So Long Autodesk Shotgun | Hacker News Reader

So Long Autodesk Shotgun

thejackjam.medium.com

92 pointsaprdm32 comments

> The high-end visual effects industry that we dominated is notoriously slow to adapt to change, and operates for the most part on razor-thin margins. There were billions flowing into film and episodic production every year, but there wasn’t (and likely still isn’t) an obvious opportunity to further monetise our existing customers.

The visual effects industry operates fundamentally on not paying people. It is symptomatic of Hollywood broadly. It is absurd that the savings of 60 and 70 year olds are, via the children in the industry they support, subsidizing Disney.

I'm not sure I understand why you say it's about not paying people? VFX jobs are usually reasonably well paying and usually with good overtime rules.

The visual effects industry is structured around a bidding process that cements VFX studio compensation before film production has begun. However, film is a collaborative medium with financial and production risks with decisions often pushed to a later date "to be solved in post". That post production VFX stage has grown in size, grown in the types of production solutions provided, all without additions or modifications of compensation. They simple get to keep the ongoing contract, regardless if doing so bankrupts the VFX studio. Which it often does. The production digital artists receive their contractual compensation, they get to keep their jobs, but the production becomes 14 hour days, 6.5 days a week for 6 to 9 months. Meanwhile, this is taking place on all 6-12 productions flowing though the VFX studio at the same time. At some unknown time, the bank and investors simple say "no more" and the studio implodes. This is the concrete reality of the VFX industry. This is how people do not get paid.

That seems like a huge jump from the "VFX industry being based on not paying people"

Yes, there are some bad business practices that lead to companies going out of business. There are lots of successful VFX studios too though. That's true of any business.

I'm unsure how this means that VFX in specific is based around not paying people.

There are tons of VFX houses that have been the receipent or the giver of Trump-esque renegotiations. At the beginning, everyone agrees to terms. At the end, something somewhere happens which means no more money (if the money actually existed to begin with), and people get screwed. Most VFX houses I've been in/around have paid on hourly rates, so the artists get paid. If the producer bringing the work to the VFX reneges, then the VFX house eats it (or sues, or something). However, if your a contract artist and did the work expecting to be paid at end of project, you tend to be left hold the bag since the VFX studio didn't get paid then the contractors don't get paid.

So in the last decade, how many VFX studios have gone under leaving people holding the unpaid bag? It's been a long time since DD 2.0 and R&H.

I'd hardly take the edge case of the industry (not exclusive to VFX) and say that it's representative of the norm.

> There are lots of successful VFX studios

For relatively short time periods. Only Weta and ILM are able to get equity in their projects, and that is largely due to their owners being the filmmakers themselves. In the end, it is a work for hire industry, with no equity participation, yet carrying significant risk that tends to sooner or later blow up at least once, taking the entire studio with it.

That's a completely different goal post now though from saying the industry is based on not paying people. The majority of film roles, even outside of VFX, has no equity. Equity is the exception not the norm.

Other VFX studios like Imageworks have been doing pretty well without equity.

I don't believe Weta or ILM get any equity in their projects though. At least, the majority of their projects are work for hire.

Either way, this is a another jump from the initial statement that the VFX industry is based on not paying people.

If I have the market power through mono- or oligopsony to force my vendors to sign fixed-price contracts for variable-scope work, I guarantee you they aren’t getting paid for a lot of that scope, whether that comes out of their bottom line or their employees’ lives. Hypothetical me is going to improve my own bottom line by pushing scope (“just fix it in post!”) to the part of the budget that’s contractually-guaranteed to stay fixed.

Contract negotiations have changed quite a bit since/due to the R&H bankruptcy thankfully. Still not great, but the power dynamic has become a little more balanced in terms of accounting for variable scope work.

It’s not the norm, but this type of stuff happens more in this industry (labor of love) than in other industries.

Read up on the sausage party scandal

https://variety.com/2016/film/news/sausage-party-animators-u... overtime-1201838425/amp/

The animators were even reluctant to raise objections over fears of being blacklisted.

I definitely acknowledge it happens. I guess this goes back to the initial post saying that the VFX industry is based around on not paying people.

That implies norm. I know there's been several issues throughout the years at various studios, but those have largely been exceptions not the norm thankfully.

Perhaps the correct line is the VFX industry is based around paying studios less than the expense of doing the job, after the job balloons post contract negotiations and post VFX vendor's ability to re-negotiate. More words, but same net effect.

I didn't know that Sausage Party was a movie :

https://en.m.wikipedia.org/wiki/Sausage_Party

the Working Conditions section covers the controversy over pay and conditions.

edit the article you refer to without any Google AMP nonsense (sorry for not catching that before) https://variety.com/2016/film/news/sausage-party-animators-u...

Not to be dismissive, but how does that differ from any other consulting/work for hire job?

Most projects in embedded devices are exactly the same.

- a price tag is fixed by negotiation around estimates

- more and more stuff is pushed to firmware and software

- everyone works way more than they should to deliver in time

I think it's different because the embedded systems clients are usually not as deadline driven as the movie industry are. You have deliverables basically every week that you need to show to the client for approval, and, if not good enough, then re-iterate.

Embedded systems usually the scope doesn't change nearly as much after the SoW, where movie is very creative/subjective and it can change every single week.

I worked in both industries for ~5 years, in VFX now

The article doesn't mention what the product is for. Even the sales page for it is rather vague. It's some kind of project management for visual projects. But how much of the job does it do? Does it do revision control and asset management, like git or Alienbrain? If not, it has to connect to some system that does, so it can find whatever someone wants to preview.

A product of this type is that it has to interface to a lot of stuff and present some kind of unified interface. That's hard.

I contract a lot of work to build manage and build tools that directly integrate into Shotgun/ShotGrid. At it's most basic, Shotgun/ShotGrid is just a really nice front end to a database that has collections of schemas that align with the task workflows at studios that produce content for TV series, movies and video games.

Asset management is not as built in but there is some infrastructure there that can be leveraged.

I think a reason it doesn’t as well is because the majority of the industry already uses it. Therefore as he says there isn’t too many opportunities for growth, it’s a small industry vfx all things considered.

> The high-end visual effects industry that we dominated is notoriously slow to adapt to change

this is patently false. They were the first to jump onboard the GPU rendering bandwagon, and have been pushing cuda since it was first usable (there is of course a debate about how usable it is..)

The issue with shotgun is that its really fucking powerful, but also hilariously configurable. Shotgun is there to track each shot as it goes through every stage of production. Think of it as jira, but far more integrated. Before shotgun was a thing(and f-track) people used access, or filemaker. or worse Excel.

> and operates for the most part on razor-thin margins. There were billions flowing into film and episodic production every year, but there wasn’t (and likely still isn’t) an obvious opportunity to further monetise our existing customers.

This is true. Unless something makes a significant impact, its out the door. Shotgun is/was expensive per seat, but allows you to cut down loads on time management. You need less production staff to run a lighting/animation/comping department. However thats a fine line. A production assistant is £25-35k, but per seat licenses of some software is £1-5k.

Shotgun when it first came out was a massive behemoth. at Framestore it was running on the biggest machine we had (from memory it was a quad opteron with 64gigs of ram[might have been less]) It was ruby on rails, slow unreliable and generally a pain, BUT, it totally sped up production for desperaux.

A lot of this sounds very familiar. I was working on the in-house tools doing the same thing for a very big VFX house from 2012-2018. Early on in that job, we only really had one pair of tools for production management - one for the project planning (a big Gantt chart) and the other for timesheeting. There was nothing really in place for the artists to use to know what was going on and as such their willingness to enter timesheets was pretty low. In order to drive awareness, reduce work for coordinators and generally try to help everyone, I was made lead to bootstrap a new app in 2014 called... ShotGrid. It did help to a degree, it was still mainly only used by coordinators, but helped them do their work slightly more efficiently. We went down that rabbit hole for a while, but still, the artists lacked good tools for connecting with their PMs. Fast forward to 2016/2017 and my team was sectioned off to completely redesign and rebuild the entire project management tool suite, with the help of some external contractors. Years went by as we did the research, design, technical investigations and review. It was a genuine real struggle to get anyone in the company to engage with us and talk about what they wanted in the product; they'd much rather just struggle on with the existing tools (including endless Excel sheets) and turn their focus to actually doing VFX instead. I left the company in 2018 amid a lot of frustration about nothing being progressed. I heard about 9 months later that the entire project had collapsed and I think they were then heading towards implementing ShotGun instead. Which now, ironically, is apparently to be renamed ShotGrid. I'm taking that as a huge coincidence, but it's nice to know that some of the success I had in that industry lives on in name alone.

> The problem though, was that no-one on the team actually cared about growth. “Get people home in time for dinner” had to become something more like “get even more people home at some time”.

Once you see this pattern you cannot stop seeing it. It shows up in startups, in the video game and entertainment industries, and apparently used to show up a little bit at Autodesk, at least until the author quit. The people with non-monetary drives are the easiest to exploit. It makes perfect sense to a business. If someone is will to do more work for the same or less money it's a no-brainer to encourage that behavior. The end result is a bunch of people willing to put up with substandard pay and poor working conditions. Run from those industries. It never pays off. If a product can't be made outside of these conditions, maybe it doesn't deserve to exist. Survival of the fittest products and all.

In general, this applies to most industries in the US these days. It's a reflection of the lack of value that employers see in employees at present.

That's fair. I don't know too much about other industries. In software, if you're using passion as a hiring indicator, you're contributing to this pattern whether you want to or not.

Yeah, well, what Autodesk gobbles up, Autdesk turns into excrement...

There were bits and pieces of familiar experience there that anyone working long time on a consumer software product had, but honestly I find it hard to sympathize with this sort of "I wish them well, but all goes down now that I left" articles - especially naming and framing the company, product, people (indirectly in this case). There are hundreds of nuanced reasons why decisions were made like this within a huge company like Autodesk, things the author does not know about, and we as readers even less. There are very, very few straight up success stories in this industry, all achievements are "successful failures", processes where individual members see things fall apart or not improving compared to a personally important point in time from their PoV - and for themselves, this is the truth, even if the product/company - an abstract thing, for them - achieves a form of success.

Is that the division of Autodesk that was behind the ruthless ceiling tile attack on Ton Roosendaal?

https://www.youtube.com/watch?v=eJwG-qt-sgk

I’m so relieved that Altium showed the middle finger to Autodesk…

Every time you see the trope about a startup inside a big company: run.

Big companies don't look at startups as healthy seedlings to nurture to maturity, they see them as nitrogen-fixing cover crops to plow into depleted fields.

That is brilliant, and extendable. Plowing in some random nitrogen-fixer allows them to keep all their industrial machinery in place, and make no fundamental changes. Switching to learning how to grow seedlings is an entirely different frame of mind. One is generative and nurturing, where each seedling has potential to be something big and important, and thus matters; the other is extractive and exploitative, where the seedlings are just biomass for the machine.

Really lovely metaphor.

I appreciate your kind words.

Reply on news.ycombinator.com