sony MDR-XB500 equalizer settings

these equalizer settings come from a microphone calibrated at a professional sound laboratory.

the only catch is the equalizer used and the differences between them.

for my equalizers..

i just used the 'audio creation mode' from the creative x-fi elite pro software.

i used the graphic equalizer and TWO parametric equalizers to get my results.

to get the second parametric equalizer for the front channel, you need to select 'parametric EQ' from the drop-down list up top where it says which set of speakers you are adjusting.

graphic equalizer settings:

31hz ~ -6.4dB

62hz ~ -10.5dB

125hz ~ -12.0dB

250hz ~ -12.0dB

500hz ~ -8.0dB

1k ~ 0.0dB

2k ~ -8.2dB

4k ~ -8.4dB

8k ~ 0.0dB

16k ~ 0.0dB

preamp @ 0.0dB

parametric equalizer 1 settings:

low cutoff ~ 110.0hz @ -10.0dB

mid center 1 ~ 2610.0hz @ 6.5dB (0.50 Q)

mid center 2 ~ 3540.0hz @ 18.0dB (0.33 Q)

high cutuff ~ 7900.0hz @ 18.0dB

parametric equalizer 2 settings:

low cutoff ~ 50hz @ 6.0dB

mid center 1 ~ 1330.0hz @ -5.4dB (0.27 Q)

mid center 2 ~ 5500.0hz @ -18.0dB (0.51 Q)

high cutoff ~ 9950.0hz @ -18.0

this is the frequency response before any equalizer:

this is the frequency response after the equalizer settings above:

regardless of ear shape or size.. there can only be one standard between us.

that standard is known as a flat frequency response, a flat phase response, and a flat timing response.

those settings give too much treble dont they?
that is what i thought after some hours of listening.
i switched my microphone calibration file from 0 degree angle to 90 degree angle, put the assembly back together and re-ran the analyzer.

i came up with some different levels that are much more appropriate to my ears, but if these are actually WORSE for you.. i left the old values in the blog above.
doesnt everybody save trash?
and isnt it a good time to place divide between trash and junk?
lots of people confuse the junk for trash.. they dont realize the value, and they see other people giving praise for the messy house.
then they go home and start to live with trash around them because the other people said it was okay by not complaining (or even standing up for the junk).

confusion isnt a mistake, but i wasnt happy and attacked the problem again to find a better solution (this very point many people fail to do, and i am still just a baby but believe people always falter eventually - hence the word)

anyways..
on to the better? settings.

graphic equalizer:

31hz ~ 0.0dB
62hz ~ -2.5dB
125hz ~ -12.0dB
250hz ~ -12.0dB
500hz ~ -5.7dB
1k ~ 2.2dB
2k ~ -1.6dB
4k ~ -6.6dB
8k ~ -3.9dB
16k ~ 12.0dB

preamp @ 0.0dB


one ONE parametric equalizer this time..!

low cutoff: 65hz -18.0dB
mid center 1: 1500hz -8.9dB (1.00 Q)
mid center 2: 3500hz 18.0dB (0.57 Q)
high cutoff: 5500hz -18.0dB

the revised frequency response looks like this:

enjoy..

i dont see any reason to do it again.

below are the studio standard settings for the xb500 headphones when your x-fi soundcard is in 'game' mode.

cmss-3d must be turned on for these equalizer settings to be considered valid.

since many video games require rear speakers, it only makes sense to do the calibration with the cmss-3d on.

i also included the bass boost feature as turned on to give the ultra lowest bass some attention.

31hz :: -12.0dB

62hz :: -10.0dB

125hz :: -12.0dB

250hz :: -8.9dB

500hz :: -7.8dB

1k :: 4.0dB

2k :: 0.0dB

4k :: -12.0dB

8k :: 5.7dB

16K :: -2.7dB

bass boost @ 10hz :: 11.0dB

 remembered i didnt get a chart for the last set of equalizer settings i did in 'game mode' on the x-fi soundcard.

so i went back to get a chart.

instead of leaving the real time analyzer running and letting the frequency response bounce all over the place, i used the 'forever' averaging to calm things down.

well...

what i seen was something i didnt particularly like.

it looked like this..

i seen a dip at 60hz that i didnt agree with.. but look at how low the midrange is compared to 30hz.

next thing i did was look at the equalizer settings, and i could clearly see there was TONS of room to keep boosting.

so i did the equalizer sliders AGAIN.. and kept using 'forever' averaging to clean up the analyzer results.

what does it look like now?

i raised 60hz a bunch, and then i went to attack the midrange and lift all of that area between 400hz and 2,000hz

now it looks like this:

it sounds better AGAIN.. because i was doing averaging by eyeball the first time.

that just means the TOOL does it better than i ever could.

i cant raise 8,000hz because it wont go anywhere much.. raising the 8k slider raises ALL of the treble to obnoxious levels.

once again, these headphones sound like the new 'high-resolution' generation.. but the $79 price tag is showing itself each time i go looking for the floor.

i find it like jumping into shallow water.

but

at least everything is smooth in the small amount of water that is there.

i look at it like this..

if you connect a lightbulb onto the speaker.. moving the cone in or out is supposed to light up the bulb.

some speaker motors will make the light flutter, and sometimes the light bulb will completely go out as you are moving the cone in or out.

i dont want the light to go out at all.

getting dim is annoying.

but

the new generation of speaker motors will be following suit.

the light will stop going out, and the level of dimming will be significantly reduced.

it should be all one solid color.. and trust me, staring at that bulb for micro pulses has been the truth.

because if you are pushing the speaker in and the bulb is staying bright.. what if you suddenly pull the cone out?

chances are, the bulb goes completely out.. or the light gets very dim.

we dont want that.. we cant do anything with a dead bulb, because music is energy and we need the energy to be functional.

that means a bunch of headphones are gonna be sold for cheap, given away, and completely thrown out.

they've been making a huge movement about energy and recycling for the last two decades.

what they havent been telling you..

all those things that are not worth recycling.

people wanting EXTRA bass would probably be upset to hear the XB500 headphones with my equalizer settings.. because the bass is there like a 12 or 15 inch subwoofer over in the corner pounding.. but it isnt louder than everything else like when you are standing right next to it.

the point is.. if the woofer is there and pounding, if you can hear it.. it should be good enough.. because it isnt one of those towers with three 6.5 inch woofers TRYING to be a subwoofer.

i could verify.. it could pass as two 18 inch subwoofers pounding.. but from a distance of at least 10ft (and NOT by the wall)

i am the type of person to appreciate the bass sounding like an 18 inch woofer and not some tiny 10 inch woofer.

but

some people are too deaf to be an audiophile?

nope.. you can be going deaf and realize how much better it sounds by listening with my equalizer settings and looking at the before and after chart.

i swear..

the water that is there, it's cleaner than speaker motors of the past.

cleaner than the $500 headphones too.

add up the vocals, the reverb, the panning.. the late-er order of harmonics are missing again.

but finally.. the young-est harmonics are FINALLY there.

strings from an instrument demand fast response times from the young harmonics.

large chunks of air, like something from a tuba, require late-er harmonics.. but these headphones can do a tuba, and can probably do a tuba with an echo too.

i think the truth is, the speaker doesnt move in and out much when it does what it does.

getting these kind of results with extra long stroke would be the final stand here.

but

i think i finally got it..

the light bulb could be brighter when switching from cone in to cone out.

but hey, at least the light is lit up.

it reminds me of a signal to noise ratio problem, one that is eating away some of the slew.

and that brings me to say.. the distortion is less with these headphones than the many others.

i think the expensive headphones have more slew, but the distortion is higher too.. and nobody complains because the slew is there and they dont know any better.

i'm hoping the 'high-resolution' revolution will teach the followers a thing or two.

**edit**

crap..

forgot the new settings..!

bass boost @ +12dB - 10hz

31hz .. -12.0dB

62hz .. -5.4dB

125hz .. -12.0dB

250hz .. -8.9dB

500hz .. 0.0dB

1khz .. 9.5dB

2khz .. 6.4dB

4khz .. -2.5dB

8khz .. 0.0dB

16khz .. 3.5dB

there wont be ANY complaining about the vocals being over powered by the bass.

not a transparent sound yet, but closer.

if you are looking for transparency (that thing that happens when you have absolutely no idea if there are headphones covering your ears or if you are really listening to the person) these headphones are many steps towards that direction THANKS TO MY EQUALIZER SETTINGS.

grab a pair of these and wait for the new 'high-resolution' audio gear to be released.

some of those 'high-end' headphones are TWENTY and even THIRTY years old.

we need the gear that is for the next ten or twenty years.

personally, i'm good for gear that is ment in the next fiftyfour years. (that would make it rain.. lmfao)

dont let my previous 'slipped mistake' be of error to you.

i realized something, and i will tell you how it happened.

first of all..

the first set of equalizer adjustments with two parametric equalizers and the graphic equalizer is perfect.

then, the second set of adjustments with the single parametric equalizer is perfect.

however,

i was doing a loop-back test with the soundcard and viewed the frequency response change a whole bunch, and the phase response changed a bunch too.

i ticked the cmss-3d on and off while doing the frequency response from the real time analyzer window, those amplitudes didnt change.

well..

this is what happened.

because the frequency response didnt change in the RTA window, i continued with the equalizer calibration.

but

when i went back to view the frequency response a week or two later to see if it was the same or much different.. it was much different.

that is the slanted line with the 'average' clamp.

so what did i do?

i re-ran the loop-back test and turned cmss-3d on to add those changes to the soundcard calibration file.

(the calibration file removes all changes added by the soundcard to make the equalizer calibration clean and perfect)

the second set of equalizer adjustments (the graphic equalizer only settings) are for cmss-3d turned ON.

the first set of graphic equalizer adjustments are for cmss-3d turned OFF.

i know it says 'cmss-3d must be on for the settings to be valid'

but

you can see in the graph what happens when you dont add cmss-3d to the soundcard calibration.. the frequency response slanted from left to right.

now..

you've got real equalizer settings for two parametric equalizers + graphic equalizer.

you've got equalizer settings for one parametric equalizer + graphic equalizer.

you've got equalizer settings for a single graphic equalizer with cmss-3d turned off.

and you've got equalizer settings for a single graphic equalizer with cmss-3d turned on.

but hey..

you dont have it all yet..!

go into your dolby decoder settings and turn prologic decoding OFF. (digital EX should be turned off too)

it sounds a ton better in the upper midrange and treble areas (this is the area i heard the trash the most).

if you need an example.. i would compare two pieces of paper, with one totally clean and no creases or wrinkles - and the other one has been crumbled up too many times, making the text hard to read because of interference from those wrinkles.

the data isnt even present with prologic on.

if you cant hear it, you need a better amp.

my home theater receiver sounds better when i turn the equalizer off.

obviously the equalizer is based on math function.

either the math's bit-depth isnt enough, or simply getting inbetween the power rails and the output is enough to change the output clarity. (either one makes perfect sense to me)

i listen to music from the digital cable box connected to the SPDIF input of the soundcard.

with the prologic turned off and the cmss-3d turned off.. it sounds as good as it is going to get while waiting for the audio's bit-depth to increase.

increasing the bit-depth is going to be like going from 480p to 720p .. and i would expect to see another bump in the bit-depth (unless we skip 720p completely).

but

you should be getting the idea that bit-depth changes the clarity like changing the SD / HD resolution.

the sample rate isnt as strong of a change, and is comparable to calibrating the colors of the television.

however, higher sample rates will require lower bit-depths.. and you can 'trick the system' by using the highest sample rate.

why is this true?

because the smaller the sample, the less data contained in the sample.. less data in the 'packet' means the bit-depth of the packet is lower and the bit-depth of the conversation goes up.

that is what the difference is between 24bit audio and 32bit Input/Output
use your head..

192khz is FOUR TIMES more than the 48khz of DVD audio.
you simply cannot keep the same bit-depth when the sample is 1/4th the size.
in fact, the number of 'differences' can go down.

differences are also known as 'colors available' in a box of crayons.

each time the analog to digital chip 'samples' the audio.. it is looking at the audio and saying what 'crayon' it is.

48khz might pick two or three crayons @ 16bit

and at 24bit.. 48khz might pick three or four crayons.

but

192khz is 1/4th

so @ 16bit .. it would be picking only one or two crayons

and at 24bit .. you need to realize the difficulty of picking only half of a crayon.

192khz recording isnt common in 'consumer-level' cards.

but 96khz is.

and if the recording software DOESNT re-sample the 96khz down to 44.1khz ... your chances and efforts amount to what?

overclocking the results..!!

the quality is defined by the preset you select.

you are locked into what bit-depth is given to us.

and if you can stretch the quality while staying in the same bit-depth, raising the sample rate is the ONLY chance of doing it.

anyways..

this applies to recording only.

whatever sample rate used to record is the 'be all that ends all' when talking about the quality results.

another way to look at it?

how many shades of color can you fit between black and white?

say you can do 60 shades.. but you dont have enough pixels to show all 60.

what does this mean?

it means a television with a high number of colors available cant show depth until the number of pixels goes up.

that is why they used to 'overclock' old CRT televisions by adjusting the electron gun to scan with more pixels.

to be honest with you, i realized long ago that i didnt put up any settings for cmss-3d off.. yet i've turned off cmss-3d with those settings and it sounded better with the cmss-3d off than it did with the feature turned on.

after hearing it, i figured i would simply let people realize themselves the improvement in sound with cmss-3d off.

the entire thread was based on facts, i realize that.

and for what it is worth,

i think the phase response for the setup i used is goofy like cmss-3d turned on.

the setup i used to record the frequency response goes as follows:

i took a used ovaltine tub (the new plastic ones with a screw on cap) and drilled out a hole for the microphone tip.. then i placed the earmuff on the plastic lid with the microphone through the hole.

lets look inside where the microphone see's:

there is a direct wave from the speaker cone to the microphone tip, but there are reflections from the earpad AND the plastic lid.

if anybody knows omega math, you know there is a ring with a divide that adds up to a pocket drunk.

if anybody knows room reflections as to how they add up and cause phase cancellations or increases, then you know reflections cause the soundwaves to bump into eachother.

what if i said cmss-3d has a goofy phase that isnt perfectly flat from 0hz to 20,000hz ?

i know it does because when i do a soundcard calibration to remove all of the frequency differences for a perfectly flat calibration, i could see the phase in the calibration result.

what if i said the cmss-3d phase was almost the same phase response of the chamber the microphone was inside of?

what if i said turning cmss-3d off removes all the phase problems that was picked up by the microphone when i calibrated the equalizer?

that is kinda what the final result sounds like when i turn off the cmss-3d feature.

what if i said the result sounds really good and it made me forget if i did the calibration with cmss-3d on or if i did it with cmss-3d turned off?

(i really dont remember if i did it with the feature turned on or off, but i would like to think i was straight and narrow until the point i simply lost the memory because of confusion and time gone by)

let's be honest with eachother, the program i used to calibrate the equalizer is digital and therefore those results can be different if somebody changed the data inside the software.. because my computer is always connected to the internet.

let's be honest with eachother, the settings i posted were input by me.. but the data is digital and viewed on a digital website, therefore the settings you read on your screen might not be the same data i input into the text box.

let's be honest with eachother, you should get what you deserve and i've told you two ways the data can change because you deserve more or less.

for what it is worth,

i've calibrated those headphones with a calibrated microphone.

i also own some floorstanding speakers that i've calibrated with the microphone.

the results are not all that different, and since there isnt a huge difference i say turn off cmss-3d and worry less about the settings being perfect because the results are a magnitude better than going without the calibrated settings.

to be honest, i prefer listening without cmss-3d turned on.

when i turn on cmss-3d i am annoyed by the final result because there is a noisefloor introduced.

some say cmss-3d is specifically designed to recreate the sound of soundwaves bouncing off the four walls, ceiling, and floor of a room.

i can hear that effort, but it sounds like the ceiling is missing (or maybe the floor is missing).

whatever the raw data inside cmss-3d is, i think it sounds annoying because the noisefloor is to be low to create the factor of invisible.

i cant close my eyes or be in total darkness with reflections and drift away far or deep to the point where i believe i am actually there (or anywhere close).

i read somewhere that cmss-3d is designed to pull the sound out from inside of your head.

i dont remember if it was the same article that said something about trying to recreate the reflections from walls.

the only time i turn the cmss-3d feature on is when i am playing a video game (and the positional audio is nonfunctional with the games i've got).

i know what pink noise is supposed to sound like when the calibration is correct.

i have only hit that sweet spot less than a dozen times, and it appears the program i use gives different results each time i use it.

with that much said.. i dont see any reason to do another calibration with cmss-3d turned off.

i've been able to use the real time analyzer feature and see a flat line that is similar to what i calibrated, but it hasnt ever been exact (and i know the difference in the air is enough to prevent an exact repeat, but i still think the difference is high.. especially when i run the pink noise again while changing equalizer settings)

it's almost like the program yells at me for touching it again after already doing the calibration.

it makes experimentation less than exciting.

what i can say about the different results each time,

each one sounds better than not doing it at all.

what i really think..

i think the program finds a frequency that is ringing in the room and the frequency the software picks is different each time.

typically that should be known as the 'gate' rather than the 'window'  ... but i might have that statement backwards.

what i know of 'window' and 'gate' settings..

one of them is the actual frequency (or harmonic) and the other one is how long or short the recording of the echo is.

the program i use doesnt let me adjust those two settings, and i think it makes the adjustment for me all on its own.

for people that deserve more, they should get more accurate settings for the room being calibrated.

for those that deserve less, those settings should be less accurate for the room being calibrated.

i believe there are more ways to prosecute people other than paying money for a fine or spending time in jail where your freedom of conscious, freedom of thought, freedom from torture, freedom of worship, right to fair trial is all taken away from you (unlawfully might i add since those freedoms are FREEDOMS bound by law's more complicated than politics, more to the count of natural primal instincts of the human species because of the chemicals inside of us that give us arms, legs, fingers, toes, as well as a neck)

...look

i turned off the equalizer on the soundcard and used an equalizer everybody could use.

the settings are in the picture .. same numbers for the left as the right.

i'm attaching the impulse response file in this post.

waterfall before impulse response:

waterfall after impulse response:

group delay before impulse response:

group delay after impulse response:

frequency response before impulse response:

frequency response after impulse response:

equalizer|convolver|spatial impulse settings:

**edit**

cmss-3d was OFF for those equalizer settings.

if it tickles your head .. omnisone wasnt included with the equalizer settings.

but it brings out the sound more solid for a 3D soundstage.

those new settings with this really shows the difference with the equalizer on|off

the last settings i posted use a VST equalizer that can be used in foobar.

the impulse response file can be used in foobar too.

i turned off the equalizer on the soundcard.. there wasnt any audio effects going on for those equalizer settings.

i posted the vst equalizer and equalizer settings for people that dont own a creative soundcard, so they could get in on the listening too.

i dont think more bands on the equalizer is going to help boost that dip at 900hz .. and if it did, the difference would likely show itself again after applying the impulse response.

you seen the dip get bigger in the second frequency response graph (the waterfall too).

to get the equalizer working in foobar you need to download and install the vst foobar plugin.. then foobar can host vst plugins jst like hosting any of the other DSP plugins.