| Author
|
FAT
|
Speakafreaka
IsraTrance Junior Member
Started Topics :
18
Posts :
779
Posted : Mar 9, 2011 02:53
|
Aye, it is subtle.
Here is some photo evidence for you.
You may not think it looks like much to worry about ... and on one part you'd be right, but it really really adds up, and I find it definitely noticable, unpleasant and unwanted, especially when K & B are in isolation!
The cursor is over the problem area. The original sample goes straight in at the transient with silence before hand.
  .
http://www.soundcloud.com/speakafreaka |
|
|
|
willsanquil
IsraTrance Full Member
Started Topics :
93
Posts :
2822
Posted : Mar 9, 2011 02:55
|
This thread is great.
Anyone who argues with Speakafreaka about filters obviously doesn't know how quickly they are digging themseles into a hole they can't get out of.
Taking notes.
PS speaka your efforts are much appreciated.
If you want to make an apple pie from scratch...you must first invent the universe
www.soundcloud.com/tasp
www.soundcloud.com/kinematic-records |
|
|
|
daark
IsraTrance Full Member
Started Topics :
58
Posts :
1397
Posted : Mar 9, 2011 03:03
|
OK LOL 
Now exactly a second before you posted i found it
It is fucking small i have to say
Big thanks for the info bro'
Discovered it on the Q-pro first it seems bigger there but i can be wrong i didn't check that much.
Now i tried the same deal with monofilter and what i found is that on Waves it is -55 db and monofilter is barely -70 db.
http://soundcloud.com/magimix-1/chilling-forest-whispers
Wierd shit happens :) |
|
|
|
Speakafreaka
IsraTrance Junior Member
Started Topics :
18
Posts :
779
Posted : Mar 9, 2011 03:05
|
Lol, yes, my DSP ability is only exceeded by my ability to come across as a cunt despite my best efforts.
I don't really think that internet forums are a good medium for me, although I'll persist.
You can't see the wry grin in my eye. Deadpan just doesn't work when its typed. 
.
http://www.soundcloud.com/speakafreaka |
|
|
|
PoM
IsraTrance Full Member
Started Topics :
162
Posts :
8087
Posted : Mar 9, 2011 03:09
|
|
dudes if i argue it s cause to me speaka is wrong thats all ,to me there is room for linear phase and phase filter.using best of both worlds at what they are good at should lead to the best results to me. i got some times better result with using some bands in lp mode insteed of phase for example.i know about pre ringing... but i just trust my ears. |
|
|
|
Speakafreaka
IsraTrance Junior Member
Started Topics :
18
Posts :
779
Posted : Mar 9, 2011 03:09
|
Aye, small it is ... but it adds up, I promise. There have been situations where I've ditched Q's because of it although I liked the tone I was getting, as ever let your ears guide you!
Of course the other issue is the sequencer delay it produces which when you've got a few linear phase EQ's racked up really starts to grate severely to me at least.
I'd just say, if you are using linear phase EQ on the kick and bass, be really careful about listening way back into the mix, because it can come back and bite you in the ass!
.
http://www.soundcloud.com/speakafreaka |
|
|
|
monno
Grapes Of Wrath
Started Topics :
9
Posts :
454
Posted : Mar 9, 2011 03:09
|
@ Speakafreaka....Well pretty much anyone can come in here quote heavy technical material and come off a winner can´t they? At least that seems to be your claim so far. Back it up dude or back off on the all-knowing rants. Helping eachother out is the main aim here, not condescending people who might not be mighty designers of filters! Your argument might be valid but try to use truly well designed filters in the digital domain and repost your listening tests. (you would probably like algorithmix or weiss a bit more, but that is very expensive code! In almost every case Eq´s have been designed for a non realtime environment (your precious pc has to devote so much of it´s time on non dsp functions) making it inevitable for some compromises to happen regardless of type, make or colour (it seems to be the norm to be able to run 512 crap plugs as compared to recording proper sounds and using heavy stuff sparingly when needed). The beauty in a good digital design must then be to mask and/or overcome those inherent limitations no? The whole point of inventing is to go where no one else had gone, not categorically stating it cannot be done. Anyway i deal with the music bit of things but i have heard some pretty damn good filters regardless of their so called limitations.
Analog still has most digital designs beat except for recallable settings (and i can live with that)
Answer me this btw: Why is is that almost every digital eq i have tried have such a hard time boosting high frequencies? Could it be again that ressources must constantly be wrestled from the system making neat tricks like heavy oversampling too cumbersome for most designers? Straight answers without too much smugness would be appreciated.
In the end i don´t care for perceived limitations, but only how to get past them and to that end, if i like the sound i go for that.
Mastering available here:
http://www.bimmelim-soundlabs.com
http://soundcloud.com/onkeldunkel
http://www.myspace.com/onkeldunkelownz
http://www.parvati-records.com |
|
|
|
PoM
IsraTrance Full Member
Started Topics :
162
Posts :
8087
Posted : Mar 9, 2011 03:19
|
Quote:
|
On 2011-03-09 03:09, monno wrote:
you would probably like algorithmix or weiss a bit more, but that is very expensive code!
|
|
look like he know better than weiss and the guy at algorithmix ,speaka you should make a company you re gonna get rich man. |
|
|
|
daark
IsraTrance Full Member
Started Topics :
58
Posts :
1397
Posted : Mar 9, 2011 03:31
|
|
Speakafreaka
IsraTrance Junior Member
Started Topics :
18
Posts :
779
Posted : Mar 9, 2011 03:32
|
OK. I'll try to answer here.
I think I provided proof above - if you need more, let me know, and I'll do what I can within reason.
On the real time thing. No PC be it OSX or Windows is actually realtime because they are based on Von Neumann architecture, This is based on interrupts, and there is absolutely no garauntee time wise when these will return and allow the system to continue, so hard real time is out.
On a soft real time system ... well it is arguable (not that I would neccesairily do so) that this is the situation of the modern DAW. It can do everything it needs to do quickly enough that it appears to be realtime with cycles left over. But here we run into a big difference between linear and non linear phase EQ. Because linear phase EQ needs a sample buffer for each instance where it is loaded (normally at least a few hundred samples) it is possible to quickly reach a situation where real time playing of keyboards or even knob tweaks have also been pushed back - I hate this passionately. This isn't a sound related thing, but a usability related thing. Personally that drives me nuts. Feedback (non linear phase EQ) doesn't suffer from this, which is another big reason in my head to give it preferential treatment.
Sooooo anyhow, my understanding of Linear phase EQ is that every one that currently exists uses a feedforward mechanism to phase allign the signal, and wouldn't be possible if it didn't do this.
I don't have access to the algorithmix or weiss EQs, but I would strongly suspect that they are following the same feedforward route for their linear EQ as everyone else, just implementing it better. A funny thing is that it is in filter design terms relatively simple to build a linear phase EQ, but really hard to make one that sounds good.
On boosting. I suspect you are approaching the right answer. The issue to my mind is simulating channel distortion accurately which is notoriously hard. In fact, accurately modelling any distortion is notoriously hard, due to precisely the difficulty you hint at which is aliasing.
Very little out there gets anywhere near simulating the effect you might get of boosting the tits off on analog channel. The nearest I get to it in the box is using sound toys decapitor as a straight run through box, and plugging a nice vintage EQ in before hand and absolutely yanking the highshelf boost. This does indeed produce the same sort of soft saturation you'd get from a channel - fizz without destroying the bass or complex harmonies on pads, although I wouldn't say it sounds authentic just reminiscent.
.
http://www.soundcloud.com/speakafreaka |
|
|
|
PoM
IsraTrance Full Member
Started Topics :
162
Posts :
8087
Posted : Mar 9, 2011 03:37
|
Quote:
|
On 2011-03-09 03:32, Speakafreaka wrote:
OK. I'll try to answer here.
Sooooo anyhow, my understanding of Linear phase EQ is that every one that currently exists uses a feedforward mechanism to phase allign the signal, and wouldn't be possible if it didn't do this.
I don't have access to the algorithmix or weiss EQs, but I would strongly suspect that they are following the same feedforward route for their linear EQ as everyone else, just implementing it better. A funny thing is that it is in filter design terms relatively simple to build a linear phase EQ, but really hard to make one that sounds good.
|
|
my memory suck but i doubt it s true, the algo red is one of the most praised eq and dont use feed forward technic as far as i remember.(could be wrong)
pre ringing is not necessery a bad thing,i dont know why you are stuck with this, in some case it could probably make something sound better, who know... like aliasing artifacts for example .
what i know for sure is sometimes i would pick it other phase shift and timing issue.
the plpareq 3 was free when they devoleped it and man ,everyone was loving this plug at kvr forum saying it s one of the best eq ever... it s feedforward design i don t think it s necessecary a bad thing.
there is also some eq like ddmf lp10 that let you go from phase to linear phase to minimise artifacts and get the sound the most suited for the source, i dont know how it s effective... |
|
|
|
Speakafreaka
IsraTrance Junior Member
Started Topics :
18
Posts :
779
Posted : Mar 9, 2011 03:54
|
Reading between the lines on the Red EQ it uses FFT, doesn't induce prering no, and indeed, I'd be interested to try it out. FFT on the other hand does induce latency.
I don't like prering on fundamentally percussive parts like kick drums and bass lines, because it makes them sound shit.
.
http://www.soundcloud.com/speakafreaka |
|
|
|
Nectarios
Martian Arts
Started Topics :
187
Posts :
5292
Posted : Mar 9, 2011 04:27
|
Quote:
|
On 2011-03-09 00:29, Speakafreaka wrote:
Quote:
|
I merely asked if there is a steep filter that does not change the sound.
|
|
Nice. I'm quite certain you know this is retarded. What are you playing at?
|
|
I am not playing at anything. I have lost touch with plug ins as I spend most of my time using the stuff I know works to make music and spend very little time to search for new things. You suggested the fab filter which I am going to try out. 7 years ago I read in my analog electronics books that steep filter slopes cause phase distortion. I have been out of the loop for like 6 years, and I asked if there is something out nowdays, in the digital realm, that does this...yet you think its retarded to ask?
http://soundcloud.com/martianarts |
|
|
|
Speakafreaka
IsraTrance Junior Member
Started Topics :
18
Posts :
779
Posted : Mar 9, 2011 04:32
|
You want a filter that doesn't change the sound.
So far as I can tell, this is an oxymoron, as well I think you know. That is why I suggested your original question was retarded.
I apologise. It was needlessly harsh to frame it like that.
I am however genuinely curious as to the logic behind the thought - you're too good a producer to not know that looking for a filter that doesn't change the sound at all makes no kind of sense!
.
http://www.soundcloud.com/speakafreaka |
|
|
|
PoM
IsraTrance Full Member
Started Topics :
162
Posts :
8087
Posted : Mar 9, 2011 04:53
|
a good linear phase filter ,depending the source can be the most transparent haha sorry to argue speaka .
last time you were arguing digital is just as good as analog man...make me wonder if you are not the best troll around here.i just got trolled again  |
|
|
|
