Next project: Cocoon

[boody]

Hi,

for anyone who's intrested in giving some advise, here's a song of the next project I've got to mix. This will be for a record (no demo this time). The artist is called Xilke, the music is a mixture of a lot of styles.

The song starts out with programmed drums & real cymbals. The finale is where the real drums kick in.

This mix is a close setup, automation is only very roughly done.

link: Cocoon


All critique is very welcome! Thanks in advance, Budy

ps: this time I tried a "lame" mp3 encoder with variable bitrate

[NeoVXR]

{** the following comment is to an earlier version of the above song, that has been replaced **}

hi boody,

your level is high, as mr.jazzius is saying, too.
He will see what I dont see...

the mp3 is quite good. be aware that LAME sometimes eats up a little bit mid-range, this is always for short single notes or certain sounds.

from the second half the voice begins to become somewhat thin. (maybe this could be from mp3).
if you see it also, you might try something like a soft C4 setup to enhance mid-range where it drops, or ride with the EQ.
or try a very miminal stereo delay 15-30msec.
also mastering can change this a lot.

referring to jazzius' remarks about high-end, also here a little bit. I think it is quite better than "obstruction" in this. but you have a boost above 8k that is to check out if it fits the music and the different listening equipments. the right channel ist also brighter there.
do you listen kruder & dorfmeister? on their double album I found it similar. maybe the genre made up a habit? seems to be the drum samples, and the listening habit. seems due to the wide sampling culture, that a "limited" kind of sound came out as a fashion. (very old vinyl samples e.g.)
I would see it as a matter of taste, not only a technical issue.
maybe you can talk to DJs that are into the style...

when a hihat or cymbal has a very flat xtreme high end, it does not help whether it goes to 21kHz. it is becoming like white noise, and losing personality. this is exactly what the first electronic drums were made of, and became a fashion.
so, when you place mics, or choose the hihat if there are a few in the transport case you can let a little metallic sharpness in!

only an idea for what could become the scope of experimenting.

who left something like vinyl noise at the end?
is this intellectual attitude or so?
I would prefer some sounds of nature. (the sea, the wind, traffic, restaurant ...)


good success!

[boody]

Hi Neo,

thanks again for the comments; always good to hear from you.

Right now I finished the second version; uploading it in a minute. Hopefully the voice is better now? It's a difficult track. I also added a short dub-delay.

The triangle is tricky. I tried to put it in mildly, I like the sound.

The lack of high end is a gear issue I think. My gear is quite low budget. I'll have to figure out a way to make it more transparent, but I guess my tascam dm-24 isn't helping. I have no idea how the good or bad the conversion & summing in that device is. Anyone?

I track my drums in a small vocal booth with the only mics I have: two large condensers, one small, a sm57 and some sm58s. No good room ambience, so I guess that could make the high end a bit flat as well.

The vinyl is the end of the record with the whales on it... I thought it was a nice ending, it happened spontanuous.

I'm always trying to improve, so any suggestion for experiments are welcome and will be tried (within reason that is )

regards, Budy

[NeoVXR]

don't change too much!
don't worry too much about the gear. the music is speaking for itself anyway. it just needs more time and more projects.
try a boombox and a bar/pub (the DJ issue). try listen with 9khz lowpass. this is real music, it should even survive without HF at all.

what if you get the whale from a CD?

[boody]

Originally posted by NeoVXR
what if you get the whale from a CD?

then I wouldn't get all the nice crackles...

[NeoVXR]

[NeoVXR]

much better now! everything is more transparent.
in the second half, you made it brighter, which is very good, but now the bass cannot catch up enough and the low mid balance between the tracks reveals some questions.
also, effects in the second part take away still some transparency. the idea is to lessen high mids or "presence" from the effects.

(I did an experiment in the meantime (first version) that gives a fat and transparent sound to this. I admit its duller than yours, but the optimum would be the strengths of both. In case you want to compare, please tell me.)

[boody]

I just listened to it on my cheapo hifi dvd system... still not too happy, especialy with the first part.

You're probably right about the fx, I'll look into that. I would love to hear your experiment!

[NeoVXR]

Here we are. This one is serious.
(sorry for the big file, mp3 translates bad with my experiments
on my test webspace its 9MB full length. those interested send me a note I'll return a link)

seems I found out something about the transparency

[boody]

thanks: I'll check it asap!

[boody]

Checked it

so what exactly did you do? I hear it sounds a bit warmer, indeed a bit duller too. I assume you took away some presence, around 8-10khz?

This weekend I'm not able to mix but I'll continue my quest on monday. I'll experiment with the fx presence and see if I can get things to clear up a bit. Still too much overlapping frequencies I guess.

regards,
Budy

[NeoVXR]

FYI - the versions in my link and that one above in this thread sound somewhat different!

I am testing a sophisticated process. Its totally from scratch.

Split at a chosen frequency (also split-width to consider).

Apply a special hand drawn-dynamic curve (the dynamics of the HF band have to be observed and critical dB values recognised as parameters. Lookahead time is also critical. Time constants are ultra fast.)

Put the thing together again. (this is only a short description, there is more to consider and tweak)

A little final equalizing.

I took away presence, but in a dynamic way, so the snare hits punch through, while the mess of "all too much" goes away.
The 2 parts of the song were processed differently.
The idea is also to nail HF levels to where they belong, and create kind of "layers" of dynamic action.

Part of the dullness comes from 16bit processing. (Don't like to wait so long.) Above 12 kHz the quality is not good, but it rewards with nice translation in boomboxes and radios.
The warmth might be a result of the process resembling kind of tape-dolby.


Poor mp3 translation seems a good sign here. It is telling that I reduced redundancy and enhanced the density of signal information (in the representation of samples). The WAV is definitely better and less dull.

[boody]

wow... sounds like a lot of work

I once heared of a guy who restored old audio recordings in a simular way. The results were said to be amazing and to good to be true.

First I want to try to get it right in the mix though

Thanks so far for the effort thumbsup The quest will continue monday

have a nice weekend,
Budy

[NeoVXR]

the work is to listen and choose parameters. good for training

>have a nice weekend
same 2 U

PS
watch out for the rimshot. is it too "snappy"?
also the sensitive processing seems to reveal some distortion in a couple of tracks.

----
here is a louder version
I can promise it sounds interesting. somewhat more pop but - bad?
(tried how far I can compress. some distortion seems already inherited - results would be better with less saturation and compression from the source mix. I hope you readers who are interested kept the file to compare, boody replaced it in the meantime)
hope you love it!

** sorry I replaced the file with other stuff - it's gone now ** (19.7.04)

[boody]

.. and I replaced it yet again

new version up. Any thoughts on this one?

Thanks NeoVXR, I got a lot of hints for improvements from your examples and tried to get them in the mix. The processing you do is quite intresting. Very different approach, taking advantage of the digital era . Remarkable how the vocals got more personality in your last version.

Regards, Budy

ps; this time I compressed with the frauenhofer algorithm, again vbr.

[NeoVXR]

boody, your LAME 3.92 files were very fine!
I jumped on a weblog telling in short that fhg sucks with high bit rates. this corresponds with some of my observations, that HF clarity did not become better after a certain ceiling, when I increased bit rate. OTOH it is possible that newer versions have corrected this problem already.
I am using LAME with the razorlame frontend, so it is easy to tweak.

I'm happy that I was on a hot track with the voice, but I liked also what happened to the drums. thumbsup
will listen the new one...