I post the noise reduced version. Wow the problem with the AI might partially be not accurate enough transfers. This sound waaaay better.
@knar
Deleted after EQ version below
1 Like
I EQ above.
FUCK~!!!
1 Like
I realize that since I tear our tone arm wires from internal board I actually wiring them wrong. So pretty much all the transfers since then were hallowed out of bass due to impedance mis matches and maybe some cancellation down there after I jerry rig a balanced circuit without changing polarity on one of the terminals.
This all confirmed with an adapter I bought today. This explain a lot.
Hmmmmm
Sounds like a mid 1950’s recording now.
Its not bad!
Big progress on noise reduction and other things even though I mildly fucked up the OG transfer path.
1 Like
It’d be cool to here some noiseless Cortot playing Chopin and/or Schumann.
1 Like
I was thinking of picking up some cortot. I’m 90s percent sure I can something listenable with little or no noise now.
That would be sick. Do the 1930ish Chopin Preludes…pweez!
2 Likes
I got a czerny variations from horowitz for a few bucks. Its thrashed and super warped but plays. Going to see if this stuff works with that legendarily shit sounding rec.
Also just bought a cortot preludes set I was looking at. 
3 Likes
Amazed this turned out listenable considering how fucked the record was.
This the thing making a real balanced circuit now.
This in case it sound nicer than commercial which I dun check yet
https://drive.google.com/file/d/10KD2jjlQeTI90_XWq_z0GdUbEagFc2qx/view?usp=drivesdk
Wow, checked. This do sound better. Shit never updated commercial release from Sony below.
1 Like
I just realize I can finally do my play it backward and forward mashup now that I figured out time alignment stuffs using boom / lav mic correction tools.
Both cancel out fucked elements of playback or just smear both ways? Who knows?
Will do in 5 hours when done with hour 12 of nodding in meetings I’m not listening to.
I transfer forward and backward, then flip and time align. Then use AI to clean this.
It maybe capture more detail but also some more noise. I think the best way for this to make sense is different diameter stylus, so I catch different dept of groove where noise slightly different so less likely to amplify and more likely to cancel. That also sound like pain.
1 Like
Did you especially enjoy the lab portion of science classes in school?
No way! I was busy ditching and having fun so avoided classes that were heavy on shitwork vs testing - which I could always cram for and eek by. Nothing has really changed, tho I find myself doing lots of shitwork now and having zero fun.
3 Likes
I see. It just seemed that way to me based on this hobby of yours. Do you do this stuff in your kitchen?
Yeah its getting cluttered. Considering throwing out the vauum chamber etc after the molding and casting thing was meh.
I just took 4 quick transfers of this valse oubliee. Try compositing AFTER I clean them up. Poss only capture goodness instead of noise that way.
1 Like
My freq response record arrive. Sorta shocked I got such tight tolerances with just fucking around. I also ran this through the AI noise cleaner.
Before I correct.
After I correct
Curious if it make any difference.. I’m already way below tolerances for half of the completely fucked recording chain they have back then.
Also too much noise and variability on the test disc to venture into impulse response / deconvolution. At least for now.
This a mildly thoughtful attempt using calibrated shits. Will need to figure out artifacting or poss not get trash records and melt them. Hmmmm
Whoops, this have the right curve and a properly balanced tone arm. Same issues tho
Valse boobliee 3 take composite using no spectral shifting to freak out the AI noise reduction
Def a little less fucked. On right track down the wrong road poss.