Re-Ran my Audyssey Set Up on Denon 4311 ... look at this crazy Chart.

[fuzz092888]

Now we get my biggest complaint about Audyssey, it is correcting the top end to make it flat at the listening position.

Unless I'm mistaken, only Audyssey Flat tries to make the top end flat at the LP. I believe that the regular Audyssey curve has a gentle roll off of the top end, but I don't recall how much or if they even say since I think that's proprietary information.

 

[KEW]

Now we get my biggest complaint about Audyssey, it is correcting the top end to make it flat at the listening position. This should NOT be done and sounds awful.

Can you elaborate on this?
Normally, I would think EQ'ing for a flat FR at the LP would be desirable (whether the low end or top end).

Thanks!

 

[fuzz092888]

Can you elaborate on this?
Normally, I would think EQ'ing for a flat FR at the LP would be desirable (whether the low end or top end).

Thanks!

TLS has said it before, and in a much more detailed and thorough explanation, but I believe he said that high frequency sound naturally rolls off with distance. So EQ'ing for a flat FR at the LP produces an unnatural sound. You want a speaker that will have a flat accurate FR anechoic, and then in room you want it to have that natural roll off of the top end, but with minimal variation, think the +- db, to give the most realistic sound.

 

[PENG]

Unless I'm mistaken, only Audyssey Flat tries to make the top end flat at the LP. I believe that the regular Audyssey curve has a gentle roll off of the top end, but I don't recall how much or if they even say since I think that's proprietary information.

They said roll off but did not specify how much. I plotted Audyssey off vs Audyssey flat vs Audyssey for my main speakers, following are the ones for the left channel measured from listening position. It did level the highs somewhat in both cases but still maintain a roll off.

 

[fuzz092888]

They said roll off but did not specify how much. I plotted Audyssey off vs Audyssey flat vs Audyssey for my main speakers, following are the ones for the left channel measured from listening position. It did level the highs somewhat in both cases but still maintain a roll off.

View attachment 15499

I used to have a graph of the Philarmonic 3's with Audyssey vs Audyssey flat and the roll off was much more pronounced than on your graph. Curious how close those two look.

 

[PENG]

You have the AV8801 as well so your Audyssey should really work the same way as mine. It does depend on how you position and support the mike. I tried many times and the graph I posted were the best I could achieve in my room. I took some measurements at 1 meter from speakers with no Audyssey and they didn't look better. Oh, don't forget I have the graphs at 1/6 smoothed, if I used 1/12 or 1/24 they would look more pronounced too. My speaker are also different, they are Energy V2.3i (the second gen Veritas before bought out by Klipsch) that are probably softer on the highs than the Phil 3s but I really am not sure.

 

[fuzz092888]

You have the AV8801 as well so your Audyssey should really work the same way as mine. It does depend on how you position and support the mike. I tried many times and the graph I posted were the best I could achieve in my room. I took some measurements at 1 meter from speakers with no Audyssey and they didn't look better. Oh, don't forget I have the graphs at 1/6 smoothed, if I used 1/12 or 1/24 they would look more pronounced too. My speaker are also different, they are Energy V2.3i (the second gen Veritas before bought out by Klipsch) that are probably softer on the highs than the Phil 3s but I really am not sure.

Actually when I ran the test I believe it was with XT32 on the Onkyo I had at the time. I haven't run that same test with the 8801, however it had nothing to do with the smoothing. IIRC the roll off started earlier and was much different by the time both graphs hit 20kHz. With the flat engaged, it was nearly flat until 18kHz before it started to roll off. However, by that point it could have been the mic, soundcard, or both.

I'll have to try it again and see if I can replicate the results with either the 7008 or the 8801.

 

[TLS Guy]

Can you elaborate on this?
Normally, I would think EQ'ing for a flat FR at the LP would be desirable (whether the low end or top end).

Thanks!

Basically as you move away from the speaker, the proportion of reflected sounds increase in proportion to direct. Since the reflected sound has HF roll off the HF content decreases with distance. This is what the ear expects, and if you correct for it, it is not an improvement, just the opposite.

I agree with Billy Woodman, of ATC, that apart from perhaps taking the edge of troublesome room modes, correcting frequency response at a distance is not a reasonable proposition. When it comes to it equalizer a speakers response is a very incomplete solution. Frequency response errors are so often just the symptom of an underlying problem which still remains after the correction

As I have said many times, you can not make a good speaker out of a bad one with an equalizer.

 

[Swerd]

As I have said many times, you can not make a good speaker out of a bad one with an equalizer.

Amen to that!

You couldn't do it old school analog equalizers, and you still can't do it computer-guided digital equalizers.

 

[bizmord]

As I have said many times, you can not make a good speaker out of a bad one with an equalizer.

First of all ... I so wish you were my neighbor Much appreciate all the feedback even though I understood about half of it. Became too technical for me.

I always thought that for price paid, these Sierra-1 NrT were one of the best bookshelves around. Not so?

 

[TLS Guy]

First of all ... I so wish you were my neighbor Much appreciate all the feedback even though I understood about half of it. Became too technical for me.

I always thought that for price paid, these Sierra-1 NrT were one of the best bookshelves around. Not so?

Well, first of all, Ascend make no attempt to "pull the wool" and present honestly what you are getting and to a very large extent why. This is all too rare and I applaud them.

So it is up to you to decide if you had value for money. With most speakers it would be impossible for me to discern what you had. My guess is that in this price range most are worse and some better.

The real issue is how would it sound if I designed it using the same drivers.

It would have BSC and the box would be aligned differently.

The result would almost certainly be lower sensitivity, and lower total output. It would appear to have less bass although the bass would be more accurate. This might or might not be to the liking of pop and rock fans.

The impedance would be lower in the range where the bulk of the power is, and some cheaper receivers might not like that.

So, it then comes down to a commercial decision on the trade offs. For my taste and choice of music and available amplification, I would have made different choices. I don't have to make a living selling speakers, just enjoying them immensely.

 

[bizmord]

What Audyssey has done is actually explainable. It also gives insight into the world of commercial speaker design.

I have not really looked closely at Ascend speakers before, to be honest. I would say their site is more forthcoming than most, and more honest.

Here is the frequency response of the unmodified and modified versions of the speaker you have.

Here is the impedance curve.

What can we see from this? First the speaker is likely not BSC compensated, or at least very inadequately.

The lack of BSC is attempted to be masked by improper tuning of the enclosure to give a peak in the 100 Hz region.

The impedance curve does not show the drop that it should if the speaker were properly BSC compensated.

Now Ascend are quite frank about this to their credit.

[Overall curve is remarkably linear, coming in at an amazing +/- 0.5dB throughout the bandwidth of the speaker. (note, port tube output is not factored into the anechoic response, which will fill in the mild dip in output below 500Hz)]

The reason given for this is linear impedance and easy drive. (Damn receivers again!)

[Linear impedance response for consistent performance regardless of amplification source.]

Other probable reasons for this but not stated would be: -

Driver at that price point not able to handle the power requires to achieve goal spl.

Lower overall spl and a reduction in sensitivity spec.

The other issue despite talk of careful phasing there are significant phase discontinuity issues at crossover which really shows up in the waterfall plot.

In the t version you have this is considerably improved.

High end response is smooth and extended.

So what of Audyssey?

It looks as if you have a room bass peak problem.

Audyssey is attempting to correct the lack of BSC. This is not a good way to do it.

Now we get my biggest complaint about Audyssey, it is correcting the top end to make it flat at the listening position. This should NOT be done and sounds awful.
For me this rules out using Audyssey frequency response correction.

So here you see what a commercial manufacturer is up against and the choices they think they are forced to make.

I highly doubt I would like that speaker. It definitely caters to the pop crowd, where these choices would be less adverse and some like the port thump.

All in all another good reason to learn to design your own speakers!

So, I copy and pasted what you said and asked the owner of Ascend speakers, David F. Here is what he replied with

This might be the most absurd thing I have ever read regarding our speakers, and shows a complete lack of knowledge about our products.

I could easily disprove every single comment this person made, but I really don't have the time. I really get a kick out of these DIY'ers who build a few pairs of speakers that *sound wonderful to their ears* and all of a sudden they think they know more or can do things better than engineers who do this fulltime and as a living. Just the fact that this person makes the assumption that our Sierra-1 NrT lacks baffle step compensation is laughable... Or that the "enclosure is improperly tuned causing a peak at 100Hz". The Sierra-1 enclosure is tuned to 40Hz. The peaks you are seeing on your graphs at around 100Hz is floor bounce. Even an amateur designer should know this. Raise the height of your speakers by about a foot and then see how it changes those peaks...

 

[TLS Guy]

While it is certainly possible that the 100 Hz peak is floor bounce, the the curves I posted and the explanation came from his company, on his blog.

[Overall curve is remarkably linear, coming in at an amazing +/- 0.5dB throughout the bandwidth of the speaker. (note, port tube output is not factored into the anechoic response, which will fill in the mild dip in output below 500Hz)]

I can tell you it is NOT kosher to fill in a dip below 500 Hz from port output!

What you posted in terms of the correction Audyssey is trying to make are more compatible with the curves from his site and my explanation

I would have him know that DIY speakers are among the best I have heard, but I have to agree I have heard some really awful ones as well, just like in the commercial world.

That is all I will say further about the matter absent measuring the speaker in question.

 

[Alex2507]

I really get a kick out of these DIY'ers who build a few pairs of speakers that *sound wonderful to their ears* and all of a sudden they think they know more or can do things better than engineers who do this fulltime and as a living.

I really get a kick out of a for profit company dishing dirt on DIY'ers. Last I heard, TLS was the Vice Chairman of AES' midwest chapter to start with.

The established and original DIY speakers I have heard at a couple of PETT DIY GTGs all sounded pretty good and some were exceptional. That might be because they were not at all concerned with profit and sometimes esthetics didn't factor in either.

One pair of speakers I have heard was voiced by ear. The designer/builder went through the trouble of getting them to some measurement gear in order to improve them. A designer's amateur status doesn't mean he's only got ears to voice a speaker with. The main difference between the hobbyist and professional is objective.

As the owner of Ascend, Mr. F. might want to look up Dale Carnegie. Making speakers that *sound wonderful to their ears* isn't that hard, especially when you are not answering to an accountant. DIY'ers are a collective and they collaborate ... and like anybody else, they don't like being categorically insulted.

The only thing that happened "all of a sudden" was me not being interested in supporting Ascend until they give me a free pair of speakers to make up for my hurt feelings.

 

[PENG]

The only thing that happened "all of a sudden" was me not being interested in supporting Ascend until they give me a free pair of speakers to make up for my hurt feelings.

Coming from you, never, I just don't believe you. I can partially agree with you but I can also understand why DF made that kind of comment (assuming he did). I will never posted any such conversation unless I have the explicit permission. Main reason is to avoid inadvertent distortion by omission, i.e. taken out of context and there are obviously other reasons.

 

[Alex2507]

Coming from you, never, I just don't believe you. I can partially agree with you but I can also understand why DF made that kind of comment (assuming he did). I will never posted any such conversation unless I have the explicit permission. Main reason is to avoid inadvertent distortion by omission, i.e. taken out of context and there are obviously other reasons.

I understand that I may have been played but would still consider taking possession of some Towers w/ RAAL tweeters.

I've been contacted via PM by a friend concerning a Troll Alert. Thanks to you too, PENG.

 

[billy p]

Coming from you, never, I just don't believe you. I can partially agree with you but I can also understand why DF made that kind of comment (assuming he did). I will never posted any such conversation unless I have the explicit permission. Main reason is to avoid inadvertent distortion by omission, i.e. taken out of context and there are obviously other reasons.

Excellent post Peng. I have sent Dave a PM. I hope he can find the time this holiday weekend to elaborate on some of things being discussed here or at some point in the up coming days.

 

[bizmord]

to avoid speculation, here is a link to the response

http://forum.ascendacoustics.com/showthread.php?5852-Re-Ran-my-Audyssey-Set-Up-on-Denon-4311-for-Sierra-1-NrT-look-at-this-crazy-Chart/page2

 

[billy p]

to avoid speculation, here is a link to the response

http://forum.ascendacoustics.com/showthread.php?5852-Re-Ran-my-Audyssey-Set-Up-on-Denon-4311-for-Sierra-1-NrT-look-at-this-crazy-Chart/page2

Dude you used a different alias at Avs & Ascend forums. You ask the same redundant question on every one of those forums...Now you have gone out of your way by using copy & paste remarks from other sources to create some sort of controversy. Really get a grip and stop trolling.

 

[bizmord]

Dude you used a different alias at Avs & Ascend forums. You ask the same redundant question on every one of those forums...Now you have gone out of your way by using copy & paste remarks from other sources to create some sort of controversy. Really get a grip and stop trolling.

Billy, different alias was not on purpose. Just different time when account was created. And, did not mean to start a controversy. Heinsight, maybe should have messaged Dave instead of posting publicly, that i agree on.

Sometimes we do things at a spur of a moment, i am sure we're all guilty at it. Didn't mean to upset.

On a separate note ... Turned off Audessey and sound is better. Less harsh heighs and more mids. My only concern is that maybe the sub is now not performing as good as it did. Not sure