Build 1701ok / 1702 cpu hog

Has anyone noticed that the new build 1702 seems to be more of a cpu hog than 1701?

This morning I ran a project with 4 tracks. 3 vsts and while it was running about 80-85% cpu load. it did play fine and there was no problem.

This evening I downloaded the new 1702 build and the same project now locks at 100% and stutters, hacks, coughs and wheezes when trying to play it.

Anyone else notice this?


...4 tracks. 3 vsts and while it was running about 80-85% cpu load...

That's not normal unless you have an underpowered system or something else is wrong. What's your specs?

I agree neither build should be loading this bad yet. (I wouldn’t think.)

System is
1.8Gig P4 512 Ram 40Gb Free HDD space. XP Home (sp2)
Running 4 - ~12 minute tracks (44.1,16 bit) Pristine Space Impulse reverb(Master w 6sec impulse file),Space Effect vst(Aux1) These 2 plugins are taxing for any system I know but not that bad I wouldn’t think.

This will play but NT reports 90-95% cpu load (NT 4.01 build 1702)

Adding another vst kills it.

Was about 75-85% in build 1701.


What’s it do without effects, especially the impulse reverb (maybe they are that bad)? Those definitely can suck up the CPUs.

It is the Pristine Space Impulse reverb that is sucking up cpu cycles. (I knew it was a hog but that much no.)


With Pristine Space, Space effects plugins Build 1702 reports 90-95% usage.

With SIR1008, Space effects - 50%

With just Space Effects 30-35%

With no effects at all - 15-18%

I will only be using Pristine effects for final mastering if then.

I guess my main point of the original post was to point out the slightly increased cpu usage of build 1702 vs 1701. Around 10% increase.

With all else being pretty much equal.



CPU usage can be based on the size of the impulse. A long impulse will use a lot more and 6 megs is pretty big, though it probably sounds good.

Are those stereo tracks? If they are then that’s almost the same as playing 8 mono tracks (not quite but close).

You usage still sounds a little high but it’s hard to pinpoint what it could be. It could be normal for your overall config.

A trick I started using was to save a submix that has nothing but the effect in is (like the output of the impulse reverb) and import that track. Then it doesn’t use any more CPU than another track. The down side is that the track sometimes needs to be slid left or right to compensate for effect latency - n-Tracks compensates and when the effect is gone the compensation is different.

I have the same problems with the vsts and high cpu usage. I try and mix some tracks separately and import them into the final mix with vsts added, so using less vsts in the final mixing.