If you’re a visualist, working with digital video, then shit is in the process of becoming real.

Toby*Spark initially announced the “*Spark D-Fuser” Sub-$1000 DVI mixer project 8 months ago, and followed up with some more specifics. If you Follow Toby on Twitter then you will have seen some sporadic details and titillating photos appear over the intervening months.

Now, of course, VisualBerlin Festival is happening, and it’s an auspicious time for an announcement:

*spark d-fuser: dvi mixer project presentation [2010] from toby*spark on Vimeo.

Toby’s presentation is a beautiful enunciation of why this kind of hardware is important and useful, as well as how it came to be. It’s a great 30 minutes of viewing, but if you’re at work and don’t have that kind of time, here’s the executive summary:

Locked In Functionality

  • Commercial-quality hardware (Toby has a masters in product design, so this isn’t just a DIY looking box)
  • Two channels of DVI input
  • Plug and Play setup – common resolutions work straight away
  • Crossfade between channels, fade to black knob, tap (“take”, or “cut”) buttons
  • Resoutions: 640×480-1920×1080, up to Triple Head at 640×480
  • Sends sync to projector, no matter what the inputs are doing: No more “Searching for input” showing up on your screens

Possible Extra Awesomeness

These  extra bits aren’t set in stone, but are high on Toby’s feature list for the first public release of the mixer:

  • Crossfader curve control between Alpha and Additive blend
  • Additive blend mode
  • Custom firmware for the video processor to make it more VJ-friendly
  • D-Fusers to be chained with ethernet for more screens, Master/Slave control
  • OSC control via ethernet
  • OLED screen and menu control
  • Open Source Software powered by Arduino – control software will be community editable!
  • Community-built software updates should allow the box to be compatible with higher-end models of the video processor for greater numbers of inputs and different capabilities

Pricing/Availability: US$1000/£750 is the intended price point. Available soon.

If you haven’t already, check out the product page, and hit the expression of interest form so Toby knows how many of these things he should be making!

  • VJ Air

    awesome news, i thought it might never appear but im realy glad to be prooven wrong. time to start saving my money!

  • Awesome. Can't wait to start using this. It's the magic box we've always wanted/hoped and dreamed for. This was like the iphone 4 keynote for visualists!

  • M7

    Really stellar. I can only imagine the hours and hours that went into such an amazing device. Can't wait to pick one up. Truly a more simple and elegant solution than my v4.

  • Well, don't go throwing your V4s away quite yet. I don't know of any video cameras which output DVI/VGA, so there's still plenty of space in our world for the "traditional" devices. But most of the V4s I see get used for precisely this: Fading from black at the start of the set, and switching between VJs between sets. So it perfectly fills the small niche for what most visualists are doing at most of their gigs.

  • Sterling

    Fun fact: the 'processor' he keeps talking about looks like its a variation of the TV One 1T-C2-750.

    TV one has a few other products with RS-232 ports as well.

  • sleepytom

    Yeah – the only remarkable bit of the story is how bloody long its taken toby to do anything other than hype on about this.

    The box is a TV One 1T-C2-750 or TV One 1T-C2-760 (either will do) – the Rs232 controller is built from an Arduino.

    Toby (MA/BS product design) claims the software will be opensource (its probably based on GPL code anyway as most Arduino code is) Why is there no release of the Arduino code now? Why is he keeping everybody waiting with stupid functions such as LCD screen, ethernet, whatever whatever. Fader curve, remote control, whatever… Nobody cares about these really, they just want the basic DVI mixer for fuck all money that they were promised YEARS ago.

    Its just pure designer wankery – never before has such a good idea been so delayed by such a small amount of changes.

    At the end of the video he talks about the bigger products from TVOne – these are key to the success of his controller and potentially offer features that VJs already want (such as other input formats). Weirdly TVOne already make a TBar remote controller for these products which is pretty good for VJs, which means the only advantage his box has is low cost and opensource software so you can tinker with it.

    Of course his box has a major disadvantage in that it doesn't exist and you cannot buy one (at the moment) Toby will not commit to a release date, he won't even commit to a functional spec for the release. So by the time your able to buy this he might of built in a coffee machine into the box.

    Each time the spec changes the design changes and then the manufacturing process doesn't start. He should of released a version 1 design in a short run production (like 5- 10 units) – put the damn code up on the web and allowed some feedback from users. Actually he is more concerned about the box looking professional than he is about it working well, so we won't see any physical units until Toby deems them perfect. This leads to other issues as without the help of others we have to trust that Toby has thought of everything.

    Net result of Toby's hype machine is that we still haven't seen any working examples of the mixer, Toby has decided that his ideas about ease of use and elegance are absolutely correct and will be put into the final unit before ANYBODY else gets to use one of them for any length of time. This is not how opensource should work as the major benefit of community projects is the meeting of minds that creates a product better than any individual could conceive on their own.

    In actual fact anyone who has wasted 30mins of their life watching the above video will of noticed the one glaring omission from the video – yeah, where is the bit showing 2 computers being mixed with this box???? Why is there no shot showing the mixer in front of a LCD screen so we can see the operators hand and the output and watch the thing fade between two sources?

    Come on enough of this hype. Get the product to some users and be prepared for some feedback. If your serious about opensource then make it opensource and draw on the opensource community to help you improve the device.

    Stop adding functions / features and just get some of these things out there for real testing by real users.

  • xnay

    no good deed goes unpunished 🙂

  • Peter Kirn

    Tom: news flash. There is no Toby hype machine. Jaymis and I have pushed Toby, in fact, to release information because we're excited about seeing some activity here.

    Given that this is Toby's personal project, I'm just unclear on to what you feel entitled here. It sounds from your email as though you feel somehow personally slighted, so help me to understand why that is. Did you invest any time or money in his solution? (You refer to the time you spent watching the video, but beyond that?)

    Honestly, it's comments like this that often discourage people from open sourcing projects or giving back to the community. *Constructive* criticism is absolutely, absolutely welcome. But trashing someone essentially for not having completed a project — come on.

    I'm often in the situation of trying to encourage people to release information in advance of when something's done. If people aren't supportive of that, it makes it harder for me to make that argument. So I really do want to hear your criticism – seriously – but then, if you want to see Toby act on that criticism, it has to be constructive.

  • deepvisual

    I just don't get it.
    Toby has, or rather hasn't yet, only designed a bespoke remote control for a retail DVI mixer.
    As he seems to be doing this for next to nothing and there is almost no money in it for him, it would have made more sense to just open it up and let people make their own. I'd rather pay him $100 shareware for a DIY kit of the RS232 code and arduino plans than wait two years for him to develop 'perfect' knobs, which I can do myself. .
    Besides which, you know how the market is. The manufacturers will probably drop DVI in two years time. under the present business model, all it takes is for Toby to walk under a bus and the whole thing evaporates – and that includes the currently non existent customer support for anyone who buys a finished model..

  • deepvisual, sleepytom, its more than just RS232. We are working with the manufacturer to get custom firmware implemented to give more options and frequencies included. Things like standard triple head timings, a different EDID subsystem and blend controls.

    sleepytom, to put it concisely, get a fucking clue.

  • To elaborate, this box is *not a mixer*, its a presentation switcher and scaler that has a custom designed image processor that TVOne built for being able to do mixing, keying and transitions at 4:4:4 quality. If you have ever played with any professional hardware like this you know it is *not user friendly* and you can completely destroy any chance of non technical users being able to literally plug and play and go.

    Specifically the EDID subsystem is not designed for a box to be moved around to different venues and switched from projector to projector. Its meant to be put in place in existing infrastructure and left there once everything is set. Not much about the box or the hardware is meant to work dynamically from the unit. Really the only thing remotely useful on the box is a fade that does a pre-timed transition. The firmware / software this thing runs is really designed for a "facility", not club a, basement b and some guys roof. This makes some things a bit more complicated for detection of the proper output frequency from the mixer (ie: DVIout to output device). And since the menu/screen is an OSD, that means you may have a hard time setting the box up and getting output if you don't know the subtleties of all of the video specifics your output device needs. Is it 59.94, or 60HZ, is it tlYUV, RGBHV, RGBsB, RGBBs, interlaced or progressive, etc. There is not one button for any of those settings on the system, and thus to set those, you need to see the on screen display. To see it, you need to get most, if not all, of those right. See the problem there? There are some short cuts in the menu, but every device I have tested is amazingly stubborn about getting just about all the settings dead on.

    Thats one reason for the LCD display on the controller, so you can go through common presets that work with the most hardware. I do professional video engineering, and work with high end broadcast HD Decks that have less options that this thing has. I'm not really exaggerating. That, plus the windowing controls and the some specifics to how they chose to put their pipeline together means things are not as straightforward as we, or you, would like them to be. Also, you know the unit and model number. If you are so fucking antsy, go and get it, and build your own.

    As for the DVI, you can go HDMI out of most modern HD camcorders and DSLRs, although there are some nuances on the RGB to UYV conversion (HDMI handles both), as well as aspects of HDCP that can cause issues. The HDCP issue alone was very annoying to troubleshoot.

    As for the lack of video in the presentation, thats partly my fault. I was at WWDC when Toby needed the video from our testing get together here in NYC, and I did not have everything, let alone an easy way to get those to him in time. I'm happy to throw some stuff up. Peter is meeting me this week to look at the unit, maybe we can put something together.

    Anyway, ill stop ranting.

  • sleepytom

    I simply disagree about the ease of use of the TVOne products – i've been using their products since before they were even called TVOne and they are easy to use. Setting up the switcher is not rocket science, and whilst it might not be a totally plug and play solution its good enough for the vast majority of users. Anyone using this box with a th2go will simply do the setup once for the th2go resolution and then not change it. That and XGA would be plenty of choice to cover almost all of the situations a regular Vj wil encounter, a simple toggle switch could be used to change the mode, no need for LCD menus. I think that many VJs will actually want to plump for a better tvone switcher to control anyway – for around £1000 there are a few options for switchers which have more inputs in other formats, these switchers have LCDs built in so can be setup for whatever format screens without needing an LCD on the controller.

    Yes its possible that some users will like an LCD, ethernet, and whatever bells and whistles toby thinks are needed, but most users won't need these things and would prefer a basic unit in their hands than a perfect unit in their imagination.

    Yes we'd love TVOne to tweak the firmware and add additive mixing, but we'd be just as happy to have it as an option to upgrade our existing units. Really if we're going to get TVOne to do anything then cutting down on the control latency would be the best thing they could do to please VJs. But actually the stock unit is perfectly usable when setup properly, and an improved firmware could be added at a later date.

    The main cause of frustration is seeing the project advertised (as in "sign up here to buy one for £750") then seeing the project stall, then seeing the project renounced and realising that we might have to wait forever for toby to perfect functions that many people simply never will use. I could add a whole list of functions which may or may not be useful to some all or none of the users of this mixer, but that is the point of opensource – if we had access to the basic code / arduino instructions then the community could be adding and testing functions which would then give a sensible base of user feedback to toby for incorporation into the final commercial version of the product. Net result would be a better product for everyone.

    Had Toby started out from the point of "here is an rs232 controller i've made for a dvi switcher, I'm thinking of opensourcing the code and maybe producing a short run of commercial off the shelf ready built units" rather than "Buy the dFuser for £750" then we would probably be less frustrated.

  • I'm honestly not understanding your complaints. If you've used it, and are happy with it as it stands, then why are you remotely interested in this project? You already have a box that does what you want, and already state most users would want a different controller.

    Sounds like you are just complaining to complain. If you think the project is stalled, whats stopping you from learning RS232, getting an arduino and reading the manual for the 750 and, well, *doing it*?

    It sounds like you have issues with how the project is marketed and run, and what I would say to you is simply "try doing this yourself". If you don't like waiting you have all of the information and hardware literally, at your fingertips.

    I disagree that the TVone is user friendly, and I also disagree that XGA is adequate for a toggle button/two resolutions. I can tell you I certainly would not be setting it for triple head and walking away. Not all venues have 3 projectors, let alone 1. Not all VJs have 3 projectors, and for the VJs who need to be able to accommodate changing parameters (different output devices here and there), being able to easily set options without needing to use the OSD (again, the catch 22 is to set it up, you need to see it, which means you need to have it set up to some degree).

    Anyway, Toby is going to continue to run this how he see's fit – and personally I am on board with how he is going about doing it, and I suspect you will continue to complain, and do little to remedy your own situation. Oh well.

  • Please ignore Tom – although he is informed he does just represent a small, overly critical group of just 1 person.

    Sure it's vaporware until theres more than a couple of them but its coming and there's still no real digital mixers in the cheap(ish) end of the market so this is still exciting and if Tobys guilty of anything – its getting excited about it and telling us how much it truely could cost for something like this.

    I never want to learn RS232, build or solder anything. I'm a creative above all else – I love my toys and equipment and understand it better than most VJs but I've got better things to worry about – mainly what I do with the gear.
    It's certainly not that I build it myself or that I make sure you involve SleepyTom in the planning and execution.

  • massta

    Iotech, you are right: this is still exciting!

    Honestly, I'm glad a VJ is building VJ Gear, which alone is worth waiting for.

    The V4 is still special but we all want something better and Toby is doing his best while sharing his progress. His presentation is genuine and my hat goes off for getting Vade involved. This unit, Berlin, is going to be very appreciated. Nice job on the firmware! I'm sure adding the ethernet connection made it easier for updating.

  • I think Toby and what's been done is cool, custom PCBs, design and everything: but SleepyTom is damn right (maybe rageus, but right): I can't see any "Arduino" or "TV One" mention on toby's site (on which I looked amazed some 9 months ago).

    The idea you get from the site is that a mixer is being manufactured by toby and d-fuse group.
    They made a cooler thing, in a way: they adapted some previous hardware, and took its design and functionalities to a next level.

    Why not writing it down on the site?
    Why not sharing the code?
    Why having the financial delirium to sell everything and not just the Kits, as mentioned above? Maybe pre-assembled for the people who have "better things to worry about"?

    Please don't take the "critical part" of the comments in the wrong way: maybe Toby will do that.
    As this video and this post happen to be a "behind the scenes" news. We all wait for the sale of the mixer. And maybe some codes ;)?


  • sleepytom

    why do you think ethernet has anything to do with updates? (arduino updates are natively done via serial / usb)

    ethernet is just for (potential)OSC and remote control (hmm remote control of the remote control – maybe i'll make a remote controller for the remote controller controller…)

    i'm not just complaining to complain. My complaints have (right from the start) been accompanied by offers to help Toby (partially taken up, he did at least let me pass on some contacts to him) , and a desire to see the product be successful.

    Why don't I just make my own? Because Toby's already done it and has already "sold" his product to loads of people. Making a competing product makes no sense given their is such a tiny market for this device.

    People really should check out Toby's ideas for how this is to develop, when i last spoke to him he was talking about essentially opensourcing the whole project and effectively selling a DIT kit – though it would come assembled and working there is no on-going support planned. He was (still is afaik) talking about essentially just selling this as an individual, he isn't starting a hardware company and isn't interested in fixing your unit when it breaks down. When I last spoke to him he wasn't even going to supply the TVOne box with the controller, suggesting that people should buy their own from another dealer. (confusingly he was also talking about the carrot of a bulk order being some how important for TVOne to make changes to the firmware)
    Mainly he is working on his PHD.

    This is the troubling bit for me. People like Lotech say things like "I never want to learn RS232, build or solder anything. I’m a creative above all else" which is fine IF your buying a product with a company behind it who can offer support, training, advice and speedy replacement in case of breakdown. Toby isn't planning on offering this support, instead he's asking the opensource community to provide his support for him. There isn't very much wrong with that in principle at least, so long as end users know what they are buying into and their is a real community of knowledge to provide the support. What troubles me with Toby is that he is announcing it like its a real product from a real company, when it simply isn't. (Hype once again, making full colour poster adverts with a big price tag at the top, rather than actually being honest about the situation) This combined with the delay is alienating the very people best placed to provide the community support and development.

    Vade – re the OSD catch 22 – yeah i get it, but i don't see it as a problem. A toggle switch to select XGA or TH2GO would cover 99% of VJ situations i've ever experienced. OK so if you plug it into a projector with a different native res you'll get scaling on the display – but every computer monitor, projector, plasma, and LED controller supports XGA, so you always need to be sure you can get back to XGA (which oddly is not recalled by doing a factory reset as this uses uXGA for some unknown reason) A switch to select between XGA and TH2GO costs nothing and avoids a whole load of dev needed to get the LCD working, its inherently more reliable as you not dependent on callbacks to display text and let you navigate the menu. A switch to reset to XGA lets almost any monitor be plugged in and used for OSD so even if you want / need to change the output format its possible.

    What about the "artsy" users? They won't change the format very often anyway, if they do need to change it then a simple toggle switch to select the preset output format is MUCH easier to use than an LCD display menu system. You only need to take a look at the number of pioneer DJVs out there which still have the pioneer logo / screen saver turned on to understand that artists do not use the menus of their devices.

    The main point though is that if the community is going to be asked to provide the long term user support, then the community should be asked for contributions to the development. Without getting some units out to people for field tests its impossible to say if the menu is a good idea for the intended users, user feedback has to come from real users not be made up based on a "i know best" attitude.

    Thats why I feel that prototypes should be in the hands of users, so that refinement of the spec can be done before any production runs. This step is hugely important to avoid blunders (eg FTB on the V4 which also fades the PVW out) and would go a long way to making the unit better.

    I'm not just complaining for the sake of it, I'm trying to help this idea not only become reality but also be a success.

    Sorry if anyone feels i'm being unreasonable..

    The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man. — George Bernard Shaw


  • deepvisual

    if you want to get tvone excited dont talk about VJs talk about churches. far more churches bought V4s than VJs ever did.

    and like I said, no customer support is going to be a real big problem as I suspect you may bound by EU regulations to provide this, even if you don't want to.. and that could turn out to be a real problem on a low priced project like this.

  • kimba23

    The critics are really lame. unreasonable men who are changing the world, please go do something and stop complaining. If i was one making this thing, i would never give sleepytom or deepvisual any testing units (what would be the point?). In the other hand I think it is a great project, and when/if it comes out, i take a look at it and may get one, without costumer support!. So keep up the great work toby. keep up the hype!

  • there really is no pleasing some people.

    the bottom line is that *if* the unit had been sufficiently plug'n'play, and *if* the company behind the arduino variant had not stopped producing it, or even had ever replied to me, then just publishing a 'this is what I did' would have been great. I'm all for people learning how to solder stuff themselves, as I hope the presentation showed: change the world to the one you want.

    as that wasn't the situation, we're where we are. a proper controller and custom firmware.

  • sleepytom

    you could use a max232 shield on a standard arduino

    so you've dropped all the opensourceness and have just produced a custom commercial product?

    where is this custom firmware? have TVOne actually agreed to do it? Has it sorted out the latency?

    release date?

  • Do you have reading comprehension issues? Because It appears as though you do.

  • Pingback: Create Digital Motion » Mobile Video Out Everywhere: iPhone 4, HTC Evo, and Droid X()

  • Luxcollective

    it sucks to see so much nastyness… I certainly don't know enough about the underlying hardware and software issues Toby is dealing or what Sleeptom's angry entitlement status is all about, but the tone really sucks. If a useable version of a desperately needed VJ product comes out of Toby's personal effort, then great… if the issues prove to be too daunting to move it beyond one unit that meets his own needs then no harm done… if people don't like the hard work of others then do it themselves.

    For the record Toby, I appreciated your presentation and sharing of all the challenges… since 90% of VJ's know nothing about RS232 or any of the other components that go into this, I for one admire the effort you have put in to it whether it is ultimately successful or not.

  • Pingback: Create Digital Motion » Inter-App Visuals on Mac: Syphon API Opens Up Visual Collaboration()

  • massta


    Anyone purchase one? Would like to see a review.

  • Chakracorona

    Master blaster the best in the market super hardware felicidades