Members TonyCrazyMan Posted July 27, 2005 Members Share Posted July 27, 2005 Hi! I just came from my home studio. Today I was experimenting with mixing, mastering and effects. I recorded everything as dry as I could in 44,1KHz/24-bit and normalized each track to get every sound equally loud and downsampled + dithered the demo sample (hi-hat,snare,kick drum, bass, piano) into 16-bit. Then I analyzed how big the difference was between a completely dry material and a mixed and mastered complete product. I used the best effects I knew and really tried to do everything I could to make the mixing and mastering as successful as I could, I tried to avoid bleeding, don't apply too much reverb etc. When the final mastering was finished I compared these two versions 5 times. The result was 5-0. The clean version won, big time! Then I asked a friend to take a listen to the two clips. First he didn't really know. Then I said maybe the volume is not loud enough so I raised it. Then he suddenly said: "Now I know!" Which one is it, I asked... The first one! (the dry version) Then I thought for my self, he is correct, that's true, the dry mix is much better! I thought this was pretty interesting, because I have really never liked any digital effects. I use Waves Native Platinum Bundle and the effects in Neuendo + Sonar. I'm sure this was also due to my incomplete knowledge in mixing and mastering, but I really think that the more digital effects I apply on a track, the worse is the final result! As a matter of fact, I thought the dry version I recorded today was one of the best sounding samples I've done so far! The kick drum had a nice mellow punch, the snare was very clear, the piano was much more clear too, and it was not even mixed with volume yet! Completely dry! The effects I used in the mixing and mastering were: Waves C1 compressor Steinberg Reverb B Waves 10-band Parametric EQ Sonitus:fx Equalizer Waves Ultramaximizer + Anybody else than me that think these effects don't do any good? In what area do you think I probably made the final result worse when mixing and mastering? What should I do now? Start recording dry?! Here are the two clips: Demo Sample Clean - 16bit.mp3 Demo Sample Final - 16bit.mp3 Which version do you think is better? Link to comment Share on other sites More sharing options...
Members rickenbacker198 Posted July 27, 2005 Members Share Posted July 27, 2005 Well I think the dry version is a bit louder sounding is the volume higher??.. there also seems to be a lot more dynamic range. Dry sounds best IMHO... but it does sound a little rougher than the wet version, so some may not like that. NB Link to comment Share on other sites More sharing options...
Members TonyCrazyMan Posted July 27, 2005 Author Members Share Posted July 27, 2005 Originally posted by rickenbacker198 Well I think the dry version is a bit louder sounding is the volume higher??.. there also seems to be a lot more dynamic range. Dry sounds best IMHO... but it does sound a little rougher than the wet version, so some may not like that. NB Hehe, another one that thinks the dry version is better! In the clean version the volume is at max. But since I apply effects on the wet version the dry signal is not as loud. Also the wet version includes some panning too which further makes the result not quite as loud. I also did a little mastering trick: I overdubbed with a track that only included very soft round bass- I think that affected the loudness negatively even though I got some more heat on the bass+kick drum hits. But I completely agree, the clean version has more sound, more dynamic range. It is a little wider sounding and I like that even though it feels and is very panned to the center. Overall I have so far had a hard time getting the result loud enough. I have had much problems with the loudness/wideness being too low/little. The dry mix is more powerful. I would like to keep that just to keep the brightness, but get a better balance in the mix, some more reverb and more roundness in the bass of the overall mix. I would also like the treble to be a little softer and thinner. What should I do to achieve this? Link to comment Share on other sites More sharing options...
Members TonyCrazyMan Posted July 27, 2005 Author Members Share Posted July 27, 2005 What happens if I take a 10-band Parametric EQ and add 18 dB on the whole frequency range on all the tracks? I wanted to test that and did it. At the same time I added a B3 organ an analog synth sound and some digital pianos on the sound sample to make it a little more like a song. This time I edited the effects very sparingly and only in Sonar. However I noticed clipping on one part of the demo, must have been due to one of the effects that didn't include a output meter/clip indicator. What you will notice on this clip is that the drum stick sound in the song is a little fatter and has pretty much reverb. I thought it was ok like that. I also felt I wanted to increase the loudness of the kick drum and the snare drum, which were both too low in the mix I did earlier. I added some more bass in the snare too. I'm pretty satisfied with the drum sounds (except I think another snare type would be cool too) since it's all coming from a keyboard. What I have to continue learn is how to produce a good warm bass sound, that is very important and pretty difficult... Here is the remix: Remix Link to comment Share on other sites More sharing options...
Members AudioMaverick Posted July 28, 2005 Members Share Posted July 28, 2005 Yep, Tony... The effects aren't matching the music. I think that is what you are running up against. Before adding any effect, think of what you want it to accomplish. The human brain has accumulated years of audio experiences from real life. If the effect doesn't match, it isn't *natural*. Plus, the piano effect is coming across as a stereo effect. This seems to be causing a volume drop from phasing. Try it as a mono effect. YMMV... Link to comment Share on other sites More sharing options...
Members TonyCrazyMan Posted July 28, 2005 Author Members Share Posted July 28, 2005 Originally posted by AudioMaverick Yep, Tony... The effects aren't matching the music. I think that is what you are running up against. Before adding any effect, think of what you want it to accomplish. The human brain has accumulated years of audio experiences from real life. If the effect doesn't match, it isn't *natural*. Plus, the piano effect is coming across as a stereo effect. This seems to be causing a volume drop from phasing. Try it as a mono effect. YMMV... Interesting, when it comes to the piano I think it is the only instrument in the mix that is coming from a stereo sample. It's "full grand". This means it feels a little wider. But to be honest I don't understand phasing and why it's causing volume drop. But I will test split up the track into a mono track and see what happens. If I get a nice loudness increase I've learned something new! Link to comment Share on other sites More sharing options...
Members Kiwiburger Posted July 28, 2005 Members Share Posted July 28, 2005 Phasing isn't hard concept. Imagine two sine waves being mixed together. If they are perfectly aligned (phase coherent) then the two waveforms will add up to create one big sinewave. The amplitude will be doubled - which in binary terms means an extra bit, or in peak level terms a 6dB boost. If - however - one of the sine waves was shifted along 180 degrees, so that the negative cycle was opposing the positive cycle, these two sine waves would add up to zero. Complete cancellation. Different phase shifts create different boost or cancellations at different frequencies. With multiple sources of essentially the same sound (e.g. multiple mics, or delays/reflections) there is the potenial for serious 'phase' issues. Link to comment Share on other sites More sharing options...
Members TonyCrazyMan Posted July 28, 2005 Author Members Share Posted July 28, 2005 Originally posted by Kiwiburger Phasing isn't hard concept. Imagine two sine waves being mixed together. If they are perfectly aligned (phase coherent) then the two waveforms will add up to create one big sinewave. The amplitude will be doubled - which in binary terms means an extra bit, or in peak level terms a 6dB boost.If - however - one of the sine waves was shifted along 180 degrees, so that the negative cycle was opposing the positive cycle, these two sine waves would add up to zero. Complete cancellation. Different phase shifts create different boost or cancellations at different frequencies. With multiple sources of essentially the same sound (e.g. multiple mics, or delays/reflections) there is the potenial for serious 'phase' issues. Wow, this phenomenon is completely new for me. What I come to think of is overdubbing and "phase" issues. If two tracks are overdubbed, the waveform is on top of each other. That would create phasing that in turn would work against the loudness of other instruments in the mix because the phasing causes a 6dB amplitude boost if I don't do anything about it? So it is the same with stereo samples? I try to imagine what effect this has to the overall mix. Well, first of all the instrument that is in coherent phasing is 6dB louder, that means normalizing all the tracks would correct the peak amplitude issue. But now the problem is left in terms of boosts of some frequencies on the frequency range? Right? Let's say the coherent phasing caused by a stereo sampled piano sound causes a boost on 700-1000Hz. Would this then cause other instruments in the mix to dissapear even though the peak amplitude of all these instruments are the same? A fader volume adjustment on the "shadowed" tracks to compensate for this would then yield a negative result on the overall mix loudness, right? If this is the case it would be nice to have a plugin that analyzes the overall dB/frequency over the whole frequency range so that you would more easily discover phase issues and correct them and also to be able to know which instrument might add much too much/little amplitude on some frequency on some instrument causing the total mix not being balanced enough. What do you do in order to control the amplitude of frequencies on different input tracks? Do you have any EQ analyzer tool? Link to comment Share on other sites More sharing options...
Members greendoor Posted July 28, 2005 Members Share Posted July 28, 2005 To get a 6dB boost (full doubling in power) you need to add an exact clone of the original wave. If overdubbing or mixing two non-identical audio waveforms together, there will be some cancellations and some boosts, so probably the sum will be around 3dB hotter. In a complex mix - especially with multi-mikes and delays - it's hard to predict exactly what is going on. Piano's that have been mic'd with a spaced stereo pair can be problematic. Often the midrange is weak, because of phase cancellations between the two mics, but the bass and treble notes can be fine because they are concentrated more in one mic. I have this complaint about many piano sample libraries. And stop it with this normalisation fetish! The solution for phase problems is to eliminate the problem sources, or shift them. A polarity switch gives a 180 degree phase shift (semantics aside) that can solve some problems. Small delays can provide more accurate control. EQ might be a last resort fix. A spectrum analyser or phase meters are the graphical eq tool that you might want to use, but I think ears are much more sensitive to phase problems. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.