Sibelius / NotePerformer 4 Technical Issue?

krummholz · July 14, 2024, 07:57:40 AM

I am absolutely stymied.

On 21 June, two things happened on my system: (1) Sibelius suddenly hung on startup, a problem that could not be fixed even by trashing the file containing my global Sibelius settings. I was about to give up, when I was informed by Avid of a new Sibelius release, which (2) I promptly downloaded and installed. And voila! Sibelius started up normally and all appeared to be well. I had started a support dialogue with an Avid chatbot named "Alex", which I promptly abandoned when it was clear that (a) the problem was fixed and (b) the bot's recommendations were unlikely to be successful if I downgraded back to the previous Sib version to try them, as its main recommendation was indeed to trash the settings file.

But then I started noticing differences in the instrumental timbres, specifically of the solo strings, and especially in renderings of my Sinfonia Solenne, of which I was in the middle of finalising the postlude. The violins now sounded more distant, the cello more "present" and reverberant. I assumed that I had inadvertently made some change, perhaps via a MIDI message or forgetting to undo a con sord, in the score. But this is definitely NOT the explanation as I carefully checked the score itself and the mixer and Performance settings. More to the point, the change does NOT affect renderings made before 21 June, and it affects ALL renderings made with the score since then. For clarity: I can make a new rendering with the exact same version of the score that I used for a pre-21 June rendering, and the solo strings sound will now be different from that in the pre-21 June rendering.

After discovering the issue, I found an update to NotePerformer 4.5.1 and installed it. No change.

The renderings that I've posted here of the most recent version of Sinfonia Solenne (Symphony No. 1 in C Minor) were all spliced together from multiple renderings using Audacity, and the passages that involve solo strings were all taken from renderings made before 21 June.

Obviously, because I made two different changes to my Sib installation on 21 June (trashed my settings file and updated Sib to version 2024.6, build 4260), I have no way to tell which one caused the change in timbre. Since the same score has produced different results, it cannot be due to a setting change made in the score (the .sib file) and would have to be a global setting in Sibelius. Is there a global setting that would affect the sound of NP4's solo strings, and (apparently) no other instruments? Or was there some change in the way Sib interacts with NP in the latest Sib release? To answer this question, I typed in a support question on Wallander's website. The support tech who responded was Alexander (no relation to chatbot Alex, I'm sure!) At his request, I sent them both the .sib file and the two clips linked to below, for comparison. That was last Wednesday and I have heard nothing from him since, even after asking if there was any news early Friday. This makes me wonder if the Wallander folks concluded that I was either deluded or a troll.

You be the judge: is it my imagination, or are the timbres and instrumental balance different between the two versions? The difference IS subtle, even to me, but I believe it is there, and there is enough difference that I would not want to splice new renderings with old ones. If this is not my imagination, I'm effectively stuck spinning my wheels for now. This is making me seriously consider moving to Dorico.

"Before" audio file

"After" audio file

krummholz · July 14, 2024, 12:57:27 PM

The tl;dr version: could someone please listen to the two (short) audio clips linked in the OP and tell me if you notice a difference in the timbre and balance of the instruments? You shouldn't need even to listen to each clip to the end: to me, the difference is apparent as soon as the solo strings enter.

lunar22 · July 14, 2024, 11:26:58 PM

to be honest, despite trying to focus on what you mentioned, I can't hear any significant difference. Any differences between how things sound in Sibelius and Dorico are likely to be as Dorico will use a different reverb (is it possible that setting has changed in Sibelius?) but whenever I've compared NP Sibelius output against Dorico, there's otherwise nothing really of significance to separate the two from memory.

As you know, I'd recommend moving to Dorico for a host of reasons-- not least the generally excellent community support. If you want better editing control over the NP output, then Dorico is also better but out of the tin, the differences should not be very significant.

krummholz · July 15, 2024, 05:19:03 AM

Quote from: lunar22 on July 14, 2024, 11:26:58 PMto be honest, despite trying to focus on what you mentioned, I can't hear any significant difference. Any differences between how things sound in Sibelius and Dorico are likely to be as Dorico will use a different reverb (is it possible that setting has changed in Sibelius?) but whenever I've compared NP Sibelius output against Dorico, there's otherwise nothing really of significance to separate the two from memory.

As you know, I'd recommend moving to Dorico for a host of reasons-- not least the generally excellent community support. If you want better editing control over the NP output, then Dorico is also better but out of the tin, the differences should not be very significant.

Thanks for the opinion. I just received an email from Alexander @ Wallander - apparently they were not working Thursday and/or Friday as he just received my attachments, but it seems likely then that they will not notice any difference either.

To me it is night and day, and independent of the audio system I listen with - non-audiophile quality headphones on computer vs. audiophile amplifier / headphones. I burn renderings onto CD so that I can hear what NP output sounds like on a dedicated stereo system, and that string quartet passage in newer renderings stuck out like a sore thumb to me - that was before I had an inkling that there might be a technical issue with Sib.

But I guess it could be my imagination.

Yesterday I decided to begin the crossgrade process to Dorico, and this morning they "approved" my request, based on my Avid documentation. I'll wait for an opinion from Wallander before proceeding.

krummholz · July 15, 2024, 05:22:05 AM

Oh, and no, the reverb *setting* is exactly the same. (It is possible, of course, that Avid changed the *semantics* of each reverb setting in their latest release.)

lunar22 · July 15, 2024, 05:29:46 AM

I have quite good Dynaudio speakers which I only tried out after struggling to hear the difference on headphones. But you should wait for other opinions and if others can hear and describe a clear difference, I'll try again! Not having an up to date version of Sibelius, I couldn't comment on any possible changes to the reverb -- in that department, Dorico uses a standard Steinberg plug-in also used by Cubase so that wouldn't change with Dorico versions (and anyway, I use my own)

Perhaps I'll have the pleasure of welcoming you on the Dorico forum before long!

krummholz · July 15, 2024, 05:37:54 AM

Quote from: lunar22 on July 15, 2024, 05:29:46 AMI have quite good Dynaudio speakers which I only tried out after struggling to hear the difference on headphones. But you should wait for other opinions and if others can hear and describe a clear difference, I'll try again! Not having an up to date version of Sibelius, I couldn't comment on any possible changes to the reverb -- in that department, Dorico uses a standard Steinberg plug-in also used by Cubase so that wouldn't change with Dorico versions (and anyway, I use my own)

Perhaps I'll have the pleasure of welcoming you on the Dorico forum before long!

Yes, I'm going to wait - at least to see what Alexander says. I'm a little sick of Avid's customer service and licensing headaches regardless, but the Dorico crossgrade is still $300 which is not an insignificant sum.

As to the timbre and balance difference: I'm not sure anyone would be able to describe clearly what the difference is. I do NOT notice any difference in the solo violin "mini-cadenza" that occurs later in the work (before letter GG), and just created a splice last night where the preceding solo strings passage was from pre-June 21 and the "cadenza" was rendered yesterday. No perceptible difference at all. It seems that more than one solo stringed instrument has to be playing before the difference becomes noticeable.

I had the thought that the left and right channel outputs might be out of phase in the new version, but that should affect more than solo strings.

This is very weird.

Thanks again for your input.

krummholz · July 17, 2024, 04:55:24 PM

To update this thread: Alexander finally replied today. He does hear less bass in the "after" clip compared with the "before". His only suggestion was that there might have been some change in the Sibelius mixer settings. But as I've said, that's impossible: the mixer settings are specific to (and stored with) a particular score (.sib file), and here the exact same score (.sib file) is producing a different result since I trashed my original Sibelius settings file, and on the same day updated my Sibelius to the latest release.

It has to be something global.

I suspect something changed in the latest Sibelius release, but getting any reliable information out of Avid is like pulling teeth, and especially since their paid tech support (apparently, since I paid them $30 for help with this) consists of chatbots.

It looks like my days as a Sibelius subscriber are numbered. I'm going to wait a couple of days to see if anything else shakes out before finishing the cross-grade to Dorico, but unless a solution miraculously appears, I think that's the way I'm going to go.