[Joe Bloggs]

Quote:

See Bloggy the one thing that really proves to me that EQ is BS is the one thing you fail to recognize. You shouldn't EQ the signal you should EQ the equipment. The signal should be assumed to be perfect. Current EQ's always work on the signal, the signal is HOLY, you do not *** with the signal! PERIOD.

What is this, some kind of religion?

/points
[nelson]hahaha[/nelson]

Quote:

If I took your EQ'ed disks and put them in my system there is a good chance they would sound REALLY bad.

Yup, probably

In fact I only use that CD on one specific set of equipment only--the set I tailored it for. I mark down this set on my CD so that I don't mix them up. This sucks, yeah--every time I buy new phones I have to build up my collection of custom CDs again, unless most original CDs sound fine through it--but no worse than having to tolerate unbalanced music

And look, what's your problem with changing the signal? All you want is for the source music to reach your ears as accurately as possible, right?

The basic principle and rationale for 'messing with the signal' is this:

Let
S = Source
SG = Signal
a = difference between signal and source
O = Output
b = difference between output and signal

Now, if S = SG = O, then 'you hear it perfectly, flawless, with nothing but the character of the original recording to guide you'.

But we know that in real life while S ~ SG and can get quite close to S = SG, O != SG (never equal). You can try to make O as close to SG as possible, but you'd never quite make it. Driver and housing engineering and manufacturing processes have their limits. Besides, what about room acoustics

However, if we can measure [size=medium]b[/size], the difference between O and SG and formulate the difference mathematically, we can *deliberately* modify SG so that the difference between S and SG, [size=medium]a[/size] = [size=medium]-b[/size].

What we then have is
SG - S = -b, S = SG + b
O - SG = b, O = SG + b
----
S = O!!
Source = Output!

In the computer age, with powers of computing ever increasing, we can modify the signal in ever more subtle ways to compensate for the limits of the reproduction equipment. And Moore's Law means that more often than not the investment required to compute the changes required to make S = O by changing SG is much less than that required to try to make S = SG = O. But I suppose you'd be one to completely dismiss digital room compensation, among other things

Signing off:
Joe Bloggs,
The audio signal atheist

(oh, and of course I keep the original CDs with me and make new custom CDs from there, not the old custom CDs. I'd only apply EQ to the source once each time, of course.

)

 

[Tim D]

Are you talking about equalization or more powerful DSP's? Cause there is a difference between simple frequency band equalization from winamp plugins, and a powerful DSP device that probably costs a pretty penny.

You say that you would only EQ your source recording once. WHY is that? If EQ is a transparent process than your EQ'd recording is just as good as the original.

If EQing is a hi-fi process with no signal loss, than there is no real reason not to apply an EQ to an EQ'd disc since the EQ'd disc is equivalent to the original under that assumption. Or do you consider the signal to be better if you EQ'd it only once from the original. Is there some premise for which you believe that the less processing that goes on, the better signal integrity? If you are fine with processing the signal all you want, than again there should be no problem with processing an already processed signal.

I believe some people just think that recordings that are the least processed through good audio equipment just sounds better. My $20 buck SRS Wowthing DSP...basically a fancy smancy equalizer...makes cheaper bad sounding equipment sound better. Makes good sounding equipment sound worse.

Also driver technology is of more importance than you may believe. An incredibly powerful DSP or equalizer would not make a rock with wires hooked up to it sing. Driver matching still matters. Transients still matter. Accurate high performance transducers still matter just as much as the microphones that pick up the recording. If you believe transducer or microphone technology is limited, than to what extent would feeding an altered signal to a poor performance transducer be able to reproduce the original intended recording?

Sensor and measurement technology is also getting much better as well just as much as computer tech. This is a major factor in producing higher performance transducers which in themselves are reverse sensors. Microphones are basically sensors. Sensors take real-world measurements and convert it into information, i.e. signal. Transducers takes a signal and converts it into real-world response. DSP technology is basically the interconnects between the processes. Some people want simple interconnects that just gets the signal through as intended because it sounds better to them than hooking up some spiffy black box. Other people would rather have the black box.

Why hook up crappy transducers to good DSP hardware/software processes, when you can hook up good transducers to good DSP processes?

DSD recording is probably a very powerful DSP recording technique...lets use cans and strings for microphones cause the DSP portion of the process will hold everything else up. SACD is a better digital recording process, so hooking up Sony streetstyles to an SACD player with an SACD....hmm where exactly is the bottleneck. After all SACD technology is utilizing better DSP in order to play back the original recording, so does that mean we don't need better speakers or headphones...just better DSP technology?

Hmmm you know this is really up Bose's alley...they are really into active equalization and DSP processing coupled to sub-par transducers but at least they fit in little cubes. But man...with equalization technology...and moore's law...I bet Bose offers the best bang for the buck in price/performance because of their reliance on active EQ technology and DSP.

 

[Joe Bloggs]

For room equalization you need more than, um, an equalizer. The DSP work required in the kind of room equalization I linked to sure costs a pretty penny, but you could do worse spending money on speakers alone...

Quote:

Why hook up crappy transducers to good DSP hardware/software processes, when you can hook up good transducers to good DSP processes?

Of course both transducers and DSP should be of high quality. Saying that with a good DSP I can just use a rock for a transducer is no better than saying that with the best transducers in the world I don't need a DSP.

Of course there are things that DSP can't do, like compensate for random noise

That's why a clean source and good interconnects are still important. And of course if the DSP tells the speaker to do things it can't do, that's no good either.

The '$20 Wowthing DSP' falls outside the strategy of modifying the signal to match output to source. It tries to make good sound out of cheap speakers (by adding 3D stuff that is not there, thereby also compensating for poorly placed cheap speakers

), not accurate sound out of good speakers. That's why you can't use it to make your high-end system sound better.

Quote:

If EQing is a hi-fi process with no signal loss, than there is no real reason not to apply an EQ to an EQ'd disc since the EQ'd disc is equivalent to the original under that assumption. Or do you consider the signal to be better if you EQ'd it only once from the original. Is there some premise for which you believe that the less processing that goes on, the better signal integrity? If you are fine with processing the signal all you want, than again there should be no problem with processing an already processed signal.

There's no need to take things to extremes... but digital stages *are* cleaner than analogue stages, would you not agree? If you don't agree, why don't you try to record the signal output of the analogue line out of your best system with the line in on your best recording system, and loop it over a few time and see if the results are any better than several stages of digital processing? I think the minimal degradation that occurs in one stage of digital processing is well worth the corrections it makes.

In any case, there are many limitations in just equalizing in loudspeakers:

http://www.roister.com/eng/comp-faq.htm
(note: this link also happens to show how far DSP has got ahead of transducer technology! Just look at the graphs!!

But it was an early forerunner in the attempt to get things right by manipulating the signal.

*Moreover*, many of these shortcomings do not exist in the case of EQ for headphones, that's why I'm enjoying it so much!

 

[Tim D]

Actually according to that article...it is actually showing that there is more BENEFIT for Digital EQ in a speaker based system than in a headphone system. It doesn't say that there are less shortcomings from EQ in a headphone system than speakers. In fact it is saying there are things that digital EQ can compensate for in a room-based situation.

Also if you have good audio equipment and don't need extra digital EQ stages and your resulting sound is already extremely close to ideal...can it be possible that the ever so slight degradation of the added digital stage can outweigh the ever so slight benefits of EQ. I have played with different DSP technologies, and also played around with different headphones. The time, effort, and cost of trying out different headphones paid off much more than from optimizing encoders/decoders, or using different DSP's or EQ's. Hmmm most people that have done BOTH would probably favor getting the right equipment over getting the right EQ's. I've also tried Dolby Headphone before which is quite a bit above SRS WoWthing.

And that page does not show how equalization technology has outdone transducer technology(it shows the difficulties of room-speaker interactions). Hell that company would probably like to say that the speakers they sell are still much better than a competitors regardless of their digital EQ technology. I've looked at the graphs...impressive for speakers in a room. Nothing special when it comes to what already existing hi-fi headphones already do.

I mean what is it exactly that you are trying to prove? That it is more cost effective to just EQ and adjust what you got to what you want than actually auditioning and trying different audio equipment? That just doesn't fly. It really doesn't make sense why I would ever want to EQ and burn CD's to make them sound right for certain equipment when I am already content because I have taken the time to find equipment I like instead of trying to change my recordings. Also where does cost-effectiveness come into play when it comes to high priced DSP equipment, especially when it comes to headphones.

Ok I don't hate DSP's or spatializers or EQ's...but burning and reburning your collection especially if large...is just silly. Even if I paid myself minimum wage at 5 bucks an hour to redo all the recordings I wanted changed because I didn't get the equipment I wanted would cost MORE than just getting the right equipment in the first place. And not to mention I KNOW I wouldn't be as happy if I didn't get myself the right equipment in the first place. Is it truly strange to "waste money" and have more than one pair of headphones, or headphone amps depending on your taste? Or is it strange to have multiple and different EQ'd burns of your entire music collection. Time = Money, and both is an investment...wheter or not you view one investment as smarter than the other...well you don't know til you invest both ways.

 

[Joe Bloggs]

Quote:

Actually according to that article...it is actually showing that there is more BENEFIT for Digital EQ in a speaker based system than in a headphone system. It doesn't say that there are less shortcomings from EQ in a headphone system than speakers. In fact it is saying there are things that digital EQ can compensate for in a room-based situation.

Umm excuse me... the digital EQ I'm talking about is the poor man's good old-fashioned parametric / graphic EQ's

that basically does the same job as analog EQ's

(I believe the thing you're referring to should be called 'digital room compensator'...) So all the detractors of analog EQ mentioned there basically apply to my digital EQ. What they did *not* mention even for analog EQ was the 'disadvantage' of 'adding a black box between the source and the output, degrading the signal'

The one big disadvantage of analog EQ that I don't have to worry about when using headphones is that they don't compensate for reflections in the room. Big deal

Quote:

I've looked at the graphs...impressive for speakers in a room. Nothing special when it comes to what already existing hi-fi headphones already do.

???

^^Measurements for the 'average good HiFi speaker'...

--------------------------------------------------------------------------

You telling me there's a single transducer on earth that can manage this without the Compensator??

PS. I'm all for getting better equipment too

I'll be buying a HD580 soon...

 

[morphsci]

Will a moderator please close this thread. Some people like equalization, some do not, others do not realize the difference between DSP and equalization. It has been discussed to death and is now at the point of wasting bandwidth. None, of the DSP products are capable of making two different headphones or speakers sound identical at this point. Was that not the point of the original post, to use digital equaliztion to make a set of headphones sound like a different pair (hopefully better?)? It will work for some, but not others. If you like to EQ great, go for it. If you don't that's great to. Can we now move on?

 

[Dusty Chalk]

Quote:

Originally posted by morphsci Will a moderator please close this thread.

Might I recommend not reading this thread? Personally I find the whole discussion fascinating, even if only at an academic level.

 

[morphsci]

Quote:

Originally posted by DustyChalk Might I recommend not reading this thread? Personally I find the whole discussion fascinating, even if only at an academic level.

You certainly may. I was just expressing an opinion that the discussion was going nowhere fast and we had covered the same ground many times over the course of the thread. But hey, whatever floats your boat.

 

[rickcr42]

jeez,missed all the fun

Well I DO use EQ.Not much but it is there if needed.Bass boost,shelving high filter,fixed treble boost in 1db increments,crossfeed,and a single band parametric that can be varied from 250 hz to 5khz.all defeatable in mass or individually
WHY ?
Because I have some Cds that are damn unlistenable unless futzed with,mostly early jazz recordings and some 50s rock.
Rather than kick my own ass and be a snob I eq the sound to what I want it to sound like.Why should I not listen to a great musical selection that sounds like **** if I can enjoy that selection as the musician intended ?Early recording techniques were not always the best and some musicians were lucky just to lay down some tracks,no matter how badly recorded.And what about "basement " recordings ?We all have some "would be stars" that we know.They try to make the best recording they can but the limits of available equipment (CASH !!!) means not always the best sound,though the performance may kick ass.
No "purist snobery" here guys,it is all about the enjoyment of the music for the rickster.Screw political correctness,i never was one to follow

FLAME THIS,I LOVE IT (uh oh,moderator fanning the flames

)

 

[kwkarth]

The Rickster said:
"Screw political correctness, i never was one to follow"

Hey, you either follow, lead or get out of the way. Nobody stands around. Fact of the matter is, the best leaders also know how to follow better than anybody else, so go figure that out!

That said, EQ when performed properly in the digital domain can be pretty cool. When done in the digital domain, you don't introduce the phase anomalies that you would otherwise incur if you eq'd using conventional analog notch/shelf filters, and you can correct some pretty nasty problems that used to be impossible to deal with before the advent of digital. As with any technology there will always be some hammerheads that screw things up and mis-apply the technology and give it a bad rap.

But, hey, that's life and shouldn't prevent the rest of us from enjoying the technology. I predict we'll see a lot more application of digitally manipulating sound in the near future.

So, Rickster, I'll bet you didn't realize you were leading, did you!

BTW, the only full range single transducer I've seen that has a published F/R graph that's ruler flat when unequalized (save for the bottom octave) is enclosed for your viewing pleasure. Guess what it's from?

 

[Tim D]

Joe...time for a little flaming whupass. I have no problem with people using EQ especially for *****'d up recordings. I have no problem with spatializers or processors...I like playing around with them.

I do NOT believe however that they can make a poorer sounding transducer sound more hi-fi.

First a few lessons in graphs. Have you yet noticed that the impulse graph of the EQ'd signal is of much less tolerance both amplitude AND time wise than the non-EQ'd signal? Do you notice the range of some of the FR graphs on the page being 30-10khz? Do you know that quite a few headphones will be pretty flat in this area, and be able to produce good impulse graphs on a much smaller scale than msecs? Ahhh how strange 20 hz got lopped off in the EQ'd chart...wow the powers of EQ are truly amazing in producing different looking graphs aren't they? If they can't even get the graph parameters to match...what other parameters did they take liberty to change? BTW the second graph seems to be of a pure signal or amplifier...sure some transducers have difficulty getting close to it but some can get a lot right, enough to not need EQ. But the graphs are comparing apples and oranges. Lol actually the website says that the second graph is of a speaker...yet if you LOOK at the title you can see that the measurement is of a signal with a voltage measurement instead of SPL! Jeez try to drum up some good looking graphs and they forget to white-out the titles LOL. There are companies that try to sell junk to people under the premise of audio voodoo. That doesn't mean there aren't companies that will sell junk to people under the pretense of scientific measurements LOL. They're selling something...and you're buying it. If I see graphs that don't have same parameters it as useful as seeing FR response specs on the back of Sony earbud blister packs...attempts to sucker people into believing in DIGITAL too much. Its so much easier to sell stuff with graphs and numbers than it is based on real sound.

If you USE GRAPHS to support your arguement....LEARN TO READ THEM. If you can't do this...than it shows that you aren't actually doing anything but appealing to authority. An authority which can be flawed if you don't understand it yourself. Whats the point of slapping up tons of graphs and not even trying to explain them for yourself.

Do you KNOW that the DIGITAL compensation you refer to actually requires the use of MEASUREMENTS with microphones which are an ANALOG device. Did you know that microphones function very similarly to headphones...and have similar technical constraints and limitations. Perhaps a reason why the graphs shown are limited.

The FR graphs are also narrow in bandwidth, but extremely wide in amplitude...again to ensure a flatter looking graph. Plenty of hi-fi headphones will measure better without any digital compensation. Now maybe your defense is that your method of EQing has nothing to do with what was linked...and my question of course would be why did you bring it up in the first place if not to say "hey EQ's can make things better, here are some graphs and products to show it". You are right, your EQ process is different and less complex as well and "cheap" in terms of money. But no one who has a decent setup they are content with is going to go off and create EQ'd CDR burns of everything they like.

Finally if you are EQ'ing the headphones by ear...there is no guarantee you are making the transducers more "accurate". In fact it is more likely that you are just EQing the phones to sound better to YOU which is FINE, but don't drag all these graphs and links to show how EQ makes things more accurate when you are just using your ear. Finally I have heard a lot of pretty accurate headphones but I would NOT be able to plot a very good looking chart by ear and paper and pencil. If I don't trust my own ears to KNOW exactly how to plot a highly accurate flat sounding response...I am not going to trust yours to do it for me. What exactly is your "reference" for knowing when things sound flat by ear? Some people are pretty good at doing this...they are however sound engineers with a lot of experience with recordings as well as a variety of GOOD monitoring devices. Forgive me if I assume that a veteran sound engineer that is present at recordings would know what "accuracy" and "flat" is compared to someone EQing their Aiwa phones by ear for their own personal use. If you do it by microphone...what are the limitations of the microphone. Buying expensive higher quality recording tranducers and DSP for poorer cheaper playback transducers, from a company publishing biased apple and oranges graphs = dumb. Apparently you didn't get my implications before but microphone = recording transducer, headphone/speaker = playback transducers. If you say that these playback transducers are limited by housing and driver constraints...well guess what microphones are limited by.

These "digital" compensations that require microphones are definitely adding another "stage" that is also analog due to the measurements of the microphones. Hurray I just made one pair of tranducers more accurate because they measure more accurate to another pair of transducers lol.

Also optimizing your transducers for single sweep tones still does not do much compared to optimizing transducers for playing full-range music. Did you also know that most headphones amps do not produce much measurable difference in frequency response with a sweep? Joe you don't need any headphone amp, because they all sound the same and don't affect the headphones sound because they all can produce very exact frequency responses.

Have you noticed how many headphone manufacturers also double as microphone manufacturers? We buy their products because we assume they have better testing equipment and methodology instead of buying and spending time with some expensive "home-kit EQ optimization" to do it for us. Also your b + s arguments is just an incredible simplification and attempting to water down something that is incredibly complex into simple mathematics with simple variables. Let me ask you this...lets record the response of a Sennheiser driven by a Max. Lets hook up a TA, and record the response. Lets compensate the difference by EQ. Wow you don't need Sennheisers + MAX you can just get Sennheisers + TA + digitally EQ'd CDRs! Maybe I'll fudge up some graphs so some people will believe it too.

There is a difference in EQing a few aberant recordings, and EQing everything you got because of aberant equipment.

I don't doubt that EQing makes your Aiwa phones sound better, I absolutely believe you. Do you doubt that some people just don't feel the need to EQ every recording with the equipment they have? If you EQ by mic's you are still utilizing mechanical driver technology and limited by the accuracy of the drivers themselves. If you believe there are some mic's that are able to convert real-world response into a signal accurately, can you believe that there are transducers able to create real-world responses from a signal accurately. If you EQ by ear you are basically going by your own taste. If it is all about taste...perhaps some people prefer the taste of most their music unprocessed with good equipment?

 

[rickcr42]

So, Rickster, I'll bet you didn't realize you were leading, did you !

Hehehe,busted again

coupla points :
trying to equalize loudspeakers and headphones usually ends up a disaster.Phase anomolies,overloads,unatural sound.
The best workaround is to match the electronics to the transducer.
Too bright ? Mate with softer sounding amps.Too dull sounding ?Again add an amp that has the opposite sound quality.I am not saying you should not try for dead accurate but not everyone has the funds to go for the best so they make do.
Loudspeakers are usually best equalized by shifting the room position or modifying the reflective characteristics of the room itself.Kind of a passive EQ without the additional circuitry.You would be surprised at how different a system can sound with as little as two inches of movement or changing the toe in.

And another point about componant matching- say you have a preamp that has a speced response variation of 1 dB from 20 hz to 20,000 hz.Not too shabby right ?
Say the amp also has the same spec.
If both have that response variation at the SAME exact frequency you end up with either a bump or depression that is an audible 2 dB !Add in some other devices in the audio chain and you could for instance have one nasty bump or loos of frequency.
But match that with just the right headphone or speaker and you get an even response curve.just one of those things that some do not consider when making choices.Most would rather ask "what is the best......." and purchase blindly.
You have to do the gruntwork yourself,not preaching but it is what it is.

Digital EQ shows promise down the road when the price for an actually usable system is within financial reach of the average consumer.Maybe DSP is the answer for those with the programming skills.
For now I will keep using my combination "active/passive" box when needed for the musical enjoyment of a performance.But NOT for altering the basic nature of the sound of my system.

One place where EQ does little damage but can make a tough situation workable is inline with a powered subwoofer.Some rooms just do not like subs and standing waves can totally destroy the sound of the system.Multiple eq controls can work wonders here.

Oh yeah,that graph might be a Quad ESL

 

[kwkarth]

"Oh yeah,that graph might be a Quad ESL"

Nope, guess again y'all! This is a single driver now, but I'm not saying if it's dynamic or electrostatic.

Hint:
Think near-field.

 

[rickcr42]

wouldn't have anything to do with your avatar would it ?

 

[dhwilkin]

The fact that that graph filename starts w/ the letters k1k wasn't a dead give-away?