kus has quit [Read error: Connection reset by peer]
<znf>
I have 2 GPUs in a machine, a Tesla P4 and a GTX 1070
<znf>
According to all the stuff I read, as far as I understand, the encoder/decoder performance should be the same
<znf>
yet the P4 is really struggling with the same video transcodes (count and quality), having it's decoder at 100% all the time
<znf>
turns out, this is due to the lower default frequency, apparently, as if you boost the clocks, it increases the encoding/decoding performance
<znf>
The P4 docs say: Tesla P4 can transcode and infer up to 39 HD video streams in real-time, powered by a dedicated hardware-accelerated decode engine that works in parallel with the GPU doing inference.
<znf>
wtf is "inference" in this case?
<CounterPillow>
machine learning stuff
<CounterPillow>
i.e. running model weights to give you some output, as opposed to training it
<znf>
Ah!
<znf>
So it's saying that you can do encoding/decoding _and_ doing ML at the same time
rossome has quit [Quit: WeeChat 4.2.1]
<znf>
but I'm clearly not seeing 39 Video Streams at the same time :(
<znf>
~24-25 and the decoder is at 100%
<furq>
those numbers are generally pulled out of somewhere
rvalue has joined #ffmpeg
<furq>
for example i'm looking at a different official nvidia page that says it's 35
<znf>
I could take 30 at least
<furq>
but also "hd video streams" could mean 720p24
<znf>
that would have been my next question
<CounterPillow>
or like constrained baseline
<znf>
wonder if it's 720p
<furq>
i doubt you'll get to do more than wonder
<furq>
this stuff is always intentionally vague
<znf>
meh, and I can't keep the clocks at max, as it's reaching 85°C
Gaboradon has quit [Read error: Connection reset by peer]
theobjectivedad has joined #ffmpeg
lexano has quit [Ping timeout: 252 seconds]
waleee has quit [Ping timeout: 272 seconds]
carpenter has joined #ffmpeg
treefrob has quit [Ping timeout: 264 seconds]
Suchiman has quit [Quit: Connection closed for inactivity]
anticw has quit [Remote host closed the connection]
anticw has joined #ffmpeg
treefrob has joined #ffmpeg
Kei_N has quit [Read error: Connection reset by peer]
Kei_N has joined #ffmpeg
JanC has quit [Ping timeout: 255 seconds]
JanC has joined #ffmpeg
minimal has quit [Quit: Leaving]
AbleBacon has quit [Read error: Connection reset by peer]
HarshK23 has joined #ffmpeg
Tinos has joined #ffmpeg
\\Mr_C\\ has quit [Remote host closed the connection]
Vonter has quit [Quit: WeeChat 4.2.1]
rvalue has quit [Ping timeout: 256 seconds]
Muimi has joined #ffmpeg
rvalue has joined #ffmpeg
Vonter has joined #ffmpeg
hussein1_ has quit [Ping timeout: 255 seconds]
hussein1_ has joined #ffmpeg
hussein1_ has quit [Remote host closed the connection]
hussein1_ has joined #ffmpeg
Vonter has quit [Ping timeout: 260 seconds]
junaid_ has joined #ffmpeg
Vonter has joined #ffmpeg
junaid_ has quit [Client Quit]
acovrig60 has quit [Ping timeout: 240 seconds]
rpthms has quit [Remote host closed the connection]
rpthms has joined #ffmpeg
FH_thecat has quit [Quit: Leaving]
junaid_ has joined #ffmpeg
Kei_N_ has joined #ffmpeg
Kei_N has quit [Ping timeout: 252 seconds]
FH_thecat has joined #ffmpeg
Capstan has joined #ffmpeg
Muimi has quit [Quit: Going offline, see ya! (www.adiirc.com)]
hightower2 has quit [Ping timeout: 260 seconds]
JanC has quit [Ping timeout: 256 seconds]
JanC has joined #ffmpeg
Capstan has quit [Ping timeout: 250 seconds]
Muimi has joined #ffmpeg
Muimi has quit [Quit: Going offline, see ya! (www.adiirc.com)]
Ox7C5_ has joined #ffmpeg
rv1sr has joined #ffmpeg
Capstan has joined #ffmpeg
rsx has joined #ffmpeg
Vonter has quit [Ping timeout: 272 seconds]
Vonter has joined #ffmpeg
<aaabbb>
for an estimate which kind of bitrate should i expect from ffv1 if given 3840x2160 yuv444p10le 30fps? i just want a very very rough estimate
<aaabbb>
i'm trying to compare it to prores
<furq>
you'd probably just have to test it
<furq>
from memory it compares favourably to prores for 1080p 422
<furq>
favourably meaning not quite as small but not lossy either
<furq>
but unfavourably because nothing supports it
<aaabbb>
furq: ok. i'm thinking about using it as a capture codec, and will later release as hevc, but i want a good mezzanine
<aaabbb>
and there are only so many hard drives i can bring with me
<furq>
well if it's too big you can always convert it to prores
<furq>
for whatever it's worth i just encoded a 1080p60 422 test sample and came in about 80mbps below the prores 422 hq bitrate
<aaabbb>
testing by upscaling a representative video from 720p to 4k and 444 10bit (i know it's not the most accurate to test by upscaling) shows about 500mbit
<furq>
prores xq is like 1700mbps for 4k
<furq>
so that sounds good to me
<aaabbb>
cool. just wanted an estimate of how many hard drives to bring, going to do the encoding to distribution format at home
<furq>
generally it's one more than you think you'll need
<aaabbb>
is there any compression benefit to manually selecting level 3 if i'm not using 2 pass or slicecrc or any level 3-only features like that?
Some_Person has joined #ffmpeg
<furq>
i don't think so
<aaabbb>
ok. i'll test to confirm
<furq>
never mind it does default to 3
<aaabbb>
furq: since this is just a capture codec, would nvenc lossless for hevc outperform ffv1 in compression ratio?
<furq>
no idea
acovrig60 has joined #ffmpeg
Ogobaga has quit [Quit: Konversation terminated!]
Ogobaga has joined #ffmpeg
Suchiman has joined #ffmpeg
Jhonny2x4 has quit [Quit: WeeChat 4.2.1]
Jhonny2x4 has joined #ffmpeg
Jhonny2x4 has quit [Client Quit]
Jhonny2x4 has joined #ffmpeg
lavaball has quit [Remote host closed the connection]
kus has joined #ffmpeg
hightower2 has joined #ffmpeg
hightower2 has quit [Remote host closed the connection]
hightower2 has joined #ffmpeg
Jhonny2x4 has quit [Quit: WeeChat 4.2.1]
Jhonny2x4 has joined #ffmpeg
Ogobaga has quit [Quit: Konversation terminated!]
Ogobaga has joined #ffmpeg
alexherbo2 has joined #ffmpeg
fling has quit [Ping timeout: 255 seconds]
fling_ has joined #ffmpeg
fling_ has quit [Remote host closed the connection]
<yrc>
Hello! Is ffmpeg able to author an MKV with multiple titles, where each title’s chapter comes from its own source video? In short, the idea would be: mergemkv --title-from=1_1.mkv,1_2.mkv,1_3.mkv --title-from=2_1.mkv,2_2.mkv --output=two_titles.mkv
<yrc>
(when I said source video, I meant source file, with video+audio+subtitles)
fling has joined #ffmpeg
<yrc>
Hum… I realize that there are no “titles” in MKV :-D I’ll need to wait until I’m at the step where I handle Ordered Chapters… Please ignore the above question.
fling has quit [Remote host closed the connection]
fling has joined #ffmpeg
DonTrackMe has joined #ffmpeg
<DonTrackMe>
Quick one, I need to shift sideways the content of an image without changing its size, like opening in paint, select all and moving it around 20px with keyboards arrow (will have a 20px white vertical strip to the left side of it and that is what I want) Can't find the appropriate filter to move around stuff. I know crop & pad to size tho pad to size isn't "add border" and the command will specify some frame size... the filter chain will
<DonTrackMe>
be used to modify png images, to add larger margins to one side of the sheet to print a booklet, every odd number will be shifted slightly to the right and even to the left.
lexano has joined #ffmpeg
fling has quit [Remote host closed the connection]
JanC has quit [Ping timeout: 255 seconds]
JanC has joined #ffmpeg
<BtbN>
crop the original image, then overlay it onto an image of the desired background color and resolution at the right spot
<BtbN>
or even just overlay it. I think the overlay filter will cut off the overhang
<DonTrackMe>
overlay
<DonTrackMe>
i'll give a look at that filter in the documentation.
<DonTrackMe>
also, not sure tho how to give a folder of png's to ffmpeg and have it process only odd or even input file names so that margins gets added on the binding side
fling has joined #ffmpeg
<DonTrackMe>
reading the overlay command description, they say it has two input but what if i simply want to drag my thing 20px to the right, single input? whay would be a simple command? I'm looking at google right now and it seems nobody wants to just move sideways an image...
beaver has quit [Quit: sudo ./give_me_another_relay.sh]
beaver has joined #ffmpeg
<BtbN>
Like I said, overlay it over a static background with your desired specifications, and then shift around the overlay
FH_thecat has quit [Quit: Leaving]
<BtbN>
"Move to the side" makes no sense in of itself
FH_thecat has joined #ffmpeg
<DonTrackMe>
hmmmm
<DonTrackMe>
that's the point, desired specification ( example size of the static white background ), I kinda don't want to input any fixed frame size in the command itself like if i want to move to the right different size images, i don't want to edit the command each time... u know with paint, open, select all, 20 clicks on the right arrow, save... done...
Sakura`Kinomoto has joined #ffmpeg
<DonTrackMe>
I think many years ago I used to batch/macro such stuff in photoshop
militantorc has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
pikapika has joined #ffmpeg
SakuraChan has quit [Ping timeout: 255 seconds]
moviuro has quit [Read error: Connection reset by peer]
kmikita has quit [Ping timeout: 260 seconds]
moviuro has joined #ffmpeg
iconoclast_hero has joined #ffmpeg
<iconoclast_hero>
while this isn't an ffmpeg question, I'm hoping someone can point me in the right direction: i have 2 versions of a song that seem to sound the same but have very different spectrograms....and I'm trying to find some place to ask about. (https://imgur.com/a/uQc8Rid)
omegatron has joined #ffmpeg
<DonTrackMe>
hi there
<DonTrackMe>
the spectrograph are in deed different and one thing i see to the left side of it is that ther is a short purple ish thing in the high frequency at begining and the channels seems to be swapped from one song to another
<DonTrackMe>
and then, i would need to see the MediaInfo of both files and have an explanation of where they both come from to give an opinion about it
kus has quit [Read error: Connection reset by peer]
<DonTrackMe>
are they like two version of the same beatles songs released 10 years apart on different albums / compilations?
<iconoclast_hero>
the MI info is there w/ the 'grams
omegatron has quit [Quit: Power is a curious thing. It can be contained, hidden, locked away, and yet it always breaks free.]
<iconoclast_hero>
as for what they are, they're two versions of buddy miles's them changes
<iconoclast_hero>
(i.e., not a band of gypsies version)
<iconoclast_hero>
i'll drop the flacs on mega.
<DonTrackMe>
10-4
<DonTrackMe>
channels are obviously swapped...
<DonTrackMe>
where did you get those two version of the song? p2p? Some ripping softwares / encoders sometimes swap chanels by mistake...
<iconoclast_hero>
the pirate bay
<iconoclast_hero>
let's just call a spade a spade, lol
<iconoclast_hero>
(sorry if that's not allowed)
<DonTrackMe>
from the spectrograph they look like lossless unless someone painfully encoded one in opus at high bitrate and back to flac, they would have similar spectrograph to full bandwdth with minor differences
<DonTrackMe>
I don't know if it ain't allowed or not. Free speech is free speech to me
<iconoclast_hero>
k
<DonTrackMe>
I have an account on TPB, i'm allowed to upload i mean...
<iconoclast_hero>
so was one always lossless?
<iconoclast_hero>
one's jamal the morrocan
<DonTrackMe>
sometimes two version of the same song have been remixed or remastered on different album releases...
kmikita has joined #ffmpeg
<DonTrackMe>
they usually use multiband compressor and some eq to change the color and the dynamic of a mix already made without having to remix the whole song from the multitrack that sometimes does not even exist anymore
<iconoclast_hero>
the other doesn't say it but the star suggests pmedia
<DonTrackMe>
what do you mean like "one was always lossless"
<DonTrackMe>
if you rip a cd with EAC to FLAC... it will be lossless, period.
<iconoclast_hero>
i did `$ vlc song1.flac & vlc song2.flac & ` and i couldn't tell a difference w/a mono bluetooth speaker (jbl xtreme2)
<DonTrackMe>
tho you can have two different mastering of the same song
<iconoclast_hero>
you said flac > opus > flac which is lossless > lossy > lossless
<iconoclast_hero>
i think my $20 stereo jbl bluetooth headphones are dead.
<iconoclast_hero>
so they're charging.
<DonTrackMe>
yeah yeah but c'mon... I have 10$ bookshelf speaker with a crappy amp connected with a cable to the computer that sounds better than a bluettoth shit
<iconoclast_hero>
girlfriend's house: not my environment.
<iconoclast_hero>
i'm lucky to have my server here.
<DonTrackMe>
thriftstore find, if you know what to shop for, i have small panasonic and lg speakers that are ugly but sounds decent.
<iconoclast_hero>
anyway...
<DonTrackMe>
I would drop the girlfriend...
<iconoclast_hero>
they could be different remasters
<DonTrackMe>
I did it 10 years ago, happy living by myself now
<DonTrackMe>
yeah probably, tho chanel swap is obvious
<iconoclast_hero>
they're probably several.
<iconoclast_hero>
but what makes you think one of them went through opus and which of them?
<iconoclast_hero>
you mentioned a purple artifact...
<iconoclast_hero>
ok, so now i see why you say there was a channel swap.
<DonTrackMe>
none of them probably when to opus, i was just ... painting a worst scenario possible while keeping the spectrum showing up to 22khz...
<DonTrackMe>
usually if the file has been thru mp3 or aac before and back to flac, it's obvious by cutoff at 16 khz ....
<iconoclast_hero>
but it also looks like in the first one, the top channel is ...louder? than the bottom channel of the second one.
<iconoclast_hero>
that's what caught my eye.... why's the second spectrogram, bottom channel so much quieter?
<DonTrackMe>
song bgin with i don't kow some cymbals or high pitched something louder on one chanel...
<iconoclast_hero>
yeah, i got that now.
<iconoclast_hero>
and about opus.
<iconoclast_hero>
yeah, i was looking for the cutoff shelf which is why i started this.
<iconoclast_hero>
really i just wanna know which one to delete!
<DonTrackMe>
u know, remaster people sometimes are old folks that don't even look at a spectrograph and a waveform to see if their end result is properly balanced before writing the thing down for production... you know, analog equipment might be 1-2 db off on left chanel and people don't notice it on analog vu meter or by ear tho it shows on spectrums...
<iconoclast_hero>
i guess i do now.
<DonTrackMe>
do you hear any difference?
<DonTrackMe>
with a mono speaker i guess you cannot...
<iconoclast_hero>
let me put it this way: your point is doubly valid for people who have only looked at a spectrogram a few times.
<iconoclast_hero>
no, i didn't notice anything.
<iconoclast_hero>
so, looking at it, it naturally brought up the questions you were considerate enough to help me with.
<iconoclast_hero>
the spectrogram is far more powerful than my ears will ever be.
<iconoclast_hero>
or even an expereienced audio engineer remastering.
<DonTrackMe>
yeah, if you plan to listen to it thru mono bt speaker, help yourself and the planet and save storage space, encode your library to mp3 320 / aac 224 / opus 192 and you'll never hear a difference
<DonTrackMe>
ears need training like everything else
<iconoclast_hero>
I have better speakers.
<DonTrackMe>
and training with crapp listening devices is difficult as not every color of the pucture is being drawn.
<iconoclast_hero>
but my listening habits are such that i don't listen in environments where quality would make that much difference.
<iconoclast_hero>
if I'm playing volleyball or driving my jeep without doors when i listen, it's not quite so important.
kron has quit [Quit: kron]
<DonTrackMe>
I have my whole library transcoded from flac to opus 192 2 years ago... it's * near lossless at this bitrate in opus
<iconoclast_hero>
and these days that's mostly when i get a chance to listen...but I still want a flac-quality library
<iconoclast_hero>
i know...
<iconoclast_hero>
not only that, but i run my audiobooks at the bare minimum.... 17kb/s and a filter to strip out what I can. I run it through a different sampling rate that saves a few extra %.
<DonTrackMe>
like 192 opus is .... above many content distribution standards! like DVD AC3 384 stereo or 160 aac youtube...
<iconoclast_hero>
i have 5 years of audiobooks...but I "indulge" my music collection with flac.
<DonTrackMe>
17 kbps what format?
<iconoclast_hero>
opus
<DonTrackMe>
hmmm
<iconoclast_hero>
not sure which line it is, but something like ffmpeg -n -nostdin -hide_banner -loglevel error -stats -i "$file" -filter_complex "compand=attacks=0:decays=0.12:points=-70/-900|-40/-90|-35/-37|-21/-18|1/-1|20/-1:soft-knee=0.03:gain=1.00:volume=-90, firequalizer=gain_entry='entry(0,-99);entry(140,0);entry(1000,0);entry(8000,1);entry(10500,4);entry(12000,5);entry(14500,-4);entry(19000,-20)', volume=1.0" -c:a libopus -b:a 17k -ar
<DonTrackMe>
i made some experiments with low bitrates and I found that Libfdk_AAC He-AAC V1 27 kbps mono is not bad
<DonTrackMe>
I sometimes encode low res movies for friends in africa... and i use x264 and mono he_aac and seriously,,, it's mind blowing for such a low bitrate
<another|>
`-ar` twice
<DonTrackMe>
and it's compatible with their old phones and devices down to android 4.1 I think...
<iconoclast_hero>
oh that's why you're not using HEVC
<iconoclast_hero>
or opus for the audio
<DonTrackMe>
well, i use opus for myself.
<iconoclast_hero>
@anotherj, are those -ar's doing the same thing?
<DonTrackMe>
but when i need compatibility with old devices for low bandwidth, he-aac is the hidden gem!
<iconoclast_hero>
yeah, that's what works with my headset in the jeep.
<iconoclast_hero>
JVC boat head unit with a USB stick.
<DonTrackMe>
and reason I use 27 and 35 is that after testing every single value, 24, 25, 26, 27, 28, I found that the frequency kept "real" below sbr changes with the bitrate by steps every 8 kbps increment. Wanting to give the most bitrate to that portion without a next step up, i use bitrate values just below the next frequency step up in the non sbr portion.
<DonTrackMe>
I know, i have too much time to loose sometimes... LOL
<iconoclast_hero>
i guess in answer to both, i basically tinker around with my conversion script every now and again... i really have no idea about the bulk of the ffmpeg functions and there's no way I can write a filter like that.
<DonTrackMe>
do you know ffQueue?
<iconoclast_hero>
so since i only have a cursory understanding of how it works, i do the best i can trying to [re]learn what i'm doing each time.
<iconoclast_hero>
no
<iconoclast_hero>
i just don't use it enough to really retain what i would need to to master it.
<DonTrackMe>
Give it a try, you'll be amazed. you can build presets and it handle many of the common filters with gui, generate and run your scripts in a breeze
<DonTrackMe>
I have presets for various conversion I do
<DonTrackMe>
if one fiter insn't supported in it, in the filter chain of a preset you can add a manual video or audio filter command
beaver has joined #ffmpeg
<DonTrackMe>
I help the programmer of the tool sometimes to add features or fix bugs, the guy is a chef!
<iconoclast_hero>
huh, cool
<iconoclast_hero>
ok. so i have another question: i've been running jamesdsp to do some graphic eq.
<DonTrackMe>
and works windows and linux
<DonTrackMe>
let me google that thing
<iconoclast_hero>
and i look at the presets ostensibly for the type of music i listen to...
<DonTrackMe>
that's a system wide equalizer thing?
<iconoclast_hero>
and they make no sense to me... which, i get, is a matter of taste, but where does one go to look at the theory behind why those eq curves are the way they are...
<iconoclast_hero>
yeah
<iconoclast_hero>
that's not specifically the question.
<DonTrackMe>
well, equalizers... nobody seems to understand their purpose in the first place.
<iconoclast_hero>
seems a reasonable place to start?
<DonTrackMe>
i have an audio engineer background, mixed live bands since i was 14
<DonTrackMe>
well, you start flat...
<DonTrackMe>
lol
<DonTrackMe>
and you take your crapp headphones + a song u know sound right...
<DonTrackMe>
and you find what's annoying and turn it down and what's missing and turn it up a bit... it's a multipass process to have your headphones sound right. And you save this preset as MyHeadphones1
<iconoclast_hero>
the two largest folders in my music collection are Bob Dylan and Grateful Dead fwiw
<DonTrackMe>
if you have a set of speakers, you start flat again with the same song and equalize your room / your sound system
<DonTrackMe>
eq's are made to flatten a sound system / a room in the first place. once you have a room you know sounds flat, you can start mixing something and ovbiously when mixing you have some eq on each instrument input to tweak stuff individualy if lets say the guitar is too bright. you won't adjust the main eq of your system in this case as you know the room has been fixed and doing so might mellow the mix later on too much as you'll arrive
<DonTrackMe>
adding the keyboards in the mix.
<DonTrackMe>
when at home listening to mixes other people made prior, an eq should only have the purpose to flatten the room or give the sound, once flattened, the color you like .
<iconoclast_hero>
ok.
<iconoclast_hero>
that's what I'm looking to do: the color i like.
<iconoclast_hero>
i just don't understand why this seems like what i like
<DonTrackMe>
use a third octave ( 31 band ) eq to flatten the room, and on top of it, use a single octave eq ( 10 band ) to give coarse color
<DonTrackMe>
color adjustment can often be better adjusted with low number of bands eq, like 5-6 bands one as they affect wider range of frequency, generating less annoying spikes, the one you're actually trying to kill out with a finer eq (31 band)
<iconoclast_hero>
like i assume there's some theory for generating these curves.
<DonTrackMe>
the reason for it is that many systems don't compensate for human hearing curve.
<iconoclast_hero>
oh you mean where we hear better in the frequency spectrum?
<DonTrackMe>
the volume of air containted in your ear canal between the outside pavillon of your ear and your eardrum creates a reasonant chamber around 3 khz (average) varies from one person to another.
<DonTrackMe>
those frequency are the first ones to be scratchy / annoying / painful for this reason as they stimulate the natural reasonant frequency of your ear canal.
<iconoclast_hero>
and that resonant chamber can set up a standing wave or something?
<DonTrackMe>
when wearing hearing aids, they fill up that chamber and thus, this have to be compensated by the audioposthesis. the frequency response + the reasonant response of each person hear can be measured by an audiologist...
<iconoclast_hero>
(as a side note, buddy miles, played with hendrix in band of gypsies, was the voice of california raisins)
<DonTrackMe>
yeah, any cavity, elevator room, public toilet metal cabinet have a reasonant frequency, if with your mouth you make sweeping sinewaves from low to high pitch while in those rooms you can find that at some point, you can almost whisper this frequency tho it's naturally amplified by the nature of the room you'Re in
<iconoclast_hero>
yeah, not exactly the same but i took physical chemistry 2 (quantum) so i have some concept of wave mechanics.
<DonTrackMe>
when mixing a band in a hotel room, sometimes i have one note being played by the keyboardist that naturally ring the room or something in the room, i stop everybody and ask the guy to keep playing the note as i go kill it with i notch parametric eq filter
<iconoclast_hero>
can you guess by looking at the room approximately what note that will be?
<DonTrackMe>
not really... because it ait't only the size of the room but let's say the room have 4 identical hollowed boxes at the cieling above which some glass sky light thing are installed to let natural light come in the room... sometimes the room itself is so large, its natural reasonant frequancy is really low and won't cause a spike, low frequencies are easily corrected while eqing with a 31band eq... tho those small boxes might have a higher
<DonTrackMe>
pitch themselves and be the cause of a note you might need to kill with a notch filter at 2k
<DonTrackMe>
you never know in advance.
minimal has joined #ffmpeg
<DonTrackMe>
get in a room, hook a system, put every eq flat, put some known music, start eqing with the 31 band eq...
<DonTrackMe>
sometimes in conventions, you have a translator booth in the back and they use a system of headset in which you feed microphones from the stage... well, if you want to blow their mind, you take one of those headset that sound like a can, put it on your head, start an eq for the aux output you feed their system with and you basically eq their headset... now their crappy headset have been fixed for today and they hear stuff from the stage
<DonTrackMe>
with all the richness it should have...
<DonTrackMe>
and without the annoying spiky 3k thing that those half inch crappy drivers naturally have...
<DonTrackMe>
also, since your hearing system in your body naturally amplify 3k ish, if you abuse your ears, the first place damage will ocur is in this area.
<iconoclast_hero>
yeah.
<iconoclast_hero>
I'm going to assume that we've evolved to hear best human speech.
<DonTrackMe>
That for me also explain why some old man mixes are sometimes harsh for my ears I always took care of. they don't hear that 3k thing much so they even put a boost there when mixing... those bastards...
<DonTrackMe>
yeah, i guess that's part of evolution...
<iconoclast_hero>
and one thing i should be concerned about is listening to audiobooks. when i told my dr. about that, he was like great, you;re going to lose human speech first.
<DonTrackMe>
tho fundamentals in human speach are lower than 3k a bit...
<iconoclast_hero>
now he's in federal prison on narcotics distribution charges.
<DonTrackMe>
phone stuff is capped at 4 sometimes 3
<DonTrackMe>
loose human sppeech ?
<DonTrackMe>
what kind of medical advice is that?
<iconoclast_hero>
that will be the range of hearing loss i'm likely to suffer first if i listen to audiobooks at damaging volumes.
<DonTrackMe>
hmmm...
<iconoclast_hero>
and the advice was to not listen to books at such volumes.
<DonTrackMe>
actually, the 3k thing is more like baby crying...
<iconoclast_hero>
ok, well then perhaps my guess was off.
<DonTrackMe>
and from an evolution point, you can understand why for a mother it's relevant for her to hear her baby crying from a distance!
<DonTrackMe>
and that explains also why from close up, a crying baby IS annoying
<iconoclast_hero>
yeah, i mean we didn't develop sight in the wavelengths we did because we need to see UV flower patterns.
<DonTrackMe>
nah, we aren't bees
<iconoclast_hero>
so [literally] naturally there's a reason we hear the way we do.