For some time now I have been experimenting with XTC-mode and enjoying it much. Once you set it up correctly it is really quite stream-lined.
The point of this thread is to capture some broader interest (hence posting in the general forum) for issues raised in http://www.planetz.com/forums/viewtopic ... 2&start=20
With the rise of modular ASIO-based VST-hosts like energyXT and Tracktion (rack-filters), I have become very psyched about the capabilities that XTC-mode could provide with some (in fact very little) additional work.
With hosts like this the SFP environment could be a thing of the past. We could have the possibilities that we have wished for a long time and is why there ever was interest in TripleDAT or any audio sequencer for Scope in the first place.
The big argument here is FULL and infinite automation of all effect parameters.
Basically what we need is a DEV/VST-wrapper that can dynamically recognize and handle plugins that have more than two inputs and more than 16 outputs. The ones we have now only handle synths, mono and stereo effects with up to 16 outputs. These are located in your App/Bin directory and are respectively called VstSynth.dll, VstMono.dll and VstStereo.dll.
With the source-code for these it would be able for the community to do something about this and compile a proper VST-plug that will handle i.e. the STM 2448.
This will probably never happen as these most likely contain sensitive code for communicating with the Scope hardware. We can draw lines to the topic of Scope for Linux and this. Same "problem", really.
So, our job here would be to influence CWA into seeing the possibilities and importance of why they need to further develope XTC-mode.
I encourage you all to take part in a discussion on this matter. FULL AUTOMATION people, come on!
<font size=-1>[ This Message was edited by: voidar on 2005-05-21 14:11 ]</font>
Thoughs on XTC mode
Personally I tend to use Energy XT and Tracktion within SFP rather than use XTC mode. I think their modularity compliments SFP's nicely rather than seeing them as a replacement for it.
Mind you as a long time Logic PC user I can tell you that Modular hosts go back a long way - the Logic environment is one example - programmes like Audiomulch have also been around a fair length of time.
Mind you as a long time Logic PC user I can tell you that Modular hosts go back a long way - the Logic environment is one example - programmes like Audiomulch have also been around a fair length of time.
Personally I have begun to like the idea of being focused on one window. I also like being able to load plugins and treat them like regular vst plugs in that same window.
I will not debate the modular legacy, but I refer to hosts that can load plugins like objects and integrate them seamlessly and let you govern every input and output the way you choose to, very much like SFP mode.
The difference being that SFP mode has very little in regard to automation, which is the key issue here.
Also, it is the aspect of continuity which XTC provides. Dealing with a host window and then the SFP window, having to patch up stuff through ASIO drivers - meaning full-recall can be troublesome - is a very cumbersome proceedure, and it really does not add to the fun of making music. As a producer and engineer it is ok, but it becomes cumbersome to musicians.
I will not debate the modular legacy, but I refer to hosts that can load plugins like objects and integrate them seamlessly and let you govern every input and output the way you choose to, very much like SFP mode.
The difference being that SFP mode has very little in regard to automation, which is the key issue here.
Also, it is the aspect of continuity which XTC provides. Dealing with a host window and then the SFP window, having to patch up stuff through ASIO drivers - meaning full-recall can be troublesome - is a very cumbersome proceedure, and it really does not add to the fun of making music. As a producer and engineer it is ok, but it becomes cumbersome to musicians.
Automating stuff in SFP thru MIDI works fine here. I've been doing it in Logic since forever, with a Logic environement knob for each parameter I want to control in SFP. Then I never have to switch screens. You can draw and automate MIDI CC values just like standard VST parameter values. There is the slight MIDI-thru-software-drivers latency issue, but I find it's alot less annoying than stacking up ASIO latencies and relying on PDC when using XTC mode.
There is also a remote control in SFP that will let you play/stop/fwd/rwd your sequencer if you are doing something that is mostly SFP-centric.
Also, loading STM2448 as a VST plugin would be pretty terribly useless. You'd need to have sequencer-side support to change the mixing engine completely, otherwise you end up with a whole mixing engine sitting in a single VST slot, which lead you to seriously more bumfangled signal chains and not much advantages (besides the automation stuff.) You might as well mix straight in SFP then, since it'll let you do any kind of weird routing that you might imagine/want anyway.
Alternatively, you could just get a dual-head video card and a pair of monitors, or UAD/Powercore cards.
There is also a remote control in SFP that will let you play/stop/fwd/rwd your sequencer if you are doing something that is mostly SFP-centric.
Also, loading STM2448 as a VST plugin would be pretty terribly useless. You'd need to have sequencer-side support to change the mixing engine completely, otherwise you end up with a whole mixing engine sitting in a single VST slot, which lead you to seriously more bumfangled signal chains and not much advantages (besides the automation stuff.) You might as well mix straight in SFP then, since it'll let you do any kind of weird routing that you might imagine/want anyway.
Alternatively, you could just get a dual-head video card and a pair of monitors, or UAD/Powercore cards.
Yes, automation by midi "works fine", but it is tedious, and 128 controllers for a console like the STM2448 is really not enough.On 2005-05-31 10:40, symbiote wrote:
Automating stuff in SFP thru MIDI works fine here. I've been doing it in Logic since forever, with a Logic environement knob for each parameter I want to control in SFP. Then I never have to switch screens. You can draw and automate MIDI CC values just like standard VST parameter values. There is the slight MIDI-thru-software-drivers latency issue, but I find it's alot less annoying than stacking up ASIO latencies and relying on PDC when using XTC mode.
There is also a remote control in SFP that will let you play/stop/fwd/rwd your sequencer if you are doing something that is mostly SFP-centric.
Also, loading STM2448 as a VST plugin would be pretty terribly useless. You'd need to have sequencer-side support to change the mixing engine completely, otherwise you end up with a whole mixing engine sitting in a single VST slot, which lead you to seriously more bumfangled signal chains and not much advantages (besides the automation stuff.) You might as well mix straight in SFP then, since it'll let you do any kind of weird routing that you might imagine/want anyway.
Alternatively, you could just get a dual-head video card and a pair of monitors, or UAD/Powercore cards.
I do come from a Logic background, but droped it when I discovered mixing with the STM2448.
First, delay compensation usually works, though I know it is cluttered in Logic (I am on PC, never tried anything above version 5.2.2).
Second, Logic is not as modular as you might think compared to eXT and Tracktion. And loading a single STM2448 would effectively mean that you are working in hardware domain. You can load whichever hardware plug you wish (perhaps not synths) in the mixer and only have to deal with one layer of vst-shell, which every track will be passing through anyway.
Third, we would only have to deal with a single ASIO-driver layer, one for recording and one for monitoring. It would not be necessary to route back into ASIO for mixdowns etc. In effect we would only have to deal with a single float/fixed/float conversion for this particular scenario, given that float is the hosts format engine.
I do not see how this is terribly useless,as you put it. It would rather mean increased possibilites. If you don't see that then you are ignorant to the modular approach of i.e. eXT and Tracktion that I speak of.
I am perfectly aware of what this hardware can do and not do at this stage, I am only trying to broaden your minds towards what would be possible with little effort in developement.
Well, as far as I know, Creamware is going the other way, i.e. dropped XTC support and isn't working on it anymore, so good luck convincing them to do anything, especially if it means opening up source code.
I was just saying you can get pretty close already with SFP, so there's no real use to investing lots of development time/effort to just get a tighter integration with sequencing software. Also, as far as I know, you'd need some serious developement on the sequencer side/VST side, I'm not even sure the VST standard supports lots of inputs. So you'd probably need to somehow extend the standard and somehow get the sequencer developers to support it (good luck.) And it's not made to replace whole mixing engines I don't think.
Also you are wrong on a few points, you'd still have as many ASIO layers as if using SFP, as the VST wrappers for XTC-mode definitely go thru ASIO. You'd still get a sequencer -> ASIO -> STM448-loaded-as-VST -> ASIO -> sequencer signal chain, thus having the same kind of latency and same number of float/int integers, it would just be hidden from view. And you could not load normal VST effects into the STM2448 either. As far as I know, no one else made a VST mixer anyway.
Not that what you suggest isn't interesting conceptually, just not worth the effort at this point.
I was just saying you can get pretty close already with SFP, so there's no real use to investing lots of development time/effort to just get a tighter integration with sequencing software. Also, as far as I know, you'd need some serious developement on the sequencer side/VST side, I'm not even sure the VST standard supports lots of inputs. So you'd probably need to somehow extend the standard and somehow get the sequencer developers to support it (good luck.) And it's not made to replace whole mixing engines I don't think.
Also you are wrong on a few points, you'd still have as many ASIO layers as if using SFP, as the VST wrappers for XTC-mode definitely go thru ASIO. You'd still get a sequencer -> ASIO -> STM448-loaded-as-VST -> ASIO -> sequencer signal chain, thus having the same kind of latency and same number of float/int integers, it would just be hidden from view. And you could not load normal VST effects into the STM2448 either. As far as I know, no one else made a VST mixer anyway.
Not that what you suggest isn't interesting conceptually, just not worth the effort at this point.
Developement of XTC mode is definitely not on the program at the moment, but you never know.On 2005-06-01 07:59, symbiote wrote:
Well, as far as I know, Creamware is going the other way, i.e. dropped XTC support and isn't working on it anymore, so good luck convincing them to do anything, especially if it means opening up source code.
I was just saying you can get pretty close already with SFP, so there's no real use to investing lots of development time/effort to just get a tighter integration with sequencing software. Also, as far as I know, you'd need some serious developement on the sequencer side/VST side, I'm not even sure the VST standard supports lots of inputs. So you'd probably need to somehow extend the standard and somehow get the sequencer developers to support it (good luck.) And it's not made to replace whole mixing engines I don't think.
Also you are wrong on a few points, you'd still have as many ASIO layers as if using SFP, as the VST wrappers for XTC-mode definitely go thru ASIO. You'd still get a sequencer -> ASIO -> STM448-loaded-as-VST -> ASIO -> sequencer signal chain, thus having the same kind of latency and same number of float/int integers, it would just be hidden from view. And you could not load normal VST effects into the STM2448 either. As far as I know, no one else made a VST mixer anyway.
Not that what you suggest isn't interesting conceptually, just not worth the effort at this point.
VST2.0 supporst multiple inputs and outputs, although multiple outputs was added first as a though for multiple output instruments (drumsamplers etc.). For sequencer developers to stay behind on the VST standard is not really an issue.
I am not sure how XTC does the exact wrapping, but I would not go as far as claiming any relations to ASIO. Perhaps technically in some way. But only CWA knows.
And no, you would not be able to load VST-effects/instruments in the CW environment (i.e. into STM2448). But XTC mode would be the only way to actually use software and hardware seamlessly.
Well, actually someone did create a VST-mixer.. http://perso.wanadoo.fr/lesitenm/04page ... lugins.htm
Well, as far as XTC goes, if you load XTCProject.pro and look in it, you can definitely see stuff going thru ASIO. UAD and Powercore work the same way too, which is why you kind of need PDC to do anything useful with them.
Anyway, I tried those VST mixers quickly in Samplitude, they look nice but I haven't managed to get sound into them yet =P. I'll try again tonight when I get home. I'll also take a look at energyXT as it looks pretty fun. Thanks for those links/pointers
Anyway, I tried those VST mixers quickly in Samplitude, they look nice but I haven't managed to get sound into them yet =P. I'll try again tonight when I get home. I'll also take a look at energyXT as it looks pretty fun. Thanks for those links/pointers
Yes, inputs and outputs go through ASIO, but when you load a CW plug in XTC-mode the VST-wrapper makes the software-hardware connections automatically to whatever device in your DevicesXTC directory that shares the filename of the plug. You don't have to setup ASIO channels for each plug you use, that would be like using SFP-modeOn 2005-06-01 11:44, symbiote wrote:
Well, as far as XTC goes, if you load XTCProject.pro and look in it, you can definitely see stuff going thru ASIO. UAD and Powercore work the same way too, which is why you kind of need PDC to do anything useful with them.
Anyway, I tried those VST mixers quickly in Samplitude, they look nice but I haven't managed to get sound into them yet =P. I'll try again tonight when I get home. I'll also take a look at energyXT as it looks pretty fun. Thanks for those links/pointers
.The interface in those VST-wrappers is a secret, but CWA have supplied one for mono, stereo and synths.
How to link a VST in/out to a device in/out is really what I would like to know because then I could basically code a multiple input/output VSt-wrapper myself using the VST SDK.
Anyway, hope those links come to some use to you.
