At first, using a custom weighting curve in Smaart seems like a brilliant time saver, but it turns a guide into a rule and could make your optimization procedure too rigid.
A custom weighting curve is like a filter for your data, similar to a microphone correction curve. Its intention was to add A and C weighting to Magnitude and RTA graphs, but Smaart makes it really easy to turn your target trace into a weighting curve as well. Then, instead of matching your measurements to a target trace, you can simply match them to the 0dB line.
Here’s an image of me comparing a measurement to a target trace.
And here’s the same measurement with the custom weighting curve.
At first, this saved me time. My normal procedure of capturing multiple measurements at once means that I’m also hiding multiple measurements at once. It usually goes like this:
- (G) Start generator
- (shift+space) Capture all
- (G) Stop generator
- (command+shift+H) Hide all traces
- Scroll down to find target trace. Unhide it.
Using the the custom weighting curve allows me to skip step 4 and work with a cleaner graph. I tried it on a show and it worked great.
Here’s how to create your own custom weighting curve:
- Load the appropriate target curve. (you can download mine here)
- Right-click on the name and choose Export as Weighting Curve
- (option-G) Open Measurement Config and choose one of your measurements.
- Under Global Settings, choose your weighting curve.
If you want to test this out on saved trace: right click > Info > Weighting
Pretty cool, right?
Here’s the problem.
Your target curve should be used as a guideline and not a rule. The guideline can trigger your intuition when comparing two traces.
Eg. The high shelf at 10kHz reminds you that measurements at different distances will show different amounts of air absorption. The low shelf at 50Hz reminds you that you should expect to see more room gain in the low end when you are inside rather than outside.
Here’s an annotated target curve to give you an idea.
Unfortunately, when you use the custom weighting curve, you don’t have those cues and you can forget that you are looking at filtered data. Plus, if you ever want to review past measurements, you have the added complexity of trying to remember which ones were made with the weighting curve on or not.
I recommend that you try it out because it’s easy and quick, but I don’t recommend that you use it in your normal tuning processes.
Have you used custom weighting curves in Smaart? What were your results?