Ideal Curves v2.0

There are many features of Har-Bal we still haven't discussed in this forum. Below we will start sharing a few items. Please feel free to add yours.
Post Reply
uncajesse
Posts: 27
Joined: Tue Apr 17, 2007 10:53 am

Ideal Curves v2.0

Post by uncajesse »

The Ideal Curve is back. This time with different versions, and it warrants a full topic for itself. I don't think many people had found it in the 2 topics it was posted in before.

What are the Ideal Curves? Well... what I did was take an hour of maximum intensity white noise, loaded it into Har-Bal and EQed it along what I call "Ideal Curve". Basically flat until above 1kHz where it starts to slope downwards so that 20kHz is 10db less than 1kHz, and it seems to always produce very natural & flat sounding results. :mrgreen: Then I rendered the EQed white noise and loaded them into Har-Bal to create the ANL files you are about to download. 8)

How you use them is simple.

1. You just drag whichever ANL file you want onto the Har-Bal window, and make sure the "Toggle reference spectrum plots" button in the bottom-right is enabled, and you will see the ANL curve you just loaded as darker colored lines of the Peak, Mean, and Average plots... whichever of the three you have enabled at the time.

2. Not needed but I highly recommend it, that you "Select the parametric EQ cursor" and selecting from 1kHz make it as wide as it will go (past the edges, and then some) and adjust the whole level of the track so that it's average relatively matches the level of the Ideal Curve you loaded. Usually it will be lower. If it's not, then you probably should ask for an unlimited mix, or a remix. ;)

3a. Either do what I do, and use the curve as a visual reference. Don't underestimate the power of just having a quick visual reference there. Or...

3b. "Select the intuitMatch cursor" and try dragging it from just above where the bass peaks are (usually not much higher than 100 Hz) and drag it up to where the treble drops a lot. You can also try dragging it all the way to 20kHz, just experiment, see what happens, and don't be afraid to Ctrl+Z (Undo) as much as you need. Once you have a good starting point, then experiment with how intuitMatch cursor reacts around the track's "problem areas" that it may have after the initial whole-spectrum intuitMatching. Remember not to overdo it, and always Ctrl+Z if the change didn't make it sound better.

That's about it. Enjoy the references, and let me know how it goes with using them in this topic.

JesseG
Attachments
Ideal.Curves.v2.0-JesseG.zip
(20.9 KiB) Downloaded 1122 times
HarBal
Site Admin
Posts: 761
Joined: Mon Apr 19, 2004 8:18 pm
Contact:

Re: Ideal Curves v2.0

Post by HarBal »

That basically tallies with my mental picture but it has built in assumptions about the type of music and the instrumentation involved. For the average pop/rock/hip-hop track it may work quite well but if the track lacks cymbals for instance, using such a reference will make it sound too bright, similarly, if it lacks drums the bottom end may end up too prominent, making it sound woolly. There are also cases where producers deliberately, for example, want to set the drum kit in the background near the limit of perception (ie. almost being masked by other instruments), in which case, applying such a reference will completely destroy the intent of the producer.

Hence, I never use referencing in this way, but rather in a very loose sense and usually only to make a compilation of tracks sit well together. I basically don't use the intuitMatch tool at all, but instead use manual editing, principally on the average power spectrum trace, to make the 1/3 octave view of the spectrum look smooth yet still basically retain the general trend of the original spectrum. That way, I generally end up with a track that honours the intent of the recording but improves on the clarity and cohesiveness. IntuitQ attempts to do this, though I can do much better manually, and that is what I generally do.

Of course, if the recording is severely tainted by the bad acoustics of the environment where it was recorded and mixed, then severe adjustment is likely necessary and in such cases you need to take an "educated guess" as to what the original intent of the artist was. Generally speaking, if the modified spectrum over/underlays the original in an average sense then it is a fair bet you are honouring the original intent and compensating for bad acoustics.

That is my point of view anyway.

regards,


Paavo.
uncajesse
Posts: 27
Joined: Tue Apr 17, 2007 10:53 am

Re: Ideal Curves v2.0

Post by uncajesse »

"The vast majority of EQ that we do in mastering is to compensate for the errors in the mix engineer's playback system."
- Bob Ludwig
SkipB
Posts: 21
Joined: Fri Jun 11, 2004 10:07 am
Location: Gray,La

Re: Ideal Curves v2.0

Post by SkipB »

Thanks Unca Jesse!! I have been using the Ideal Curve from the first one You Made.I pretty much use the Flat one but the others are interesting for experimentation. I also Have to agree with Paavo in his method, Mainly because he built the program then he should know how better to use it LOL!!. No just kidding. I having been working intensly with Harbal Trying to Master its Capabilitys,and any advice or new method for its implimentation to Audio makes me hunger more to learn.Regards to all. Skip Balltzglier
"Who's worried about the marsh when your up to your waist in Alligators"
cgibson7
Posts: 1
Joined: Wed Sep 01, 2010 1:14 am

Re: Ideal Curves v2.0

Post by cgibson7 »

Thanks for these, I nice additional tool (I have only just installed on my mac so am eagerly learning the different the approaches) :D
Post Reply