A Better Design Process - Planning Like Pixar
Вставка
- Опубліковано 28 чер 2024
- Support Wintergatan:
- Patreon ► / wintergatan
- UA-cam membership ► bit.ly/4cQVM7C
Big Thanks to Louis Barclay for recommending this book on the Marble Machine engineering discord!
Thank you to all Wintergatan Patrons regardless if you are backing through youtube memberships or Patreon!
Marble Machine Engineering Discord Server:
/ discord
Video edited By Martin and Hannes from the Trainerds UA-cam Channel:
/ trainerds
-
WINTERGATAN RECORDS ► www.wintergatan.net/#/shop
SPOTIFY ► bit.ly/2oKxXWd
ITUNES ► apple.co/2ntWNsZ
COMMUNITY DISCORD ► / discord
-
Big Thanks to Louis Barclay for recommending this book on the Marble Machine engineering discord, Prophetic book recommendation :)
The shark not working and forcing the film to work around it built such amazing suspense and made for a much more satisfying reveal 🦈
And thank you for soliciting advice from Discord. It was a very enlightening conversation!
As a production technician, I will say that this is great news.
You have been doing quite a few of the necessary design steps already, with the testing of different gates and such, but it is fantastic to know that you have a better understanding of why this is imperative now.
Reiterating ideas thoroughly and testing things on the small scale before incorporating them on the full machine increases the chance of getting to the finish line without getting stuck in the "break and fix" loop _massively._
Another strategy to project management is called agile and it complements waterfall strategy but allow you to be able to adapt in the middle of a project one of the key factors is putting our consumers in the driver seat of what do they want out of this project ie a working machine that is ascetically beautiful. The other factor is allow all different types of people with different backgrounds to input into all the problems and not just their strengths. Love you Martin ❤
For an actual Method on how to execute System Design + Project Design, I recommend the book Righting Software by Juval Lowy. Yes, it is about software projects but the Method he describes is treating software development as an Engineering effort. You won't have issues finding the parallels to any project in real life: from requirements to design, volatility based decomposition, to detailed design, estimations, planning, tracking (earned value), costs, risk (unique method of quantifying risk!), valid options of building the project, construction.
glad to see Martin finally learning CAD (cardboard aided design)
93 Colin furze viewers :)
Bad Obsession Motorsport did it first!
The next step is MAD - Meccano Aided Design. Maybe not actual Meccano, but building a prototype out of parts that can be easily reused and reconfigured. Don't start by welding everything together!
It worked for skunk works with the SR71 and the F117 prototype. It will work here.
Long time sub. I understand some of the frustration viewers are expressing about planning to plan to plan. Many of us enjoy the blend of engineering, progress, and pure creativity that building the actual machine affords. I also understand your point of view as a creator. I am a software engineer working on a complex product and planning is essential in what I do. But I have to start with and ship MVPs or the company I’m working on goes out of business. Similarly, I have no doubt that movies like Up and Jaws had target release dates that couldn’t really be moved. There are real deadlines for these projects that become creative constraints and prevent them from getting stuck at any particular phase. Not having a consistent, realistic deadline for MM3 makes the project feel doomed to never-ending planning drudgery. I still believe in this project. I think the approach you’ve taken over the last few months (of validating core assumptions with quick prototypes) has been worthwhile. Now I think you take a second to clarify your vision like you described in this video and make sure you know where you are going. But then LET’S GET TO WORK. Don’t get stuck here for too long. Love ya and wish nothing but the best for this project.
to be honest, martin stands to make more money by making more and more of these videos than if he actually finished the project
@@MCasterAnd I don't think you have any way to actually know that. Sounds more like an old lady gossiping with nasty, unverified stories.
At this point it almost feels like procrastinating seeing Martin taking yet another look at how to design things in general. All critical technical questions have been answered and the requirements concerning size, playability and look have already been defined. Just make a handful of sketches, refine them, pick one, make a digital twin of it and start building already.
@@MCasterAndmartin doesnt care about money. Thats why he shut down the patreon for the mmx when he thought he had enough to finish the project
@@D3nn1sCheck the description of this video…seems he’s opened Patreon back up…
Here is the why or purpose of your project: Go watch your original marble machine video. It blew everyone's minds. It has this steam-punk atmosphere, this strange person steps up to this crazy machine, starts cranking it and boom this amazing whimsical music comes out. The only difference between the video and now is that instead of faking it with 50+ takes with the video camera, you want to recreate the same experience in real life in front of a live audience. End of story. That is your purpose. Am I right?
I would agree with you. But at the moment Martin seems to want a lot more than that. It's not just the same machine but working. Now the bass is to be self playing (both fretting and plucking), drums are to be full size, a whole new dimension added by introducing dynamics into all instruments. All those things increase the complexity of the project immensely. While the current prototyping approach is a much better way of creating a working machine, the problem is that with better approach comes growing appetite for new features.
@@piokul Martin is the true definition of feature creep
Martin wants so many features, that no one but him cares about. No one need a self playing bass, no one need 20 rhythms in one song and most importantly, no one needs 100% tight music. If I want tight music, I listen to electronic music ant not to a marble machine
@@LovisPlatz To be fair, he's not building it for you, or for anyone else really. You say that no one wants these things, but that's simply not true. It's easily falsifiable by just one person wanting it. The growing list of patreons suggests that there are already quite a few that do want it.
If we're working to the logic of not doing something that nobody wants, Martin would have just said 'who wants music played by marbles?' when the idea for the first marble machine came into his head, and it would never have been made. The individual instruments exist after all, and he is part of a band. What does any machine do that those people cannot?
It didn't matter if no one wanted it. Martin did. What makes the marble machine great for many people is the fact that he actually built it. It's a crazy idea with almost no practical purpose, but he built it, it worked, and it was awesome.
As for what you're going to listen to, tight vs marble - this simply highlights what Martin is trying to do here. For him, tightness is an element of good music. He wants to make good music. You should want to watch the machine, you should geek out about it's mechanisms and engineering. But first and foremost, it should produce music you would want to listen to *even if you couldn't see it being made* or *you didn't know how it was made*.
This all speaks to the 'why' of the project. Looked at another way, would the original marble machine have gone viral and come to be considered so successful if it had played 'Mary had a little lamb'? 'Happy birthday'? Or was it the fact that the song was awesome and original, and happened to be played in a unique way? If the latter is closer to the truth, then what does it say that the machine was built to play that one song well, and the end footage still needed to be cut together to get a complete song? It could, of course, be programmed for other songs, but much of the design was geared towards being able to achieve that song, and that song was constrained to what the machine could do.
To me, it says that the original machine is a beautiful thing, but it introduced constraints on his ability to make the music he wanted to make. He wants to flex his artistic muscles, and these requirements that make no sense to anyone else are the very aspects that he feels restricts his creative freedom.
So where does that leave us? To make you happy, he should compromise his artistic desire so that you can see a machine play music with limited expression? Just rest on having made the machine to show it was done, and go back to the full expression of manual playing? Just go make electronic music? I mean, why even do that? If we are just going to discard Martins' intent and ability to express himself, why not just remove him completely? Anyone can create music with AI, laying down new tracks faster than you can listen to them. But what is the point without the artistic intent? Why 'music' at all?
@@LovisPlatz Daft Punk asked Giogio Moroder to go into a vocal booth and speak about his life. In the studio were multiple microphones of various vintages from the 60s to today. When Moroder asked the engineer why they had so many mics, he replied that the mic they would use would depend on what decade of his life he was speaking about. When Moroder asked if anyone would know the difference, the engineer replied "They will know".
If you cannot understand the artist's point of view, that's on you, but the artist has a point.
Pixar planning is nice! But can it play tight music? ;P
Down to 1 femtosecond.
Do Pixar use Huygens drives?
@@thomasphillips885you can bet their Bowden are slicker than a slug
first it was tight music, then it was lookling good music instrument. He'll find a way I'm sure.
To be fair, your test driven aproach was, to a certain extent, a form of planning/prototyping. It was just more expensive and time consuming than cardboard
It was more of a R&D exercise than anything else. Now the real planning can start.
And much more accurate and also necessary
POCs for various bits that will likely end up in the final design is a very important part of planning. You can't plan something without understanding the pieces. The important thing is to realize that you are testing a concept, not designing your final product, and I think he got it right.
On one hand I understand the general vibe of frustration of annoyance in the comment section, but I’m at least glad that this has shown Martin that he can keep the form factor of MM and MMX. The full stage monstrosity scared me lol.
I'm concerned about his use of Bowden cables. I know he put out two videos about their function and the ability to play tight music, but there were 10 engineers who commented on it when he first announced it and they were all like "those cable suck. Trust me, I've been using them for 20 years."
I guess time will tell.
@@robertdascoli949 hey maybe in 20 years well have a video of martin trying a different solution!
I don't think that's what he said. He said the form factor wasn't the root problem, not that it wasn't a problem.
@@pvic6959 hahaha 💀
Has Martin become a marble in a machine? Stuck in a loop, taking a slightly different path some times, but always ending up in the same place again over and over?
As long as it doesn"t land on the floor...
the only escape from the endless loop is getting out by falling on the floor (death)
"One must imagine Sisyphus happy".
The real meta here is that Martin's series documenting this "planing to plan" work is an art peice unto itself, because it has got me thinking deep about all the ways I also hem and haw when tackling big projects. Maybe here on youtube, creators and fans alike are all just fish in an ocean of procrastination? 😂 I know the reason i'm here is that i'm procrastinating right now 😂
Thoughtful comment 👍🏻
true, at first it was marvelous to watch him just create and it was inspiring to see an artist making a sculpture. Then it became a lesson in perseverance and killing your darlings (pain is temporary glory is forever). Then it became a lesson in managing large projects by delegating tasks, making a planning, requirements, designs and acceptance criteria. After that, prototyping became way more important and using a approach by using digital designs. Now it is fully meta and looking onto his own processes in the last years, learning from which aspects caused certain results and going full circle by understanding why he started doing this in the first place. Now he has all the tools (both technical as his new mindset) to really 'start' over again (this whole process could be seen as not 3 MM's but the first 2 where needed to create the 'production' edition). This all shows how much know-how is needed to really follow up on such a big idea and what personality Martin has. This must have taught him so much about himself, because I am damn sure I learnt a lot about him.
This has been a commentary on following this man and his mission for over 5 years, really rooting for him and also using his experience in my own life as inspiration.
These videos ARE the art piece and the MM-whatever will mostly be the cherry on top.
I suspect that's actually the main reason he doesn't stop making the videos, because he wants people to learn from his mistakes without having to go through the same process.
At this point, if he wanted just money from content, he would've started getting some sponsorships like every single youtube content creator that does that for a living.
If he was just interested in finishing the project, he would've stop giving people platforms to be toxic against him (not to be confused with the valid criticism) and would focus on working, with no videos. Specially now that he FINALLY started accepting financial support through patreons and yt membership AGAIN.
So I suspect his main objective is indeed to inspire thought and self-reflection.
I'm also procrastinating some French homework right now.
Every time I think the project is moving along Martin finds a way to start over. I imagine Martin 80 years old saying “today is the first day of the project!” And his band mates all skeletons at the table waiting for him.
Wintergatan hasn't released an album in over a decade. I would have stopped waiting for him already if I were his bandmate.
@@Alpha99333 I always wondered what happened to them? Like are they in another band? Pursuing their own career? Or no musical career at all? Are they in stand-by and waiting for Martin's green light? Or will he find new band members after the project?
@@hundredfireify they're in other jobs, different bands, different careers.
This guy reads a new self help/productivity book every couple months and completely changes his mindset every time
I suspect this may be the one that does it for him, and I say it as a designer with two degrees myself. Notice how much more comprehensive and credible it was, with all the examples given and the validation it gave to all the feedback Martin got over the years from an entire community of designers and engineers alike… And sure, while it’s no substitute for a design or engineering course, Martin may make use of a little bible of sorts to give proper credibility to those concepts in his head and guide him through the process; it’s the next best thing and may just be enough to turn him into a competent, amateur designer.
Isn't that what they call evolution..
@@Mainyehc Yeah, compared to Web3-fueled autonomous DAO organizations, whatever those were, this seems pretty awesome. I found it useful for thinking about my current projects at my job and a fun video to watch.
The problem is that he *doesn't* actually ever change his mindset. He just starts using new terms to describe his process, which is basically random and whim-based. That's fine, actually, but his constant need to "solve" it like a puzzle leads him to the next in an endless series of self-help books and techniques to justify what he wants to do next once he realizes that what he's doing now isn't working.
It gives him the illusion of change and renewal without the burden of doing things any differently.
lol I agree but ultimately I would have to say it is more of a strength than a weakness
You don't need cardboard, you've already been using 3D printing for quick prototyping! It might take a little longer than cutting cardboard but when you need to test something with relative accuracy, cardboard won't cut it (pun intended)
Cardboard is useful for blocking out large components before trying to design an entire 3d printed assembly. It's also a lot cheaper.
I cant imagine the anxiety you must be feeling about failing, it shows through your procrastination
Thats a lot of words to just say you're starting again. Again.
How many design process concepts is that now? 10? 12? 15?
Are we now planning to plan a plan for a planned plan?
I thought we were at 1? He never really had one before.
You have been doing this. Where you make prototypes, test them, optimize, execute.
Honestly at this point you should just start a career as a management consultant. The only music you produce these days is audiobooks for micromanagers.
Ive had hope that one day Id get to see this live show but after this video ive lost almost all hope that it will ever happen. I feel like without a hard deadline to aim for martin will just keep restarting the project and it will never get finished.
I love this... Please remember the Name: THE MARBLE MACHINE!!! The marbles play the music. The story of the machine is how the marbles travel around the machine and produce the music.
Marble Machine: Martin is an artist
Marble Machine X: Martin is also a engineer
Marble Machine 3: Martin is also a project manager
also?
Marble Machine 62: Martin is a deep sea diver
Wait, what? Are we back at square one again? This has become a channel about not doing a project in as many ways as possible.
You kinda need to stop reading those books, they keep moving you backwards, every time you read one you get the idea that this single thing is going to solve your problem, that this magical solution is going to solve everything, and you fail to realize and continue a process, you can't move forward if you keep scrapping everything and going back...
Martin I love you, I want that dream of yours to be played in half the world, and then end its days on a museum dedicated to wintergatan with everything else you did.
But at this point I'm completely sure you just need to stop overthinking everything, you were on the right path, the mix of engineering and art, so you got 2 paths now if you really want to make it happen, go the hard engineering route and have a team of fans helping you design. OR stop overthinking it and start iterating, do a small machine of carboard make it work, do a bigger one of Wood make it work apply that to the true machine, youre going to end up with over 200 prototypes for the museum. BUT you gotta get out of the planning zone.
That's the point. To not finish.
Another video on how you are figuring things out? Let me guess, the next video is going to be an improvement on some part of the NEW marble machine.
lately i feel like every video is the first video of the project :D
Is it just me or has Martin said he read "How Big Things Get Done" like three separate times now
I think he missed the part where it says how to get things done.
All while getting approximately 0 big things done
it genuinely feels like weve been here before. i thought your new way of figuring out parts of the machine one by one was the culmination of this mindset. turns out i was wrong and you feel like youre only just at the beginning :D
I think both are important aspects which he figured out separately. While this larger planning process is helpful, I imagine the other part will be used with this new idea for a better result.
I think his previous videos were just determining if the machine was even worth building. Now he has to actually design the thing and integrate all of the knowledge he's built up over the last few months.
These Atomic Habits type books will be the death of the machine
The endless loop of perfectionism and insanity
I’m honestly entertained by it
No perfectionism seen yet
Not even close. What he's doing right now is the start of every successful engineering project. He's designing a complex system, that doesn't just happen by sitting down and doing it, you have to take it step by step.
There is something that I think you either realized already, or are close to realizing and that is: until you have a fully working prototype, you can’t begin in earnest with the final product. And there are a few things that follow from that.
For example, you have to accept that everything you build prior to the point where you approve of the final prototype is a prototype that is disposable, no matter how much time and effort you have sunk into it. In fact, what you think is the final version may provide you with lessons that will make it clear that it was also a prototype worth discarding.
If you had not gotten so committed that everything that you did for the Marble Machine X was going to be a part of the final product, and thus make sure to prototype more before fully commuting to an idea, I think the project would have succeeded. Or at least gotten a lot further.
I think that the change in perspective would have helped too. Assuming that you are further ahead in the process makes it feel like whenever something doesn’t work it is a failure. If you view what you are doing as exploratory makes it feel like a lesson learned at a stage where lessons are expected to be learned.
I am always surprised at the amount of haters in here comments, but the ones for this video seem to be a step above normal.
I have to assume they have never worked on a large project or that they didn’t learn anything from it if they have.
I work as a Software Engineer and every lesson that you have learned when it comes to achieving a large project is one that I have come across myself.
I don’t claim to be good at it, but everything that you have struggled with are things that I have seen even very large companies struggle with even in cases when the people in charge have the experience to know better.
Those things ARE difficult! But you are learning. And I would argue that you are learning the correct lessons.
Is because at this point he isnt commited to do an actual project. His channel now is doing content about how he thinks should do step 0 on a project.
I work as a mechatronics engineer and i have some experience with projects under my belt and one thing that experience has taught me is that the more time you spend planning the less is going to be made.
+7 years of building marble machines would point martin to taking such little time on design since he already should have in his mind what he wants for his machine, not only his endgoals (what he plans to do with it) but what hes actually making. He went from a "compact" one man machine to a multi-crew big ass machine with the intentions of every instrument having every type of way a note can be played on them.
The more he is planning this project the more it looks that nothing will be made because his designed product keeps getting bigger in size and will be logistically impossible to even use it without being filthy rich.
"If an MVP is not possible...." To me the MVP is a modular setup with a drive module and vibraphone module. Go play live music with it at small venues and see how it actually works in practice. Meanwhile develop new modules and slowly add them to your live performances. I could imagine that this may lead to a machine that doesn't have the playful visual appeal that Martin referred earlier to. Yet when it's finished you have been playing music for a few years and have been able iron out any flaws along the way. Then in 10 years from now you build the 4th Marnle machine that combines technology that has proven itself on stage with your wildest creativity.
Also consider why most software projects are not run in waterfall mode but in iterative development these days.
Yes yes yes, a thousand times yes. Fully agree with you!
As s software engineer of almost 4 decades I watched this with the horror of a slow motion impending train wreck. Waterfall only works when the requirements are very well understood at the outset, and don't substantially change during the development. The examples presented capture the error - they are architecture projects, where:
- Iterating is impractically expensive.
- Enormous experience on every minutia is already available from the literally billions of buildings we've built as a species in the past.
This leads to a situation where the unknowns are low, and the cost of not getting it right first time is nearly total. This is not generally true, most especially when trying to do an ambitious thing that hasn't been done before. Ironically the successes claimed in the book aren't in my opinion triumphs of planning, but *anti planning* - cheap iterations, feeling your way and a humble realisation that you only have a vague idea of where you want to be and what problems you'll encounter. The costs of successfully planning up front (and here I include the "digital twin" , ie. cad model) under those conditions are cripplingly high, and will suck all the joy out of the process, which is the real value of the end result anyway. The problem of how to make programmable music was solved long ago - use a digital tracker. What makes the marble machine wonderful is that it works at all, the joy is in how far from practical and sensible it is.
For the non-software engineers who argue that iterative approaches can't work for hardware, I present the object lesson of Blue Origin vs. SpaceX. Blue Origin started two years before SpaceX with plans for reusable orbital rockets, landing on ships, but have yet to fly an orbital rocket. Blue Origin's motto: "gradatim ferociter," or "step by step, ferociously", captures this idea of methodically planning, being sure at each step, getting it (hardware) right first time. In the mean time, by adopting the lessons learned from software processes, SpaceX have rapidly iterated hardware, failing early, failing often, optimising for low-consequence quick turnaround iterations, have soundly nailed the once thought impractical goal of reusable orbital rockets, flown the largest rocket ever built (by far) and are iterating that monster so quickly that there are already 4 future design iterations pretty much built ready to go.
Is this is the "No, no, no. THIS will be the design standard to rule over all the other 15 design standards", but you just end up with 16 design standards?
It's Martin. He always does this. Gets hyped about whatever selfhelp hustle culture book he's reading that gives him a great new philosophy on how to tackle this project. It will just lead to overcomplications in the project and we will get yet an other video telling us this instalment of the marble machine will not be finished. He needs to get out of his own head, write down a simple plan, and finish that plan including all flaws. He can still grow from there. But that won't happen.
Yep, but only until next week, when the 17th method comes in.
I feel like all of his videos are about some revolutionary new design approach that he uses for all of fifteen minutes.
@@NotLegato Martin: "No no THIS time will be different! Really!" Next week "How the ultimate penultimate super omega design standard I wasted a week studying will solve all my problems!"
Obligatory XKCD.
The road to hell is paved with supposedly bulletproof and magically effective management theory books. The road to success is paved with **discarded** supposedly bulletproof and magically effective management theory books.
praying that one day your soul will be delivered from design hell
He's not even speedrunning being an engineer at this point. It would have been faster to actually get a bachelor's in mechanical engineering and then get 4 years of job experience
That's usually how the "School of Hard Knocks" goes. It's actually the longer route.
Lol lol lol. So true. He could have finished a school, get some experience, then design it in a year or so.
The problem with making planning into something you want to cuddle with is that you might never leave the planning stage.... 😬😬😬😬 Sounds a little familiar
Have u even watched the video?
Hes already planning for a few years 😅
Yeah he's done this about a dozen times already.
@@coin777 maybe he was planning but he wasn't iterating on his plans, he was iterating on the machine itself
The problem is not the design process. The problem is, that you truly believe you found the solution in it. Like you did with the former solutions you found. :)
I doubt this would reach the ears of Martin, but I was just enjoying watching the epic video that started this all and I had this thought. Keep It Simple Stupid. Build the same machine, but make it work this time. Figure out what was wrong with MMX and fix it through your awesome design and planning processes. It needs to be 10 cm taller, make it taller. It shakes too much, so design a frame that will handle it! I think the marble lifting and the top dividers are workable and beautiful, they just need the proper design/engineering. The original concept is amazing and was surely a stroke of mad genius. Why not take all you've learned and make it work.
The problem is that the tinkering videos are fun videos to watch! 🎉
He'll go back to them once his metrics dip.
From now on he will be tinkering with CARDBOARD, so they aren't gone yet
Up was a successful movie because it was actually finished and the audience got to see a finished a product. Not continuous pushing the goalpost farther back every time with a new process. Either make the damn thing or don't
I need to be honest this time. Some of the comments are spot on, because I've always been supportive of MM3 and eager to see a working machine but however, now I think this has become an endless prototyping loop. I believe art has a factor that it takes its final form in the process, but you just need to start assembling at some point to see its birth. The machine will look good in its style anyway Martin builds it because it is a goddamn music playing automaton. However, it seems like we're stuck in the designing process.
Forget about cardboard, man. Instead, you can use plywood and make another MM1.
"This video is part 1 of the new marble machine project."
Come on... this is groundhog day at full force.
Martin, is this really about the machine? We haven't gotten any "finished" music from you in literal years...
Ok, so lets try asking why! If it would be for the sounds alone, a cheap keyboard would easily do the trick. But - the first marble machine did not only have the sounds forming music - it was the mixture of naive wood cutting with "standard" marbles rolling, the surprising control possibilities that were implemented and creatively used by Martin, all together in a rather compact machine. The playful looking mechanical thing helped Martin to do great music, that would normally have been played by 3 Persons. The success of the first video shows that there is a pretty surprising high demand for such a thing. The marble machine X was obviously built to make it even more playful looking (which worked great) more relieable (which also worked pretty good) and a little more versatile to make it an instrument that can play different music styles and rythms. That didn't work, as the aim for perfection grew out of hand (my opinion) and almost no music was ever developed for the machine -rightfully so, as the machine was never finished (until recently).
While I understand the quest for perfection, the word "good enough" also comes to my mind. Also, automatic playing of all instruments should not be the primary goal - every keyboard can do that better.
So what can be the primary goals for a stage usable marble machine? I can only give my personal Ideas: It should look like a useful instrument. I can't see the demand for a huge monster (your 3rd design) that takes hours to be transported and put together to then be just a stage that plays automatic music. Let's take a look at a typical on stage keyboard use: You pick a solo intro melody, let percussion step in, maybe divide it in two to have one hand change the key of the chords and the other hand plays the melody. So there is some Automation going on, but the main control is done live, by hand. Most Keyboards could have everything recorded and do great music parts without any human input as well, but no one would call that a Live concert. A guitar played by a human alongside would not really help to improve the boring and inhuman music. On the other hand - a DJ just mixes prerecorded musical instruments live - and done right he can do highly creative and successful music.
What is my conclusion? I'd say: stay small and humble. The vibraphone sound works great with marbles (with all the improvements you have found along the way) and looks like an instrument. If you can have some control over the melody and other human played instruments join in on stage, that can be a great and stunning addition to a live group of musicians. The percussion? Its reasonably small and versatile - its a nice mechanical addition as well and works great with marbles. The base guitar? Honestly? Not in the same machine. Its pretty boring if it works on its own an ergonomic nightmare if controlled by a human hand and mindblowing complex if it would do real musician like music. I'd say don't keep it - its nice to know that you could automate it as well, but it doesn't help the experience of the audience. Let alone if you have an actual human base player as well.
What else? The programming wheel should be very well visible and easily changeable with fixed meoldies on it. And a free programmable one, to do completely new music on the fly. You already had thought of that in the Marble machine X! A Video screen on top of the machine showing how the marble divider and the internal marble lanes do their work would be way more important for the whole audience experience, than a mighty complex machine, that can replace several human musicians. Little gimmicks like the wintergatan star turning for every marble is more helping, than a amchine that could pretend to be as mighty as a CD Player, if you install an electric motor and have complete automated playing instruments covering every aspect of the music. Prince did not get famous for being able to play all instruments by himself - he got famous for writing and singing great music! And performing it on stage with musicians that added to the whole experience!
So my answer would be: Concentrate your efforts and experience to build a great programmed marble Vibraphone in conjunction with highly versatile percussion instruments (like you had done on the marble machine X) and make it look compact and nice. The programming wheel could be under the main mechanics, as most concert hall stages are elevated and a screen on top would give a youtube like immersive view way better anyways - the side view that shows how the programming fins actually "control" the music that you hear. For the guitars, flutes, mouth harp - take them into your hands or the hands of other musicians. If you want to get more complex without cramping: Build a vibraphone that can be swapped with a wooden Xylophon to be played by marbles as well. And if you still want to make a one man show as well: Build your automatic base guitar as a second marble machine, that can be temporarily synchronized by a shaft. And than add a bell playing programmable machine and one for flute and so on. But not in a Single oversized overkill machine, that would not result in a great live instrument.
A lot of people will avoid this comment due to its size, but I think it's the most valuable comment on this video.
All of your suggestions are pratical, possible, and good advice. Which means he won't take it. He does not want to finish. MM3 looks more like a museum piece, and it would lose all of those 0ms tolerances in 1 or 2 concerts. It's pretty obvious too. He won't compromise on things that don't work well like the bass guitar either because MM2 would have actually gotten completed (Remember that around 20% of the time was spent trying to get the bass to work)
This comment right here should be sent by mail directly to Martin
“I have never ever been conscious about what design process I’ve been using”
You at one point had an entire team and project manager dedicated to planning, with a sheet that updated as you designed with real time data and percentages to completion.
There were multiple people on calls from all over the world working alongside you to help plan. Literal professionals who designed and implemented strategies to plan.
I’ve seen this reconceptualization influenced by a self help book so many times on this channel.
In fact, the last video you just released had you reusing the dragon slaying analogy and now it’s a fish in water to conceptualize but ultimately it’s the same sentiment. Could be a man rolling a stone up the hill, etc.
Long time watcher, but I just want to see a complete machine at the end of the day, and when I pop my head back in to check up and see a video like this, it’s just disheartening to see you in the same loop.
If you this seemed like you weren’t doing this out of fear or procrastination I wouldn’t have said anything but like many of the commenters here, the pattern is obvious unfortunately from an outside perspective.
The thing is, for people who have never made something large and complicated before, those steps LOOK like planning, but they're not planning. It's like making a list of navigation directions before deciding where you actually want to arrive. Planning (a.k.a. preproduction) happens BEFORE you make a huge list of components and start outsourcing design work. Just because production has been formalized (which makes it look a bit like pre-production) doesn't mean pre-production actually happened. It's a big pile of documents and meetings, yes, but without proper pre-production, it's still the same thing as just plowing ahead based in intuition
I feel like every new ”Heureka moment” forces Martin to take a step back and further away from building the machine. This is very close to where the planning turns only abstract.
Also, this 30 minute presentation is so detailed and well put together... I wonder how much time it took to plan and make all this?
I think this is a super edifying process for martin and all the viewers to go through, and while maybe it delays the final finishing of the machine, it will ultimately lead to a better final project. In the end, the human experience is more significant than the machine.
@@MTG_Musicwatching Martin preform art in front of me is why I started watching. The goal was a machine, the plan was to do. All the support to Martin on his personal adventures but it feels, to me, he is being pulled into a management & engineering world. Plan, plan, plan rather than the beautiful act of creating art. ✌️
It's by design.
I really think he's going to absolutely nail this after this video. If you've been involved in large projects through work or elsewhere, you will know he is right on the money.
He is setting himself up for success, I'm really excited for him.
While it’s probably ok to do this much planning at this stage, please set yourself a deadline for your MVPs. Doing so will force you to keep iterations short and acts as a creative constraint to grow from
I think what would really help is having a deadline. Movies like Up have deadlines, which leads to committing to decisions. I think that would be very useful for MM3 - otherwise this process can go on forever
THIS. In the time it has taken Martin so far in attempting to follow up the original marble machine, Pixar has turned out multiple feature length films from concept to in the theaters.
That would keep him from grifting and scrapping everything again in 5 years
Pixar have money incentives in making a final cut and sell that to movie theaters or streaming services. The MMX has the opposite problem, the adsense/patreon/sponsorships money is already rolling in pretty nicely, and it will keep rolling in as long as the youtube channel is producing content, so there is not an incentive to have a final working machine. Would Pixar ever release a movie if it had basically infinite funding?
Design processes are collaborative, whereas an artistic process is individualistic. This seems to be the hardest transition.
🎵 It's the circle of liiiiiife..... 🎵
I think at this point Martin needs to hire an engineering team and just serve as creative director.
This is a never ending cycle.
100%, he would likely love to work with a talented engineer if he'd accept it.
Unless he's willing to pay a good rate (60+/hr), a good engineering team would get frustrated with him pretty quickly.
Very good suggestion!
Pretty sure at some point someone was hired, back in France. No idea what happend to that dude.
Martin IS the engineer needed to realize his artistic vision. He just needs remedial classes. Don't worry, he is much closer then you fear. Keep an eye on the comments. When he finally makes a model that gets ZERO mechanics in the comments to wince... he will be ready to go!!!
Maybe just pause the whole project and fire up composer Martin... I LOVE Wintergatan music! Would love to go to a concert, also without a marble machine. You already built so many interesting machines
Yes!
As an engineer i LOVE this series. Seeing you work out your own design process and follow steps I had to come up with myself after being with a disaster of a design company just feels great. thanks for making awesome content about it, this is going to be an incredible resource to engineers and makers in the future.
Haven’t we seen this episode before?
Every six months or so.
Like Taylor Swift once sang, "I think I've seen this scene before, and I didn't like the ending".
I'm glad people are finally catching on to this.
"First time here?" meme
It's true. Mock ups and computer simulations are a fast, cheap and low-rise way to work out a physical end product. But, it's hard to gain access to things you don't know about. Sometimes you need a bit of failing to find out what you need to know and prepare.
I never comment on these videos, but this is absolutely a project where a Minimum Viable Prototype makes sense and will answer so many questions. I would avoid investing time and effort into digital simulations wherever possible.
Didn't he do two minimum viable prototypes already?
I mean, yes and no. He’s dealing a lot with physics, namely those bouncing marbles, and while a lot of those interactions must be physically prototyped, they could also be roughly simulated beforehand.
the way i understand "minimum viable prototype" is that the product is meant to be in its simplest and cheapest it can be. this project is definitely not that.
@@jacobzimmerman3492I don’t think so. At best, they can only be called POCs. Both of those machines validated that this concept is something people are interested in, but neither were not in a state to be moved to an actual stage for multiple songs. That was proved when the live recording on the MMX with original song was done in the museum (it was bad; loud noises, timing issues, missed notes). That was also after a lot of repairs and improvements were done.
The steps in the engineering he had done so far for MMY seem to be good: he’s battled the first principles, focused on timing, off-the-shelf products, quietness and the like.
An MVP **now** will be if he were to put it all together and not focus on form at the moment, but pure function. This will be to validate whether these modules work together well. He can keep iterating on that until he gets to a finished product.
He's done that like 5 times
I’ve been watching every video of him since 2 years ago. It never gets old every Wednesday ❤
I think GTA 6 will be out before the planning is finished. 😅
25:57 "What you don't see can be scarier than what you do see" is the same mentality Ridley Scott used when directing his 1979 movie, "Alien"
Hope this doen't translate to building suspense by never showing the mm3...
One day I see Martin building prototypes, trying and learning and experimenting solutions... Yeah, making some progress even in small steps.
And then there is a break and long talkings about some book and start again...
I just think about cardboard isn't good for any serious prototyping. Better you keep 3d printing instead.
Cardboard works very well for visual prototypes, but not so much for functional prototypes.
The cardboard was just an example
Exactly, he's moving backwards
23:30 "Planning should maximize experience."
You know what also maximizes experience? Variety, diversity and creativity.
I remember from when you started on the MMX, you did a lot of other videos in between, making other small and cool contraptions and making music with those. Your wall is a testament to that.
Recently though, you've been VERY focused on the MM3. Sure, you've been doing very different facets of that project, like creating prototypes, thinking of design requirements, imagining the machine layout and now tackling planning, but all those were for the MM3. What you're describing in this video is very important for the MM3, but you might also need more variety to experience more different things and get more creative juices flowing. A thing you learn from those other smalller projects might be applicable to the MM3 too and help you realize that project.
This is a mental illness
"I really had this epifany about what this project should be and how important the playfulness about the project is. The Marble Machine is not a functional project it's an artistic sculpture" - Martin, March 2024.
The collective cheer that went up could be heard for miles.
Within four weeks we're back here.
Tight music is not what the MM is about. It's about the joyful amateur adventure of discovery.
Gotta be honest, this is getting exhausting. I think I am going to take a step away from this channel for a while until something is actually built. And thats really sad because I have been following this band for like 10 years at this point. Good lord, nothing is getting done. Just dump the project and do something else if you dont want to make it, but at this point I feel like I am chasing a carrot on a string.
I can relate to you sooo well! I've been watching this from before the first machine was finished and I just went back to these videos these days. And tbh I was shocked how different the whole spirit was back then which was what I loved so much about this project.
Back then it was Martin as a part of wintergatan. A cool swedish band with a very unique sound that I loved. And they were building all kinds of funky custom music intruments like the modulin, the music box thingy and then the marble machine.
Now it feels like he's trapped in an endless circle of wannabe project management and engineering without having even the slightest idea of why he is doing all that or what he wants. It just hurts to watch by this point and I have to leave this for a while now too :/
martin is trying to solve every problem that he might face rather than finding what doesn’t work along the way
Welcome to Design Hell. The marble machine will never tour.
@@StylishHobo MM3 is so huge that it would never tour. Its intentional.
Martin, in the recent time you are always hopping from design process to another one, as if you've found the holy grail now and know what to do - only for some video later to come again at that point. It doesn't really matter so much which design process is the best, because there are many out there and everybody has its reason why it came into existence. There is no design process out there to rule them all, keep that on mind! What matters most is that you are realising your dream machine the way you want it, stick with a well known design process and just do it!
Working more on the design process than the actual designing at this point
There's money to be made by prolonging the problem.
@@thomasbecker9676I can only give you one like.
@@mrbfox1775 That's alright, I appreciate the sentiment.
@@thomasbecker9676 So you think that martin has some meta-plan for making this project longer, just because he makes money with it? Just want to know if you think he is intentionally doing that for the reason of profit?
@@lorddenti958I think it’s not intentional but rather a side-thing. When he wants to make a new video he probably asks himself, what does the marble machine need improving in and thus ends in another cycle of chasing minor details.
The biggest challenge to planning is discovery. Planning can only anticipate known unknowns, and unknown unknowns will almost always be discovered throughout the entire design process. Consider that time is one of the most valuable resources that you are paying, and planning needs to keep the time cost as low as the materials cost.
I'm kind of in awe of the fact that you're basically reinventing the wheel here. The things you're learning the hard way are all things I learned in my 5 years of Mechanical Engineering College education.
What is he reinventing here ? he is sharing his progress in the field of project plannification wich is a science in itself.
@@guillaumevincent716 The Engineering Design Process, torque loads, ways to transfer and maintain energy, machine design, tolerances, why CAD has become so commonplace of a tool, designing for manufacturing, etc.
These engineering / build videos sure are entertaining, but do you know what I keep coming back to week after week? Your music Martin, the original marble machine, the full concert and Victoriateatern, the random mash up during MMX stream and many more.
WOW, that was awesome. Should be in everybodies playlist who does start or is in a project.
It’s not that I don’t like these types of videos and would prefer more testing…
It’s that Martin seems to be cyclical where he gets in his own way from actual working on the project.
That being said; my wife and I plan on going to the closest show with the marble machine - whenever that happens…
Martin may be the modern day Sisyphus.
Im in for the journey. Long time follower and love watching you learn and learning with you.
Martin,
As much as it is painful to admit, I'm very glad to hear you express your opportunity cost of not composing music for the past 8 years. I've had that same thought. I am a music lover and engineer. I was immediately drawn to your content from the moment I first watched the original marble machine video. I've latched onto every single that you've released in the past 8 years (regrettably, it's not been many 😭 and Sandviken Stradivarius is a favorite) and have wished that you would take a little more time to keep making music. I would take delays in the next marble machine if it meant getting more music from you!
I love to see your journey of self-discovery and betterment. I know it can feel like you're just spinning your wheels or going in circles, but if you evaluate where you are today to where you were last year or 10 years ago, you can see that these circles you're going in are actually a spiral and you are elevating yourself at every round.
I'm familiar with the feeling that you're jumping from one solution to the next and that, at every jump, you feel so passionate and confident that THIS one will be the answer. Do not be discouraged if you find that this stops being motivating in a couple weeks and you seek another one. If you haven't looked into it, I can confidently recommend studying up on adult ADHD. It's been an eye-opening experience for me and my life will never be the same (for the better). It's even made my past much better as I've been able to see my life experiences from a new perspective of understanding.
Keep up the good work. I'm glad to support your work. Very glad you can bring Hannes back on.
Martin, I've been watching your videos since the first marble machine. While you were working on the second marble machine it was awesome. It was logical step from first machine to perfect it. And then you abandoned the second machine while it was 99% finished. All that years watching you create it, it went from building mega hype greatest marble machine, to literally nothing, at least to us viewers. You didn't even record a nice single song like it was with first marble machine. You just gave up, and then you started to plan another machine that will also take years (it already does), and you even want to scale it up that machine will pretty much became entire stage . If you think that you can travel with that around the world to make concerts, then good luck. Stuff like this breaks easily in transport, not to mention what logistical nightmare that would be, and how much spare parts you will need to carry with it. Not to mention setting that thing up before every concert.
Dude, you are talented musician that created amazing music with far, faaaar less. And you are also a great engineer, it is awesome to see how you design mechanics of machine and how you test it, but you should really consider the weight (literally) of the machine that you plan to build. Machine you plan to build should not only bring joy to your fans, but also to you. Don't put too much weight on your shoulders. Keep it simple and keep it awesome, or otherwise you will end up as just another youtube machinist, creating nice little videos about mechanics and mechanical design, always aiming for perfection but actually never achieving the main goal. We both know that you are far far more than that. Don't get entangled in machinist megalomania.
I hope I didn't offend you with my comment, but that's how I see this. Good luck Martin, and I hope that you will make great machine and more great music!
I think he knows its not feasable and is just grifting. It would be made up of over 1000 parts, maybe 200 or 300 moving parts with stupidly tight tolerances to get 0ms variance as well (just the simple variance in the gravitaitonal field makes this impossible for travel)
Not to mention grime, dust, and damage from transport. And this one is HUGE so it would have a million issues in the sections coming together perfectly.
@@BaronCreel why can't he ? I get that YOU wont care anymore at some point, but is chasing a dream inherently bad, wrong, criminal ? Even if it never finishes ? You dont get what being an artist means, nor will you stay around to see if it works out or not, just move along, noone owes you anything, especially not that guy you've seen on youtube and that you never met, talked to or even understood through what was shown to you.
If he never finishes anything but explores his creativity and has a life full of wonders and engineering problems, so be it, that's what he chose for himself. Why the F would you want to imply it's not "productive" enough ?
Do you even understand why humans create anything ?
Feels like the movie "Groundhog Day". We are forever stuck in an endless planning loop. Sure hope Martin finds a way out of it and gets back to building.
The Sydney Opera House may be a success on the outside but the inside, away from the theatres, is a real challenge. Lots of little rooms and corridors that don't flow and a heritage listing meaning that they can't drill holes in the walls for electrical and computer cabling.
Martin this really reminds me from the early planning stages you did when you finally started thinking about the MMX and I think you are taking your time to solve all the problems as well as better yourself and I really enjoy watching this journey and am here for all of it thanks.
Bunch of nice words... written in ivory tower.
Hey Martin, just wanted to share that our company has adopted the working backwards model, inspired by Amazon. It’s really changing the way we approach projects and problem-solving!
One of my go-tos as a project manager.
I finally realized that I must finish the beginning of the thought process that will lead me to conceive the beginning of the idea, of a title for this clip.
I'm always amazed how my job as a software engineer overlaps so much with the one of a "regular" engineer. Next time I'm struggling to convince a customer that more planning is necessary, I will show him this video!
Just build it, man...
I would love if you could make the marble machine as quickly as possible to go on a world tour, but I would rather wait 10 years for a machine that lives up to your true vision than see a rushed project fail again. I have been watching since the original marble machine music video and I've tuned in every week, and I know you will make the machine that everyone will love.
This is my favorite of your videos so far! I can't wait to watch your design process unfold in this way. It reminds me of the fun I used to have in these stages during my career. I just loved it :D
Here is another design process for you: build something that's good enough and works and then improve it along the way.
And don't listen to morons on Discord.
Or UA-cam.
That's what he did with MMX
@@Snaketown1337 Untill he trashed it and started over in pursuit of "perfection" which is a never ending process....
@@nuneke0i think because he made it too over-complicated. If you watch the scrapping video of MMX the thing basically kept jamming and break down even if you dont look the machine at the right angle. He pushed for a lot of questionable design decisions and it backfired big time.
If he cut the bs and start doing something that actually works and then add his aesthetic choices safely it would be a different story.
Loving the references to Solo Leveling. It's a good comparison. Can't wait to see what comes next!
Thank you, someone noticed lol.
planning this device has always been getting done on this channel, and you have dedicated your life to creating this damned machine. you are a creative thats always changing the story like a writer or story teller. there are so many problems that you see in your work that only you see as the creator, and you will never be satisfied. at one point, you've gotta travel the world with a machine and know where to go after that. I would not be surprised if you've got another idea floating around in your head that your dying to get started on but cant because this machine is not letting you move on with your creative life.
This came at the perfect time for me. I've been feeling a little frustrated because my spring projects keep getting pushed back because spring refuses to spring, so I feel that I'm stuck in the planning phase. After watching this, I feel more optimistic now because planning is cheap, and more planning time means being more prepared and should mean more chances of having a good outcome.
Hey martin, I have being following your channel for a while now and I love how you get in detail into the planning and the "less sexy" part of engineering. I am myself a french engineering student and I have only now made the connection between your videos and another youtube channel called "Benjamin workshop". Benjamin is an engineer and he made a gt40 sports car from scratch in his garage. Today he is in between the detailed design (cad) and production design of his hypercar project. All of this to say I think the reason he is able to tackle such complex projects mostly by himself is because he mastered what you are currently learning. So I think if you can understand french or if subtitles are available you should really go take a look at the first videos of the hypercar project to see how a professional engineer approaches such a project. Sorry for the lengthy comment I got carried away. Thanks for providing us such detailed videos of your learning process and may this project be a success.
You are never going to finish. You have too much enjoyment constantly seeking, never achieving. the idea of perfection is a trap. There is nothing this side of the grave that will ever be perfect.
The light goes on! Best design practices are SOOO important. Most of a project is completed by contemplation and planning. Even if that means looking at a blank wall and just thinking.
Once a good, solid design is formed, bringing the project into reality is the reletively easy part.
I'm a graphic designer, and I'm so glad to see this research into planning.
Yes, a lot of people are saying "Just get started" - and a lot of these people don't understand the design process.
But you're gaining that understanding, and with every video your structure for the project gets more concrete.
You HAVE to plan before you start building, if you want the pieces you build to work together.
And the people in the comments will probably get more frustrated when you start doing iterations. Work on a design for two weeks, say "It isn't working", and go back to the drawing board.
"Just finish it!" They'll say. But you will know that it will take more time and resources to make a broken version work than to start a new iteration that addresses the questions properly.
Personally, I'm excited to see what different forms these iterations take. Which ideas won't work? Which ideas will kind of work but don't quite make the cut? These are just as interesting and educational as the final design.
Starting over with these prototypes means you get to avoid starting over with the entire machine. This new planning structure for the third Marble Machine means you won't need to build a fourth. And that is more valuable than many people in these comments realise.
never ending story
I cried at the introduction to this video
Every week is the first week of the mm3
glad that you are figuring it out!
i am very happy to see this project being worked on
Another week, another wasted overthinking and overanalysing every single thing.
You've been doing nothing but planning for the last 2 years, since canceling the wildly successful MMX. The problem is TOO MUCH planning.
CAD-CAM in "old school " speak meant "Cardboard Aided Design- Cardboard Aided Manufacture "
Worked really well in steel sheet.
You could go from a cardboard prototype, reverse origami it into a pattern and turn it into metal within a day.
Our fantasy was to have a material that you could cut with scissors and put together with a material like masking tape. You would then spray "Component "B" on it and it would transmute into something remarkably like mild steel with a zinc coating.
That’s really inspiring. I love you new style of videos where you explain the process itself, mainly because you seems so much happier and stress-free than last year. Take care of you @Wintergatan that the best gift you can give us, even better than a world tour.