Discrete Opamp Vs Monolythic!

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Which file is which?

  • File "A" is Discrete and "B" is Monolythic

    Votes: 0 0.0%
  • File "A" is Monolithic and "B" is Discrete

    Votes: 7 100.0%

  • Total voters
    7
  • Poll closed .

abechap024

Well-known member
Joined
Aug 8, 2009
Messages
2,303
Location
Provo, UT
Hello,
Fun little poll here....
Which do you think is which??

Preamp is based on a API style topology. Reichnbacher(sp?) transformers (early cinemag)

Monolithic chip is LM318

Discrete chip is yours trulys Dip 8 API style opamp.

Files:

www.acsoundstudio.com/DIY/discrete_opamp/audio/test_file_A_first_mp3.mp3
www.acsoundstudio.com/DIY/discrete_opamp/audio/test_file_B_first_mp3.mp3


 
Hey,

I'm not into A vs B threads but I will say that if it's to have any worth at all that Mp3s won't cut it.  Even 320K MP3 destroys the kinds of details we may hear in an A/B.
 
its just for fun.
I checked these vs the original WAV files and I think there is definitely enough sound left to tell the difference. its 320k Mp3

Plus if you can't tell the difference in MP3 format than who cares anyway ;) ;) 


BUT that being said. I shouldn't keep these 96khz 32bit audio files all to myself. They are here if you like:

www.acsoundstudio.com/DIY/discrete_opamp/audio/test_file_A_short.wav
www.acsoundstudio.com/DIY/discrete_opamp/audio/test_file_B_short.wav

 
It would help if the sound bytes were identical, the second one sound like the speaker is closer to the mic, plosives are louder and the proximity effect is there...BUT I think B is the discrete.
 
Hello yes its not optimal

But the general idea is just to get a feel for the different sounds. I tried to make sure the distance from my mic was the same and tried to keep it the same volume etc.
 
I have mentioned this once before - this community is full of some of the most highly intelligent, extremely logical, very scientific, right brained blokes. These skills are admirable and mandatory for perfecting circuit design; - but what amazes me is that so many people cannot have a bit of fun with something like this without picking apart the lack of science behind the idea and taking things so seriously. Surely this small thread in the brewery is not posted for definitive electronic theoretical proof of one circuit format being better than another - its for someone to post recordings of circuits they have put their time and love into and have fun with other people. Its about sharing and enjoying interacting. In its crudest sense, the shootout is for people to have a little fun!!! Am I the only one who feels this way?

Surely it's a given, and I mean a given, that without true reference material, high res .wav files, etc. etc, there can not be any real scientific comparison. It must also be a given that anyone who forms opinions on op amps vs monolythic based on a sample on a website is a confirmed idiot! But with that being said, why are people unable to simply enjoy a little fun listen without taking the whole process so seriously?

I mean is anyone going to be quoted for the rest of their lives and have their reputation damaged because they contributed to having a bit of fun commenting on what they liked about a web sample?

Im almost worried about partaking for fear of being identified as insipid:)
 
Haha! Yes!!. This is a very scientific based forum. But I think thats what I like so much about it.

That being said I'll just tell which is which so people can form their own conclusions:

A is monolithic

B is discrete

Now for my personal feelings, and impressions:

The high end is were I can tell the most difference, which is were honestly I expected the least difference. Still want to do some actual sessions with them, to see how well they actually sound in the mix, but so far I like them. I have some sessions coming up, so I only did 2 preamps. I really like the original lm318 in the mix, so Its not like I'm eager to get rid of them, but the discrete seem to offer a little more "life" or mojo. Anyway. enough feelings!! ;)
 
I like listening tests! Everything audio is subjective anyway. There is no scientific way to define what sounds good to one person and bad to another. Deuce42 made a good point. It's just fun!

Haven't listened to the files, but I will check them out later today when I can monitor them properly.
 
wow, I got them correct, listening to them in mp3 and on my macbook laptop built in speakers... for fun of cause...

I just went for the one that sounded better to me and assumed it was the DOA. I like how these sound.... Gonna have to get some...

bless
 
Cheers!

It is interesting on my laptop speakers the difference is even more noticeable...weird how things work like that sometimes!
 
ruairioflaherty said:
I'm not into A vs B threads but I will say that if it's to have any worth at all that Mp3s won't cut it.  Even 320K MP3 destroys the kinds of details we may hear in an A/B.

If the difference isn't big enough to even be heard on a 320kbps MP3, I'm not interested. Now lets give these files a listen...


 
Well that was interesting! My first reaction was that there was something about clip A that I didn't like... but the differences weren't vast. I kind of think its like clip A had a bit of fizziness in the high mids and highs....?
 
I just heard a slight addition of second harmonic in the loud parts of clip b.... Other than that they were very similar to me.
 
Huge difference, and I vastly prefer the discrete one. But which one is more accurate? How much does it matter in an area of the console that only passes audio (no heavy lifting)? If it is coloration, how does it accumulate? And finally, have you tested the discrete op amp vs. a modern quality IC like the OPA1611 or ADA4627-1?
 
Hi,


  I am very surprised at the considerable difference. I have never tried LM318, or even heard of it. I would find a comparison between 5534 more revealing. I am surprised that an ic with such a high slewrate sounds as mucky in the sibilance. I am a huge 2520 fan, but I have some pres with 47910 instead of 2520 specially FOR the cleaner top end. Maybe there are other factors here. It is not an equation with constants . . . . . besides, there is far more proximity effect on sample A, detectable even on my Lappy.


    Sounds bloomin astonishing at first listening - nice one Abe


  AndyP
 
strangeandbouncy said:
Hi,


  I am very surprised at the considerable difference. I have never tried LM318, or even heard of it. I would find a comparison between 5534 more revealing. I am surprised that an ic with such a high slewrate sounds as mucky in the sibilance. I am a huge 2520 fan, but I have some pres with 47910 instead of 2520 specially FOR the cleaner top end. Maybe there are other factors here. It is not an equation with constants . . . . . besides, there is far more proximity effect on sample A, detectable even on my Lappy.


    Sounds bloomin astonishing at first listening - nice one Abe


  AndyP

The LM 318 was an early high slew rate opamp that predated the Bi-FET technology (which gets it's speed from lower FET input stage transconductance). and predated the 5534 (with it's variant on dominant pole compensation to gain speed). The LM318 got it's speed by degenerating the input stage long tail pair with IIRC > 1K resistors in series with each emitter. There is noise and other negative consequences from this  extra R effectively in series with the input devices.

I used the LM318 in the front end of a home brew power amp I designed back in the early '70s and it sounded good enough to me at the time (note amp still works and still sound Ok, I think). I don't listen to it very critically, and it doesn't measure up to modern amp performance specs, for reasons other than the LM318 (like slower power devices available 40 years ago). 

Not to be a broken record but when opamps are used with negative feedback, audible differences between otherwise identical NF networks using different opamp cores are caused by either inadequate loop gain margin that results in deviation from that proscribed response, or in gross cases from non-linear behavior (slew limiting, oscillation, etc).

A null test can help identify the reason for the deviation from proscribed transfer function, of course you want to control all variables, so use the same source.   

JR
 

Latest posts

Back
Top