Cheapest Full Range 20hz - 20khz Speakers?

[andreasmaaan]

If you have Foobar and a few extra minutes, I'd definitely suggest following @detlev24's advice and trying with Foobar's ABX comparator.

Please also send me other samples of music if you'd like them processed. I would suggest something with vocals or sustained sounds rich in harmonics would likely give people a better shot of identifying differences than percussive samples like the last two.

Also, @Duke gave me the idea of posting the filter's response. Here it is:

 

[andreasmaaan]

Also thought people might be interested in seeing the group delay and step response (note that these are estimated in VCAD so are not 100% accurate, but are very close):

Group delay:

Step response:

Step response (zoomed):

 

[tuga]

Even better! @tuga @Duke, keep in mind you'll have to do this a bunch of times and correctly get it right in order to establish an actual confidence level. Good luck!

Do you know of an ABX software for MacOS? I can't think of a more tedious task than comparing snippets but it would make more sense to do it properly...

 

[andreasmaaan]

Do you know of an ABX software for MacOS? I can't think of a more tedious task than comparing snippets but it would make more sense to do it properly...

I think these should work (don't have a mac to test it on though):

https://www.foobar2000.org/mac

https://www.foobar2000.org/components/view/foo_abx

 

[detlev24]

Do you know of an ABX software for MacOS? I can't think of a more tedious task than comparing snippets but it would make more sense to do it properly...

foobar2000 for macOS, which still is "work-in-progress", did not support any plug-ins. This might have changed in the meantime, though.


Here's one that works cross-platform, on Linux\macOS\Windows:

Lacinato ABX/Shootout-er or web-based: Lacinato WebABX/Shootouter

 

[tuga]

This might not be audible but wouldn't you rather not have it altogether?

from "Perception of temporal response and resolution in the time domain" by Michael J. Turner (too big to attach)

 

[andreasmaaan]

This might not be audible but wouldn't you rather not have it altogether?

from "Perception of temporal response and resolution in the time domain" by Michael J. Turner (too big to attach)

How is that not equivalent to looking at DAC spuriae 120dB below the signal and saying “These might not be audible but wouldn’t you rather not have them?”

 

[detlev24]

[...] from "Perception of temporal response and resolution in the time domain" by Michael J. Turner (too big to attach)

 

[tuga]

How is that not equivalent to looking at DAC spuriae 120dB below the signal and saying “These might not be audible but wouldn’t you rather not have them?”

What is the chance of it it producing audible artefacts through some form of "intermodulation" (perhaps not a correct term)?

 

[vavan]

What is the chance of it it producing audible artefacts

audible but not measurable?

 

[andreasmaaan]

What is the chance of it it producing audible artefacts through some form of "intermodulation" (perhaps not a correct term)?

I can’t think of any equivalent process to intermodulation in this context.

The research suggests this kind of distortion will be most likely to be audible in the following scenarios:

• Shift from linear phase to distorted phase (or indeed any kind of phase shift) happens in real-time (not applicable to loudspeaker phase distortion)
• Listener is wearing headphones or listening in an anechoic chamber
• Signal is a square wave or impulse (or one of various artificial signals)

Like anything in psychoacoustics, audibility is somewhat listener, stimulus, and environment dependent.

The research suggests that, if you can’t hear a particular degree/kind of phase distortion with a particular signal on headphones, you won’t be able to hear it in-room with speakers.

 

[pozz]

pdf

Thanks for posting it.

Summary:

• The topic in the presentation is specifically about LF, and the author notes that most LF designs on the market are bass reflex and there is a penalty involved, but has no strong experimental evidence for subjective impact. Concludes that active designs and servo control are a good fix for the technical problem.

Not an entirely foreign conclusion to our parts. We do have Rythmik and Grimm as good examples of thoughtful bass tech. There is only so far that you can push an LF driver without getting into feedback mechanisms.

 

[tuga]

The topic in the presentation is specifically about LF, and the author notes that most LF designs on the market are bass reflex and there is a penalty involved, but has no strong experimental evidence for subjective impact.

At least I'm not alone, there's another fool speculating on the issue...

 

[richard12511]

I haven't been able to get Foobar working on Mac OS . Any suggestions for a good(cheap) windows HT PC? I've been meaning to buy one anyway.

 

[Bear123]

Even if you found speakers that play down to 20Hz or lower, who cares? Its optimal to cross them to a good pair of subs around 80 Hz anyways for improved sound quality. My inexpensive tower speakers ($1400/pair) play down to 25 Hz in room easily when run full range. However, I would never give up the fidelity and improved sound quality that I get when crossing them to good subs.

 

[raistlin65]

Even if you found speakers that play down to 20Hz or lower, who cares? Its optimal to cross them to a good pair of subs around 80 Hz anyways for improved sound quality. My inexpensive tower speakers ($1400/pair) play down to 25 Hz in room easily when run full range. However, I would never give up the fidelity and improved sound quality that I get when crossing them to good subs.

I know what you mean. I used to have towers. Now I have Sierra 2EX with two PSA V15 series subs. The PSA's put out fantastic bass below 80hz.

 

[richard12511]

Even if you found speakers that play down to 20Hz or lower, who cares? Its optimal to cross them to a good pair of subs around 80 Hz anyways for improved sound quality. My inexpensive tower speakers ($1400/pair) play down to 25 Hz in room easily when run full range. However, I would never give up the fidelity and improved sound quality that I get when crossing them to good subs.

Judging by the F328Be review, it seems we still have many that disagree with this.

 

[andreasmaaan]

Just another little post re: the phase problem we've been discussing (really should have started a new thread for this).

Some of you might be questioning whether the phase response of the filter I used to convolve the audio samples with is truly representative of a real bass-reflex speaker tuned to 40Hz with an LR4 XO at 1kHz.

I don't have exactly such a speaker to measure and compare. However, I do have phase meaurements (taken using the ground-plane method and ungated/unsmoothed) of a 3-way bass-reflex speaker tuned to 30Hz with LR4 crossovers at 150Hz and 1250Hz.

I've used the same software I used to generate the FIR filters with which I convolved those two recordings to generate a filter with the predicted phase response of this speaker.

Here is the predicted phase response:

And here is the measured phase response (NB: measurement stops just above 2kHz as this was done using the ground-plane method and the high-frequency driver was not connected):

The correlation between the model and the reality is excellent between 30Hz and 500Hz. Note that the fact that the HF driver was not connected likely explains the less perfect tracking above 500Hz.

Just posted as I think it's important to demonstrate that the model we're using matches reasonably well with reality.

 

[detlev24]

I haven't been able to get Foobar working on Mac OS .

Have you given this a try?

Any suggestions for a good(cheap) windows HT PC? I've been meaning to buy one anyway.

If you are interested in a "barebone", which would save you a lot of money, I can suggest an 'ASRock DeskMini X300'. The minimum you need to buy additionally comprises of:

• APU
• RAM
• SSD
• WLAN module (e.g., "Intel WLAN-AC-9462")
• furthermore, it is advisable to upgrade the cooler, as even the slowest APU: "AMD Ryzen 3 4350G" will make the stock one get noisy quickly.

 

[raistlin65]

Judging by the F328Be review, it seems we still have many that disagree with this.

The room doesn't always cooperate with bass output for where you want to put your speakers relative to the listening position. Whereas you can always move your subs.

That being said, it would be interesting to see CEA 2010 data for the F328Be to really find out how it performs with bass.