How do you decide if a mod was "right"

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Maybe a little awkward thread title here, but i am going in rounds.
First TDA1541 Non-os, then back to original oversampling, tried different opamps in cd player: from NE5532 to OP275, LT1469, AD823, OPA2132, but replaced NE5532, and i am happy again??!! Nice warm and deep bass notes, i like that.

For output cap incdp i tried 10uF MKT, now i have a 7.2uF MKP starter cap. (cheap)
Tried a BG Nx there: sounds crap! Tried those BG Nx near opamps: crap!!. Standard Philips lytics do a good job near the opamps.

Replaced standard MKP audyn coupling cap in tube pre-amp for a Audyn cap plus with double series winding: not good, replaced the standard cap again.

How do you guys here decide if the mod was done right, besides listening tests. Wat gear are you using? Spectrum analysers, distorsion meters or whatever. Just only curious.
 
tubee said:
Hi jackinnj

I didn't look at it that way, interesting vieuw. My taste does the final decision indeed.
What gear do you use besides you ears?


Ears are the ultimate arbiter -- there was an interview in the most recent issue of EDN discussing measurement of audio chips -- turns out that you can leave the data sheet at home -- many of the folks making amplifiers want you to show up with a completed design and they will go from there -- the fellow from Audio Precision said that in almost all cases where a difference could be detected by "golden ears" they could pick up some difference with their instrumentation -- i will rely on my ears but have to admit that the range gets more restricted as the decades advance.
 
Every audible difference must have some manifestation, somewhere, in the physical world. Everything audible is theoretically measureable. Whether the measuring gear is up to task and whether the measurer is looking in the right place remains, in any instance of measuring, an open question. The former might be a hard limit to measurement efficacy.
 
Good question, tubee. I'd say as long as mods addiction lasts. I would not depend on ears-brain factor only - if one is after the highest fidelity rather than best sound.

The problem with perception based best sound concept is that it varies over time not only due to ageing but due to heaps of factors affecting chemistry of the brain. And each brain is a touch different. So not only drugs alcohol, nicotine and coffee affect our perception but all other moods generated by the environment we live in. And in cities it is also noisy. Golden ears belonging to some young person living in the nature and depending on his hearing for survival could be of great assistance.

On the objective side one can use a number of tools (high frequency/resolution scopes, distortion and other meters etc) but here new technologies bring progress as well so no limit in site.

In my case it is when I get enough of doing something. It is brain dependent.

cheers,
 
Mod is a matter of your need in a sonic improvement.

Having said that, I think a common method is to compare the sonics against a benchmark. I always visit hi-fi stores or my friends' places to hear their gears. If you find a "right" player or component within your budget, then just buy one !

If you choose not to buy a new one, then you have to mod the one you have to become the one you like. Sometimes the benchmark is composite, the combination of the bests.

Any measurements and theories will help/guide you which mod and how much I should mod it. The final tool is still your mind, not ears. Enough or not enough mod. :D
 
AX tech editor
Joined 2002
Paid Member
jackinnj said:



Ears are the ultimate arbiter -- there was an interview in the most recent issue of EDN discussing measurement of audio chips -- turns out that you can leave the data sheet at home -- many of the folks making amplifiers want you to show up with a completed design and they will go from there -- the fellow from Audio Precision said that in almost all cases where a difference could be detected by "golden ears" they could pick up some difference with their instrumentation -- i will rely on my ears but have to admit that the range gets more restricted as the decades advance.


Hi Jack,

You have a link to the EDN piece?

Jan Didden
 
Thanks for all replies.

jackinnj said:
i will rely on my ears but have to admit that the range gets more restricted as the decades advance.

Experienced that myself also. But my ears are getting more critical too with the years. :xeye:
When i was young i listened more easy to a slightly distorting tweeter driven from 2800Hz @6dB. My ears filtered the distorsion away or some thing like that. Nowadays i even refuse to listen to a diy speakers-tweeter if it has no complete resonance dampening circuit in it's Xover, even if the tweeter had ferrofluid dampening itself!

janusz wrote:
"The problem with perception based best sound concept is that it varies over time not only due to ageing but due to heaps of factors affecting chemistry of the brain."

1 fine glass of red wine does a lot of good to the music perception. :D


An extra explanation from my side why i came up with this thread afterall:

Music is my interest, my brother has build his speakers himselfs long time ago, and wanted to do that also was my decision as a kid. Well, made quite some speakers indeed. Also did a lot of mods on sources, CDP's, analog tuners, cassete players, TT's, build a TT myself, made an arm for it, tried non-os with PCM56 and TDA1541. Have build tube amps, transistor amps, and a tube pre amp. Learned to recognise the sonic difference between bipolar transistor, fet and tube.

*What i found out in time now is that linear distortion is dramatical important. When i swap an opamp, treble is maybe 1dB raised, because the mids are displayed some laid-back (opa2132!) Not a big problem, but i have a hole in my hearing treshold from 2.5 to 4K of -35dB (popconcerts,loud music?) and when treble is perceptably raised 1 dB, the whole system sounds out of balance. Sound gets annoying. My purist tube pre has no tone control, should add a baxandall. I only can lift tweeter level slightly on speaker with switch.

*What i found out also is that it is better to buy a reasonbly good source from the beginning to start with. Not a cheapo plastic unit and mod it until it has X added diy PCB's. The designs have been tuned for good balance in sound (have a CD304mk2. and a Kenwood KT1100 tuner) Swap an opamp/coupling cap and there it begins, a (never ending) cascading of mods after eachother.


I am interested in EDN discussing measurement of audio chips too.
 
rdf said:
Actually, a quite incredible read.

It is, and it made me wonder just where audio is going. "High Fidelity" in the old sense of the "piece of wire with gain" seems to be being replaced by a synthetic processing of the sound signal to re-create, rather than preserve, the original qualities, and fool the listener's ears into accepting that it's the real McCoy.

But listeners said that the sound stage was inaccurate. Further exploration showed that the cause of this problem was the digital-filter algorithms for FIR (finite-impulse-response) filtering. The most common algorithm does exactly the right thing in the frequency domain, in which engineers check the filter response. However, observing the impulse response in the time domain shows that the response centers on—rather than trails—the impulse. In other words, preringing occurs. This effect interferes with the human ear’s method for keeping track of the spatial relationships between the sounds from the speakers. A new FIR algorithm was in order.

Does an algorithm have a soul?
 
cpemma said:
Does an algorithm have a soul?


Actually, there is a wide variety of filter algorithms out there. The German magazine Stereoplay, for instance, has characterized different filter settings for emmLabs, Marantz and T+A DACs.

According to them, a simple filter that chops off part of the audible spectrum with no passband-ripple and low phase distortion, gives a "warm and relaxed" sound, whereas a sharp filter will give "accurate" sound. The trend nowadays seems to be exactly what the Wolfson guy mentions: pre-ringing-free filters. The guys at emmLabs have a few nonlinear filter patents to create exactly that kind of filter, IIRC.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.