Official JTR speaker thread - Page 108

why is there no frequency response available for the jtr speakers?

is it that the typical on axis response will scare customers away even though the power response is good? i am really curious about the trip 8 and trip 12.

both speakers have a tremendous amount at work behind them. they do many things right. where are the distortion vs. frequency plots?

i am a big proponent of high sensitivity and controlled directivity--both really matter.

for realistic level playback, most folks should just forget low sensitivity brands and get something like the trip 8 or trip 12...

---

more curious than *why* folks go jtr, i'd be curious to know why folks *don't* go jtr

Good question, one that's been asked before. It's certainly possible I've missed it, but I don't even recall ever seeing an owners in-room FR graph.

I was at RMK!'s home theater gathering a couple years ago, Keith Yates was there and someone asked him to share his "house curve" (how he eq's a system). Keith said that he would share it if they would pay him for his Stanford degree and his 25 years of research.

Very few large audiophile speaker manufactures (Krell, Wilson, Goldmund, etc) display graphs because they have spent large amounts researching and developing their ideal response curves. Yes, someone could measure the response but then they wouldn't know "why" it looks the way it does. The frequency response is only one fragment of what makes a speaker sound good. What makes a subwoofer sound tight? What makes a subwoofer sound muddy? You certainly won't see these characteristics in a response curve.

I remember that conversation well. When Jeff Meir EQ'd the audio system he was looking at a response curve but did not give me a copy of the FR in graph form. He did give the the corrections by frequency and I have that data and would be happy to share it. I also have my Audyssey Pro graphs that I could/would share.

Of course. these (with the exception of Audyssey) are single point in space measurements and as well all know, a slight movement of the mic can produce very different results. Anechoic chambers are the best way to establish the base line FR of a loudspeaker and take the room out of the equation.

I suppose there will be some probing comments to Jeff's post but really, it does ultimately come down to how these speakers sound to us in our rooms. I believe Jeff would like to have professional measurements done for all of his products but that is an expensive and time consuming proposition for a small manufacturer.

I can only say how they sound to me in my room. Having had some pretty good speakers to compare them to, and some very experienced ears listen to the system, my positive feelings have been validated. Are there better sounding speakers out there? No doubt there are, but I am really happy with what I have and for the first time since I began this journey in early 2003, I have no desire to upgrade.

Take it for what it's worth ...

I actually agree with most of that and I respect your opinion, but anechoic FR curves and polar charts do provide an objective means to evaluate a speaker prior to purchase. Combined with subjective evaluation, ie- listening, the two are better than either alone for those of us that may not have the ability to confidently evaluate accuracy by ear alone. As well, they give an idea of what there is to work with when one is considering the room the speakers will reside in and what challenges may be ahead for EQ and application of house curves. It is no different than choosing the proper sub, ie - sealed vs. ported for the room it will be used and the FR that is required. If we never measured subs how could we know that information? Regardless, EQ and house curves are entirely different subjects than raw speaker response and the two should not be considered together in the context of this discussion IMO.

As to the "why" a speaker measures as it does, it is through listening and measuring that helps us learn what matters and what doesn't in frequency response graphs and polars. It is but one piece of the puzzle when evaluating speakers. But it is certainly not a useless piece.

I suppose by your response you consider the measurement of your speakers proprietary data?

And to be clear, I don't doubt the SQ and build quality of your speakers as attested to by many, many owners on this forum - nor do I think you're hiding anything - its just that I am a believer in measurements too

I understand where you are coming from and also respect your (and many others) opinions on this. I think Jeff's simple point is that a ruler flat FR does not a good sounding loudspeaker make. It is however interesting data to have once you understand what that means in terms of SQ. That subjective interpretation of the data is the important stuff. Really a right brain vs left brain discussion don't you think?

I just upgraded to Audyssey Pro and spent yesterday doing the measurements and tweaking. Audyssey was able to get my JTRs completely flat. After Audyssey was done all of the speakers worked very well together, did I like them flat, no. I put in a small boost for 300Hz and below and brought down 12kHz a bit and applied this to the Audyssey. While that may bother the flat curve purists I love the sound I am getting 'for me in my room'.

Did you measure independently of Audyssey after the calibration? As you may or may not know, the post-calibration graphs in Audyssey Pro are simply target curves generated by Audyssey based on what it thinks the applied filters will achieve. It doesn't mean that is what actually happened. Independent measurements are the only way to know the actual in-room FR after the calibration.

Frankly, I'll take Audysey's target curves as at least they are averaged over a larger (more realistic) listening area. The in-room single point in space based measurements are for me (in a 7 seat theater) ... useless. Sometimes I recline, sometimes I sit up straight, sometimes I actually change seats .

If I wanted a single chair, head in a vice, two channel setup then maybe.

I haven't run an independent REW to see what it is like at a single LP. I know that Audyssey's corrections are based on a listening bubble so I shouldn't get a completely flat response at a single LP. I can definitely tell when I am in and out of the bubble though.

I am going to be working on getting my subs eq'd properly with the DCX, still need to order some cables. When I am working on this I will check out the FR at a couple listening positions.

Sorry, I can't agree with that. So what you're saying is that if Audyssey post graphs were ruler straight and you took independent single point measurements at points where you might reasonably have an ear when listening and there was a consistent deep 15 dB null at 100 Hz at each or several positions, you would consider that ok?

Personally, I would then use that data to move my subs and/or mains, adjust sub distances and perhaps crossovers to eliminate the null. Then re-run Audyssey and confirm it's gone. To think that would not be audible is silly IMO. Audyssey is excellent IMO, but it is not going to fix that regardless of what the graphs say. The Audyssey post graphs I've seen *always* look great.

This pretty much corresponds to what I have read, Audyssey can flatten the response but cannot fix a null. The null(s) are going to be what I look to fix when I start doing the REW FR sweeps. Hopefully there won't be many and I can fix them with little position adjustments.

Hopefully . I have improved existing calibrations and SQ dramatically by taking independent single point measurements and positioning subs and mains to get the best FR at as many points as possible, then re-running Audyssey. Of course, rooms are unique and YMMV.

Any change made to accommodate a single point measurement will have an impact on other locations. What you described is an ad hoc multi position methodology that will at some point yield an average response across those multiple locations. Outside of speaker placement and room treatments, this would require using a separate EQ before running Audyssey (not a good idea IMHO). I find that Audyssey works pretty well by itself and their results have been validated by other sources.

To each his own my friend ...

My point is you can't use Audyssey to optimize your speaker positions in the room as there are no post EQ measurements performed by Audyssey. Audyssey does not do a good job when faced with poor in-room placement of speakers, particularly subs. Of course many systems with poor placement will have SQ improvement with Audyssey, but it's like polishing a turd - in the end it's still a turd. Whether any particular room will see a dramatic improvement depends on the problems presented by the current speaker positions and whether there are options to place them in better positions, but optimal positioning has the potential to have a far greater overall impact on SQ than any EQ solution. Even Chris from Audyssey recommends optimizing room placement for the speakers prior to calibrating since no EQ system can fix significant nulls, or even peaks for that matter. All EQ systems are limited by headroom constraints.

I don't have any problem with you not wanting to independently measure and optimize your speaker positions and rely on Audyssey to do the best job it can - as you say, to each his own . But to say that Audyssey can make a room with poor speaker positioning sound as good as one that is optimized is simply not true and I've never read any source that would validate that position.

But this I think is OT at this point so I will politely bow out and agree that we will disagree. Thanks for the discussion.

Just so that you don't get the last word ...
There is much conjecture in your post regarding my approach. My system has been measured several times by qualified people. In some cases, my speaker locations were dictated by other considerations (AT screen etc). I was fortunate that the room and my acoustic treatments allowed me to use the more convenient and in some cases, esthetic locations for the loudspeakers and subs.

I have no problem with measurements and would never try and dissuade anyone from doing this. But there are many variables at play here and getting it really "right" requires knowledge and precise equipment.

A basic understanding of correct speaker placement, boundry issues, crawling for bass and then Audysey might get you as close to perfection as is possible without engineering a room and having a real pro measure/tweak it. REW, an SPL meter or cheap mic may not help much and may in fact exacserbate some problems.

But it is a hobby and trial and error are great teachers.

You just *had* to get the last word!...

I could have swore they was a response graph on a set of Jeff's speakers. Though now that I think about it, it may have been the Cap.

I was playing around with the XTZ Room Analyzer Software last night and here are screen dumps of the room graphs. These are the "high resolution" graphs of <250hz.

The room response looks pretty good but I am unclear on the Spectrogram waterfall...
Measured Response


Result ()


RTA graphs

With Audyssey engaged



And no EQ

RMK!, I believe "result" is if you were to apply the recommended PEQ to the room modes it found in your measured response.

Makes sense ...

Here is another interesting XTZ graph:

Full Range 80db:

Interesting. . . below 4k looks nice. I am being lazy so I will ask instead of looking up, do you use an acoustical screen? I have heard that acoustical screens will interact with the upper frequencies and display what you are seeing.

Yes, I use an AT screen but it's affect is 1 or 1.5 db across all frequencies. If you look at the RTA graphs they show a reasonably good response up to 12K or so. AT that point Audyssey is doing a high frequency roll off and that shows in the graph.

These single point measurements can be deceiving (especially the first attempt). I plan on taking more tomorrow and see how things look from slightly different locations.

I've never seen RT60 displayed like that, but it's pretty cool. I could be reading it wrong, but I think the initial pulse has got to be at a higher dB since it appears you're so close to the noise floor that you have insufficient sensitivity to get any useful information as to reverb. Again, it's possible I'm reading it wrong, but that's my take on it.

That's not a FR graph, it's RT60. The 80 dB is simply the scale and as I said, the initial pulse is, I believe, too low to be useful.

Have a question for any 2009 triple 8 owners.
I have a line on three of these fine speakers and i was wondering what your
impressions are after having these speakers for three years or more?

Of course those of you who have sold them and moved on or upgraded please feel free to chime in.

I emailed Jeff and he said they were "nice speakers" and that there were many
happy users of "older" JTRs
I would love to hear from you "happy" owner/users of this particular vintage.

Also, my AVR is a pioneer elite SC-37, I know there are some conflicting opinions on whether the ice amps in them can handle 4 ohm loads.
Any elite users with JTRs please comment.

cheers

I think you are correct. The XTZ Manual recommends taking these measurements at close range (0-50cm) for the lower frequencies (20-500Hz). Also. my chosen mic location 4ft off the floor at the primary LP was less than ideal (too high) so they really don't tell much. The XTZ is an interesting product and although easy to use, in the hands of a novice the results can be inaccurate or misleading.

The latest Triple 8's use a new CD and crossover and do provide more resolution/detail in the high frequencies. So for music priority systems, they would be preferred. For HT applications they are too close to call (IMHO).

No problem . There's no way to learn it if you don't start taking measurements. I'm certainly a novice with my Omnimic, but perhaps just slightly farther along the learning curve than you. So don't think you're getting feedback from an expert .