• 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!

How to find the difference in two recordings taken from same source?

OP
U

UltraFine

Member
Joined
Jan 18, 2023
Messages
16
Likes
1
@Digby , allow me to explain more clearly.

I buy vinyl on regular basis and I rip it. My system had not a Scarlett2i2 until recently. With my system I always had a feeling that the sound is sub-par. Like a bit dull. It is not only with these few examples that I uploaded.

I also understand that this accessment sounds like woodoo. So I started comparing. I started downloading rips of the same vinyl of other people. If I could not find a vinyl rip I went for copies from digital origin. I had to start somewhere, right? All the way I found my rips dull comapred to other recordings I could get. It is not easy to find other vinyl rips of the records that I buy, so it is hard to find good material to compare.

Recently I bought Scarlett2i2 because I thought that the problem of dullness is with my SoundCard and because I got a discount on it. But it seems it is not. The dullness is still there. I would argue that the sound improved a bit comparing to my SoundCard, but not significantly, but this is not the point here.

So then I started showing my rips to other people (and arguing with my brother) asking them if they also hear the "lifeliness". My gf for instance said: the (dance) music does not sound moving. And I agree with her. My brother was of the opinion that the difference is in the margin of error. So I came here to ask for a way to CLEARLY show that there is a difference and that there is a quality issue. I also want a programatic analysis because it would also help convince by brother to invest in new hardware like new phono pre-Amp. We are sharing the costs. Hardware is expensive.

If you have further questions, feel free to ask.

I'm confused about the purpose of your post.

You are comparing your recordings (of vinyl) made to digital recordings (yes, they will sound different) and vinyl transcriptions done on other equipment (yes, they will sound different too). Does the vinyl sound dull before it is recorded to digital, if so the problem is cart/pre-amp/interface, if it only sounds dull after conversion it is probably the fault of the interface.

FWIW I briefly tried a Scarlett interface but found the sound quality wasn't acceptable. I upgraded to RME and have had no qualms since.

I do sometimes wonder if my digital recordings lose a bit of 'air' compared to direct playback. However, the difference is so small, I doubt I would notice in a blind setting, which suggests it may just be my imagination.
 

Digby

Major Contributor
Joined
Mar 12, 2021
Messages
1,632
Likes
1,566
Your turntable is fine. Which cartridge are you using and how old is the stylus? Your phono preamp seems quite old (1970s?), so even if it is working, there are probably much better ones available. I put my money on it being down to those two things. The Scarlett interface is probably fine.

Many people use compression and noise reduction (perhaps eq?) on their transcriptions, so this will change the sound too.
 
OP
U

UltraFine

Member
Joined
Jan 18, 2023
Messages
16
Likes
1
I use the CONCORD system by Ortofon. The stylus is not that old. It is not a high end setup, but still I would expect it to have decent frequency responce with normal top end.

@Digby , @Blumlein 88 , @dasdoing , @HarmonicTHD - I was thinking .. maybe the best way to go is to buy an "Analogue Test LP" with a sweep from 20hz to 20Khz. I could record that sweep and maybe see if it shows anomalies. But if the anomaly is proportional, I assume, I would not be able so see from that recorded sweep at what frequency ranges the defect actually is.


Your turntable is fine. Which cartridge are you using and how old is the stylus? Your phono preamp seems quite old (1970s?), so even if it is working, there are probably much better ones available. I put my money on it being down to those two things. The Scarlett interface is probably fine.

Many people use compression and noise reduction (perhaps eq?) on their transcriptions, so this will change the sound too.
 

HarmonicTHD

Major Contributor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,846
I use the CONCORD system by Ortofon. The stylus is not that old. It is not a high end setup, but still I would expect it to have decent frequency responce with normal top end.

@Digby , @Blumlein 88 , @dasdoing , @HarmonicTHD - I was thinking .. maybe the best way to go is to buy an "Analogue Test LP" with a sweep from 20hz to 20Khz. I could record that sweep and maybe see if it shows anomalies. But if the anomaly is proportional, I assume, I would not be able so see from that recorded sweep at what frequency ranges the defect actually is.
The most likely causes which I see are:
1) clipped recording. Easily avoided by monitoring the green light in the 2i2 and setting gain correctly.
2) unmatched impedances between your preamp and 2i2 (see above). This can be fixed by using the 2i2 line input (which you should use anyhow for this kind of application) and buying two XLR to TRS cables at ca 10EUR each (see Amazon).
3) old phono preamp not working properly. This is the most costly fix and therefore I would try it last after having eliminated possibility 1 and 2.

(By far the most unlikely possibility is the 2i2).
 

SSS

Senior Member
Joined
Jan 12, 2023
Messages
328
Likes
221
Location
Germany
What UltraFine wants to achieve will have no real success. Even between different vinyl LPs there may be differences because of varying press quality and how long the press master was used. And different turntable with the audio chain after the sound will differ anyway. Comparing to a digital version is misleading depending on whether it was taken from the master tape or it was digitalized from the vinyl LP. Both happened in the past and maybe still today. So, if the sound of the self digitized vinyl is not good enough it needs to find out the reason and modify the setup.
 

LtMandella

Member
Joined
Jan 13, 2021
Messages
67
Likes
44
Location
Las Vegas
Hello all,

I am looking for a way to compare two different recordings (samples) of the same vinyl record in an audio software and find out what the difference is. I can hear the difference, but I want to know in which frequency range it is exactly.

Details:
My brother and I, we are trying to digitize some of our vinyl records. To do this, we are using a Technics turntable, which delivers the sound to a USB interface Scarlett 2i2, via an EELA Audio 804 pre-amp. The Scarlett delivers the signal into a Windows PC.

The problem:
In my opinion, all recordings made with the system described above have a distinct "coloration" . They sound somehow slightly dull, lifeless, as if some frequency range is reduced. I hear it clearly in the headphones, but my brother claims that this is normal. Now I got vinyl rips from other people as well as digital versions of some tracks and compared them by listening test. I hear the coloration clearly. My brother does not want to see it however. I want to show him therefore it black on white.
In order to be able to grasp the difference better, I merged small samples in a WAV file, in order to be able to make a fast A/B hearing test. I attach three MP3 files as an example: A-B-Test.Vinyl.Ripp.zip. The samples are more or less the same content, just from different recordings. B version is ours.

Addendum: I have uploaded a collection with full tracks in FLAC and ready to go samples here.

Question:
I am looking for a way to use software to "overlay" the samples and see in which frequency range the difference is. The samples are from the same original. Which software would accomplish that?


Cheers
I think Audacity can invert a track? And then blend. That should work right? But matching levels could be tricky. Also, I had what I felt was awesome results transcribing my vinyl to digital. I did ultransoncially clean every LP, that reduces the noise floor incredibly well, even for brand new unplayed discs. And I used a new MC cartridge with IFI phono preamp (very accurate RIAA eq). Also, I make the transcriptions with only headphone monitoring, so no loudspeaker microphonics to deal with. You may need to experiment (cartridge, preamp, recording levels, encoding, etc) to find the most satisfying results, I sure did.
 

DVDdoug

Major Contributor
Joined
May 27, 2021
Messages
3,092
Likes
4,082
They sound somehow slightly dull, lifeless, as if some frequency range is reduced.
Most old viny sounds dull to me. I haven't heard any "new" records but back in the vinyl days "good sounding" records were rare. I don't play records anymore but when I (occasionally) digitize a record I'll sometimes boost the highs. (I don't publish/share my music on the Internet.)

I just hear the difference between my rip and other peoples rip of the same source vinyl.
You don't know what the other people are doing or what cartridge they are using. They might have a different or newer or older copy than you. Sometimes copies released in different countries were different. You don't know if the other person is using EQ or other processing.

I think Audacity can invert a track? And then blend. That should work right?
No, that won't work. Two related reasons - The waveforms/data has to be aligned perfectly and there are 44,100 samples per second (or more). If you align it at the beginning the timing/speed will drift to some extent. But you can't "perfectly" align it (even at the beginning) because you can't control exactly where the waveform is sampled. You are only sampling/reading the waveform at one instant and the digitized data will never be the same twice (even though the restored analog waveform can be very-close to identical).
 
Top Bottom