When optimizing, go for Lowest 3rd harmonics .. not THD!!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Lineup,

my software development is temporarily on hold as i am a new father.

when i do get going again in need to teach myself some of Farina's methods so i can incorporate them into my work.

i am mostly doing this so i can learn the first principles of amp and loudspeaker measurement. i will likely be strongly influenced by software like ARTA.

i use a piece of software called igor pro from wavemetrics. it is similar to matlab or labview but more suitable to things i do most of the time.

as far as hardware goes, i have yet to invest in a nice sound card but i am building a nice mic with a panasonic capsule as well as a preamp.

i am also waiting the arrival of a book on loudspeaker measurement and keeping my eye on ebay distortion analyzers which i would use as a reference.

thanks for asking,
Tim
 
Hi,
antiquated tube technology: harmonic distortion components of a directly heated, low-mu triode amp, not too faraway from the ear`s preference. Unfortunately, I don`t think it`s attainable with solid state, no kind of magic trick help, just something to dream of.
 

Attachments

  • harmonic distortion.png
    harmonic distortion.png
    40.9 KB · Views: 286
Hello,

Actually, there has been many theories about the earing process! Each one ruling out some old ones over time again and again since 19th century. The only thing that is sure is that the inner ear is a tapered closed pipe so it's prone to a lot of harmonic generation, decreasing in power as order increases. The brain makes up for it and masks the lower harmonics. That could be why if the harmonics are too high (the awfull 7th...) the brain cannot correct it because it is not used to it and so it is less ear friendly.

Regards
 
Lumba Ogir said:
Peter,
I`m talking about how the human ear perceives sound. Hearing is not linear in any respect.

Ok, I think I undeestand what you're getting at.

Yes, listening at pure sinewaves there is a masking effect that means a spectrum like the one you showed is tolerated much better than the same power of distortion spread out at equal level at the harmonics. Playing music will result in lots of intermodulation products though which are non-harmonic and those are not masked the way low order harmonic products are masked.

My preference is as low distortion as possible until it's not audible anymore. The distortion performance that you posted will not likely be transparent but will result in audible coloration.

That the hearing is nonlinear in many aspects does not change the things mentioned above.


/Peter
 
darian said:
Hello,

Actually, there has been many theories about the earing process! Each one ruling out some old ones over time again and again since 19th century. The only thing that is sure is that the inner ear is a tapered closed pipe so it's prone to a lot of harmonic generation, decreasing in power as order increases. The brain makes up for it and masks the lower harmonics. That could be why if the harmonics are too high (the awfull 7th...) the brain cannot correct it because it is not used to it and so it is less ear friendly.

Regards

Yes but the pipe (cochlea) is full of water and water is not very compressible. Wihtout actually knowing where the nonlinearities (as in harmonic distortion) are produced in the hearing apparatus I'd guess the Membrana tympany and the "inner tympany" at the cochlea.

Then there are other nonlinearities as well for ex. the relation of pitch to frequency which depends on SPL.


/Peter
 
1. when i asked for a reference i was trying to gather the level of hearing was being thought about when talking about the ear. when the ear was mentioned as being non linear i wanted to know if they were really talking about our perception of sound and not, for example, the effect sound on the basilar membrane. in addition, i was also interested in the experiments they were considering when making this statement because this might be helpful in determining what type of distortions we might be most sensitive to. i was not trying to be disagreeable. since i don't know what experiments were being referenced i cannot "look them up".

2. Lumba, there was good reason to think that hearing might correspond to "made up" linear scales. the early work of Georg von Bekesy for which he won the nobel prize in 1961 argued for linearity at the level of the basilar membrane. Bekesy showed that each spot on the basilar responds to sound by vibrating linearly. given the structure of the basilar membrane this is perhaps expected (maybe too strong a word). unfortunately, he did this work on human cadavers. on live animals it was later show by Ruggero and others that hair cells actually tune the response of the basilar membrane to make it non linear. there are some cool experiments using ferosimide which selectively knocks out hair cells showing this.
 
This thread started with a statement of preference of a particular characteristic in an amplifier. It has gone round 360 degrees twice maybe more and it would appear that there is no rational conclusion.

I think from what I read here, that individuals will perceive the same amplifier differently which may well be related to that individuals hearing, which when processed by the brain - perception.

Therefore the the best amplifier in the world is the one that you like listening to most. Now I might like to add one more characteristic, that if the amplifier is also aesthetically pleasing, it even sounds better.

regards

Nico
 
Administrator
Joined 2007
Paid Member
Which is what I am doing now :D Listening to the best amp in the world, with Alfred Brendel playing Mozart just for me.
You guys are all still searching are you not ?

Oh yes Lumba ;) ( Bit slow on the uptake ) What did you mean a couple of months back " By chance you got it right " Chance ? Chance doesn't even come into it :D
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.