The DIY'rs Cookbook

[kstuart]

The main reason that I never opted for the Project Sunrise is too many options - too many tube choices, too many adjustments.  The "tube rolling" becomes infinite when you can use almost any tube and you can change other settings as well.
 
With the Bellari (similar circuit design), I was limited to 12AX7, and in that world, I found that 1960s tubes were simply much better than any current production.  You can find used Telefunken for around $30 (singles are cheaper than pairs) on ebay (the last time I checked).

 

[landroni]

  The main reason that I never opted for the Project Sunrise is too many options - too many tube choices, too many adjustments.  The "tube rolling" becomes infinite when you can use almost any tube and you can change other settings as well.

 
With the Bellari (similar circuit design), I was limited to 12AX7, and in that world, I found that 1960s tubes were simply much better than any current production.  You can find used Telefunken for around $30 (singles are cheaper than pairs) on ebay (the last time I checked).

I know the feeling. I own an E12DIY, and the choice of SS opamps is simply bewildering. This is why I simply looked at the community filtered choices for optimal opamps for this device, and simply tried that:
http://www.head-fi.org/t/674808/fiio-e12diy-limited-edition-version-diy-your-own-sound
 
Didn't go wrong.

 

[johnjen]

User @Stillhart has reported on a previous occasion that:

http://www.head-fi.org/t/763905/finding-a-dac-for-the-cavalli-liquid-carbon-only-four-months-to-go/1005#post_11872598 "The GMB’s SE output sounds noticeably worse (mids are very recessed) than the balanced output. This is very easily tested on the LC [Liquid Carbon]."

Have you noticed anything similar? Does this mean that Bimby is "noticeably worse" than Gumby balanced output (and by extension Yggy)?

We only tested the SE outputs in the second test while in the first test we used Bal to SE transformers to convert the signal to feed the SE inputs on the amp.
I noted that the SQ of the signal pathway using the transformer didn't match up to the SE output of the Gumby.
So we didn't test for the sonic differences between SE and Bal on 'equal footings', since we weren't using an amp with both Bal and SE inputs.
Thus I can't really address your question.
What we were paying attention to were the strengths of that $1500 system as the main focus and using these 2 dacs as the sources.

Mike Moffat noted in one post that instantaneous blind A/B testing could lead to erroneous insignificant results, whereas long-term blind A/B testing (e.g. continuous 1h sessions with each individual setup) could allow to reliably identify the source. This has also been discussed in this post:

http://www.head-fi.org/t/779572/r2r-multibit-vs-delta-sigma/30#post_12041382
Did your own testing go more along the lines of instantaneous blind A/B testing, or that of long-term blind A/B testing? If it's more like the former, could this explain your difficulties to tell which DAC was which?

We/I used a combination of 'quick' and longer term listening to each source, based upon what we were paying attention to.
And it's interesting that you would bring this up, because I mentioned that the only 'real' way to determine actual suitability in any system to gain a 'more complete' grasp of any 'new device', is at home, over the long haul, using a variety of different music that one is already familiar with.

This makes it long term test which we didn't have the time for.
Nor was that our primary goal in these tests.

We were testing for scaleability all the while paying close attention to dollar amounts to keep the system as affordable and as kick-ass as possible.

It's interesting that you mention this, as Jason Stoddard mentioned in the Bimby announcement the importance that the design team attached to the DSP:

http://www.head-fi.org/t/701900/schiit-happened-the-story-of-the-worlds-most-improbable-start-up/8295#post_12031225

"The burrito is the DAC."

I alluded to this by saying that the hardware that is used in these 2 dacs are quite different from each other.
But the family resemblance was strikingly similar.
Thus my supposition was that the “supercomboburrito” filter is what we were hearing as the main contributor to this similarity.
Same idea, approached from 2 different angles, reaching the same conclusion.

JJ

 

[johnjen]

Looks like a good approach to A/B testing...

As  for the Electro-Harmonix 12BH7, does it have any harmonic distortion worth mentioning?

Um, I don't mean for this to be blunt nor rude or nuth'n, but…

It’s a tube, and added harmonics are what they do.
What makes one sound 'good' or 'better',
or not,
is strictly a matter of personal choice.

And that is both the allure, fascination etc. and the frustration, since that little round hole in the chassis into which you can throw as much or as little $$$$ as you deem appropriate, can seem like a bottomless hole, if you let it.

But there are others who have 'been there - done that' already and some can help avoid the same mistakes they made…
If you ask.
Just be careful of whom you ask. :atsmile:

Think sailboat, as in, a hole in the water, where $$$$$$ gets tossed, only to sink out of sight, about as fast as you can toss it.

Only completely different. :atsmile:

JJ

 

[johnjen]

The main reason that I never opted for the Project Sunrise is too many options - too many tube choices, too many adjustments.  The "tube rolling" becomes infinite when you can use almost any tube and you can change other settings as well.

With the Bellari (similar circuit design), I was limited to 12AX7, and in that world, I found that 1960s tubes were simply much better than any current production.  You can find used Telefunken for around $30 (singles are cheaper than pairs) on ebay (the last time I checked).

Yeah I know what you mean, especially the number and variety of tube choices is, or can be, just crazy.

But what I've learned is for a particular setup usually once a degree of 'synergy' has been reached the search for better ceases to be important.

Like the settings tend to settle down once a degree of familiarity is reached. which implies a degree of fussing is needed.
Well, unless you're like me and don't have a single bone in your body that doesn't DEMAND that it needs be tweaked still further.
In which case my sympathies, 'Been there - Done that' and by now its ingrained and I'm not sure I WANT to change it.

Oh well, I've lived with it for low these many years so far, what's a few more?
Eh?

JJ :atsmile:
ps I now run all SS, because it does simplify the sheer number of possibilities down to a more manageable number. But tubes are or can be magical, but then so can SS…

 

[kstuart]

 

It's interesting that you mention this, as Jason Stoddard mentioned in the Bimby announcement the importance that the design team attached to the DSP:

http://www.head-fi.org/t/701900/schiit-happened-the-story-of-the-worlds-most-improbable-start-up/8295#post_12031225

"The burrito is the DAC."

I alluded to this by saying that the hardware that is used in these 2 dacs are quite different from each other.
But the family resemblance was strikingly similar.
Thus my supposition was that the “supercomboburrito” filter is what we were hearing as the main contributor to this similarity.
Same idea, approached from 2 different angles, reaching the same conclusion.
 

 
That is easily tested.
 
Bimby does not use the "supercomboburrito" filter on 24/192 material.
 
So, if you get a top quality 24/192 file ( such as Alan Yoshida's excellent mastering of "Blue Train" ) then you can eliminate the filter as a factor and just test the multibit implementation.

 

[bimmer100]

Hoo Boy is back for a 2nd test of the differences between these 2 Schiit DACS.
And they are even smaller than I thought, while running in SE mode.

User @Stillhart has reported on a previous occasion that:
http://www.head-fi.org/t/763905/finding-a-dac-for-the-cavalli-liquid-carbon-only-four-months-to-go/1005#post_11872598
"The GMB’s SE output sounds noticeably worse (mids are very recessed) than the balanced output. This is very easily tested on the LC [Liquid Carbon]."

Have you noticed anything similar? Does this mean that Bimby is "noticeably worse" than Gumby balanced output (and by extension Yggy)?

 
 
 
I'm quite curious to know about this as well. I tend to agree with stillhart on many issues and we have a similar ear when it comes to gear.  I've heard the gumby a bit and the balanced mode is quite good, but didn't get a chance to try SE. Stillhart did mention the quality was lacking in a noticeable way, not just subtle.  A great amp like the liquid carbon should bring out all the details and nuances in the Schiit dac's like the yggy, gumby and bimby.  I'm most excited to get a chance to spend some time with the bimby and have it paired up with a few amps and hear it. If only I had a gumby at the same time. At least i'll be able to compare the yggy and bimby side by side.  Either way, i'm sure the bimby sounds pretty fantastic for it's price point.  I know Stillhart is partial to his DAC19 and I understand why! I used to own one and think quite highly of it for a SE DAC... that's why I ended up selling it and upgrading to the M11.  
 
     I would love to get a cheaper DAC for work, and hoping to hear this Bimby soon! So the above reviews really intrigue me and I thank JJ, AtomicBob and Big Poppa for taking the time to make a write-up review on it. I just hope it's not so bright like most all the Schiit Dac's i've heard. I feel that they might be better at lower volumes if I was to pair it with any of my amps, but nothing short of deafening and sibilant once the volume is up to 90db or so. I think I would like to try the PSIII amp as it might be the ticket to tame the sibilance from schiit dac's. Maybe i'm just super sensitive to higher frequencies. Having a pair of HE560's likely doesn't help, but they are much better after mods. 
 
What tube choices are ideal for the PSIII to keep treble smooth when paired with a bimby? And most importantly, not losing details. Unfortunately my experience with most tube amps leaves me unhappy as many sound very colored and overly smooth or lacking some detail.  The Zana Duex was one of the best sounding tube amps i've heard and would likely be happy with that. But the pairing with the HE560 wasn't the best, definitely better with HD800's. I am considering getting a new pair of headphones, maybe Ether C ( I need a good closed back), HE-X or possibly HD800S (depending on price).
 
 
Last question:
Does anyone know of a good A/B switch that would be able to handle at least two separate 4pin XLR balanced inputs and one 4pin XLR output that would be reliable and high quality?
 
 
 
Thanks in advance
-T

 

[gefski]

   I would love to get a cheaper DAC for work, and hoping to hear this Bimby soon! So the above reviews really intrigue me and I thank JJ, AtomicBob and Big Poppa for taking the time to make a write-up review on it. I just hope it's not so bright like most all the Schiit Dac's i've heard. I feel that they might be better at lower volumes if I was to pair it with any of my amps, but nothing short of deafening and sibilant once the volume is up to 90db or so. I think I would like to try the PSIII amp as it might be the ticket to tame the sibilance from schiit dac's. Maybe i'm just super sensitive to higher frequencies. Having a pair of HE560's likely doesn't help, but they are much better after mods.

IMO

Can't say that I have encountered frequency response aberrations with quality DACS I've had on my desk, and think I've never had a "bright" dac. Once I deal with the f/r "personalities" of various cans, and to a lesser extent, amps, it's time for long dac listening sessions seeking timbre, texture, transparency, attack and damping, music in space (or perceived space).

But maybe my ears are too old.

 

[johnjen]

That is easily tested.

Bimby does not use the "supercomboburrito" filter on 24/192 material.

So, if you get a top quality 24/192 file ( such as Alan Yoshida's excellent mastering of "Blue Train" ) then you can eliminate the filter as a factor and just test the multibit implementation.

That's a really good distinction to bring up.

I think we used a 176.4 track (the Firebird Suite) and that was as high as we went.
I might be wrong on that but we didn't test for this specific condition.

So if the multibit architecture is a major contributor in and of itself, their results are all the more astonishing taking into account the VERY different hardware architecture between these 2 dacs.

Excellent!

JJ

 

[landroni]

That's a really good distinction to bring up.

I think we used a 176.4 track (the Firebird Suite) and that was as high as we went.
I might be wrong on that but we didn't test for this specific condition.

So if the multibit architecture is a major contributor in and of itself, their results are all the more astonishing taking into account the VERY different hardware architecture between these 2 dacs.

Excellent!

JJ

For info, here's what the Bifrost FAQ has to say:
http://schiit.com/products/bifrost
"How the heck can you do Bifrost Multibit so affordable? Gungnir Multibit and Yggdrasil cost a lot more!
Because it’s a much simpler DAC than either of them, using less parts, a more basic power supply, a single dual-channel DAC, and a 4X digital filter, rather than an 8X digital filter. However, it still offers amazing value, from its advanced bitperfect clock management system to USB Gen 2 input to upgradable, modular design.
 
What’s this about a non-oversampling (NOS) mode?
For 176.4kHz and 192kHz input sample rates, Bifrost Multibit passes the music right along—no digital filter, no oversampling. Have fun! Upsample lower-rate music with another algorithm, and compare to our “comboburrito” filter."
 
It looks like the burrito uses 4X oversampling in BMB, and 8X in GMB and Yggy. And no oversampling in both 176.4kHz and 192kHz in BMB.

 

[johnjen]

So while I work on the next few topics in the 'Better' series, I figured some might be interested in the tube tester I am repairing.

You may have heard of the Transcendent Audio Tube analyzer.
It is a rather interesting device that measures Gm (Transconductance), Mu (Amplification Factor) Ip (Plate Current) and from which Plate Resistance and Plate Dissipation can be calculated.

Between these measurements and characteristics, any tube can be described and even characteristic performance plots can be created.

And I don't even own any tube gear! :atsmile:

But I figure there might be those who live in the seattle area who would like their tubes tested and matched pairs checked etc.

But first I gotta find out why this unit won't analyze 12 volt and power tubes.
I've got a schematic and everything… :atsmile:

JJ

 

[atomicbob]

Any progress on the tube tester?

 

[duncan1]

Johngen it wont analyze  12 V tubes as it has only a 6.3 V supply. small double triodes that are dual heater/filament working are wired in parallel.

 

[johnjen]

Any progress on the tube tester?

I fussed with it last night and will test the results tonight.

Johngen it wont analyze 12 V tubes as it has only a 6.3 V supply. small double triodes that are dual heater/filament working are wired in parallel.

Um, the dual triodes do have dual filaments which are wired in parallel using their center tap on this tube analyzer.
To use a 12 volt filament supply the dual filaments would be connected in series.

JJ

 

[ericr]

So while I work on the next few topics in the 'Better' series, I figured some might be interested in the tube tester I am repairing.

...

But I figure there might be those who live in the seattle area who would like their tubes tested and matched pairs checked etc.

...

[Waving hand] Pick me, pick me!

There are a fair number of tubes I would appreciate having tested. Some 6SN7s as well as various 9-pin tubes - both 6 & 12 Volt (so I hope you're able to work that out).

-Eric