View Full Version : Grand Hungarian Overture

Andy Brick
04-09-2003, 03:43 PM
Hi Everyone!

I have just finished a new piece call Grand Hungarian Overture. I would love to hear your comments and critiques.

About 90 percent of the strings came from GOS and, if you follow the link below, it will bring you to a page with the mp3 and my string patchlists for this piece so you can follow along and hear which is patch is playing and when it is playing.

http://www.andybrick.com/projects/hungarianoverture/ (\"http://www.andybrick.com/projects/hungarianoverture/\")

Thanks, I hope you like it!

-Andy Brick

His Frogness
04-09-2003, 05:50 PM
All I can say is....WOW! Excellent composition. You definitely have a knack for instrumentation. I\'d offer more critiques if I knew what it was for, but simply as a composition, I loved it. If you\'re looking for constructive criticism, I thought it could use some more verb in spots, but I guess that\'s a personal preferance thing. So, in all honesty, how long have you been working on that? A piece like that would\'ve taken me, ohhhhh, a month. To get it to sound that authentic, anyway. Good job!!! images/icons/smile.gif

04-13-2003, 12:06 AM
Ah what the hell. I just sent you an e-mail but might as well say it again.
Really like it. I t makes me very happy to have received my GOS lite in the mailbox this weekend. I got some serious work to do!

Good work


Tom Hopkins
04-14-2003, 02:13 PM
Excellent work Andy. The detailed patch info should be a big help for anyone interested in achieving results like this. It\'s rare to see such a generous gesture.


04-14-2003, 06:05 PM
Sounds Awesome Andy! Great work!

The patchlist idea is great too!


04-14-2003, 06:46 PM

Great piece, but there is just something weird about the overall sound.


In the beginnning it sounds like its being played through some small speakers then re recorded, then when the cellos come in, it feels more \"mixed\" but still weird. The mono returns on the woodwinds sound a bit strange too.

It sounds like an \"overdubbed\" recording. If you know what I mean. Just a bunch of instruments recorded in different rooms seperately and then mixed together.

I mean its a really good piece, but it just doesn\'t sound like one orchestra at one time IMO. Sort of like a big composition but a small recording. Sometimes there\'s not much to be dont abou tit tho.

My own opinion tho, and I\'m not saying I could do better (especially in terms of composition/orchestration). Much of the MIDI mockup-ing is great too.

04-14-2003, 07:05 PM
Hello Andy,

First of all the piece is great and I\'m sure you could fool about 99.9% of the populace!

Could you give use some additional details on the other samples? Are you using Dan Dean for brass and woods? What about percussion?

Could you give us some info on your general rig and how many computers you have running together? Also, the screenshot is from DP, is that your sequencer of choice?

Andy Brick
04-14-2003, 10:26 PM
Hi Everyone,

Thank you all for the gracious words. I am very humbled.

Let me start with Kings critique.

Perhaps it sounds weird to you because I wound up digitizing all the midi tracks in mono. The problem for me is that midi is a 2 dimensional sound whereas a live orchestra is a 3D sound. Not only do we hear Violins I on the left and low brass on the right (in a modern live orchestra) but we hear violin 1 closer than we hear the low brass because the low brass sit in the back. Without the 3rd dimension, stereo samples dont sound right to me and, as such, I render midi to audio in mono so I can have more control over the placement of instruments. The draw back is that you dont have the natural bleed through the stereo image. To sensitive ears, that can sound very strange. I am in favor of samples with the natural reverb in the sample. As far as I can tell, its the only time I hear things in 3D. Certainly Altiverb and the like sound great but they dont solve the 2d problem to my ears. A poor mans solution to this problem is to lightly mix each instrument with a similar reverb then, at the mastering level, add another layer of larger reverb. Ive even experimented with 3 layers of reverb. It seems to have a positive effect but undoubtedly, some folks like King will hear beyond that solution. Yes King, it does sound strange. MIDI sounds strange.

Frog, the piece took me 4 days to write and about 6 full days to sequence and mix. I didnt quantize a thing. Ive found that recording at about 70% tempo yields a tight performance at full speed and a much more natural feel. It also helps to diminish the maladies of latency.

Christian and Tom, thanks for the kind words. I dont mind sharing my techniques with others. In fact, after my original post I added the short score for the piece to my website. If you are interested click on the \"scores\" button on the home page and you will find it there. It contains the entire score on 5 staves, all my orchestrations in shorthand and I even left my little notes about harmony, rhythm and form in the short score for those of you that care about things like that.

Midphase, my rig is pretty simple. 3 PCs that I control from a Mac using Digital Performer. DP is my sequencer of choice. Ive been with MOTU since Performer 1.2 and I really like their products. I am watching MachFive with great anticipation. I also like to have my orchestra on line. I feel very strongly that this is necessary for me. It seems I am sort of alone in this perspective but I just cant imagine spending the time creating GSPs for every project.

I wont get into the specifics about my libraries right now because I have spent a lot of time tweaking them. As such, without having my version of a given library, you wouldn’t get the same sound. The one exception is GOS which I use straight out of the box.

I think just as important as the library is what the composer brings to it. A flute gliding in unison with a high trumpet wont be heard as a flute. It will be heard as a silky shimmer glistening in the power of the trumpet. A clarinet doubling the violas wont really be heard as a clarinet. It will however give a depth to the violas that will draw the listener inside of the sound. These and many many other orchestral colors and textures enrich not only the live orchestra but the midi orchestra as well. Good composition and orchestration is a good start to a good sequence.

As well, try to work with the samples, not against them. In the middle of the overture when I go into the harmonic sequence with violins taking the double dotted quarter note theme accompanied by flutes in triplets above and bassoons and clarinets in triplets below there is a short little trade of 2 bars with the trumpets. Those trumpets were slightly too short for what I had originally intended. The problem was, my next longest trumpet instrument was too long. So instead of trying to make the shorter trumpets into something they aren’t, I let them be. They sort of give the passage a cute, lighter feel that I have come to like very much.

Well thats it for now. I hope I have been at least a little helpful.

All my very best regards,

04-14-2003, 10:50 PM
MIDI is definitely not the problem here Andy.

you\'ve gotten some really great performances out of the samples, and that is usually extremely difficult with MIDI.

04-18-2003, 09:16 AM
Awesome Job, Andy. Over the year or so that I\'ve been visiting this site I have listened to quite a few midi mock-ups from users here. Yours is among the best of the best, not just from a compositional/orchestration perspective, but also it\'s production value.

I have long held the point of view that stereo samples can be counterproductive inside a thick orchestration, producing numerous summing problems, cancellations, and adding to an \"unnatural\" sound.

I honestly don\'t have enough experience with ambient samples yet to have an opinion on them, but my sense is that this may also become problematic inside thick orchestrations. They would be great for small ensemble and soloing.

I think it would be wise for people to follow your advice of rendering to mono audio tracks from stereo samples in some instances.

Thanks for providing an inspiring piece to this forum and some valuable insight into your approach.

04-18-2003, 12:22 PM
Personaly I think it is an utter mistake to render to mono with GOS. There is too much ambient content, and you will get get some phasing issues as well as send all reflection data to sound as if its coming from \"farther back\".

Stereo samples are useless for orchestral mockups if they have no ambience. Dan Dean woodwoonds and brass for example take well to losing the stereo feild completely. GOS, SAM, DDBE, etc dont.

Another issue about mono-izing a string section is that sections are not small. They are fairly wide, so if you want more \"placement\" you should collapse the field slightly using both left and right panning pots, or a plug in like Waves S1 or use Soundstage with stereo inputs. Making a section recorded instereo into a mono track effectively turns it into nearly the same as a single speaker output being reverberated (or in a room with impulses)

I also dont agree with the whole \"MIDI sounds weird\" argument since I\'ve heard some GREAT midi mockups using current and older libraries. I also believe this one could be one of those since the MIDI work and composition/orchestration is quite good.

I\'m not trying to be a jerk, its just what I hear and I doubt its because I have sensetive ears, but even more so if it *IS* because I do, then I\'m pointing out things that could make it sound better overall.

I also think that an influence to render out to mono with GOS will make alot of future GOS demo\'s sound weird, especially if its done by people who cant hear the problems this causes (as there are people that cant hear even the most problematic phasing problems). Its something I wouldn\'t like to see happen.

If you do want to get more focused placement use the DSP mixer (works), Waves S1 (better), or Sound Stage (my favorite with good trapping/absorbtion settings).

His Frogness
04-18-2003, 06:19 PM
Originally posted by KingIdiot:
If you do want to get more focused placement use the DSP mixer (works), Waves S1 (better), or Sound Stage (my favorite with good trapping/absorbtion settings). <font size=\"2\" face=\"Verdana, Arial\">I can\'t see how you could use S1 as a master effect. Wouldn\'t you have to insert it into every channel to really have it work for you? If that\'s the case, I can\'t say for sure, but I\'m pretty confident my 2.8GHz couldn\'t handle an S1 on every channel of an orchestral set-up. Additionally, are you refering to the Stereo-Imager or the No Shuffler. I\'m not sure what the difference is. Sounds like you might know images/icons/smile.gif

04-18-2003, 06:49 PM
you dont really have to use S1 on every channel, mostly just larger sections. The mono trick works well with solo instruments that aren\'t recorded in a large hall, so things like Dan Dean woods would take well (and generally mix better) in mono.

a 2.8 should handle quite a few instances of S1. I dont use the shuffle algorithm, jsut because I dont like it. I believe it works by slightly \"shuffling\" lower frequencies, split via crossover, across the width you create, to make for a more level sounding \"spread\" accross all frequencies, it jsut never was something I liked personally, your mileage may vary.

Another option is using stereo pan pots (both left and right pan sliders) to create the width you want instead of rendering to mono. Now I\'m not saying that rendering to mono is the \"wrong way\" to do things. Each person should go with what they hear, and if you like it, more power to you. Its jsut that I dont believe it can be just my ears that hears the problems this creates, not to mention what it does can do in subtlties to every listener.

04-18-2003, 07:14 PM

I don\'t disagree that larger sections should probably stay in stereo. I certainly render them in stereo. I believe I am suggesting that in the case of thick orchestration, stereo samples from every section may work against you.

Now, I\'m not trying to be a jerk when I suggest that the entire notion of a stereo recording, even one of a live orchestra, sounds \"unnatural\" compared to being in a live hall during a live performance....it does. 5:1 audio is going to help the recording world sound more natural.

I\'m wondering how many midi mock-up masters like yourself, are working samples in 5:1, and what enhancements, or disappointments, you are discovering. And furthermore, what stereo ambience in the samples themselves, are doing to a 5:1 mix. I can see this as a potential problem.

04-18-2003, 10:26 PM
I\'ve only toyed around with 5.1, using an extra reverb with a lager predelay and having it sent to the surrounds.

Its not something I have in wave, since it was at another studio. It worked really well, but in no way way it like a live listening experience.

I\'m thinking that QLSO will get us to a good 5.1 mix, but not a truly live listening experience. That will be up to Franz and headphones.

Anyhow, I\'m thinking that delay lines will help too, but in the end I think it will only sound like a \"surrounding\" mix and not a live sound.

I think I\'ll be getting more and more into surround this year so I\'ll keep you all up on my esperiences. I still however think this will be comletely new territory for me, so my comments will only start from very little experience, I\'d rather people who\'ve been tyring surround a bit to give us some ideas. I\'m a stereo guy images/icons/wink.gif

now when 3-d sound modeling starts to show up... images/icons/smile.gif as in ray tracing for sound images/icons/smile.gif

Andy Brick
04-20-2003, 08:36 PM
Wow, who knew such an innocent post would create such a commotion?

I hope I didnt lead anyone astray with my comment about digitizing in mono. When I did the GOS review for music4games I originally mixed by rendering in mono then decided to remix by rendering in stereo and without a doubt, all else being excluded, stereo sounded better. For me, when things start getting thick, as they do in the Hungarian Overture, it just becomes a matter of what process will allow me the most clarification in my mix. I dont profess to be a technical wiz but trade off in sound vs. clarity favored a mono solution for me.

Quite a few people have taken a listen to the piece so far and I have been receiving all sorts of emails asking for tips and suggestions. The best I can do is just tell people how I did it and encourage everyone to experiment. My techniques may not be technically perfect and I suppose there are folks out there who can do better but by showing everyone one way that gets a good result, regardless of how imperfect it may be, perhaps we can raise the bar a bit and get more people to produce great sounding music.

Maybe Ill try mixing GOS stereo and winds and brass mono in my next piece and see what happens.

Thanks for the words of wisdom King.


04-21-2003, 01:42 AM

I truly believe that what makes the piece sound so good is both the way you did the midi sequencing and the orchestration, as well as the composition itself.

These things really standout in the cue, the mix, as always, is secondary (third?...fourth?), I\'m jsut pointing out things that I think really should be mentioned. This cue CAN sound better IMO. If it is due to sensetive ears, then I guess its good to be pointing it out for people to read jsut so they might take a listen to the things I\'ve pointed out. In the end it makes us all listen to things from a couple different POVs.

Still great work.

04-25-2003, 11:46 PM

I mostly agree with King that the composition and orchestration is what stands out in this piece, in addition to the overall performance. At the risk of sounding like I have a bad attitude, I have never believed in the concept of striving for perfection. I don\'t think that \"perfection\", however one defines it, gets us anywhere worth going.

What resonates with me is a certain passion and invention in the playing. Or an intelligent take on an otherwise cliche\'d musical technique....be it playing or composing.

Even world class orchestras and soloists are not \"perfect\" like their recordings in a live concert. But if a certain passion comes through, no recording can touch it....mistakes and all. The more polished a recording becomes, it often sacrifices raw passion.

I can\'t tell you how often in the past, I wish I could go back to a take that had some mistakes, but felt more honest in interpretation. In my haste, I would often delete it and try to capture the take again....the ol\'...\"I know I can do that better\" syndrome. Alas, I would end up with a more technically perfect take, but one that also lacked that extra \"something\". Call it more reserved, or whatever.

Sometimes now, with a simple part, I\'ll record direct to an audio file, bypassing midi altogether.

I guess for me, when it comes to samples, and sample libraries, the MOST important thing is to get as convincing as possible with as little tweaking as possible. If I want to spend days on the production and mix of a cue, one can get great results from Miroslav or the other older libraries. The newer libraries seem to be working toward this end....and for that, I am grateful.

05-01-2003, 05:41 AM
Hi Gang,

Wonderful composition, Andy. And generous of you to share the details behind the sound.

The discussion regarding the mono versus stereo rendering of MIDI tracks to audio, for subsequent mixing, is of great interest to me. I have been puzzling over this, in trying to achieve a more realistic sound, and especially so since Bruce Richardson recommended the mono approach on the main forum. His suggestion (I hope I\'m not twisting his thinking here) was to \'use mono tracks for samples without necessary or appropriate stereo content\' (my words). A dry solo clarinet should be set in a mono track. Or, with SAM horns, use a mono close mic\'d instrument backed by a softer far mic\'d version, an approach that helps localize the horn section\'s sound naturally, while also giving a realistic depth/ambience.

I find that getting a coherent overall ambience is more difficult using ambient samples from different libraries, because the samples were recorded in different sized/sounding rooms. I suppose this is the thinking behind VSL\'s approach, which is proving very successful - eliminate ambience mismatch as a mixing factor.

And ambience rich samples are sometimes very difficult to work with. Michiel Post\'s Bosendorfer 290 is an excellent example. The \'wet\' version sounds wonderful. But, it is spread from far left to far right. If you try to move the image back a ways, to put it in the middle of the stage in a concert hall, simply panning the left and right samples more to center does \'narrow\' the image, but the ambience now sounds whacky and unnatural. Hence, I use the dry version, and add my own reverb.

My own conclusion is that stereo samples rich in large-space ambience should be used only in an appropriate context (i.e. large space application), or they add to an out-of-kilter sound and image. As I recall, this was exactly the reason that in the GOS development it was decided to keep the library relatively dry - in order to increase its general usefulness. I have also begun using Bruce\'s tip, to only use mono audio tracks for instruments whose samples don\'t contain essential or appropriate stereo content.