Bohemian Series Expansions Thread, Violin Expansion 2 Improv Video

OP
Virharmonic

Virharmonic

Senior Member
This is probably the feature I wanted more than anything else. My main issue with Bohemian Violin is it decides for me when the bow changes happen. With my way of playing, it is often sooner than I want a bow change. So this is amazing.

As a former violinist, the other thing I would like to be able to do would be to play a musical line without vibrato, followed by a sustained note with vibrato. Is this possible in Expansion 2?

Thank you so much for this upgrade! I can't wait to get it.
EXP2 does add the Lasting bow like we have in Cello (more advanced in EXP2 Violin), so you can extend the length of any legato type without a true rebow, but of course most true to real world would be forcing the right legato type when you need longer notes. Also the new rebow no longer uses mid arc for rebowing, so the rebows in general are really sweet now :) . In other words ... Lots of options based on your needs.

You can force on non vibrato that smartly merges to legato patches. As non vibrato isn't a main feature of the Violin, it isn't as deeply sampled as all the rest (adding over 10 real world non vibrato legatos in true speeds and dynamics would be insane :D ) but it does allow you to force on non vibrato and it comes in three true dynamics as well as rebowing in nonvib and of course you can then rebow to vibrato. Also faster bow changed legatos are almost non vibrato in the lower dynamics as that is how I would play it naturally :)

Lots of discoveries ahead for those who own EXP1 :) To be clear the Content is almost doubled with EXP2 (Exp 1 violin has over 30k samples and EXP 2 pushes it over 50k total sample count, but sample count isn't what drives me. I pay most attention to the fluidity and phrasing).
 

holywilly

Active Member
I wish the Violin EXP 2 won’t be too resource hungry like Cello EXP 2.

When having Cello EXP 2 loaded, I always see the beach ball rolling (I’m on Mac Pro 6,1) when saving my session.
 

midiman

Active Member
Thank you everybody for the kind words and questions :)

In the video I show how much more is possible in EXP2 without using words or forcibly drawing attention to it in a technical manner, but rather playing what came to my mind to showcase how easily can the performer play it in a musically accurate and fluid way. I'll leave the technicalities to the Overview and tutorials ;).

I know that a lot of you love EXP1 already and improving on something that is reaching performance levels is hard to do, so we paid attention to the musical flow to reach higher levels of True Performance that stays in correct motion as close to what real violinist would or could do.

Every time you see Bow Change or Rebow it is an EXP2 sample that fits the flow of the phrase (Chapbot referred to it as more real sounding). IE a sample captured in motion including the interval and the new note without extra crossfade for each bow change in dynamics and speeds. You can see and hear the difference in playing the phrase with bow change or onbow as I do use both in the improv. When onbow is played the notes are connected on the same bow stroke and these legatos sound fundamentally different to Bow Changed (you already have onbow samples in EXP1), but when you see bow change the change happens at the same time the bow changes direction and all of these samples are brand new in Expansion 2.

Especially the Part at 1 minute onwards you can hear very nicely how I switch between the two different types of legatos to enhance the flow exactly how I would do it if playing my violin and decide nicely which notes to accent, as bow changed legato will be always sharper then onbow. In phrasing one could see is as TEAY or TATETY (Each T is a bow change) but including all the magic that happens when in motion from note to note.

Bow changing is no longer something to be careful about, but something to be used as it adds so much colour to the performance and to me as a violinist it is an absolute must to have. And in Violin EXP2, you decide which one by simply overlapping (onbow) or not overlapping (bow change), so you can do it on the fly and get the musical flow you desire without breaking the flow.

In the non VI world, every note you see in manuscripts that doesn't have connecting legato line above would be played using what we call Bow Changed legato unless there is a notated pause and if you look at any manuscripts you will see that there is often more bow changed notes then onbow, but VIs over the years relied more heavily on Onbow legatos due to many restrictions that made it compulsory. With Soul Capture we work towards being true to real world, so we had to make sure that we have the more often used bow changed legatos captured in the same detail as our onbow which of course applies to note repetition as well. It is this subtle yet crucial variety that makes this improv and Violin EXP2 sound so much more real.

Now it is back to prepping beta for me. Cheers

Ondrej
The play through of the EXP 2 sounds great. Did you fix the tuning/intonation issues that existed before on quite a lot of notes? I love this VI. It has great features, and sounds great, and fits really well into a mix. But I ran into so many out of tune notes that it was frustrating. Always had to find a way to pull from another legato sample by hitting key switches so that it would pull from an alternate legato note that was I tune. That was disturbing a lot my workflow when using this VI. That was my only issue with this lib and I am really hoping it is fixed on EXP 2. Did you do any work in terms of tuning some notes?
I remember there being an argument made of keeping the instrument with imperfections, but in my opinion, bad intonation is not something that really makes it sound more real. Just sounds not in tune. And it is frustrating when you are writing a beautiful tune, and the VI can't can't pull it of without having one or two out of tune notes in a phrase.
That aside, this is indeed one of the best solo violin libraries ever made.

I look forward to trying out the new EXP 2!
 
Last edited:

markleake

Recovering sale addict
Did you fix the tuning/intonation issues
This stopped me using it also. I ran into issues where with a lot of the legato note changes, the player took a half second to get in tune. It was an issue introduced in EXP1 I think. It was OK by itself, even desirable in a way, but when other instruments were in the mix, I just couldn't get it to play well enough in tune.

I'm hoping this expansion fixes this. The bow change legato do sound great here. It's going back to earlier behaviours in a way, where in the first version it didn't use slurs at all - something I think actually worked better. So yeah, less pitchy slurs is a thumbs up from me. :thumbsup:
 
OP
Virharmonic

Virharmonic

Senior Member
The play through of the EXP 2 sounds great. Did you fix the tuning/intonation issues that existed before on quite a lot of notes? I love this VI. It has great features, and sounds great, and fits really well into a mix. But I ran into so many out of tune notes that it was frustrating. Always had to find a way to pull from another legato sample by hitting key switches so that it would pull from an alternate legato note that was I tune. That was disturbing a lot my workflow when using this VI. That was my only issue with this lib and I am really hoping it is fixed on EXP 2. Did you do any work in terms of tuning some notes?
I remember there being an argument made of keeping the instrument with imperfections, but in my opinion, bad intonation is not something that really makes it sound more real. Just sounds not in tune. And it is frustrating when you are writing a beautiful tune, and the VI can't can't pull it of without having one or two out of tune notes in a phrase.
That aside, this is indeed one of the best solo violin libraries ever made.

I look forward to trying out the new EXP 2!
Hi Midiman and MarkLeake,

I remember us having this discussion after Expansion 1 with Midiman and I stand by what we said back in the day. Also until today nobody else has emailed us regarding intonation, so if we believe that it is correct to include these samples and we don't get it reported by a percentage of the users, we do not change it. We believe in choice rather then 100% pristine intonation perfect samples. We don't micro tune parts of our samples, they are parts of real performance as captured. The main reason is that Intonation perfection is highly unnatural. When I hear a pitch perfect performance where each note lands exactly right within x ms I know it is virtual instantly. I know that some people prefer pitch perfect at all times, but we are strong believers in variety and with the vast choice of our legatos as well as bow directions there is always a pitch perfect variant available as we have excessive true legato landing note options compared to what is usual and all of these are of course chromatically sampled, but more importantly they are true to the performance that they were part of and not really touched up by us to be better then the real performance which is important part of our concept.

Every Concept has it's tax and avoiding paying it will break the concept!

In EXP2 we have replaced a few samples from EXP1 that we deemed in need of replacement, but it wasn't many, but much more importantly EXP2 brings over 20k more variants to use which I feel is a much better solution then micro tuning samples, which is more commonly used approach in VIs as they usually have a lot less sample variety for legatos. It will always contain some samples with odd landings as well as all other things that happen when playing in real phrase and instead we allow the user to force something different on to get closer to their desired performance.

Special note on Iterative development process.

Main issue with iterative development is that you can never satisfy everybody, especially if you replace/remove parts. This would be double true about something we haven't received complaints about, hence we rather add variations then take stuff out.
Main comments about EXP1 that drove EXP2 development were in order of requests -
1. Loading time is desired to be better ( insanely hard to achieve when adding 20k samples, but improved in EXP2 using our scripted background loading for as much as possible as well as improvements that UVI made for us that enabled us to map and call on samples differently which we are very grateful for),
2. Desire for longer bows after legato (resolved using Lasting Bow now added to Violin EPX2 and big variety of really nice rebows which are longer allowing for smoother playing).
3. More Plonkability ( This is further enhanced beyond original version as in combination with all the legatos available now as well as lasting bow it is by far the most plonkable library I have ever used)
and those were the main requests.
The rest we have worked out with our friend composers as well as what we ourselves saw as in need of evolution.

We have also added 2 new moods that will be unveiled during Beta and one of them puts very strong preference on Fingered over slurred legato, so that in effect leads to more pitch perfect performance.

I hope the above explains in depth why it is so hard to make changes that could potentially break the instrument even though it might not appear as dangerous to the users.

I wish the Violin EXP 2 won’t be too resource hungry like Cello EXP 2.

When having Cello EXP 2 loaded, I always see the beach ball rolling (I’m on Mac Pro 6,1) when saving my session.
Performance overall is improved in EXP2 when it comes to loading, saving ect..

Passionate Regards :)

Ondrej
 
Last edited:

markleake

Recovering sale addict
Hi Midiman and MarkLeake,

I remember us having this discussion after Expansion 1 with Midiman and I stand by what we said back in the day. Also until today nobody else has emailed us regarding intonation, so if we believe that it is correct to include these samples and we don't get it reported by a percentage of the users, we do not change it. We believe in choice rather then 100% pristine intonation perfect samples. We don't micro tune parts of our samples, they are parts of real performance as captured. The main reason is that Intonation perfection is highly unnatural. When I hear a pitch perfect performance where each note lands exactly right within x ms I know it is virtual instantly. I know that some people prefer pitch perfect at all times, but we are strong believers in variety and with the vast choice of our legatos as well as bow directions there is always a pitch perfect variant available as we have excessive true legato landing note options compared to what is usual and all of these are of course chromatically sampled, but more importantly they are true to the performance that they were part of and not really touched up by us to be better then the real performance which is important part of our concept.

Every Concept has it's tax and avoiding paying it will break the concept!

In EXP2 we have replaced a few samples from EXP1 that we deemed in need of replacement, but it wasn't many, but much more importantly EXP2 brings over 20k more variants to use which I feel is a much better solution then micro tuning samples, which is more commonly used approach in VIs as they usually have a lot less sample variety for legatos. It will always contain some samples with odd landings as well as all other things that happen when playing in real phrase and instead we allow the user to force something different on to get closer to their desired performance.

Special note on Iterative development process.

Main issue with iterative development is that you can never satisfy everybody, especially if you replace/remove parts. This would be double true about something we haven't received complaints about, hence we rather add variations then take stuff out.
Main comments about EXP1 that drove EXP2 development were in order of requests -
1. Loading time is desired to be better ( insanely hard to achieve when adding 20k samples, but improved in EXP2 using our scripted background loading for as much as possible as well as improvements that UVI made for us that enabled us to map and call on samples differently which we are very grateful for),
2. Desire for longer bows after legato (resolved using Lasting Bow now added to Violin EPX2 and big variety of really nice rebows which are longer allowing for smoother playing).
3. More Plonkability ( This is further enhanced beyond original version as in combination with all the legatos available now as well as lasting bow it is by far the most plonkable library I have ever used)
and those were the main requests.
The rest we have worked out with our friend composers as well as what we ourselves saw as in need of evolution.

We have also added 2 new moods that will be unveiled during Beta and one of them puts very strong preference on Fingered over slurred legato, so that in effect leads to more pitch perfect performance.

I hope the above explains in depth why it is so hard to make changes that could potentially break the instrument even though it might not appear as dangerous to the users.



Performance overall is improved in EXP2 when it comes to loading, saving ect..

Passionate Regards :)

Ondrej
A wonderful explanation. Thankyou, and thankyou for your passion on this project. I've always been a believer. :thumbsup:

Some of those slurs did stop me from using it sometimes, yes. But I probably exagerate the issue, given I've tended to push it beyond its purpose (orchestral work better suited to something less soloistic, for which I now have a good alternative anyway). Not something I would report to the dev, as any VI in those conditions is going to have a few quirks. It's good to know there are many easy options around the slurs; I didn't realise that.

All 3 of those priorities I think are exactly spot on. And a mood for using bow legato more?... just perfect. :dancedance:
 
OP
Virharmonic

Virharmonic

Senior Member
I wonder if any violinists might comment on how they might vary intonation when in an ensemble vs as a soloist?
Well I can answer that really easily. There is no difference in approach to intonation when playing in an ensemble vs Soloists. We always as all violinists aim for maximum accuracy. The difference between playing in ensemble vs Soloist is in Expression, in how you approach the swells and vibrato, how intensive you make said expression and in the type of transition you might be happy to use. IE in Ensemble you will prefer to use Fingered legato where possible to increase the accuracy and you would naturally avoid slurs, while as a soloist you might be more liberal with these.

This also means that when capturing more then 1 violinist you should always and I mean always capture at least 3 if you want them to play in unison, otherwise they will always sound out of tune and phase to each other. Violin is a microtuning instrument and thus it will never be perfectly in Pitch per say especially when playing in unison. There are many other rules for recording violinists as well as being violinist, such as avoiding over shadowing each other in ensemble and instead learning to listen to each other and form a unibody when playing in ensemble vs when in Solo where you are aiming to stand on top of the ensemble, thus you must be more liberal with your playing and more "over the top". Sometimes this might mean more vibrato, sometimes more pronounced arcs, sometimes simply higher dynamic to the ensemble .... The list and approaches are long :)
 

muk

Senior Member
I wonder if any violinists might comment on how they might vary intonation when in an ensemble vs as a soloist?
As Ondrej wrote the intonation does not depend much on wether it's an ensemble or a solo piece. However, in tonal music intonation depends on the harmonic context. A major third, for example, will be intonated slightly lower than in equal temperament tuning. Conversely, a minor third will be intonated slightly higher. As every note can be a third in a chord it's not possible to take this into account during sampling. Cubase has a feature called 'Hermode tuning' that tries to imitate the natural adjustments of tuning a real ensemble/orchestra would use.
 

ism

Senior Member
As Ondrej wrote the intonation does not depend much on wether it's an ensemble or a solo piece. However, in tonal music intonation depends on the harmonic context. A major third, for example, will be intonated slightly lower than in equal temperament tuning. Conversely, a minor third will be intonated slightly higher. As every note can be a third in a chord it's not possible to take this into account during sampling. Cubase has a feature called 'Hermode tuning' that tries to imitate the natural adjustments of tuning a real ensemble/orchestra would use.
Thanks! The more I learn about real violins, the more sample libaries I need to buy. :)
 

ism

Senior Member
Well I can answer that really easily. There is no difference in approach to intonation when playing in an ensemble vs Soloists. We always as all violinists aim for maximum accuracy. The difference between playing in ensemble vs Soloist is in Expression, in how you approach the swells and vibrato, how intensive you make said expression and in the type of transition you might be happy to use. IE in Ensemble you will prefer to use Fingered legato where possible to increase the accuracy and you would naturally avoid slurs, while as a soloist you might be more liberal with these.

This also means that when capturing more then 1 violinist you should always and I mean always capture at least 3 if you want them to play in unison, otherwise they will always sound out of tune and phase to each other. Violin is a microtuning instrument and thus it will never be perfectly in Pitch per say especially when playing in unison. There are many other rules for recording violinists as well as being violinist, such as avoiding over shadowing each other in ensemble and instead learning to listen to each other and form a unibody when playing in ensemble vs when in Solo where you are aiming to stand on top of the ensemble, thus you must be more liberal with your playing and more "over the top". Sometimes this might mean more vibrato, sometimes more pronounced arcs, sometimes simply higher dynamic to the ensemble .... The list and approaches are long :)
Thanks, very helpful!

Follow up question though - and this is really the use case that I struggle with in all solo string sample libraries - what about mixing a violin soloist with an ensemble?

Is this the same case as you describe above? Does the soloist need to carefully match the extent of intonation to the ensemble? Can a soloist go hyper expressive to pierce through the ensemble? Can it depend on the harmonic or contrapuntal relationships?

My own experience is certainly that's got the best results with the Bohemian Cello by writing the cello line first, and then matching the dynamics and vibrato of the accompanying ensemble to the expression of the cello. Though I feel by taking more control via keyswitchs I might be able to take this farther (which now that that keyswitching bug if fixed I'm looking forward to experimenting with).

I guess maybe the pre-release isn't the time for this level of detail (so feel free to defer any answer) ... but these sorts of issues are very much the next things I need to get my head around in order take what I'm able to do with the Bohemian instruments to the next level.
 
Last edited:
OP
Virharmonic

Virharmonic

Senior Member
Thanks, very helpful!

Follow up question though - and this is really the use case that I struggle with in all solo string sample libraries - what about mixing a violin soloist with an ensemble?

Is this the same case as you describe above? Does the soloist need to carefully match the extent of intonation to the ensemble? Can a soloist go hyper expressive to pierce through the ensemble? Can it depend on the harmonic or contrapuntal relationships?

My own experience is certainly that's got the best results with the Bohemian Cello by writing the cello line first, and then matching the dynamics and vibrato of the accompanying ensemble to the expression of the cello. Though I feel by taking more control via keyswitchs I might be able to take this farther (which now that that keyswitching bug if fixed I'm looking forward to experimenting with).

I guess maybe the pre-release isn't the time for this level of detail (so feel free to defer any answer) ... but these sorts of issues are very much the next things I need to get my head around in take what I'm able to do with the Bohemian instruments to the next level.
I'll use a general rule of a thumb. When a soloists works with an orchestra they commonly have rehearsals to make sure that the performance clicks. The main reason is because the Soloist has to work with the Conductor and orchestra to agree on the approach to the performance. It isn't that the violinist dictates to a conductor , but a good conductor will often take points from the violinist or other soloist and follow their lead to support them with the orchestra and agree on any areas of conflict up front on how to best approach them. General rule of any solo performance even in small bodies as bands, soloists leads and Orchestra, band follows and supports when the soloists is playing their main lead part. With solo libraries it will be always easier to match an Orchestra to the will of the soloist then the other way around, so it is always easier to arrange a piece if you leave the Soloists to lead, but if you are happy to use force keys and CC11 for a bit of extra expression you can also bend the performer to the will of the orchestra, even if it is a slightly backwards way. The only thing the Solo violinist should bend to is the Melody and Phrasing which is what we keep working on ;)
 

ism

Senior Member
I'll use a general rule of a thumb. When a soloists works with an orchestra they commonly have rehearsals to make sure that the performance clicks. The main reason is because the Soloist has to work with the Conductor and orchestra to agree on the approach to the performance. It isn't that the violinist dictates to a conductor , but a good conductor will often take points from the violinist or other soloist and follow their lead to support them with the orchestra and agree on any areas of conflict up front on how to best approach them. General rule of any solo performance even in small bodies as bands. Soloists leads and Orchestra, band follows and supports when the soloists is playing their main lead part. With solo libraries it will be always easier to match an Orchestra to the will of the soloist then the other way around, so it is always easier to arrange a piece if you leave the Soloists to lead, but if you are happy to use force keys and CC11 for a bit of extra expression you can also bend the performer to the will of the orchestra, even if it is a slightly backwards way. The only thing the Solo violinist should bend to is the Melody and Phrasing which is what we keep working on ;)

I've been butting a lot of work into playing with solo string libraries lately. And with some of the best one (which certainly includes the Bohemian) I occasionally get these moments where I hit such transcendently convincing sweet spot of the library that at the same time it both justice all the effort I put into them and inspires me to forget about samples altogether and just lear to play the cello or (depending on the day) violin.

But before the current generation (Bohemian, Joshua bell, spitfire etc) I don't think there was much need to think about the nuances of real performances at this level of detail. (Or, more accurately, I don't think there was any point, as there was nothing previous generations of solo string libs could do about it ).

Again, thanks for that insight. If you ever get a chance to make a video fleshing out such techniques of blending it would be helpful.

In particular, I think the two directions you describe require very different techniques in crafting a performance, and where this didn't matter with the previous generation of instruments, with sampling at this level of nuance I think it becomes crucial to understand the difference.
 

ism

Senior Member
Hmm, actually, I think I can refine my theorization of "plonkability" in this light. Let's call it a 'directionally colour' of 'plonkability'.

Lets call the first colour of directionality 'soloist directed' plonkability. Capturing the expressiveness of a soloist first, with the intent of crafting the performance of the ensemble to support the soloist. Bohemian is clearly state of the art in this kind of plonkability.


The second colour of directionality, lets call it 'elbow directed' plonkabiliy, referencing a player attending to the motion the other ensemble member(s)' elbow(s) in crafting an accompanying performance.

My sense is that with all the keyswitch control, it should be possible to develop technique for this direction of plonkability in the Bohemian instruments. My sense is also that the non vibrato of this update should significantly enhance the plonkability of the violin in this direction. As crafting the phrasing of the vibrato is the biggest challenge.

(This is a technical nuance, but arguably the plonkability in this direction here lies more in the ability to plonk down the ensemble first, and then lean on the inherent performability of the soloist to match the expressiveness to the ensemble. In this sense we have a kind of duality between plonkability and performability, and the entanglements this implies).


Note that in both 'directionally coloured' notions of plonkability, plonkability becomes a fundamentally relational concept, achieved only in conjunction the enselmbe, in the form of the result of the soloist and ensemble playing together in context.


My sense from these pages, however, is that people (including myself) generally approach the Bohemian instruments from the perspective of the first direction of plonkability. And a certain amount of frustration arises when a workflow calls for a solo line to be noodled in after the fact, but the expectation is that the Bohemian can deliver the first colour of plonkability here, when in fact, the context very much calls for a performance to be crafted based on the second colour of plonkabilty.


Again, in previous generations of sampled solo strings, this directionally didn't matter much, simply because there wasn't enough nuance captured in the samples to make much difference one way or the other. So its unsurprising that they now risk being conflated.


But this directional asymmetry (soloist first vs ensemble first) in crafting and blending of performances suddenly matters quite a lots in the current generation of libraries.


[end of theorizing, carry on].
 

LamaRose

Gato Mighty!
Sounds really good... but I have to wonder what the CPU hit will be with this update. And to be honest, in retrospect, I truly wish that a basic viola version had been released, as with the violin and cello, just to have had something to work with and to have had more than one version to call upon.