cassettes instead of just flipping over the read-head. There were so many mechanisms, they were engineering miracles of their times. Though unwieldy, and prone to breakage, they belong in an engineering museum.

Exactly!

And I had a Nakamichi RX-505 in the 80's that flipped the tape. Paid a stupid amount of money for it new......

 

[Xyrium]

So, it appears that though measurements indicate a "winner", these two are audibly very similar in performance. In fact, the Magni has more power, and lower output impedance. Though the latter might not be useful if I'm stuck in low gain mode on the Magni when using 16-32 Ohm IEMs, but very little volume control due to the high power design. So, this probably leaves them at a subjective draw for those testing with their ears.

Sounds like a lot of qualifiers, but I'm pleased to see the Topping do well, though it is actually more expensive. I own it, and it drives my DT880 Pro 250 and DT880 Prem 600's perfectly.

Another great review Amir, thanks!