Four Horn patch vs. Four Solo Horns...

[mtryon]

I am running Gold XP Pro and am a newbie to EWQLSO. I have the following question; I have the four French Horns loaded in the Kompakt player and this is fine for unison horn parts and sounds fantastic. If I want to stack a chord with the horns, then I take it I must load four solo horns, each in their own instance of the player??? Is this correct, or am I missing something. Otherwise I'm stacking a chord and end up with 12 horns! Yikes!
Thanks in advance...

[Jeff Hayat]

Otherwise I'm stacking a chord and end up with 12 horns! Yikes!


Exactly. The same holds true when you want to do a divisi string section. If you do divisi 1st violins to make a 3-note chord, you can't stack indivdual notes - you stack the section, and wind up with 54 violins. That's the way it works. For horns, you can stack individual notes, but you are probably better off using the sections, and lowering the volume. Do a test, and see what sounds better.

Cheers.

[mtryon]

Thanks for the quick reply, and now it makes total sense to me. I also see why the heavy users are running the orchestra libraries on multiple PC's for each "section". I can see I am going to quickly run out of processing power or memory by loading up individual instruments!
Thanks again for the clarification. This forum is amazing!

[playz123]

Personally, when I am creating a 'horn section' for 'pop' music, I always like to do each horn part separately, and 'play' each horn, thus creating subtle timing nuances for the horns in the section, and of course playing 3rds and 5ths, depending on the harmony required. Chris Hein Horns 2 is ideal for this sort of thing. Of course, once one starts working with sections in an orchestral setting, creating a track for each and every instrument is really not practical for most folks and that's when 'section' voices which may include several horns in each patch become more useful. But the more individually played parts one can mix together in the end, the more authentic things may sound................playz

[digit]

One thing to watch is possible phase issues if the part crosses and you're playing the same note on two horns using the exact same sample. Also, the tuning is usually never exactly the same with real players.

[johng]

I wouldn't worry about having 12 horns in many cases. Unless you really want a chamber music sound, it is no big deal, in my personal opinion.

I think that sequencing requires one to struggle with so many shortcomings that it's ok to indulge oneself a bit.

[mtryon]

Thanks for all the information guys...this is great.
While laying in bed this morning thinking about this I had an idea I'm going to try. (after coffee!) For my stacked four horn chords that require four instances of the Kompakt Player be loaded (one for each horn) how about I try this? I load ONE instance of the Kompakt Player and then load into that one player FOUR of the same articulations for that French Horn passage on MIDI channels 1 through 4. THEN, Play my stacked chord into the player. Then open up the editor and change the stacked chord channels to MIDI Channel 1 for the first chair, 2 for the second chair, etc and move them around in the editor so they are all not perfectly timed or lined up. OR, maybe better, I could play all four horns in separately by playing MIDI channel one on the first pass, then go back and play the 2nd horn part on MIDI channel 2, etc. This should work right? At least I hope it does. I think I'll run a little memory test and have the performance viewer on so I can see if this method saves me PC resources as I don't know how much memory each instance of the Kompakt Player uses doing it the other way. Also, by doing it this way I think it might be much easier to VIEW the horn section in one editor window with all the horn section parts in one editor window, as opposed to viewing four different editors the other way. If this works I could see breaking up the string sections this way too with the smaller groups of four, or one, etc.

I think I see already that the four instances of Play loaded may use less resources than the four instances of the same articulation. (unless it only gets loaded once?)
...fingers crossed. Now for my coffee....

[mtryon]

OK, I tried both experiments. In the first one, I did as recommended here, and I loaded three separate instances of the Kompakt Player and loaded the QLEG solo French Horn into the Player for each French Horn. This gives me three separate horns in three separate Kompakt Players on three separate tracks in my DAW. I then recorded a simple four bar passage with three separate recording passes playing the root on the first pass, third on the second pass, and fifth on the last pass.
The results of resources used on playback were: CPU Averaged 7 % use. RAM used: 657 Mb.

Next I loaded ONE instance of the Kompakt Player but loaded THREE instances of the same articulation in this one player, on MIDI channels one, two, and three I loaded the same QLEG articulation. I noticed that three instances of this articulation DID NOT LOAD! Only one instance loaded and totaled only totaled 268 Mb!
Then I recorde the same parts but did them on ONE pass playing the same triad. (1,3,5)
Then, and here's the GREAT part. I opened the MIDI editor, and changed my horns to separate MIDI channels, first horn on channel one, second horn on channel two, third horn on channel three. ALL the notes were now lined up perfectly...but that is NOT what I wanted, so... I just slid them around a little, changed the velocity on all three and BINGO!!! I have all three horns playing slightly separate parts in one MIDI editor and using only one instance of the Kompakt Player!!!
To my surprise; the resources used? 8% CPU and only 268Mb or RAM!!

How about that!?! Can anyone offer any reasons why I shouldn't do this for all my sections as I'm only running one PC (AMD Athlon quad core, 8Mb RAM - 10K RPM raptor drives)

[A.Leung]

Answer:

Because using the same exact sample for every part gets boring and static. Players all play differently, in expression, dynamics, tuning etc. You might get away with it for short samples but I certainly wouldnt try it with say for example strings.

Even the same player him/herself never plays the same note the same way twice.

[mtryon]

I certainly understand what you are saying, and please, no newbie musician here, (40 years pro bassist in L.A./studio scene---Mike Posts TV/Film stuff---that will make a few here laugh! Ha!), so I do understand individual sounds/techniques, etc., however, in the tests I just ran, my three horns in one MIDI editor did sound more like, individual players, and was much easier control and achieve this effect since I didn't have to open up three editors at once. BTW, if I loaded the QLEG horn samples in three different instances of the Player won't I get the same results, since again I am using the same samples?... only using up more RAM?

It was very easy to select each group of notes, change the MIDI channel, and slide the notes and velocities around a bit thus making them sound like three individual players. I'm wondering now, based on your advice not to do this, just what I will run into if I do, say, divisi the first violins. I'll give it a try. One thing for sure, the three horns in the single instance of Kompakt used one third the RAM, though used nearly the same CPU resources. That was a bonus for me on my limited PC. So I was VERY surprised to see a more efficient way to work that also used 1/3 the RAM.
Thanks again for all this help....like I said before this is an amazing and wonderful forum and I really appreciate all the advice. And, if I'm missing something, I would really appreciate a deeper explanation if anyone has the time.

[A.Leung]

If it sounds good - do it. There really is not a whole lot more to it than that. Saving resources is important of course. Samples by themselves present a very challenging problem in Midi mock-ups, ie. they tend to give themselves away after about 15-20 seconds so its always been my goal to disguise that fact by:
1.) Alternating samples as much as humanly possible
2.) Stacking or layering samples
3.) changing articulations and variations more often then I would if I were writing for a real orchestra.
3.) Using Round Robin type samples as often as possible.
4.) Editing the living He!! out of my controller data by always having different velocities, expression, volume, other dynamics, etc.

Cheers,

[mtryon]

THANK YOU for confirming #3 (changing articulations more than if writing for real ochestra)... I was wondering about that being a viable way to achieve realism, and I am going to experiment with that a lot as I try to get better. Until I get more experience I do love being able to use the MIDI editor with the stacked parts in this one editor rather than several editors cluttering the screen and me being confused with time lines on the piano roll, etc. Also, the saved resources this way is a plus for me now. I'm not yet ready to commit to multiple PC's running the orchestra. So much to learn...so little time...