Schiit Freya S Preamplifier Review

[confucius_zero]

So the SAGA is still the better pre-amp?

 

[gvl]

Dang, I was considering it and was hoping for better performance in unity gain given their AP results. Seems they ran this at 550mV input, looking at the IMD plot distortion goes downhill past that point so their choice for the input level wasn't random it seems.

 

[stereo coffee]

With the majority of testing at 4v in and 4 volts out in passive mode, I am unsure if this helps greatly to reveal anything about distortion, at normal listening levels.

Some measurements at a range of normal listening levels that purchasers would be using, might assist those contemplating purchase, and reveal the products, actual daily use characteristics.

Also is there any indication of how it sounded in your audio system in terms of reproducing music, was it able to reproduce as example, the depth evident in Jan Garbareks live recording "Dresden, particularly with Manu Katche's drums, and the harmonics of the other instruments being used ?

 

[majingotan]

So the SAGA is still the better pre-amp?

Like I predicted, anything more complex to the signal results to worse performance and unacceptable channel measured performance matching. Only well competent engineers can pull off such feat including dCS, Mola Mola and Holo Audio May. Best off to use the Magni Heresy as preamp in this case as its design is far simpler than the Freya S.

My OG Schiit Saga performs better than Freya as its much simpler design. Schiit tries to make complicated designs yet can’t engineer them right as you can see from measurements here.

 

[garbulky]

Amir,
What is the input/output impedance in passive mode?
You should also test common-mode rejection in balanced mode, in this case in the 0dB mode. The passive VC would have no CMRR.

@amirm
Yes I too would like to know. Also interested in output impedance in active as well. Looks like this would be a great match for the Ygdrassil. I've been looking for a good solid state balanced preamp for it. BTW its preamp uses a resistor ladder so perhaps the resistor ladder has got a bad set of matching resistors.

 

[Tks]

https://smile.amazon.com/Poniie-PN1500-Electricity-Electrical-Consumption/dp/B07VPTN8FZ

Anything with digital I/O interfaces to computers or such? Or storage that could be transfered?

 

[LTig]

With the majority of testing at 4v in and 4 volts out in passive mode, I am unsure if this helps greatly to reveal anything about distortion, at normal listening levels.

In passive mode the unit does not distort at 4V, so it cannot distort at lower voltages. Any distortion in the plot stems from the AP.

Some measurements at a range of normal listening levels that purchasers would be using, might assist those contemplating purchase, and reveal the products, actual daily use characteristics.

See the IMD vs Level graph. Best performance in active low gain mode is at 500 mV.

Also is there any indication of how it sounded in your audio system in terms of reproducing music, was it able to reproduce as example, the depth evident in Jan Garbareks live recording "Dresden, particularly with Manu Katche's drums, and the harmonics of the other instruments being used ?

With a performance like this those sound qualities depend only on room and speakers.

 

[LTig]

Conclusions
If you can live with passive mode, you get superb, transparent performance with a very accurate volume control. Active low gain mode, if operated as good as the better channel would also be excellent. Alas, one channel is worse there to the tune of 9 dB. I wonder if Schiit has a specification for channel differential and would fix this under warranty. You should not need the high gain mode if you are fronting a DAC with proper output levels.

As it is, the Schiit Freya S is good but given the channel distortion issue, I can't fully recommend it. If that is resolved, then it is a very good preamp and one that I could recommend.

Even with the slightly worse channel (EDIT: spec is 0.002% at 2V so it's still in specs) I would recommend it, given the price. I may not have a good market overview but here it is difficult to find an excellent stereo preamplifier below € 1000 at all, so getting one for € 600 and with those feature set (XLR, stepped control, passive/active gain selection) is quite an achievement.

 

[amirm]

Some measurements at a range of normal listening levels that purchasers would be using, might assist those contemplating purchase, and reveal the products, actual daily use characteristics.

That is what the IMD versus level is for.

 

[FrantzM]

Seems that when they put their Schitt together, they can swing with the best... Good Schiit this product ...
, I would like to see the manufacturer response to the lone underperforming channel issue.. Could this be a problem with this particular sample? .. Schiit happens

 

[JDragon]

How audible is that channel difference between passive and active mode? I have the Freya+ tube version and feel like imaging is a bit worse with active or tube mode on.

Any easy way for me to measure if mine has the same issue? I don’t know anything about electronics but have a RME ADI-2 Pro which I think I’ve seen used to generate other measurements.

Would be happy to send mine in but I have no other pre-amp for this system.

 

[amirm]

It should be very difficult if not impossible to tell the difference in use.

Yours being tube based, likely has much worse performance and channel differential. There is no easy way to measure this. If you are in US, you are welcome to send it to me and I will measure.

 

[stereo coffee]

In passive mode the unit does not distort at 4V, so it cannot distort at lower voltages. Any distortion in the plot stems from the AP.

As seen here: https://pinkfishmedia.net/forum/threads/alps-potentiometers-gain-and-distortion.68076/ distortion is best with some attenuators when there is no attenuation.

Set like this, with this style of passive they are just measuring wiring, and little else @amirm" Let's start with our usual dashboard, feeding the unit 4 volts and measuring 4 volts out in passive mode (volume set to max " ): In the process as I referred to, it reveals little about needed attenuation characteristics, as few purchasers would actually listen at full volume.

Music remains a great tell tale of product ability, and should always be part of every test.

 

[JDragon]

It should be very difficult if not impossible to tell the difference in use.

Yours being tube based, likely has much worse performance and channel differential. There is no easy way to measure this. If you are in US, you are welcome to send it to me and I will measure.

That’s a relief, I can keep on using active mode with no issues. As far as I can tell, the only magic the tube section adds is some hiss and a headache dealing with a mishmash of faulty tubes Schiit sent me over two batches. But they sure look cool...

 

[pkane]

Anything with digital I/O interfaces to computers or such? Or storage that could be transfered?

I've seen a few with bluetooth interface and smartphone apps that let you see power consumption over time. Don't know if these allow data to be extracted or downloaded.

 

[LTig]

As seen here: https://pinkfishmedia.net/forum/threads/alps-potentiometers-gain-and-distortion.68076/ distortion is best with some attenuators when there is no attenuation.

Set like this, with this style of passive they are just measuring wiring, and little else @amirm" Let's start with our usual dashboard, feeding the unit 4 volts and measuring 4 volts out in passive mode (volume set to max " ): In the process as I referred to, it reveals little about needed attenuation characteristics, as few purchasers would actually listen at full volume.

Music remains a great tell tale of product ability, and should always be part of every test.

Difficult to understand what you want to say with this scramble of bad quoting (intermix of my posting and your answer). Mabe next time use the Preview button before hitting Post reply?

 

[restorer-john]

But can I suggest something other than "design language" - design could mean engineering

or ergonomics

design language is often used to mean visual esthetics, and I think it denigrates real design, i.e. engineering - here circuits

Aesthetics and ergonomics, definitely preferable. "design language" is a meaningless marketing term. It's not telling a story or talking to me.

 

[amirm]

Aesthetics and ergonomics, definitely preferable. "design language" is a meaningless marketing term. It's not telling a story or talking to me.

It is the proper industry term (design language). Would you prefer for us to not say capacitor and instead say charge storage device?

 

[gvl]

@amirm, can you easily produce THD+N vs. output level at a fixed input, say 4V or a bit less, for the 0dB mode?

 

[Sir Sanders Zingmore]

In passive mode the unit does not distort at 4V, so it cannot distort at lower voltages.

showing my ignorance here, but why is this the case?