KVR Audio

duncanparsons · Post by **duncanparsons** » Sat Sep 25, 2004 10:52 am

..I think this thread itself is getting aliased..

There's certainly alot of pollution and crosstalk going on..

DSP

arguru · Post by **arguru** » Sat Sep 25, 2004 12:00 pm

I agree this thread is just scratching inharmonic spectra. Center the discussion on the topic, and if somebody doesnt agree someone else arguments, our mothers dont have nothing to do here (spoonboiler...).

I have to give an opinion, (i should open thread for this) that this was not the first time the classic quote of some user posting:

'Since you dont agree my opinions, I saw you saying something I dont like or you dont want to kiss my ass everytime I wished, you suck, your company XXX suck, and I will not buy your crappy products)'

This does a lot of hurt, yes. Software developers have against them that there's warez, stolen cards, refunds, easy ways to stole products, we're getting f**ked everyday, and of course, the crappy minded ppl who thinks 'hah, I'll add more f**k to those poor guys posting to cause some free hurt'.

I saw not only against discoDSP , also in some others companies/cases.

NOTE: Before go a public forum and feel the 'force' that make you post something like that, think about it. If you dont want buy x product, just STFU and think like a smart person. Just limit to try the products demos and if u like it and want to buy it, go ahead, remember after all that if the developers of X company of the plugin(s) that you like, think you're an asshole (and i guarantee you 100% compnies here had the classic 'user' doing "dev-f**king"), or you think they are assholes, doesnt mean their software share that opinion. Software dont think by itself yet.

Now why not center the discussion on the topic? or better: why dont you just close your browser and write music?

PD: I kindly will receive any insults for my english! =)

nuffink · Post by **nuffink** » Sat Sep 25, 2004 12:16 pm

arguru wrote:I agree this thread is just scratching inharmonic spectra. Center the discussion on the topic, and if somebody doesnt agree someone else arguments, our mothers dont have nothing to do here (spoonboiler...).

I have to give an opinion, (i should open thread for this) that this was not the first time the classic quote of some user posting:

'Since you dont agree my opinions, I saw you saying something I dont like or you dont want to kiss my ass everytime I wished, you suck, your company XXX suck, and I will not buy your crappy products)'

This does a lot of hurt, yes. Software developers have against them that there's warez, stolen cards, refunds, easy ways to stole products, we're getting f**ked everyday, and of course, the crappy minded ppl who thinks 'hah, I'll add more f**k to those poor guys posting to cause some free hurt'.

I saw not only against discoDSP , also in some others companies/cases.

NOTE: Before go a public forum and feel the 'force' that make you post something like that, think about it. If you dont want buy x product, just STFU and think like a smart person. Just limit to try the products demos and if u like it and want to buy it, go ahead, remember after all that if the developers of X company of the plugin(s) that you like, think you're an asshole (and i guarantee you 100% compnies here had the classic 'user' doing "dev-f**king"), or you think they are assholes, doesnt mean their software share that opinion. Software dont think by itself yet.

Now why not center the discussion on the topic? or better: why dont you just close your browser and write music?

PD: I kindly will receive any insults for my english! =)

I must have been reading a different thread to you. With the exception of the last few pages, this has mainly consisted of DEVS ripping lumps off OTHER DEVS. Yet you choose to blame the users.
It has also included a masterclass in the arcana of dsp programming by some of the most talented developers around.
All in all a classic K-v-R thread. Just the kind of thing that keeps me coming back.

arguru · Post by **arguru** » Sat Sep 25, 2004 12:33 pm

I must have been reading a different thread to you. With the exception of the last few pages, this has mainly consisted of DEVS ripping lumps off OTHER DEVS.

With my last post , I was not referring to that. Devs arguing with Devs is fine, I agree some used some hot ways, but i didnt saw insults as spoonboiler did.

Yet you choose to blame the users.

Some users deserve it.

TheWall · Post by **TheWall** » Sat Sep 25, 2004 12:44 pm

arguru wrote: Some users deserve it.

Some devs deserve it too.

I'm sure some of us still remember who said "we don't give it a damn if you guys buy our products or not"(something like that, not the exact quote) then came back two weeks later pleading for forgiveness due to no sales at all.

Ugh. This is not my business so I'm outta here... I know whom I won't do business with though.

Kriminal · Post by **Kriminal** » Sat Sep 25, 2004 12:46 pm

Dunno about the rest of ya, but i have no idea what 'aliasing' is, or care, i just try to write tunes.....

arguru · Post by **arguru** » Sat Sep 25, 2004 12:58 pm

I'm sure some of us still remember who said "we don't give it a damn if you guys buy our products or not"(something like that, not the exact quote)

100% Right, I said that and I didnt changed opinion. My goal is write software, and it must be sold by itself, and not because if I like to ppl or not.
That's why demos exist.

then came back two weeks later pleading for forgiveness due to no sales at all.

100% Wrong as far as i know, I dont remember -me- doing that.

TheWall · Post by **TheWall** » Sat Sep 25, 2004 1:03 pm

arguru wrote: 100% Wrong as far as i know, I dont remember -me- doing that.

It wasn't you, but your associate did.

george · Post by **george** » Sat Sep 25, 2004 1:06 pm

TheWall wrote:It wasn't you, but your associate did.

I don't remember neither.

TheWall · Post by **TheWall** » Sat Sep 25, 2004 1:13 pm

george wrote:
TheWall wrote:It wasn't you, but your associate did.
I don't remember neither.

Hehe. Whatever you say. I'm sure I'm not the only one who remember the whole incident, as that was one hell a big fiasco at the time.

I'm really out of this thread now.

stefancrs · Post by **stefancrs** » Sat Sep 25, 2004 1:40 pm

Ok, I don't care that you guys keep slipping off-topic. I will now try to explain my suggested method a bit better.

We have a sample, I'll refer to it as an array called S.
We integrate this sample into Si

Code: Select all

loop-i-over-sample-length:
Si[i] = Si[i-1] + S[i];
end-of-loop

Now, if we want to play back S, we instead use Si like this

Code: Select all

out[i] = (Si[i] - Si[last_i]) / (i - last_i)

And hence we have now played back the average output from last sample position to current sample position. This should (atleast in my head) be equal to downsampling from "infinite sample rate" (due to good calculation of Si and interpolation when reading from it) using box filtering. Box filtering is not optimal, since it does not remove everything above nyquist, but it heavily reduces it. By trial and error one might be able to get very good results using some oversampling and decent interpolation as well.

What do you guys think about this? Anyone tried it? (If the sample does not have a dc offset the value of Si won't get out of hand.)

autloc · Post by **autloc** » Sat Sep 25, 2004 1:46 pm

So you're saying that samples should be lowpass filtered before processing (and then high-pass filtered for playback)?

stefancrs · Post by **stefancrs** » Sat Sep 25, 2004 1:49 pm

I'd hardly call integration "lowpass filtering", nor would I call "taking the average over time" for high-pass, but lowpass. Averaging filter. Box filter.

I'm saying that playing back the average of what has happened since last output sample might be a good idea, but I don't now since I've only used this method in synthesis.

arguru · Post by **arguru** » Sat Sep 25, 2004 2:14 pm

loop-i-over-sample-length:
Si = Si[i-1] + S;
end-of-loop

Dont you mean : Si=S-S[i-1]; ?

and if i'm not wrong ... in that method, having step incrementations higher than 1.0 may result in unstable DC even if the sample has no DC bias.

Anyway, i may be missing stuff, did u tried it already stef?

stefancrs · Post by **stefancrs** » Sat Sep 25, 2004 2:21 pm

arguru wrote:
loop-i-over-sample-length:
Si = Si[i-1] + S;
end-of-loop

Dont you mean : Si=S-S[i-1]; ?

and if i'm not wrong ... in that method, having step incrementations higher than 1.0 may result in unstable DC even if the sample have no DC bias.

Erm, no I do not mean that. I want to calculate the integral of the sample, not the derivative

And the method is only of interest when having steps over 1.0 But why would it cause unstable DC bias? I have only used this method on calculated waveforms, but it works really good for those (but you don't have to precalc the integral there etc). All I do is divide the sum of the sample from last position to the current position with the time between these positions.

The aliasing thread