I just thought I would add a little activity to this site by posting my latest rendering of an older work, my first (and so far only) string quartet, which I began as a student at Michigan in 1975 and finished, except for minor revisions, in 2020. It's a dissonant, rather Bergian work that drifts in and out of a kind of free atonality, though to my ears there are passing references to key centers everywhere, especially in the Tippett-like 3rd subject and in the newer material (roughly the second half) that was written with the benefit of notation software. It begins and ends in A minor, but very little of the piece is actually in that key, or even in any key. Toward the end the "Tippett" theme is, however, transformed into a lilting, and very diatonic, duet for the violins against a pizzicato accompaniment.
There are no changes to the actual notes since I last posted this work here, but numerous changes to phrasing, articulation, dynamics, and the lengths of fermatas and other pauses as I ported the score from Sibelius to Dorico earlier this month. This is basically the same rendering that I posted recently on a different site (GMG), except for shortening the caesura at letter K from 275% to 200%, which clips off a little less than a second from the rendering.
The score has not been proofread thoroughly; there may still be collisions and extraneous rests left over from Sibelius - the score was exported first to XML and then read by Dorico, which apparently doesn't do any cleanup during the import process - if my impression is correct, you have to edit a measure directly before it will consolidate rests there.
Comments are welcome.
Edit: about two weeks ago I bought a copy of the VSL (Vienna Symphonic Library) Studio Solo Strings library and just finished "porting" this quartet to render in Dorico via that library. I think the sonics are better, and some parts are musically superior as well. I've included links to the new rendering and score below - there are very few changes in the score other than as needed to correct dynamic imbalances and some invisible ones, like the lengths of some of the pauses, but there is a biggie: from letter G to the first beat of m. 106, I had to swap the 2nd violin and viola parts because the VSL library doesn't have legato samples for the F6-E6-Eb6 triplet figure in m. 105. There are some glaring flaws, most notably a "whooping" effect in places where two notes connected by a slur are literally slurred together, but I think the superior sonics and the fact that the four instruments have very different, distinctive timbres in the VSL library more than compensate for those flaws. A comment below mentioned that in places the counterpoint disappears because of parallel motion - but that's allowed in counterpoint, and I think the real problem is that the two instruments have nearly identical timbres in NotePerformer, which was used for the first rendering.
The main changes to the *music* in this version have to do with the lengths of pauses, which I tried to tweak to reduce the "sectional" quality that the work can have if they're too long, as someone rightly pointed out below. Hopefully this version will be a little more convincing - if you can overlook the flaws in the rendering.
String Quartet No. 1 in A Minor, Op. 1 (VSL Studio Solo Strings)
The files below are the original ones that I posted here before adapting the work to VSL.
String Quartet No. 1 in A Minor, Op. 1 (NotePerformer)
Replies
Well, that definitely sounds better than NotePerformer's solo string sound! I'm impressed, and maybe this is one of the improvements that Arne & crew have made since NP v3, which was the first version that I had and certainly the one that I used, several years ago., to experiment with single player sections strings vs. solo strings My understanding at the time was that there *couldn't* be any difference between the sounds, since (I think Arne said this, but don't quote me) the NP section string sound was built up from their synthesized solo strings. But maybe I misunderstood, or maybe that has changed. What version of NP did you use for this recording?
I'm definitely not as turned off as David by anything other than NP's solo strings, and will now have to revisit that experiment. BTW I only got through the 8th movement, or maybe it was the 7th, of your 5th Quartet before YouTube cut in with an ad. I normally don't listen to anything on YouTube for that reason - the ads are always much louder than the actual content and are annoying beyond description... even when it lets you skip them after a few seconds, which isn't always. I hasten to add that I really enjoyed what I heard of your piece. The idea of building a string quartet as a series of short character pieces reminds me a little of Holmboe, though yours are much shorter than anything in Holmboe's later quartets. I don't hear a lot of Webern in your piece otherwise, but the sound world is well imagined and realised. I have been thinking about writing a similarly constructed work, a series of brief "sketches" for various instrumental combinations in a chamber context. Whether this idea ever really gets off the ground is another question - recently I tend to reach a point where I can't hear what I want to write clearly enough, get frustrated, and move on to something else. But we'll see, now that I'm done with this VSL rendering (for now anyway).
I think also that NP does better with glissandi in Sibelius. The BBCSO solo strings do better in Dorico as well. I had to fake some of it, though. In SQ3, the performance version of the score uses 8 lines because you can't tie into a gliss. The glissandi all appear on separate staves from the rest of the music, where the tie goes the starting note and reattacks the finishing note softly. I noticed with the section strings that in Dorico the glissandi don't seem to work at all. Sibelius handles them differently, so they work. Although in Far From the Fading Light, there is a long glissando that sounds a little ridiculous, better with the stock NP strings sounds than with the BBCSO section strings. It's a smooth gliss, but it stops sounding like violins after about a 5th.
I used NP 4.5.1 for the rendering of String Quartet No 5 in the YouTube version. There is an NP4.5.1 version of SQ2 there alongside a NPPE version.
I use the DuckDuckGo browser which has its own Duck player that blocks YouTube ads.
The NPPE with BBCSO Pro does give you different timbres between Violin I and Violin II. Different players recorded the samples. I'm very disappointed that NP has dropped NPPE. I spent a lot of money upgrading my computer and purchasing samples to use it. I wrote Arne to give him a piece of my mind yesterday. I haven't received a response yet.
Which version of Dorico do you have? True gliss in Dorico was only implemented somewhat recently - I know it was there in Dorico 5 because that's the first version I had and I waited to buy until they had added that feature, and in the NP version that viola glissando in my quartet sounds the same in Dorico as in Sibelius.. But as David wrote below, VSL apparently doesn't support the range of the gliss that I call for there.
It sounds as if NP's section strings model was improved by NP 4, then, well after I had concluded that single player section made no difference. One of my "sketches" is for violin and piano - I'll have to replace the solo violin and compare using NP 5.0.1, as that would be easy to do since I've only written about 10 bars so far.
I know a lot of people swear by Duck Duck Go for privacy reasons, but it sounds as if their browser does things that would break most sites where if your browser blocks ads, the site blocks you from accessing the content. I'm surprised that YouTube doesn't do that, since they're very aggressive with their ads.
Of course NPPE with BBCSO (or any of the supported libraries) can give you different timbres for vln1 and vln2 because they're sample libraries. The sounds were generated by actual instruments. I was talking about straight NP synth.
I hadn't tried using section strings in NP3, so I don't know. Certainly, NP4 was adequate, if not spectacular, for the section string solos. As far as glissandos go, it depends on the sample library. BBCSO Core orchestral strings don't allow them in Dorico, while in Pro, the solo strings do. Here is a link to my SQ3, in case you didn't find it in the Music section. It has loads of glissandos, even ones to and from a quartertone. I really had to finesse to get that, but I finally succeeded. Sibelius, on the other hand forced the glissandos, regardless of whether they were allowed in the sample library. Hence the demos of Far From the Fading Light and The Seventh Trumpeter of the Apocalypse were both realized in Sibelus, although the definitive score for the latter is in Dorico. I tend to compose into Sibelius and then port it into Dorico because I like its condensing feature (with some reservations). I find it much easier to manipulate material in Sibelius.
BTW, you don't need to use a CC code to change the sound of solo strings globally in Sibelius. All you need to do is redefine each instrument to have the section (one player) sound. It's easy and takes only a few seconds. It's a little more complicated to redefine the word "Solo" within the context of the music, though.
My problem with Dorico is that I use whatever software my publishers want, and nobody is asking me for Dorico, so if I'm doing a lot of work, then I'm away from Dorico for long stretches of time. It's so complicated to use that I have to re-learn it each time. Right now, I'm in a big job using Score, so I haven't touched Dorico, since probably early March. I've done a lot of Finale, as well as some Sibelius in the meantime, but Dorico is just the opposite philosophy from them.
FWIW, I just heard back from Arne, and as I expected, he has no plans to resurrect NPPE. It looks like I'll be frozen in time with NP4.5.1, just like I am with Score 4.01 (for DOS, ca 1995) since the Windows version of Score is so unstable that it is nearly unusable.
You wrote: "BTW, you don't need to use a CC code to change the sound of solo strings globally in Sibelius. All you need to do is redefine each instrument to have the section (one player) sound. It's easy and takes only a few seconds. It's a little more complicated to redefine the word "Solo" within the context of the music, though."
I'm not sure what you mean by defining each instrument to have the one player sound. The only way I know of to do this is to choose a string section and then tell NP to only use the single player sound. In Sibelius NP supplies a pull-down menu that has a selection for specifying the number of players, but it's done with CC codes under the hood, so for Sinfonia Solenne I just used the menu to create one CC code, and then copied, pasted, and edited to get all the others. The menu doesn't seem to be available in Dorico, so I just created a CC editor (one for that, and another for adjusting vibrato) for my Dorico port of Sinfonia Solenne.
Unless, of course, we're talking about different things...
And yes, that's my understanding re: NPPE, that it's dead. I never took the plunge as I was waiting for them to support VSL or something else that was both high-end and within my hardware capabilities. The latter is a problem even with VSL solo strings: I've noticed that my audio occasionally cuts out during playback. At first I thought my computer's sound chip was beginning to fail, but it ONLY happens on playback within Dorico, never when listening to an existing audio file - and it can usually be fixed (temporarily) by restarting Dorico - so it's pretty clear that the resources needed to host VSL are putting a strain on my CPU. Maybe it's an available memory issue and there's page swapping going on - I would have thought that 32 GB was enough, but Apple's memory management seems to have problems in other areas as well. Anyway that I missed NPPE is probably for the best, given Arne's eventual decision to kill it.
What I meant re: glissandos is that prior to Dorico 5, true gliss wasn't supported at all, even using straight NP (not NPPE). That delayed my move to Dorico by over a year, until David informed me that true gliss was now available. From what you say there are apparently still limitations with some libraries that actually do support them - and from what David said, apparently there are few if any sample libraries that could properly render the one in my SQ.
You don't have to set up a CC code. In Sibelus, go to Instruments->Edit Instruments. Choose your string instrument, violin (solo) or in a string quartet, just violin I. Click Edit instrument, best sound ->Choose->NotePerformer->Strings->Violin->Violin (section member), OK, save, etc. Do the same for all members of the quartet. Then you will have as the default for that score, the section member solo strings, which are much better than the Solo strings. That is all you have to do.
I haven't tried it in Dorico.
Sorry Stephen, but the steps you say you use don't correspond to anything that's available in my version of Sibelius (Ultimate). There's no Edit Instrument button or menu item. There's a Change button, for changing an instrument to a different one, but you don't choose NotePerformer through the Change dialog. That dialog shows many possible string instruments including Violin I and Violin (solo), but not Violin (section member). The first is a violin section, with a default number of players that isn't documented (to my knowledge anyway), and that NP's literature directs you to change from the Plug-ins menu on the Home tab, which inserts CC codes. The second is, of course, a solo violin, which if you're using NP for playback, will be their synth solo violin. To play back through NP, you choose it from the Audio Setup dialog, but there's no access from there to the Change dialog or anything that functions the way you describe - as far as I can tell.
In the Instruments Section of the Ribbon, you see a little icon (a box with a little arrow that points down and to the right) at the bottom right. That takes you to Edit Instruments dialog. I forgot you were on a Mac, so it might be a little different. Just type Edit Instruments in the search box (upper right in Windows). That will take you right there with the first option that shows.
Thanks Stephen, I see what you mean now. I had used that dialog before but never followed through the Choose... menu to see if it gave you any new instrument choices. That's very interesting! If that actually gives you a realistic single section player, I wonder why Arne's documentation for adjusting the number of players in a section doesn't mention this? He only describes the procedure I outlined above. I shall have to experiment with that and see if I can discern a difference in the string sound.
So I tried this in my quartet, and it definitely gives a different timbre to the string sound - different both from straight NP synth solo instruments, and the string section with CC 104,1 to reduce the section size, which interestingly is NOT any longer identical to straight synth solo either. But in my estimation it doesn't sound remotely like a real soloist, and all three of these routes lead to output that sounds synthy, electronic. I much prefer VSL solo strings, though I wish there was a way to get rid of the "whooping" effect at transitions between notes. They're not limited to legato passages either - the 1st violin's second solo passage as the main climax unwinds in my quartet has three or four of them, at least one between notes not connected by a slur, real conspicuous bloopers.
But most likely, pursuit of perfection is inevitably doomed to failure at some point...
-
1
-
2
-
3
of 3 Next