Schiit Freya Modes: Subjective vs Objective Differences


Longtime lurker, first time poster here, thanks to all the contributors to this community over the years.

I’m curious if anyone has had experience with a firm subjective impression being negated, or at least complicated, by objective testing.

I recently purchased a Schiit Freya Noval unit and have been driving it hard and testing its output modes for the last month. I very much enjoy it, regardless of the following.

Ignoring the SS buffer for now, my impressions of the passive mode are that it is clean, clear, detailed, and transparent, at the cost of the lower midrange down feeling a little thin, as if the backing band is a few steps farther away from the singer. Vocals are prominent but the soundstage is a tad narrow. The tube mode on the other hand feels a smidge richer and fuller, with male vocals and acoustic guitar in particular having extra presence, weight, and warmth. The bass is more articulate and extended, and a pleasant haze falls atop the presentation, softening harsher details and making the soundstage seem wider and airier.

Or so I thought.

I decided to record the outputs of the Freya with a high end video capture device I use for work that has analog inputs. Using test tones, scopes, and the stepped attenuator, I’m able to achieve a dead-nuts exact level match between the two modes prior to making 24-bit/88khz recordings of a handful of songs spanning different genres. I sync these up in my editor program, and am able to instantly A/B them live listening with my Sundara headphones, sighted or blind.

The result: no difference. No difference as in truly zero difference between the passive and tube outputs. I’ve done this test in a few other contexts with other components before and always find some little detail I’m able to latch onto to establish a contrast, but not this time. Classical, bluegrass, rock, reggae, doesn’t matter: the passive and tube modes give the same output in my test.

Assuming for the sake of discussion that my methods and analyses aren’t flawed, I see two possible explanations:

Either A) The differences exist, it’s just that they are due to the interplay between the Freya and all downstream components in my main listening area, and therefore cannot be heard directly from the Freya to the capture device.

Or the dreaded B) There are actually no differences at all and this is purely one’s brain and ears, the collected influence of reviews and audio forums, confirmation bias, and the imprecision of level matching the modes while listening, playing tricks on me.

What do you all think? Is this something you have experienced as well? Thanks for reading and for sharing any impressions.

nixanthrope

Showing 7 responses by nixanthrope

Hi @mattw73 , no worries happy to provide an update.

I didn't end up pursuing my findings with Schiit, though I continue to be a happy camper using this preamp and would encourage you to pick one up if you can still get one, especially with the balanced options. It's a steal.

I learned shortly after my last post that my microphone was no longer compatible with my recording software so I put a pause on further recording and have just been enjoying the music.

However, I have a UMIK-1 in the mail right this second as I was interested in exploring convolution filters within Roon and HQPlayer so I might pick this back up and can let you know what I find.

Regarding my experiences in the subjective realm, I've tried out some additional NOS combinations, and I feel there are differences, but they are definitely subtle. Most notably, there seems to be less energy and dynamics whenever you deviate too much from the recommended 2.4A heater current total of the four tubes. Using 4x 6922s just feels too mellow for instance, but adding 2x 6N6Ps to the output side restores the excitement while letting some of the character of the 6922s come through, which with my Bugle Boys, I interpret to be a tiny bit of warmth, with a slightly wider and more holographic soundstage compared to the SS Buffer. As tempted as I am to spend $300+ on a pair of NOS Mullards, I just feel that this application wouldn't do them justice. Luckily 6N6Ps are plentiful and cheap though.

So in short, I think you're right to call this an entry point to tubes, but it's still worth it in my book given the range of features and strength of the solid state modes.

@mattw73 I think since you're already stocked up with options, it's a no brainer and you should go for it.

Buying second hand might break the warranty, so maybe have a look at some QC issues others have experienced here before pulling the trigger (I personally haven't had any, mine is a good unit):

https://www.head-fi.org/threads/schiit-freya-impressions-and-tube-rolling-thread.832177/page-210

And if you're stateside, ebay seller "kambremer" has US stock Russian 6N6Ps and 6N1Ps for good prices. For the latter, you can get a matched quad for cheap, which is better than what the Freya N ships with. I also recommend trying 4x NOS 6CG7s which will also give you the full 2.4A.

Good luck and Happy New Year!

Thanks for your replies, @kota1 and  ​​​​@zlone .

I haven't gone down the REW rabbit hole just yet but that would be the logical next step to try to measure the actual perceived differences I originally thought I heard through my equipment and from my listening position.

Do you know if REW only works with recordings you make within it, or if you can use its tools on audio files you create elsewhere in order to analyze harmonics and frequency response?

And yes, I did the test in two parts, the first time with the tubes warmed up for well over an hour, followed by the passive test, then for the second round I did the passive first so the tubes were cold...same result. This makes me less keen to roll tubes on this unit, which is perhaps a blessing in disguise.

@zlone @jeffrey125 I also notice the SS buffer appears to have a little more oomph than the passive, but question if this would hold up under scrutiny given it is a few db quieter than the passive and again difficult to level match in real time (running all RCAs at the moment as my vintage Accuphase amp appears to invert phase for the XLR inputs).

@kota1 which room correction device do you like best?

Amazing, thanks @kota1 I hadn't heard of the first two and I'm going to research them further. No sub in my setup right this second but when I do use one I do REL high level so all good.

@kota1 Very cool! Can you only use ARC with the analog outs/ML wireless speakers, or is it possible to use that feature in conjunction with an external DAC? I would think not but worth asking.

Thanks, @ejlif , despite this test I, like you, just can’t shake these same subjective impressions while listening: passive feels sharp but lean, tube soft but full. I came across some NOS Tesla E88CCs that I threw on the right side and continue to have the same reaction after this swap.

If I feel inspired to dive back in one more time, it will be with a USB microphone from my listening position. I’d love to verify my observations, but I won’t lose sleep if I can’t. And I won’t stop using tube mode either.

If nothing else, Schiit should be commended for creating a gain-adding signal path with four Soviet light bulbs that is so true to the source. I would love if there were more obvious differences between the modes, but still admire this unique preamp and will continue to enjoy it.