how cpu heavy is VM?

[tamarack]

I suppose the answer is that it depends on what you build? But I'd sure love some advice before I buy.

I'm just about to pull the trigger on buying the core, but I'm wondering if I'll actually be able to make use of it? Are simpler generative patches out of reach? The sampler? Or something like the Drift modules?

I've got Intel(R) Core(TM) i7-2620M CPU @ 2.70GHz, 2701 Mhz, 2 Core(s), 4 Logical Processor(s) and 8GB of ram.

I just used the nucleus bundle to make a really fun little sequencer patch with one oscillator and some lfo modulation and i'm already bouncing around 10 percent CPU.

[huggermugger]

I can't say I've ever hit the ceiling, although I don't make those huge patches that you sometimes see in this group. But here's a moderately complex generative patch.

1 OSC, 1 waveshaper, 1 three-band filter, 1 stereo mixer, 1 stereo delay, 9 LFOs, 2 remote control units, 3 S&H , 1 audio suboctaver, 1 VST plugin (Things Tilt), 1 stereo chorus

The waveshaper is being heavily modulated, as is the 3-band filter. The delay fbk is also being modulated.

I'm running a 6-core iMac i7 3.2GHz, 16GB. You can see the CPU usage in the Activity Monitor display, upper right. Not really heavy - one core about 40%, two cores about 20%, one core about 10%, the last two barely registering. This patch would probably push pretty hard on your system though.

[huggermugger]

BTW, I'm not using Hyperthreading - my six virtual cores are inactive. I think you are, though. If so, try turning it off. It may improve your performance.

[huggermugger]

BTW, I think you can demo the Core Bundle (and the Drift Bundle too). That would be a great way to figure out if you want to commit.

[tamarack]

Thank you for the detailed reply! That's helpful and it was interesting to see your patch... I didn't realize I could demo, that might be the best way to see what's achievable. I'm still hopeful I can do something but i think i'm a bit under-gunned. I'm a bit reluctant to get into the bios and mess with hyperthreading, i'm a reluctant computer guy at best. I should go bloatware hunting though.
I'm not that worried about performing live so I could also perhaps break pieces down into several independent patches and record them as audio tracks individually, if i can figure out how to sync tracks to the daw tempo, and i think i can... they just wouldn't interact with each other on their own accord...

[huggermugger]

As for hyperthreading, just turn it off in VM, as you can see in the image I posted. No need to mess around with the BIOS

[huggermugger]

Synching sequencer playback in VM to your DAW is easy. Here's what it looks like in Reason, but it looks the same in any DAW. I'm using green cables for Start and Reset, red cables for Stop, and yellow cables for Clock data. Once you've hooked this up, the Play and Stop functions in your DAW should start and stop VM's sequencers too.

[tamarack]

Thank you! ( i will test with the hyperthreading, sorry i thought that screenshot was from your apple system, didn't realize you could access those settings right in vm )

I'm running the core demo and having so much fun, without hitting CPU ceilings just yet. I'm using field recordings through the sampler to a sample and hold and a quantizer to make Stuff. Super fun. OK, fine, just take my money then C audio. Today's project is trying to figure out polyphony. we'll see how my system handles that.

Now I need to figure out a way to generate a more interesting ext trigger for the sample and hold, i'm just using it's internal clock and it's a bit static. Oh and I might need to buy the vintage resonator too. Damn.

Anyway, thanks again for the tips.

[huggermugger]

Here's how you can make a random trigger...
Patch an LFO (LFO1 set to random) into the FREQ CV of another LFO (LFO2 set to non-random).
Patch LFO2 into the EXT TRIG of the S&H. Set the SHH Clock to EXT.
Set both LFO rates to approx the same, around the speed of S&H you want. Then adjust the FREQ CV amount on LFO 2. The triggers will become random.

[huggermugger]

And here's a variation on the above, which will create trigger 'patterns'...
Set LFO1 to a non-random shape (SIN is a good starting point to hear the effect).
Set LFO1 FREQUENCY to a much slower value than LFO2.The triggers will speed up and slow down predictably in the shape of the LFO1 waveform.