KVR Audio

YBaCuO · Post by **YBaCuO** » Tue Dec 28, 2004 12:00 am

Thanks Christian for your post and demos.

While clearly big brassy orchestral works is not Synful's current strength, I thought your mock-up was okay and certainly promising. Much better than I expected, and I bet it was fairly fast going.

It will be interesting to see this technology develop.

Since I am guessing that this is resynthesis, so there are real "samples" somewhere, hopefully Synful/Eric will team up with a sample developer to get better recordings, better distribution and be able to spend more time on coding. But who knows ...

Thanks again
YBaCuO

ultrassom · Post by **ultrassom** » Tue Dec 28, 2004 12:20 am

Christian Schüler wrote: The other is a version where I tried myself at physical modeling, using my own developed algorithms. Overall it sounds a little too clean, but I am quite satisfied with the brass articulation. Too bad I cannot make a VSTi from this, because of the patent situation .

http://www.ariescode.com/scratch/physical_robin.mp3

Christian
I really think your PM version sounds even more fat and full than synful's (and Roland's!), and the expression sounds very nice here... Congratulations, and pray tell us more about how you came about doing your own Physical Modelling - was it with reaktor? or what tool? And tell us about the patent issues too... I would love to have access to this kind of technology!!

Since I do not own GPO, VSL or any other of the "big" sample libraries, maybe somebody else could go for a try at this piece. I would definitely be interesed in a GPO rendition of this Robin Hood passage. (No, I don't have a score, you need to guess the notes for yourself, or mail me, and you can get my EXT file.)

Just a question here - you say you don't have the score, but can't you share a midi file??

thanks

Ruben - São Paulo - Brasil[/quote]

Christian Schüler · Post by **Christian Schüler** » Tue Dec 28, 2004 1:40 am

For Roland old version, I lost the midi file (Hell, I owned an Amiga back then ...)

The physical version is kind of "tracked" (internal sequencer), with the score embedded in the source code of the program...

The Synful version was done in EnergyXT, but while EXT can import midi files, it seems it cannot export them

.

The EXT file is here:
http://www.ariescode.com/scratch/robin.ext

(it uses AriesVerb of course

)

Christian Schüler · Post by **Christian Schüler** » Tue Dec 28, 2004 1:59 am

The Stanford and Yamaha patents covering waveguided synthesis are listed here:

http://www.sondiusxg.com/patent.html

So if your PM is not waveguided (like is Tass), you should be ok, but it eats way more CPU that way.

Lunch Money · Post by **Lunch Money** » Tue Dec 28, 2004 2:30 am

I could be wrong here, but it seems as though it's an unfair comparison using the same MIDI file for both. I was under the impression that Synful responded to a different style of playing/programming.

I'm not saying that the sound quality is better or worse, just pointing out that it seems like a pointless way to compare the two things.

Greg

realmarco · Post by **realmarco** » Tue Dec 28, 2004 5:17 am

Christian Schüler wrote:The Stanford and Yamaha patents covering waveguided synthesis are listed here:

http://www.sondiusxg.com/patent.html

So if your PM is not waveguided (like is Tass), you should be ok, but it eats way more CPU that way.

how does Fm fit in all of this?

FM was patented by Yamaha ...so how come we got stuff like fm7 which boast loading sysex from the dx7 and its offspring ?

Yossarian · Post by **Yossarian** » Tue Dec 28, 2004 12:32 pm

A bit of a side note: I noticed that so far, the only other developer I’ve seen commenting on Synful Orchestra is Gary Garritan and you can’t help admiring the guy. Gary has been working on a new technology which combines sampling and physical modeling for a while and here, from absolutely nowhere, comes this new Synful plugin and creates all this buzz. In addition, posts are made with A/B comparisons between this new technology and Gary’s GPO library for all to hear.

Does Gary start slugging it out in the forums in defense of his product? Does he take pot shots at Synful Orchestra, pointing out its weaknesses (easy enough to do, even Eric Lindemann himself admits it’s not a finished product)? Nope, he posts this in the Northernsounds forum:

There is nothing to defend and nothing to be frightened about as Thomas_J and Drew suggests. Quite the contrary. This is good news. Eric has done groundbreaking work and his work should be supported. This opens up digital orchestration to a whole new realm. Physical modeling and Additive Synthesis are the future of digital orchestration. I do not view Synful as a competitor but as a partner in bringing digital orchestration to many musicians.

In September of 2003 I posted that the goal was to "bring people into the digital orchestration fold - hobbyists, novices, the curious, pop artists, students, starving artists, etc. People who were once excluded who don''t have the budget for more expensive libraries will literally be players....As more people come into the fold, everybody will benefit." Eric's work furthers that goal and Eric is to be applauded for providing tools that benefit more musicians. The more colors in the artist's palette and the more tools and options, the better.

Physical modeled instruments are not new. There is HarpTime, CelloFan and SmartFlute, but Eric's work takes it further. Physical modeling and Additive Synthesis solves many of the performance issues that are problematic in sampling. But there is still more to do as this technology is in its infancy.

As Jkerr points, it is no secret that we have been working on physical modeling for the past several years. Quite different from what Eric is doing since it will be a hybrid of sampling and modeling, but another valuable tool in the palette. This is something we've been discussing on the GPO Forum. I believe that the ideal approach in the future may be a combination of technologies that are best suited for particular instruments,

I welcome Eric to the Northern Sounds community and wish him much success. There are many people here who will offer great input as they have helped other developers. We are going to see a very interesting year ahead with new announcements, more orchestral libraries and the convergence of technologies.

Gary Garritan

A class act if ever there was one.

/Yoss

Christian Schüler · Post by **Christian Schüler** » Tue Dec 28, 2004 2:14 pm

FM has expired 1995 or so, which by this date you have seen the proliferation of FM soundcards for the PC

Some of waveguide stuff will expire over the next years, but as of now it's not enough...

olepro · Post by **olepro** » Tue Dec 28, 2004 2:51 pm

For fast passages, Synful does do something interesting.

But a lot more work needs to be done on the sustain.

A Franck Violin Sonata with Synful Violin
http://www.wusik.com/song.php?id=403

Franck Violin Sonata with GPO
http://www.wusik.com/song.php?id=362

My ears still hurt from working on this piece with the Synful ...

(other GPO examples at http://ybacuo.wusik.com/

YBaCuO

There´s no comparison at all, the synfull version is so much more real sounding, the synfull violin sounds like a instrument, not a sampling!
The GPO violin sounds like the rest of the strings, dead and static.
It´s funny how the synfull violin in the front can get the GPO strings sound way better !!

griels · Post by **griels** » Tue Dec 28, 2004 3:32 pm

Christian Schüler wrote: The other is a version where I tried myself at physical modeling, using my own developed algorithms. Overall it sounds a little too clean, but I am quite satisfied with the brass articulation. Too bad I cannot make a VSTi from this, because of the patent situation .

http://www.ariescode.com/scratch/physical_robin.mp3

Damn that brass sounds nice

You couldn't even distribute a VSTi of the algos for free?

If so, Yamaha get off yer arse and make that VL70-m emulation

And yes, the Synful strings sound amazing!

Motion · Post by **Motion** » Tue Dec 28, 2004 3:45 pm

From the Synful site:

RPM Phrase Database
The Phrase Database contains musical phrases for each instrument. These are not recordings of isolated notes but complete musical passages that represent all kinds of articulation and phrasing.

Searching and Splicing
When Synful Orchestra receives MIDI input it looks at the pitch, velocity, amount of separation or overlap between notes, note duration, volume, pitch wheel, and mod wheel to determine what kind of phrase is being played. Synful Orchestra searches the RPM Phrase Database for fragments that can be spliced together to form this phrase. These fragments represent transitions between notes, slurs with portamento, lightly tongued transitions, aggressive fast bowing, rapid runs, long sustain regions with graceful vibrato, and noise elements like flute chiffs and bow scratches. Synful Orchestra stretches and shifts these fragments in time and pitch so that they combine to form the final output.

Synful RPM Additive Synthesis
A single note in Synful Orchestra may be built from three or more rapidly spliced phrase fragments. Splicing ordinary PCM sampled sounds in this way would create unacceptable warbles and clicks. Synful Orchestra uses a patented form of additive synthesis in which sounds are generated from combinations of pure sine waves and noise elements. This gives Synful Orchestra the ability to rapidly stretch, shift, and splice phrase fragments while preserving perfect phrase continuity.

Personally I'm absoloutely smoked by what Mr Lindemann has achieved....Garry Garritan is right on the money with what he says regards the benefits to digital musicians.

I sit here with the demo of Synful and have never played in realtime off the bat so easily such an expressive instrument that is effectively virtual or hardware based,up until now it was AAS Tassman or the Ultra analogue.

The expressivity is wonderful and I can only see it get better with more phrases for the database.

The fact that people compare this to high end sample libraries is evidence enough of its abilities and whilst it can't compete in various ways it more than makes up in other ways.

But what is more important is that electronic musicians in all its many genres now have the ability to employ quality expressive traditional orchestral instrumentation in our music making so easily and this is just the beginning...R 'n' B,Jazz and other electronic/synthetic based vsti's planned.

It also gives many the opportunity to use this technology with respect to low cpu and memory use on what is regarded as a modestly powered PC(80-160 real-time voices are obtained on a 1.5 GHz Pentium IV with 512K memory dependining on note speed and other factors.30-60 real-time voices are obtained on a 1 GHz Pentium processor without SSE2)(Pentium III or lower)dependining on note speed and other factors.)

This almagamation of resynthessis mapped to realtime playing dynamics is just incredible(my understanding of the above anyway)along with how this is also implemented in conjunction with the mod and pitch wheels is just great...sure it's not perfect but none the less quite remarkable,spectacular in fact...the technology and how it's been implemented has had an 8 year pregnacy and the way I see it is that right now it's a helluva baby,where this will be in a few years will be even more impressive.

I also asked him this a few days ago...BTW....any other plans to apply the technology elsewhere re other areas of electronic synthesis in particular?

From the dev:

Yes, The technology lends itself to all kinds of morphing and strange new sounds.I hope to get there one of these days.

Eric

Most folks are probably familiar with the likes of Cameleon,White Noise and Cube's morphing abilities.So lots of potential other creative uses hopefully on the way/to be added to the Synful vsti's...anyone for violas turning into pianos,percussion melting into atmospheres as you play..mmm

well lets hope so anyway.

So the future for Synfuls own take on this is looking very promising.I'm looking forward to see how things develop further.

Lunch Money · Post by **Lunch Money** » Tue Dec 28, 2004 6:18 pm

I'm glad to see there are people who "get it". It's not even about what it currently sounds like, but about the potential and the theory. The sound itself isn't half bad already, though.

I'm especially glad that Gary Garritan wrote such encouraging words. I mean, I have no stake in it at all, and hadn't even heard about Synful until this thread, but it's sparked my imagination a little bit. In the meantime, seeing Gary's words makes me want to buy HIS product even more.

Greg

Dufay · Post by **Dufay** » Wed Dec 29, 2004 7:55 am

Has anyone had trouble installing it? I have the system requirements and I accept the default locations..after I press next, nothing happens...

mtyas · Post by **mtyas** » Wed Dec 29, 2004 9:38 am

Draagen wrote:Is it just me - or does the sequencer reset all the instruments each time I press "play" and wanna hear what I just did?

Yep, always goes back to the flute sound.

Oh, and by the way, try playing in tune at 48kHz

useruseruser · Post by **useruseruser** » Wed Dec 29, 2004 9:54 am

Yossarian wrote:www.synful.com

The demos are amazing – and this thing only takes up 75 MB (the orchestra just 32 MB)!

Off to download the trial version now.

/Yoss

LOL.

I'm a classic freak regulary visiting symphonic and chamber concerts and also recorded them at pro level.

Those sounds meight impress people, who never in their live were in a real concert hall or a classic music recording studio ...

But it sounds definitively terrible and absolutely strange, if you actually know how classic music *should sound* and (gladly) mostly sounds.

I think, that a decent symphonic sound library has much more to offer in expression and realism, than this instrument...

I am somehow amused (about the *big words* on this website). Sounds like a joke to me, if I hear the examples then...

Sorry, but I believe this instrument (at current state) may be able to fake some "classical feeling" but never to satisfy us "classic freaks" even for purely digital pre-productions. So I will continue prefering high quality megabyte heavy sound libs for pre-productions in the future.

So remains the hope, that this peace of work will be continued. Maybe one day it will become an alternative, but not at the moment.

Only my personal opinion.

.

Synful Orchestra - wow!