Hi Vincent, 

Hm, I wonder what would makes the difference to other instruments... Not sure about that, sorry to say. 

Best, 

Paul

If you are using different articulations, there is no phasing.  They are totally different recordings. Also, if you do the transposition/pitch shift, it is not phasing either because they are completely different sample recordings being put together.   You are probably hearing the natural chorusing of similar timbres which is normal, not actual phasing which is the same exact waveform doubled.  You can sometimes hear a similar thing happen with two different brass instruments.  Their timbres are so close that there is a chorusing effect, though it is not at all the same waveform.  This chorusing effect in fact lies behind the reason for a large ensemble of doubled instruments such as 14 clarinets in a symphonic band playing the same line. 

O.K., enough of this, back to waiting for the Dimension string download...  I've only got 50 hours to go! 

@William said:

If you are using different articulations, there is no phasing.  They are totally different recordings. 

I think there is a problem with this position. Two recordings of the same flute playing together do result in a level of artificiality that is most certainly audible to those with an ear for it. There is a timbral difference between overdubbing, which will share a timbral relationship, as opposed to two different instruments. That's not even counting the fact that by being forced to make adjustments to EQ, etc to acheive a Violin II sound- that you are losing a great deal of flexibility in how you mix that section. Personally, every time I play with a Mach Violin II section, it has never sounded even remoately effective. Although, the concept of different strings and mroe options may improve this a great deal. I still feel that it is just better to record the total amount of instruments, that a Violin II section should always be sampled. But regarldess of that preference, I am excited for Dimension Strings nonetheless.

-Sean

 I think I need to point a couple of things out:

1. In real life you do hear phasing between players. This is really common in a trombone section, for example.
2. One of the reasons a string section sounds the way that it does is that the players are constantly phasing with each other. This is natural and part of the sound.

Having said that it should now be clear why you hear unacceptable phasing, even with a transposed patch. The only way to avoid this is by timing, different controller values and most importantly, by tuning. Even then, you have to understand that 12 x 6 players does not equal 12 players and never will, until some software genius manages to analyze how to account for the sonic differences.

DG

 Yes DG is correct in there being real-life phasing.  That is what happens with a big ensemble of brass, winds, etc.  and what makes a string section or band clarinet section sound so big.   I was referring to "phasing" in this sense as what happens when an exact reproduction of a soundwave is electronically duplicated on top of another, resulting in weird harmonics, bad chorusing, etc.  In a synthesizer done deliberately for a special effect, with samples done accidentally.  But phasing  does not happen in this bad or unuseable way when you use the transpose pitch-shift technique because if you look at the waveforms on an oscilloscope they are totally different even though they sound almost the same.  You might hear some more CHORUSING than you want, but that is NOT the same thing as phasing in this electronic sense. 

 If humanize is used in Dimension Strings to randomize delay and pitch in one instance of 8 players, and then a different set of humanize parameters used in another instance of 8 players to create a fuller 16 section first violin section, and then another 8 players humanized for the second violin section,  will this create enough difference in the players to avoid phase cancelling when they happen to play in unison?

Or more to the point - will this create a convincing 24 player violin section? 

That's assuming this method doesn't bring my computer to it's knees in terms of processing power !

The phasing that I hear with the chamber violins and the transposed one playing ine unison is really unacceptable.  Very present and dirty.  [:D]

Thank you DG for the information.  [;)]

@andyjh said:

Or more to the point - will this create a convincing 24 player violin section?

This is the real question before spending a grand on a new library. If you have to combine dimensions strings with other libraries to create larger sections it seems a little pointless to me.

While the amount of sonic detail is incredible I doubt this will still stand out when the libraries are combined.

@DG said:

In real life you do hear phasing between players. This is really common in a trombone section, for example.

That is true, but doesn't negate the need for a Violin II section. Consider this: two trombonists on two different trombones produce two very different timbres. A trombone may be made from Nickel Silver or Rose Brass, etc. These two instruments have unique sonic qualities and two trombonists have unique playing characteristcs even down to the most subtle of differences in attacks, releases, sustain consistency, etc. There are differences and no amount of computer programming, eq'ing, etc. will ever be able to account for this. You may consider it as effective to overdub, but others don't. This alone is enough of an argument for a Violin II section. But the real problem with the phasing comparison amongst trombones is that cyclic phasing between two different instruments sounds very different between the phasing that takes place in overdubbing. The sonic differences between the two different trombones means that you are ONLY getting the cyclic phasing. When you overdub you not only get the cyclic phasing, but the artificiality of having the same instrument play both parts. If I were wrong, then the same violin could be used to record the entire ensemble and it should be JUST as effective. The problem is that it isn't as effective.

Not to mention the fact that VSL users have been requesting a Violin II section and while that remains unanswered, other sample libraries are offering it. Don't get me wrong, I still consider VSL libraries to be of the highest quality and I'm setting aside my money to buy Dimension Strings. I'm all for it. I just think that a Violin II section is more than justified and that not sampling one is less logical.

-Sean

P.S. Please keep in mind, this is all my friendly opinion. [;)]

@andyjh said:

Or more to the point - will this create a convincing 24 player violin section? 

Andyjh,

Is it convincing? That's up to you. For me it isn't. Humanizations such as delay aren't enough though. Dimension strings has multiple options for this, such as the string used and the different expessive patches, etc. But the real question of 'convincingness' is whether you could take those 8 violins and build a rendition of 32 players playing different patches, different humanizations, all in unison, and have as convincing of a result as you would have had if you had recorded as many players together. In all my experience, there is always a noticable difference. I don't have dimension strings yet so I will have to play with it to really test it out.

The demo video does show some degree of mixing articulations, but not in creating a Violin II or a larger Violin section. I think it would be great if someone demoed this. [:D] Perhaps, if with Dimension Strings, it is convincing enough then it may settle some of the concerns people have about this.

-Sean