Schiit Fire and Save Matches! Bifrost Multibit is Here.
Oct 20, 2015 at 3:25 PM Post #751 of 2,799
  I think I'm experiencing something similar.
When I first tried to play a 96/24 file it was terribly distorted. A MB reboot (off/on) fixed the problem.
 
This morning (after playing all night) the sound wasn't right. Etched, grainy, hard on the ears. Another reboot also fixed the problem.
Macbook -> wyrd -> Bifrost MB.
 
Don't understand the cause at this point.


Reset the SMC, try a different USB cable (non-audiophile, USB 2.0, 2M or less in length.)
 
If that doesn't work, turn off App Nap system-wide: https://www.ihash.eu/2014/12/disable-app-nap-system-wide-os-x-yosemite/
 
Schiit Audio Stay updated on Schiit Audio at their sponsor profile on Head-Fi.
 
https://www.facebook.com/Schiit/ http://www.schiit.com/
Oct 20, 2015 at 3:30 PM Post #752 of 2,799
I think I'm experiencing something similar.
When I first tried to play a 96/24 file it was terribly distorted. A MB reboot (off/on) fixed the problem.

This morning (after playing all night) the sound wasn't right. Etched, grainy, hard on the ears. Another reboot also fixed the problem.
Macbook -> wyrd -> Bifrost MB.

Don't understand the cause at this point.
[/quote

Exactly what I ran into last night and this morning. The on/off fixed it and it was fine for the next near hour before I left for work. I'm just running a coax from my soundcard on a Windows 7 PC to the Multifrost. Would drivers still possibly be an issue if I'm not using USB? We'll see what happens tonite. Sounds incredible when it's working right. Never had a comparable issue with Uber...but it sure didn't sound nearly this good with Uber either. I'll keep my fingers crossed. Any advice, let me know.
 
Oct 20, 2015 at 3:33 PM Post #753 of 2,799
Oh I didn't see Jason's response. Will that work with both coaxial and USB?
What was interesting was when I turned it off/on this morning after it being on for about
12 hours since I got it back from Schiit, the 3 lights kept going on and off in sequence for a good 45 seconds
before it finally stopped on the coax light. Then it worked fine. Any thoughts on that? Like I said, no distortion
on any flac files after that for the hour or so I listened.
 
Oct 20, 2015 at 3:40 PM Post #754 of 2,799
I understand Different OS have different requirements and this was pretty painless (minor frustration) , A minor frustration can lead to some malcontent with the product, I am not convinced blaming windows would be my route to explain why my product was having difficulties

Get the frustration but Windows does in general incur the majority of reported gremlins. This is not even limited to computer audio. How many forum feedback threads do you read where the poster has viewing issues due to Windows or ms browser
 
Oct 20, 2015 at 3:40 PM Post #755 of 2,799
A tad over 1 hour in and the Bit Frost is singing cleanly
 
Pitch black and zero noise at max volume while the Havana can produce some tube hiss
 
Bifrost 2  Havana 1
 
 
 
Initial listening says as far as Sound Quality goes the decision will not be an easy one, and the final keeper decision may end up being features........................Going to take some time with both before commenting on sound impresssions  
 
Oct 20, 2015 at 3:43 PM Post #756 of 2,799
 
Reset the SMC, try a different USB cable (non-audiophile, USB 2.0, 2M or less in length.)
 
If that doesn't work, turn off App Nap system-wide: https://www.ihash.eu/2014/12/disable-app-nap-system-wide-os-x-yosemite/

Thanks for the suggestions.

I'll try the SMC reset and app nap tonight.
 
I use a 3M cable from the macbook -> the wyrd. And <1 M from the wyrd to bifrost MB.
I can try a different wyrd -> bifrost USB cable, but don't really have a choice regarding the 3M usb cable.
This exact USB setup has been working flawlessly with the uber for a while.
 
Oct 20, 2015 at 3:45 PM Post #757 of 2,799
Get the frustration but Windows does in general incur the majority of reported gremlins. This is not even limited to computer audio. How many forum feedback threads do you read where the poster has viewing issues due to Windows or ms browser

Me personally I dont recall seeing any one complain of viewing issues.............I must admit though I only frequent a few threads that interest me or concern my personal voyage 
 
For example this thread, I am only on here to learn more about Multi-Bit and improving my DAC (or not) 
redface.gif

 
Oct 20, 2015 at 3:55 PM Post #758 of 2,799
   
And, to be fair to Windows, there are certainly frustrations on the Mac side as well. Macs have some of their own power management issues (hello, App Nap.) Disabling power management on Mac USB ports may or may not "take," as with Windows. And now they're moving into some freakish netherworld where a single USB 3.1 connector is all you need (?!)

 
Yea, saw the 3.1 USB "only" connector while shopping for a laptop for my granddaughter.  Steered away from that Mac and stayed with the Air that has 2 x 3.0.  Seemed like a safer bet...
 
Oct 20, 2015 at 4:08 PM Post #759 of 2,799
For what it's worth, the only way I could get the drivers to install on Windows 7 when I moved my Bifrost to my work system was to both be running as an Administrator AND then launch the installer using the "Run As Administrator" option.
 
Prior to that the installer wouldn't even detect the Bifrost, even though Windows had already tried to install a default driver (and only half succeeded).  This was true even after scrubbing all traces of the initial driver and restarting.
 
Oct 20, 2015 at 5:23 PM Post #760 of 2,799
When my Bimby arrived at work yesterday I plugged it into USB on my Win7 pro 64 bit PC and Windows asked if I wanted it to find the drivers, I was busy and for a laugh I let it; to my surprise it went through an automatic multi-stage process of installing drivers finishing with the Schiit Gen 2 - it all worked just fine.
 
When I got it home and plugged it into my Win10 PC nothing much seemed to happen, the lights cycled round for a while but no sign of recognition. I went off for a while and when I got back no sign of Windows activity, went into foobar hoping - and there it was - I guess it was Windows plain vanilla driver but it all worked OK.
 
The only real complaint was that I got nothing done because I was listening to music until past bed time. 
redface.gif

 
Thank you Schiit.
 
(I guess I'll try to load the Schiit drivers at some point)
 
Oct 20, 2015 at 6:09 PM Post #761 of 2,799
OK a few hours in now, the Bit Frost does sound great, But it has a slight edge above what the Havana presents (causing some slight fatigue)....................Everyone states the Bifrost is basically broken in after 4 hours, I am at about 5 hours now of straight playing while my listening has been on and off
 
Without the Havana setting right here next to the Bit Frost I doubt anyone would notice  They are VERY close.  If I had to guess I would bet the NOS western Electric tubes are the difference in the slight edge
 
Oct 20, 2015 at 6:16 PM Post #762 of 2,799
Oct 20, 2015 at 7:23 PM Post #763 of 2,799
The original high standards of digital audio were 16 bit at either 44.1KHz or 48KHz sampling rate.  These standards still exist today.  Hundreds of thousands of these tracks exist and are still being recorded today.
 
Today, there are a number of D/A converter chips and A/D converter chips that brag about 24, or even 32 bits of output bit width (A/D converters) or input bit width (D/A converters).   The promise of more bits is alluring, but is only a hand job at best.  Higher numbers of bits have become new buzz numbers.
 
The heart of the lie is that all of the sd A/Ds or ds D/A chips have digital filters; all high res recordings not natively recorded, and many if not most of the high res recordings are all recorded with digital filters, which are required so the recording will not violate sampling theory. Kinda like physics.  Like why it hurts when you crash cars, etc.  Similarly, all oversampling D/A converters have some kind of digital filter. There are also “hidef” recordings natively recorded live or from analog at higher sampling rates – those are cool. The “hidef” recordings made from original redbook digital sources – they go through someone else's filter.
 
Now all of these filters multiply samples of the original by a series of coefficients; this is typically thousands of multiplications and as they go, the numbers get longer and longer.  If the calculated coefficients start out at 24 bits, the numbers get longer from there.  All of the above is irrefutable.  So we get bonus bits that may be beneficial, according to how good the filter is.  This is an important point - the bonus bits can be slop (always the case when the original bits are discarded) to boring, or occasionally even really good.  Most of the filter derivations do not scratch my sonic itch.
 
So all filters do not sound the same.  The megaburrito filter has a time domain and a frequency domain optimization.  The time domain dating from Bell Telephone Labs (1917) and the frequency domain from a much later evolution of a similar filter from the same era.  If you buy a Schiit multibit product, you get that filter.  It optimizes more than anyone else’s, and keeps the original recording in its entirety.  A lot of people really like it.  I bet with the Yggy, Gumby, and Bimby that you may well like it a lot, particularly with redbook recordings which have not been sullied by other filters.  In any kind of system.  
 
Back to the bits - I know of no DAC that decodes any kind of reality beyond 20 bits and those get really expensive (Like the Yggy).  You can use two per channel and get 21 real, honest to God bits.  The megaburrito filter's DSP chip runs at 32 bits - not that is any bragging right other than it puts the computational errors at least 12 bits below the rounded output of Yggy, 14 bits below the rounded output of Gumby, or 16 bits below the rounded output of Bimby.
 
Numbers for numbers sake are fairy tales translated into marketing.
 
Schiit Audio Stay updated on Schiit Audio at their sponsor profile on Head-Fi.
 
https://www.facebook.com/Schiit/ http://www.schiit.com/
Oct 20, 2015 at 7:38 PM Post #764 of 2,799
  The original high standards of digital audio were 16 bit at either 44.1KHz or 48KHz sampling rate.  These standards still exist today.  Hundreds of thousands of these tracks exist and are still being recorded today.
 
Today, there are a number of D/A converter chips and A/D converter chips that brag about 24, or even 32 bits of output bit width (A/D converters) or input bit width (D/A converters).   The promise of more bits is alluring, but is only a hand job at best.  Higher numbers of bits have become new buzz numbers.
 
The heart of the lie is that all of the sd A/Ds or ds D/A chips have digital filters; all high res recordings not natively recorded, and many if not most of the high res recordings are all recorded with digital filters, which are required so the recording will not violate sampling theory. Kinda like physics.  Like why it hurts when you crash cars, etc.  Similarly, all oversampling D/A converters have some kind of digital filter. There are also “hidef” recordings natively recorded live or from analog at higher sampling rates – those are cool. The “hidef” recordings made from original redbook digital sources – they go through someone else's filter.
 
Now all of these filters multiply samples of the original by a series of coefficients; this is typically thousands of multiplications and as they go, the numbers get longer and longer.  If the calculated coefficients start out at 24 bits, the numbers get longer from there.  All of the above is irrefutable.  So we get bonus bits that may be beneficial, according to how good the filter is.  This is an important point - the bonus bits can be slop (always the case when the original bits are discarded) to boring, or occasionally even really good.  Most of the filter derivations do not scratch my sonic itch.
 
So all filters do not sound the same.  The megaburrito filter has a time domain and a frequency domain optimization.  The time domain dating from Bell Telephone Labs (1917) and the frequency domain from a much later evolution of a similar filter from the same era.  If you buy a Schiit multibit product, you get that filter.  It optimizes more than anyone else’s, and keeps the original recording in its entirety.  A lot of people really like it.  I bet with the Yggy, Gumby, and Bimby that you may well like it a lot, particularly with redbook recordings which have not been sullied by other filters.  In any kind of system.  
 
Back to the bits - I know of no DAC that decodes any kind of reality beyond 20 bits and those get really expensive (Like the Yggy).  You can use two per channel and get 21 real, honest to God bits.  The megaburrito filter's DSP chip runs at 32 bits - not that is any bragging right other than it puts the computational errors at least 12 bits below the rounded output of Yggy, 14 bits below the rounded output of Gumby, or 16 bits below the rounded output of Bimby.
 
Numbers for numbers sake are fairy tales translated into marketing.

I like the fact that frequency and time domain are both optimized.  I've always found that phase/time is as important to frequency response.  Ruler flat in one domain can cause havok in the other.  From my days in digital communications and sound repro, I've learned how much phase can either help or screw up everything.  I'll gladly live with some f ripple to make the time alignment more kosher.
 
Oct 20, 2015 at 8:17 PM Post #765 of 2,799
   
Reminds me of the old Onion classic:
 
Apple Introduces Revolutionary New Laptop With No Keyboard
http://www.theonion.com/video/apple-introduces-revolutionary-new-laptop-with-no--14299


Isn't that an iPod Touch or iPad? :wink:
 

Users who are viewing this thread

Back
Top