Welcome, Guest
Username: Password: Remember me
NOTE: This is a "Community" forum. Please be mindful that community members are here to help as part of a community effort. We therefore appreciate your effort in keeping this forum a happy place!

If you have a specific issue (e.g. hardware, failure) and want help from our support team, please use our tech support portal (Support menu - > Contact Us).
Thanks a lot of your help in making a better community.
  • Page:
  • 1

TOPIC:

MiniDSP Flex Jitter 1 week 5 days ago #61467

  • DanBenkman
  • DanBenkman's Avatar Topic Author
  • Offline
  • New Member
  • New Member
  • Posts: 3
  • Thank you received: 0
 
I'm overall impressed with my balanced output version of the Flex device.  I was also impressed with Audio Science Review's excellent review, including the low jitter they found.

But I found that the Flex sounded different and better when using its SPDIF input than when using its USB input.  This led to discussions and research of the sonic effects of jitter.

This led me to the purchase of a USB Bridge device with low jitter, called the Topping U90.  When connecting my laptop's USB output to the Topping U90's USB input, and the U90's output to the SPDIF input of the Flex, the sound was definitely different and better than with the Flex's USB input connected directly to the same laptop's USB output.

I'll admit that 'different' isn't necessarily 'better', but in this case the sound is pretty obviously better when using the U90 between my laptop and the Flex.  The sound is clearer, more defined, and has a wider and deeper soundstage -- precisely the kinds of things that lowered jitter is known to improve.

The question is why the AS review found so little jitter on the Flex's USB input.  Perhaps an older version of Flex firmware worked better (I have v1.25), or AS used an extremely good USB interface in a fast computer, whereas I have an older HP laptop.

Being an engineer, I looked inside the Flex and saw that there is no ASRC (Asynchronous Sample Rate Converter) chip, so it must be done in firmware inside the DSP, most likely using Analog Devices ASRC library code.  The question is if there may be a bug in their firmware, or if their ASRC code could be improved.

The problem with sample rate conversion is the sampling and interpolation process.  Jitter on the input could be 'interpolated into' the steady, precisely clocked output data .  If the effective internal software PLL (if there is one) lacks sufficiently low bandwidth, jitter on the input clock/data stream will end up as slightly 'off' values in the steady stream of interpolated data going to the digital filters and DACs.

Some expensive high end DACs claim excellent internal de-jittering.  The MiniDSP Flex has key features at a much lower cost (FIR filters and woofer-tweeter outputs among others) that even expensive DACs don't have.  But I'm sure I would not be the only Flex owner interested in lower jitter in the Flex.

In the meantime, I'll continue using my Topping U90 to send a low jitter signal to my Flex, and will continue to see the Flex as an amazing and useful product.

Thank you, devteam, for any comments.


 

Please Log in or Create an account to join the conversation.

MiniDSP Flex Jitter 1 week 4 days ago #61473

  • dreite
  • dreite's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Posts: 1572
  • Thank you received: 708
A conclusion based on subjective evaluation is hardly conclusive.
I hardly think recommending a USB/bridge that costs more than the Flex unit itself is a reasonable solution.  :)

You're an engineer with test equipment, yes?  Why not measure your own unit and see if its jitter rejection ability is compromised in some way?  It's possible the issue is external and associated with a substandard cable....or similar.

Dave.

Please Log in or Create an account to join the conversation.

MiniDSP Flex Jitter 1 week 3 days ago #61474

  • DanBenkman
  • DanBenkman's Avatar Topic Author
  • Offline
  • New Member
  • New Member
  • Posts: 3
  • Thank you received: 0
Dave,

I don't have equipment to measure jitter, that is pretty specialized and expensive equipment.

"A conclusion based on subjective evaluation is hardly conclusive."  This is relative.  Do you need a thermometer and rain gauge determine if it is cold and rainy or hot and sunny outside?  You can just step outside and observe the difference, correct?  Would you need to measure the frequency response of your audio system in order to realize that both your tweeters stopped working, or would you notice just by listening?

"It's possible the issue is external and associated with a substandard cable....or similar."  I've eliminated those possibilities to my satisfaction.  If such things are possible, is it not also possible that the ASRC code inside the DSP does not remove jitter as effectively as you think it does?  Not only is that a possibility, it seems that it is indeed a certainty -- and also not surprising.  Complex, external ASRC chips exist for a reason.

"I hardly think recommending a USB/bridge that costs more than the Flex unit itself is a reasonable solution."  You are correct, unless it is the ONLY solution.  I use the phase coherent crossovers and EQ inside the Flex, and 2X4 HD before it, and there is no cost effective substitute.

With an issue like jitter, the audibility of which is disputed, I accept the fact that you have already concluded that it is not an issue.  I accept the fact that you are convinced that what I'm hearing is all in my imagination.  On the other hand, from my point of view you can't possibly know how audible jitter is unless and until you do such a comparison for yourself, which no one at MiniDSP will probably ever do.

We thus have two sides with opposing opinions.  One side has investigated the issue, and the other side has not and will not.

Thank you for your reply, and no hard feelings.  MiniDSP has to run as a company.  Jitter is officially a non-issue.  Therefore, my email is a non-issue.

If you ever take a trip to Kuala Lumpur, Malaysia, I'll demonstrate the audibility of jitter and its removal for you.

Thanks for your reply,
Dan

Please Log in or Create an account to join the conversation.

MiniDSP Flex Jitter 1 week 3 days ago #61475

  • dreite
  • dreite's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Posts: 1572
  • Thank you received: 708
I didn't conclude anything one way or the other.  'Just a couple of comments/suggestions regards subjective and objective evaluations.

There's no need for the ridiculous gaslighting.

Have fun.

Dave.

Please Log in or Create an account to join the conversation.

MiniDSP Flex Jitter 1 week 3 days ago #61476

  • DanBenkman
  • DanBenkman's Avatar Topic Author
  • Offline
  • New Member
  • New Member
  • Posts: 3
  • Thank you received: 0
I am having fun.  Your command of colloquial English is very good.

The lesson learned is that those of us concerned with jitter on the Flex (whether it's all in our imaginations or not) need solutions external to the Flex.  Wouldn't you agree?

Please Log in or Create an account to join the conversation.

  • Page:
  • 1