Review and Measurements of Schiit Aegir PWR Amplifier

[bunkbail]

These might work with my 4367.

I thought you were satisfied with the Akitika's already?

 

[direstraitsfan98]

I thought you were satisfied with the Akitika's already?

Begone, frogposter.

 

[tomchr]

First, targeting a very small niche market if that was indeed by design is not a winning strategy. Second, it is not doing the right marketing even for tech geeks correctly. This forum is over-represented by tech geeks and yet most don’t like this company’s products. Because tech geeks also look for proof of technical quality of the product, not just shop talk.

Schiit markets themselves as high end performance at an affordable price. Great! Now deliver! I have yet to see any of their products deliver high end performance. Many of their products have performance issues stemming from Schiit's implementation of the circuit. Now, I'm not suggesting that I expect $10k performance from a $99 product. That would be silly. But it would be nice to see a product from Schiit that delivered the data sheet performance of the parts they used.

As for the question about where on the THD vs output power curve to rate the output power: I'd say it has to depend on the amp. If it's a solid state amp that clips hard, I'd go with the point where the THD shoots up. If it's a tube amp (or low loop gain solid state amp) with a softer clipping response, I'd probably take the intersection of the low-power THD slope and the high-power/clipping slope (so basically the knee of the curve) to be the max power output. This will mean that my solid state amps will be rated as, say, 100 W at 0.0001 % THD, whereas a tube amp would be more like 10 W at 2 % THD. I'd be fine with imposing a maximum of 1% THD, though many tube designs (especially single-ended ones) will struggle to provide any output power below 1% THD.

Tom

 

[Panelhead]

Rated power is good to know. Whether 0.1% or 1%.
What I really care about is seeing the THD + N spectrum at 10 mW. 100 mW, and 1 watt. But the low power specs are not usually listed.
I thought about purchasing one of these. The lack of balanced connection was a deal killer.

 

[JohnBooty]

I am guessing you are more in a technical career than a marketing career who thinks what the latter does is mostly dishonest.

I was thinking the same thing about you.

First, targeting a very small niche market if that was indeed by design is not a winning strategy.

Yeah, because audiophiles aren't a niche or anything. Surely you're joking? A $799 stereo amplifier is the very definition of a niche product.

There is a difference between dumping everything about the product and only putting the best part of your equipment in the description. The section I quoted only appeals to hard core tech geeks.

Hard core tech geeks? Sure. Or, y'know, audiophiles. Not sure if you've noticed, but (waves hands at where we're posting, which is a shrine to obscene levels of technical detail about audio products) audiophiles kiiiinnnnnnnnnnnda like the technical stuff.

Saying it still runs hot is unnecessary in that section. Have a faq somewhere if that happens to come as a question that people ask. Or link to a technical white paper.

Or be really honest about it up front, because that builds trust. And also helps head off costly returns and support calls from buyers who failed to read the fine print, and are confused why their new amp is running hot. Margins in the electronics industry are not huge, and a few support calls/emails to Schiit's (US-based) staff and product returns quickly annihilate those margins. I see you've never shipped (successful, profitable) product.

The latter would certainly look refreshing to tech geeks but not sufficient if the products are “Schiit”. Except perhaps for a certain group of people that talk meaningless tech porn in the units they buy like the size of their subwoofers, the number of speakers and amps you can embed in a car

Technical information that flies over one's head often looks like "technobabble" or "tech porn."

I don't find a quick heads-up that "hey, this is a variation on Class A and it runs hot" to be an overwhelming amount of technical information, and I don't think most audiophiles would either. That's overwhelming to your average person, but it is not overwhelming to anybody who's looking to drop money on a very niche product like a $799 stereo amplifier.

At the very least, "it runs hot" would be overwhelming to precisely nobody, and it's a good thing for somebody to know before they buy this thing.

 

[amirm]

Schiit goes after what audiophiles think they need. They want tube. They give it to them. They want "class A" they give it to them.

 

[restorer-john]

Schiit goes after what audiophiles think they need. They want tube. They give it to them. They want "class A" they give it to them.

Audiophiles want a turntable, they do that too. Seems to be working for Schiit and their loyal followers.

 

[audimus]

I am repeating myself here from another thread. Audiophiles and audio tech geeks are not synonyms as often confused here. There are a lot of audiophiles (in terms of looking for good equipment) who aren’t necessarily into tech porn or even understand any of it. Then there are people who fall for equipment based on technical porn who aren’t necessarily audiophiles in that they don’t necessarily have the ability to differentiate good sound from ok sound. There are people of both kinds here and in the spectrum between the two.

Schiit marketing isn’t targeting the former but the latter where being Class A becomes the end than what Class A does for you. Try telling the former that it still runs hot as part of your brief product description and see where it gets you. Anyone who has done any marketing will get this.

Then there is the cult aspect to Schiit (not unlike Bose) where it isn’t rational anymore. They could say their device melts if driven hard and people will say it proves how great and honest they are.

 

[yodog]

Are you ever going to measure the Schiit vidar?

 

[March Audio]

Schiit markets themselves as high end performance at an affordable price. Great! Now deliver! I have yet to see any of their products deliver high end performance. Many of their products have performance issues stemming from Schiit's implementation of the circuit. Now, I'm not suggesting that I expect $10k performance from a $99 product. That would be silly. But it would be nice to see a product from Schiit that delivered the data sheet performance of the parts they used.

As for the question about where on the THD vs output power curve to rate the output power: I'd say it has to depend on the amp. If it's a solid state amp that clips hard, I'd go with the point where the THD shoots up. If it's a tube amp (or low loop gain solid state amp) with a softer clipping response, I'd probably take the intersection of the low-power THD slope and the high-power/clipping slope (so basically the knee of the curve) to be the max power output. This will mean that my solid state amps will be rated as, say, 100 W at 0.0001 % THD, whereas a tube amp would be more like 10 W at 2 % THD. I'd be fine with imposing a maximum of 1% THD, though many tube designs (especially single-ended ones) will struggle to provide any output power below 1% THD.

Tom

Whilst you could argue about the audibility of the way distortion sets in on SS V tube, I dont agree that different standards should apply to the two.

 

[amirm]

Are you ever going to measure the Schiit vidar?

One was offered but there was no follow up. Let me circle back with him.

 

[Panelhead]

The Aegir and Vidar remind me a little of the two Forte amps designed by Nelson Pass. One was high bias class AB, the other class A.
The AB model was higher power (of course) and matched with more speakers. The class A model more for limited applications

 

[tomchr]

Whilst you could argue about the audibility of the way distortion sets in on SS V tube, I dont agree that different standards should apply to the two.

How would you rate the max output power of a soft clipping amp then?

Tom

 

[March Audio]

How would you rate the max output power of a soft clipping amp then?

Tom

If 0.1% (for example) THD is our limit, then that is our limit.

Its a specific technical test. Never take it in isolation, but it is what it is.

 

[tomchr]

But isn't that just as arbitrary as 1%?

If the purpose is to catch an amp that misbehaves under the limit, maybe some kind of deviation spec (similar to ±3 dB) should be used.

Tom

 

[Panelhead]

One area this bank of testing misses is overload recovery and settling time. Hitting an amp with an impulse to drive into clipping and recording the output after as it settles.
These contribute the amps character. Most amps can be overdriven. Some hide it well. Others it is glaring.

 

[DonH56]

One area this bank of testing misses is overload recovery and settling time. Hitting an amp with an impulse to drive into clipping and recording the output after as it settles.
These contribute the amps character. Most amps can be overdriven. Some hide it well. Others it is glaring.

Amir measured some amps at clipping and found some very ugly waveforms, much worse than I'd have expected in this day and age. He showed plots over on AVS, he may still have them around. At least the polarity did not invert like with some of the early op-amps -- that was unnerving to say the least.

 

[amirm]

Here they are:

 

[boXem]

But isn't that just as arbitrary as 1%?

If the purpose is to catch an amp that misbehaves under the limit, maybe some kind of deviation spec (similar to ±3 dB) should be used.

Tom

Since @amirm always puts a reference curve at the same time (here it's the ABH2} it quite easy to separate the good from the average in terms of behavior before clipping.
In terms of limit, I vote for 1%. 0.1% may simply exclude some amplifiers from the test, and 1% allows the amplifiers with clean behavior at the limit to be rewarded with a nice power score.

 

[March Audio]

But isn't that just as arbitrary as 1%?

If the purpose is to catch an amp that misbehaves under the limit, maybe some kind of deviation spec (similar to ±3 dB) should be used.

Tom

I think the purpose is to measure the maximum power output with a distortion level that can be reasonably considered inaudible. I think 1% is pushing it a little too far. Ive already said its a bit arbitrary.

Also someone else made a very good point about the resolution of the plot. You dont really want to be on the vertical part of the slope.