3L has been in the wild for 2 weeks now, so I’m expecting to see some work from new users appearing online soon.

Since launch the artificialeyes crew have kept themselves busy. ExiledSurfer has been posting plenty of new content, example clips and screencast tutorials. I particularly enjoy the talking head screencast recorded from within 3L, using the Apple webcam to drop a little talking head into the preview window.

3LU Sophomore: Capture an Object + Feedback from professor thrill on Vimeo.

These videos have been collected into a Thrill University section of the site. This is being supplemented with live tutorials, the first of which happens 12 hours from now (GMT1300, Thursday) on the 3LU page (which will even remind you via email when the session is about to start). I’m going to be there to increase my 3L knowledge, it would be great to see some other CDMers as well.

There’s also a 3L channel on CDMo’s favorite online video service, Vimeo. CDM makes a little appearance in the 3L channel, as I was able to finally edit the bulk of the “3L Sessions” videos we shot with Michael in Perth last year. There’s over an hour of video in the 3L Sessions album, covering the history of 3L, and an extensive tour through the interface and capabilities of the software. Through the magic of Vimeo, these are all downloadable, so you can use them to guide your first steps with Thrill.

3L Sessions 1 – History from Create Digital Media on Vimeo.

3L Sessions 2 – The Slider from Create Digital Media on Vimeo.

3L Sessions 3 – Interface Tour.
3L Sessions 4.1 – Shapes, Models, Mapping, Deform.
3L Sessions 4.2 – Processes, Particles, Colour, Objects, Effects, Camera.
3L Sessions 4.3 – Lighting and Multiples.
3L Sessions 5 – Font Synth, DVR, Audio + Graph Modules, Graph Morphing, Step Sequencer.

It’s great to see a developer working so hard on documentation, which is often lacking on “indie” projects. In fact, the only thing lacking now is 3L output from other users! I’m as guilty of this as anyone, having had a big head start with the software I still don’t have any output online. I aim to change that this evening, see you there!

  • Bert

    Am I the only one who finds 3L absolutely impossible to understand and use?

  • Hi Bert. There's a live chat going on right now which can help with that:


    Already quite a lot of information and many people in the chat, should be an interesting session.

    It is definitely an extremely confusing piece of software though, however once you start to understand it, very powerful.

  • Bert, you are not the only one. Trust me. The UI needs to be completely re-thought out IMO. Too much on screen at once.

  • yup vade, you made that clear during the beta test. However, it was our choice from the beginning to do a one piece interface and not a moveable modular one. People may think that the UI is confusing, but it is only comprised of 3 elements: sliders, buttons, and menus.The reason that the GUI is the way it is is because I dont like the way other interfaces are. There are a jillion parameters in 3L, and they are ALL always at your fingertips, not hidden away in popups. I / we are perfectly satisfied tolose users who are offended by the interface the same way that arkaos or modul 8 offends my sensibilities, or the frustration i feel when looking at the patchcords in qc or jitter or pd. It was a total conscious choice to design the interface the way it is and let the chips fall where they may.

    and yes, jaymis is right, it is confusing because it is a totally new paradigm for a performance app. Heres the thing, that i have sort of settled on as a realization in the last week about what makes 3L different:

    there are two types of programs out there:

    one that allows you to play clips. and affect laering and filtering, and the timing/timeline of preproduced material, and another type that allows you to code your outcome.

    in my opinion, most vj's are playing either material that they preproduce, or that they search and download…those vj's let the program do all the work for them.. and they are just interpreting pre existing content…you dont really have to "think"..u just "look" at it and TWEAK it

    on the second type of program, the type of people who use things like qc aor processing, "think" about what they want the program to do, and then design a solution to achieve that….and THEN they TWEAK during the preformance.. but that involves coding or patching

    needless to say, the first prgram type possesses 90% of the market the second type there are 10 percent creating patches and sketches, and a gazillion people scouring the net for those patches and sketches.

    what's different about thrill, is that it is between those two types of programs…

    the first type is planar/and linear
    the second type is dimensional / non linear

    to use 3L, people will HAVE to think… but they will have to think less than if they try to do something in qc, processing et al. For a performance app, that is a compeletely new paradigm than what 90% of the market is now doing. but, that 90% of the market hasnt been doing it as long as people like me have, and maybe arent quite as bored of that paradigm yet as i have gotten

    so, it will probably just take some time for peope to catch up to CREATING and TWEAKING in real time. but if u look at the user forums , people spend hours tweaking aroundon whatever sofftware theiy are hobbying with it just needs some time for people to tweak around on 3L.. the point of the 3LU and made with 3L clips is to show how many different ytpes of things you can do with the program.

    So the gui design reflects all of these decisions, and our experience n the beta phase, and now with the release, is that new, fresh people interested in BECOMING visualists dont have a problem with the interface, because they come with no predisposed opinions. We are interested in THAT generation. The second type of people are the peers of my generation who are already set in their ways and have their own ideas about how something should look. they don't work for apple or another vj software company, but they have tons of opinions and express them frequently on comment boards and are prolific posters on vj forums, and they rarely have anything good to say about anything, cause they know everything better. These are the "potential users" that we could give a fat rats ass about because they will never be satisfied with anything.

  • Appreciate the long and thought-out ideas here, exiledsurfer. And I'm personally delighted that all of a sudden, after the late 90s' look-alike soft synth interfaces with fake knobs and identical VJ app clip banks, we're seeing interfaces that look wildly different from one another, in music and visuals alike. I think creating some controversy is a good thing.

    I'm constantly thinking about UI and performance, and to me you also create some things in performance that are different than how you might work in the studio. That's why I'm interested in how the iPhone "remote" interface works, or how people do controller assignments. It's very possible this UI *won't* work for people onstage/in a club, as with most of on-screen UIs. That opens up other possibilities for how you adapt.

    Anyway, I was able to navigate 3L in Australia with a couple of drinks in me and very little in the way of tutorials … then there are other UIs that were supposed to be "easy" that I found baffling. So I think people do respond differently.

  • Bert

    To exiledsurfer: I'm not a naysayer and I do not think I know everything better, I'm just asking an honest question. Your rudeness and disrespect is completely uncalled for.

  • I'm just putting this out there:

    I've been testing 3L since February. It took me A LONG TIME to figure the UI out – and I'm still digging up cool new things almost every time I use 3L – that's how deep this software is. I can say, (now that I've spent X amount of hours) that the UI choice here is very well executed; I can grab anything I want quickly – 3L is 100% like an instrument. No one walks up to a piano and can play a song, and I'm sure that a person looking at a piano for the first time would have the same reaction for a VJ looking at 3L for the first time – they just don't understand. AND heck, if you dont like green – then press (c-trl+option+apple+8) to invert your screen πŸ™‚

  • "To exiledsurfer: I'm not a naysayer and I do not think I know everything better, I'm just asking an honest question. Your rudeness and disrespect is completely uncalled for."

    I think it is pretty clear that i have strong opinions, and that i am not afraid to express them, nor to make enemies in doing so. If i was directing my comment at you, i would have done the same as you did: "To bert, etc…."

    If you take the time to notice,

    i responded directly to vade's statement, about the gui in the first paragraph

    and i added a bit of my point of view to what jaymis had to say. in the second paragraph.

    i didnt respond disrespectufully to you, because i did not NAME you. You are correct, you just "asked a question. And you did not even mention the GUI, you mentioned "impossible to understand or to use"

    vade raised the issue as to why the program is not understandable as being related to the gui. Just because my opinion about the gui is expressed here on this page does not mean that what i have to say is directed at you unless i specifically direct my comments at you. i apologize to you specifically if you considered yourself to be included in the description of those who's opinion's i choose to either disagree with, or even further, to DISDAIN with strong language.

    It is my further opinion that it is in vade's self interest to criticize the gui, as he offers a different solution to interfacing with max, which can be seen at his website. It is his answer to the same set of problems that we addressed in our own way with 3L which was developed in max. I fully recommend his solution for people who want to have a minimal interface to their patches.

    If one pays close attention to what i DID write, one would notice that vade is ALSO not included in my last ranting paragraph, because vade DOES write software, and more importantly he writes solutions which he distribute's freely and documents and supports well to the vj community at large

    Thrill, however, was developed to be exactly the way that it is… it is HALF of the point of the whole program. Every individual thing that one can do with thrill has its basis in max and jitter behind the gui. Hell, one can just grab the jitter examples from the trial version of Max/jitter and slap vade's gui and module solutions on top of em and duplicate a lot of what thrill does with out the gui that he think's needs to be completely rethought and redesigned. Vade has his right to take issue with the gui, which he has also done with me in the past in private chats, and i expressed to him the exact same position then: That is the gui and it aint going to change.

    Furthermore, BERT, I AM FULLY AWARE that you are not alone in not understanding how to use thrill which is the reason why i have spent an inordinate amount of time in writing a manual, and creating the 3LU series of screencasts, as well as answering forum questions , and WRITING FULL descriptions on exactly how each example was made, and why back in november jaymis and peter sat down with us on the other side of the planet to videotape me explaining the program. I fully anticipated that people would have difficulties with getting up to speed with 3L because 3L totally fucks my brain and challenges it every time i use it. That's the point, That what it was DESIGNED to do it was designed to FORCE me to discover new things to do and new ways to do them, because I was bored with everything else available. Part of forcing myself to rethink what i was doing included making a new interface for that process, rather than a familiar one.

    And that, is why i am adamant and non-apologetic about the result which we released to the public.

    And also why i am totally willing to take as much time as it takes either publicly in this forum, or on our website, or our forums or through emails or our screencasts to help people to understand how to use the program and encourage them to get the most out of it, and more importantly, the most out of themselves. That however, demands an investment of time and effort, as any complex or worthy endeavor does for those who undertake it.

    Understanding the gui is the first step along that path, an that has been stated unequivocaly on our website, in the user manual, in our press releases, and obviously now here in this comment thread.

    Putting 3L into the public domain as a commercial product is a statement of intent on our part to match or exceed a user's commitment to getting the most out of their machine and their creativity, and to assist them in doing so.

    If you want my help in figuring out the path to 3L, then just contact me, i am pretty easy to find.

  • Its the worst UI Ive ever tried to use. I can see why it might find a niche where it could be really popular and useful, so I wish it luck.

    I was bound to be disappointed because it was one of the best hyped & marketed VJ apps Ive seen. But from the moment I tried it, I wanted to runaway, the load time was enough to put me off for a start.

    Once the kineme 3d plugin for Quartz Composer comes out, the combination of VDMX and Quartz Composer will do far more and with a saner interface. Thats just my personal opinion, about what suits me.

  • @ steve:

    "from the moment I tried it, I wanted to runaway,"

    totally legit. everything is subjective, that's why we call it art. or religion. and its the same way i feel about arkaos or ANYTHING that runs on a pc. That's my suit.

    "the load time was enough to put me off for a start."

    and how long is that? takes about 80 seconds on my machine… faster than loading photoshop. Also, if you load vdmx with a full preset with a few media banks… i have seen it take up to 5 minutes to load with complex settings.

    "Once the kineme 3d plugin for Quartz Composer comes out, the combination of VDMX and Quartz Composer"

    cool man. the more that party in this dimension, the better. FLAT video is dead. πŸ™‚

    "I was bound to be disappointed because it was one of the best hyped & marketed VJ apps Ive seen"

    That's definately cool to hear in a funky sort of way πŸ™‚

  • Well I didnt time it but yeah 80 seconds sounds about right. I havent had VDMX take anything like that long to load, but it might be the way Im using it, I havent done anything too complex with it yet.

    I dont think flat & 2D are dead, there is so much difference in style of what is created, and different techniques, and lots of people who are not fans of the 3D stuff they've seen. I can understand their point, but it still stings a little when some are so instantly dismissive of all things 3D.

    Take for example the objection that the sorts of things we can do in realtime in 3D with VJ apps, look a decae or 2 behind the times. That is true in some ways, but I like to try to turn that into an advantage, I am trying to create stuff that looks old fashioned but with a modern twist, and I dont particularily care for ultra-realistic 3D as used in movie effects. Plus back in the 70's and 80's primitive 3D was only accessible to a small number of people, and took a long time to do, I dont see anything wrong with embracing it now thats its easy. There is a continual challenge to avoid stuff from looking cheesy, but Ive decided not to let that bother me πŸ™‚

  • @ steve
    well, i am unapologetically old fashioned, the gui reflects that commitiment, as well as what i "make" with 3L. actually, after now looking at your website, i am doubly bummed that our gui turns you off, because you are doing the type of stuff that 3L does so well and so easily.

    i understand that many people are unwilling to move away from the quartz composer environment, as it has received widespread support across applications. We had decided early on not to include qc support because when we started, we didnt think quartz was that hot, and it was just another "port" do go through to get to pure gl commands, and we wanted to do something closer to native, in our own way. In any case, way more time went by in creating 3L than we ever thought possible… it was always a "month away" from release, and in the end, 2 and half years intervened on what we set out to accomplish,, and in that time frame qc acceptance became pretty dominant. But still, even looking at what most people are doing with qc, i haven t seen anything that 3L can't duplicate, and many things which qc can't even do. (and probably wont be able to). Someone will duplicate some of the features for qc however in whichever way possible, and that is not a bad thing…

    And really, what i have chosen to create in 3L is no measure of what can actually be done with the program… if one spends some time tweaking, basically any style can be achieved in a 2d or 3d "look", and i have tried to show that in my examples, but i really believe that 3L is capable of far cooler stuff than what i do with it, and i am reall looking forward to what other people can make it do who have a different visual sensibility than i do.

    anyway, steve, i really think that if you gave the program a bit more of a chance, and spent a bit of time trying to understand the logic of the ui ( which is really pretty simple as opposed to how many different things are displayed in the one-piece gui) that you would end up having more time for exploring your own creative vision inside of the 3L environment than you would with various qc patches… the set of variables that you immediately have at your fingertips without having to go back "into the patch" allows for pretty much an infinite set of possibilities.

  • Well although it isnt reflected in much of my work yet, my real interest is in being able to do quite complex 3D animation in realtime, for things like virtual 3d instruments, 3d bands, virtual puppets etc.

    If I could program well then I should have made my own solution using a game engine. If I had more money & any faith in future development, I should have gone with Touch Designer on Windows.

    Instead I tried various windows programs such as VJO and Pilgrim R2, which in the end did not have enough 3D flexibility for my needs. So I moved to Mac 3 years ago, and was underwhelmed by the few VJ apps for OS X that had 3D object support. Quartz Composers UI always pleased me for designing stuff because I am used to the node-based paradigm. But out of the box its UI is not good for live performance, and it doesnt have very many useful 3D functions. But various plugins by the likes of Kineme are improving what it can do with 3D dramatically, and software such as VDMX largely takes care of the performance UI.

    To look at 3L again, I would need to know that it has features that lend themselves to more complex 3D animation, and that the framerate would be pretty good at resolutions of 960×540 or higher.

  • STEVE:To look at 3L again, I would need to know that it has features that lend themselves to more complex 3D animation, and that the framerate would be pretty good at resolutions of 960×540 or higher.

    Here we go:


    3L is definately not MOTION (the app), but it does most of what TOUCH does on the PC side., and far more than any QC patch i have come across, but just to give you a quick rundown:

    Simultaneously you can

    create 3 individual 3d objects, which can be either a primitive, a super shape, a 3d .obj file, or 3d text word/letter sentence.

    Each one of these objects can morph between two independent selections of the same type of object

    Each one of these objects can move independently of the other around its own axis, or the axis of its individual "object space.

    Each one of these objects can have up to 40 multiples of itself in a variety of manners (particle, grid, etc)

    Each one of these objects can be independently rendered either only with its exterior faces or its interior faces, or both faces., and as a combination of wireframe or solid independently for each face.

    Each one of these objects can be independently mapped with up to 4 composited textures. in 24 different transfer modes, including keying. The texture can be text or a mask, generated by 3L or a movie clip/ photo / live input, and each one of these textures can have up to 12 fx applied to them with each individual fx having RGB wet/ dry and a selection of 24 different transfer modes prior to being mapped to the object.

    Each one of these objects can have one of 100 different independent alpha mode combinations. and independent rbg ambient and specular qualities which determine how that object responds to the six scene lights which are individually positionable each with rgb controls for specular, diffuse, and ambient qualites, and these six scene lights can be animated individually.

    Each one of these objects can either have their depth turned on (so it is fully 3d) or their depth turned off (so it looks "flat" and you only see th front face and it's outline)

    Each one of these objects can independently have shape distortion applied to it whose values can come from either:
    – a random value generator
    – one of the four textures pre or post fx
    – any one of the other two objects
    – or the scene pre and post fx
    and this value of distortion can have one of 14 different math operations applied to it (subtract, divide, addd, etc)

    THEN, These 3 objects can be then animated as a group (scene)

    THEN, After each one of individual objects (along with it's multiples) is moving and being distorted, AND each of the individual objects "space in the scene" is moving in relation to the other two objects, and all three objects are moving as a group, you can further use the camera to navigate through the scene by:

    -using the wide/telephoto lens,
    -using the look left/look right//look up/ look down camera controls
    -using the pan left/pan right /pan up / pan down camera controls
    -and by using the zoom in / zoom out camera buttons.

    ALL of the above mentioned animation values can be individually assigned to one of 3 global lfos or 3 global audio analysis bands, or one of 3 video analysis bands, or mapped to midi (and in 2.0 release OSC) to automate their motion based on the bpm clock. If you want one of the variables to be out of time with the clock,

    EVERY SINGLE one of the variables (sliders) has an independent step or interpolate timing lfo. with up to 1028 divisions of the values. And every one of these individual incoming values can be independently smoothed from the others, or can have one of 5 different math operations applied to the incoming value to change its quality. And if you like the motion rules you have drawn in the graph in the INFO MANAGER for the slider, you can just copy and paste that value to another slider.

    Aside from the fact that you can assign automation to the oscillating value, you can also assign automation to the RANGE OF VALUE that is represented by the min and max value, if that range selection is >0 and <100, and this range selection can also have any of the above numerated automation modes and independent smoothing value applied to it independent from the value oscillation for each of the values.

    Oh, and did i mention that each one of these values can be chained to another value, and that this value can be mixed, and that THAT mixed value can be chained to a THIRD oscillating value and mixed with that value, and that the third slider can be linked to a FOURTH oscillating slider and mixed with that slider's values, and that at each step of the way independent smoothing values can be applied, as well as one of the aforementioned 5 math operators, and that this can be done with as many sliders as you hardware can handle the calculations for.

    So basically, you can create a scene of 3 objects and it's multiples and have each one of them morphing between two types and while they are doing that, tell each object where you want it to be and how you want it to move in the scene, and then you can animate that scene as a group, and then you can navigate through that scene with a camera.

    By my count, that is 4 (5 if u include sizing for fake perspective) levels of motion animation which basically replicates any type of motion that exists in physical space, or that can be done with a camera in virtual space and the possibilty to have up to 6 models of instruments which should be enough for anyone to create a virtual band.

    And hey, most of the best bands are just the classic guitar / bass / drum trio.


    In te prefernces of 3L,The second screen AND the DVR both support INDEPENDENTLY the following NATIVELY RENDERED (so not UP-sampled) output resolutions at 25/30 fps depending on your broadcast standard:


    Which pretty much covers native support for any connection to a projector over composite, component or dvi

    You should have no problem with your desired resolution of 960×540 with a laptop running at 2gz.

    Any of these second screen output resolutions can then be resized and stretched across as many multiple monitors as your machine supports, and THEN the output can be UP-sampled (zoomed) independently on either it's x or it's y axis to and the output will distort to whichever format that is. One could then turn around and "rescale all of the objects" so that they are no lnger distorted if it is a multiple screen output.

    And if your hardware and hard dirive are not fast enough to capture at the resolution you want to render at with the built in DVR (or would prefer going directly "digital") you can use the THIRD INDEPENDENT OUPTUT over firewire directly to your recording device, and you can select at which point in the render chain it is being grabbed: Texture 1-4 pre and post fx Scene pre and post fx, and switch between them. Which means that one can also have TWO INDEPENDENT OUTPUTS from 3L. Route the firewwire through a cheap A/D converter like the canopus, and you have two outputs to plug into your favorite mixer. Generate 3d on one, and play clips with the other using the built in media grid for movies —

    and all of this (and more) from the one-piece interface, the 3L gui, which one only has to learn THREE ELEMENTS in order to use:

    The Slider
    The Buttons
    The Drop Down Menus.

    Granted, there are a lot of icons in the interface. The slider alone has 5, (or 6 if it is an rgb or an xyz slider) and each one of those 5 icons hides a set of further icons with their functions. But that is exactly the power of 3L: all of these capabilities are built right in to the slider, and not shown in another pop up window or "module" across your screen: all of the possibilities to determine animation are right there on the slider doing the automation, with it's title at the top of each slider. You never have to move your eyes away from the variable that you want to determine and tweak it's behaviour.

    I hope that answers your question adequately.

  • bgrggfe

    The City Council is examining a request to open a Louis Vuitton Handbags and retail shop at 11502 Middlebelt in the Livonia Crossroads shopping center on the southeast corner of Plymouth and Middlebelt roads.The council heard at a study session on Monday from Taylor Bond, president of Children’s Orchard, who wants to open a 7,500-square-foot Louis Vuitton Handbags Sale store at the site of the former Family Buggy restaurant, which was closed several years ago.