This is a known problem that Bryston is working on but so far without success. Therefore, I thought I put it out on the forum to see if anyone has any ideas for a solution. For example, perhaps there is an obscure setting in Roon that might work.
The problem is that my connection from the Roon Nucleus Plus to the Bryston SP-4 surround processor via HDMI will only upmix 2.0 channels to my 7.2 channel system; for example, 3.x and 4.x and 5.x channels will not upmix. This applies to both the FLAC and DSF files that I have in my library.
Gary Dayton, VP of Bryston actually visited my place some time ago and tried to solve the problem. He even phoned one of the engineers who agreed with his opinion that these files should be able to be upmixed if desired to 7.1 channels via this connection no matter the configuration.
I rarely use physical discs other than for system calibration or movies but recently had occasion to do so as my Roon core was inaccessible for a couple of days. I used some of that time to compare upmixing using physical discs via HDMI from my OPPO BDP-205 to the SP-4. The results were as follows:
1. 2.0 channels upmixed properly to 7.x
2. 3.0 channels would not upmix at all
3. 4.x channels would upmix to 6.x channels (no center channel)
4. 5.x channels would upmix to 7.x channels
This leads me to think that there might be something in the Roon settings that might be causing the problem as upmixing obvious can work more or less properly (the 3.0 channel discs or files being the most notable exception).
I mostly used Auro-matic for my tests as that is the one that almost always produces the best sound for me. However, I did also try it with Dolby Surround and DTS Neural X. The results were all the same.
OK. If you get the same results it is not one of those. I wonder if it how Roon packages the output. What number of channels is Roon set for outputting to the Bryston?
I know that #2 looks suspicious and I will try it set to “Yes”. It is a long time since I played around with the settings but if I remember correctly that was the setting that worked the best. To complicate matters, I have had major problems with Roon over the past 2 days and although most of the settings look O.K., I know that at least one setting was changed during the down time. Perhaps others were changed as well.
I will try changing #2 to “Yes” in a few minutes and get back to you in the next couple of hours.
Almost as soon as I finished the previous email, I realized that #2 above is irrelevant to this discussion. My apologies for wasting your time; it has been a very long and tiring day.
OK but 1. is relevant. It means that Roon is sending 7.1 to the Bryston all the time although many of the channels are empty/silent. The Bryston thinks they are all 7.1 and so will not upmix them.
Try changing Roon’s output format to 5.1 and see if that doesn’t help.
I tried a number of experiments and upmixing occurs for both 4.x and 5.x. The results are spectacular. For example I just listened to the superb Dutton Vocalion 4.0 SACD of Boulez conducting Stravinsky’s Petrushka and the listening experience was significantly enhanced. Thank you so much!
This is one of those cases where one cannot see the forest for the trees. The documentation is not exactly intuitive and indeed often ones eyes glaze over after reading a few sentences. This struck me as a situation where a fresh viewpoint was required. That is why I posed the question on this forum and that is why forums like these are so useful.
Unfortunately, similar to my experience with the physical discs, 3.0 surround material such as the wonderful RCA Living Stereo and Mercury Presence SACDs still won’t upmix. In this case, the choice is between having only the discrete front 3 channels in the 3.0 version as opposed to having the 7.x channel upmix from the 2.0 version. I usually favour the 2.0 version but I can see situations (e.g. recordings with prominent soloists such ad concerto recordings) where the 3.0 version might be preferred. I always hedge my bets by ripping both versions.
If someone with Roon is watching this thread, I wonder if they could investigate the possibility of treating the 3.0 material in a similar manner to the 4.0 material. That would truly be the icing on the cake if that could be done.