Schiit Magnius Balanced Headphone Amp Review

[JohnYang1997]

If I ever make a habit to lay on the couch at night after the kids are in bed an listen to IEMs, I'll certainly consider something from your company with the three output levels. You obviously considered and invested in IEM performance with your design implementation. I only use IEMs for teleconferencing and plugging into my phone or tablet right now, but this will be an important criterion for some consumers.

The thing is that for IEMs you probably only need a couple of millivolts to be loud enough. I don't think I need more than 100mV for my hd650. So really for almost all use cases 50mV performance is relatively more relevant. Not that it is an issue as it's probably under audible threshold anyway. Also I'm not asking you to buy certain products but to think a little bit differently. Atom is also better in this regard. So unless you need the full power of Magnius, Atom and heresy are the actual better performer under real world use case.

 

[Helicopter]

I see dashboard as competition as L30 is also the top of SINAD. I also don't really like the 'manufacture' badge on top of me. I was just trying to make a point. But it's impossible not to mention a90/l30. 50mV and multitone(preferably with load) are the two measurements (channel balance too if not for pure performance) I actually care if I were to choose a product. I'm not saying schiit or other brand was bad. But as you can see from 50mV snr chart, the Atom was much better still. The heresy is also better than magnius. If you look this way. Immediately the view completely changed. Magnius's strength is 199USD powerful BAL amp with tons of power(probably the cheapest amp to power HE6 and K1000 "properly"). But for normal use I don't think it's better than Heresy and Atom. The channel balance is not even that good here regardless having more expensive pot.

Multi-tone is my favorite measurement to evaluate a product for transparency, but it doesn't give a nice number for ranking like SINAD (or Harmon Preference for speakers). I'm going to take 50 mV more seriously now. You've made persuasive arguments. Thanks for teaching me more about this. I didn't mean to imply that you are not objective; I was just teasing and pointing out your perspective, hence the wink-face. I intended mentioning your perspective and products in the context of these charts as a back-handed compliment. I don't think you are wrong about the value of the 50 mV scenario. It is still a specific point, but it is one that makes better sense than Best Case for headphones like I use most often (Focal Clear, LCDX), and as you point out, less sense for HE6, and closer but still way high for IEMs.

 

[JohnYang1997]

Multi-tone is my favorite measurement to evaluate a product for transparency, but it doesn't give a nice number for ranking like SINAD (or Harmon Preference for speakers). I'm going to take 50 mV more seriously now. You've made persuasive arguments. Thanks for teaching me more about this. I didn't mean to imply that you are not objective; I was just teasing and pointing out your perspective, hence the wink-face. I intended mentioning your perspective and products in the context of these charts as a back-handed compliment. I don't think you are wrong about the value of the 50 mV scenario. It is still a specific point, but it is one that makes better sense than Best Case for headphones like I use most often (Focal Clear, LCDX), and as you point out, less sense for HE6, and closer but still way high for IEMs.

Actually multitone can produce some single number outputs you can even apply weightings. The single number is often referred as TD+N (total distortion + noise). This is less use due to the noise contribution being overwhelming. It's still a usable evaluation of the overall performance. (the overwhelming noise contribution would also indicate the importance of noise performance on the other hand) To me personally, as it produces a graph with much more information than a single number, directly reading it is better. And I do understand it being less easy to compare.

 

[T.M.Noble]

To what can we attribute the significantly worse or much less good single-ended performance? I am afraid that many 'audiophiles' will latch on to this specific
piece of equipment as general proof that Balanced is miles ahead of single-ended in terms of performance.

We only use one phase of the balanced output so there is no cancellation of distortion between phases. Hence the difference between SE and XLR. It was a design choice. Our goal was to make a beast headphone amp to go with Modius. We already had a fantastic performing SE amp in Heresy.

 

[Thomas savage]

Why even include the unbalanced outputs then, if the performance is so wildly different?

Functionality, it would impede sales not to do so. Anyway, audibly different?

 

[ElNino]

Functionality, it would impede sales not to do so. Anyway, audibly different?

This probably isn't the thread to discuss audibility thresholds, but with IMD in the -73dB range, draw your own conclusions. (Though I note that T.M.Noble has basically conceded that it is audible, in that s/he said they blind tested this amp against one that has a proper SE output stage and thought this much higher distortion version sounded audibly preferable.)

I don't doubt that it contributes to sales, but people buying it for the single-ended output are likely buying a higher-distortion amp than whatever they already are driving their headphones with. In that sense, it's pointless.

 

[itayoron]

Flip a coin or somthing.

Yup- coin toss, or get the Magnius as cheaper and (depending on where you are) maybe easier to obtain. I dont think you would be missing anything getting one over other- except the 789 has a "better" single ended performance- but you have the Heresy.

I got the Magnius over the THX based on aesthetics (especially with my other Schiit gear), and because I am a big fan of Schiit. I like how Jason participates here, and how Schiit is meeting the challenge of consumers who demand products with excellent measurements you can't hear.

As long as nothing is broken or incorrectly configured, you shouldn't be able to hear a difference in any of this gear. Cost is also a benefit of the Magnius where I live. However, I would have paid another $100 for the Magnius without hesitation; I even waited for Schiit to release the Magnius over buying something else a couple months earlier because I gathered something like this was next in their lineup.

Tossed the coin three times - all three THX-789...

 

[T.M.Noble]

This probably isn't the thread to discuss audibility thresholds, but with IMD in the -73dB range, draw your own conclusions. (Though I note that T.M.Noble has basically conceded that it is audible, in that s/he said they blind tested this amp against one that has a proper SE output stage and thought this much higher distortion version sounded audibly preferable.)

I don't doubt that it contributes to sales, but people buying it for the single-ended output are likely buying a higher-distortion amp than whatever they already are driving their headphones with. In that sense, it's pointless.

Concede is a strange word here. It directly implies an acknowledgement of truth after a denial. I said I think the SE sounds great and only commented on the reason why there is a distortion difference between the balanced and SE.

And I agree, if you want a better SE amp, buy Heresy. Its cheaper and supplies awesome measurements and plenty of power.

 

[ElNino]

Concede is a strange word here. It directly implies an acknowledgement of truth after a denial.

I never said that you denied that the distortion was audible. You basically acknowledged from the outset that these levels of distortion were audible in your tests. It was Thomas who questioned whether it is audible.

 

[T.M.Noble]

I never said that you denied that the distortion was audible. You acknowledged from the outset that these levels of distortion were audible in your tests. It was Thomas who questioned whether it is audible.

The comment was to point out the word concede is defined as acknowledging truth after a denial. You stated that I conceded the difference was audible as if I denied the difference at a previous time. Just wanted to clarify in case there was ambiguity.

As far as I can see, my acknowledgement was that there is a measurement difference not that I ever audibly heard differences. However if I said that or implied it, that is certainly not the case. The vast majority of my critical listening of the amp was balanced but I never noticed a qualitative difference between the two. Magnius/Modius is the current stack in my office so perhaps I will be able to pick out a difference to more I get used to the setup. I will get to polishing my golden ears.

 

[Racheski]

This probably isn't the thread to discuss audibility thresholds, but with IMD in the -73dB range, draw your own conclusions. (Though I note that T.M.Noble has basically conceded that it is audible, in that s/he said they blind tested this amp against one that has a proper SE output stage and thought this much higher distortion version sounded audibly preferable.)

I don't doubt that it contributes to sales, but people buying it for the single-ended output are likely buying a higher-distortion amp than whatever they already are driving their headphones with. In that sense, it's pointless.

No rational consumer would buy the Magnius for the SE output. You buy because of the balanced architecture, and the SE output is there for convenience.

 

[ElNino]

As far as I can see, my acknowledgement was that there is a measurement difference not that I ever audibly heard differences. However if I said that or implied it, that is certainly not the case.

When you posted earlier "We could have made the SE measure the same but to be honest it adds parts which adds cost, we didn't like how it sounded when tested..." what did you mean then?

For reference, that was post 481789:

We could have made the SE measure the same but to be honest it adds parts which adds cost, we didn't like how it sounded when tested...

 

[T.M.Noble]

When you posted earlier "We could have made the SE measure the same but to be honest it adds parts which adds cost, we didn't like how it sounded when tested..." what did you mean then?

I meant that adding components in order to sum the balanced signal to SE changes the sound in a way we did not prefer during testing. The differences we heard were between different versions of the product using different components. Adding an opamp or any of the multiple ways to sum the signal changes the way it sounds. We went through multiple iterations of this product until we like the way it sounded. Measurements mattered but we also had to enjoy the product as well (hard to get rid of that pesky subjectivism). Just to be very clear, once volume matched I cannot tell the difference between the XLR and SE on Magnius.

I hope I made that a bit clearer. Apologies for the misunderstanding. Not intending to mislead anyone.

 

[Asylum Seeker]

This thread is making me wanna try out the Balanced output of my DAC|Amp which I've used exclusively single-ended...

 

[BAMCIS]

Dont believe Amir tested the rca inputs. Might have to wait for user reviews to come in.

Of course, if you use RCA in@2v , XLR out, it wont be as loud for any point on the knob as if used XLR in @4v. Not sure why performance should suffer though.

Can I use a phono pre amp into the RCA input, switching from it to XLR fed by my Modius? The Modius has all my digital sources under control. Reason I ask is that Magnius' SNR measures way better than Freya S at a third of the price, I only need one XLR in and one RCA in and I don't need a remote. Not sure how the alps blue pot can match or beat their highly touted film-resistor volume attenuator, can anyone speak to that? I asked Shiit how many volts it makes via stereo XLRs and they told me to find a formula and convert the headphone juice /db to volts. Me no engineer. Not sure why no mention is made of the Magnius for stereo pre amp duty to a power amp or measurements provided by anyone, I think we got this info on the Jot and it wasn't good.

 

[Helicopter]

Can I use a phono pre amp into the RCA input, switching from it to XLR fed by my Modius? The Modius has all my digital sources under control. Reason I ask is that Magnius' SNR measures way better than Freya S at a third of the price, I only need one XLR in and one RCA in and I don't need a remote. Not sure how the alps blue pot can match or beat their highly touted film-resistor volume attenuator, can anyone speak to that? I asked Shiit how many volts it makes via stereo XLRs and they told me to find a formula and convert the headphone juice /db to volts. Me no engineer. Not sure why no mention is made of the Magnius for stereo pre amp duty to a power amp or measurements provided by anyone, I think we got this info on the Jot and it wasn't good.

This works fine. You can use the XLR/RCA selector of Magnius as a source selector without any issues.

I have phono preamp and modius RCA going into Sys so I can select sources on Heresy, and then RCA output from Heresy and XLR from Modius going into Magnius and RCA from magnius to power amp. For power amp I set Sys on phono and then I use Magnius to switch from RCA phono to Modius XLR. Heresy and Sys are just for TRS headphones since Heresy has better TRS performance than Magnius.

 

[BAMCIS]

I hate to say it but Magnius is the equal of the Benchmark HPA4 by the ASR headphone SINAD measurements and Schitt claims 125 db SNR from it's XLR pre amp outs. No other pre amp performs this well as far as I know, not to mention it's $199 vs $2,999.

"It goes without saying that the Benchmark HPA4 garners my strongest recommendation for a headphone amplifier and preamplifier. It doesn't get better than this folks" and yet no mention of the Magnius performance as a pre amp. I rarely use headphones but this looks like a world-beater pre amp unless I'm missing something. Considering how so many today have only digital sources so the DAC handles switching anyway.

 

[maverickronin]

I hate to say it but Magnius is the equal of the Benchmark HPA4 by the ASR headphone SINAD measurements and Schitt claims 125 db SNR from it's XLR pre amp outs. No other pre amp performs this well as far as I know, not to mention it's $199 vs $2,999.

The HPA 4 is stupidly expensive, but it's stepped relay volume control kicks any pot to the curb and makes a much bigger difference to usability and audibility than a few dB of already inaudible SINAD.

 

[Racheski]

The HPA 4 is stupidly expensive, but it's stepped relay volume control kicks any pot to the curb and makes a much bigger difference to usability and audibility than a few dB of already inaudible SINAD.

And a remote (can't believe they charge extra for it btw), a digital touchscreen color display, a mobile app, additional XLR and RCA inputs, mono XLR out. Are there any other headphone amps with all of those bells and whistles?

 

[Helicopter]

The HPA 4 is stupidly expensive, but it's stepped relay volume control kicks any pot to the curb and makes a much bigger difference to usability and audibility than a few dB of already inaudible SINAD.

Makes me wonder what Schiit would have come up with if the only target they changed was price point, and made it a $499 HPA. Obviously a Freya attenuator, Heresy TRS... they probably could have included a Modius board and still been less than 20% of HPA 4 price.