That's simply not true (outside at least.) Without later arriving lateral reflections there is no "spaciousness" as defined by the renown authors of subjective acoustics. (Blauert, Kuttruff, etc.)
This is why artificial reverb does not work very well, because the simulated "reflections" do not come laterally, but from the same direction as the direct sound image. The reflections have to be spatially distributed as well as temporally.
Not really. I think that you have some misunderstandings about acoustics that you will need to get resolved.
As someone suggested look up Greisinger and his views on this topic. The entire idea of multichannel sound is to simulate lateral reflections. It's an improvement, to be sure, but still not a perfect solution. I prefer natural room reverberation to a simulated one.
This is mildly true, but the fact is that the loading does not have a strong effect on displacement. That's because loading, as an impedance, is much lower than the internal impedances of a high BL driver. So yes there is some effect on displacement from the resonances but not to the extent that one might expect.
The problem is that the discussion was about Camplo's waveguide, which does not have a diffraction slot. Hence Marcel's comments were on topic, but yours were not.
I'd like to think that I have way too much headroom to be worried about 115db peak performance, in a small room, at 1m-2m
Not sure.....
We went through conversation about scaling once before, I am trying to gain that clarity again...
At the moment my minds eye sees White noise and Pink noise.....Z scale and C scale. Some how this damn Fork drop is confusing me lol....I'm like; "Is this 115db transient of the fork drop in Z or C scale...
Furthermore, What does this Fork drop mean to me versus the gain levels I will monitor at..... I can likely conclude that Z scale is how the measurement was taken/displayed for the Fork drop, but where does that put me on the Volume knob is the question...
So from there it might make sense to drop back and say that transients can be 20-30db over average.....And to that I have to point out that average has not been clarified. It should be feasible to use Pink noise establish the averages as it can be seen as equal tonal balance to the ear. It also needs to be pointed out that these Peaks that maybe 20-30db over average do not happen in sub bass....
So probably the most important generalization that needs to be made is;
where do these 20-30db, over average, transients dominate, within the spectrum.....
I know that it tends to be in the mid to high frequency but I wonder does anyone have a more respectable and empirical stance on this. (3: capable of being verified or disproved by observation or experiment)
Do not forget that we long ago had the discussion about what transients can be....yes a transient can be low frequency. The thing is, Bass can commonly carry a very high average, much higher than we would ever tolerate from mid or high. Examples would be a bass heavy song or some type of bass dominated scene in a movie. What seems to be, is that, Bass, average and Peak, levels, have the same Max level. So you might have a Bass note Averaging at 115db or a Bass transient that Peaks at 115db.
What would have been helpful would be if Peak could be taken with REW, with scaling....We only have access to Zpeak unfortunately.
Balanced tone, the scientific definition of sound quality... will be the basis for most of the sound we here in music and movies.... In other words, the tone of things will not be extremely far off from this, generally speaking. Once again pointing to Pink noise for a metric.
Another inhibiting factor in this part of my journey is the lack of ability to see true peak vs spectrum....With the RTA, I clap, I see 120 on the spl meter reading of Zpeak....Yet I have no visual reference of where this 120db is on the RTA..... Pointless information without the whole picture.
How do I fix this, as in, where is the RTA that reads True Peak.
Not sure.....
We went through conversation about scaling once before, I am trying to gain that clarity again...
At the moment my minds eye sees White noise and Pink noise.....Z scale and C scale. Some how this damn Fork drop is confusing me lol....I'm like; "Is this 115db transient of the fork drop in Z or C scale...
Furthermore, What does this Fork drop mean to me versus the gain levels I will monitor at..... I can likely conclude that Z scale is how the measurement was taken/displayed for the Fork drop, but where does that put me on the Volume knob is the question...
So from there it might make sense to drop back and say that transients can be 20-30db over average.....And to that I have to point out that average has not been clarified. It should be feasible to use Pink noise establish the averages as it can be seen as equal tonal balance to the ear. It also needs to be pointed out that these Peaks that maybe 20-30db over average do not happen in sub bass....
So probably the most important generalization that needs to be made is;
where do these 20-30db, over average, transients dominate, within the spectrum.....
I know that it tends to be in the mid to high frequency but I wonder does anyone have a more respectable and empirical stance on this. (3: capable of being verified or disproved by observation or experiment)
Do not forget that we long ago had the discussion about what transients can be....yes a transient can be low frequency. The thing is, Bass can commonly carry a very high average, much higher than we would ever tolerate from mid or high. Examples would be a bass heavy song or some type of bass dominated scene in a movie. What seems to be, is that, Bass, average and Peak, levels, have the same Max level. So you might have a Bass note Averaging at 115db or a Bass transient that Peaks at 115db.
What would have been helpful would be if Peak could be taken with REW, with scaling....We only have access to Zpeak unfortunately.
Balanced tone, the scientific definition of sound quality... will be the basis for most of the sound we here in music and movies.... In other words, the tone of things will not be extremely far off from this, generally speaking. Once again pointing to Pink noise for a metric.
Another inhibiting factor in this part of my journey is the lack of ability to see true peak vs spectrum....With the RTA, I clap, I see 120 on the spl meter reading of Zpeak....Yet I have no visual reference of where this 120db is on the RTA..... Pointless information without the whole picture.
How do I fix this, as in, where is the RTA that reads True Peak.
Hello Camplo
Look up M noise as far real world requirements for PA applications headroom wise. The average is arbitrary depending on the circumstances. Clapping and expecting to see it on an RTA in real time??? Why don't you set your sample size higher to average the spectrum and slow down the visual response and just try repeated clapping to capture the spectrum. You should see the peak in the average spectrum.
https://m-noise.org/
Rob 🙂
Look up M noise as far real world requirements for PA applications headroom wise. The average is arbitrary depending on the circumstances. Clapping and expecting to see it on an RTA in real time??? Why don't you set your sample size higher to average the spectrum and slow down the visual response and just try repeated clapping to capture the spectrum. You should see the peak in the average spectrum.
https://m-noise.org/
Rob 🙂
The REW "Scope" (oscilloscope) has up to a 20 second buffer, so you can go back and visually observe the frequencies of the waveforms and their amplitudes after starting the SPL meter/logger and the scope.
Set at 1ms per division, a 1kHz waveform would take up one division.
Set at 10ms per division, a 100Hz waveform would take up one division.
Easy to see the difference between a "dead" little drum (dies out in just over 100ms), and the high frequency dominated hand clap, both peak at 120dB.
If you do the same with your real drum set, you'll start to get an idea of the difference between different SPL meter dynamics and peak frequency content.
Art
Camplo
You can also look directly at the recorded waveform and find the peaks. Any good software can then tell you the average. Then you know, for that sample, exactly what the peak pressure was (in a calibrated system - not so easy to do with low cost equipment.)
But here is the thing, I've done these tests with my system and here is what I conclude. 115 dB(C) of a live Cream concert sounds like its real life - level wise - ear splitting, not tolerable in the long run (hearing safety wise not sound quality wise,) and the sound never "lost it." I gave out before the system did. Where the peaks are is not really relevant with most music, (live being more demanding.) But remember that even on the recording side there are level limits, and they tend to knock down any excess (overloads) peaks in the recording as issued. Home mix stuff can probably be a little greater dynamic range, but not much I suspect. Your proposed system is far beyond mine in terms of power handling. I think that you are way over the top for what you need.
You can also look directly at the recorded waveform and find the peaks. Any good software can then tell you the average. Then you know, for that sample, exactly what the peak pressure was (in a calibrated system - not so easy to do with low cost equipment.)
But here is the thing, I've done these tests with my system and here is what I conclude. 115 dB(C) of a live Cream concert sounds like its real life - level wise - ear splitting, not tolerable in the long run (hearing safety wise not sound quality wise,) and the sound never "lost it." I gave out before the system did. Where the peaks are is not really relevant with most music, (live being more demanding.) But remember that even on the recording side there are level limits, and they tend to knock down any excess (overloads) peaks in the recording as issued. Home mix stuff can probably be a little greater dynamic range, but not much I suspect. Your proposed system is far beyond mine in terms of power handling. I think that you are way over the top for what you need.
There's a lot of research from Griesinger about how we hear/perceive, for instance what envelopment is etc. His body of work is a wealth of information.
It's about more than Multichannel alone I'd say.
looking at the Oscilloscope doesn't tell me frequency, nor does looking at the recorded waveform.
An RTA that is fast enough to capture True Peak is what I am after.
I want to see Peak information, through the RTA....there is no other way to put that lol....
Here we can see that my clap has reached 125db on the Peak meter yet on the RTA only reaches 75db....
How can I fix that?
An RTA that is fast enough to capture True Peak is what I am after.
I want to see Peak information, through the RTA....there is no other way to put that lol....
Here we can see that my clap has reached 125db on the Peak meter yet on the RTA only reaches 75db....
How can I fix that?
Try changing the mode settings, RTA 1/48 octave could well be hiding the peaks in the the 1/48 smoothing.
I fixed it!
You have to click "Adjust RTA levels"
So nooowwwwwaaaah, I can use my mic and play music and media content at whatever level I want to speculate and watch the RTA and see where the Peak information takes me in real time. The system can limit this activity if the volume exceeds its ability....
Whos going to be the first to develop software that compares and contrast source signal vs microphone signal to expose Compression and attenuation or any other distortion??? Maybe display the source signal superimposed over the Microphone signal?
Are there any recommended settings for anything else, you guys might suggest? Like for windowing???
Edit: Ok I didn't "fix it" all the way, but peak is showing much closer to the spl meter vs the rta after changing that option...it might be a matter of available windows at this point. Even more frustrating is that I don't get the same result every time I clap....Except for the Peak reading on the SPL meter...
You have to click "Adjust RTA levels"
So nooowwwwwaaaah, I can use my mic and play music and media content at whatever level I want to speculate and watch the RTA and see where the Peak information takes me in real time. The system can limit this activity if the volume exceeds its ability....
Whos going to be the first to develop software that compares and contrast source signal vs microphone signal to expose Compression and attenuation or any other distortion??? Maybe display the source signal superimposed over the Microphone signal?
Are there any recommended settings for anything else, you guys might suggest? Like for windowing???
Edit: Ok I didn't "fix it" all the way, but peak is showing much closer to the spl meter vs the rta after changing that option...it might be a matter of available windows at this point. Even more frustrating is that I don't get the same result every time I clap....Except for the Peak reading on the SPL meter...
Last edited:
This might help you understand your confusion between different results:
I think some may have a misunderstanding of FFT. You cannot see true peak levels with FFT because the FFT averages the values during the FFT interval. As Geddes said, the oscilloscpe is the way to see the true peak levels. That peak level contains the sum of all the frequencies in the measured bandwidth.
ZPeak reading is not an average...Im trying to Get my RTA to read True Peak, I am not using any averaging....the speed of the RTA even with Averaging turned off isn't as fast as the Spl Meter....thats about it.
Depending on the Windowing I use the I get more consistent readings on the RTA.....Rectangle seems to be the fastest, with the truest to actual SPL per test....
Depending on the Windowing I use the I get more consistent readings on the RTA.....Rectangle seems to be the fastest, with the truest to actual SPL per test....
As I said, I want to see True Peak with Frequency ..... If FFT can't do it, it just can't do it....
Peak is just a measurement of the highest signal that the device can capture regardless of where that highest signal may fall within the measured bandwidth...Thus not an average or sum
sounds like the lower I can get the FFT interval, the better. Likely why 8k gives me the best results.
Last edited:
Another option to find peak power used is to put a bridge rectifier on the amplifier output and feed it to about a 10 microfarad cap. Measure the voltage on the cap after playing a bit to guess what was the peak level.
Remember you are measuring a peak so you need to multiply by .707 to get RMS. Also add normal voltage for the rectifier.
Remember you are measuring a peak so you need to multiply by .707 to get RMS. Also add normal voltage for the rectifier.
"As the FFT length is increased the analyser starts to overlap its FFTs, calculating a new FFT for every block of input data. The degree of overlap is 50% for 16k, 75% for 32k, 87.5% for 64k and 93.75% for 128k. The overlap ensures that spectral details are not missed when a Window is applied to the data. The maximum overlap allowed can be limited using the Max Overlap control below to reduce processor loading at higher FFT lengths."
I think this is a better answer of what REW is doing and why 8k works best for my task.
After a quick trial with music, it seems that even 8k has too big of a sample window achieve my Goal....
To view Peak SPl vs frequency if anyone continues to miss that point.
Not every RTA is FFT is it?
I think this is a better answer of what REW is doing and why 8k works best for my task.
After a quick trial with music, it seems that even 8k has too big of a sample window achieve my Goal....
To view Peak SPl vs frequency if anyone continues to miss that point.
Not every RTA is FFT is it?
Last edited: