Gerard

Thanks very much for the clarification. I read the other thread you started on this subject. If I understand correctly, this is only a problem if you are running Express on a slave computer and want to use more than one physical midi in port. You could still (theoretically of course) run as many as 64 instances of VI in Logic Express if you were using it as the master DAW. Right?

How come only 16 tracks?

I just checked the Logic Express specifications:

http://www.apple.com/logicexpress/specs.html"="" target="_blank" title="http://www.apple.com/logicexpress/specs.html">http://www.apple.com/logicexpress/specs.html">http://www.apple.com/logicexpress/specs.html

and it says 64 software instrument tracks!

I am also considering Logic Express, so I would be very grateful for clarification on this issue.

Interesting texture. I like the fourth/fifth harmonies and orchestration, especially the use of varied string articulations and the percussion. A little bland and monotonous melodically. The tempo also strikes me as mechanical. This is perhaps intentional, but varying the tempo a bit more could help build the tension. More dynamic variation could help too.

Unfortunately I have had almost no time to compose recently. To help me get going again, I wondered if anyone would be so kind as to offer comments on the following two extracts of incomplete pieces. Samples are mostly VSL, with a few obvious exceptions. Both rather contemplative in mood, but each in their own completely different way.

1. Romantic, scored for solo violin, piano, alto flute, french oboe, and chamber strings (2 mins.)
http://www.buist.org/david/KSMSeq25.mp3

2. Pseudo-ethnic, scored for mixed orchestra and solo sarangi (1'30" - sorry about strange ending - work in progress!)
http://www.buist.org/david/SarangiConcerto3_4.mp3

An interesting piece. Thanks for letting us listen. The unexpected emergence of a massive tutti out of what seemed to be a calm almost desolate stillness is striking. Or at least it could have been. I don't think the effect was nearly as boldly done as it could have been. The tutti was not loud enough in relation to the intro, and it was too fleeting to leave any lasting impression. The dynamic contrast needs to be bolder IMO, and if it were me, I would have drawn out the tutti just a little longer before returning back into the calm. I can't say for sure, but I may just be imposing my expectations on your piece. Your intentions might have been completely different for all I know.

The strings? They need to be wetter and darker, don't you think? More reverb would help and perhaps you could darken the tone with EQ, or try using the con sordinos. The note transitions in the moving parts (before the tutti) could be more convincing. Are they supposed to be legato or what?

The ending is OK as far as it goes, but the piece as a whole needs to be longer IMO. Or is this only the beginning of something longer?

I would like, if I may, to put in a word for two more WW instruments: Oboe d'amore, Basset horn?These surely have at least equal claim to be included in the VSL orchestra as those listed above (which I am also glad to see included)

You will save money if you buy and register the Horizon solo strings first and then get the Pro ed. performance set later. Then you will get a discount for the samples you already have. It only works this way round. There is no discount if you buy the Pro ed. performance set before the Horizon solo strings (unless you also buy the Pro ed. string set).

I really wish that VSL had included fpf samples for the brass in the Pro Ed. They did for a few of the woodwinds.

In my experience it makes no difference whether you have the PRE button on or not. So far, I have been unable to obtain any significant panning effect with either the Position or Width controls when the signal is directed 100% wet into GP. Unless, that is, I use the workaround described in my first post on this thread.

I have not tried it yet, but I don't think using mono mode with independent L and R controls is going to make any difference either. This is simply another may of doing the same thing as the Pos and Width controls. In either case all you are doing is controling the amount of each channel of the sample being sent to each side of the stereo output pair. If you want full stereo, you send 100% of L to L, and 100% of R to R (that is the default setting in GS in either mode). If you want to convert a stereo sample source completely into mono, then you send equal portions of L to both L and R, and equal portions of R to L and R. This is all that is happening when you set the Width control to zero. If your sample source is mono to start with, then none of this is of any relevance at all, and you might just as well use MIDI panning (and forget about this thread entirely!).

I'm glad you found my workaround useful.

It seems to me that some systematic tests are required in order to figure out exactly what is going on here. In some cases audiable panning effects can be heard using the mixer controls in the same layer as the GP insert/aux send, and in some cases not. In my experience, it is generally not. I don't have time to do tests right now, but I would like to see if the following make any difference:

1. Which GP preset you use
2. Which sample library you use (different micing techniques in the original sample recordings might produce different results)

I have generally found panning effects to be very unpredictable when using Gigapulse. You can never be sure where the sound will seem to be placed just from the numerical value.

Ideally of course we would entrust all panning to GP, but that requires a separate instance for each position, which is not realistic for orchestral arrangements with the current level of CPU power.

The difference as I understand it (I may be wrong though) is that MIDI panning simply alters the relative volume of the two channels of the sample, so for example panning hard left effectively obliterates entirely the right channel of the sample. On the other hand, panning in the mixer allows you, for example, to retain the right channel of sample but send it wholy or partially to the left output. You can thus retain the full stereo image of the sample but simply move it in space. The mixer also allows you to narrow the width of the stereo image (or even to reverse the image if that is what you really want to do -- try pushing the knob beyond zero and it will go "negative"). You can't do that with basic MIDI panning.

I also noticed the phenomenon William describes. However, I have found a way round it: load GP onto an Aux bus(es); direct the inputs to Group bus(es); use the Aux sends in the Group buses to direct the signal to GP. Now you will probably find that using the position and width controls in the INPUT buses makes a lot of difference. However, the pan controls in the Group buses from which you send to GP still don't work. It seems you need to use all three layers of the GS mixer to get any pan effects with GP (assuming that you are not resorting to MIDI panning -- which indeed has its problems and is best avoided if possible).

I don't understand the logic of this at all, but just from experience, it appears that cannot pan and send to GP in Aux on the same mixer layer.

Thanks everyone for listening, and especially to dcoscina and Gugliel for your comments. Certainly the higher notes of the violin can sound a little excessively bright at times. Actually, I should mention that I EQed the violins -- the first and second differently in an attempt to make them sound like different instruments. Different EQ settings might help. And, as you say Gugliel, gigapulse adds its own colouring too. Some adjustments there might also help.

I cannot hear the change in the cello timbre you mention at 1:11. The only thing at that point is that the viola takes over the leading line (very briefly). Is it perhaps the viola you are hearing, not the cello? If so, is it a problem with the panning making the viola sound as if it is in the place of the cello? I used the same GP instance for both the cello and viola, but with different position settings in the mixer.

My next idea for experimentation with this piece is to play around with the tuning (a la Gugliel). I would imagine that this would contribute significantly to producing a more authentic solo string sound.

It started life as a keyboard improvisation, then I divided it into four parts and scored it for string quartet. Rendered in GS3 with GigaPulse (two instances). Comments of any kind are most welcome.

http://www.buist.org/david/StringQuartetteVSLGP.mp3
(playing time: about 1 min; file size: 2MB)

To return to the topic:

If we can have two flutes and two harps, then we can surely also have two solo violins. The logic is already there, so if VSL are consistent it should only be a matter of time.

Likewise, if we can have an additional larger horn ensemble, we can also have an additional and larger violin section. Like "Epic Horns" it would probably be a costly add-on product, but given the likely demand very worthwhile. "Epic Violins"?

In the meantime, let's strive to make full use of the great samples VSL already offer. As always the only reasonable "criticism" (if you can call it that) of VSL is that we still cannot get enough of it!

I had a listen to both demos. They generally sound nice, although perhaps a little too drawn out, especially the second one. If there could be a little less repetition and more development/variation of the themes it might be more satisfying perhaps.

I thought the piano in the first demo was too forward in the mix. Since it seems to be more of a supporting part than a solo, I think in this case it could be blended in more. Also I think I would have liked a less brightly toned piano for this context.

I was intrigued by some of the percussive effects at about one minute into the first demo, and also in the second demo just before the end. I could not make out exactly what they were.

Thanks for listening and commenting, Gugliel. My earlier version was criticized for not being "thick" enough. Are you suggesting that I have "overdone" it now? Or is your comment directed more at the melodic/harmonic/rhythmic content than at the orchestration?

Actually the whole piece is intended as an "apotheosis" of sorts. It is meant to be played at the end of the marriage ceremony to accompany the newly weds exit from the church.

I don't have the Pro performance edition (yet) but I do have the First edition with the free update. If I understand correctly, the free update increased the violin performance legato to the same range as the Pro edition. The range is G3 to D7. (I say this from memory. I do not have the thing in front of me right now, so I cannot be absolutely categorical) Actually, I am pretty sure that is correct, because I remember I wanted a high D#7 recently, and I had to use pitch shifting. I agree it would be nice to have an even higher range. As far as I know, 4 octaves (actually one semitone short of 4 octaves) is the absolute limit for performance legatos using the present tool (because of the way the key groups are arranged).

The other strings have shorter ranges. Cellos are only up to E5 for example (in the first edition at least). I wish they went higher!

The main intention of my post was clarify whether the discount calculator was actually working in the way intended by VSL. From the above reply I conclude that is. I was thinking that it might be a mistake, but apparently it is not.

It is definitely not my intention to accuse VSL of unfairness. I just wanted to clarify how the discount system works (and that it is really working how it should).

However, regardless of the issue of fairness, I will say that the system as it stands at present does not make it any easier for me to purchase VSL products. I will be a little more cautious about buying the products when I know that the order of my purchases could have a significant impact on the price I pay. Buying one product before another could mean that I have to pay more for the other product I buy later. In fact I may not know yet what products I will want to buy later, and I may be surprised to find that the price is more than it would have been if I had bought it before one of the products I already have.

I realize that pricing policy is a difficult issue in this industry. I also appeciate that VSL is making efforts to make the system work for customers as well as for themselves. Above all, I would like to emphasize that overall VSL offers really excellent value for money considering the outstanding quality of their product!

I think I have found what may be an anomaly in the VSL discount calculator.

When I check under "registered products" the following products:
Horizon edition solo strings
Horizon edition Vienna harps
...and check under "purchase" the following:
Pro edition performance set
I get the following result:
Pro Edition Performance Set Full Price 2,870 $
Discount on Pro Edition Performance Set for Solo Strings -90 $
Discount on Pro Edition Performance Set for Vienna Harps -45 $
Total: 2,735 $
So the total for all the products is like this:
Hor solo strings 495
Hor harps 295
Pro. ed Performance set 2735
Grand total: 3525

However, when I check under "registered products" the following:
Pro edition Performance set
...and check under "purchase" the following:
Horizon edition solo strings
Horizon edition Vienna harps
I get this result:
Horizon Series Solo Strings Full Price 495 $
Horizon Series Vienna Harps Full Price 295 $
Total: 790 $
So the total for all products is:
Pro. ed. performance set 2870
Hor solo strings 495
Hor harps 295
Grand Total: 3660

In other words, if I bought the Pro ed. performance set before getting the solo strings and harps I have to pay twice for the samples which I already have in the Pro ed. performance set. But, if I buy the solo strings and harps first and then buy the performance set later, I do get a discount for those samples. Surely the order in which you purchase the products should make no difference to the price you pay?

I should add that this is not a purely hypothetical case. I cannot buy all the products I want at one time, so I was thinking of getting the Pro. ed. performance set first, and then the solo strings and harps later. However, it seems that I would be penalized for making my purchases in that order. I would have to pay twice for the same samples! Is this the way it was intended?

New version up:

http://www.buist.org/david/WeddingOverturefullGS3mod3.mp3

Comments welcome!

I have also been getting stuck notes with earlier versions of GS3. After installing 3.03 the problem continued with about the same frequency as before. Then I tried using Maple (Marble Sounds MIDI program) instead of the VSL MIDI router, and I got much fewer stuck notes: not completely zero, but few enough to make the program useable again. This is on a .gsp set-up using about 70 channels spread over 7 ports, with the VSL tool only on ports 1-4 (MIDI goes directly from the sequencer to ports 5-7). The sequence I am currently working on takes the polyphony peak value up to about 80. I had stuck notes all over the place before, but since ditching the router I now only get the odd one or two.

Thanks, Ackerdemiker, for those comments. As I already mentioned, the first version was mixed to be played in a church: that's why it is so dry. The second version is for listening at home or in the studio.

I think I understand your point about adding a fp< crescendo to the longer notes of the "fanfare" passages. Unfortunately, that is one articulation I find lacking in VSL. There are fp, sfz, sffz, and also crescendos of various lengths, but no crescendo notes beginning with a sharp attack. To create this kind of effect you have to layer two articulations together: a staccato or sfz note to give the attack, and a crescendo for the swell. I tried to do this in the trumpet melody at 01:07 and 02:38, but I did not succeed in getting exactly the sound I really wanted. It is probably possible to obtain a good effect by layering, but it would be much easier if VSL had recorded some samples of brass instruments playing fp< (like in Dan Dean's Solo Brass library which I also have).

Here is a different mix of the same piece (more ambient, for studio listening):

http://www.buist.org/david/WeddingOvertureGS3.mp3

Thanks, Evan, for listening (despite the apparent discomfort caused by the terrible mix!) and for you comments (all of which I find very helpful and thought provoking).

Could you clarify the expression "MIDI pumping effect" with respect to the horns?

I should have mentioned that this recording was mixed to be played in a church, through speakers located high above the congregation. I left it relatively dry in the expectation that the building would add more ambience. Getting the right ambience, especially on percussion is always tricky, I find. Maybe I would do better with ambiently recorded percussion samples? I did actually add some extra NFX reverb to the percussion in this mix (in addition to the global outboard reverb that went on the whole thing), but I guess that may not have been sufficient.

As regards the orchestration, specifically the matter of doubling brass parts with strings, I am not so sure. Wouldn't layering too much together simply lead to a uniform "orchestral" tone color, rather than allowing the different instruments and divisions to stand out in contrast to each other? When orchestrating, I tend to be conscious of the fact that there are many individual instruments there, rather than treating the whole ensemble as a single "orchestral instrument". I like to bring out those contrasts. Perhaps, however, I do this at the expense of the overall blend (which admittedly is important too).

Besides the mix and the orchestration, would anyone care to comment also on the compositional and thematic structure of the piece?

Looking forward to further comments