Not apples to apples. Gets pretty bright without a descending target slope. Targeting the harmon slope (not curve) with the new adjustements for my ear that smooth everything out. Fascinatingly boring stuff
Unrelatedly, I love looking at the Spectogram graphic that CanOpener implemented.
Got it to work with Roon. Good times!
Michel Béroff’s recordings are giving me an eargasm right now!
1 Like
My favorite preset on CO is Lifelike with the amount brought down to 100-125% for modern recordings. Also, make sure HQ! is on in the advanced settings (makes a noticeable difference).
1 Like
This ear mic thing fascinating and gives you clues to what might need to get adjusted by not much different than EQing by ear at end of day with all the tweaking. Possibly need to think of a new method that is a little more accurate.
1 Like
I c
1 Like
duz ur plugin haff gammatone spectrogram ?
Gammatone filters are a popular linear approximation to the filtering performed by the ear.
allegedly
2 Likes
Ooh will tinker with this after I’m back in
1 Like
Fark… this probably the only way to do this right. It looks kind of complicated.
After an incredible amount of fucking around I finally figured out how to take consistent in ear mic measurements in REW. Will use this guys correction method. His video reminds me why I didn’t do well in school but I will tough it out.
1 Like
FUUUUUUUCK
This is the truway. Some of the better recordings really sound like you are at a concert with the headphone/ear correction and can opener doing the crossfeed and directional correction. JESUS FUCK… Will probably need to concentrate on taking better measurements next go but is worth the trouble.
I can’t describe what I’m hearing it is that much better. I have no reference for how fucking amazing this is.
I’m listening the later horowitz rach 2nd sonata. Its like a kaleidoscope is skull fucking me. Wow… Those fucking nerds measuring nats ass against blind averages really missing the point until that accuracy tuned to your ear.
If anybody insane enough to do this I tweaked a few things.
I used the following instead of a harman target in the target settings:
"PS: New research shows that you might enjoy a slightly different implementation of the Harman Target Curve better: Delete any “House Curve” uploaded in REW, Preferences tab. in EQ Target Settings, “Add room curve” with the following values:
LF rise start: 200 Hz
LF Rise End: 10 Hz
LF Rise Slope: 0.7 dB/octave
HF Fall Start: 200 Hz
HF Fall Slope: 0.7 dB/ octave
Target Level: Calculated from response"
and on impulse wav export I used minimum phase and normalized to peak.
Convolving in J river.
Dudes drive file with the particulars.
So weird with can opener engaged and the perfect response corrected for me ears, “reverb” is more of a finite distance / spatial sensation on good recs instead of haze. The insane performance of these overkill headphones finally making some sense.
1 Like
Looped the YouTube audio through Jriver so I could convolve it. Checked out transfers holy god! They sound terrible! Except for recent rach one maybe
Gonna redo a recent one but checking things through headphones that now aren’t a comb filter festival.
1 Like
Amacha cuz didn’t want it spend forever cleaning it up, but EQd in corrected headphone
Probably a dumb idea but I have a test record with pink noise and a digital reference pink noise from interwebs. I tried to EQ from it once but didn’t have a lot of luck. Was thinking of making an impulse response file to correct instead of an EQ curve… I think it’d be pretty easy in voxengo’s deconvolver or REW. Lord knows couldn’t make this laser TT sound worse! There some weird shit going on in that machine that maybe this the only way to accurately unfuck. The theoretical curves I’m sure are great but sound like an ass potato.
Wtf?
All this theoretical curve shit and mastering dudes at time guessing the amount of energy in treble on a whim to account for what they think will be hard for cartridges track at the time and/or were biased toward? Those fucks!
This makes so much sense now… The galo write about this, but mainly to say that magnetic carts have a 6db slope factored into RIAA. Sounds like there are differences worked into early mastering compensation outside of the scope of the science or emphasis standards. This maybe why mono recs sound so dull even if perfectly transferred and restored. Another 6DB slope might be on the floor.
Da galo article
Fuckin A
This a test disc with white noise and NO curves, so this just the bias of the laser TT and of course wear and cut of test disc but looks pretty minty based on high freq response. Will generate corrective impulse filters and run those before running the insanely accurate wurcer biquads for RIAA and all that shit I spent months figuring out.
After fixing the TT shit above used the Galo reference from RCA guys 1935 letter instead of commonly recommended curves and rolls.
I didn’t cut above 7500 cuz I like the psychoacoustics of surface noise up there and think they were worked into the decision making.
Was worth it 
