VSTFX: La Ride

[lalalandsynth]

The main issue is that it will not sound the same when rendering faster then realtime.
Second issue is that the ticker can not be relied upon , so the sound might change according to the load on your cpu or project.

[wlangfor@uoguelph.ca]

So, as I had mentioned earlier in the post I'd realized that there was an issue in applying a global redraw event that is in the frequency of 25 - 100 etc etc, but I'd realized that a tick 5 is just right.

furthermore, By setting a wireless out to a global wireless receiver named collected I was able to provide a way to turn off this feature. But therein lies a probable dilemma. And that is redrawing using the redraw area.

In FL studio 12 or 20 it's a realization that there becomes a redraw issue, images and knobs etc become distorted due to the fact that DAW's have changed the way in which they redraw. I guess it really came down to that maybe the old programs forced redrwing at old times to compensate, and I suppose the new programs just expect that to be on a vst to vst basis.

I'm not sure if I agree with the rationale but regardless of that, this method could be hindered if using instead a redraw area. Who's to say that if there's an unaffected area that it might "break" due to the same redraw issue. And I have witnessed that too, it's a serious issue for plugin developers.

But in regards to nyquist, I guess My only reply is that frankly people forget that trigonometric functions are frankly some of the most math intensive. It's why i have always said that it's a girdle for distortion, and it's the very reason why I have often pointed out that division is not as fast as even multiplication of a float.

So, I figure that this math is unable to correctly be factored (at least immediately) for 44,100 samples a second much less 48,000. So, in consideration of this realization and latency that is often forced by compressors. I guess I am frankly unworried about high latency application.

I used to until I tried applying some, and I find the result pleasing. I frankly doubt that the waves vocal rider is running any faster because I have used it on a regular basis. It's always dawned on me that using envelopes seems to be a slow thing.

So, I mean in the way of speed I've always considered the larger game and the finer points of CPU usage. I guess it becomes pretty un-nerving when people see that My new plugs are using only 0.5% CPU, and I still use trigonometry, envelopes too; Wholly realizing that there is a hindrance in the way of latency.

So, in regards to the changes I wish to make to this plug, I'd really like to get a system down for preventing numbers moving around when they change. It's very distracting. This is why I made sure that in the image You can see, on the left there is a bunch of zeroes. I used these to prevent changing and moving numbering. So, on the right I've been using a new system. I'm separating the negative (and positive) symbols from the number and furthermore making sure that (in the case of 2 digit numbers) that there is only one preceding zero.

So, that seems to solve the issue in regarding to numbers jumping around.

The other issue I've had is that I set the float(0-1) output to an absolute. because it is absolute it denies the bipolar negative values and I can just add a conditional check. So, the thing is I wanted an LED light image to start to light as soon as the knob is moved. And I'd realized that it's possible with a nice idea like barak had for taking a rational mapper to the next level. It means that the image can switch (by means of opacity) at an accelerated rate as soon as it gets into the low decomals (0.1 and on) which is perfect because then I can make sure that the LED does'nt look grayed out,

And I think these finer details are what make a plug really pop, obviously to add mouse support too does as well. So I'll be making some tuts to cover that hosted on My new forum. It's good for Pros to figure out all of these little details.

So, while I figure these slow methods are a little worrisome, so's an envelope, and division, and trig. But it's nothing new. People even like latency. They love it.

And by all means make a plug, call it the best I won't diss You, lol.

[lalalandsynth]

You are being too defensive in my opinion , the plugin will not render correctly, That is hardly what you are aiming for. People will not love that !? And the plugin will not always sound the same , people will not love that and unless you disagree that it will not sound or render correctly I cannot understand why you are being defensive?
it could also cause the user some headaches as one would never expect it not to work the same when rendering

You can fix this pretty easily as far as I can tell , you can add the floats to the stream like this.

streamin in, contrl;
streamout out;

// '+' Operator
movaps xmm0,contrl;
addps xmm0,in;

// Assignment to 'out'

movaps out,xmm0;

or just use a stream add ? unless I am missing something ?

And then just use a multiply stream for the 0.33 mult.

Same result , just sample accurate, apart from the rest of the circuit. ...

fix2.jpg (151.15 KiB) Viewed 18264 times

[wlangfor@uoguelph.ca]

I've tested the output with Youlean and have not had any issues with phase.

It's not a matter of defensiveness. I'm literally just moving on with this development. I'm plainly trying to make a soft gain rider. Something like from the forties or fifties that simplifies the sound for radio.

It does exactly what i wanted it to do.

And while You thought of all this (thanks BTW, I'll look into this) I think You maybe second guessed what i was attempting.

I like slower sound and it's oldschool, and I just made a great track with it.
OK, so - If You can throw down a track is good with Your own plugin, I'd like to see it. because Here's what i came up with on youtube.

Another product that tried to achieve the same effect was the Thrillseeker VBL by Bootsy (Variety of Sound). And now You know LOL. They used to refer to them as governors and I'm not sure whether or not they used envelopes to the same degree We use them in this day and age.

And LOL, I understand lalaland, I'm a fan of Yours, I just dislike Brickwall limiters. They, instead of averaging and making the sound one dimensional just cut it off. And so do compressors limiters. I don't like to do any cutting at all. I guess that's an old school mentality and it must seem strange to want to put all of the sound in the near threshold.

Well, You might not think so when You're mixing rap and You need to chop your vocals into bits because I'm not the best vocalist and when I don't want My vocal chop tail ends to be discernible, I use a gain rider. It's like normalization. It prevents dips. In exact contrast there is pumping, which is something I use after (in My esquire limiter) to bring out the peaks and differences by using some surgical latency.

I think sound in movement sounds best. Which is why I like pumping Comps, limiters, but they require the vocals not to have dips, thus the LA Ride. That's what a vocal rider is for. But.. I'm not a fan of the fact that waves never has theirs on sale. I will only charge max 10 dollars when I go commercial which should be soon.

Regards Trog, Lalaland just a misunderstanding. I wrote this Rap a long time ago and I'm happy with the result so far.

[check youtube video in latest post]

[trogluddite]

...being a pedantrist

If you meant "pedant" then, yes, guilty as charged!
[Sorry, it's the first commandment of linguistic pedants that any unintentionally ironic mistyping/misspelling of "pedant" must be responded to with a pathetic linguistic pedant joke, on pain of death. ]

If it sounds right, it is right!

If you like what you hear, and you enjoy making the stuff, that all cool by me. And I certainly find your technical writing interesting; though whether in the sense you intend, I'm a little less sure (it's post-modern deconstructionist free-form poetry, yes? Am I close?)

And please, feel free not to take me too seriously, my sense of humour is warped to the point that only I recognise it as such most of the time (maybe folks should read my posts in the same silly voice that I do?) And my advice is sometimes opinionated and a little bluntly put, especially when the spammers make me grumpy - but I mean well.

Peace!

[lalalandsynth]

[lalalandsynth]

I am not discussing the validity of the approach , i.e using a leveler rather then a limiter, that is fine.

I dont know if you are just ignoring what I am saying but I will say it again, when you do a faster then realtime render in Reaper for example, which I do all the time , the plugin will not behave in the same way , i.e the rendered file will not level out in the same way as what you hear when playing.

I find it hard to believe that you find that acceptable ?

[wlangfor@uoguelph.ca]

I am not discussing the validity of the approach , i.e using a leveler rather then a limiter, that is fine.

I dont know if you are just ignoring what I am saying but I will say it again, when you do a faster then realtime render in Reaper for example, which I do all the time , the plugin will not behave in the same way , i.e the rendered file will not level out in the same way as what you hear when playing.

I find it hard to believe that you find that acceptable ?

well, so far it sounds good. And as far as I know Youtube uses a similar algorithm. Like this product slowing down peaking, theirs does too.

But, I mean a better tick, sure. Something that tests for db In a faster way, yes I could use one if You have one, We all could. but, it Works so far lol.

[wlangfor@uoguelph.ca]

LOL, now You have Me second guessing Myself. You see, being that tick is always used for crossfades what have You, more than just things that are visual, I'd merely assumed that it would speed up. But now... How do I test if that's true?

If it sounds good... what does that mean..... ?

I think there's a blue linear db someone made with dsp code I could use I guess. Maybe it'd require asm.

[trogluddite]

I'd merely assumed that it would speed up. But now... How do I test if that's true?

This was worked out many years ago, in the days of the ancient SM guru's who are no longer part of our little club. The tickers have been part of SynthMaker/FlowStone right from the beginning. In fact, in a way, they're even older than that...

The limitations of tickers arise because they are actually a (very old) component of Windows. All FlowStone does is to request the time interval and set them running, and Windows then calls back to FlowStone when each tick is due. Unfortunately for us, Windows timers have a fundamental limitation; 10ms between ticks is the shortest interval that is allowed. This is why Tick100 is the fastest one that we have. Even the general purpose Timer primitive works the same way; so duration values less than 10ms will be rounded up. There is nothing that FlowStone can do to make them go any faster, as Windows simply won't let them.

Windows timers also have a relatively low CPU priority, so the duration of a tick can be longer than requested if the CPU is very busy with other things. So when you're running a busy mix, a Tick100 might run substantially slower than 100Hz and with lots of jitter.

To confirm this, what you could do is set up a simple synth, then have a Ticker connected to a MIDI Note primitive to produce a very short pulse from the synth each time the Ticker ticks. Export this as a VST, and plonk it into your sequencer. Then, assuming your DAW allows it, do both a real-time render and an offline (faster than real-time) render, and compare the two. You should see that the pulses are further apart in the offline render than in the real-time one.

I think there's a blue linear db someone made with dsp code I could use I guess. Maybe it'd require asm.

I'm pretty sure that someone did make an ASM optimised version of this years ago - I'll have a rummage when I get back to my Windows PC to see if I can find it.