Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Sampling techniques (Read 7400 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Sampling techniques

1. Is convolution reverb used for mimicing a real concert performance only? For film scores, soundtracks, game soundtracks, cartoon/anime music etc... is algorithm based reverb better?
2. other than the reverb, and articulations/dynamics (supported by the libraries), I don't think there is anything else to do for standard music? I looked at the boasted kontakt4's manual, most of them are what FLstudio does. Like emulate old tapes, fm radios, synths, multi channel mixing etc... If you look at the players from Vienna Symphonic Library,

http://vsl.co.at/en/211/497/1686/1673/1673/115.htm

it comes with very basic features because it doesn't emulate disco/analog music.

For kontakt 4, they have

compressor
limiter
inverter
AET
Surround stuff
Saturation
Lo-Fi
Stereo Enhancer
Distortion
Send Levels
Skreamer
Rotator
Twang
Cabinet
Phaser
Flanger
Chorus
Reverb
Delay
Convolution
Gainer
Filters


out of these I only need the compressor, limiter, send levels, chorus, reverb, convolution, gainer, filters. Most of these exist in cheap players like the Garritan Aria player. So is computer based sequencing a pretty simple thing?

http://forum.image-line.com/viewtopic.php?t=45272

Image line says that "amateurs" exaggerate the features of their equipments.

Sampling techniques

Reply #1 –
Quote
So is computer based sequencing a pretty simple thing?


I find it is. I in the past used to use the plain Yamaha XG softsynth, XGEdit for doing the XG specific sysex/effects, and Cakewalk Sonar 4 for the environment. Didnt need anything else. Now use Milkytracker with a set of samples. Modplug when experimenting with IT, though it tends to render things like low pass filter cutoff wrong.

Dance music to me just seems very heavy on the chorus/reverb/echo/and filters. LOTS of filters. You can very easily do a MIDI sequence with dance in mind, as long as you use plenty of filters and plan your patch choice wisely. I bet with FLStudio or something, its even easier as you get to play around with decent samples at least.

Heck, a TR808 kit, a TR909 kit, plus some saws and triangles can do something kickass. Like this:
http://tssf.gamemusic.ca/Remakes/rocksolid.it
http://www.youtube.com/watch?v=FVIcm_o1Fmo

Sampling techniques

Reply #2 –
Yes, newbies exaggerate the gear factor a lot. For example, I've heard/read many times how a certain DAW sounds better, even tho it's been demonstrated that the output is exactly the same (at least in the major ones that have been tested). Placebo works wonders, as usual.

That being said, there are differences between effects. You can definitely find some reverbs that sound bad. But if you know what you're doing (which takes time to learn), you can make a professional sounding product even with freeware stuff. There's a whole sea of effects out there, though, it can take a while to find the good ones.


Quote
1. Is convolution reverb used for mimicing a real concert performance only? For film scores, soundtracks, game soundtracks, cartoon/anime music etc... is algorithm based reverb better?

Better is what sounds better to you. But in general I'd agree (although it should be noted that convolution can also be made from other reverb units, not only from real spaces). Real space convolution can sound very realistic indeed, but this realism can be boring sometimes.
I suggest you take a look at the KVR Effects section, there's also a sticky thread with some links to IRs. A decent free IR plugin is SIR. There's also Freeverb3, which works without latency.
A couple of good free algo ones: Mo' Verb and jb_Omniverb (you might need to search around for this one).

Sampling techniques

Reply #4 –
Quote from: Brand on
Better is what sounds better to you. But in general I'd agree (although it should be noted that convolution can also be made from other reverb units, not only from real spaces). Real space convolution can sound very realistic indeed, but this realism can be boring sometimes.

http://www.youtube.com/watch?v=oeYH-2_93qY
http://www.youtube.com/watch?v=IT12DW2Fm9M
http://www.youtube.com/watch?v=ZWuNf4gxwuM
http://www.youtube.com/watch?v=6_e3oldjZbw
http://www.youtube.com/watch?v=wrrWWvLAWTU...nel_video_title
http://www.youtube.com/watch?v=-f3gGtSuKxg
http://www.youtube.com/watch?v=_VayXWsCjdY...feature=related
http://www.youtube.com/watch?v=Yw_Gp-3HHTI

Are these done with convolution or algothrithm reverb?

Sampling techniques

Reply #5 –
It's very hard to say this, especially on a finished mix. Even with a single instrument you'd have to be a reverb expert to recognize a specific reverb or IR. But if you want my guess and if by convolution you mean convolution of real spaces, then I'd say mostly algorithm reverbs were used, since video games music generally doesn't aim at simulating real spaces.

Sampling techniques

Reply #6 –
Quote from: Brand on
It's very hard to say this, especially on a finished mix. Even with a single instrument you'd have to be a reverb expert to recognize a specific reverb or IR. But if you want my guess and if by convolution you mean convolution of real spaces, then I'd say mostly algorithm reverbs were used, since video games music generally doesn't aim at simulating real spaces.

So with algorithm reverb, do all DAWs/Players' implementations sound pretty much the same? They are all called hall, room, church, reverse, non-linear etc... These are all programmed by the audio engineers at the respective companies right?

Yes, I meant are these based on the IR profiles of real spaces or not.

Sampling techniques

Reply #7 –
Quote from: HTS on
So with algorithm reverb, do all DAWs/Players' implementations sound pretty much the same?

No. As I've said before, the differences can be big. But it's up to the listener to spot and evaluate them.

Sampling techniques

Reply #8 –
Quote from: Brand on
Quote from: HTS on
So with algorithm reverb, do all DAWs/Players' implementations sound pretty much the same?

No. As I've said before, the differences can be big. But it's up to the listener to spot and evaluate them.

for orchestral music, other than the reverb, what other effects are needed?

Sampling techniques

Reply #9 –
http://www.secaudio.ch/side122.html

Quote
Users of convolution reverbs have found themselves limited by incompatible library formats and cumbersome import facilities, restricting their ability to mix and match impulses between reverbs. Finally, the heavy CPU overhead of native convolution reverbs has made their use in live sessions difficult due to high latency.


What does Avid mean by this? That sample libraries must be programmed for each type of convolution engine and IR?

Sampling techniques

Reply #10 –
Quote from: HTS on
http://www.secaudio.ch/side122.html

Quote
Users of convolution reverbs have found themselves limited by incompatible library formats and cumbersome import facilities, restricting their ability to mix and match impulses between reverbs. Finally, the heavy CPU overhead of native convolution reverbs has made their use in live sessions difficult due to high latency.


What does Avid mean by this? That sample libraries must be programmed for each type of convolution engine and IR?

Probably:
1. That there are several file-formats used for the impulse responses that you load into the convolver, and that any given convolver may read some but not all formats.
2. That convolvers tend to have high latency (but is this due to cpu usage, or algorithmic aspects? I would think the latter)

Sampling techniques

Reply #11 –
Quote from: knutinh on
Quote from: HTS on
http://www.secaudio.ch/side122.html

Quote
Users of convolution reverbs have found themselves limited by incompatible library formats and cumbersome import facilities, restricting their ability to mix and match impulses between reverbs. Finally, the heavy CPU overhead of native convolution reverbs has made their use in live sessions difficult due to high latency.


What does Avid mean by this? That sample libraries must be programmed for each type of convolution engine and IR?

Probably:
1. That there are several file-formats used for the impulse responses that you load into the convolver, and that any given convolver may read some but not all formats.
2. That convolvers tend to have high latency (but is this due to cpu usage, or algorithmic aspects? I would think the latter)

1. What kind of file formats? audio formats like wav, or library formats like nki, fxp, exs? Or something totally different?
2. Is latency important for sequencing (i.e. not real-time generation+playback)

Sampling techniques

Reply #12 –
Quote from: HTS on
1. What kind of file formats? audio formats like wav, or library formats like nki, fxp, exs? Or something totally different?

I believe that impulse responses are usually put into specialized files, even though wav is sometimes used. You'd usually want floating-point precision and perhaps some metadata. I have been away from this game long enough that I dont remember the file acronyms. Anyways, this was Avids claims. Do they have a commercial interest in portraying IR-reverb as worse than it is, or do you think that they are trustworthy?
Quote
2. Is latency important for sequencing (i.e. not real-time generation+playback)

Often, no. It depends on your way of doing things though. If you are laying down a drum track using a midi controller, depending on live feedback to hear what you are doing and want to hear your drums reverberated while recording, you may want total latency in below 10ms.

If you are inputing your music clicking and pointing with your mouse, then hitting play to hear the rendered result, you may be able to live with seconds of "latency" (rendering time).

-k

Sampling techniques

Reply #13 –
What are the best impulse libraries? Every DAW has some convolution plugin right now and claim to be as good as altiverb.
 

Sampling techniques

Reply #14 –
Quote from: mudlord on
Quote
So is computer based sequencing a pretty simple thing?


I find it is. I in the past used to use the plain Yamaha XG softsynth, XGEdit for doing the XG specific sysex/effects, and Cakewalk Sonar 4 for the environment. Didnt need anything else. Now use Milkytracker with a set of samples. Modplug when experimenting with IT, though it tends to render things like low pass filter cutoff wrong.

Dance music to me just seems very heavy on the chorus/reverb/echo/and filters. LOTS of filters. You can very easily do a MIDI sequence with dance in mind, as long as you use plenty of filters and plan your patch choice wisely. I bet with FLStudio or something, its even easier as you get to play around with decent samples at least.

Heck, a TR808 kit, a TR909 kit, plus some saws and triangles can do something kickass. Like this:
http://tssf.gamemusic.ca/Remakes/rocksolid.it
http://www.youtube.com/watch?v=FVIcm_o1Fmo

I'm going to resurrect this thread.

I'm wondering how far we have "come" since the old days. The companies are in the business to make money, but are they making honest money? They will tell you that Wendy Carlos' stuff are obsolete now, but I don't see any major digital composer endorsing the new expensive stuff. The question is, what would Carlos or Uematsu do with the VSL at their hands?

In fact, most of the rants come from losers complaining that the market is too "dumbed down" to appreciate their stuff. That's why they are dead poor. not because they blew too much money in their 30 different sets of beats.

http://www.northernsounds.com/forum/showth...ample-libraries

I get annoyed when I encounter too many shills in a single day online, trying to either coax or simply insult you hoping to get you to buy the new releases.