The dipole is far better below 100 Hz, but also better from 100 to 200 Hz, although room modes start to show up in that area. But still less than with monopoles.
I have tried to integrate dipole woofers with closed box subwoofers. It never sounded right, so I went the brute-force route and build large dipole H-baffles and went crazy with cone area (sixteen 21").
I have tried to integrate dipole woofers with closed box subwoofers. It never sounded right, so I went the brute-force route and build large dipole H-baffles and went crazy with cone area (sixteen 21").
@ Mandrake- i agree, and i do understand digital encoding and multiplexing well enough.
@Markus- yes i can distinguish 16 and 24bit recordings. Yes, cd is16 bit. However, i do not exckusively listen to 16 bit recordings, since i master in 24 bit 192 kbps. I agree that MOST would only hear 16bit and 'quality' is more due to DAC type and quality.
@Markus- yes i can distinguish 16 and 24bit recordings. Yes, cd is16 bit. However, i do not exckusively listen to 16 bit recordings, since i master in 24 bit 192 kbps. I agree that MOST would only hear 16bit and 'quality' is more due to DAC type and quality.
Last edited:
I'm not sure if a dipole can be "far better below 100 Hz" when compared to a properly placed and equalized multisub configuration. No usable output at very low frequencies (needed for home theater applications) and low efficiency (massive hardware requirements). Where's the benefit?
Here's my room with two subs, equalized with Audyssey MultEQ XT32 (left + sub, 8 locations as far as 60cm apart, red = main listening position):
Again, I would consider a dipole midrange if it had a proven advantage over a monopole but there is no data.
Better frequency response without EQ; does not trigger standing wave modes. And this also results in a cleaner and shorter decay, which indeed is audible.
This is how it measures with dipole compensation EQ and XO only. No room correction EQ at all.
Attachments
Last edited:
by recording in 24 bit, 192kbps. Mastering 'copy' down sampled to 16 bit 48kbps. Granted that by doing this, sampling rate is also an influence on the sound. Playback through each channel, eg 24 bit in RHS ear, 16bit in left.
Is this at the main listening position? Why only one location? How good is the frequency resolution? Any smoothing applied?
Mono bass or stereo? If stereo then you would need to show left, right, together separately.
Last edited:
Because I listen only in one location. And of course the response changes when I move several meters, because of boundary reflections.
Measurement is taken with one channel only. The other channel, and both channels together measures the same, within less that half a dB.
Sadly, there is smoothing applied. I lost the measurement data in a harddrive crash sadly, and the only thing I have is this graph. Let's see if I get the urge to do a new measurement ..... 😉
You played it mono through two different loudspeakers? And I guess this was a sighted test?
But you do have two ears, do you? And they are mounted about 20cm apart from each other?
Not in my setup. And, you don't move while listening?
Yes, isn't that sad again 🙂 Sorry but your graphs doesn't show what you're trying to show.
I will try to do some measurements tomorrow, I hope that will clear things up. 🙂
Like to add that I usually sit still when listening critically.....
Like to add that I usually sit still when listening critically.....
yes markus, mono. One sample in left, one in the right earpiece. It wasnt blind in the sense that i was alone, eyes open. I did not know which sample was which so i didnt have a pyschosematic pre emptive answer. As i said, sample rate was also different, so at least some of the difference would be due to that. As an aside, all modern soundcards are 24bit at least 96khz, why? Maybe DVD 5.1 requires it? Digi TV sound is appalling! Must be 3bit ladder DAC or something equally crude, with piano notes i can hear the quantisation 'stepping' amplitude errors very obviously. That side of the digital encoding is where my knowledge ends.
Last edited:
Greatly appreciated 🙂 Frequency resolution should be about 1Hz with no smoothing.
Nevertheless you should measure multiple locations to show how smooth the response really is. If dipoles categorically show less detrimental modal effects then the frequency response across a certain area should be very smooth, right?
Lots of variables you first would need to control for a proper listening test...
What software did you use for resamping and on what settings ?
Any type of resamping has at least some generational loss, as it is not a bit for bit copy of the original. So even assuming that 16 bits at 44Khz is enough, you are still comparing an "original" with a generational copy, which is not a fair comparison.
A more valid test would be feeding the original analogue source material into the best 16 bit 44Khz A/D you can find, and also feeding it into the best 24 bit A/D you can find. Of course that is not going to happen because the source material is nowadays sampled at higher rates, mixed that way and down sampled at the end of the chain.
So just how good or bad are resamping algorithms ? You might find this thread a real eye opener:
http://www.diyaudio.com/forums/ever...ads-why-they-make-no-sense-2.html#post2937482
There is a link to the following site where comparisons of sample rate conversion quality of numerous free and commercial software packages are made, and the results vary from excellent to downright awful, with almost no correlation to the price of the software either....(my guess is that programmers in general simply dont have the digital signal processing theory background to know how to do the algorithms properly, and make some elementary mistakes...)
SRC Comparisons
It could easily be argued that the difference you hear is the loss in quality caused by potentially substandard resamping software.
Last edited:
...or a difference in the earpieces or a difference between the left and the right ear or...
Of course, but even if everything else is controlled, you still have the issue of signal degradation due to imperfect resamping if one test signal is the original and the other is a conversion of the 1st, instead of having both be direct samples of the original analogue signal.
It's as if you did a comparison between LP and tape, by mastering the LP from the tape that you're comparing it to. There is no way the LP is going to sound better, or even as good in that circumstance...
I didn't want to express any disagreement, just wanted to add to your comment. The way he did the test, he cannot conclude that a difference between 24bit and 16bit is audible. Let alone 18bit vs. 24bit at low frequencies which was the original claim.
sorry for the out of topic but if you could help me :
are any of these any good for working with REW (connect to mic input on PC) ?
http://www.ebay.com/itm/Measurement..._Speaker_Parts_Components&hash=item46062fa6e8
http://www.ebay.com/itm/Behringer-E...451?pt=LH_DefaultDomain_0&hash=item564bdcab63
are any of these any good for working with REW (connect to mic input on PC) ?
http://www.ebay.com/itm/Measurement..._Speaker_Parts_Components&hash=item46062fa6e8
http://www.ebay.com/itm/Behringer-E...451?pt=LH_DefaultDomain_0&hash=item564bdcab63