[KLJTech]

  I just got my Multibit in, but is it normal that the input is this far off center? https://www.dropbox.com/s/or4emk79hwjwm0c/20160823_142801.jpg?dl=0

 
I have components that cost well over 2K that the connectors/jacks on the back of the case are not "perfectly" centered within the hole on the back of the case. It doesn't matter at all, the connectors/jacks are soldered to the board inside and can't always line up perfectly with the hole on the back of the case. No worries.
Enjoy your music! 

 

[iJimmy]

   
I have components that cost well over 2K that the connectors/jacks on the back of the case are not "perfectly" centered within the hole on the back of the case. It doesn't matter at all, the connectors/jacks are soldered to the board inside and can't always line up perfectly with the hole on the back of the case. No worries.
Enjoy your music! 

All this time I was not using a DAC. I just plugged in the Modi Multibit, connected to my Vali and TH-600. Holy Schiit.

 

[KLJTech]

  All this time I was not using a DAC. I just plugged in the Modi Multibit, connected to my Vali and TH-600. Holy Schiit.

 
Life is good, enjoy your music! 

 

[iJimmy]

   
Life is good, enjoy your music! 

Will do! I think I know what I'm doing for the rest of the day... Does make me wonder if I'd notice the difference between the Multibit and the Modi 2 for example (never heard it).

 

[KLJTech]

Since you already have the multibit you don't have to worry about it, with headphones or good speakers I imagine you would indeed hear the difference. I've never heard either of those DAC's so I can't be sure...I wouldn't worry about it as I'd be too busy listening to my favorite music. Congratulations on your new DAC!  

 

[matthewpartrick]

Modi and Vali descended today; setting up now and will post impressions here later.
 
Thx!

 

[srisaikat]

  I just got my Multibit in, but is it normal that the input is this far off center? https://www.dropbox.com/s/or4emk79hwjwm0c/20160823_142801.jpg?dl=0

 
I believe it is not quite normal, the insulation or plastic buffer (from body/chassis) around the coax input is either missing or misplaced. You should send picture and confirm from the Schiit itself.

 

[iJimmy]

   
I believe it is not quite normal, the insulation or plastic buffer (from body/chassis) around the coax input is either missing or misplaced. You should send picture and confirm from the Schiit itself.

They said it's correct!

 

[jamor]

  I just got my Multibit in, but is it normal that the input is this far off center? https://www.dropbox.com/s/or4emk79hwjwm0c/20160823_142801.jpg?dl=0

 
They are probably all like that.  I've had 2 Modis now and both had it so I guess it's normal.

 

[yage]

  Think of if this way. The original samples are (x,y) points where x is the time axis and y the amplitude for the sample. Suppose now that that x0 and x4 are the times for two consecutive 48 kHz samples (meaning that x4 - x0 == 1/48000 of a second), with amplitudes y0 and y4. To oversample 4x, we need to add 3 new equally spaced synthetic samples (x1, y1), (x2, y2), and (x3, y3) with x4-x3 == x3-x2 == x2-x1 == x1-x0 == 1/192000 of a second. The Schiit filter computes y1, y2, and y3 so that the resulting waveform minimizes additional time and frequency distortion from an ideal waveform that would have produced the original samples. Most other oversampling setups are unable to do this without also adjusting y0 and y4, thus failing to preserve the original samples.

 
Are you sure this is what they do? y1 -> y3 would seem to be interpolated values then. Why wouldn't they just zero-pad the data?

 

[earnmyturns]

   
Are you sure this is what they do? y1 -> y3 would seem to be interpolated values then. Why wouldn't they just zero-pad the data?

The R2R chips need to get a y for each target rate sample to produce a corresponding analog amplitude. The synthetic samples y1, ..., y3 need to be chosen to yield amplitudes that lead to minimal distortion from the ideal source waveform, meaning in particular that their relationship not only to y0, y4 but to the samples before y0 and after y4 -- both the original ones and the interpolated ones --must minimize spectral and phase distortions from the ideal source waveform. Thus you need a filter that ensures those relationships, not just a simple local interpolator. I'm not a digital filter expert, but I know a bit about interpolation from my day job.

 

[jimmers]

When I first read earnmyturns (original) explanation I didn't think it likely to help anyone who didn't already know what was happening
 
Quote:

Originally Posted by yage /img/forum/go_quote.gif
 
Are you sure this is what they do? y1 -> y3 would seem to be interpolated values then. Why wouldn't they just zero-pad the data?

I guess this supports the supposition, 
The values y1->y3, as you put it, are interpolated - it's how the interpolation is calculated is the thing. 
 
I don't understand what zero-padding the data has to do with it.
 
edit. cross posted.

 

[Tuneslover]

I have been burning in my Modi MB during the night throughout our vacaction in Tennessee and North Carolina and have about 52 hours on it now. we are on our way home now and should be home by Wednesday night. I'll plug it in overnight when we get home and give it a comparative listen to my Modi original, as well as, my Bifrost 4490 and Bifrost MB the next day. Really missing my X5ii coaxial out > Bifrost MB > HD650 setup.

 

[earnmyturns]

When I first read earnmyturns (original) explanation I didn't think it likely to help anyone who didn't already know what was happening

Fortunately I stopped working as a professor over 8 years ago, so no one is paying to be confused by me 

 I have this bad habit of expecting everyone to have learned and retained those wonderful Cartesian x's and y's from secondary school, and maybe a teeny bit of trig to visualize a waveform in Cartesian coordinates... I stand defeated.

 

[slex]

This just arrived. Gonna squeeze the juice out hella of this little mimby