Essential tools for proper/accurate audio calibration? - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 48 Old 10-14-2012, 09:20 AM - Thread Starter
Advanced Member
 
metallicaband's Avatar
 
Join Date: Feb 2007
Posts: 993
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 98
Hello all,

I was wondering what would you guys consider as absolute essential tools (hardware or software) to get a proper and very accurate audio calibration for home theaters and stereo systems, a calibration that's not just about sound levels, but involves EQ and knowing where to acoustically treat your room for high and low frequencies, basically deal with everything related to how your system sounds.

Here's what I use so far with my basic calibration skills:

- Measure tape
- Laser pointer to adjust where the speaker (tweeter) points at
- Radio Shack Analog SPL meter
- Internal receiver test tones

I've got DVE:HD basics but I use that for video calibration only, heard it's better to use the internal receiver tones for audio calibration but I'm not sure if that's true or not.

KRP-500M | Kuro 9G 5020FD + e-lite mod | B&W 683 Fronts | B&W 685 Center | B&W DM601 S3 Rears | HSU VTF3 MK3 Sub | Yamaha RX-V663 AVR
metallicaband is offline  
Sponsored Links
Advertisement
 
post #2 of 48 Old 10-14-2012, 09:47 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,371
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 975 Post(s)
Liked: 546
The best tool by far is knowledge of what to measure and what to do about them! smile.gif Here is the one and only book I recommend: http://www.amazon.com/Sound-Reproduction-Acoustics-Psychoacoustics-Loudspeakers/dp/0240520092/ref=sr_1_1?s=books&ie=UTF8&qid=1350229225&sr=1-1&keywords=floyd+toole.

Once there, a really nice piece of software is Room EQ Wizard which is free to get with registration. Note that it is very easy to get misleading data and conclusions out of the tool (as with any other) so I can't recommend highly enough that one learns science per above.

You need a computer to generate the tones and run REW. Laptop is the weapon of choice since it is easier to move in and out of your listening space.

Next you need a measurement mic& amp. Prices vary as does quality although for some work, it may not matter. If you have an AVR and it came with a measurement mic, you can plug that into your PC's mic input and start there for now. That way, you can be sure that want to spend the time and energy to go there before investing more.

Good luck smile.gif.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #3 of 48 Old 10-14-2012, 10:16 AM
 
diomania's Avatar
 
Join Date: Dec 2008
Posts: 1,389
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 49
Quote:
Originally Posted by metallicaband View Post

Hello all,
I was wondering what would you guys consider as absolute essential tools (hardware or software) to get a proper and very accurate audio calibration for home theaters and stereo systems, a calibration that's not just about sound levels, but involves EQ and knowing where to acoustically treat your room for high and low frequencies, basically deal with everything related to how your system sounds.
Here's what I use so far with my basic calibration skills:
- Measure tape
- Laser pointer to adjust where the speaker (tweeter) points at
- Radio Shack Analog SPL meter
- Internal receiver test tones
I've got DVE:HD basics but I use that for video calibration only, heard it's better to use the internal receiver tones for audio calibration but I'm not sure if that's true or not.
Try this thread, it's free: http://www.avsforum.com/t/1413173/does-sound-sounds-better-in-a-room-full-of-furniture-and-stuff-or-without
Once you go through it, you will know a good bit of programs for such task. Also, you will know which poster to take seriously and which one not to.
diomania is offline  
post #4 of 48 Old 10-14-2012, 12:06 PM
AVS Special Member
 
Ethan Winer's Avatar
 
Join Date: Apr 2003
Location: New Milford, CT, USA
Posts: 5,748
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 134
Quote:
Originally Posted by metallicaband View Post

I was wondering what would you guys consider as absolute essential tools (hardware or software) to get a proper and very accurate audio calibration for home theaters and stereo systems, a calibration that's not just about sound levels, but involves EQ and knowing where to acoustically treat your room for high and low frequencies, basically deal with everything related to how your system sounds.

The Room EQ Wizard program mentioned is excellent and free, and it works on all major operating systems.
Quote:
Here's what I use so far with my basic calibration skills:

- Measure tape
- Laser pointer to adjust where the speaker (tweeter) points at
- Radio Shack Analog SPL meter
- Internal receiver test tones

The RS meter is fine for low frequencies, but it's not accurate above around 1 KHz. And test tones are not useful unless they sweep to include every frequency. These two articles explain all of this in detail:

Room Measuring Primer
Comparison of Ten Measuring Microphones

--Ethan

RealTraps - The acoustic treatment experts
Ethan's Audio Expert book

Ethan Winer is offline  
post #5 of 48 Old 10-14-2012, 12:48 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,371
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 975 Post(s)
Liked: 546
Quote:
Originally Posted by Ethan Winer View Post

Comparison of Ten Measuring Microphones
--Ethan
I recently landed on that page and found it to be really good work Ethan! It is the only comparison report of its kind that I have seen.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #6 of 48 Old 10-15-2012, 02:09 PM
AVS Special Member
 
OtherSongs's Avatar
 
Join Date: Jan 2009
Location: Chicago-ish
Posts: 1,663
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 15
Quote:
Originally Posted by amirm View Post

The best tool by far is knowledge of what to measure and what to do about them! smile.gif Here is the one and only book I recommend: http://www.amazon.com/Sound-Reproduction-Acoustics-Psychoacoustics-Loudspeakers/dp/0240520092/ref=sr_1_1?s=books&ie=UTF8&qid=1350229225&sr=1-1&keywords=floyd+toole.
Once there, a really nice piece of software is Room EQ Wizard which is free to get with registration. Note that it is very easy to get misleading data and conclusions out of the tool (as with any other) so I can't recommend highly enough that one learns science per above.

Thanks for your above comments and link.

22 owners on amazon give this book ("Sound Reproduction: The Acoustics and ...") an overall rating of 5 stars. I went with a new copy from the low cost 3rd party seller at $32 plus $4 shipping. smile.gif The downside is the very open ended time for delivery. frown.gif

The best is the enemy of the good. Voltaire (1694-1778)

OtherSongs is offline  
post #7 of 48 Old 10-16-2012, 08:29 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 762 Post(s)
Liked: 1175
Quote:
Originally Posted by metallicaband View Post

Hello all,
I was wondering what would you guys consider as absolute essential tools (hardware or software) to get a proper and very accurate audio calibration for home theaters and stereo systems, a calibration that's not just about sound levels, but involves EQ and knowing where to acoustically treat your room for high and low frequencies, basically deal with everything related to how your system sounds.
Here's what I use so far with my basic calibration skills:
- Measure tape
- Laser pointer to adjust where the speaker (tweeter) points at

IMO, sort of like putting a telescopic sight on a dust mop. ;-)
Quote:
- Radio Shack Analog SPL meter

Echo what everybody else says about this being more like a dust mop than a laser pointer. ;-)
Quote:
- Internal receiver test tones

Echo what everybody else says about this being more like a dust mop than a laser pointer. ;-)
Quote:
I've got DVE:HD basics but I use that for video calibration only, heard it's better to use the internal receiver tones for audio calibration but I'm not sure if that's true or not.

If you're really serious about setting up a room, you're going to bit the bullet to the tune of a couple of hundred dollars for the kinds of analytic tools that others have mentioned. Like everybody says, its money well spent. The exact form of these analytic tools will vary. One thing is that the software which is all freely downloadable, is more dissimilar than the hardware which will actually cost you the bucks.

The first tool is a PC laptop that you hopefully already have. It doesn't have to be much, but it has to be available.

While you may be able to get away with one of those little calibration mics that comes with self-adjusting receivers plugged into your laptop's mic input, just about everybody goes with an external pro grade mic and a USB interface with regular pro grade mic inputs.

For mic - the Behringer ECM 8000 is probably the cheapest and just fine. Keep an eye on yours as I find that some of them tend to lose sensitivity, even if kept stored nicely.

http://www.behringer.com/EN/Products/ECM8000.aspx



The ART stereo USB interface is probably the cheapest way to get a regular mic input hooked into your laptop. I've never had one, but many people seem to find them useful.

http://artproaudio.com/art_products/signal_processing/usb_audio_devices/product/usb_dual_pre_ps/



For software, there are at least 3 freebie choices that make some sense:

(1) The speaker test function of the Audio Rightmark program

http://audio.rightmark.org/download.shtml



(2) Room Eq Wizard

http://www.hometheatershack.com/roomeq/



(3) Holme Impulse

http://www.holmacoustics.com/holmimpulse.php



If you are in a hurry, cut to the chase, REW hauls the mail.

I've listed them in order of my opinion of their professionalism, which is not a complaint or a detriment to any of them.

If you want to fool around comparing alternative views, try all 3.
arnyk is offline  
post #8 of 48 Old 10-16-2012, 12:02 PM
AVS Special Member
 
localhost127's Avatar
 
Join Date: May 2009
Posts: 2,286
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 40
  • your RS SPL mic can be used for setting relative gain between the sources
  • the dayton eccm6 mic is comparable in price to the ecm 8000, and includes the cal file.
  • +1 to the art dual pre usb (can generally be found for ~$69 shipped from B&H photo) - and allows one to utilize the hardware loopback correction (required for accurate time-domain analysis while taking into account hardware propagation delay/etc).
  • REW is free, but ARTA can still be had for relatively low cost - and the capabilities speak for themselves.
  • monoprice.com for all required cables and adapters
localhost127 is offline  
post #9 of 48 Old 10-17-2012, 10:23 PM
 
anwaypasible's Avatar
 
Join Date: Aug 2010
Location: illinois
Posts: 391
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
i've used the radio shack digital SPL meter and i've heard a difference in volume before the number on the display changed.
i've literally been able to adjust the level of my back speakers by a single number and hear the difference, but there wasnt any change on the display of the meter.
i dont remember if i was watching only the number or if i was watching the little bars on the bottom, but i thought the bars on the bottom were for each single decibel.. and if the bars reach all the way to the right, then the 'tenth' number changes .. like it goes from 70 to 80.

i seen that and i wasnt overly sad, because i know i can still use the meter to grab a level that is safe for the microphone.
just because the receiver says it was +1dB .. that doesnt mean that it was really +0.5dB .. and i kinda just went with that.
(if i change it twice, the number on the meter goes up)

i remember trying to see what kind of results i could get from the SPL meter and they sounded horribly awful.
i looked down at the grill on the tip of the microphone and thought 'is that what i am listening to with these equalizer settings?'

but anyways..
i've used the SPL meter to get the numbers for the back speakers close, and then did a single number up or down to hear if it blends any better.


if the time delay was working correctly, then identical front and rear speakers shouldnt need a gain increase or decrease .. unless the speaker is really far away, but even then it makes me look at the time delay as a simple delay .. not a delay with a filter for the air the soundwave is flowing through.
but that is what is good about the calibration microphone, it will look at the ringing of the room, but it is also forced to see the resistance of the air inside the room too.

if time delays had a filter for the air the soundwave was flowing through, the level of loudness would go further. and then you wouldnt need to turn up the gain for the speaker.
but
not all speakers are identical, even with the same ohm rating, when there is pink noise flowing through them.. because their impedance curve is different, maybe the sensitivity is different, maybe the speaker cone is smaller.

what i did was trust the receiver and left the rear speakers at 0dB because they are at about the same exact distance from the listening position.
i learned that the rear speakers arent really loud like seperate speakers, they just fill the rear of the room with sound that blends with the front.
i've tried listening with the rear speakers up louder and it sounded annoying because the audio from the rear speakers wasnt 'solid' or it was too solid and the audio didnt sound like it was clean and clear, it sounded like it had a bunch of audio effects like reverb added to it.
i've kept the dial down low ever since without ever wanting to look at the SPL meter to see if the rear are the same as the front.
it was one of them times when my ears were happy, i turned it up and i was less happy, and i didnt care to use the SPL meter to see if louder than 0dB was what all the 'reference' guides tell us to use.
i want to avoid a 'reference' guide telling me my happy ears are wrong and i should put it back to something that made my ears not happy.

i guess the worst that could happen is i pull out the SPL meter and it tells me the rear speakers are still louder than the front, but i can use my brain to tell me they arent.

but nevertheless..
a lot of people have a powered subwoofer and it wouldnt suprise me if the number on the SPL meter goes up higher than expected with the subwoofer going.
i would use the SPL meter to always check to make sure the dB level is safe to pull out the microphone and prevent damage.
or at least do yourself the favor of knowing how loud it is, because they said the ecm8000 distorts more and more as the volume goes up.
that way you can see what the dB level is to know how much distortion you are getting from the microphone, to avoid the distortion and get a cleaner reading.. or to push things to the limit without going over.



anyways..
nobody said anything about a microphone stand.
i use one and i like it because i can run out of the room without causing a reflection of soundwaves off my body.
it worked really well to keep the microphone at the ear level.
and it worked really well again when i had the microphone in the corner of the room behind the front speakers.. i wouldnt of been able to do that measurement with any satisfaction without the microphone stand.
anwaypasible is offline  
post #10 of 48 Old 10-21-2012, 06:42 AM - Thread Starter
Advanced Member
 
metallicaband's Avatar
 
Join Date: Feb 2007
Posts: 993
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 98
Thank you all for the detailed replies (specially arnyk that was very helpful)! Looks like I have a lot of reading (& investment lol) to do!

KRP-500M | Kuro 9G 5020FD + e-lite mod | B&W 683 Fronts | B&W 685 Center | B&W DM601 S3 Rears | HSU VTF3 MK3 Sub | Yamaha RX-V663 AVR
metallicaband is offline  
post #11 of 48 Old 11-06-2012, 04:45 PM - Thread Starter
Advanced Member
 
metallicaband's Avatar
 
Join Date: Feb 2007
Posts: 993
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 98
I have these stuff lying around (bought 'em like 6-7 years ago for some basic music recordings) and was wondering if they'll be useful for audio calibration and cut some cost on the hardware stuff:

Mic tripod stand, it's pretty big
Behringer mixer http://www.behringer.com/EN/Products/UB802.aspx
Mic cable http://www.zzounds.com/item--SHUC25J
Shure SM57 mic http://www.shure.com/americas/products/microphones/sm/sm57-instrument-microphone

Now I doubt that I can use that mic since it's made for recording instruments, but what about the other stuff, can I just get the ECM 8000 mic and see if I can use it with what I have?

KRP-500M | Kuro 9G 5020FD + e-lite mod | B&W 683 Fronts | B&W 685 Center | B&W DM601 S3 Rears | HSU VTF3 MK3 Sub | Yamaha RX-V663 AVR
metallicaband is offline  
post #12 of 48 Old 11-06-2012, 05:06 PM
AVS Special Member
 
DonH50's Avatar
 
Join Date: Feb 2010
Location: Monument CO
Posts: 6,275
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 223 Post(s)
Liked: 291
^ Yes.

"After silence, that which best expresses the inexpressible, is music" - Aldous Huxley
DonH50 is offline  
post #13 of 48 Old 11-06-2012, 07:18 PM
 
anwaypasible's Avatar
 
Join Date: Aug 2010
Location: illinois
Posts: 391
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by metallicaband View Post

I have these stuff lying around (bought 'em like 6-7 years ago for some basic music recordings) and was wondering if they'll be useful for audio calibration and cut some cost on the hardware stuff:
Mic tripod stand, it's pretty big
Behringer mixer http://www.behringer.com/EN/Products/UB802.aspx
Mic cable http://www.zzounds.com/item--SHUC25J
Shure SM57 mic http://www.shure.com/americas/products/microphones/sm/sm57-instrument-microphone
Now I doubt that I can use that mic since it's made for recording instruments, but what about the other stuff, can I just get the ECM 8000 mic and see if I can use it with what I have?


the mixer said it is good for 24bit 192khz
but
i would wonder if that means it doesnt zoom in well enough for the regular 16bit areas.
and because 24bit audio might include 16bit audio or all of the third 8 bits is enough of a layer to change the other two rows of 8bits underneath.
those types of situations could be the difference between an invisible lens and one you can see dirty.

if there is time to look at what is in the junk drawer because you are bored, then look.
you would almost need to record the room from a different order of harmonic using the audio measurement software if the circuit really did bog out for a 16bit recording.
...just a question of both halfs full or one half is empty to compensate for the premium quality of the other half.

i'd go nuts about the three band equalizer knobs .. i wouldnt want 'em at all.
anwaypasible is offline  
post #14 of 48 Old 11-07-2012, 06:14 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 762 Post(s)
Liked: 1175
Quote:
Originally Posted by metallicaband View Post

I have these stuff lying around (bought 'em like 6-7 years ago for some basic music recordings) and was wondering if they'll be useful for audio calibration and cut some cost on the hardware stuff:

Mic tripod stand, it's pretty big
Behringer mixer http://www.behringer.com/EN/Products/UB802.aspx
Mic cable http://www.zzounds.com/item--SHUC25J
Shure SM57 mic http://www.shure.com/americas/products/microphones/sm/sm57-instrument-microphone

Now I doubt that I can use that mic since it's made for recording instruments, but what about the other stuff, can I just get the ECM 8000 mic and see if I can use it with what I have?

You've got it right. Just swap out the SM57 (which does not have particularly flat response and is also not very sensitive to even medium intensity sounds) for an ECM 8000 or competitive mic and you are good to go.

I have some practical experience with the UB 802 and because it has tone controls unh equalization knobs, its frequency response can be a little suboptimal, but dealing with that is a refinement, not an initial need.

arnyk is offline  
post #15 of 48 Old 11-07-2012, 06:29 AM
AVS Special Member
 
localhost127's Avatar
 
Join Date: May 2009
Posts: 2,286
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 40
skip the ECM 8000 unless the cal file is offered. dayton (eccm6 emm6 - i'll never get that right) cal can be obtained free of charge from their website (enter serial number; download cal file).
localhost127 is offline  
post #16 of 48 Old 11-07-2012, 06:45 AM
AVS Special Member
 
NicksHitachi's Avatar
 
Join Date: Jul 2007
Location: Wilmington, NC
Posts: 3,721
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 141 Post(s)
Liked: 166

For basic room response measurement I use and recommend the Cross Spectrum Calibrated EMM6.  Just the Basic version is good enough if you don't measure off-axis responses and design crossovers wtih it etc.

 

The stock EMM6 is good and does come with its own cal file although i've heard the cal file can be less than accurate, is it worth another $25-30 for a calibration treaceable to NIST?  Thats up to you.

NicksHitachi is online now  
post #17 of 48 Old 11-08-2012, 09:48 PM - Thread Starter
Advanced Member
 
metallicaband's Avatar
 
Join Date: Feb 2007
Posts: 993
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 98
I'm a bit confused about the whole calibrated mic stuff, I'm not sure if I'll be using those or not, also I don't understand how a cal file can help calibrating a mic, I mean wouldn't the manufacturer just apply it from the get go if that was the case?

The ECM 8000 is being sold for pretty cheap now on Amazon

http://www.amazon.com/gp/product/B000HT4RSA/ref=ox_sc_act_title_1?ie=UTF8&smid=ATVPDKIKX0DER

I'll definitely just get a calibrated version though if that will save me a lot of headache.

KRP-500M | Kuro 9G 5020FD + e-lite mod | B&W 683 Fronts | B&W 685 Center | B&W DM601 S3 Rears | HSU VTF3 MK3 Sub | Yamaha RX-V663 AVR
metallicaband is offline  
post #18 of 48 Old 11-09-2012, 04:04 AM
AVS Special Member
 
NicksHitachi's Avatar
 
Join Date: Jul 2007
Location: Wilmington, NC
Posts: 3,721
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 141 Post(s)
Liked: 166
Quote:
Originally Posted by metallicaband View Post

I'm a bit confused about the whole calibrated mic stuff, I'm not sure if I'll be using those or not, also I don't understand how a cal file can help calibrating a mic, I mean wouldn't the manufacturer just apply it from the get go if that was the case?

The ECM 8000 is being sold for pretty cheap now on Amazon

http://www.amazon.com/gp/product/B000HT4RSA/ref=ox_sc_act_title_1?ie=UTF8&smid=ATVPDKIKX0DER

I'll definitely just get a calibrated version though if that will save me a lot of headache.

Sure, they could at additional cost.

Each mic performs appreciably different and needs custom correction. Most of the deviation occurs in the three highest and lowest octaves from graphs ive seen. If your serious about bass measuremen you MUST have a calibrated mic.

Dont save $20 on a measurement rig and end up with something thats inaccurate.
NicksHitachi is online now  
post #19 of 48 Old 11-09-2012, 06:22 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 762 Post(s)
Liked: 1175
Quote:
Originally Posted by metallicaband View Post

I'm a bit confused about the whole calibrated mic stuff,

Mic calibration is not the most important thing if the mic has reasonably good frequency response.

IMO, Ethan did a real service to the audio community by posting this article comparing a lot of likely measurement microphones:

http://www.realtraps.com/art_microphones.htm

In that article he posts the following comparison of measruements of a real world system as performed using a number of likely mics:



He publishes other info, so you definitely need to read the article.

My summary of the above graph is that from 50 Hz to 10 KHz, which is the range where frequency response is most important, the mics gave results that are so close that as a practical matter, they are interchangeable. This is really pretty amazing for a bunch of more-or-less randomly chosen samples of mics that have such different prices and look so different on their spec sheets.

The background behind this is that about 30 years ago Panasonic started making 1/4" (6 mm) electret mic capsules that sold for $2 and had phenomenally flat frequency response. They have been cloned by many manufacturers, and the cheap measurement mics are basically $2 electret capsules with the makings of a pro mic wrapped around them. The trouble with electet mics is that they are internally charged with a static charge that tends to leak away, unless they are very very carefully made. Their output is proportional to this charge, so if it leaks away the mic becomes less sensitive but tends to keep its frequency response the same.

Even $2,000 lab measurement mics are electret mics under the skin, but they are made super-carefully using expensive materials and put together and checked very carefully by skilled technicans. The $2 ones are made out of cheap materials by a machine. However, cheap materials and machines can be pretty good.

Now you may know why I wince a bit when people start talking about expensive mics and calibrated mics.

On balance I know from practical experience that there are such things as defective mics and mics that become defective while sitting on the shelf. It has happened to me several times. The two most common failures that I have seen is that the mics either stop working entirely or become far less sensitive (IOW like a 10 dB loss) to sound. This will happen whether or not the mics start out being individually calibrated.

If you are a believer in calibration, then the logical approach is to recalibrate them before every use. I know people who do this and if you have the resources, its a good idea. BTW that's the big practical difference between a $50 measurement mic and a $500-1000 measurement mic. The more expensive mic is often far more durable and stable. However, all measurement mics under about $300 are very similar under the skin. If they last they are great but watch out for spontaneous shifts, particularly in sensitivity.

Mic calibrators cost about as much as 5-50 inexpensive microphones, so maybe there is some way to get around using one and still get reliable results.Mic calibrators generally just provide a stable sound source at a single frequency. They are somewhat based on the idea that sensitivity is a decent indicator of the health of a microphone.

To me the above situation makes a point being that if you have say $100-200 to invest, you might be better off getting two nominally identical mics than one calibrated mic, and compare them every time you use them. If are very similar and stay very similar, then they are probably both OK. If one becomes significantly different from the other, then either one or both have become defective. Buy a replacement and compare again and repeat until you again have two mics that are very much the same.
Quote:
I'm not sure if I'll be using those or not, also I don't understand how a cal file can help calibrating a mic, I mean wouldn't the manufacturer just apply it from the get go if that was the case?

The ECM 8000 is being sold for pretty cheap now on Amazon

http://www.amazon.com/gp/product/B000HT4RSA/ref=ox_sc_act_title_1?ie=UTF8&smid=ATVPDKIKX0DER

That is actually about what I paid for most of mine. But in these times that is indeed a competitive price.
Quote:
I'll definitely just get a calibrated version though if that will save me a lot of headache.

It won't.
arnyk is offline  
post #20 of 48 Old 11-09-2012, 06:42 AM
AVS Special Member
 
localhost127's Avatar
 
Join Date: May 2009
Posts: 2,286
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 8 Post(s)
Liked: 40
Quote:
Originally Posted by arnyk View Post

Mic calibration is not the most important thing if the mic has reasonably good frequency response.

the cal file is to account for deviations from the factory per individual unit.

Quote:
Originally Posted by arnyk View Post

IMO, Ethan did a real service to the audio community by posting this article comparing a lot of likely measurement microphones:

not sure how this is relevant to the discussion of cal file - that graph is but merely 1 sample of each of the listed mics. he did not test 10 units of the same mic and overlay...


Quote:
Originally Posted by arnyk View Post

This is really pretty amazing for a bunch of more-or-less randomly chosen samples of mics that have such different prices and look so different on their spec sheets.

see above.

Quote:
Originally Posted by arnyk View Post

Now you may know why I wince a bit when people start talking about expensive mics and calibrated mics.

expensive mics you say? the cited dayton emm-6 can be had for the same cost (or cheaper) than the ecm8000. the reason it was recommended above is because for the same cost you are also gaining access to the cal file for that particular unit (vs generic cal file) at no charge.
localhost127 is offline  
post #21 of 48 Old 11-09-2012, 06:56 AM
AVS Special Member
 
NicksHitachi's Avatar
 
Join Date: Jul 2007
Location: Wilmington, NC
Posts: 3,721
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 141 Post(s)
Liked: 166

What Arny is saying is correct and a good read over at Ethan's site.  That data however represents small sample sets(1mic each) and does nothing for inter-unit variability.

 

Yes a "generic" cal file will get you within a few dbs one way or another, I however fret over a few dbs, especially in the bass region.

 

Ethan's testing does not address inter-unit variability(since they only tested one) which is the only benefit IMO for calibration.  A german "lab" posted an article(cited by Cross Spectrum) testing 55 Behringer ECM8000s.  Sensitivity differences at the extremes are the main offenders but also pay attention to the scale on the graph which is different.  Typical variability appears centered around the 5db point so if thats good enough for you, then go for a generic cal file.

 

This is the article.

 

Here's the graph from article posted on CS labs site.  The german image wouldn't past for me.

 

 

ecm8000_frequency_response_large.jpg

I'm not trying to come off fanboyish for cal'ed mics.  For 99.9% of folks measuring home setups the generic cal files are "good enough."   Furthermore one would never know the difference.

NicksHitachi is online now  
post #22 of 48 Old 11-09-2012, 07:14 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 762 Post(s)
Liked: 1175
Quote:
Originally Posted by localhost127 View Post

Quote:
Originally Posted by arnyk View Post

Mic calibration is not the most important thing if the mic has reasonably good frequency response.

the cal file is to account for deviations from the factory per individual unit.

Of course!
Quote:
Quote:
Originally Posted by arnyk View Post

IMO, Ethan did a real service to the audio community by posting this article comparing a lot of likely measurement microphones:

not sure how this is relevant to the discussion of cal file - that graph is but merely 1 sample of each of the listed mics. he did not test 10 units of the same mic and overlay...

You've missed the point. If random samples of that many different mics come out so close, its safe to assume that random samples of any of them will be the same. I've done this experiment with several kinds of cheap measurement mics and that is indeed what happens.


Quote:
Quote:
Originally Posted by arnyk View Post

This is really pretty amazing for a bunch of more-or-less randomly chosen samples of mics that have such different prices and look so different on their spec sheets.

see above.

Been there, done that both decades ago and in just the past few years.
Quote:
Quote:
Originally Posted by arnyk View Post

Now you may know why I wince a bit when people start talking about expensive mics and calibrated mics.

expensive mics you say? the cited dayton emm-6 can be had for the same cost (or cheaper) than the ecm8000. the reason it was recommended above is because for the same cost you are also gaining access to the cal file for that particular unit (vs generic cal file) at no charge.

Again you keep missing the point that other than obviously broken samples, these mics are about as alike as adjacent peas in a pod. Many of them probably come out of the same factory, many that come out of different factories use the same capsule. Even when the capsules vary from the original Panasonic from 30 years ago, to Chinese clones made last week, they are still very much the same for practical purposes.

Also, just because a sheet of paper looks like it was individually made, doesn't mean that it contains any especially useful wisdom. This is particularly true given that the most likely failure is that two years from now, maybe 2 out of 10 samples of any of them will have become significantly less sensitive.

Also," expensive, you say?". You obviously don't know what the Earthworks QTC01 and Joshephson C612 cost. ;-)

QTC-1 price info (mic has been renamed the QTC-40) http://www.frontendaudio.com/Earthworks-QTC1-QTC40-Ultra-Quiet-Omnidirectiona-p/1696.htm

Most people I know who use Earthworks mics for measurement use the QTC-30 which is a bit easier on the wallet.

Josephson C617 price info: http://soniccircus.com/Josephson-C617

And not mentioned but probably one of the gold standards in this area:

DPA 4006 - $1795 street price.

For a real thrill, mount a 4006 and a ECM 8000 side by side, capsule to capsule and observe that all differences fall within the calibration curve for the 4006. Been there, done that.

Of course beyond these, there are the truly expensive measurement mics! ;-)
arnyk is offline  
post #23 of 48 Old 11-09-2012, 07:37 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 762 Post(s)
Liked: 1175
Quote:
Originally Posted by NicksHitachi View Post

What Arny is saying is correct and a good read over at Ethan's site.  That data however represents small sample sets(1mic each) and does nothing for inter-unit variability.

You also seem to be somewhat missing the point. When so many samples from different sources are so close, how much sample variation can you reasonably expect in any of them.

You kindly shed massive light on that question:
Quote:
ecm8000_frequency_response_large.jpg
I'm not trying to come off fanboyish for cal'ed mics.  For 99.9% of folks measuring home setups the generic cal files are "good enough."   Furthermore one would never know the difference.

I analyze that as follows:



What I see is that if you picked the curve for the average microphone, you would be within 1.5 dB of it from 50 Hz to 10 KHz for 98% of the 85 samples.

With acoustical measurements, +/-1.5 dB is about as good as perfect.

If you just assumed the mics were flat, no big deal.

Ever move a mic a few inches and see what effect that has? ;-)

Ever come back and repeat a measurement a half hour later? ;-)

Get things right from 50 Hz to 10 KHz, and in the process things will very likely be fine for an octave either way beyond that.

Given the sensitivity of the ear for small differences, it won't matter anyway.

BTW, thanks to Cross System Labs for their very believable and valuable evidence.
arnyk is offline  
post #24 of 48 Old 11-09-2012, 08:01 AM
AVS Special Member
 
NicksHitachi's Avatar
 
Join Date: Jul 2007
Location: Wilmington, NC
Posts: 3,721
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 141 Post(s)
Liked: 166
Quote:
Originally Posted by arnyk View Post

You also seem to be somewhat missing the point. When so many samples from different sources are so close, how much sample variation can you reasonably expect in any of them.

 

 

I'm not sure I understand what your asking....  I don't "expect" or "assume" anything.  I also dont extrapolate anything from a sample set of one.
 

 

I analyze that as follows:

 

You conveniently cut off the regions of the highest variability in your "analysis."  If 50Hz and 10Khz is good enough for you, thats fine.  Like I said several posts up the highest variability occurs at the extremes.  I measure 5-20K so those extremes are important to me.  A 5db hump in my bass makes my eye twitch as does your chopping this graph to suit your fancy.



What I see is that if you picked the curve for the average microphone, you would be within 1.5 dB of it from 50 Hz to 10 KHz for 98% of the 85 samples.

 

Really though who only measures 50-10K?

With acoustical measurements, +/-1.5 dB is about as good as perfect.

 

AgreeWithin the variation introduced by measurement error.  However error is additive 50% of the time assuming equal probability of error in both directions.  If you start with 5db calibration error and have 1.5db measurement error in the same direction it compounds 50%(eq. prob.) of the time to 6.5db.

If you just assumed the mics were flat, no big deal.

Ever move a mic a few inches and see what effect that has? ;-)

Ever come back and repeat a measurement a half hour later? ;-)

Get things right from 50 Hz to 10 KHz, and in the process things will very likely be fine for an octave either way beyond that.

 

Disagree, but thats ok too.

Given the sensitivity of the ear for small differences, it won't matter anyway.

BTW, thanks to Cross System Labs for their very believable and valuable evidence.

NicksHitachi is online now  
post #25 of 48 Old 11-09-2012, 08:24 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 762 Post(s)
Liked: 1175
Quote:
Originally Posted by NicksHitachi View Post

Quote:
Originally Posted by arnyk View Post

You also seem to be somewhat missing the point. When so many samples from different sources are so close, how much sample variation can you reasonably expect in any of them.


I'm not sure I understand what your asking....  I don't "expect" or "assume" anything.  I also dont extrapolate anything from a sample set of one.

I'm not asking you to extrapolate from just one.

 
Quote:

You conveniently cut off the regions of the highest variability in your "analysis." 

No, I weighted frequencies by the sensitivity of the human ear to frequency response differences. They may look the same to you, but that might be due to your lack of appreciation for this issue.

Quote:
If 50Hz and 10Khz is good enough for you, thats fine.

That looks like being dismissive to me. What I see is dismissal of a very important issue - the sensitivity of the human ear to frequency response differences which is well known to be strongly frequency dependent.
Quote:
  Like I said several posts up the highest variability occurs at the extremes.

Like I said and you seem to be continually ignoring, you've got to weight frequency response differences by frequency if you want an accurate view of the signficiance of the differences.

If you think that the ear is equally sensitive to frequency response differences at all frequencies, then that is simply wrong. If you think that we should worry about all dB differences regardless of frequency, then you are ignoring at least 50 years of audio wisdom and accepted practice.
Quote:
I measure 5-20K so those extremes are important to me.

You can be as unreasonable as you wish to be as long as you don't get sarcastic in public about people who are being reasonable. ;-)
Quote:
A 5db hump in my bass makes my eye twitch as does your chopping this graph to suit your fancy.

Bass is commonly thought to be the range below maybe 100 Hz. it is excruciatingly well known that the ear's sensitivity to both sound at all, and changes in sound intensity varies tremendously over this range. I suspect that your eyes don't twitch for 5 dB humps at 5 Hz if you do bias controlled evaluations.

However, looking at frequency response curves with 5 dB variations at 5 Hz may cause your eye to twitch. It can be cured! ;-)

All you need to do is apply a little generally accepted practice that might not be aware of yet...
arnyk is offline  
post #26 of 48 Old 11-09-2012, 09:04 AM
AVS Special Member
 
NicksHitachi's Avatar
 
Join Date: Jul 2007
Location: Wilmington, NC
Posts: 3,721
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 141 Post(s)
Liked: 166

Dude, I gotcha and your not keying me in anything I'm not aware.  Equal loudness curves and frequency dependent hearing sensitivity are a little out in left field IMO.

 

No, we don't hear very well at the extremes nor register minute differences as well in those regions, which is why I measure.  If audible differences are all that matters why not just tune by ear and not measure at all?

 

Fact:  The mics exhibit significant(IMO) variability within whats considered the human audible range, allthough the majority occurs at the extremes.

 

I consider it worth the extra $20 for the extra accuracy.

 

Arny don't, NBD......

 

Have a nice day.

NicksHitachi is online now  
post #27 of 48 Old 11-09-2012, 10:47 AM
AVS Special Member
 
DS-21's Avatar
 
Join Date: Jun 2005
Posts: 3,553
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 208 Post(s)
Liked: 244
To answer the OP's question, FuzzMeasure, a USB mike preamp, a measurement mike, and a tripod.
Quote:
Originally Posted by NicksHitachi View Post

For basic room response measurement I use and recommend the Cross Spectrum Calibrated EMM6.  Just the Basic version is good enough if you don't measure off-axis responses and design crossovers wtih it etc.

The stock EMM6 is good and does come with its own cal file although i've heard the cal file can be less than accurate, is it worth another $25-30 for a calibration treaceable to NIST?  Thats up to you.

FWIW, I have a comparison of my CSL calibrated EMM-6 in both on-axis and grazing incidence with the factory curve here. Here, FWIW, were my concluding thoughts:

The only reasonable answer to [the question of whether or not to get a mike individually calibrated] is, "it depends."

IF your primary use is going to be optimizing the response of your system in the modal region (hopefully using multiple subwoofers, and possibly EQ on top) then the factory calibration is probably adequate. All of them look similar enough through the lower midrange for the "differences" to boil down to experimental error.

If you want to do in-room measurements, then it is probably a good idea to get the mike calibrated for grazing incidence, because otherwise one could think her/his system doesn't have enough highs and tune it to be ear-piercing.

Also, if you're using it for crossover design work in mains speakers, I think calibration is more than warranted.

Another way to look at it is, an individualized professional calibration is cheap insurance. It does not cost much to get a microphone calibrated; well under $100 at the time of this writing. If one is choosing between doing something that could be beneficial, such as getting a measurement mike calibrated, and something stupid, like buying "high end" AC power wires, then get the mike calibrated!

I certainly don't regret getting it done.

--
"In many cases there aren’t two sides unless one side is 'reality' and the other is 'nonsense.'" - Phil Plait
Serious Audio Blog 
Multichannel music (and video) urban loft living room system 
DS-21 is offline  
post #28 of 48 Old 11-09-2012, 11:20 AM
AVS Special Member
 
Ethan Winer's Avatar
 
Join Date: Apr 2003
Location: New Milford, CT, USA
Posts: 5,748
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 134
Quote:
Originally Posted by arnyk View Post

If random samples of that many different mics come out so close, its safe to assume that random samples of any of them will be the same.

Exactly. More to the point, even with a perfect microphone what you measure will be an approximation of the actual response at your ears. Especially at higher frequencies where speakers become directional. The main reason people measure their rooms is to assess acoustic problems, and to compare the difference after adding acoustic treatment. Even a coarse measurement is "good enough" for that, and for comparison all that matters is the relative change.

--Ethan

RealTraps - The acoustic treatment experts
Ethan's Audio Expert book

Ethan Winer is offline  
post #29 of 48 Old 11-09-2012, 11:21 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 762 Post(s)
Liked: 1175
Quote:
Originally Posted by NicksHitachi View Post

Dude, I gotcha and your not keying me in anything I'm not aware.  Equal loudness curves and frequency dependent hearing sensitivity are a little out in left field IMO.

Equal loudness a little out in left field?

Wow!

So you're dubious about Fletcher and Munson?

Where I come from (a realm often chastised by golden ears for their conservatism) they are motherhood and apple pie. I hate to think what kinda space cadets you think that Zwicker and Fastl are. ;-)
Quote:
No, we don't hear very well at the extremes nor register minute differences as well in those regions, which is why I measure.

News flash guy, we don't even hear large differences at the frequency extremes.



The way to interpret the curves above is as follows:

Let's take the octave from 15 to 30 Hz which is centered around 20 Hz more or less. If you increase or decrease response in that band by 4.5 dB, there is no chance that anybody will hear a difference presuming low distortion and clean test signals. In fact it takes a difference of several times that (maybe 9-15 dB) for that kind of difference @ 20 Hz to be readily heard. I've actually done this experiment with a parametric equalizer and a very large clean subwoofer, and it works out.

Of course the equal loudness curves say that a clean 20 Hz tone has to be at something like 75 dB SPL to cross the threshold of hearing. That also works out in real life. The big caveat is having a clean enough subwoofer because any harmonics that are generated will vastly change the outcome.
Quote:
If audible differences are all that matters why not just tune by ear and not measure at all?

Same reason I use a very nice ruler when I do woodwork that is supposed to be better than rough carpentry. Measuring is faster and more accurate than trying to do it by eye. If I want the errors to be invisible, I have to use something more accurate than "by eye" when I do the actual construction. But, that isn't an excuse to obsess over cutting wood within ten thousandths of an inch on my table saw.
Quote:
Fact:  The mics exhibit significant(IMO) variability within whats considered the human audible range, allthough the majority occurs at the extremes.



Fact is that the 8-15 dB it takes for a just perceptible change in the octave @ 20 Hz is something like twice the actual variation in the 85 mics tested above. The 85 ECM 8000 mics tested were on the average something like 3 dB down at 20 Hz, and that's half of what we know for sure is a completely inaudible change.
Quote:
I consider it worth the extra $20 for the extra accuracy.

Since it is your project and your money, that's a choice that is clearly up to you. ;-)

But if we are making generalities for lots of people including newbies, we might as well tell it like it is.
arnyk is offline  
post #30 of 48 Old 11-09-2012, 11:27 AM
AVS Special Member
 
Ethan Winer's Avatar
 
Join Date: Apr 2003
Location: New Milford, CT, USA
Posts: 5,748
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 134
Quote:
Originally Posted by arnyk View Post

we don't even hear large differences at the frequency extremes ... I've actually done this experiment with a parametric equalizer and a very large clean subwoofer, and it works out.

This is a great point. I've never tested this myself to pin down dB levels, but I've noticed that at very high frequencies it's impossible to hear level differences even when they're 3 dB or more. Compared to 1-2 KHz where it's easy to hear changes less than 1 dB. At 10 KHz, just moving your head half an inch can change the level by 5-10 dB.

--Ethan

RealTraps - The acoustic treatment experts
Ethan's Audio Expert book

Ethan Winer is offline  
Reply Audio Theory, Setup, and Chat



Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off