The King of Bass is Dead, Long Live the New King

Well guys, the results are in. Comparable testing of the W15GTI vs a 2205 recently remagnetized and reconed (in Dec 2005) as a 2235 show the W15GTI the clear and present winner in the reproduction of bass. Furthermore, the W15GTI has a more extended high end.

Both speakers are in the same enclosures and there is no reason that the results wouldn't be directly comparable.

It is also interesting to note that the W15GTI has a nice "hump" at 60 Hz whereas at that point the 2235 is starting to loose ground. From my experience with WinISD, I recognize that with the W15GTI, this hump could easily be smoothed out by further lowering the tuning of the cabinet which would further help the W15GTI go lower, and flatter.

Sorry that the format is a little hard to read. It's the fault of the web site tha twill not let me space the values, even though I've added blanks between them. If anyone would like to see the data, I have it in an excel sheet.

Frequency W15GTI 2235 Difference
Hz dBL dBL dBL

1500 91.9 84.5 7.4
1450 93.7 78.9 14.8
1400 95.6 87.9 7.7
1300 96.4 89.2 7.2
1200 98.8 90.5 8.3
1100 97.9 94.2 3.7
1000 92.3 101.4 -9.1
950 93.1 101.8 -8.7
900 95.3 101.9 -6.5
850 98.1 99.0 -0.9
800 103.0 98.3 4.7
750 101.2 98.7 2.5
700 100.0 96.6 3.4
650 99.1 98.2 0.9
600 98.7 99.1 -0.4
550 98.1 100.8 -2.7
500 98.8 100.1 -1.3
450 98.8 99.3 -0.5
400 97.9 99.6 -1.7
350 99.0 98.9 0.1
300 99.2 99.8 -0.6
250 100.2 100.0 0.2
200 100.0 100.0 0.0
180 100.0 100.0 0.0
160 99.8 100.5 -0.7
140 99.5 100.3 -0.9
120 101.5 101.4 0.1
100 101.3 100.7 0.6
90 101.6 101.8 -0.3
80 102.3 101.5 0.8
70 103.1 101.6 1.5
60 104.3 100.9 3.4
50 103.8 99.5 4.3
45 101.0 96.9 4.1
40 98.0 90.4 7.6
35 97.0 89.7 7.3
30 96.4 89.6 6.8
25 95.4 90.4 5.0
20 93.4 89.3 4.0

http://www.largescaleonline.com/eima...235-w15gti.jpg

Hi

- Interesting ! Nice to see these comparisons being posted .

May I suggest that you ;

- Post the Excel spreadsheet as a ( zipped ) attachment .
- This way someone can take that info and inturn display a viewable chart ( using Excels' builtin charting capabilities ) . I know a couple of folks around here, who are proficient at that ( including displaying in a logarithmic format ) .


:)

I'm sure others will chime in with technical questions to validate your results, but I'm just going to state that this is no surprise to me. That's one p-fine driver. :yes:

Quote:

---

Originally Posted by Earl K

Hi

- Interesting ! Nice to see these comparisons being posted .

May I suggest that you ;

- Post the Excel spreadsheet as a ( zipped ) attachment .
- This way someone can take that info and inturn display a viewable chart ( using Excels' builtin charting capabilities ) . I know a couple of folks around here, who are proficient at that ( including displaying in a logarithmic format ) .


:)

---

Anyone who wants the spreadsheets with the raw data is welcome to them. Just e-mail me at [email protected].

Maybe you could keep us posted, Todd, on how they continue to behave over time. Your 2205/2235 is fairly new...how many listening hours do you think you have on them? The 2235 might wake up a little more down low with more break in. Just a thought.

Quote:

---

Originally Posted by Earl K

- Post the Excel spreadsheet as a ( zipped ) attachment .

---

:yes:

Quote:

---

Originally Posted by edgewound

The 2235 might wake up a little more down low with more break in. Just a thought.

---

My 2245's sure did - far, FAR beyond my expectations.

Simple chart of the results...

:thmbsup: :cheers: :applaud:

Cool Bo!

- You are definately my hero ( at least for the next few hours ).

:p

Quote:

---

Originally Posted by Earl K

- You are definately my hero ( at least for the next few hours ).

---

Nice! I'll take it!

I love data analysis - although this ain't much to work with. I actually prefer the scatter plot - although it doesn't quantify the difference in response, it shows overwhelmingly that the results - as measured - favor the W15GTI.

ummmmm,

- You may need to walk us mere mortals through the basics of scatter plots . :D

- Also, triangles vs squares ( or circles ) might be more demonstrative ( along with which is which ) . :)

;)

Quote:

---

Originally Posted by boputnam

I love data analysis - although this ain't much to work with. I actually prefer the scatter plot - although it doesn't quantify the difference in response, it shows overwhelmingly that the results - as measured - favor the W15GTI.

---

Please don't take this the wrong way, but I have a real problem with this thread's premise... Here we are comparing a subwoofer to a remarkably well balanced woofer. Other than seeing "as measured" that the subwoofer does indeed have greater output we really don't know anything about the relative quality of the two drivers in question. You probably could get a cheapo no-name sub from Parts Express and have similar results... that doesn't mean that it is a better woofer.

That said, I wouldn't be at all surprised if the W15GTI is a better subwoofer... the point is that the 2235 isn't a subwoofer. It is a woofer that is so capable that JBL found it acceptable to use it all the way up to 1000Hz in one application and use it as a sub in another... very few drivers are that capable.

Widget

Quote:

---

Originally Posted by Earl K

- Also, triangles vs squares ( or circles ) might be more demonstrative ( along with which is who / like a legend ) . :)

---

I'm with Widget on this one...

But, to your question - the scatter plot are the paired data, by frequency.

For example:

for 1450 Hz, X = 93.7 and Y = 78.9, thus the point will plot below the Y = X line, biased toward the W15GTI

Quote:

---

Originally Posted by Earl K

- You may need to walk us mere mortals through the basics of scatter plots . :D

---

One of the best uses of a scatter plot is for comparing agreement (correlation) between two data sets.

Here, we have data from two woofers. Plotting the pairs for each frequency gives the scatter plot. If the paired data were to cluster exactly along the Y = X line, then the data are highly correlated (R^2 = 1.0). Anything less than R^2 = 1.0 evidences less correlation.

These data - as measured - DO NOT cluster about the Y = X line, therefore are very dissimilar (FWIW, the R^2 = 0.32), and are strongly biased (cluster) toward the X axis.

Mind you, this analysis says nothing about the quality of the data - it merely analyzes the data, as is. Data integrity and validity is a whole 'nother discussion. I for one, would like to know more about how the data were collected... :)

In the car audio world, "woofers" are almost nonexistent. The nomenclature almost excludes them by definition. If you peruse JBL's car audio pages, the largest "woofers" are either the 6.5" diameter drivers as part of coax or triax or component systems, or 6"x9" woofers in coax or triax designs. Anything that's 8" or bigger is a "subwoofer."

So the naming conventions must be considered.

As such, the W15GTI is listed as one of the largest subwoofers, yet an examination of its characterisitics is instructive.

http://jbl.com/car/products/product_...at=SUB&ser=GTI

It's a sub that goes to 1kHz at the high end or down to 16 Hz at the low end. It'll handle 800W with 5000W peaks. To really get the measure of this beastie, plan on having a solid 500W just to warm it up.

As I wrote earlier, I'm not at all suprised it can whack the nuts off a 2235, and I think that given enough power it can beat the 2235 at both the low and the high end.

Now, whether you'd like the sound is another question. My opinion is that in a blind test, most people would pick the W15GTI, but in a sighted test, it'd probably lose, due in part to everyone knowing it's NOT a pro or high end consumer product--just a freaking low rider, ghetto blaster, ass-thumpin', tuner freak car speaker.

Really, it's so much more. :bouncy: