• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Schiit Magni 3+ and Heresy Headphone Amp Reviews

makmeksam

Member
Joined
Nov 13, 2019
Messages
59
Likes
44
Do like I did and buy both, try them out, then send one back. Consider reviewers recommendations, both Amir and the author at headphones.com, but it's your money, your ears, your preferences and ultimately your decision. From what I heard, both are fantastic amps so there is little risk of making a bad decision.
If both are fantastic what could be the bad decision?
 

makmeksam

Member
Joined
Nov 13, 2019
Messages
59
Likes
44
I don't trust the impressions in that review. The are cliche predispositions against well measuring gear. I specifically listened to the effects he mentions and there is none there. Please see my listening impressions in the review.
Following is from the listening tests.
I tried with a few tracks to tell the two amps apart but I could not. In more careful ABX scenario where I can focus on specific segments I might do better.
Here @amirm said that two amps are identical in sound in a quick comparison but he said he might see differences if listened more carefully. This means they can sound different being so close in measurements around -110dB levels! How come?
 

Mikey

Member
Joined
Oct 27, 2019
Messages
33
Likes
48
I don't trust the impressions in that review. The are cliche predispositions against well measuring gear.

... that persisted in blind testing, which isn't usually a thing baseless predispositions do.

I specifically listened to the effects he mentions and there is none there. Please see my listening impressions in the review.

I did. They say, "I wanted to get this review out quickly so did not spend extensive amount of time listening.... I tried with a few tracks to tell the two amps apart but I could not. In more careful ABX scenario where I can focus on specific segments I might do better."

That doesn't seem at all incompatible with someone finding differences when spending many hours over multiple days doing careful comparisons. If you don't hold a strong preconception that it's impossible to hear the difference between amps, the evidence here seems to point toward it being possible.
 

Ratatoskr

Active Member
Joined
Nov 22, 2018
Messages
125
Likes
83
Well USB for audio was pretty much 'solved' a decade ago.
It took them long enough and proper operation still has to be verified independently.

Jason Stoddard wrote that Unison USB was UAC2 compliant and Schiit is an official USB-IF developer. Who would do independent verification?
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,880
Likes
2,032
Location
Tampa Bay
So, wait... are we getting performance near the THX 789 for 99 usd? Or I misinterpreted the numbers?
Yes we are actually passing the performance in some cases for $99.
But the downside is less power output; although it is really not needed at all with headphones.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
... that persisted in blind testing, which isn't usually a thing baseless predispositions do.
Nope. That is the conclusion that is stated, not what we can see for ourselves. Statistical analysis needs to be performed and data shown to know if the results are valid or not. In any forced choice test, the person could be guessing. You need to determine the probability of that or the results are not valid.
That doesn't seem at all incompatible with someone finding differences when spending many hours over multiple days doing careful comparisons.
It is incompatible. If I told you in blind tests of two ice creams I found one to be sour, you would not just believe me. Such is the case with his report. Measurements show that telling these two products will be extremely difficult let alone have the effects he mentions.

Having performed the test and given my formal training in hearing such differences, I can assure you his results are not reliable and cannot be repeated.

If you don't believe me, buy both, match their levels and shoot a video of you telling them apart at least 8 out of 10 times blind. If you can I will pay for both of them and you can keep them for free!
 

Tks

Major Contributor
Joined
Apr 1, 2019
Messages
3,221
Likes
5,494
... that persisted in blind testing, which isn't usually a thing baseless predispositions do.



I did. They say, "I wanted to get this review out quickly so did not spend extensive amount of time listening.... I tried with a few tracks to tell the two amps apart but I could not. In more careful ABX scenario where I can focus on specific segments I might do better."

That doesn't seem at all incompatible with someone finding differences when spending many hours over multiple days doing careful comparisons. If you don't hold a strong preconception that it's impossible to hear the difference between amps, the evidence here seems to point toward it being possible.

While I agree with you more than the bossman in this instance. The problem is, the testing methodology is unpublished. We cannot replicate his experiment, so it's tough to grant full backing. But in that same fashion, the bossman's is similarly, if not more non-descript about the testing (but based on how he does tests with other things in the past, I can be sure his tests are rigorous, but in this case we have no idea what he actually did either).

Like if I said "yeah man I can blind test THX 789, 877, Archel Pro 2, Atom, Element 2, Heresy" and "I level matched them all within 0.1dB of one another, and I could tell the difference between all of them using FLAC files on my computer". You wouldn't take that to the bank I hope. Even if I had a review site to publish my review similarly.

When one person is doing a whole review with no one to at least attest to any of this actually happening. Most of this is taken as truth of non-truth with some level of faith. But if for instance, the person doing the review regularly participate in ABX tests with other people helping him, and has a history and publishes is work for proper archival purposes of re-running the experiment if others want to try themselves.. Then a review he conducts one time without a person, you can probably be sure he's most likely not lying or conducting the experiment in some shady or inadequate manner.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
If you don't hold a strong preconception that it's impossible to hear the difference between amps, the evidence here seems to point toward it being possible.
There is no "evidence." We have a writer saying something. Evidence would look like this:

File A: C:\Users\Amir\Music\AIX AVS Test files\On_The_Street_Where_You_Live_A2.wav
File B: C:\Users\Amir\Music\AIX AVS Test files\On_The_Street_Where_You_Live_B2.wav

18:16:24 : Test started.
18:17:21 : 00/01 100.0%
18:17:45 : 01/02 75.0%
18:18:10 : 02/03 50.0%

So far, the above says in two out of three times I could the two files apart in double blind ABX test.

Lay assumption would be that audible difference must exist then. But let's continue:

18:18:24 : 02/04 68.8%
18:18:44 : 02/05 81.3%
18:34:34 : 03/06 65.6%
18:35:28 : 03/07 77.3%
18:36:09 : 04/08 63.7%
18:36:32 : 04/09 74.6%

Now I have a score of just four (4) out of nine (9) times (75% chance of guessing). Clearly lost in the woods with some lucky guesses of differences I thought were there, but they were not.

Heck, unless you know me, I could tell you that I passed such a blind test. How would you know if I am telling the truth without posting at least this level of detail?
 

Mnyb

Major Contributor
Forum Donor
Joined
Aug 14, 2019
Messages
2,633
Likes
3,592
Location
Sweden, Västerås
Anyone can make a mistake , I’ll trust the measurement in these cases . What do I mean by that .

These measures really good, very very likely beyond anyone’s ability to tell them apart by ear .

So if you do detect a diff in ABX please get help and investigate your test methodology, it’s very likely that a mistake creeped in somewhere ? More than 0,1 dB diff in level to few trials ? Humans have problems with what 50/50 chance ie random looks like.
 

makmeksam

Member
Joined
Nov 13, 2019
Messages
59
Likes
44
Anyone can make a mistake , I’ll trust the measurement in these cases . What do I mean by that .

These measures really good, very very likely beyond anyone’s ability to tell them apart by ear .
We do not have IMD and multi tone tests for these two amps to see how they perform when more than one tone is fed in. I think IMD and multi tone tests are very important to simulate something at least close to music and see how the gear perform.
 

Tup3x

Active Member
Joined
Nov 17, 2018
Messages
162
Likes
130
Location
Finland
If Schiit would combine Magni Heresy and DAC that would do it justice in one package I would definitely be interested. Sticking the O2 to my work laptop sadly doesn't make that piss poor Realtek sound any better.

Well... Techically Schiit Hel could be it but it's still a mystery.
 

Human Bass

Addicted to Fun and Learning
Joined
Nov 25, 2018
Messages
653
Likes
660
Yes we are actually passing the performance in some cases for $99.
But the downside is less power output; although it is really not needed at all with headphones.
2.4 watts at 32ohms is a lot of power, like 4 times my q5s balanced output. So Im sure it can drive 99,9% of headphones no problem
 

ClosDeLaRoche

Active Member
Joined
Jun 21, 2019
Messages
136
Likes
131
If Schiit would combine Magni Heresy and DAC that would do it justice in one package I would definitely be interested. Sticking the O2 to my work laptop sadly doesn't make that piss poor Realtek sound any better.

Well... Techically Schiit Hel could be it but it's still a mystery.
Schiit posted their own measurements on Hel. They have the same measurement tool Amir does. Looks like 0.0008 THD+N on low gain and 0.002 on high gain. Not bad!
 

Warbles

Member
Joined
Mar 9, 2019
Messages
35
Likes
4
'I started with the Drop/MrSpeakers Ether CX 25 ohms. Both amps drove it to ear bleeding level in high gain. While I could hear increasing distortion, both were identical in this regard so I am amusing it is the headphone complaining. There was no sonic difference that I could detect."

Don't forget to change that 'amusing' to "assuming" when you get a chance!

Love from der Grammatikbahnführer
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
'I started with the Drop/MrSpeakers Ether CX 25 ohms. Both amps drove it to ear bleeding level in high gain. While I could hear increasing distortion, both were identical in this regard so I am amusing it is the headphone complaining. There was no sonic difference that I could detect."

Don't forget to change that 'amusing' to "assuming" when you get a chance!

Love from der Grammatikbahnführer
Ich bin dankbar fur deine Hilfe. :) Corrected.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
So what does the heresy look like on high gain?
You mean in measurements?

index.php
 

makmeksam

Member
Joined
Nov 13, 2019
Messages
59
Likes
44
Nope. That is the conclusion that is stated, not what we can see for ourselves. Statistical analysis needs to be performed and data shown to know if the results are valid or not. In any forced choice test, the person could be guessing. You need to determine the probability of that or the results are not valid.

It is incompatible. If I told you in blind tests of two ice creams I found one to be sour, you would not just believe me. Such is the case with his report. Measurements show that telling these two products will be extremely difficult let alone have the effects he mentions.

Having performed the test and given my formal training in hearing such differences, I can assure you his results are not reliable and cannot be repeated.

If you don't believe me, buy both, match their levels and shoot a video of you telling them apart at least 8 out of 10 times blind. If you can I will pay for both of them and you can keep them for free!
Will there be an audible difference between listening through iPhone headphone jack and using something like Heresy + modi? Let’s say power is not a concern. If there is a difference what would the difference look like?
 
Top Bottom