Schiit IEMagni Review (Headphone Amplifier)

[soundwave76]

Shiiiit, this is brilliant!

 

[Robbo99999]

The blue curve has its vertical scale to the right. 0 dB is both channels equal level, and deviating up and down is channel mismatch.

The red curve has its vertical scale to the left. It starts max volume (0dB attenuation) and decreases levels until the end at the right side.

The cursor shows 0.75 dB channel mismatch (blue curve), which is quite bad, already at -15 dB (red curve), which is fairly high in level (bad). Usually channel mismatch happens more often below 30 or 40 dB attenuation, but in this case it happens quite early.

So yes, they should pick their pots better ("binning" is the name for testing and picking the good ones from the bad ones).

Thanks for the explanation, yep, that's how I understood it, but it wasn't until yesterday that I had truly worked out how to read that graph, but wanted to check I was reading it right, thanks. Yes, so the channel mismatch starts happening at -15dB....is there any way to work out where on the volume dial that would be happening? As in I know that -6dB is about half the perceived volume level, so -15dB would be another -6dB (and a bit more) on top of that....so are we saying that the mismatch starts to occur from about 25% downwards, so 9'oclock downwards assuming the volume pot zero is at 6 o'clock? I don't know if it's possible to work it out this way.

 

[amirm]

.is there any way to work out where on the volume dial that would be happening?

The x axis is time. It is how long it took me to rotate the volume control gradually from max to min. Using that maybe you can guess where each point is?

 

[Robbo99999]

The x axis is time. It is how long it took me to rotate the volume control gradually from max to min. Using that maybe you can guess where each point is?

Cool, I had a suspicion that might be the case, but do you try to rotate the dial at roughly consistent speed? If you try to do it at a constant speed then we can certainly derive some rough information on where on the volume dial the channel mismatch occurs.....so in this reviewed headphone amp the channel mismatch is happening in the middle third of the volume range.

 

[rkbates]

I have a few Schiit products - happy with all of them. The best part is when my partner asks "what sh!t are you listening to now?", I no longer feel that my taste in music is under fire.

 

[vkvedam]

Quite a decent one. Price is very good in particular if you happen to live in the US.

 

[CedarWind108]

Sennheiser IE 900 just announced. This amp would have been perfect if not for the channel imbalance

 

[respice finem]

I guess that channel imbalance is slightly overrated. The channel imbalance of most headphones will be more (bad luck if in the same channel).
An analog pot practically never has perfect channel balance, and if it does, not forever...
A balance regulation in an analog amp is sometimes a good thing.

 

[scooter]

Thank you Amir, I'm happy to see a huge influence you have made to audio industry. And I don't want to imagine what manufacturers would be doing nowadays without you.
Great work! You deserve to get at least 1$ from each member here and additional income from the reviewed product sales.

 

[daftcombo]

The 0.8dB difference in SPL is equal to moving your head 9cm from the speaker centre!

I can hear an imbalance of 0.8 dB more than -90dB H2 distortion. So according to you we can close the website.

 

[daftcombo]

I guess that channel imbalance is slightly overrated. The channel imbalance of most headphones will be more (bad luck if in the same channel).

And what if the imbalance of the amp has to be added to the imbalance of the headphones? At the end it makes a lot.

 

[amirm]

Cool, I had a suspicion that might be the case, but do you try to rotate the dial at roughly consistent speed?

I try but obviously this is not easy to do. I have to take my hand off at times to reposition as well.

 

[YSC]

And what if the imbalance of the amp has to be added to the imbalance of the headphones? At the end it makes a lot.

completely agree, and that for the better headphones except in HF region mostly the balance is pretty good, and while it's near impossible to produce 2 completely identical full range transducer, the channel balance in the volume control and dac is much easier to achieve and thus should be kept an eye on more.

the arguement of the drivers itself deviates more than the amp is like saying a car won't have completely flat road surface or balanced weight distribution so wheel alignment doesn't matter

 

[respice finem]

And what if the imbalance of the amp has to be added to the imbalance of the headphones? At the end it makes a lot.

That's exactly why I wrote about the balance regulation.

 

[respice finem]

...the arguement of the drivers itself deviates more than the amp is like saying a car won't have completely flat road surface or balanced weight distribution so wheel alignment doesn't matter

The problem with car analogies is, they rarely fit... What will you do if your headphone has 1 dB L>R up to 1 kHz and 1 dB R>L above? Nothing is ever perfect, and never will be. Our own "channel imbalance" is also present.
Last but not least, we hear music and don't compare stuff all day (at least so I hope). And with music, does it really matter if an instrument is "virtually" 5cm one side or the other? Not for me.

 

[rohilx]

This is a review and detailed measurements of the just announced Schiit IEMagni headphone amplifier

This amp vs jdslabs atom? bought the atom stack 1 day before new schiit was announced :\. Was going to buy schiit as well but decided last minute for atom stack

 

[mmicko]

Another headamp with 120dB performance and price around 100USD , but I'm still missing the same performing (117-121dB Sinad) DAC with similar price (or max 150 USD) for pairing with these amps (iemagni, L30, atom etc.)

 

[MCH]

The x axis is time. It is how long it took me to rotate the volume control gradually from max to min. Using that maybe you can guess where each point is?

you can just make the attenuation the x axis and the imbalance the y. more clear.

 

[YSC]

The problem with car analogies is, they rarely fit... What will you do if your headphone has 1 dB L>R up to 1 kHz and 1 dB R>L above? Nothing is ever perfect, and never will be. Our own "channel imbalance" is also present.
Last but not least, we hear music and don't compare stuff all day (at least so I hope). And with music, does it really matter if an instrument is "virtually" 5cm one side or the other? Not for me.

well this analogy never convince me...
same for car analogy, "what if my steering is off to left by 1 degree and weight difference steers the opposite direction by 1 degree"

and for music, YES, it does from personal experience, especially if you are listening to those pop where the image is dead center, I once had an amp that way and below certain volume setting, the singer always appeared to standing in my front right side singing which feels weird! not to say at current day the same setup can as well be used in FPS gaming (I definitely switch to headphone when gaming), that could make the very advanced virtual sound clues off also. you can try having some flat speaker with poorer deviation tolerance like the lower range JBL 305P vs Genelec/ neumann, 0.5db of channel imbalance isn't obvious, but if it's ~3db the imaging will become weird

 

[mSpot]

Wonder if one of the other two is going away. They're starting to compete with themselves in this price segment.

That is exactly Schiit's "Thunderdome" approach. Create variations that compete with each other, and discontinue the versions that sell poorly. Jason Stoddard discussed it in his post.