TD snare sounds

[Dimitri]

Sometime ago I have heard a sales person say that on the TD-6 module, no matter how soft or hard you hit the pad, the same sample will be played always. Only the velocity will be different. (I thought this applied to all TD modules)

Well, now that I got me my own set (V-Club with TD-6), I have reasons to doubt his words... If I hit the snare softly, I hear one kind of sound. And as I hit it harder, it seems like another sound pops up from the background. On most kits, its kind of a ringing sound. Therefore, I have come to the conclusion that on Roland modules, a snare is always composed of two samples that are being played at together, at different volumes. Sounds like a neat way to do it...but I have never heard anyone mention that before I actually bought the set when I was doing research.

Is this so? Or am I just hearing things?

[clymore]

I think there are several overlapping sounds on the TD's. These sounds dominate or overlap based on your striking force.

Another Kentucky drummer!!!



------------------
TD-8, TD-7, Roland rack, Pintech AX14, ConcertCast and Roland PD-9/7/5 pads, Yamaha and Pintech cymbals, Dingbat, Nimrod, Vertikik & KD-7, FD-7 , DrumKat DK-10, Fender Bassman 60.......2000 Eclipse GS

[puttenvr]

There's only 1 sample but with Roland's COSM technique some variations in sound are made. Normaly you would need multi samples for this.

[dr. kildrum]

Smooth professional quality velocity switched samples with full positional and dynamic sensing across all playing surfaces is the Holy Grail (IMO). Not to be confused with the elusive cabling technique.

[This message has been edited by dr. kildrum (edited November 07, 2001).]

[Dimitri]

There's only 1 sample but with Roland's COSM technique some variations in sound are made. Normaly you would need multi samples for this.

Hmm. Though your point is valid, I find it hard to believe that "Roland COSM technology" can produce two such realistic and distinctly different sounds from just one sample.
I mean, if one thinks of it, what is COSM?... just a bunch of EQ'ing/Compression/Distortion/Resampling and the likes, where needed...

Or anyone, please prove I'm wrong.

[gingerbaker]

Hello dimitri,

Let's see, Putt has 1690 posts, you have 2.

Perhaps the onus is on you to prove your argument is correct?

Especially since you are from,,well...you know...Kentucky?

P.S. - Welcome to the board!

[Dimitri]

Thanks, Gingerbaker!

I've been enjoying this board for a few weeks now and found it very helpful. Especially when it came to picking out a set to buy. By the way, have you ever been to Kentucky?...I didn't think so (Actually, I was born and lived in a BIG city in Russia until i was 15).

And, of course, you may disagree with this... But had there been someone to say that potatoes grow on trees...and had the post said that they had a hunred thousand posts...I would still argue

[Dimitri]

Within the 3 years of my residence in this fine nation, I had time to travel all over central and eastern US, and I must admit that most parts of this country resemble one another like a mirror. 1 or 2 story residential areas...gas stations...wal-marts and krogers... It's everywhere. There is just as much of it in Kentucky as there is in Vermont. Except, perhaps, they clean up their streets in the south a little better

Also, I have had a chance to meet the most intelligent and open-minded people as well as the most ignorant *******s in all parts of US...and Russia.

Sorry for the off-topic.

[szvook]

Originally posted by Dimitri:
Actually, I was born and lived in a BIG city in Russia until i was 15

Hmm... interesting.....



------------------
szvook

[Blades]

When it comes to sample switching, velocity crossfades, COSM modeling, etc, I think Roland would say

"Pay no attention to that man behind the curtain".

The marketing rap says that it's all COSM, some of the sound suggests samples. My thought is that each sound must start with sampels and COSM is used to "mold" those sounds.

But, I think such things as the sound that is produced by the edge of the head vs. the center...that ringing quality that helps the realism...could be done with either method.

But is the actual method important? The sales guy was really wrong either way. Maybe it is the same sample, but it's not the same "sound"...not like an old Alesis DM4 or something (except for a few of the sounds)...that was true machine gunning...try triggering a poorly done GM module, like an old Boss Dr. Synth or the like...that's one sample at differnt volumes and it sounds like poo.

In short...who cares.<g> they sound good.

redbrick
Who's a geek and actually DOES care. www.mp3.com/Sublte_Nuance

[gingerbaker]

Ah Mitya,

I'm afraid I am on a tight schedule this morning and really can't give your two provocative responses the attention they deserve.

However, the pristine innocence of The streets of Vermont must be defended at any cost! Including removing any Blue Grass generated horsesh*t which has been playfully deposited upon them.

While it is interesting that a defence of my spurious (and good-natured) charges of bucolic unsophistication would be mounted from a Russian battlement using a reference to potatos, it really doesn't address burden of proof of your argument.

The explanation of the Roland company for how the V-drums generate sounds is, well, at least as far as I can understand it, something like this.

A sample of, say, a snare is used as a starting point.

COSM is more than just an effects loop series, it is supposed to be a way of manipulating the characteristics of a sample's wave profile using algorithms which also incorporate the dynamic nature of the input signal.

So, ( I think), that as velocity changes, so does the algorithm(s). As position on the drum head changes, so do the algorithm(s) change various aspects of the original samples wave form.

Additionally, with the V-editable sounds, we can go in and tweak the original instrument characteristics, yet the same types of dynamic changes with velocity and position occur.

It must be pretty complex because different samples, tweakings have different dynamic changes.

That is, different snares react differently - perhaps the different samples with their different sonic characters are "interpreted" quite differently by the algorithms?

After all this occurs, one can then add all the different effect chains for even more variation.

So, I think, one sample can indeed generate two very different sounds according to where and how hard the pad is struck. Indeed, this is purported to be the advantage of COSM versus pure sample-layered sound.

( Now, Putt communicated this idea in about 23 words!)

Roland's literature has been talking about this technology for the drums, guitar, voice and keyboards for many years.

As I would posit that the likelyhood that this technology exists is far greater than that of "potatos growing on trees", I still submit that the onus of proof is on you.

I await, open-minded, and hopefully, not perceived as the other option.

Gingerbaker

[Dimitri]

Greetings, Gingerbaker.

Perhaps I have made a mistake by not explicitly specifying that my "potatoes on trees and COSM" analogy was not a direct comparison. I did not mean to equate the two.

I cannot doubt the fact that COSM must have a couple of ingenious ideas behind it (technical as well as *marketing* ideas).

What alerts me is that nobody here or anywhere else seems to really know *what* COSM is, and how it does what it does, and most importantly, WHAT exactly it does to the sound

No doubt that "manipulating the characteristics of a sample's wave profile using algorithms which also incorporate the dynamic nature of the input signal" sounds very cool.

But man, morphing one realistic sample into another realistic sample by means of just algorhythms... Sounds to me like one heck of a tough task.

Now, I can see how this can be done with guitar amp modeling... Here is my general idea of how it's done...

1. They take a simple waveform, and put it through a guitar amp of choice.

2. They mic the amp while slowly, say, turning up gain on it and digitize the miked sound.

3. They look for patterns in how the recored simple waveform changes as the the gain was turned up. Since the initial waveform is primitive, they come up with a formula that would apply these changes to the waveform when it is already digital.

4. They digitize your beloved guitar pickup sound and put that waveform through the *same damn* formulas.

5. Ta-da...you get a digital emulation of a mic'd amp.

Now, the problem I see with doing this for DRUMS...

Now, you CAN run a simple waveform through a guitar amp... but unfortunately you CAN't do that with a drum (it ain't got an input jack!!!)

Also, while you can *gradually* turn up the gain on the amp... and you can *gradually* fiddle with the EQ... You *cannot* *gradually* increase a drum's diameter. Besides, the sound of a drum is constantly changing every millisecond, for drums are very dynamic instruments. Therefore, you'd have to come up with a *different* formula for every little bit of a drum hit's sample. Therefore, it would be EXTREMELY hard to figure out any patterns in the way a digitized waveform of a drum's sound changes.

I hope this makes at least some sense...

Well, if you even got to this point, then hail to you

Dimitri

[This message has been edited by Dimitri (edited November 08, 2001).]

[suuntov]

Great explinations.

It's all perfectly clear now.

[gingerbaker]

Hi Dimitri,

your quote:

"What alerts me is that nobody here or anywhere else seems to really know *what* COSM is, and how it does what it does, and most importantly, WHAT exactly it does to the sound "

Well, I do think you have hit it on the head!

I was basically going from memory from Roland User Group readings. How it works is for computer/math whizzes to explain, certainly not me.

However,

""manipulating the characteristics of a sample's wave profile using algorithms which also incorporate the dynamic nature of the input signal"

does sound very cool!

I am quite proud of that one.

Roland, do you need a freelance copy writer? How about me and Dimitri doing "point-Counterpoint" for the Bullsh*t Files?

[dr. kildrum]