- Joined
- Oct 20, 2019
- Messages
- 353
- Likes
- 434
No. That is laughably wrong. I suppose a 100 watt bulb would also light to equal brilliance when presented with a waveform
of the same peak voltage and greatly differing duty cycle. How do dimmers work?
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Amount of power needed for dynamic peaks?
- Thread starter Trdat
- Start date
No. That is laughably wrong. I suppose a 100 watt bulb would also light to equal brilliance when presented with a waveform
of the same peak voltage and greatly differing duty cycle. How do dimmers work?
Again , wrong analogy , the 100 watt bulb will consume less rms power when presented with a lower duty cycle signal , it will also be dimmer, if you want to get the same brightness and power consumption with the lower duty cycle you will need to increase the voltage (peak) which will stress the bulb more.Normal people do not hookup a dac without a volume control directly to an amp. Just like rms power for the same brightness of a bulb can be changed by varying the duty cycle and amplitude , SPL can be changed by varying the compression (duty cycle) and volume/gain ( full scale power) You are making a fool of yourself.
Last edited:
The highs are compressed/limited around the bass peaks
No volume changes by me.
Track before was Daft Punk RAM, track 1
Here is Michael Jackson Dangerous, track 1:
It is counter intuitive. High pass filters tend to increase the crest factor (the ratio between peak voltage and RMS voltage), thereby often increasing the heights of the peaks, even though the high passed signal is lower power (i.e. have a lower RMS voltage). Below is taken from ANSI/CTA-2034A.
Thanks, makes sense and points out a problem with overly compressed music.
- Joined
- Oct 20, 2019
- Messages
- 353
- Likes
- 434
That is what I said. Lower duty cycle=dimmer, ie less power. The higher duty cycle in the first graph requires more power.
What did you think I was saying? Dimmers work by reducing the duty cycle...which is what a much less compressed track
would be compared to the first.
You are acting as if volume controls don’t exist. For a given SPL the volume control for the compressed track would be lower. You are confusing the digital signal for what the amp must do at a given listening level.
I think you summed it up best in another thread:
cistercian said:
“Your ignorance should be embarrassing. Dunning Kruger much?”
- Joined
- Oct 20, 2019
- Messages
- 353
- Likes
- 434
I made no such assumption. I was commenting on the ridiculous compression of the first track.
It is a waveform display. My comment was simply related to how much
power it represented. Compare the first track and the second. Which represents more power?
That's all.
Agreed that the first track is pretty compressed which I also find undesirable . You’ve been conflating power in the digital signal with power requirements for a power amp. If you are not assuming that the volume (system Gain) is fixed then they are not the same.
Here is what you said that started this:
“Look at the average power in the first track...like a TV commercial. Max compression! You need serious RMS power to
play that loud. Yuck.”
meanwhile you’ve been telling me to look things up and that I’m laughable wrong. What kind of serious RMS power were you talking about:
1) digital signal power? This makes no sense
2) amplifier power? I explained why for a given SPL (“play that loud” in your post) the amplifier would not be more taxed. In fact for a given average SPL an uncompressed track would tax the amplifier more because while the average power would be approximately the same the peaks would be far higher.
why don’t you go back and re-read your posts?
- Joined
- Oct 20, 2019
- Messages
- 353
- Likes
- 434
I understand what you are saying now. I strongly agree a less compressed track would involve much higher peak power.
I was simply stating (very unclearly) that the first track would be similar to a sinewave or other high average power
hence yuck...because it would just be loud. The compression is obnoxious. It is why I cannot stand broadcast FM...too
much compression.
My statement about SPL assumed that a dynamic track with the same peak power and hence same peak SPL
would tax an amp less than a ultra compressed track or sinewave would at the same peak SPL...but I am not sure
peak applies to the mess in track one. I am sorry I was unclear!
- Joined
- Oct 20, 2019
- Messages
- 353
- Likes
- 434
Horrible day at work + wine = poor clarity. Again, sorry. Lots of people fired today. Some of them
just bought houses. I am a bit rattled.
Have a good night.
just bought houses. I am a bit rattled.
Have a good night.
Alcohol is a lethal mix with audio science.
With Jazz and Blues and Rap and Country music, ok.
In bars during a pandemic, lethal too.
That's where the peaks came from, high dynamic energy flow.
With Jazz and Blues and Rap and Country music, ok.
In bars during a pandemic, lethal too.
That's where the peaks came from, high dynamic energy flow.
Last edited:
Data point:
56th Annual Grammy Awards
Production
Best Engineered Album, Non-Classical
Random Access Memories
Peter Franco, Mick Guzauski, Florian Lagatta & Daniel Lerner, engineers; Bob Ludwig, mastering engineer (Daft Punk)
Oh well...
It’s works for some music, and I’ll pick music over Sonics.
What really burns me up is the Johnny Cash American albums: why the compression?
- Joined
- Sep 4, 2019
- Messages
- 1,172
- Likes
- 2,321
My two cents based on my observations over time. Let me preface this by saying I am pretty fastidious about my listening levels as I have tennitus and too loud, too long can really aggravate it. As a result, I have carefully calibrated for 74db per speaker at -20dbfs on my desktop and 78 db per speaker on my main system and use volume leveling to adjust music and albums to an average of -20dbfs. These levels are based off of Bob Katz's recommended levels for monitoring while mixing (which can and should be adjusted for room size). This is my normal maximum level listening, but most of my listening is on my desktop while working during the day at about 60-65 db average.
What I have found over time is that compressed music tends to sound pretty loud at the reference level and I often have to turn it down. Conversely, with highly dynamic music, I want to turn it up 3-6 db above my reference. I understand physically, this is because our ears sense loudness based on average levels, not on peaks. The side effect of this (for me at least) is that I play compressed music at lower volumes where the overall peaks may be only 85 db per speaker, not even a watt with my 88 db efficient speakers (a bit more with my very inefficient Magnepans).
Conversely, with highly dynamic music, I find that my average level may be 83 db, but now I am getting peaks of over 100 db. I was curious and measured with REW on a dynamic piece that I was playing above even my normal loud level and shocked to be seeing short peaks of over 110 db. Now we are talking some real power, 400 watts needed to not clip the amp (fortunately my amp puts out 300w/ch into 8 ohms and 600w/ch into 4 ohms), so when turned up, I was right at the clipping point for a pretty big amp with moderately sensitive speakers.
So my theory on power is have lots of it, as most people likely clip their amps when they turn it up without realizing it. It is only when you hear dynamic, unclipped music at higher levels that you realize that you were clipping the amp, you just didn't know it. When in doubt (unless you using high efficiency speakers like horns) err on the side of too much. The reality is that with the very affordable prices on Class D amps, you don't have to sacrifice on power. For $700 you can purchase a Nord with an NC502 with each channel putting out 350 watts into 8 ohms and 500 into 4 ohms. For $900 you can purchase a Nord with Icepower AS1200 modules putting out 600 watts into each channel at 8 ohms and 1200 at 4 ohms.
What I have found over time is that compressed music tends to sound pretty loud at the reference level and I often have to turn it down. Conversely, with highly dynamic music, I want to turn it up 3-6 db above my reference. I understand physically, this is because our ears sense loudness based on average levels, not on peaks. The side effect of this (for me at least) is that I play compressed music at lower volumes where the overall peaks may be only 85 db per speaker, not even a watt with my 88 db efficient speakers (a bit more with my very inefficient Magnepans).
Conversely, with highly dynamic music, I find that my average level may be 83 db, but now I am getting peaks of over 100 db. I was curious and measured with REW on a dynamic piece that I was playing above even my normal loud level and shocked to be seeing short peaks of over 110 db. Now we are talking some real power, 400 watts needed to not clip the amp (fortunately my amp puts out 300w/ch into 8 ohms and 600w/ch into 4 ohms), so when turned up, I was right at the clipping point for a pretty big amp with moderately sensitive speakers.
So my theory on power is have lots of it, as most people likely clip their amps when they turn it up without realizing it. It is only when you hear dynamic, unclipped music at higher levels that you realize that you were clipping the amp, you just didn't know it. When in doubt (unless you using high efficiency speakers like horns) err on the side of too much. The reality is that with the very affordable prices on Class D amps, you don't have to sacrifice on power. For $700 you can purchase a Nord with an NC502 with each channel putting out 350 watts into 8 ohms and 500 into 4 ohms. For $900 you can purchase a Nord with Icepower AS1200 modules putting out 600 watts into each channel at 8 ohms and 1200 at 4 ohms.
Hey, that's my theory, but I'm happy to share it.
Have you listenend to the album? It sounds to me like the compression in this track is deliberate, if not tongue in cheek.
Similar threads
