Quantcast
FFT Analyzers not relevant? - Gearspace.com
The No.1 Website for Pro Audio
FFT Analyzers not relevant?
Old 28th January 2009
  #1
Gear Addict
 
laboso's Avatar
 
🎧 10 years
FFT Analyzers not relevant?

hey guys, i read an article about metering and voltages in audio last night in which it was said that the Fast Fourier Analysis is not relevant enough as a visible monitoring tool and that especially software analyzers cant compete with hardware analyzers which can get really expensive very soon if you consider buying a relative good one.. (he explained why but i didnt really got it!) my question to the mastering engineers is if this really applies and if i could go on relying on my Pinguin ! i know in the end you have to let your ears decide but i cant live without my FFT analyzer especially when im studying music or if im doing a little mastering job for a friend. thanks in advance!
Old 28th January 2009
  #2
Lives for gear
 
MAzevedo's Avatar
 
🎧 10 years
I've been mastering for 10 years, and I've never found situation where an analyzer gave me something useful enough to work from. RTAs don't have the resolution to tell you if the midbass is a half dB hot, or if the guitar's midrange presence is stepping on the vocal clarity.

Excellent monitors and well-trained ears are all you need. I've got various metering to idiot-check myself, but 98% of the decisions are purely based on listening.
Old 28th January 2009
  #3
Lives for gear
 
Alexey Lukin's Avatar
 
Verified Member
🎧 10 years
Quote:
Originally Posted by laboso ➑️
software analyzers cant compete with hardware analyzers
I don't think it's true. Have you got the source?
Old 28th January 2009 | Show parent
  #4
Gear Guru
 
Ethan Winer's Avatar
 
🎧 15 years
Quote:
Originally Posted by MAzevedo ➑️
RTAs don't have the resolution to tell you if the midbass is a half dB hot, or if the guitar's midrange presence is stepping on the vocal clarity.
Great words of wisdom there.

--Ethan
Old 28th January 2009 | Show parent
  #5
Lives for gear
 
wado1942's Avatar
 
🎧 10 years
My father told me about a college course he took on sound back in the 70s. One day, his teacher put up a bunch of charts and graphs which analyzed a recording of a musician playing a certain brand of instrument. He then put up similar charts of the same musician playing the same note on another brand of instrument. They spent the whole hour looking at the minute differences between both the recordings, noting a tiny little peak is higher here than on the other recording etc. At the end of the hour, the teacher played the samples. One was a violin and the other was an oboe.

The purpose of his lesson was that the best analysis instrument is your EARS. YOU have to find out what is what, the electronics can only HELP you understand why. So I'd put aside the FFT and do a ton of listening on your monitors. Your ears can detect a +1dB spike at 55Hz right next to a -1dB dip at 53Hz. I don't think any FFT can do that and even if it could, the chart would be very difficult to read.
Old 28th January 2009 | Show parent
  #6
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by MAzevedo ➑️
I've been mastering for 10 years, and I've never found situation where an analyzer gave me something useful enough to work from. RTAs don't have the resolution to tell you if the midbass is a half dB hot, or if the guitar's midrange presence is stepping on the vocal clarity.
I've been mastering for almost 20 years and engineering for more than 26 and I can tell you that yes, you can. When you have trained your eyes for many years to a sample accurate stereo RTA yes, you CAN. But, you need to confirm it with your ears as well. It's a 50/50 kinda deal {you listen, you watch}.

Quote:
Originally Posted by Ethan Winer ➑️
Great words of wisdom there.
Ethan, I showed you once mine and you were baffled by it. You didn't even know such software existed. Remember?

Best regards,
Old 28th January 2009 | Show parent
  #7
Lives for gear
 
MAzevedo's Avatar
 
🎧 10 years
If you could Edward, post me a screenshot that shows that a guitar is obscuring a vocal. If possible, maybe an image that shows a large bump at 80 Hz that is OK because the bass is very dark and needs that much low-end level to properly balance with the other elements of the mix. Maybe a picture of a high end with an average amount of energy, but made up of too-bright cymbals and too-dark everything else.

All these are spectral problems, and I've never seen an RTA or anything else that can visualize these simple things that can be easily heard.
Old 28th January 2009 | Show parent
  #8
Gear Guru
 
Ethan Winer's Avatar
 
🎧 15 years
Quote:
Originally Posted by Edward_Vinatea ➑️
Ethan, I showed you once mine and you were baffled by it. You didn't even know such software existed. Remember?
I do not remember. Showed me your what? Please remind me.
Old 28th January 2009 | Show parent
  #9
Lives for gear
 
Masterer's Avatar
 
Verified Member
🎧 15 years
I'll show you mine Ethan. It'll baffled ya alright.
Old 28th January 2009 | Show parent
  #10
Lives for gear
 
MAzevedo's Avatar
 
🎧 10 years
Quote:
Originally Posted by Masterer ➑️
I'll show you mine Ethan. It'll baffled ya alright.
I was very specifically not going there after reading that exchange. You folks can show each other whatever you'd like, I am going to go back to listening.
Old 28th January 2009 | Show parent
  #11
Gear Nut
 
🎧 15 years
Damm...another RTA discussion...I'm selling my monitors...going RTA and ipod earbuds! My second monitoring system will be my iphone speakers... What did anyone do before they could see the music? Recordings have become noticeably worse since then.
Old 28th January 2009 | Show parent
  #12
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by MAzevedo ➑️
If you could Edward, post me a screenshot that shows that a guitar is obscuring a vocal. If possible, maybe an image that shows a large bump at 80 Hz that is OK because the bass is very dark and needs that much low-end level to properly balance with the other elements of the mix.
Quote:
or if the guitar's midrange presence is stepping on the vocal clarity...post me a screenshot that shows that a guitar is obscuring a vocal. If possible, maybe an image that shows a large bump at 80 Hz
Honestly Mr. Azevedo, I don't see how anybody can demonstrate that. When you said that the vocal was being "obscured" by an electric guitar, I actually thought you were referring to regular electric guitars where sometimes the guitar is so loud that takes over the lead vocal {depending on the singer's vocal range 'course} As per your comments, can you clarify the quote right above? I don't get it.
Quote:
Maybe a picture of a high end with an average amount of energy, but made up of too-bright cymbals and too-dark everything else.
Actually, let's do it the other way around, upload a track with "bright cymbals" or even better, loud tambourines and I will make you a screen shot at my earliest convenience. Thank you.
Quote:
All these are spectral problems, and I've never seen an RTA or anything else that can visualize these simple things that can be easily heard.
Actually, it's not a spectral problem but it would be more like a mix problem in that case. The RTA is merely interpreting the frequencies which usually bears a correlation with the "problem".


Quote:
Originally Posted by Ethan Winer ➑️
I do not remember. Showed me your what? Please remind me.
OK Ethan. Sorry that you forgot. I have condensed our conversation which you appeared unable to answer questions regarding the use of certain spectrum analyzers and where you mentioned to favor scopes. So, I hope it ends here. And to other members, the picture only shows the software doing analysis in mono mode {L+R Summed} and not in stereo. There are two sections missing: The phase scope and the sample accurate reference meters.




Quote:
Edward Vinatea: I just want to make sure that you understand what it does; It displays all 25 bands' amplitudes in real time both in mono or stereo, it has peak markers to show the highest amplitude. This is just a slice of time or 1/1024ms. So - in this 'variable state' context, you say that the term "transients" is not applicable to these frequency movements? {On each band from the magnitude axis}. Please, correct me if I am wrong


Ethan Winer: You should be asking this in the Geekslutz section at Gearslutz, and maybe PM or email Andreas Nordenstam and ask him to reply. He knows much more about this stuff than I do! I think Andreas goes by the user name Lupo atGearslutz.
Old 29th January 2009 | Show parent
  #13
Lives for gear
 
macc's Avatar
 
Verified Member
3 Reviews written
🎧 10 years
Ed, have you got email notifications on for any time 'RTA' is posted? How did you do it?
Old 29th January 2009 | Show parent
  #14
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by macc ➑️
Ed, have you got email notifications on for any time 'RTA' is posted? How did you do it?
I just stopped by for Nicky and to thank folks for the info at the baking tape thread and there it was. I now regret clicking on it and replying though. But you know me, Macc, you speak ill about RTA's and I will debate you. heh

Take care I really got to go.
Old 29th January 2009 | Show parent
  #15
Lives for gear
 
wado1942's Avatar
 
🎧 10 years
Quote:
I'm selling my monitors...going RTA and ipod earbuds!
A while back, an aquaintance sent me a 16-track song for me to mix using only a VU meter plugged into a headphone amp and an FFT window. No monitors, no headphones, never heard a single note of a single track till after I uploaded it to my website. The results were somewhat interesting.
Old 29th January 2009 | Show parent
  #16
Lives for gear
 
macc's Avatar
 
Verified Member
3 Reviews written
🎧 10 years
I'd love to hear that heh
Old 29th January 2009 | Show parent
  #17
Gear Maniac
 
fader8's Avatar
 
🎧 10 years
What are you guys on about? Spectrum Analyzers are great if you're going for FCC approval in the 2.4 GHz ISM band.

And btw, if anyone needs this certification for their masters, I'm your man!
Old 29th January 2009 | Show parent
  #18
Lives for gear
 
🎧 10 years
I have noticed there are allot of folks who treat this issue allmost like the creationalism/Evolution debate. Do they have to be mutually exclusive???? Is'nt there room , somehow for both theory's to have merit?? If you stumble on a great sound, why would'nt you use analytical tools to increase your chances of repeating it ??

What's wrong w/ having some EAR/EYE coordination????

Can you rely soley on those golden ears????stike

( make no mistake , 90% of the mastering mafia around here is going to say yes!! things can't get done without me and my golden ears !! The rest of you are mere mortals , and a FFT anylizer can't save your tin ears !!!

( followed by a meniacle laugh , no ???heh)

Sure , the ears are always going to have the final word, and you don't want to get mesmerized in a RTA read out, But you all should'nt be attempting to completely and utterly marginalize what can be a valuable tool, especially to newbies. You know they can learn faster with them, make better decisions in the short term, and then leave them behind.


Here is what the Gear Sluts Mastering forum Mantra basically when ever this subject comes up;

Yes, immeadiate visual feedback is entirely useless. All of you will also be required to disable any RTA'S and, in addition, the graphic display of your equalizers, compressor reduction meters, transfer plot displays, and master fader meters, ect. , ect.. there all a crutch, and you don't need them; witness the total domination of the audio industry by the blind. they do it all with there ears only. so must you!! All of you are incapable of ear/eye coordination. turn off all graphic displays, NOW!


To the OP ; Betya were'nt expecting a **** storm over an innocent little question like this , Huh !??


Unfortuneatly , things around here are very polarized around here on this subject . Get one , read all you can about FFT, read the manuals , make up your own mind.

Cheers
Old 29th January 2009 | Show parent
  #19
Lives for gear
 
Virtalahde's Avatar
 
Verified Member
13 Reviews written
🎧 15 years
Quote:
Originally Posted by flatfinger ➑️
If you stumble on a great sound, why would'nt you use analytical tools to increase your chances of repeating it ??
I once mastered something someone had already tried to work on. I heard the guy had used transparent sheets to draw the FFT curve on, straight from the screen. The he tried repeating it to the other songs. It was told to sound pretty bad.

Then I worked on it, and to tell you the truth, I only used my golden ears since there was nothing wrong with the mixes to begin with.

FFT's just don't tell me anything. Good monitoring does.
Old 30th January 2009 | Show parent
  #20
Lives for gear
 
🎧 10 years
Some members would like to associate serious RTA analysis with HardBall stuff {HB in short, get my meaning?} That's not what we're talking about. If you're versed reading a RTA, you will be able to determine a whole lot of things. Naturally, it requires a good sense {knowledge} of the material at hand. For that you do need your ears! Rule of thumb: always compare apples to apples and pears to pears. As I always say, it's a 50/50 kinda deal {you listen, you watch}.

Regards,
Old 30th January 2009 | Show parent
  #21
Lives for gear
 
Franco's Avatar
 
Verified Member
🎧 15 years
I've been mastering for 8 years and...you seasoned vets & pros can stop reading right now, lol.

For the rest of you, I too use mostly my ears and also visual tools. No kidding, I just finished mastering a song right now and used my FFT to help me find this bit of harshness in the high frequencies that seemed to be living around 8-9.9k. I used some filters to get rid of it to the best of my abilities, but some of it was still there. I looked at my FFT, zoomed in around the area I suspected, noticed a spike, zeroed in and it told me that the spike was @ 10,400 Hz. I got another EQ and using my ears, I dipped a little bit around that range and adjusted the Q using my ears and now that annoying, harshness is gone.

I mean, I could have used just my ears and not bothered looking at the FFT, but I want to knock out a few more tracks on this project by the end of the night, and this tool helped me zero in on the problem faster.
Old 30th January 2009 | Show parent
  #22
Lives for gear
 
MASSIVE Master's Avatar
 
Verified Member
🎧 15 years
Quote:
FFT's just don't tell me anything. Good monitoring does.
Word to the [SELF-CENSORED]-ing peeps. "Listening" is an "auditory" thing. When I'm listening to a CD through my system, I'm hearing whatever I'd be hearing. I know some people think there's some "special talent" to "reading" analyzers, but that's for another thread. Listen. You want to look? Look. But your ears decide.

Quote:
I mean, I could have used just my ears and not bothered looking at the FFT, but I want to knock out a few more tracks on this project by the end of the night, and this tool helped me zero in on the problem faster.
See, that's totally different though. You *knew from listening* that there was something "harsh" in there somewhere. You used the analyzer as a simple, valid tool to find out the specific frequency. No one is going to argue that. But you didn't need the RTA to tell you something sounded harsh.

I use a FFT EQ when I head VDO whine. By examining the RTA, I can easily "see" whether it's at 15.875k, 15.750k, 15.625k, etc. But I don't need the FFT to tell me there's a whine... And if it gets past the production team on whatever disc is in my CD player, I don't need a RTA to tell me that the whine is driving me slowly towards madness.
Old 30th January 2009 | Show parent
  #23
Gear Addict
 
laboso's Avatar
 
🎧 10 years
well, thank y'all for your answers! especially flatfinger, yes, wouldnt expect this heh
Old 30th January 2009 | Show parent
  #24
Gear Addict
 
laboso's Avatar
 
🎧 10 years
mastering engineers act like divas
Old 30th January 2009 | Show parent
  #25
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by laboso ➑️
well, thank y'all for your answers! especially flatfinger, yes, wouldnt expect this heh
Just don't forget that some members' ignorance on the subject comes from - per your excellent perception about this thread - a place of vanity. It's foolish and totally unprofessional to show such passion against a little device that when 'mastered', only helps you achieve greater music mastering results. To believe that RTAs are useless and they should be able to read/display sound textures is as absurd as stating that you can master music without hearing. Doing both things {watching and hearing} is not different than using your vision to check on reference levels, knobs and fader positions, etc, etc. FWIW.

Regards,
Old 31st January 2009 | Show parent
  #26
Gear Maniac
 
fader8's Avatar
 
🎧 10 years
Quote:
Originally Posted by Edward_Vinatea ➑️
Just don't forget that some members' ignorance on the subject comes from - per your excellent perception about this thread - a place of vanity. It's foolish and totally unprofessional to show such passion against a little device that when 'mastered', only helps you achieve greater music mastering results. To believe that RTAs are useless and they should be able to read/display sound textures is as absurd as stating that you can master music without hearing. Doing both things {watching and hearing} is not different than using your vision to check on reference levels, knobs and fader positions, etc, etc. FWIW.
Edward,
In your defense of the FFT RTA, you've neglected to once mention what you're actually using it for. Yes, it's a useful tool, but by definition it's usefulness is limited to its settings. For temporal vs. amplitude analysis, the device is only useful if we sacrifice frequency resolution. At that point we are just as well off using a peak meter and our ears. For high resolution frequency analysis, the device then becomes temporally blind and loses amplitude accuracy. While that may be useful for diagnosing bearing noises on a motor, or looking for RF sidebands and other steady state signals, it helps us little for music where the spectrum constantly changes.

I think you'd agree that in the mixing and mastering of music, time, energy and frequency are inexorably inter-related, (recall the previous discussion we all had about transients). What makes our ears so unique as measurement instruments is that we can process these 3 dimensions all at once, at very high speed. The RTA remains terrible at that.

It isn't that the RTA isn't useful. It's that there are usually better tools available that allow us to do the analysis and still keep those dimensions in context. Even sweeping an EQ peak, or doing the same with a dynamics sidechain filter, lets us see and hear with better precision in all three domains, simultaneously, than an RTA ever could. I hope you can see my point.
Old 31st January 2009 | Show parent
  #27
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by fader8 ➑️
Edward,
In your defense of the FFT RTA, you've neglected to once mention what you're actually using it for. Yes, it's a useful tool, but by definition it's usefulness is limited to its settings. For temporal vs. amplitude analysis, the device is only useful if we sacrifice frequency resolution. At that point we are just as well off using a peak meter and our ears. For high resolution frequency analysis, the device then becomes temporally blind and loses amplitude accuracy. While that may be useful for diagnosing bearing noises on a motor, or looking for RF sidebands and other steady state signals, it helps us little for music where the spectrum constantly changes.

I think you'd agree that in the mixing and mastering of music, time, energy and frequency are inexorably inter-related, (recall the previous discussion we all had about transients). What makes our ears so unique as measurement instruments is that we can process these 3 dimensions all at once, at very high speed. The RTA remains terrible at that.

It isn't that the RTA isn't useful. It's that there are usually better tools available that allow us to do the analysis and still keep those dimensions in context. Even sweeping an EQ peak, or doing the same with a dynamics sidechain filter, lets us see and hear with better precision in all three domains, simultaneously, than an RTA ever could. I hope you can see my point.
Sorry, but again, I don't agree with anything you've said. Yes I can, yes It can, Yes, We can {I am beginning to sound like Obama here heh}. Do yourself a favor and use a professional software version like the AD-1 above. It will blow your mind. If you want to have a serious discussion about RTA technologies and using RTA with me for that matter, do the following:

Because this topic seems to be a very controversial one at GS and I want to avoid getting into arguments with strangers, update your account and write something about yourself on your bio section, put a sig file through the account edit settings and make sure that I see your real name. I hope this is acceptable to you otherwise, I will have to put you on my ignore list. And Fader8, I am saying this as respectfully as I can and it will apply to all anonymous users, so please don't take it personal.

Regards,
Old 31st January 2009 | Show parent
  #28
Lives for gear
 
Sunbreak Music's Avatar
 
Verified Member
8 Reviews written
🎧 15 years
You do not want to get into a technical argument w/ Fader8.
Old 31st January 2009 | Show parent
  #29
Gear Maniac
 
fader8's Avatar
 
🎧 10 years
Quote:
Originally Posted by Sunbreak Music ➑️
You do not want to get into a technical argument w/ Fader8.
LOL, I promise, I'll be nice. Not in the mood anyway!

Quote:
Originally Posted by Edward_Vinatea ➑️
Sorry, but again, I don't agree with anything you've said.
When was the first time?

Please feel free to refute anything I said.

Quote:
Originally Posted by Edward_Vinatea ➑️
Because this topic seems to be a very controversial one at GS . . .
I don't see it as controversial. Just that people have various opinions on the value of such things as a regular tool in mastering audio. After making a joke about it previously, (I can't help myself) I thought I'd share why I see little value in them. My question to you, as you seem to think otherwise, is just exactly what is it you're using it for?

I show you mine, you show me yours, that's how it works.

You're right about the profile thing. Updated, as requested. Not much room there to write much, is there?
Old 1st February 2009 | Show parent
  #30
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by fader8 ➑️
While that may be useful for diagnosing bearing noises on a motor, or looking for RF sidebands and other steady state signals, it helps us little for music where the spectrum constantly changes.

Nobody is advocating using this paticular tool in an exclusive manner!!!!

I'm certianly never going to advocate leaving the ears out of the loop when making decisions about AUDIO!!!


OK??

Now on to that remark... (quoted above)


Oh yes , that sounds so reasonable and intuitive! Machines and music are so very different ; what a reasonable analogy!!
I can see why "Sunbreak Music " says to watch yer butts!!
Of course we should because
when you start to use spurrious analogy and logical falacy like
that , it's going to be like getting Bill Clinton to explain to us
all what IS "IS"!!
Music (unless it's some inaccessible, post modernistic ,zero
cadence , "sophiticated" cultutral realitivistic tripe ) Has a
Beat!!! You said it , the SPECTRUM constantly changes!! Music
however , has a periodical predictableness in spades!! , When I
increase the block size to look at lower frequncies in greater
detail , I'm doing so to deal with a spectral problem , I'm not
going to make changes in the temporal content based on what I see
in the FFT at that point , I'm looking for spectral anomillies
ONLY .
If I'm going to do something temporially , then I'll do it by ear
and use the beat and pulse , the meter and tempo to do so . And
since these can be chunked down into small , repetitive units ,
and , especially since you usually give up temporal accuracy in
the FFT to gain spectral accuracy in the BASS registers , I'm NOT
GOING to have to be chasing allot of accents or wild timing
anyways since , most pop bass playing is locked straight to the
beat (kick drum , anyone ??) and is reinforcing the harmony with
chord roots and 3rds and fifths , It's not that hard .!!
Look , why don't we just let this guy try it out for himself ??
The more he studies the theory behing FFT , the more success he
will enjoy . I hope your just having one of those liberal "save
this guy from himself " moments , because god knows this stuff can
run circles around you and leave you chasing your tail easily
enough .
Why is it so important to make him think there can be no
usefullness AT All from this paticular tool !!! Let him give a go
for himself!!!
Mr. OP , I TOLD YOU SO !!

Divas, egotistical Divas!!
Let this thread die already

Edward , don't waste your valuable time , This guys already
whipping out the arsenal of college debating tactics ( maybe took
a bit of law school to!) that take the simplist of things and make
them into quadratic [email protected] equasions!!
Closed

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 56 views: 7054
Avatar for Reptil
Reptil 16th April 2005
replies: 700 views: 248238
Avatar for seby
seby 57 minutes ago
replies: 16 views: 5785
Avatar for electricsky
electricsky 16th February 2012
replies: 76 views: 12841
Avatar for scandal
scandal 7th March 2021

Forum Jump
Forum Jump