Kanthos Posted September 20, 2006 Share Posted September 20, 2006 Apologies if this is a redundant post. For some reason, my computer at work hates cookies, despite my best efforts to fix that, so whenever I do a search on the forums, I can never view more than the first page of results. I use FL Studio, but this isn't strictly an FL Studio question, which is why it's not in the FL Studio thread. I was listening today to the Zircon and Pixietricks interview on The Next Level and something that struck me about the Final Fantasy VII remix was the number of different instruments that were used, in addition to the vocals. So far, I've tried to make two remixes. The one that was intended to be orchestral fell through mainly because I didn't have enough ideas and it was turning out too similar to the original track, but I also had trouble with EWQL SO Silver. I didn't have a ton of stuff loaded, wasn't using keyswitches, and was using at most two patches per instrument, but usually one, and I was still running out of memory. I have 2 GB of memory (and 2.4 GHz CPU, if that's relevant) and didn't have much running in the background (I terminated/exited every task I could before launching FL Studio). I could easily do a small group, like a string quartet, but trying to load all four orchestral string instruments, a decent brass section, a few reeds, piano, and some percussion was too much. I didn't even have any effects applied either; when creating the mix, I was just going with EWQL and planning to do effects later. I can't remember if I was using DFD or loading everything into memory at the time. However, I expected that with EWQL, you'd be able to recreate a symphony if you wanted to and had enough skill and attention to detail, and I managed to push the resources of a decent computer enough to know that I could never pull that off the way I was doing things. What I'd like to know is what the secret is to making a mix sound interesting because there are enough instruments. What I did notice with the FF VII mix is that there were never a ton of instruments playing all at once, but overall there was a large number. Is my problem just using too good of a sample set, and should I consider using some instruments from EWQL SO and some from soundfonts or other sources? Are there any other tricks I can do, either within EWQL's Kontakt player, Kontakt 2, FL Studio, or other VSTs that help conserve memory but still allow for a large number of samples/sounds? I welcome any tips/advice. Kanthos, aka "He who will actually finish a remix once his masters is done" Quote Link to comment Share on other sites More sharing options...
zircon Posted September 20, 2006 Share Posted September 20, 2006 Well, for one thing, there was a TON of stuff going on in that mix. As pixie can attest, the project was so big that it took at least 10 minutes to load, 20 to render, and I couldn't play back ANYTHING in real time at all at any buffer setting. It was insane. Different parts of the song generally had different textural choices, and I built it around the vocals w/ pixie's guidance. Take those away and it wouldn't sound as full or as interesting. Generally, I used a lot of pad-type sounds and percussive layers (sometimes up to 10 or more at a time). Lots of doubling and layering, especially in different octaves. So, that all being said, what exactly are you asking about your own stuff? How do you pull off lots of instruments at once witout sounding muddy? How to load a lot of stuff without killing your computer? Be a little more specific and I think I can help. Quote Link to comment Share on other sites More sharing options...
Kanthos Posted September 20, 2006 Author Share Posted September 20, 2006 How to load a lot of stuff without killing the computer. I know that making things sound clear instead of muddy comes down to good panning, eq, levels, and a good sense about what to have playing and what to keep silent at any given time. I was meaning more like are there any tricks to load samples partway through a piece, whether you or others tend to stick to soundfonts as much as possible to keep memory down, whether it's better to get some parts the way you want them and render to audio tracks, that kind of thing. Quote Link to comment Share on other sites More sharing options...
OverCoat Posted September 20, 2006 Share Posted September 20, 2006 Take advantage of internal MIDI routing and automated patch changes. Though patch changes might not be very good when using QLSO, haha Also, a lot of hollywood composers use a multi-computer setup, where they would for example run GigaStudio with different libraries on one computer, and record on another.... though I have no idea how you actually do that, it's something to look into depending on how serious you are about a good orchestral setup. Quote Link to comment Share on other sites More sharing options...
The Vagrance Posted September 20, 2006 Share Posted September 20, 2006 Best CPU-saving type: Bouncing in other words, if you have a section/instrument that you're completely done with, convert each different group to audio, then play the audio tracks, it takes up a lot less CPU than a whole bunch of VSTi's and effects. Aside from that, go to the highest bufferyou can. Quote Link to comment Share on other sites More sharing options...
zircon Posted September 20, 2006 Share Posted September 20, 2006 How to load a lot of stuff without killing the computer. I know that making things sound clear instead of muddy comes down to good panning, eq, levels, and a good sense about what to have playing and what to keep silent at any given time.I was meaning more like are there any tricks to load samples partway through a piece, whether you or others tend to stick to soundfonts as much as possible to keep memory down, whether it's better to get some parts the way you want them and render to audio tracks, that kind of thing. Well, bouncing down completed tracks to audio (or even high-encoded MP3) is advisable if you're running low on memory. I had to do that with about 16 different parts in the DotH mix. Also, when I was done with, say, the vocal part I exported all of it, bounced to a new project, deleted all the vocal tracks + plugins, and then imported the WAV of just the processed vox. There's no reasonable way of loading samples mid-song, as far as I know, with good samples. However, it's always a good idea to see if you can get away with lesser-quality samples if possible. No reason to use QLSO for a big sustained string part in the background if you can use Cadenza Strings. It's only when the samples are exposed that you gotta be careful. Quote Link to comment Share on other sites More sharing options...
sgx Posted September 20, 2006 Share Posted September 20, 2006 Does FL not have freeze yet? Quote Link to comment Share on other sites More sharing options...
OverCoat Posted September 21, 2006 Share Posted September 21, 2006 Freeze doesn't matter much here, when everything's going on at once [apparently!] I thought FL Studio had a freeze function but I never really noticed! Quote Link to comment Share on other sites More sharing options...
tgfoo Posted September 21, 2006 Share Posted September 21, 2006 Freeze doesn't matter much here, when everything's going on at once [apparently!]I thought FL Studio had a freeze function but I never really noticed! It depends on the freeze function that you use. A lot of freeze functions I've seen just bounce things to audio and then disable the the track so that it doesn't eat up cpu. Some freeze functions (like sonar's) do that plus they unload the samples used by any vsti's and vst efffects thus freeing up memory and then automatically reload thinks when you unfreeze. If you don't have a freeze function like that (I'm pretty sure FL's isn't like that) then zircon's method would probably work best. Just make sure that if you're going to mixdown the whole piece into one track and continue working that you have everything the way you want it and that you aren't going to want to make any changes to things. Quote Link to comment Share on other sites More sharing options...
Splunkle Posted September 21, 2006 Share Posted September 21, 2006 Fl does have a smart disable - but thats only for effects, if I recall correctly. FX usually don't eat that much memory, though its really nice if you have a gazzilion FX that eat your CPU. Since your problems seem to be with the generators eating up memory, then it looks like bouncing is the go for you. Well, there's a suggestion for FL 7 - awesometastic freezing. Quote Link to comment Share on other sites More sharing options...
zircon Posted September 21, 2006 Share Posted September 21, 2006 The FL 6.3 beta supports VSTi smart disabling, which is as good as freezing since it does it automatically w/o bouncing to track. Quote Link to comment Share on other sites More sharing options...
sgx Posted September 22, 2006 Share Posted September 22, 2006 If the instrument is disabled and nothing has been rendered to wav, how do you hear it? Quote Link to comment Share on other sites More sharing options...
zircon Posted September 22, 2006 Share Posted September 22, 2006 FL's "Smart Disable" works like this; if the instrument has been inactive for more than 4 seconds, it disables processing on the plugin, thus conserving CPU. This means you can work on an enormous project part by part without your CPU being stressed by inactive parts, but at the same time, you can go back and edit those other things at any time. Of course, if you play everything at once, you're going to be just as lagged as without Smart Disable. However, if you want to listen to all the tracks without being able to edit them in real time, I don't see why you wouldn't just render them. Quote Link to comment Share on other sites More sharing options...
Kanthos Posted September 25, 2006 Author Share Posted September 25, 2006 Take advantage of internal MIDI routing and automated patch changes.Though patch changes might not be very good when using QLSO, haha Also, a lot of hollywood composers use a multi-computer setup, where they would for example run GigaStudio with different libraries on one computer, and record on another.... though I have no idea how you actually do that, it's something to look into depending on how serious you are about a good orchestral setup. Not serious enough to spend the extra money on a second machine, especially if I'm not willing to upgrade EWQL SO Silver to platinum, which I'm not. What do you mean by taking advantage of internal midi routing? My (possibly incorrect) understanding of the term is that it means sending midi data internally through software instead of over a hardware connection. Please correct me if I'm wrong and expand on what you mean by that. Thanks for the tip on Cadence Strings, Zircon. I'll take a look at that. I guess I approached the project with the idea that if I've spent the money for a good piece of software, I should use it as much as possible to justify the expense, which is true to a point, but the music has to come first. Now all I need is time to actually get into this regularly. Stupid thesis. Quote Link to comment Share on other sites More sharing options...
OverCoat Posted September 25, 2006 Share Posted September 25, 2006 Kanthos I mean like, Kompakt has 8 midi outs, and you can use 8 midi tracks per instance of Kompakt to save resources. Quote Link to comment Share on other sites More sharing options...
zircon Posted September 25, 2006 Share Posted September 25, 2006 Kanthos, one last thing I forgot to mention. You can also write a whole piece using small soundfonts rather than QLSO, and then once your arrangement is done, load up QLSO part by part, tweak each thing one at a time to perfection, and then continue that way until you've replaced all the soundfonts with good stuff. This can save memory/CPU while you're still working on the song; a stage where preserving inspiration and ease-of-use is critical. Quote Link to comment Share on other sites More sharing options...
Kanthos Posted September 26, 2006 Author Share Posted September 26, 2006 Kanthos I mean like, Kompakt has 8 midi outs, and you can use 8 midi tracks per instance of Kompakt to save resources. Ah. I was doing that in some places with EWQL, typically when I was using more than one effect for the same instrument. For straight orchestral stuff, that's a great idea, especially since EWQL has reverb and so on built in, but I'm not sure how well it'd work in FL Studio for other things. Someone please correct me if I'm wrong, but as far as I know, the only way to send different tracks from Kompakt to different tracks in FL Studio is to set all but one FL Studio track as MIDI Out, which means you can't use effects on it. The other drawback is that it still isn't going to save you memory for the samples in any way, only for the player. Oh well, I guess there's no magic solution that makes this all easy short of spending a lot more on more hardware. Zircon, that's a really good idea. Yes, I probably could've figured it out myself eventually, but I obviously didn't, so thanks! Right now, I have a fully complete mix in my head; I just need to find time to get it down. It's a jazz version of Endless Battlefields from Final Fantasy Adventure, mainly in 3/4 but with a fast 4/4 section and a more progressive jazz break. Other than exact notes for the solo, I have the whole thing worked out. Someone want to take over my programming job so I can actually do the remix? Quote Link to comment Share on other sites More sharing options...
Kanthos Posted October 5, 2006 Author Share Posted October 5, 2006 Someone please correct me if I'm wrong, but as far as I know, the only way to send different tracks from Kompakt to different tracks in FL Studio is to set all but one FL Studio track as MIDI Out, which means you can't use effects on it. Bump. To add to this, is there any way of somehow routing different MIDI channels from a VSTi into multiple mixer tracks that *can* be used with mix effects? Seems to me that doing so would be common and would be a fairly important feature. Quote Link to comment Share on other sites More sharing options...
zircon Posted October 5, 2006 Share Posted October 5, 2006 Set Kompakt to the desired mixer track. Then, in the FL Wrapper options (upper left corner, the little triangle), make sure "Enable Multiple Outputs" is checked. Outputs 1/2 will be sent through the mixer track Kompakt is set to. You can change the output set in the patch options; it's to the right of the MIDI channel selector. Outputs 3/4 will then go to the NEXT mixer track, 5/6 the next one after that, and so on and so forth. eg. You set Kompakt to mixer track 1, and load 4 patches; violin, viola, cello, double bass. The violins go through mixer track 1. You set the viola to outputs 3/4, and so they go through mixer track 2. The cello is set to outputs 5/6, so that goes through mixer track 3. The double bass is on outputs 1/2, so it will go through mixer track 1. Quote Link to comment Share on other sites More sharing options...
Kanthos Posted October 5, 2006 Author Share Posted October 5, 2006 Thanks once again. You in particular have helped me and so many others a lot. I should send you a cookie or something. From Subway. Along with a bottle of water, none of this cheap "tap" stuff. Quote Link to comment Share on other sites More sharing options...
Kanthos Posted October 7, 2006 Author Share Posted October 7, 2006 Kontakt has it in for me. What I want to do is to have four separate channels, each with one instrument, and have Kontakt or Kompakt (Kontakt in this case, I'm doing a jazz mix) use the input MIDI data and send to four separate output channels. To do this, I have one channel using Kontakt as the generator and the other three using MIDI Out. All the MIDI Outs are set to send on MIDI port 1 to the channels 2, 3, and 4, with the Kontakt channel being channel 1. Kontakt has MIDI port 1 selected as its input port and has my four instruments set up to get MIDI data on the appropriate channel. Kontakt only has three built-in outputs: two stereo outputs, st. 1 and st. 2, and a surround output. Since I can't use the group thing that you mentioned, like I can in Kompakt, I made two more output channels, named them st. 3 and st. 4, and set them up to output to channels 5/6 and 7/8 respectively. So far, so good. What I'd expect to happen here is that mixer channel 1 will get the piano, channel 2 will get the bass, channel 3 will get trumpet, and channel 4 will get drums, everything will be playing the correct part, and won't be playing more than one part. What's happening instead is that the trumpet is getting mixer inserts 3 and 4, and the drums are getting 5 and 6. It's not a fault of the trumpet instrument or the way it's set up; if I disable the piano and send the trumpet output in Kontakt on st. 1, it only uses one insert on the mixer (and it doesn't overlap with other mixer tracks either; I checked for that). One way around this would be to have two copies of Kontakt, one for trumpet and piano and one for bass and drums, so that I can use st. 1 and st. 2 in each, but that seems like a silly wokraround when I'm probably doing something else wrong. I looked throught the FL Studio Bible and Kontakt 2 manual and couldn't find anything to indicate that what I'm doing is wrong or that I've missed something. Any help would be appreciated. Quote Link to comment Share on other sites More sharing options...
perelandra Posted September 2, 2008 Share Posted September 2, 2008 Set Kompakt to the desired mixer track. Then, in the FL Wrapper options (upper left corner, the little triangle), make sure "Enable Multiple Outputs" is checked. Outputs 1/2 will be sent through the mixer track Kompakt is set to. You can change the output set in the patch options; it's to the right of the MIDI channel selector. Outputs 3/4 will then go to the NEXT mixer track, 5/6 the next one after that, and so on and so forth. eg. You set Kompakt to mixer track 1, and load 4 patches; violin, viola, cello, double bass. The violins go through mixer track 1. You set the viola to outputs 3/4, and so they go through mixer track 2. The cello is set to outputs 5/6, so that goes through mixer track 3. The double bass is on outputs 1/2, so it will go through mixer track 1. I am in this same exact position! I enabled multiple outputs, that was easy. But I can't find this 'patch options'. It is nowhere, at least not accessible in FL Studio. Opening Kompakt outside FL Studio though, the 'file' tab is accessible, but not opening it inside FL via fruity wrapper. Would someone be able to walk me through this? Quote Link to comment Share on other sites More sharing options...
Nutritious Posted September 3, 2008 Share Posted September 3, 2008 I am in this same exact position! I enabled multiple outputs, that was easy. But I can't find this 'patch options'. It is nowhere, at least not accessible in FL Studio. Opening Kompakt outside FL Studio though, the 'file' tab is accessible, but not opening it inside FL via fruity wrapper. Would someone be able to walk me through this? Wow, 2-year old bump . You should be looking for this here: This is Kompakt as opened within FL Studio. Quote Link to comment Share on other sites More sharing options...
perelandra Posted September 4, 2008 Share Posted September 4, 2008 WOW thank you so much! This solves all my problems! Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.