Free spectrum software?

[Stephen Perry]

I'll do some spectrograms for my next tests, post them up for examination. What's a decent free piece of software for that?

Thanks!

[fscotte]

Audacity.

[Pete Jenner]

+1 on Audacity.

[Jim Adwell]

+2 for Audacity.

[Stephen Perry]

I will get audacity and figure out how to use it!

I'm actually looking forward to testing some bridges. I have a fine player with excellent ears on tap for testing and video work. Should be a fun exercise. I will test first just mass through tests of a moderately heavy thick maple bridge that is progressively lightened by drilling holes up from the bottom.

I will then take the best mass and attempt to hit that mass with a new pattern bridge. Test that. Then "tune" it per my ears and then my player's ears, looking for progressive spectrum changes.

Should be fun.

[Marty Jacobson]

I also use Audacity. Record a sample, select the waveform, then choose the "Plot Spectrum" command from the "Analyze" menu.
You can then export the FFT data as a CSV and analyze it with whatever tools you are most comfortable, so you are not limited by viewing only one data set at a time or with Audacity's simplistic (but effective) display of the information. For example, there's no way to zoom in on the graph directly in Audacity, or to set the extents of the displayed information.

Have you seen this Campanella piccolo mandolin with a violin-like bridge?
http://www.mandolincafe.com/forum/sh...anella-Piccolo

[DHopkins]

Audacity. Probably one of the best free programs I've ever seen.

[dezguy]

If you are really serious about this, I would consider purchasing a spectrum analyzer. The problem with something like audacity is that the digitization process drops tons of data.

For $99 you can get a hobby license to Matlab. Which has some pretty awesome data analysis / spectrum analysis tools. A bit of a learning curve, but really useful.

Cheers,
Mike

[Stephen Perry]

I'm kind of an 18th C guy and use my ears. People bug me about "scientific" evidence, so that's what I'm looking for. I won't claim any great understanding.

[dezguy]

I'm kind of an 18th C guy and use my ears. People bug me about "scientific" evidence, so that's what I'm looking for. I won't claim any great understanding.

Fair enough

I was going to add, you'll probably want to use a lab quality microphone too. Something like http://www.bksv.com/Products/transdu...artridges/4160

With a flat frequency response across a wide bandwidth.

Essentially any spectrum analysis on sound captured through a computer is going to be a pretty blunt measurement.

Cheers,
Mike

[Nevin]

If you don't have access to a high quality micraphone, you could try sending sine wave signals through the mic you have so you at least have an idea what variations to expect.

[Marty Jacobson]

I would argue that if you need a lab-quality microphone to detect the changes in timbre or volume, then the changes aren't significant enough to be worth implementing. It sounds a little like saying you'll need 20x magnification to check whether or not you need to mow the lawn. If you were doing comparisons between multiple instruments or between multiple lab setups, having well calibrated mics is a good idea, if you have the funding for it, but for just seeing the big peaks and valleys, it doesn't take much in the way of special equipment to get a pretty clear picture.

[dezguy]

I would argue that if you need a lab-quality microphone to detect the changes in timbre or volume, then the changes aren't significant enough to be worth implementing. It sounds a little like saying you'll need 20x magnification to check whether or not you need to mow the lawn. If you were doing comparisons between multiple instruments or between multiple lab setups, having well calibrated mics is a good idea, if you have the funding for it, but for just seeing the big peaks and valleys, it doesn't take much in the way of special equipment to get a pretty clear picture.

It's not so much about needing a 20x magnification to see the lawn. It's more akin to needing a a clear piece of window. One without any huge warping on your view.

The problem with running a cheap mic into a sound card and then audacity is that the spectrum represented would be a really distorted picture. So much, that it's probably not even worth doing.

In that case your probably better off not trying to rely on pseudo science and instead just rely on your ear.

Cheers,
Mike

[dezguy]

"If you don't have access to a high quality micraphone, you could try sending sine wave signals through the mic you have so you at least have an idea what variations to expect."

Yeah, that's a good option for sure.

Cheers,
Mike

[fscotte]

There's not going to be a significant difference between a $20 mic and a $2000 mic for something like this. I know because I've owned some very expensive equipment in the past, Neumann mics, vintage and modern, to be exact.

[Nevin]

There's not going to be a significant difference between a $20 mic and a $2000 mic for something like this. I know because I've owned some very expensive equipment in the past, Neumann mics, vintage and modern, to be exact.

I would respectfully disagree with this. Microphones have different response curves. That is one reason why people prefer diffent mics for different applications. That is not too say that the most expensive mic has the flatest response. People appreciate the way a mic can color a singnal the same way they appreciate the way a tube amp colors a signal. I have a Rode N2 1A mic that has a rather flat response. I got mine used for under $100. It is very versitile but does not give the warmth to vocals that a good vocal mic does. On the othe hand most good vocal mics don't have flat response curves.

[fscotte]

You can find a flat response mic for $20. That's all you really need for this. We aren't looking for smooth vocals or a high end rolloff so the mandolin can sound like Sinatra, ala U47.

[dezguy]

Wonderful, if there is a $20 mic that'll work that's awesome. My comment wasn't meant to imply it had to be expensive. Rather that it has a wide bandwidth and a flat response.

Cheers,
Mike

[Stephen Perry]

This would be for comparison, not for absolutes. I do have some decent microphones somewhere. I should run across them if I keep my eyes open.

I'm actually OK at real science. The N just tends to be rather low in these shop efforts. Real science does not take special gear, just a good solid plan with some preliminaries. My interest at the moment is whether I can see as obvious what I can hear as obvious. As to science, an ear is perfectly good.

[tablaninja]

I have an omnidirectional test and measurement microphone you can borrow. If I recall it has a flat frequency response from 20-25kHz. I should be working in your area some time after labor day and was going to contact you anyway about a possible bridge installation/setup.

[Stephen Perry]

I'll be around!