Serum problem

[ShadowRaz]

Ok so i have this weird problem with Serum which i use as a sound design plugin. Some drone, growl designs are WAY more quiet in the export audio file compared to listening inside the project. Why is that? Can it be fixed somehow? Or does it require exporting the drones growls as an audio file for example a.k.a freezing that precise mixer channel as that action sometimes is called

It also presents itself inside the project also as in when i first listen to the track if i were to spot something to change anywhere, it is quiet, then i stop the play, click back the part where the drone starts playing, and play it again and all of an sudden it's loud and clear as day. It also tends to be ever so slightly different several times if i hit play, stop, play, stop, on the same part where the drone plays.. And i have no idea why

Don't even get me started on other problems with Serum, even though it is my favourite sound design plugin, it is really hard on CPU. And on weird parts also for example if i have several growls with heavy modulations, they don't mind that much but if i have two serum saw melody sound designs playing at once, both in the same mixer channel and both have both oscillators activated and all with 16 unisons which totals to 64 saw waves playing at once, CPU goes nuts and the "live play" lags really hard.

[timaeus222]

The CPU issues are expected. Serum is just not extremely efficient in that regard.

Did you check your Windows Sounds settings? I had to turn off "loudness equalization" to avoid inconsistencies with sound playback volume.

[ShadowRaz]

http://postimg.org/image/6ta3ca26n/

Do you mean uncheck that option?

[timaeus222]

That might be it, but I recall it being called exactly "Loudness Equalization". Maybe you have a different version of windows, but this is what I'm thinking of, which is in the same spot on Windows I believe (this is not from my computer):

[ShadowRaz]

Yeah it was the lowest option also on my windows, did not see it at first because did not scroll far enough. the equalizer was just the first i saw. Also it was already unchecked for me. Got to look into this. But thanks anyway.

[timaeus222]

Maybe you can try the following comparison:

• Using SilverSpike's TapeIt to record what you hear in FL Studio at full speed (no lag is recorded)
• Using FL Studio to render a specific segment where this problem seems to occur.

Then, see if there is a difference in these waveforms. They tend to not be different (other than due to synth phase differences), but it could happen.

[ShadowRaz]

6 hours ago, timaeus222 said:

Using FL Studio to render a specific segment where this problem seems to occur.

Yes, actually that is precisely what i meant by "freezing" the selected mixer channel in the first message. It is quite easy to do actually and it takes away the load from CPU quite effectively also.

2 problems solved with one solution, although with some extra time consumption on the render parts. Especially if there needs to be changes done after the render, then it's like back and forward sort of thing. But producing full tracks requires a lot of back and forward stuff to get some adequate results anyway.. usually..

[timaeus222]

I'm simply wondering if you have a difference between a WAV file constructed from recording real-time playback vs. a WAV file constructed from rendering. There isn't for me, but is there for you?.

[ShadowRaz]

Just took a screen capture of the problem what i basically mean. I don't know how much can you hear a difference with your audio equipment but through my DT880 headphones, the difference of every play, stop, play, stop, play, stop, is huge.

http://www.dailymotion.com/video/x45f4t8

Then i render the part and it remains the same of course after that. Should try rendering the same part multiple times and see is there a difference

There seems to be some amount of changes in the "final render" a.k.a. exporting the entire track with all in it into final audio file which truly makes me ponder of the reasons

Gonna try to render multiple times now

[ShadowRaz]

http://www.dailymotion.com/video/x45f5x0

Yep they change. Even the bloody wave forms are different in each and sound different. I had ASIO4ALL on at this video so no sound. Why bother capturing a new when you can hear the difference i mean in the first video when i hit play and stop repeatedly.

 

[timaeus222]

Have you had this "problem" since you bought Serum? It seems like a natural phase difference to me. If in Serum you have a running phase (i.e. it doesn't reset upon playing a new note), then each time you render the same segment, it's going to sound slightly different because a different phase offset is being rendered and so a different output is heard based on a different phase cancellation result between the sub oscillator, OSC1, and OSC2.

I do hear what you think is a problem, as I have the same headphones, but it seems like a natural result of using a synth with a running phase. Did you try unlinking the sub oscillator in Serum from the internal processing (i.e. a raw/unfiltered sub oscillator output layered with [OSC1 and OSC2 running through the filter])? I'm guessing the sub oscillator got high passed by a slow running-phase LFO? See if that's the issue by muting OSC1 and OSC2 and listening to the processed vs. raw sub oscillator.

I've never had this issue before with my copy of Serum in FL 12, so it might just be a sound design issue.

[ShadowRaz]

Actually i don't recall right now did i have the problem earlier, if it was there, i did not notice it. But that phase shift seems like an reasonable explanation although not entirely for example comparing the sound quality in rendered audio file vs. raw

Well it is not like a life and death type of thing, so if it is natural and can't be fixed, i'll just have to deal with it, but still it bothers me as i am trying to make music that people would like also and maybe someday even support so i could fund my productions as in buy stuff.. not do it just for my own fun, and for that, there are some standards people most likely expect in music (as tastes also are different etc) for example not many have liked my stuff so far and for which reason i have to consider even the small details and it is annoying when render does not match with the sound design and modulations and the over-all workflow in the project.

I turned off the sub oscillator for this video, slightly harder to notice the difference on this

especially between the two render files but the sub osc does not affect still much apparently..  but it still maybe does something. it is better to make separate bass line anyway for growls because it's easier to adjust. But on this the render files are way different than the live play even though not that different from each other so it is not necessarily the sound designs fault but rather some settings on the render for example? i don't know, but still something causes the difference

Also by the way, as you have the same headphones i have to ask: What are your thoughts on the quality of the headphones? I like them a lot but only bass, in my opinion, could be stronger. But otherwise, they are bees knees

[timaeus222]

Yeah, this recent comparison sounds hardly any different. I do SEE the difference in the waveform, but as before, I can chalk it up to a phase offset difference.

At this point, based on how without the sub oscillator, the difference is even smaller, if I wanted to be even more sure, I would turn off Fruity Limiter and see if that then fixes the problem. If that does, then I would believe that there is a change in amplitude in the sub oscillator that triggers the limiter more heavily, causing a greater duck in volume via overcompression. If that doesn't, then that's an interesting issue you got there.

I never really continued using Fruity Limiter, and switched to a soft-knee limiter about 3 years ago because of the overcompression.

-----

I actually was aware of the high impedance (a kind of "collective resistance") of my Beyerdynamic DT880s (250 ohms), which led me to realize that I needed to amplify them to get a reasonable volume and frequency response to them. Afterwards, I found that they have good bass relative to my Grado SR-60i's that I owned previously, and they have certainly improved my bass mixing drastically (for instance, something full in bass that would normally take me 8 hours on my previous pair of headphones can take me less than an hour to mix on these).

-----

This kind of phase difference can bother people if it's in an exposed context, but if you fill in the soundscape with more elements that encompass the stereo field, the phase difference becomes a minor issue in the big picture, because it sinks beneath the more prominent elements in the soundscape. For instance, I used Serum in writing this remix a few months ago (submitted to OCR already), and I really don't hear any issues with the Serum bass because the soundscape is filled out as much as I could make it.

I realize that that explanation of "minor issue in the big picture" is kind of "covering up" what is labeled as an issue, but in my honest opinion, and with all due respect, I don't think people will really care about an issue of this small magnitude. This is something that you honed in on because you know how your mix is supposed to sound, and I think it sounds worse to you because of that. A lot of sound design is quite personal because over time, you get used to how you prefer to move the synth knobs to hear what you want.

So, just because you really love how a specific sound that you made turned out doesn't mean that in context, it will make for a final mix that others will find amazing. Similarly, just because you don't like how the sound design seems a little inconsistent on live playback doesn't mean people will suddenly find that to be the major issue that they arbitrarily choose to be the reason they don't like the song.

More likely, I think it would be the collective observation that they make of the entire soundscape (the vertical layering), coupled with the length of the song that they choose to hear (the horizontal progression), and the overall arrangement excellence (soundscape + note interplay), that might tell you how much they like or don't like the song.

[ShadowRaz]

Lower notes on the more recent youtube clip also affects the amount it changes compared to the example i showed on the first clip from 'Peace part.1'  for some reason, well it is crunchier and has less mid/high frequencies and slightly different effects and vocodex settings. Tried to turn limiter off, on that second one but it did not change anything except taking away the kick snare compression, Also there was no limiter in the first clip at all, tried to take the sub osc off also in the same part at peace pt.1 and seems like it did not make a difference for the phase shift. Also took off all the effect plugins at master and vocodex on the growl mixer channel (i use it for making the growls louder and more metallic & crunchy, vocodex has certain effect on sounds) as well as Parametric EQ 2, still the same "problem" presents itself.

And yes, mostly other sounds as soundscape make the small differences near damn impossible to spot but the thing is, how can someone else who didn't make the project hear the difference as they hear the one and only version of it anyways.

So i assume that it is caused by phase shift and nothing i can do about it except render the sounds as many times until i find the best render and then use it in the final export.

 

And yes i just have to make new songs how i like them and can do them, and if people start liking them more, then it is an good sign that i have made progress on the over-all arrangement on EVERYTHING. Composing and producing music has countless amounts of things to consider. That kind of small growl problem is nothing compared to the necessity to be extra creative on the entire build of tracks while producing and also having patience to try to make cool chords and single note leads and having good selection of virtual instrument libraries and sound design plugins and hardware with efficient PC and good headphones / speakers

Also some amount of freaking promoting should be done, as people can't basically find the track to listen anyway (whether they like it or not) if they don't even know it exists. I have worked for years on music but have no idea how to promote myself properly. Other than adding some suitable hashtags to Twitter and adding my tracks into both Youtube and Soundcloud with appropriate search tags. And now as a new addition, into Bandcamp from which i can also get some support if someone wants to support my music productions at some point when and if they are that good. Also i am extremely unstable with my emotions sometimes, and i truly suck at being social and tend to write some things that aren't "normal" so to speak, which can make me kind of an weirdo hence people don't even want to like music, because they don't like me, which would be weird reason.. But it is also weird for people liking music only because they like the artist who made it and is famous etc. I don't know..

---------------------------------------------

I don't have a headphone amplifier for my DT880, and i connect them directly to my motherboards internal audio chipset, but they are the lowest impedance (32Ohm) and do not necessarily require one. Although i have noticed that the sounds sometimes tend to be quiet, and if i turn up the volume too much, distortion comes along so i should also get some amplifier at some point and a separate soundcard also if they were to also boost the bass. slightly. Not sure though.

[timaeus222]

24 minutes ago, ShadowRaz said:

But it is also weird for people liking music only because they like the artist who made it and is famous etc. I don't know..

---------------------------------------------

I don't have a headphone amplifier for my DT880, and i connect them directly to my motherboards internal audio chipset, but they are the lowest impedance (32Ohm) and do not necessarily require one. Although i have noticed that the sounds sometimes tend to be quiet, and if i turn up the volume too much, distortion comes along so i should also get some amplifier at some point and a separate soundcard also if they were to also boost the bass. slightly. Not sure though.

Yep; sometimes people like someone else's music because they like the person. I couldn't imagine finding a youtube video of a live concert without people screaming, i.e. they come to the concert for that famous person. So, it can be hard to predict how people "receive" your music. I try not to think about that and just make what I like to make, and if people like it, then great! I stay motivated to write music because it keeps me happy.

-----

I have some nice reference mixes that tells me how much bass is too boomy (http://ocremix.org/remix/OCR02093), how much treble is too piercing (http://ocremix.org/remix/OCR02177), and what loudness is too loud (https://soundcloud.com/zircon-1/level-bounce). That's what I used to calibrate my headphones, anyways.

[ShadowRaz]

On 4/19/2016 at 10:17 AM, timaeus222 said:

Yep; sometimes people like someone else's music because they like the person. I couldn't imagine finding a youtube video of a live concert without people screaming, i.e. they come to the concert for that famous person. So, it can be hard to predict how people "receive" your music. I try not to think about that and just make what I like to make, and if people like it, then great! I stay motivated to write music because it keeps me happy.

-----

Well true.. Can't argue with one sort of logic in it. For example, to go into slightly lesser thingy that i would understand even more than basic celebrity adoring, IF there is no real reason for it other than superficial (he/she is good looking for example) stuff:

If a person was in love ( what is love T.T ) with someone, they most likely will love basically everything the person does no matter of it's real "value". Pretty much how some gifts etc should work for example, rather than expecting something needed or wanted necessarily, one should respect the other for giving the thing in the first place, BUT not saying that someone who is giving the gift still should not care what to get to the other as in of course take in account what the person for whom it is for likes or especially not "buy the love of the person".. well that is rather controversial now when i think about it kind of.. but not exactly. Basically i did explain what i mean. Emotional attraction to the person or fame, makes people like anything the person does more.

But still some amount of REALLY superficial stuff is going on in the world also and a lot of discrimination just because someone looks different or behaves different for that reason, everyone hates the different/odd one and bully them. I will never understand still how shallow or especially rude can some people be also. No one cares about feelings anymore. They care about how everything looks or the benefit they get themselves, am i right? For example let's say a person has hundreds of friends in facebook and everyone likes that persons selfies and the person gets everywhere because of looks and the "fame". Then someone who works hard but is not pretty or very "likable normal" personality does not get far and is taken for granted or used.. I for example don't care that i don't have real friends because i'm not even trying to get friends as an introvert and don't care that much about what others think about me, but still slightly do because of course i would not want to insult people for no reason at all nor would i want to be hated because who would for example want to support me if everyone hated me? But still the things in society (lack of empathy for example as in people treating other people like they were shit) that seem to be really off and causes others to suffer, and if it means some hate me when i state some real facts even if they aren't pleasant and almost complain how much there are things going that aren't nice or whatever, so be it. I haven't got much to lose because i have nothing basically. I would just want other people to also at least wonder slightly how things are getting into corrupted, racist, superficial and highly unequal direction sometimes. Although i think they are slightly better nowadays still in some things compared to past but still there should be a better peace situation among all.

Whoa getting seriously off-topic.. should just stop writing and continue with other stuff. Wanted to make it clear though.

[ShadowRaz]

But to get back into the topic of the topic, as i was working on my remix and had several Serum designs, it most definitely presents itself in all the growls and drones in which i modulate wavetable positions, bends and filter with macros. I literally should just render them until i can find the best render.