[Wayne1]

There has been some mention that Bybee Purifiers "filter" out high frequencies.

The Bybee purifier works in a different way then a standard capacitor filter and DOES NOT remove ANY MUSIC.

Here are a couple of graphs I ran with TrueRTA.

One is a cable hooked up from output to input of my Audigy2 USB sound card. The other is the same cable with an inline purifier with Bybee inside added .


No Bybee


With Bybee Inline


With music playing through the cables, in most systems, you can hear the difference with a Bybee inlne. You can see that there is not any high frequency roll off with pink noise in a test situation.

[Levi]

The graph looks the similar w/ or w/o Bybee.

There has been some mention that Bybee Purifiers "filter" out high frequencies.

The Bybee purifier works in a different way then a standard capacitor filter and DOES NOT remove ANY MUSIC.

Here are a couple of graphs I ran with TrueRTA.

One is a cable hooked up from output to input of my Audigy2 USB sound card. The other is the same cable with an inline purifier with Bybee inside added .


No Bybee


With Bybee I ...

[Wayne1]

The graph looks the similar w/ or w/o Bybee.

That is what I am trying to show. There is no amplitude difference at high frequencies because of the Bybee being inserted. The are differences that show up audibly as a reduction of "noise". I do not have test equipment that is able to show those differences.

The graphs are not the same one just repeated. If you look at around 1.7K on the no Bybee graph you will see a small double dip while on the with bybee there is just a single dip.

[azryan]

I guess I don’t ‘get’ Bybee’s at all (and I’ve REALLY tried to ‘get it’ though I’m not an E.E. at all).

You know I had you put 3 Bybees in my Panny you modded but since I never heard the diff. of all your mods w/o Bybees (though I did live with the stock 45 for quite a while) I can’t personally say what I think ‘they alone’ do even though all my audio passes through them (which I still feel kinda weird about).

You seem to be trying to make it very clear that they remove NO music, but you’re also saying they DO remove noise -right?

Isn’t that the same thing though? Both are simply audible sound?
Music -good, noise -bad... but which is which is a matter of opinion?

Or are you saying that it’s only removing sound (whether called ‘music’ or ‘noise) in the band ‘above’ audible? -which the graphs you posted seem to imply at least to me.

Or that the Bybees can somehow tell the diff. between sound that’s part of the song and sound that’s recording noise or noise made by your system/power lines?

Maybe a matter of how what’s called noise is typically more a constant energy than music which is far more a dynamic energy?

It always seemed to me that the bottom line was that it’s 100% unexplainable how the Bybees work, and what they do, yet ‘sometimes’ people find them to make a diff. and ‘usually’ that diff. is an improvement?          

Also the very very slight diff. in your two graphs... could that just be a random thing that could happen if you simply retested the same cable over and over again or will ‘with’ and ‘without Bybee’ always measure just slightly diff. as your graphs show?

[JoshK]

As a statistician/mathematician by training, I'll chime in here and say that if you want to make a better comparison between the two graphs, take the second minus the first, or the second divided by the first.  The range of the two graphs makes it pretty hard to differentiate any difference.  

I think anything more than a 0.5db change would be significant as your speakers will compound this change.

[eico1]

I think Wayne stated his objective and shows the highs are not rolled off.
Following Josh's advise would make it the graph more clear.

If one was able to show that a component either does or doesn't produce audible differences, that would put about 90% of audio companies out of business. That is if anyone believed you:)

steve

[Wayne1]

The main reason I posted the graphs were to show that there is NO roll off in the highs. A customer of mine has a modded SA-XR45 with Bybees installed. A friend came over with some brand new room analysis software and proceeded to tell my customer that his system rolled off at 24 db per octave above 2 khz.

Even though the system sounded good, the customer was worried by the readout. After all test results don't lie  He said he read somewhere that someone felt the Bybees were filters and got rid of highs. I assured my customer that the bybees were not the cause of the roll-off. I posted the graphs to show that whatever it is that bybees do, they DO NOT REDUCE the highs frequencies.

My customer finally concluded his friend really didn't know how to work his new toys.

[Carlman]

Thanks for the proof that Bybees do not roll off highs.

As to not being able to prove what they do, then if that were true, they could not be developed in the first place.

I think the mystique of bybee sells more than the actual function....  I had one in my transport but never bothered to try any comparisons without it.

I know they do something, I know that no one has posted how they work, but I also know now that they don't roll off highs.  Thanks for taking the time to get us a little farther in knowing what they do.

-C

[azryan]

"-24 db per octave above 2 khz. -"

HOLY MAN!
Someone actually even considered thinking this result was true -much less from Bybee filters??

The best way to achieve this kind of 'roll off' would be to take a knife and stab out the tweeters in your speakers -or just disconnect them (though that's less dramatic/comical).

Then cover the speakers in beach towels to keep the midrange drivers from outputting any treble.

[nathanm]

Are we sure we're not talking about 20KHz instead of 2KHz?  Perhaps Wayne mistyped or the customer misspoke?  If the Bybee shorted out the tweeter you might get a result like that, right? Heh!

[Wayne1]

[azryan]

Whoa.

So uh... what do you think it was? Wrong type of mic?

Or maybe that there are diff. ways of graphing loudness and this is not the way human's hear? (it does say logrithmic at the top though so that seems right).

That sure looks like the graph of a speaker with the tweeters disconnected, but who wouldn't hear that before the test OR not hear it and instantly know the test was daffy?

[Gazza]

The graph above came from my system, which consists of a PC with an RME HDSP9632 sound card being used as transport to feed an spdif signal directly to my Bolder modified XR45 (with Bybees on the 2 front channels) and then Green Mountain Audio Europa's as my 2 main speakers. The graph came from a session with a friend of mine who had just got the ETF software along with the pre-calibrated mic they recommend. The odd thing is that we replaced the speakers with a set of small Kef's), and still got the same result, we walked through the whole process with the tech guy at ETF, and he couldn't spot anything we were doing wrong, so we ended up stumped at the end of the evening.

The results must be wrong, because, as was mentioned before, you wouldn't be hearing any high frequencies at all if they were correct, and the truth is that the system sounds great! So we still haven't found a sensible answer as to what was going wrong.

the only reason we did the test in the first place was because I'd just got hold of a Behringer DEQ2496, which I was going to use to (hopefully) sort out any of the problems that the room itself was causing. Trouble was we never actually got that far.

[firepile]

heya! did you try using another source, besides the RME sound card, to see if that was the culprit?

oh yeah, I'm asking, cause I have a similar setup: RME 98/6 PAD to Muse 192 DAC, and just wondering at the sonic impact of using the sound card rather than a standard transport?

[vikingboy]

Hi All,
I was the friend/fool of Gary's with the new ETF toy.

I've in the past used basic Radioshack SPL meters to my own system and thought the time was ripe to move to something a little more advanced to help me integrate my new sub with parametric EQ into my system. I figured ETF would be a good place to start.

I was speaking to Gary about this software and offered to go round and help him optimise his system.

ETF has proven to be the most difficult piece of software I have had to get my head round so far. Actually, its not the software, its the acoustics that are the tricky part.

I am sure my hardware setup is functioning correctly. Measuring my system I see a flat frequency curve out to about 16khz, I then see a gradual fall off to 20khz totally in line with the manufacturers claim for my speakers. When measuring Garys system I am buggered if I can understand why the graph shows the insane rate of fall off that is obvious in the graph above. I've never heard a system with that kind of curve, but I wouldnt suggest Garys system sounds like that so the problem must be with the measurement of the room. Correct?

They guys from ETF talked us through taking the measurement so I dont think we did anything wrong, although I'm happy to be shown where we did go wrong.

Gary has a Behringer EQ unit which does its own frequency measurement using a different microphone, the display is much smaller on the front of the unit, and Im not all that familiar with it myself, but I would say that it looked like the behringer was indicating a fall off through the high freq area similar to what ETF was measuring.

I'm totally clueless what could be causing this.  The question to Bolder about the Bybees wasn't meant to be suggesting his mod was rolling off the highs, it was asking if the bybees could contribute to this, and what would Wayne think of the curves we sent him. Wayne uses TrueRTA software and I have now downloaded a version to try. I am using a calibrated Microphone and preamp (the same one as suggested by the guys at ETFacoustics).

If anybody has any pointers to site that can discuss the issues of setting up a 5.1 channel system using ETF I would be only too happy to spend the time learning and remeasuring Garys system in order to provide correct results here for the sake of completeness.

At the end of the day, there is no substitute for final positioning by ear, and Gary is currently happy with the way his syetm sounds. I'm sure by effectively using ETF and correctly setting up his behringer unit we can improve things further but at the moment I have no confidence in the measurements we are seeing and therefore don;t want to mess with his setup.

Firepile: When we captured the curves in ETF, the ouput source was my laptop playing white nosie via a Soundblaster USB external soundcard via optical into his behringer (in bypass mode). Therefore if there is something in the system introducing a degree of roll off, we think it is either the behringer, the Panasonic Xr45, speakers or the room. Of course, it could also be the fool with the toys

[Wayne1]

vikingboy,

Thanks for posting and clearing up how the test gear was used.

Next time you want to try it, use pink noise. Also, just as another step in eliminating what could cause a problem, try using a test CD with pink noise instead of the laptop. Also try going straight into the Panasonic with out the Behringer or anything else in the line. Test just one speaker at a time. Then the whole system.

If there is still a problem, ask Gary to send the Panasonic back to me and I will go through it to see what, if anything, may be causing the rolloff. It sounded fine on my test system, in two channel, before it left.

I am sure it is not the Bybees, but who knows what happend on the trip over the pond.