What a long, strange trip it's been -- (Robert Hunter)
Mar 26, 2018 at 4:54 PM Post #7,456 of 14,566
Finally a new direction in protos I am building. As I mentioned on the Facebook stream the other day, I am taking a look at CAB (cheap-assed balance) technology as to explore new ways of providing value for those who insist on balanced systems.

The best way to balance in the digital domain is to use two separate DACs per channel. Since digital circuits generate horrible amounts of high frequency noise, a DAC for the other phase of the balanced signal converts all of that noise to common mode and cancels it out.

A much cheaper way is to balance the D/A converter in analog only, the way the majority of our competing balanced units do. This gets rid of the cost of a balanced (including DAC chips) digital section, which is not trivial, considering the parts we use. I’ll be looking at this more seriously in the next few months for less expensive balanced designs. Let it go without saying that the Yggy and Gumby are both fully digitally balanced (hard-way).

Does analog domain CAB compromise single ended performance (e.g. adds more electronics in the signal path to get to the CAB signal) or does single ended performance remain the same even when CAB is implemented? I like the way you are doing it now, though I had hoped for an "Uber?" true balanced biforst. I always felt those "only" 16-bit dacs can provide more performance than most people expect based solely on the number of bits.
 
Mar 26, 2018 at 4:59 PM Post #7,457 of 14,566
I've been thinking on this black (also read "clean") background a long time with video. I was a very early adopter of both BRD and HDVD as well as all of the earlier video formats. I remember when some films were "digitized" so they could remove the film grain and then "edge enhanced" outlines. Who the hell came up with that idea? And for who? It completely destroyed the look of those films.

It never dawned on me that something along those lines could crapify recorded music as well but your point is well taken by me.

My hope is that my plasma monitor will outlive me so I don't have to deal with the latest and greatest screen technologies.


I remember when THEY started colorizing classical B&W films. You couldn't just add any color, it had to be bright enough to overpower the shade of grey.
When I saw Jimmy Stewart about to commit suicide in "It's a Wonderful Life" wearing a lavender suit, I finally understood his need to die. Fortunately THEY were stopped and those abominations purged from the vaults.
 
Mar 26, 2018 at 7:04 PM Post #7,458 of 14,566
Question about Dither?
I use a laptop running JRiver vs 23 into a Yggdrasil, Should I have the dither option checked on the JRiver software? Reading Mike's post mentioning the dither in the Yggy makes me wonder if I should leave the dither option unchecked. I listen to 90% 16/44 cd burns to the harddrive. The other 10% is high definition.
Thanks
 
Mar 26, 2018 at 7:26 PM Post #7,459 of 14,566
Dithering should be used when changing the bit depth of a signal, to shape/spread to rounding error.

If you have a bit-perfect chain, then leaving the option off will let the DAC perform its own dithering.
If you have a Yggy A1, then you might want to play with the dithering options of JRiver.
If you have an A2, you should leave it off as the DAC has its own (improved) dithering.
 
Mar 27, 2018 at 12:24 AM Post #7,460 of 14,566
I'm convinced the greatest living pianist, sparing Mrs. Argerich's blushes, is Kit Armstrong
Thanks, I wasn't familiar with Kit Armstrong and it's been fascinating as I learn about him. He's a genuine genius —started composing music at age 5 and began college at 9. Earned a masters degree in math in addition to studying music. In different videos, he is seen speaking English, French, German, and Mandarin.



As an 8 year old prodigy...
 
Mar 27, 2018 at 12:40 AM Post #7,461 of 14,566
Thanks, I wasn't familiar with Kit Armstrong and it's been fascinating as I learn about him. He's a genuine genius —started composing music at age 5 and began college at 9. Earned a masters degree in math in addition to studying music. In different videos, he is seen speaking English, French, German, and Mandarin.



As an 8 year old prodigy...

Heilige Scheiße...
 
Mar 27, 2018 at 11:40 AM Post #7,462 of 14,566
Dithering should be used when changing the bit depth of a signal, to shape/spread to rounding error.

If you have a bit-perfect chain, then leaving the option off will let the DAC perform its own dithering.
If you have a Yggy A1, then you might want to play with the dithering options of JRiver.
If you have an A2, you should leave it off as the DAC has its own (improved) dithering.
JRiver does internal bitdepth conversion to 64 bit and back lower as it likes to output in the highest bitdepth the sound device driver allows. For Yggy I believe it's 24 bits. Because the Yggy is 20/21-bit the internal conversion will take place regardless of what you send to it (I suppose you can't send a 20-bit stream). So, the conclusion is the dithering should always be turned on in JRiver. And dithering will take place in the Yggy anyway, thank Mike. And dithering on top of dithering won't do any harm. If you understand what dithering is it will be obvious.
 
Mar 27, 2018 at 12:05 PM Post #7,464 of 14,566
I believe JRiver uses dither only to distribute the error from any processing. This is not 'noise' that is added ...it's just rounding errors when going from 64 bits to the DAC bit resolution (say, 24bits) are not dropped but are maintained and diffused across multiple samples.
In my resolving system I can hear dither on/off and recommend always on and set to TPDF.
 
Mar 27, 2018 at 1:11 PM Post #7,465 of 14,566
In other news, work continues on the Windoze driver for our own USB implementation. Which otherwise is running with a few bugs, but works!

Hey Mike @Baldr - glad to see you're back and blogging again for those of us that like to follow what you're up to. Also, for someone who went through so much recently, you were looking well in the video that you and Jason did for the Lyr 3 launch.

On the topic of your USB implementation, I'm wondering what you're anticipating from non-computer streamers/sources that output USB (e.g. network audio adapters, full network player devices, etc) - since the user can't just install any 3rd-party driver for these devices, are you expecting some compatibility issues with these devices, or do they generally work with the solution you're building?
 
Mar 27, 2018 at 2:07 PM Post #7,466 of 14,566
JRiver does internal bitdepth conversion to 64 bit and back lower as it likes to output in the highest bitdepth the sound device driver allows. For Yggy I believe it's 24 bits. Because the Yggy is 20/21-bit the internal conversion will take place regardless of what you send to it (I suppose you can't send a 20-bit stream). So, the conclusion is the dithering should always be turned on in JRiver. And dithering will take place in the Yggy anyway, thank Mike. And dithering on top of dithering won't do any harm. If you understand what dithering is it will be obvious.

Set the output in your software to 32 bit and disable dithering. Let the DAC handle it.
 
Mar 27, 2018 at 3:05 PM Post #7,467 of 14,566
FWIW, I listen to a lot of electronic music on my Bimby and it sounds great too. So just do your thing, @Baldr. :)
I also think the Bimby sounds great, listening to mostly rock and metal, so no complaints there. I was just commenting that at least for the stuff I'm listening to, or with my ears, I have hard time distinguishing the modi2uber from mimby or bimby. Did grab a copy of Clapton unplugged as was suggested above, so will check that out soon. (I only had the Crossroads box set as far as any Clapton I already had)
 
Mar 27, 2018 at 5:09 PM Post #7,468 of 14,566
That album has a track (Walkin' Blues) that has monster kick a** ba**.
He is thumping the riser on the stage with his foot, which is acting like a drum head.

With the MB dacs recognizing the sound of his foot hitting the stage floor is much easier, even on such a simple song (single guitar and voice).

JJ
Since you wrote that, JJ, I had to go back and listen closely...
Note, I just have a GMB, not an Yggy, listening with my BHA-1 and HD800 (unmodded).
To me, it clearly sounds like Eric's left foot, the heel (not toe), size 11.5, possibly Cole Haan.
The shoe brand is the toughest to hear. With an Yggy, you may be able to spot the brand. :)

ps. There is a ton of background noise in this recording, i.e. about which Mike just wrote. Adds to the experience!
 
Last edited:
Mar 27, 2018 at 5:27 PM Post #7,469 of 14,566
Since you wrote that, JJ, I had to go back and listen closely...
Note, I just have a GMB, not an Yggy, listening with my BHA-1 and HD800 (unmodded).
To me, it clearly sounds like Eric's left foot, the heel (not toe), size 11.5, possibly Cole Haan.
The shoe brand is the toughest to hear. With an Yggy, you may be able to spot the brand. :)
According to my Yggy, he cleaned the shoe (only that one) 2.5 days before the recording, with his spit, right after eating a burrito.
 
Mar 27, 2018 at 5:33 PM Post #7,470 of 14,566
According to my Yggy, he cleaned the shoe (only that one) 2.5 days before the recording, with his spit, right after eating a burrito.
Nice! Can you spot which brand of hot sauce...Louisiana or Tabasco?
 

Users who are viewing this thread

Back
Top