| Author
|
Isratrance Mastering Shootout 1
|
xoC
Cubic Spline
Started Topics :
10
Posts :
179
Posted : Oct 7, 2012 23:01
|
Quote:
|
On 2012-10-07 12:25, daark wrote:
....and
Overdream-
Verrry nice 
The sound is flat and clean. Problem here it is only a little bit too much air very bit but its subjective. Very good job.
Hermetech-
This is a very good one. You stayed on track and the result is a good clean sound.Great job.
To note nobody did a perfect job imho and the sound changes from person to person drasticaly each has his own aprroach so far .In one thing you nail but the other thing you forget. 
Guys some of your links went down on dropbox you should recheck them Richard , SineFreq and Cubic Spline.
|
|
hm mine is still online :
https://www.dropbox.com/s/b65atc5e2tc1pcm/Pawnsphinx%20-%20WinterFur%20-%20Storm%20Mastering.wav  http://www.storm-mastering.com |
|
|
|
daark
IsraTrance Full Member
Started Topics :
58
Posts :
1397
Posted : Oct 8, 2012 00:02
|
|
richard
Started Topics :
0
Posts :
7
Posted : Oct 9, 2012 22:49
|
|
no other comments on group2? Don't be shy guys... |
|
|
|
Babaluma
IsraTrance Junior Member
Started Topics :
18
Posts :
729
Posted : Oct 10, 2012 00:48
|
|
PoM
IsraTrance Full Member
Started Topics :
162
Posts :
8087
Posted : Oct 10, 2012 01:15
|
Quote:
|
PoM-
Such a nice sound
Bu here the mix is boomy. Just that out of the equation and its perfect.
|
|
yep the tonal balance is not good , couldnt get it right |
|
|
|
daark
IsraTrance Full Member
Started Topics :
58
Posts :
1397
Posted : Oct 11, 2012 14:06
|
|
skyhighatrist
skyhighatrist
Started Topics :
6
Posts :
101
Posted : Oct 12, 2012 12:02
|
Just out of interest, for psy tracks, what kind of RMS level do you guys look to achieve? It seems there's a fair bit of variation in the current releases. On the whole, they are much louder than releases say 3-4 years ago.
Do you guys have a target RMS in mind or just make the track as loud as possible without artifacts? If you're doing an album, how do you decide what overall volume level you will master each track to?
Thanks
http://www.catawampus-records.co.uk/Catawampus_Records/Skyhighatrist.html |
|
|
|
xoC
Cubic Spline
Started Topics :
10
Posts :
179
Posted : Oct 12, 2012 13:01
|
Quote:
|
On 2012-10-12 12:02, skyhighatrist wrote:
Just out of interest, for psy tracks, what kind of RMS level do you guys look to achieve? It seems there's a fair bit of variation in the current releases. On the whole, they are much louder than releases say 3-4 years ago.
Do you guys have a target RMS in mind or just make the track as loud as possible without artifacts? If you're doing an album, how do you decide what overall volume level you will master each track to?
Thanks
|
|
A lot of releases today get out around -5 dB RMS (AES 17 standard, which means -8 dB RMS on the old standard), and it's too loud I think. On some really well mixed materials, you could go at -5 dB without to much artefacts, but on not so well mixed stuff, it can be really awful.
-7/-8 dB RMS is a nice level for Psytrance tracks. It's just 2-3 dB quieter, but requires considerably less limiting.
My last mix was -3/-13 (peak/RMS) on full power passage, it required very little processing to get arround -7.5 RMS, which is really enough for playing in my live set, and still punchy enough
But who decides ? The artist/label
What target ? It really depends on the production quality and what they want ... as many will say, really well mixed tracks can handle a lot more abuse.
edit : those numbers can also be very misleading. A bass heavy track at -5 dB RMS can breathe better than a bass light track at -6/-7 ... thoses sub frequencies can really increase the RMS value. I really hate those modern -5 RMS productions without much subs, they are really harsh & squashed. http://www.storm-mastering.com |
|
|
|
skyhighatrist
skyhighatrist
Started Topics :
6
Posts :
101
Posted : Oct 12, 2012 13:24
|
Thanks
This is exactly what I have noticed. 3-4 years ago RMS values were between -7/-8 and have gradually increased to around -5 now.
I have always tended to prefer being be slightly quieter to having artifacts and i dont like a squashed sound but put side by side there's a fair difference between -5 and -7/-8. With the louder track often being perceived as better by some listeners. So I guess the dilemma for me is whether to stick at -7/-8 or push it a bit more say -6 ish... -5 to me often is a bit much...
I'm very keen to hear some opinions on it 
you're right about the bass content of course that does make a difference and RMS does not necessarily equal how loud the listener would perceive a track.
http://www.catawampus-records.co.uk/Catawampus_Records/Skyhighatrist.html |
|
|
|
Babaluma
IsraTrance Junior Member
Started Topics :
18
Posts :
729
Posted : Oct 12, 2012 14:20
|
there is no single answer to most of your questions, except, "it all depends on..."
i don't have an RMS level in mind when i master. i make it sound as good as i can without any limiting, and then work from there with the artist. some artists want it limited a lot, and louder, some don't want any limiting at all, and prefer a more dynamic sound.
having said all that, i've measured older Psy releases with RMS and DR meters, compared them to what's about now, and the loudness has definitely crept up, but that's been the same with most forms of popular music since the invention of Waves L1 in the early 90's. newer releases tend to have 4-6dB less dynamic range than the older Psy and Goa tracks, which is a fair bit more slammed.
if i'm doing an album, i get each track sounding as good as i can, then go back and listen to all the tracks in the correct order, making slight loudness/limiting adjustments as necessary, and making sure the flow is good. the idea is not always to get every track sounding the same loudness (and definitely not by matching RMS levels, which as i keep harping on about, are a pretty useless way to measure percieved loudness), for example if you have a ballad or an ambient track amidst a rock or psy album, you don't necessarily want that track to sound as loud as the others. it's all about the flow, and keeping the intentions of the artist and the music in mind whilst working.
if it was down to me i would never use any limiting at all. it always colours the sound in a bad way, to my ears, especially if you do more than a dB or so.
xoC, i think you have your standards the wrong way round. the AES standard goes back years, and as it's +3dB (i think because it is referenced to a sine wave and not a square wave), that means that a correct AES average RMS measurement of -5dBfs actually means it would be -2dBfs with the wrong standard that many people are still using, and NOT -8dBfs. the AES RMS measurement reads 3dB QUIETER than the other (because it's already set at +3dB, right)?
the AES is THE way to measure audio average RMS, many of the DAWs have had inaccurate meters for many years.
http://hermetechmastering.com : http://www.discogs.com/artist/Gregg+Janman : http://soundcloud.com/babaluma |
|
|
|
xoC
Cubic Spline
Started Topics :
10
Posts :
179
Posted : Oct 12, 2012 14:47
|
Quote:
|
On 2012-10-12 14:20, Babaluma wrote:
xoC, i think you have your standards the wrong way round. the AES standard goes back years, and as it's +3dB (i think because it is referenced to a sine wave and not a square wave), that means that a correct AES average RMS measurement of -5dBfs actually means it would be -2dBfs with the wrong standard that many people are still using, and NOT -8dBfs. the AES RMS measurement reads 3dB QUIETER than the other (because it's already set at +3dB, right)?
the AES is THE way to measure audio average RMS, many of the DAWs have had inaccurate meters for many years.
|
|
No. Old standard refers to the way to measure mathematically the RMS value : i.e. referenced to a square wave (the RMS value of a square wav is mathematically the same as it's peak value). In this standard, RMS square = 0, and RMS of a sine is sqrt(2) it's peak value, i.e. 3 db less, so RMS sine = -3.
In new standard (AES 17), it is posed as the peak and the RMS of a sine are equal, i.e. RMS sine = 0, RMS square = +3.
So to go from old to AES 17, you add 3 dB.
There is an option in wavelab, when you activate AES 17 standard, all RMS values reads +3 dB. The same in voxengo span, pure is old, pure+3 is AES 17.
I agree with you on the part that AES 17 is the way to correctly measure RMS values.
Why put this standard which, mathematically speaking is weird ? I think it has to do that a full scale sine is sounding pure, and as soon as you're trying to get more RMS level, you are distording it. Square waves full scale also aren't really something nice to listen, so it makes sense from the listening point to say 0 RMS is the value of a full scale sine. http://www.storm-mastering.com |
|
|
|
Babaluma
IsraTrance Junior Member
Started Topics :
18
Posts :
729
Posted : Oct 12, 2012 15:43
|
|
xoC
Cubic Spline
Started Topics :
10
Posts :
179
Posted : Oct 12, 2012 20:10
|
|
Babaluma
IsraTrance Junior Member
Started Topics :
18
Posts :
729
Posted : Oct 17, 2012 14:01
|
OK here's more from me!
First up notes on Group B, once again all tracks time aligned and level aligned in a DAW so that I could easily mute and compare with the original file:
DocDic
Nice level and clarity in the high end, smooth sound, but I feel like too much has been carved away from the bass/lower mid range, leading to a somewhat "thin" sound, compared to the original mix. You also have some digital overs, which could cause nasty distortion on some playback systems, but this isn't something many people worry about, as I have seen pop releases with digital overs in the millions!
Upavas
I liked this one a lot, powerful sound, more limiting than I would use, but sounds good. Got the bass end sound really full. Minus point: a lot of the mid range now sounds too scooped, leading to a lot of the melodic lines sounding far less prominent.
Overdream
Very nice, faithful to the original, while bringing out more detail. All ranges sound pretty spot on to me, great work. Love the clarity you have achieved here.
Padmapani
Overall tonal balance nice, but something sounds off in the stereo field, things sound very shifted about compared to the original mix. Quite a lot of limiting too, sounds a little too much/distorted for me.
PoM
Very nice and clear, mids and highs sounding great. Level great (as in you followed the artist's requirement of not over limiting). Again, not much to complain about here, good work. Maybe my favourite.
SinFreq
Really phasey, nasty highs and mashed stereo image. Didn't like this one at all.
Storm
Very faithful to the original track with more overall clarity. Nice level. Another great one!
Doesn't seem like anyone else is gonna comment (you've had AGES!!!) So, here's what I did:
Digital track went straight out the Crookwood DAC and into the analogue chain.
Chandler Germanium compressors were mainly set very clean, maximum GR about 1.5dB. HPF side-chain at 60Hz. Ratio very low. Knee quite soft, with just a bit of "squidge". Attack set for maximum impact on all those lovely transients (a little slow, letting them all through uncompressed). Release at lowest value. Make up gain set for some colour (these comps have a great line amp at the end of the chain, with two controls, Drive & Feedback, so you can achieve different tonalities), here a slight treble roll-off with a slight bass boost.
Pullet EQ was set with a slight medium bell mid boost at 2.5kHz for "popping" the melodic sounds and snare, slight medium bell mid cut at 340Hz to clean up the slightly muddy lower mid range, and a high shelf boost at 15kHz for "air". This was then followed by 40dB of make up gain from the Chandler TG2 preamp (which also colours things with a slight mid forward sound).
The Bax EQ was set with a HPF at 36Hz to filter out sub-sonic rumble, followed by a shelving 0.5 dB boost at 74Hz to get that bass and kick a bit more prominent. The combination of HPF and low shelf is always fun to play with. The high shelf was set with 0.5dB boost at 4.8kHz, again to bring out more mid range detail. Finally the LPF was set at 28kHz to filter out any ultrasonic rumble, and tame the highs a bit after all that mid/high shelving I had done!
Back into the PC via the Crookwood ADC, with almost, but no clipping. Then -0.2dB on the mid signal, for a very slight stereo widening effect. Elephant limiter was set in El Uni mode with 1.7dB of gain, with a maximum GR of 1.4dB. 4 x oversampling was used to prevent IPSs, and the output ceiling was set at -0.0 (this was experimental, I nearly always leave it set at -0.3). Then final render down to 16 bit with Sonoris dither (TPDF with no shaping).
I also thought it would be interesting to run everyone's files though the DR meter plugin in FB2K, it's a quick way to get a good feel for everyone's loudness levels, although it doesn't include ISPs (inter sample peaks).
foobar2000 1.1.15 / Dynamic Range Meter 1.1.1
log date: 2012-10-17 13:46:07
DR Peak RMS Duration Track
--------------------------------------------------------------------------------
DR13 -2.07 dB -18.36 dB 6:34 ?-Pawnsphinx - Winter Fur (Original Mix)
DR9 -0.32 dB -11.62 dB 6:33 ?-Pawnsphinx - Winter Fur (Bleeps)
DR8 -0.13 dB -10.26 dB 6:31 ?-Pawnsphinx - Winter Fur (Colin C)
DR6 -0.30 dB -8.04 dB 6:34 ?-Pawnsphinx - Winter Fur (Crn)
DR8 -0.17 dB -10.69 dB 6:34 ?-Pawnsphinx - Winter Fur (Daark)
DR10 -0.02 dB -12.27 dB 6:31 ?-Pawnsphinx - Winter Fur (Hermetech)
DR7 -0.07 dB -10.22 dB 6:30 ?-Pawnsphinx - Winter Fur (KGB 16D)
DR10 -0.30 dB -12.73 dB 6:34 ?-Pawnsphinx - Winter Fur (KnocZ)
DR9 0.00 dB -10.92 dB 6:36 ?-Pawnsphinx - Winter Fur (DocDic)
DR8 -0.10 dB -10.75 dB 6:31 ?-Pawnsphinx - Winter Fur (Overdream)
DR7 -0.09 dB -9.40 dB 6:34 ?-Pawnsphinx - Winter Fur (Padmapani)
DR7 -1.29 dB -10.61 dB 6:34 ?-Pawnsphinx - Winter Fur (PoM)
DR7 -0.14 dB -9.78 dB 6:31 ?-Pawnsphinx - Winter Fur (SineFreq)
DR9 -0.16 dB -11.53 dB 6:34 ?-Pawnsphinx - Winter Fur (Storm)
DR6 -0.20 dB -9.26 dB 6:31 ?-Pawnsphinx - Winter Fur (Upavas)
--------------------------------------------------------------------------------
Number of tracks: 15
Samplerate: 44100 Hz
Channels: 2
Bits per sample: 16
Bitrate: 1411 kbps
Codec: PCM(1-7),PCM (floating-point)(8),PCM(9-15)
So KnocZ, you loose points for returning the file as 32 bit float I believe!
Finally, the whole idea is that EVERYONE who submitted a master comments on EVERYONE else's masters, so that we can all learn from one another, otherwise it's a waste of time and unfair to those who do take the time to do it. Over on the REP forum, where we do this every few months, if you don't comment you are automatically banned from entering the next one! I dunno, it just seems like we haven't had 14 people commenting yet...
http://hermetechmastering.com : http://www.discogs.com/artist/Gregg+Janman : http://soundcloud.com/babaluma |
|
|
|
makus
Overdream
Started Topics :
82
Posts :
3087
Posted : Oct 17, 2012 14:27
|
Guys! Thanks for commenting!
babaluma, special thanks for your work and positive comment on my mastering version.
recently ive been busy with several tv projects at once so i couldnt afford listening 15 versions of mastering examples we've got! So sorry. However will try to do it when get some free time later.
only once ive listened to where Colins (which impressed me a lot!), Storms and Hermetech's. These two are pretty damn good too. Cant wait to get Poms version
will post my mastering chain later today.
www.overdreamstudio.com |
|
|
|
|
