I\'ve been getting along ok with my single-computer, SBLive system, but I\'m outgrowing it. I\'m about to get another computer to dedicate to GSt, but I\'m really confused about all the sound card options. I\'m thinking that I\'ll end up with two computers running GSt (one dedicated, one not). All midi, no microphones involved.
Trouble is, I haven\'t set this kind of thing up before - I\'ve never used any sound cards other than Creative, so I don\'t know what I\'ll need in the way of midi interfaces and sound card I/O capabilities, and I don\'t really understand how the midi and GSt output will get routed around. Is there some kind soul out there who could explain how this stuff works, or point me to a place where I can read a comprehensible explanation?
If I understand it right your setup will be 2 computers connected together, one running GSt and Sequencer and one GSt. You want to keep your SBLive in one of them and get a new soundcard. Here are my suggestions:
1) You will need to establish a MIDI connection between the two computers in order to play your GSt from sequencer. For that you need a MIDI Interface on each of the computers.
Every MIDI interface features at least one (but can have upto 8 of them) set of MIDI connectors: MIDI IN, MIDI OUT (and sometimes also MIDI THRU). These are MIDI ports. With one set of these ports you can connect two machines in both directions, for instance from a computer to a synthesiser and back from the synth to the computer. For each direction you need one MIDI cable.
With one connection in one direction (thus one cable) you can transmit upto 16 MIDI Channels. To express this in orchestral terms you can send instructions for upto 16 orchestra players at the same time (one string section being usually seen as one entity). If the players also wanted to send some information back to the conductor you would need another cable for the data in other direction. This however is not always necessary and in your case only a connection from MIDI OUT port of your Sequencer-Computer into MIDI IN port of the GSt-only computer will be needed.
Now, if you want to play 17 instruments at the same time you need two sets of the MIDI ports. The GigaStudio 96 can play 32 instruments at the same time (one says it is 32× multitimbral). Therefore, in this case you need on each computer a MIDI interface with at least 2 sets of the MIDI ports. The GigaStudio 160, I believe, can play 64 instruments at he same time and so you need 4 MIDI ports in order to direct all of the musicians.
For your MIDI interface you basically have three options:
a. You use the MIDI interface of your SB Live ( I guess you already use the MIDI IN port to connect your keyboard) and you buy a soundcard which also features MIDI port(s). The problem is that SB Live has only one set of MIDI ports and so you will only be able to send 16 instruments at the same time to your GSt-only computer.
b. You buy a dedicated MIDI interface boxes, probably ones which connect into USB port of your computers (fairly simple installation). These are offered with 2 to 8 sets of the ports.
c. The third option is a combination of the two above: you buy a soundcard with a MIDI interface and a dedicated box for the computer where your SB Live will be.
One thing to remember is that in all three cases you will only be able to play instruments on your GSt-only computer with the other computer switched-on and sequencer running. Otherwise you would need a hardware MIDI switch to route your keyboard (or you could also replug the MIDI cable every time it is necessary).
2) You will need to mix audio outputs of both of your soundcards. You can do this in two ways:
a) You buy a little hardware mixer, connect outputs of your soundcards into it and the mixer\'s output you can either record onto a DAT (or whatever) or you can route it to the input of your better sound card and record the signal onto a hard disc.
b) You could route the output of your GSt-only computer into the input of your other soundcard and mix the signals in the latter one. This means that your new sound card should have more inputs than one stereo pair and should have full duplex and multiclient drivers.
As regards the sound card itself, apart what I said above it should probably have GSIF drivers so that you would be able to use more separate outputs in GigaStudio and very low latency.
Well, I tried to express myself as clearly as I could (English is not my mother tongue, you know) and I hope it helps you at least a bit. When you have questions, don\'t hesitate to ask - I will try to answer what I can.
[This message has been edited by ondrej (edited 06-15-2001).]
I think this is starting to penetrate my thick skull. So here\'s what I\'m thinking so far:
Initially, I\'ll just use my current computer as a sequencer that drives GSt on another computer. For that, I\'d need to add a 4-out midi interface on my current computer (to handle the 64 midi channels). The new computer will need either a 4-in midi interface, or a WaMi Rack (I think it\'s the only GSIF sound card with 4 midi ins?). I could send the output from the WaMi Rack (or whatever sound card I pick) straight to my monitors. My current SBLive would only be used to record midi in from my keyboard - maybe also to do reverb, which works better than any software reverb I\'ve tried (which is really only CoolEdit Pro\'s reverb, the Cakewalk plugin, and a program called \"Pristine Sounds\").
Second step would be to upgrade my current computer so it can run another 160 voices (I hope!) on its own GSt. I guess I could continue using the SBLive on computer #1 (current computer, upgraded), and send its audio output to computer #2\'s sound card. That would be the least expensive option, since I already have the SBLive. Eventually, if I want to improve latency etc., I\'d upgrade to a GSIF card on computer #1. Then, if I wanted, I could add outboard reverb processors etc. until it\'s time for debtor\'s prison. (!)
Does this seem like a reasonable path to go from my current all-in-one setup, to a dedicated GSt, to a combo dedicated/non-dedicated GSt?
\"WaMi Rack (I think it\'s the only GSIF sound card with 4 midi ins?)\"
I don´t know if it is the only GSIF card with 4 MIDI in/out but I can tell you that with the new E-WDM driver (v1.1) GStud160 work very fast (1.5 ms/3ms latency)and appropiate for last version from Nemesys. The very new 1.2 E-WDM driver comes with the new feature called GigaWire: internal routing og the GigaStudio output to ASIO input. This great thing allow to hear your GigaStsudio sequences in your Sequencing program (tracks) using the ASIO plugins also! I have to test it but this feature is disigned for working in one machine. You can read a good explanation in www.egosys.net
Hope this help,
Thanks Richard - I had noticed that about the E-WDM drivers, and it seems interesting. I\'m not sure the ASIO thing will be of much use to a SONAR user, but maybe the E-WDM thing will make it easier to use the SONAR soft synths. As it is, I have to boot to a different setup to get SONAR-with-WDM, and use a somewhat unstable beta WDM driver for my SBLive! It would be nice to improve on that.
Your plan makes generally a good sense to me. I made only several remarks below. The specs of WamiRack look just right for what you intend to do but I have no experience with the card. Also you should get some other opinionn about your plan since I might have overlooked something or there might be more elegant solutions. Well, here are my remarks:
Reverb on the SB Live
If you want to use a reverb on your SB Live you will need to route a signal from your new sound card to it. Because the value (cheap) version of SBL does not feature a digital input you would have to use analogue IN. This would lead to a loss of signal quality and make the expense of your new C#2 high quality sound card like a bit of waste. If you owe the SB Live with the front panel you could use its digital input with result being much better. I suppose the inevitable sample rate conversion (SB Live has a fixed sample rate of 48 kHz) will not do too much harm.
Audio in Sequencer
You never mentioned whether you work with audio in your sequencer (or in some dedicated program). If so, you will need to mix sound from both sound cards because your audio wavs would come from different computer than your GSt music. As I said you can either get a small hardware mixer or route on of the signals to the second sound card and mix it on it. There might be an issue with latence accumulating.
GSt on both computers
For your future setup, apart from having to mix the outputs of both cards, there might be a small latency problem because since this is likely to be different for each card. Thus the instruments wouldn\'t come exactly together. But this difference will probably be so small that it won\'t be audible. Otherwise setting small delay on some MIDI-tracks could help.
Also, obviously, your SB Live will not match the sound quality of your other card but I think it\'s still far better option than a life in debtor\'s prison.
If you capture to disc and want to play the files in your sequencer you will also need to solve a problem with transporting files between the two computers (ZIP disc, CD-ROM, Network?). Although I think the capture should work in this setup I wonder if there can\'t be a catch. No experience here...
I would advise that you take a great care of the noise of the new computer because this adds to the noise level of your current machine. You can either buy a system from the music-computer specialist (expensive) or you should choose components as silent as possible. The most problematic things are a processor fan, hard disc(s), a power supply fan and a graphic card (no fan = no noise).
I use WamiRack 24 for 4 MIDI ins/outs (64 channels, 4 Audio ins, 8 Audio out, SMPTE in/out, SPDIF coaxial & optical and Word clock all in the same card.
It´s sounds pretty good for me, clear.
I can play the standalone B4 with only 1,5 ms of latency!
Since I upgraded my RAM to 384MB I have been able to launch Logic Audio from Gstd.160 and record a Gstd. midi track inside logic and after this routing his stereo audio (out 5,6) to input 1,2 and record it inside Logic. This is in one computer. This has be done with 1.1 E-WDM driver. After this we have the new GigaWire plus in 1.2 (buggy) and 1.3 E-WDM driver but I am not more able to have the above descripted situation??
Now I am expecting a response from Ego-Sys, We´ll see. GigaWire is functioning well with Cubase but I don´t know yet how it can be working for Logic Audio but it employes ASIO.
Try, like Ondrej says, to have more opinions of Wami owners with different audio programs maybe...
Wow, Ondrej, thanks again for the help! Regarding SBLive for reverb, mine has spdif in, so at least I\'ll get digital in. You\'re right, the 48khz sampling rate might be a problem; I\'ll just have to try it and see. Anyway, there\'s something just \"wrong\" about taking all that high-quality audio and pumping it through an SBLive, of all things, at the end! Well, there are a number of other options for reverb (ones that don\'t lead to prison), so I\'m not too worried about that.
Now comes the toughest technical hurdle of all - convincing the wife that it\'s a good idea to spend all this money.... Anybody have advice on that one?
Save money and effective latency by using a \"null modem serial cable\" ($15 at Tandys) between PCs and the Yamaha general CBX midi driver (free) available from yamaha site. This gives upto 4 ports of 16 channels each! with a lot less hardware/buffers inbetween, it is very fast. The equivalent of using two 2x2 i/o midi interfaces, not to mention their power supplies, MIDI cables etc.
Install yamaha driver in both computers, connect cable and you\'re away.
In your case between 2 PCs this IS possible. In my case from mac to PC, not possible (darn it).