Tassman 5, any words?

Modular Synth design and releases (Reaktor, SynthEdit, Tassman, etc.)
Post Reply New Topic
RELATED
PRODUCTS
Tassman

Post

whyterabbyt wrote: I'm leaning to the conclusion that maybe they're not able to. Its possible their PM algorithms don't 'do' other forms of modelling (I mean, how long have people being asking for conical bore models?)
urm, years, that's why I got Harm Vissers collection for Reaktor - amazing brass models with 2 windbores in there.

I don't want to believe there are no more interesting physical attributes to acoustically model - I'm waiting for cymbal models (circular plates?) and guiro emulators :hihi:

Post

cptgone wrote:
whyterabbyt wrote:
cptgone wrote: they could have added VST export though.
That'd undermine their own market, surely?
it might entice Reaktorheads to get Tassman instead/too.
there's already a crossgrade offer for Reaktorheads

Post

Arwa's new synth should blow anything else completely out of the water with regards to playable brass/woodwind sounds that are realistic in terms of how they compare to real world instruments that exist today - like saxophone, clarinet, trumpet, etc. But the complexity and the warm organic sound of Tassman (and hopefully an upgrade to Tassman 5) will allow anyone to create a sound that is very playable. If only they'd release a real upgrade for Tassman. It's very playable now but if AAS added some more interesting models to Tassman it would be even better.

I think they're more interested in just create "Studios" (ala String Studio, Lounge Lizard, Ultra Analog, and maybe a "Brass or Wind Studio")...

If Arwa not only develops a woodwind/brass synth (wind/brass/reed) but also a string synth (finger, pick, hammer, bow) he would really get some attention. A woodwind and brass synth is amazing in and of itself, but a string synth from Arwa would be phenomenal. Imaging a resynthesized piano, acoustic guitar, harpsichord, harp, all the bowed instruments...it would be amazing.

If he can do it for wind instruments, he might be able do it for bowed instruments too. And probably finger, pluck, hammer instruments as well.

Imagine the savings in RAM and probably CPU too, because the resynthesis is supposedly very CPU efficient. I remember Arwa saying that you could run quite a few instances of his synth on a decent PC.
Little Black Dog - 2008-Present

Post

CinningBao wrote:urm, years, that's why I got Harm Vissers collection for Reaktor - amazing brass models with 2 windbores in there.
'zactly.
I don't want to believe there are no more interesting physical attributes to acoustically model - I'm waiting for cymbal models (circular plates?) and guiro emulators :hihi:
Oh, there are interesting mels; its just that there are several ways of doing PM, and a lot are under patent. Whichever way AAS are doing it, its possible that they can't apply it to those kinds of models.

Another thing one would have hoped for by now is 'dynamics' on the component parameters (eg membrane X and Y size) but that hasnt happened either; again, could be a limitation of their method of PM, because some systems can manage that sort of thing.
my other modular synth is a bugbrand

Post

You can patent a physical model????? How on earth could this be allowed? Someone correct me if I am wrong, but mathematically speaking it's a function that's being patented. This is absurd.

Post

Yamaha has many patents on physical modeling. Here are two of them:

The following one it a physical model of a brass instrument, also including a design for a mouthpiece to operate it. This never appeared as a product:

http://patft.uspto.gov/netacgi/nph-Pars ... /5,543,580

The following one is an enhancemenet to the basic cylindrical bore model to simulate a conical bore:

http://patft.uspto.gov/netacgi/nph-Pars ... /5,438,156

Yamaha has at least 20 more, from the fundamental to the specialized.

Unfortunately they seem to have dropped development years ago with the VL1, and a limited version is still available in the VL70m. However, their patents are discouraging progress by others. The VL1 included the conical bore model, but that never made it into the VL70m.

Post

Thanks for the links. I see the point in having a patent over a specific technology. By that I mean the 'implementation' process of a certain theoritical idea.
I never imagined I'd see the day when a mathematical formula being patented.
So, if I understand what you are saying, no one can use the conical bore model, in a PM VSTi say, without Yamaha's permission, right?

Post

Yamaha's physical modeling patents are all related to digital waveguide based synthesis, which was invented by Julius Smith at Stanford, CCRMA music researcn group. CCRMA licensed it to Yamaha, and Yamaha carried on the research resulting in the other patents. (Previously, CCRMA also had invented FM synthesis and licensed it to Yamaha)

Therefore Yamaha's patents are only blocking work on waveguide synthesis. There are other methods of physical modeling. Unfortunately Waveguide synthesis is by far the most computationally efficient. Arturia Brass uses a different method, based on numerical solution of differential equations. I think it includes a conical bore model, but it uses a huge amount of computer power compared to Yamaha's method. Arturia's need to limit the model's detail to improve efficiency may be one reason the sound is poor. Actually the differential equation method should produce the best sound, if one had a computer of unlimited power.

Actually, as it turns out, a conical bore model isn't needed to produce good sounding physical modeled brass. The VL1's brass sounds are very good, but all those patches as far as I've seen were developed before the conical bore model was added to the VL1, and never updated to use the conical bore capability. The VL1 includes a resonator model and other modules that do a pretty good job of making the conical bore model sound like a brass instrument. Many aspects of the VL1 in addition to waveguide modeling are covered in patents.

It would be great if Yamaha would create a softsynth version of the VL1 (preferably polyphonic). I don't think it would be that hard to port the code, and todays computers are much more powerful than the processor in the VL1 synth.

Post

whyterabbyt wrote:
pdxindy wrote: AAS had to go the direction they did because Tassman is too complex, uses too much CPU and requires more learning time to make good use of. I doubt the company could live on a single complex modular synth.
Unfortunately, given that there were minimal changes to their PM technology between T2, T3 and T4, I'd suggest that the real reason they had to go in the direction they did (spin off subsets of Tassman into fixed-architecture synths) is because they'd reached the limits of their ability to extend Tassman in any meaningful way.
I would be very happy if Tassman 5 had no new models and just made the existing ones easier to work with.

String Studio clearly surpasses Tassman for string sounds. It is more playable, expressive and easier on the CPU and has a more sophisticated model.
whyterabbyt wrote: Another thing one would have hoped for by now is 'dynamics' on the component parameters (eg membrane X and Y size) but that hasnt happened either; again, could be a limitation of their method of PM, because some systems can manage that sort of thing.
There are modulatable parameters in String Studio that are fixed in Tassman. Inharmonicity and pluck position for example.

I would say the issue is more one of CPU use. It is only now that cpu's are fast enough to make use of Tassman as it is and still have a bit of headroom. Tassman is inefficient because it is modular.

Since String Studio is more sophisticated than Tassman, I think your conclusion that AAS is in a developmental deadend may not be right.

Post

Is something happining here? Does anybody know something new about an update about Tassman 5 ?
rabbit in a hole

Post

Last thing they told me was they were revamping the String Studio (VS1) next.
They really should have gone with doing Tassman...

Post

I heard that Tassman 5 is going to get 2 new fx, a revolutionary tabbed interface, and 150 patches designed by that one guy that does their videos with a supremely powerful eyeglass prescription.

-Sam

Post

Thiago Pinheiro

Post

nothing is happening. there are a couple posts on their forum that basically say "we don't know when.. it'll take a long time. sorry"

Post

just a guess that tassman 5 will be released April 2014

(10th anniversary since the last version) :hihi:

you think AAS have had enough time?

Post Reply

Return to “Modular Synthesis”