A Template question - theorical

I have been constructing an orchestral template for EWQL Orchestra - platinum. This template has all bells and whistles, things like expression maps and mixer views., but I have come across a problem with EWQL
My intention was to load all the master keyswitch patches, but then to remove all the samples using the two tick boxes in Play. My reasoning was that this would leave the maximum amount of RAM available and no patch would be loaded until it was required.
But EWQLSO does not like this, there seems to be some kind of technical issue which makes the reloading of patches very glitchy.

So, here is the question on my mind.

My machine has 12 gig of Ram. Loading all of EWQLSO uses 6 gig in total. If I take out the patches I regain 2 gig.

But why?

Why not leave all the patches in place - after all even with 6 gig used this means there is six is still spare.

I am wondering if thre is any downside to this?

Maybe having these samples hanging around in RAM but not being used has no consequence on performance?

Maybe somehow behind the scenes it would have an impact?

Could someone more technical than me advise?

It seems to be an interesting question for those with new 64 bit machines


Do you not regain the extra memory if, after removing the samples, you save the Project, quit Cubase, then relaunch it (or even, restart your computer)?

Hi Vic, Yes you would regain the memory, but this is not what the question is, I am asking if you fill up memory then does this have an impact on performance in the case where you have plenty left - i.e. the new 64 bit world

I only have 8GB of RAM, and my system is kinda old now, so take this with a grain of salt. What I’ve found in my testings is that Windows 7 x64 is not as flaky when hitting the RAM limits as was XP. For example, I’ve been able to load a little over 7GB of samples in RAM and the system still runs fine (no crashing, no memory warnings). I think the max I was able to do with XP was ~3.4GB of RAM out of 4GB (and that was with Virtual Memory enabled, of course) before it crashed.

Now, to put this into perspective, XP used to take anywhere from 200 - 300MB of RAM for itself where Windows Vista/7 consumes ~1.2GB. As you can see, I was able to “steal” RAM from the OS without it complaining on me, whereas on XP I couldn’t. My point is that if you have the RAM, then use it. You’ve got a total of 12GB of RAM and your template only uses half, so this should not affect your performance one bit. You still got plenty of headroom. The only inconvenience would be the loading times, but then again this would be the case on an almost finished project (unless you freeze).


So, you can fill up your RAM with samples, and if you leave them alone, i.e. dont use them, then Cubase has absolutely no issues and runs in the rest of the RAM no problem?
There are no hidden processes to consider?

Let’s put it this way, my system can handle 6GB of RAM of EWQLSO all day long, so I don’t see why yours couldn’t. It gets a bit finiky with HS though, but I think that’s because my HDD can’t handle it well.

The truth is that I use the Gold version of these libraries, but I also sometimes use 3 mic positions without issues on projects that aren’t too big. Like I said, my system is old and I definitely need a new one that can keep up with my projects. Best bet is for you to experiment and see how far you can take it with yours. That’s the only way you’ll know how much you can do with your system. Yours should handle way more than mine, so don’t be afraid :wink:

Yeap, I just confirmed that my HDD is not keeping up with Hollywood Strings Gold. My current project has 9 PLAY live instances and 4 frozen ones (3 HS and 1 HB), 4 audio tracks, and HD720p H.264 video, 13 audio FX, several tempo changes and Expression Maps. Total CPU usage is ~ 26 - 30%, and RAM usage is 6.6GB. Not pushing my system at all yet I still get interuptions on playback unless I freeze pretty much all my strings (I can only leave one string instrument live at once to get clean playback). On the other hand, I am using the powerful system patches cause they just sound so good! So that’s the price I pay :stuck_out_tongue:

I need an SSD :frowning:

If you want I will share my cpr with you - as Vic can tell you working up a template is a lot of work and each one is somewhat individual to needs and wishes of the developer, but so far this one has expression maps and mixer views, stil a work in progrss though

BTW why are the stereo double controls on as default in EWQLSO play?
and why does the left light come back on if deslected?

Thank you, Zero! I appreciate it but my workflow requires a different type of template. I transfered one I’ve developed a few years ago in Sonar, thought it is slightly different in Cubase. I had to make adjustments due to the way Cubase works, but is pretty close to what I had before. My template is made up of ~30 instances of PLAY, and includes an entire orchestra + choir. Each instance represents an instrument and they are loaded as Instrument Tracks with the articulations being triggered through Expression Maps. This enables me to keep track count very low, as opposed to the crazy amounts of tracks that is required through the multi-out method that’s commonly used. There are a couple other things I don’t like about the multi-out method (i.e. not being able to rename the VSTi’s loaded in the VST Rack and the VST tracks that are created along with the MIDI tracks, which defeats the purpose of keeping track count to a minimum).

I don’t use keyswitches either. Instead I create my own keyswitches that are particular to a specific project and which are stored inside each project’s folder (they reside inside an Expression Map folder). So far I’ve only needed to make 3 or 4 Expression Maps per project, so it’s not bad at all. I try to keep the articulations consistent with their corresponding MIDI Channel, which is the way I trigger them, and this keeps things consistent across instruments. That’s pretty much the gist of it. Very simple, IMO.

I wouldn’t mind looking at yours though, just to see if there’s anything I’d like to steal from you :wink:

It’s always been like this as far as I can remember. However, the left light thing seems odd. Some patches don’t have this problem, so I wonder why this is. Hmmm…

This my game plan - 90% completed at present…

One or two Play instances per section for the woodwinds brass and strings. This depends on how many master outputs are needed (Play having a maximum of 9 stereo pairs per instance).

Load one master keyswitch per Play instrument ( e.g. solo violin and the violin sections being ‘different’ ‘instruments’ in this respect)

One project track ‘per play instrument’ with an expression map to change the keyswitches

One folder track per tranditional orchestral group, plus a folder for ‘chromatic percussion’ inc harpichord, harp, xylophone, marimba etc) I.e. any keyboard type instrument.

Somewhat different approach for the harp which does not have a keyswitch instrument - this gets one project track and an expression map that changes the MIDI channel and an instance of Play with each articulation loaded with different MIDI channels - but the same stereo pair as an output for all articulations. Similar treatment for percussion groups (woods, metals, etc) at present I have the following groups of sounds each with an individual stereo out - timpani; bass drums; snares and field; woods; metals.No expression maps here yet, they all have their own project track. Not usre if this is the best way to go as often several percussion is on one staff.

Tonight I am considering whether to output the groups of instruments to the same stereo outputs. For example is there any point having ‘solo flute’ and ‘three flutes’ on different stereo outs? Probably not, but there may be a point in having the first (solo) violin on a different stereo pair, to the violin sections. Still thinking this over…
There is also the issue of the violins traditionally having two sections, and the matter of divisi - how is this to be reflected in the project window? Most of the time one staff should be good enough (and two violinists sharing the same stand read the same staff high and low/left and right), but if one divisi part articulates different from the other this might not be possible …hmm…

In the mixer:
create a mixer view for each section (woods, brass etc) then output each section to a group channel, include the relevent group channel in the mixer view for that section (woodwind group channel in woodwind mixer view. This gives the ability to find each section easily and adjust either individual channels for an individual play instrument, or, use the gruop channel as a local master for this section.
finally a mixer view for all the group channels for over mixing before going to the master out.

The goals are:
To emulate a classical score arrangement as far as possible
To use the minimum number of tracks (with expression maps controlling the articulations)
To keep the mixer user friendly but deliver appropriate and intuitive control
To merge the outputs of Play instruments that represent the same instrument (e.g. 3 clarinets and solo clarinet) into the same stereo outs, in order to a] simplfy the mixer, and b] allow for cohesive mixing using one fader.
To deliver maximum flexibility of articulations for any given track
To deliver a ‘master’ template that can easily be cut down into smaller sections and saves as e.g. ‘EWQL Strings’

Should have this one finished by the end of the week


These were pretty much my goals too, when I created my template. But the main goal, above all, was simplicity. I have uploaded my template if you want to check it out. Let me know if you have any question :slight_smile:


Zero, I would very interested in getting hold of your template :smiley: I use EWQLSO a lot and haven’t had the time of late to set things up properly due to building a new studio.

To be honest I haven’t even got round to working with expression maps yet!!!


Ill post it when its ready