PDA

View Full Version : PC into RTA?



Alex Lancaster
10-15-2003, 02:01 PM
Hi:

I would like to turn a PC into a RTA, I have seen programs advertised, but would like Your opinions about program, mike, and, I guess a sound card, I do not think the one I have (generic), would be even close; The PC is a [email protected], 1G ram, 15" Hi-res LCD monitor.

I will use it for my and friend´s curiosity.

Thanks, Alex.

Robh3606
10-15-2003, 03:22 PM
Well I considered that and tried it but my PC is ancient a 200MMX so it really could not run it properly. Your PC looks like a keeper to me so you don't have that issue. Depending on the cost of the software and soundcard you might want to consider a stand alone. Granted you have a lot more utillity down the road with a PC test set-up. I use a Berhringer 8024 as an RTA it was a cheap way to get a 1/3 octave RTA. The bottom line is what ever can do what you want for the lowest cost but think ahead too. If I ever get a faster PC I would seriously looking into what measurement software is available.

Rob

boputnam
10-15-2003, 08:48 PM
Hey, Alex...

Man, IMHO you just gotta go for SmaartLive 5 - Link (http://www.siasoft.com/products/smaart_live.shtml) - it's actually owned by Harman/JBL.

Anyway, it is The Standard. Pricy, but incredibly functional. You'll need some doohickys to get things running - an Earthworks reference mic (M30), and the XLR interface/preamp/phantom power, etc..

In every FOH gig I've done this is the standard. Pink noise generator, RTA, much, much, more. Click on a bar of the "spectrum mode" EQ and it tells frequency and note of the scale. Handy, handy, handy.

Ian's got a much more affordable "pedestrian priced" work-around that he's real fond of, but I've not tried it. Ian's got his head in baffles, so no telling when he'll pitch-up.

You need a duplex sound card - one that can coincidently send and receive (pink noise out, measurement data in). These are more standard in newer computers, than in the past. It's handiest on a laptop for portability (you can take it to your friend's house), but with long enough cables, you can RTA your house, surely.

So, for a no-risk set-up, this is what you'll see FOH and monitors mix, as well. It's how the pros have gone. Period.

Alex Lancaster
10-15-2003, 09:27 PM
Bo:

Thanks, I saw the page, but what I want is something with a couple less zeros.

Alex.

Oldmics
10-15-2003, 10:45 PM
Just a note about the ownership of Smaart. The S.I.A. Software Company who did the original design work for JBL was purchased by E.A.W. (Eastern Acoustics Works).With the release of Smaart Version 3.0 in March of 1999 all ties with JBL and Harman were broken off.Currently JBL refers Smaart updates to the S.I.A. Software site.It has become the standard in pro sound world even though I would say 98% of the owners that I have encountered only know how to turn the RTA window on with there laptops for "The Cool" look! Oldmics

Ian Mackenzie
10-16-2003, 03:20 AM
Na Na Bo,

Been observing the fauna in the sun today.....

Try this link for the software called Winairr by the inventor

pcbunn.cithep.caltech.edu/jjb.html

You can buy it legally from audio express for $49.00 here

http://www.audioxpress.com/bksprods/sotware/sof-airw.htm

WinAIRR is a software tool for measuring the frequency response of loudspeaker systems. It runs on a PC with Windows 95/98/NT equipped with a sound card that must feature full-duplex operation. Although WinAIRR is not a professional measurement system, useful reproducible results can be obtained with careful adjustments of the supplied controls. Version 4.02 features:
• Averaging: Time or Frequency
• Sample Record Delay
• Time Domain Cuts
• Sample Rate Adjustment
• Analysis Results: Standard, Histogram, Phase, Waterfall Full Mixer Control, Plot Scaling, Adjustment and Subtraction
• Storage of Results to Disk, and Retrieval from Disk
• Support for 48kHz sampling rate on soundcards which Version 4.02 includes Executable and software manual in HTML (Web Browser) format.
Requirements: IBM Compatible, 486+ (Pentium recommended); Windows 95/98 or Windows NT; 8Mb RAM (32 recommended); 1Mb hard disk space; sound card capable of full-duplex operation; VGA 1024x768 pixels, 16 colors. Shipping Wt: 1 lb.


After that go up to IMP or Clio, but for what you need to know this is it as far as software is concerned. Then its down to listening tests to trim the L Pads. But you will need a mic and preamp ..also available from audio express from memory

Ian

boputnam
10-16-2003, 07:15 AM
Originally posted by Oldmics
I would bet 98% of the owners only know how to turn the RTA window on with there laptops for "The Cool" look! :duck: Whew... Could be. I've seen it assist in taming some phenomenally difficult acoustics, but I tend to hang around pedantic types ;) .


Originally posted by Ian Mackenzie
Been observing the fauna in the sun today..... Knowing you, she's gotta have a name...? :yes:

Still mean to give that Winairr a try - I'm horribly behind on too many fronts. There's a few other PC RTA's I've used - more of the freeware stuff. Lemme check...

boputnam
10-16-2003, 07:24 AM
Yea, Alex - here's another...

True Audio / TrueRTA - Link (http://www.trueaudio.com/rta_abt2.htm) I've used this in the past, and it was pretty comprehensive. It has since been upgraded. I've not trialed the most recent version, but it's priced-right. :yes: I've not A/B this against Smaart.

Here's the screen shot of the "Bar Mode" - make it your wallpaper and you'll at least look cool! :rotfl:

boputnam
10-16-2003, 07:30 AM
I'm not finding too much else that I recall knowing about.

Try a google on "spectrum analyzer" and other iterations, and see where you end-up...

Alex Lancaster
10-16-2003, 07:56 AM
Thanks!, I contacted them about what microphone to use, and if my sound card is adequate, I kind of doubt it, it is an el cheapo; Suggestions?

It really looks good and affordable.

Alex.

boputnam
10-16-2003, 08:12 AM
Originally posted by Alex Lancaster
...what microphone to use - Suggestions? Ugh... They are all expensive. Earthworks - Link (http://www.earthwks.com/ns/index.html) brands the best known and most acclaimed "M30", but they are pricy - much more than say, a Shure Beta SM58 (but that's only a lowly best-of-class vocal mic ;) ).

Ol' Ian teased me with an email wherein he describes how he put-together rather inexpensively a flat-response condensor mic (Panasonic, I believe... :confused: ) and preamp confab that reportedly works terrific. That email's at home. Maybe, if he wakes early enough, he can post that before going to the dayjob...

Oldmics
10-16-2003, 09:18 AM
Howdy All,Thought I would stick my two cents in here.I see a lot of PC and stand alone measurement devices being used (mostly inappropreately)in my world.I would like to address some of the things that I see which may be benefical to new users.The measurement taken with any device is only as good as the knowledge availiable to the person doing the test and limited by the hardware used in such tests.An R.T.A. adheres to standards that are routinely being ignored as more and more people get into the measurement game.Critical measurement is done in an anechoic chamber with calibrated mics.Before computers these flat referenced mics were a hughly expensive and precious items stashed away in the labs.Well we can"t put our sound system from our listening room into an anechoic chamber for measurement so that leaves us with calibrated mics in the real world.The few fine quality measurement mics that are still manufactured today are very expensive due to there quality construction and materials.B and K microphones come to mind.The current crop of relatively inexpensive measurement mics (under $1500.00 which is cheap for a measurement mic)do not have ruler flat response(and expense) of the old calibration mics.However these annomilies can be entered into the software files of the particular program that you use to compensate and correct the calibration.That is the first mistake that I commonly see made.The omission of microphone anomilie information into the files of the user program. This lack of information will skew a response curve.Measurement taken with Shure 58s always bring a chuckle to me.A great mic,not slamming the 58 but not quite right for this application.What ever mic that you decide on,make sure that you get the calibration files with it so that they can be entered into the program that you decide upon and then load those files into the program.Without a standard to hold to whats the sense in measurement anyway?Personally I like Praxis from Liberty Instruments for driver analysis.They also have a freeware R.T.A. availiable at www.libinst.com Most of these programs will have to be purchased before the microphone information can be loaded in.Good luck,Oldmics

JonFairhurst
10-16-2003, 10:03 AM
This is the mic I use:

http://www.behringer.com/02_products/prodindex.cfm?id=ECM8000&lang=eng

It's not ruler flat, but it sure beats a 58 for this application. (Don't try mic'ing a Marshall stack with an ECM8000 though...)

The ECM has a 2dB bump at around 7k, but if you believe the chart it's smooth enough to be factored out. Of course, they don't show the standard deviation or a file with the response of your particular serial number, but what do you want for around $35?

For RTA software I found some German shareware a couple of years ago. Can't think of the name right now, but it wasn't fully developed, and was somewhat difficult to use. I can't read German, but most of the technical terms were easy to figure out. It's nice to see that there are more RTA options these days.

For a soundcard, the M-Audio FireWire 410 would be killer. It uses 1394, so it plugs easily into a laptop - and it gets the A/Ds outside of the desktop case. It has two XLR inputs with phantom power. It also has a 108 dB S/N and supports 24-bit, 192 kHz. Plus ten outputs. It's overkill if you're not doing home studio production, but overkill can be a good thing :-)

I use an EchoAudio MIA and a Mackie 1402 VLZ-Pro mixer. The MIA uses a balanced input, so noise is kept relatively low.

And noise is the killer for RTA measurements. You may expect to see some nice smooth curve that describes your system. You won't. If you're doing sweeps, any extra noise will affect the results. And then there are room resonances and possible rattles. Make the volume high enough to overcome noise, and you may blow your drivers (and ears). Plus any mis-match in impedances or overdriven section may distort, clip or limit your result.

Reading a sweep is like reading a lie-detector test. You have to know how to interpret the results.

Finally, if you have knobs on the front of your devices, you may find things to be fleeting. Multi-turn pots inside of your crossover is much better for setting and forgetting. And in the end, you will likely wish that you had a muti-band EQ, so you can compensate for your room and all the rest.

These days I've been tuning to taste. As long as it sounds good to me, what more could I want :-)

4313B
10-16-2003, 10:14 AM
"These days I've been tuning to taste. As long as it sounds good to me, what more could I want :-)"

:yes:

:cheers:

The other stuff can be quite fun though if not taken too seriously :)

Robh3606
10-16-2003, 10:22 AM
My favorite is setting up the 1/3 RTA and EQ's to get things looking real good. Then you run your sine sweep. Damn things lie like hell! Using pink noise they average whats there and don't excite the room resonances . Too narrow in bandwidth. But what can you do. Have to admit they are useful tools but like you say you need to tweek by ear when all is said and done.

Rob:) :cheers:

JonFairhurst
10-16-2003, 10:25 AM
Originally posted by Giskard
"These days I've been tuning to taste. As long as it sounds good to me, what more could I want :-)"

:yes:

:cheers:

The other stuff can be quite fun though if not taken too seriously :)

Very true. And after doing some RTA work you will calibrate yourself, as well as the speakers/system/room. I find that I want to roll off the high end slightly, and want to bump the low-end *for some source material*. It's like adding salt and pepper to your food.

My high-end tuning stays contant for most any decent recording. But I find that I want to customize the bass on nearly every CD - even if by a dB or two. The difference between weak bass and flabby boom can be very subtle. At least for my ears in my room with my speakers.

Ian Mackenzie
10-16-2003, 10:38 AM
Interesting Jon,

I made a laboratory mic using a Panasonic insert which is known to be flat within a 1dba over most of the range, it is used in the popular Mitey mic marketed for speaker testing in the speaker diy field.

Basically you put the insert into a thin length of brass or aluminium tube about a foot or so long to avoid difraction effects.

The peamp with phantom PS is nothing more than a two transister gain stage with phantom supply via the nine volt battery and a resister. (can scan the design for those interested)

The PC RTA look fun and pretty, however non appear to have variable gating windows, and non appear to use FFT pulse .

There in lies a problem for a speaker builder if you require true speaker only measurements on the mid and top end without room effects or other early reflections.

The convenience of being able to selectively remove room effects is crucial for speaker testing.

I can't possibly see any one lugging a 200lb speaker out into their drive way everytime you wana do a frequency run..muhhahah frighten the neignbours away.

This the data from JBL of the 4345 measure in open air.

macka

Ian Mackenzie
10-16-2003, 11:06 AM
Attached is a gated measurement using Winairr,

The top curve is the time response and the vertical lines are the gates or time sample I have chosen.

The immediate wiggley lines on the time curve after the initial pulse are baffle reflections (being cardboard at the time..muhhahahah) and time displacement of the various drivers (over a millisecond), everything else is chopped out.

The second curve is the response, and you can see the clear thin line except the very top end which could be further gated but we would loose details of the lower midrange.

The lower curve is the system calibration curve

If this curve were not gated, the mid and high frequences (and low ) would be an unreadable blurr and usefuless.

Also you can do far field measurements lie this one @ 2.5 metres which are more meaningful, particularly with horns when near field are influenced by the immediate pressure zone of the flare.(and still gate out room interference)

The mid range can be independantly analysed by also gating out reflections from the floor, walls and ceiling around 300-1000 hertz.

Low frequency measurements to me are a bit of a waste. The modelling of the LF on bassbox is a far more accurate indicator than any measurement. There are so many influences from your room modes its easier to work out the main resonances with a pocket calculator, and you really can't fix them properly with EQ, hence the use of tube traps and refractors to absorb LF content.

Ian

boputnam
10-16-2003, 11:14 AM
Originally posted by Oldmics
Measurement taken with Shure 58s always bring a chuckle to me.A great mic,not slamming the 58 but not quite right for this application. I certainly wasn't - nor wouldn't - use an SM58 for that. That'd be dumb-is-as-dumb-does. I must've confused by even mentioning a vocal mic in this Thread. I didn't fugire anybody here would misinterpret that. :no: The rest of your post is quite iteresting. :yes:


Originally posted by JonFairhurst
And after doing some RTA work you will calibrate yourself, as well as the speakers/system/room. As posted on a very similar thread some time back, after flat to Pink Noise I toss the RTA and fine-tune to ear. I find the RTA is most useful for identifying the large reflections/standing waves of the room acoustics. After that, nothing beats painstaking tweaking over the course of a few days. And I'm still using my Gold Line - Link (http://www.gold-line.com/30mp.htm) with it's reference mic. Good enough for now. Not gone PC, yet...

Ian Mackenzie
10-16-2003, 11:18 AM
Here is another screen dump after some tweeking of the faders.

Not bad for $50 bucks when all we need to know is if there are any major peaks, dips building an assumed design from JBL.

Granted however, if your into graphic and parametric EQ, the RTA is still a useful tool.


Ian

Oldmics
10-16-2003, 12:46 PM
Hey boputnam,I knew that you were only doing a tongue in cheek kinda comparision about the 58.Please do not think I was trying to cast dispersions towards you regarding my comments about the use of a 58 for measurement or regarding the comment about encounters that I have had with 98% of Smaart users.I think that this is a more in touch forum with the way things are correctly done.Some of the exacting measurement that I do regards arrays of systems.Its a diffrent mindset when you are measuring a stereo cluster with a six box horizontal trapazoidal array that is multiple rows hi on each side.Trying to achieve null free coverage in both near and far field listening positions in combination with high SPL is like a quest of the Holy Grail.Thank God for line source array technology.As an experiment place your home stereo speakers together and see what I am talking about.Start listening and then try to correct for that mess.Upon measuring you will see phase goes right out the window and comb filtering becomes a new factor to deal with.Add in humidity and altitude outside and you have just another day at the office for me.Thats why I need exacting measurement.While it can be fun ,I do take it seriously.Best regards,Oldmics

boputnam
10-16-2003, 01:47 PM
Hey, Oldmics...

Thanks, and no dispersions hit me :duck: - but I was compelled (after Jon's comment on the 58, too) to ensure we here were in-no-way suggesting that application! Hell, next-up, someone'd be doing RTA with a RadioSchucks Hi-Z hard-wired Karaoke mic. Boy-howdy, that'd be it.

I love your comment about line arrays - not the end-all, but heaps better than the stacks-of-the-past. My friends think I'm nutz :nutz: (OK, so they're right...) to seek seating in "null free coverage" areas, and they give me the Homer Simpson "blink" when I try and explain it all (maybe it is just me, after all...) :(

All I'd add, is after all the "ringing the house" (if you will forgive the use of the term), it's back to square-one the minute the audience arrives and adds their acoustical damping to the room. Yikes...

Anyway, if you're ever doing FOH at a show I happen to be at, know that your work is mighty appreciated! :)

JonFairhurst
10-16-2003, 03:08 PM
A Radioshack mic? No way. To really save money, just use the microphone in your telephone. Your wife won't mind.

boputnam
10-17-2003, 06:35 AM
Originally posted by JonFairhurst
A Radioshack mic? No way. To really save money, just use the microphone in your telephone. Your wife won't mind. Ha! You prompted an even better idea - a grid of cell phones! Think of the cost savings in relieving all that troublesome wiring... ;)

Can you hear me now...? :duck:

Can you hear me now...? :duck:

Can you hear me now...? :duck:

Alex Lancaster
10-17-2003, 06:45 AM
Thanks everybody!

I think I know what I want, now comes the slow process of getting it down here, but I will.

Alex.

JonFairhurst
10-17-2003, 09:02 AM
Buena suerte, Señor Lancaster.

Mr. Widget
10-17-2003, 10:09 AM
As a long time reader of "Speaker Builder" and now "AudioXpress" I have been meaning to get a Mighty Mike. Has anyone used one. They are based on a cheap (possibly the same Panasonic unit) condenser mic, but are tweaked and available in a calibrated version that seams to be quite the deal.

http://www.audioxpress.com/bksprods/kits/kd-4.htm