H9K – SPDIF 96Khz BUG

[macgee]

Hi Eventide

When the SR is set to 96Khz, I don’t get audio passing on the SPDIF, only lots of clicks. I’ve spent ages troubleshooting everything including cables but finally when I change the SR to 48Khz then audio is passed. I didn’t test other SR but from my testng this is a BUG

Can you please confirm?

Thanks

[gkellum]

Ok, I tried to reproduce this with two H9000s but I didn't hear any clicks or other distortions.  There are a couple of different ways to set this up though.  Here's what I did:

* The two H9000s were connected by SPDIF over optical.  It's possible to also connect them via SPDIF over RCA.

* One of the H9000s was the clock master, and the slave H9000 had its clock synchronized via a word clock cable.

Was your set-up different than this in some way?

Some other possibilities to keep in mind are:

* Not all algorithms run at 96K.  If you're starting with a fresh session and you load an algorithm that's not supported at 96K, both the H9000 and Emote will warn you of this, but if you switch an existing session from 48K to 96K, you won't necessarily see that there's a problem if you're only looking at Emote.

* Some users have reported they needed to restart their H9000s with the SESSION and FXCHAINS buttons pressed to clear out the H9000s current state to get it to switch over to a different digital audio mode and/or a different sample rate when using digital audio.  This has come up a few times with running digital audio at 96K.

[macgee]

Thanks for the feedback

I didn’t include enough details now that I look at it.

My setup was H9K SPDIF RCA OUT to UAD Apollo X IN and UAD Apollo X SPDIF RCA OUT to H9K SPDIF RCA IN.

The H9K was master and the UAD Clocked to H9K SPDIF RCA IN – The Apollo was locked to the H9K.

Apollo doesn’t support SPDIF over Optical, only over RCA or ADAT

These clicks were with all processing completely bypassed. Switching back to 48Khz it worked again. I was just swithing SR while in an existing session and only ever used SPDIF RCA.

I’ll get a wordlock cable and try clocking that way, you think that would make a difference?

I’ll also try testing out these other scenarios like selecting a different session that is 96k or resetting session get back to you.

Thanks

 

 

[nickrose]

Note that S/PDIF does not always work well at 96K due to ground noise (AES is balanced and floating so does not have this problem).

Try and keep your cables short and high quality, and ground the units to the same point (usually a common power strip).

 

[gkellum]

Ok, I changed our test set-up to more closely match your configuration.  I connected two H9000s with S/PDIF over RCA where one was the clock master and the other was the clock slave and was receiving clock over the S/PDIF connection.  Both H9000s were running at 96K of course.  I didn't hear any audible distortion or clicking in this configuration.  

To Nick's point, the cables we have connected are relatively short and they are cables that were manufactured with the correct impedance for S/PDIF.  Not to say that your cables don't match these specs though of course; your cables might be perfectly fine…

 

[macgee]

Ok thanks guys – I’m travelling at the moment so when I’m back I will test this more thoroughly to get to the bottom of it.

Your support is appreciated

[Author]

Posts