Building stuff, figuring out problems, researching your way through setting something up – it really doesn’t matter if it’s building a complex fracturing setup in Houdini or building a sandrail for a Baja race, it’s about how you approach problem solving, it’s about a mindset.
Sometimes you come across Houdini setups that just blow your mind. I still remember seeing this setup of a baseball and thinking how in the world did they figure out how to accomplish this? And who is this individual who posses such magic? Well, the person in question is Houdini TD Johan Gardfeldt or as many may know him online as Johnny Farmfield. It also turns out it’s just good old fashion problem solving. A concept that comes up again and again in my few words with Mr. Johan Gardfeldt.
Some great things I learned about Johan. One is that he holds nothing back. Just straight forward and brutal honesty. I like that. I also learned he rode back in the day along with how he became to be such a master of complex setups in Houdini. Grab your beverage of choice and enjoy.
990 First, thank you for agreeing to answer some of my questions.So before we get into Houdini talk, tell us a bit about yourself. Where are you from, where did you grow up and all that.
Johan My name is Johan Gardfeldt and I was born and raised in Gothenburg, Sweden. As some might be surprised by that, let’s clarify all this Johnny Farmfield business.
In 1994, after spending a couple of months in Egypt, freediving, as I was heading back towards Sweden I of course met a girl and got stuck in London for 2 years. Living in London I just had a lot of hassle with people asking how to pronounce my name as well as noting its similarity to that of a, for the period, prominent fat cartoon cat, so my then girlfriend asked me if my name could be “translated” to english and with that the moniker Johnny Farmfield was born. Even after moving home to Sweden in 1996, I worked a lot against the UK over the next couple of years, so in regard to work, the name kind of stuck.
Twenty years later I live a pretty simple life here in Gothenburg, I got two kids, I’m a freelancer who work primarily from home and I grow super-hot chilies in my kitchen, just to randomly blurt out some facts about me.
990 So do you prefer Johan or Johnny?
Johan I’m completely split personality on that one – in Sweden it’s Johan with the possible exception on some clients referring to me as “Farmfield”, but in English, it’s Johnny, always. I also try to keep to communicating in English in all regard of VFX online, even with other Swedes, because I don’t want to keep people out of those conversations. It’s a weird world we live in.
990 Do you come from a design/animation background or do you come from a different industry altogether?
Johan Well, I spent most of my youth on a skateboard, had a couple of years being sponsored as a junior, 1986-88, so I traveled around for competitions a lot. My parents had a decent SLR camera which I brought with me and that’s how I got into photography.
Then, in ‘89-90, a friend of mine who worked as a photographer for an entertainment magazine here in Gothenburg got work in New York and she recommended me to replace her at the magazine, so I started working for them. At that time I already had gotten into vector graphics in Aldus Freehand and at some point Photoshop got released, so over the next couple of years I shifted my focus from photography to computer graphics – until I just got bored with it all and took off to Egypt with a pal to freedive, that is… 😀
Then, on my way back from Egypt, I “got stuck” those two years in London before moving back to Sweden. Over these years I worked with GUI design, illustration, and I eventually got into Softimage|3D.
In the late ‘90s my 3D work culminated in doing in an intro animation for the Gothenburg Opera House Millennia celebration. I had the deadline for that and my first kid on the same day, early December 1999 – and with that I basically dropped 3D from one day to the next. To have a set income I took a job for a friends company designing gaming computers and CAD workstations for a couple of years.
After that I got back into freelancing again, worked a bit with photography again, then I got into video/TV production, editing, and shifted into compositing. I didn’t touch 3D from 2000 onwards and I didn’t get into it again until 2012 when I had the idea to complement my compositing with doing my own FX work. Of course I quickly got sucked into working mostly with the FX stuff and at that point, in early 2015, I migrated from 3ds Max to Houdini.
990 A Skateboard sponsorship! Nice! I did the bmx flatland circuit for many years. I did ride boards for a bit. Learned hang tens. LOL 😉 What type of skating did you do? Vert? Street? Flat?
Johan All street, I could do a heelflip 180 down a 3’ ledge but I could barely do a 50-50 grind on a mini ramp. I lived in central Gothenburg during those years so I had a lot of great spots to skate within 5-10 minutes from where I lived… But I also rode BMX for years as well, way longer than I skated – and also street, mostly, though no flat ground.
Then I (of course) did a bit of street riding on inlines as well – I never got why you should keep to one thing, as seems to be the idea among most people into skate, BMX, inlines, scooter – or even Parkour – it’s all fun, it’s all the same mentality, the same type of challenges…
990 Do you follow the skate scene at all anymore? Any favorite riders current or past? Do you still go out and skate?
Johan Naeh, at 47 I don’t do any of that anymore, I’m just too old, fat and lazy… I got rid of my last BMX as late as 10 years ago. But I have a really nice set of mountain bike parts I move around on different frames…
… and right now I have them set up on one of few suspension frames I really like, the older generation Mongoose Teocali – it’s so friggin short, so fun to ride, it’s basically a big BMX with full suspension… 😀
990 So it hasn’t been that long since you moved over to Houdini from 3ds Max. What drew you to Houdini? What was the attraction?
Johan 3ds Max being complete crap was the main drive. So I was dead set to migrate away from 3ds Max and I had my sights set on Blender, and then I saw the promo for Houdini 13 and started playing around with it and felt right at home. With the release of Houdini 14 in early 2015, I migrated fully – as in uninstalling 3ds Max and not having touched even once since that day.
990 Were nodes a stumbling block for you when first learning Houdini?
Johan No, not only had I just worked a lot with Pflow and Thinking Particles in 3ds Max, I also had a fair bit of experience of Fusion and Nuke, so that wasn’t an issue at all – but I do remember how immensely unintuitive they first felt when I got into node based compositing workflows in Discrete Combustion (I think it was) back in the late ‘90s – it felt completely nuts and I really couldn’t wrap my head around how that would be better than layer based compositing… Well, I do now.
Node based workflows are overall far superior to anything else, from having that overview, being able to follow how the data flows through the network, too that easy access to any data at any point, how you can just branch that off and reuse it, or whatever you want to do – there’s just no comparison.
990 What was the hardest concept for you to grasp in Houdini, if any? Working between the different contexts always felt confusing. Even today, some node trees can get a bit crazy.
Johan Lots of stuff. L-systems are confusing as hell. I sometimes get into it and get to a point I feel I know what I’m doing, kind of, but then I get back into them after a month and feel like a complete moron again.
The “deeper” gaseous stuff (some microsolvers, messing with custom fields) is also not very easy to get your head around… I’m wildly impressed by the people who mess about with custom fields, like doing gas strain setups for elasticity, jiggle type stuff, the people who are using custom fields doing good fire, etc – that’s way beyond me at this point.
But funny enough, in 6 months I may have had an epiphany and think that’s all easy – because it’s all about concepts and what you “get” and in the end Houdini knowledge kan simply be separated into “I get it” and “I don’t get it”. And there’s a lot of stuff in Houdini I don’t get. Yet.
990 Do you think someone new to Houdini needs to have prior scripting/programming knowledge to really leverage Houdini’s power and flexibility?
Johan No. It helps, of course, but in Houdini almost all experience of anything before, helps. For me, getting into Houdini was all about a convergence of my prior experience and knowledge, where everything from having experience of coding, vector graphics, compositing, 3D…
…and even my experience restoring a couple of old Land Rover’s, were suddenly very useful. Having torn apart an old gear box and managed to put it back together, or just generally having experience of building stuff, gives you a way to look at things, a practical sense of sorts, that is immensely useful in Houdini. That mindset, -What do I have, where am I going, and what do I need to get from this point to that?
…having experience of building stuff, gives you a way to look at things, a practical sense of sorts, that is immensely useful in Houdini.
990 Like all applications, there is always something new to learn but it seems you are comfortable in all areas of Houdini. Besides the programming aspects of Houdini, is there perhaps an area you are not as familiar with or hope to dive deeper into to gain a better understanding?
Johan I’m comfortable handling the data, manipulating it – that’s pretty easy – and that power comes from how I work in SOPs and with SOP solvers in DOPs – but don’t confuse that with knowledge of Houdini, because it’s really not. I’m not at all well read up on FLIP and Pyro, as an example, and the reason is I don’t want to get sucked into tweaking.
And I think avoiding tweaking is important, learning Houdini, because tweaking is scene specific, it applies to the setup and scene you are working on now, and you have no idea if it will help you on your next setup , even doing something similar. So when I do RnD, when I know I can get the setup where I want it to with tweaking, I move on. I don’t spend time tweaking it to look good and post it, that is imo a waste of time.
Interestingly enough, the people who argue against me about the above are also the people who ask me how I have time to get so much done or how I have gotten so far as I have in Houdini in just 3 years – well, it’s in large part due to me not wasting time tweaking sh!t unnecessarily. 😉
990 If someone is looking to learn Houdini coming from another package, what advice would you give them in order to learn Houdini? Is there a particular area they should first concentrate on?
Johan SOPs. Just learn SOPs inside out, because if you know that, everything else becomes easy. You need a very custom DOP setup? You set it up in SOPs and you control it using SOP solvers in DOPs.
And get into VEX early. There’s no need to get in deep, but for the small stuff like creating attributes and such. Personally I use a lot of if statements in VEX, it’s just such a simple way to control things. But just overall, it’s something you will need to learn so there’s really no benefit trying to avoid it.
990 Where would you say is an area or areas SideFX can focus on to make it more approachable to artists?
Johan To be honest, I’m not sure that ever really applied to me in the first place, so that’s a question better answered by someone else – but I will instead take the opportunity to say what I think needs to be done with Houdini. And these are just my personal opinions…
- Animation and rigging should be updated and organized to simplify migration for Maya animators – because that’s the focus group.
- Animation rigging muscle and skin should get their own context. That would unclutter the object level and make more sense streamlining these tools into a specific context aimed at animators.
- 👆🏼this, but for hair/fur.
- Update Pyro in regard to fire – because there’s no way to do good fire unless you mess about with custom fields and what not.
And I don’t know how many people have sent me renders of bad fire proposing it looks good – I’m so sorry, but no, it doesn’t. Houdini just doesn’t do good fire out of the box, it looks like crap, even if a talented artist can shade and comp it to look better – but it’s a solver issue, so it needs to be sorted by SideFX.
Just learn SOPs inside out, because if you know that, everything else becomes easy.
990 Any feature requests you would like to see for the next iteration of Houdini?
Johan There’s a new cloth and soft bodies coming in H17, so that’s huge for me, especially as they are based on the PBD solver, the current grain solver, which I’m very familiar with – so that’s a change I want that IS coming.
But what I would like is continuity with the contexts in regard to rigging/animation/muscle and skin. There was good reason for keeping the rigging stuff on object level, way back, but as Houdini is organized today, having riggin on object level is mental – it just is. So SESI should absolutely reorganize that. Objects, lights, cameras, etc, are objects – bones, groom stuff, muscles, are NOT – or shouldn’t be, at least. That stuff needs to be shifted into their own contexts, for sure. Now some people may argue against this, but I don’t care because these people are dead wrong. The Farmfield is always right, especially when talking about himself in 3rd person. 😀
990 You are strictly freelance correct? Tell us about the Houdini freelance life? Do you just take on Houdini jobs or are open to doing other type of work such as comp, look-dev, etc?
Johan The bulk of my freelance work over the last decade has been doing illustrations and mograph type animations for industrial/scientific visualization connected to financing research in emerging technologies. That’s not much fun but it has meant that over the last 3-4 years I had a lot of time to study as well as way more leeway in choosing Houdini work that fits my taste and talents rather than panicking about needing a lot of income from that…
And I come from comping but I do very little of that, and even though I do a fair bit of 2D animations and such in Fusion, it’s not compositing work. As for lookdev, no, I do my very best to stay away from lighting/shading/rendering and instead focus on my TD stuff.
990 Are you using any of the new fancy third party render engine such as Octane or Redshift with Houdini?
Johan No, I don’t mess with lighting/shading/rendering at all. You have to choose your battles and each of those areas are such you really get in deep to be good at it. They are also way more about tweaking and I try to stay out of getting sucked into that. I’m good at what I do and I just don’t understand why I shouldn’t focus on that
But I am looking forward to what will happen in GPU rendering in regard to the RTX stuff from Nvidia now. I heard from a reliable source that there’s Octane test builds with 3-5x increase in performance with RTX and I heard other people talk about stuff with the tensor cores in the RTX cards, so I think that under 2019 and moving forward we’ll see a revolution in regard to GPU rendering…
990 Houdini has been around for a while. Definitely not the new kid on the block. Why do you think Houdini has become so popular as of late?
Johan The biggest reason is Autodesk sucking complete and utter ass – it really is that simple. So when SideFX got such a great response from Houdini 13 and then followed that up by introducing the Indie license at a crazy competitive price point, it was perfect timing for everyone that was just fed up with how bad 3ds Max had become and those pissed off Autodesk just killed Softimage… Add to that how SideFX are really proud of their product and really care about their end users – in that alone they offered something Autodesk obviously had left behind…
Then you have the release of Houdini Engine for the game engines and the idea of assets, that was also a brilliant move.
990 You have been sharing your crazy setups for a while now. They are such a great resource. Tell us a bit about why you decided to start sharing your scene files?
Johan Well, someone asked me how to do something and thinking about how to explain it I just felt it was just easier to build a demo setup than try to explain it. The problem you have with explaining anything is you are dependent on what previous knowledge the recipient have and there’s a tendency that you start explaining and then you realize you are talking “above them” and suddenly you have to start explaining concepts or start teaching math, etc, and it’s just not very efficient.
990 You also have a Patreon account where you share some of your setups. Tell us a bit about that.
Johan The idea with my Patreon was to shift part of my freelance income to Patreon so I could spend another couple of days a month on doing that. Right now the focus is to get my site up with a searchable listing on my shared scene files – which are to be way better organized inside with color coded, descriptively named, nodes, annotations, etc…
990 Any plans for long format Houdini tutorial content in the future?
Johan Oh yes, that is coming. I have now solved so I have someone that will do my editing and such so that will allow me to focus on the recordings and voice overs, then he’ll put it all together – so that will make a huge difference in getting more breakdowns and tutorials out – and also get some courses out… the first of which is going to cover my approach to doing growth and as these gifs hopefully illustrate, I have some pretty cool setups that will be included in that…
Samples From the Growth course
990 Where do you find inspiration these days?
Johan Finding inspiration is probably my least problem. I’m wildly over-associative, so I just get ideas from all over. When I set something up, I’ll have spinoff ideas that gets their own life, all the time, and what do is setup the base for the idea and then copy-paste that into its own Houdini scene and save it on the desktop for later. Then I’ll revisit those scene files as I have possibility, but when my desktop gets crowded, I just shuffle the oldest into folders, and there’s perhaps 100+ ones I haven’t revisited. I have a couple of “Misc. Houdini RnD”-folders on my disks filled with these things…
990 Is there a methodology you have for deciding if you are going to invest time in learning something new? After all, in this field you have to keep learning and evolving if you want to stay relevant.
Johan As I mentioned before, my methodology in regard to learning and evolving is to not get stuck wasting time on (final) tweaking – and also avoid wasting time on RnD on very specific types of setups/scenes as you have no clue if what you learn from it will ever be useful.
And this is an opinion I know annoys the sh!t out of some people, but yeah, I see final tweaking and rendering of RnD for FX as a colossal waste of time. Now, I get you need to have nice looking renders for your reel and such (though I personally don’t have a reel) but in the time people spend to finalize their last setup, I’ve done 10+ setups just trying out new stuff and as soon as I concluded I could get it to where I want it to, if I want to, I’ll just close them and move on learning new stuff – with the exception of the 1/25 or so setups I share on Patron/Vimeo that I have to at least make “presentable”…
990 Where do you see the industry going in the next few years?
Johan VR, AR, lightfields, AI, neural network based progress in many areas – a lot of things I’m in no way interested to get deeply involved in. This is the stuff that I see as my exit cue out of doing FX work.
I think all these technologies are very cool, and I really do look forward to watching immersive movies in 5-10 years, but I really have no interest of getting into this stuff on the technical side. I will instead shift my focus towards the maker-scene instead, go from building stuff in Houdini to building stuff in a workshop – so yeah, that’s the 5 year plan.
990 So what other professions would you see yourself in if you were to move on or maybe had chosen another path?
Johan I’m only a couple of years before I’m moving on, I usually say 3-5, though it of course depends on where I find myself down the line. But again, not getting into AR, VR, lightfields, etc, and the plan is shifting towards the maker-scene with a studio/workshop, maybe slightly biased towards metalwork, but overall pretty generic. I want to be able to work with everything from building tall bike frames to carving a wooden sword in white oak to 3D printing custom parts for drones…
990 You mention exiting the industry. I would hate to see you go but I have a great deal of friends that are finding other career options. One just became a nurse. What do you think of the current state of the industry?
Johan As someone who only does freelance work, thus really not “in the industry”, I’m not really in a place where I feel I’m the right guy to answer that, at least not from an artist perspective. But what I do see, from what I hear from other artists, etc, makes me think the issues in the industry in part is created by the VFX houses themselves. There seems to be an idea that you need to maximize the amount of hours you can invoice the client, and the client responds by pushing the price per hour down – which trickles down to the artists in regard to work hours and pay. There also seems to be a lot of waste – like people telling me how they worked months on 8 seconds that eventually was cut anyway – that’s insane.
990 Do you think it’s still a good career choice for people just jumping in and what do you think is the future of this industry?
Johan I don’t know. What’s a good career? I honestly don’t know what that means, it seems to me a remnant of an idea that might have made sense 50 years ago, but really doesn’t today – or at least not to me. Who knows how the world looks in 10 years, let along 25. How about 50? That’s impossible to predict. So if you don’t know how the world looks, the whole idea of having a career is out the window…
990 When I first started, being a jack of all trades was frowned upon. Today it seems you almost need to be one to move forward while still being real good at one thing. What are your thoughts on that?
Johan Well, as the idea of having a “career” is out the window, your best bet is learning how to solve problems, because that’s a skill in itself. Looking at any problem, breaking it down, coming up with an approach to solve it, taking it step by step, not being afraid to test new things along the way, etc… And the same is true for building stuff, that is in itself a skill unrelated to what you actually build.
Building stuff, figuring out problems, researching your way through setting something up – it really doesn’t matter if it’s building a complex fracturing setup in Houdini or building a sandrail for a Baja race, it’s about how you approach problem solving, it’s about a mindset. The practical skills is the easy part. I suspect for most people it would take less time to become a proficient TIG welder than truly understanding the L-systems SOP in Houdini..
The whole point of life for me is living in the moment and making the best of what you have and try your best to enjoy the journey.
990 I love that you mention tearing apart car engines and learning from it. I did a lot of car work in my early teens. Also bmx helped me develop a different mindset and the way I approach things. This is why I go through the process of banging my head against the table to try to make sense of things before searching for that specific tutorial. I find it helps me learn the best. Do you feel in this day and age of tutorial land this has become lost a bit? People relying more on recipes than learning to problem solve.
Johan Yeah, for sure – but I’m planning a TD course with this in mind, just get away from the normal tutorial style and get into how I personally approach problem solving – and going to try to keep it way more generic than to just Houdini, because I didn’t become good at Houdini doing stuff in Houdini, I became good at Houdini from doing a lot of other stuff, so learning Houdini to me was more about how Houdini in itself worked, memorizing stuff, but how I approach problems has nothing to do with Houdini, really, that’s a separate skill that can be applied to anything from carving a wooden sword to welding your own bike frames…
990 You are now working on Linux correct? Can you tell us a bit about your setup and what has been your experience with it so far? Pros and cons? ?
Johan I am not, I’m in Windows 10. I would want to be in Linux, and I have tried a couple of times, but there has just been a lot of small annoyances that has kept me primarily in Windows, stuff like gif recording where I had a lot of graphics glitches in the stuff I recorded to not having native audio support in DaVinci Resolve (sorted in v15, though), and such…
But Linux Mint 19 and the Mate desktop is really nice and I actually did a test boot just the other day and I had everything identified and loaded, even my Wacom tablet worked straight out of the box on a LiveCD boot – that’s pretty impressive. So we’ll see…If I get the time to test stuff out and if I can get everything running smoothly, I might just migrate back – because that extra 10-25% boost you get in Linux in the pro apps is very nice, of course…
990 What other software companies do you think are really pushing the boundaries besides SideFX 😉 ?
Johan I don’t know if they are pushing the boundaries, I guess they kinda do, but Blackmagic Design seems hellbent to give Adobe a run for its money – and seeing how amazingly bad Adobe has become from being so dominant, we really need that competition in the marketplace… In my opinion Adobe, Autodesk, should have been targeted by the US antitrust laws long ago, they are perfect examples on how market dominance is detrimental to the end users.
990 If you can go back and give your younger self some advice, what would that be?
Johan I wouldn’t. The whole point of life for me is living in the moment and making the best of what you have and try your best to enjoy the journey. Now, that’s of course way easier to do being born into an upper middle class family in Sweden than being born in many other places of the world, but having been this lucky, I really wouldn’t.
That being said, don’t think I haven’t had my fair share of bad luck, from having a Rottweiler take a chunk out of my leg at 9 to literally getting stabbed in the back twice by a crazy person going bezerk at a restaurant I worked extra at when I was 18, to bad dates, working myself to death in the early 2000s, clients not paying for work, etc… But all this stuff is what made me, me – and I do like me a lot. 😀
990 Any last piece of life or career advice you can give to those just starting out?
Johan Yeah. Most people get into this because it’s fun, but then end up miserable. That’s not right. So unless you’re enjoying what you do, it’s not worth it. Don’t be miserable at work just for the money. Personally, the one time I tried that myself, I burned out and it almost broke me. It took years to work myself out of that hole.
So now I’d rather take on smaller stuff, do work in the background, than take on stressful projects just to get some credit or to be able to point to it and say “look what I did”. My first responsibility is always to myself and my well being, because anything else would not only make me miserable, it would affect my work as well as everyone around me.
990 Can you give us a quick Houdini tip?
Johan I’ll give you a couple…
Learn SOPs inside out before getting too deep into DOPs. There’s a lot of power in prepping before DOPs and controlling DOPs with SOP solvers.
If you want to be a TD, unless it’s something you really want in your reel, don’t get stuck in tweaking stuff. If you know you can get your RnD to where you want it if you put in the time, just move on, learn something else.
Finally – and most important: Pay it back. Share what you learn. There’s imo no way to become a good Houdini artist without getting a lot of help from the Houdini community, so pay it back.
Thank you Johan for your time! Make sure you follow Johan on Twitter, and Vimeo. Also, show your support over on his Patreon and learn how he goes about creating his Houdini setups. Visit his website at farmfield-vfx.com.