• Register
  • Help
Page 1 of 2 12 LastLast
Results 1 to 10 of 13

Topic: GS GVI for live playing

Share/Bookmark
  1. #1
    Senior Member
    Join Date
    Jun 2005
    Location
    Martinez, California (Bay
    Posts
    163

    GS GVI for live playing

    Hi folks. I am not a composer. Although I own Sonar 6 PE, I don't really do any sequencing any more. I used to in my last band but now I do strictly live playing with an 8 piece show band. I want to start using GS samples for my live playing. I have a pretty powerful rack mount PC system that should provide me all the horsepower I need for live playing. I just bought GVI. I also have GS 3.2.

    It seems to me that GVI would probably be a more simplistic and sensible way to use GS for live playing. But being new to this, there are some questions I have:

    1. How many individually addressable (thru midi) instruments can I have. I am not talking about memory constraints but rather how many different midi bank/program combinations does it support. Bercause I can not be loading anything in the middle of a gig.

    2. I really don't want to use Sonar live because there are no real projects. Sonar seems pretty cumbersome just to support the GVI software. So what is a good alternative to support GVI that would not be as resource intensive as having Sonar up for just that purpose.

    3. Since I will only be using GS for live playing situations, my polyphony needs are probably going to be quite low compared to those of you composing complex music. With that in mind, do you think that using GS GVI live will be reliable? Would I want to just consider using GS 3.2 stand-alone for live playing instead? Pros vs cons?

    Thanks all for the great support you give to those of us who are still just learning how all of this works.

    Rob

  2. #2

    Re: GS GVI for live playing

    Hi Rob

    This is something I'm very interested in too. I currently play keyboards in a couple of covers bands and I already use Gigastudio for most of my sounds. I actually take two 2U rackmount PCs in a rack to gigs so I have completely redundant backup, but as it turns out I haven't had any problems since I got my rig completely configured and sorted.

    I actually use Giga 2.5 still on my live rig because I could use the old licences when I upgraded to 3.0 for the studio, and don't really need the extra functionality of Giga 3 for the naturally limited area of live playing. I was briefly tempted to upgrade my live rig to Giga 3 partly because of the integrated VST effect handling, so I could use for example Scarbee's VKFX along with his Rhodes and Wurly samples. But when I tried this out on Giga 3 in the studio, I found the latency was so bad I could never use it live! Never resolved that one, and it's one reason I'd be interested in switching to GVI. With GVI in a VST host app, I could simply insert the FX further down the channel and it should work fine, not being subject to running "within" Giga. I can only presume it's something to do with having to get the signal from the kernel level of Giga to the app level of the VST effect and back again that causes this unacceptable latency, which is hardwired to four times the soundcard buffer at present. (And of course in a live situation, you don't want to be tempting fate with the lowest latency you can get away with, you want to give yourself a margin for stability.)

    Anyway...

    Quote Originally Posted by goldberg96
    1. How many individually addressable (thru midi) instruments can I have. I am not talking about memory constraints but rather how many different midi bank/program combinations does it support. Bercause I can not be loading anything in the middle of a gig.
    I don't think this should be an issue. The max is probably 16, as I think that's a VST limitation. BUT you can presumably get round this just by running more instances of GVI within your host. There shouldn't be a limit to this - you can for example run as many instances of Kontakt or Halion inside Cubase, Sonar or one of the "live" hosts like Forte as your want, up to what your CPU and RAM can handle.

    And I agree absolutely about not loading stuff. I load my entire template before the gig then put the PC to one side and forget about, selecting channels from my keyboard. It's the only way.

    2. I really don't want to use Sonar live because there are no real projects. Sonar seems pretty cumbersome just to support the GVI software. So what is a good alternative to support GVI that would not be as resource intensive as having Sonar up for just that purpose.
    Again, I agree entirely. There is however a LOT of choice here.

    I personally use CHAINER by xlutop. I'm pleased to say that it's absolutely, totally 100% stable and has never let me down. It's just a simple little thing with ten channels into which you can load "chains" of VST instruments and effects. I use it for a few instances of NI B4 and Pro-53 that I run alongside Giga. If I used GVI I'd probably stick that inside it too.

    It doesn't do much - just holds the instruments and effects, allows you to set overall levels for them and routes the whole lot to your soundcard. But it's very cheap and like I say, for a live system you want stability and simplicity, which it delivers.

    There are a few others around like this. OTOH if you want something more ambitious, a lot of people rave about Brainspawn's FORTE, which has loads more functionality for changing "scene" with different combinations of programs, settings etc. I'm sure it's probably very reliable as it's designed completely for live use, and is used I believe by some pretty big names, but I have no real need for it as a lot of that functionality I have already in my keyboard, and it's more convenient for me to operate it from there.

    3. Since I will only be using GS for live playing situations, my polyphony needs are probably going to be quite low compared to those of you composing complex music. With that in mind, do you think that using GS GVI live will be reliable? Would I want to just consider using GS 3.2 stand-alone for live playing instead? Pros vs cons?
    I've found Giga very reliable in live use, though like I said I have the backup just in case. I also have some onboard sounds on my keyboard that I could go to if really pushed. I have to be honest and say, despite my good experiences, I wouldn't want to rely on just a single computer for an important gig. I have had a couple of time where my PC has done something screwy during setup that has made me nervous. In the event I've always been able to sort it, but you never know...

    But polyphony - yeah, you'll never max out a modern program and CPU with what you can do with ten fingers at once. I never have glitches, audio breakup or anything like that. Never raises a sweat.

  3. #3

    Re: GS GVI for live playing

    Quote Originally Posted by Ouch that hurts
    ...But polyphony - yeah, you'll never max out a modern program and CPU with what you can do with ten fingers at once...
    But with one forearm and a sustain pedal...

  4. #4
    Senior Member
    Join Date
    Jun 2005
    Location
    Martinez, California (Bay
    Posts
    163

    Re: GS GVI for live playing

    Thanks for the reply. I guess I'm not as up on the technology as I thought because I still am missing some concepts here.

    So suppose I have 16 midi channels set up on a copy of GVI and I set up 16 instruments. This would be the most simplistic configuration and I guess I would control what instrument is being played by controlling which of the 16 midi channels I am sending on.

    OK, now how about stacked instruments and the ability to play just one instrument in a stack. Is that accomplished by sending bank/program messages through midi? If not, how is that done (or is it)?

    OK, now on to having more than one instance of GVI running. How do you access the different instances of GVI with your midi controller? Do you need multiple physical midi ports?

    And what about having other virtual instruments? How do you access the different sounds on the different instruments in a different configuration for each song?

    And if I use GS 3.2 which supports multiple ports, what exactly do I need to use the multiple ports? Is is multiple physical midi ports into the PC? Is it a multiple port midi controller with a USB interface to the PC? If not, what is it?

    As you can see, I am struggling a little bit with understanding how midi is used to access the various sound components. I know this is not specifically Giga related but since I will be using GS 3.2 and/or GVI, I hope no one will mind me asking these questions in this forum.

    Thanks for all the continued help everyone!

    Rob

  5. #5

    Re: GS GVI for live playing

    Quote Originally Posted by goldberg96
    Thanks for the reply. I guess I'm not as up on the technology as I thought because I still am missing some concepts here.

    So suppose I have 16 midi channels set up on a copy of GVI and I set up 16 instruments. This would be the most simplistic configuration and I guess I would control what instrument is being played by controlling which of the 16 midi channels I am sending on.
    Correct. HOWEVER:

    You'll have to forgive me because I'm going purely on how GS2/3 works, and I'm not sure that the functionality is the same for GVI. Maybe someone who has it can confirm. But it IS possible to access way more than 16 sounds even with a simple configuration like this, by using program changes. If an instrument is loaded into RAM as part of the setup, but not assigned to a slot, then you send a program change message from your keyboard at the beginning of a song with the program number of that instrument, then Giga will insert it into the slot corresonding to the MIDI channel you sent the message on. Thus you could theoretically have thousands of instruments - put it this way, you'd run into RAM limitations a long time before you reached whatever the theoretical maximum is. The only thing is you couldn't play more than 16 at once, which you're unlikely to want to do ina live situation anyway.

    I don't actually do things this way because I have three MIDI ports on each of my PCs, two assigned to Giga and one to Chainer, and that allows me to have all the sounds I need up permanently.

    OK, now how about stacked instruments and the ability to play just one instrument in a stack. Is that accomplished by sending bank/program messages through midi? If not, how is that done (or is it)?
    In GS3, it can be done a few different ways. You can do it with program changes, as you say. Or you can assign a MIDI cc number to the stack. Say you have four instruments on a stack, you can make it so if you send values 0-32 of that cc number, on the MIDI channel of the stack, you hear the bottom instrument; values 33-64 select the next instrument up; values 65-96 the next and values 97-127 the top one. Or whatever you like.

    There's another way too I think but I can't remember what it is.

    OK, now on to having more than one instance of GVI running. How do you access the different instances of GVI with your midi controller? Do you need multiple physical midi ports?
    Yes. Multiple outs on your keyboard and multiple ins on your PC.

    And what about having other virtual instruments? How do you access the different sounds on the different instruments in a different configuration for each song?
    Same basic idea. You can assign an instrument to a particular MIDI channel of a particular port, in whatever your VST host app is. Then you can use MIDI program changes to select different sounds on the instruments, either presets or patches that you have made yourself. The exact details of how this is done vary from one piece of software to another, but the idea is basically the same.

    As I said previously, I believe that in FORTE you can do much more complicated stuff than that, for really dynamic setups, but I'm not sure what or how.

    And if I use GS 3.2 which supports multiple ports, what exactly do I need to use the multiple ports? Is is multiple physical midi ports into the PC? Is it a multiple port midi controller with a USB interface to the PC?
    Either. As long as the MIDI interface is taking multiple MIDI ins, Giga will see them and you can select between them to assign them to different ports in Giga's setup routine. It doesn't matter whether the MIDI cable physically enters the PC or whether there's a USB cable between them. The USB cable can carry the information of several ports, while allowing the PC to still recognise them as discrete ports.

  6. #6

    Re: GS GVI for live playing

    To Ouch That Hurts: Thanks. That was absolutely one of the most informative answers I've read regarding the use of the program. Questions I've wanted to ask since forever but afraid of sounding too ignorant. I have GVI installed but am still intrigued by the sheer power of GS3. I'm using M-Audio's Audio 1010 and MOTU's Midi Timepiece AV. Is there some way I can get more use besides going out to other samplers? The Delta 1010 has only 1 in and 1 out.
    -Rajay

  7. #7

    Re: GS GVI for live playing

    I've used GS3 (and GS2.5) in low-stress live environments (hauling my equipment to jam with friends). I've also avoided program changes, just for simplicity and lower risk. Just send data to a given channel and hear the sounds.

    Regarding stacks, you can split by velocity and/or note range. For velocity you can do things like play the organ softly and add electric piano on accents. Or you can split with bass in the left hand and piano on the right.

    Some people gripe about GS3's reliability, but I think they're off base. Once you get your system working well, it just chugs along. With GVI it should be a no brainer to get the software running for a streamlined live situation.

  8. #8

    Re: GS GVI for live playing

    Quote Originally Posted by JonFairhurst
    Some people gripe about GS3's reliability, but I think they're off base. Once you get your system working well, it just chugs along.
    Yes, I think that's absolutely true. People confuse two different things - setup issues and problems during actual use. There do seem to be people who have huge setup issues with Giga (though I've had no serious ones at all) and sometimes these seem to be completely unfixable. But for those of us who don't have them, or those that do manage to solve them, it then just "chugs along" as you say.

    And of course on a system designed purely for live use you can keep things very streamlined so there isn't lots of other stuff to compete with it.

  9. #9
    Senior Member
    Join Date
    Jun 2005
    Location
    Martinez, California (Bay
    Posts
    163

    Re: GS GVI for live playing

    Ouch That Hurts: Thank you so much for being willing to take the time to give me such valuable information. People like you are who make forums like this so incredible.

    I have one more question. I know because GS runs in kernal mode the amount of RAM that can be utilized is limited further than it would be without kernal mode. So with GVI, which does not run in kernal mode, what is the limiting factor for loading samples into RAM. Can it use 4Gb, 8Gb? I have room for more RAM in my machine (currently have 3GB) but I don't know if it can be used effectively.

    Also, if I run multiple VST's under Chainer as I believe you said you do, do they each get a different address space and as such get a separate memory allocation or do they all run under one address space for Chainer and share one address spaces memory?

    Thanks once again to all ............ Rob

  10. #10

    Re: GS GVI for live playing

    Quote Originally Posted by goldberg96
    I know because GS runs in kernal mode the amount of RAM that can be utilized is limited further than it would be without kernal mode. So with GVI, which does not run in kernal mode, what is the limiting factor for loading samples into RAM. Can it use 4Gb, 8Gb? I have room for more RAM in my machine (currently have 3GB) but I don't know if it can be used effectively.
    There is normally a limit of 2GB to the total amount of application memory that can be addressed in Windows XP. However it's possible to make a little change during setup (the "3GB switch") that increases this to 3GB and reduces kernel memory to 1GB.

    As far as I'm aware, 2GB is also the limit for most individual applications. However there have been reports of some people using more - up to say 2.7 or 2.8 GB - in some things like the Vienna VIs. This really just depends on the individual app and how it's coded, so you'd need someone who has tried maxing out GVI to answer this with any more precision.

    The best I could say is - probably about 2GB, maybe a bit more. Certainly not 4 or 8 or anything.

    Also, if I run multiple VST's under Chainer as I believe you said you do, do they each get a different address space and as such get a separate memory allocation or do they all run under one address space for Chainer and share one address spaces memory?
    That's an interesting question and I don't know the answer. My intuition tells me that it's Chainer that gets the memory space, since the VSTIs in this situation aren't actually running as programs, they're running as plugins to a host program (ie, you don't open the .exe file of the standalone instrument, you load the .dll into Chainer).

    Bear in mind anyway that for anything that isn't sample-based, memory is not nearly as much of an issue. Soft SYNTHS, as opposed to soft samplers, work by making lots of calculations in real time, not by storing and choosing from a set of short recordings that take up RAM. Obviously they use up some RAM like any other program, but nothing anywhere near Giga, Kontakt etc. The main thing they use is CPU (though many of them are very efficient now, and you won't be using many at once for live playing).

Go Back to forum

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •