Hope you enjoy this robot as much as I do 🙂. A big thanks to everyone who helps to make these projects possible through their patreon support. If you'd like to help me make more projects like this (and get more behind the scenes content, an AWESOME discord, livestreams, etc) check out patreon.com/stuffmadehere
It's weird that I started watching Stuff Made Here, then I found your channel, started watching you, and now that I come back to Stuff Made Here you're here, too.
2 tips for wood: 1. Use a carbide tipped chain 2. File the guards down. Lots of people that don’t do it for a living don’t realize you need to file the guards every 3 or so filings of the teeth otherwise they block the cut. The more you file, the more the teeth bite. Since you have a robot with lots of strength holding the saw, you can file them down way more than a human could hold back and have it cut well at all angles.
@@Oroborus710 I got that bit but the tool parth might need some adjustment after looking at quite a few freehand chainsaw milling videos over the past 3 years Me and my father have completely changed our technique with the saw. Less is more to optimise a saws cutting ability you want to cut with as few teeth as possible this let’s each tooth do the most work possible. Not slow the saw by making lots of teeth all work at once!!! This can be quiet dangerous if the saw is held by a man but (safety first always watch for kickback) but I don’t see that problem with a robot!
You are SEVERELY overestimating how strong a robot arm is lol, that particular robot can only lift 13 lbs. If you look closely you don't see any massive hulking motors sticking off that robot or anything like that because the motors are so small they fit inside the arms and they don't even really have insane gearing ratios either because robots are quite fast, usually a 2:1 or 3:1 reduction with a small belt and sprockets is very standard. Its very unlikely he will ever get a robot that small to be able to cut wood with a human chainsaw.
With the amount of work you put into these videos, I can't believe you're able to do more than one a year! I'm a full time UA-camr and I'm in awe of effort that goes into these. 3D modeling, CNC, 3D Printing, Robotics engineering, more math than I've ever even heard of, welding, electrical, software engineering, etc., etc. You combine about 40 full time professions into every video! And I love the self deprecating humor too :) Keep up the amazing work!
I agree, it's unbalivable, the knowledge neeeded for this and the time in which he does this stuff, i mean he posts a video every month with a compltly crazy ideia that would take an expert like an year to reproduce!
Everything about his journeys is so relatable as an engineer. Except that he can write a 3D path planning algorithm in his spare time with a baby in a week. Friggin genius.
@@mgalese what doesn't mean, that this is also a slicer. my understanding is, that a slicer works layer for layer and has an relative simple strategy, while a cam has a way more complexe approach((as long as it should give a good result)
I'm a PhD student. This guy could honestly write academic papers in several different fields if he wanted to. Just listening to him describe optimization algorithms and the trade-offs behind each really shows a level of intuition that inspires me to keep going whenever my code isn't working or my results aren't making sense. He doesn't stop trying until it works. That never giving up mindset is what inspires me more than anything else.
Computer scientists explaining gradient ascend: "imagine a ball on a curved surface" Mr.s stuff explaining gradient ascend: *actually have a ball on a curved surface*
Wow, this guy is crazy smart. I’m a developer myself, that doesn’t just take a couple of minutes to program. But he also knows how to metalwork, engineer, teach what he’s doing, make you laugh, and edit a very high quality video. 👏🏼
The amount of energy you're able to put inside these projects and videos is absolutely amazing. So many creative details in the video. And all the stuff that needs to happen off-camera to get the projects done. So far watched each of your videos at least two times, Just fantastic!
I work as an engineer in a well-respected global company. I solve many complex projects for employers or customers. But watching every each of your videos makes me feels like a school grader tinker. And the storytelling of this video is at the NatGeo level (or even better). You are the model I would like my kid looks up to.
raise the bar then man. maybe a 4-6 motor dc brushless dronecraft with a chainsaw on the front of it. with radio or ir tracking that ominously stalks people
I'm just about finishing a year long dissertation that's working toward using a robotic arm on a rail for ship welding. I know of at least 2 PhD students and a professor working on the same project. You've covered a significant proportion of the work this team has done over a few years within a couple weeks. My mind is blown
As a software developer I’m gonna start applying this mindset to my work, it’s actually a great way to think which will probably make you more resilient
As a mechanical design engineer with 10 years in product development, this guy is an absolute beast. Makes me realize I need to step my game up. He would make newton, tesla, franklin etc proud the way he applies the scientific method. Mad respect.
Agreed. Makes me wish I didn't have dyscalculia. It really gets in the way of my ADHD hyper-focus on problem solving from being productive... By which I mean I have a BA in political science / sociology instead of something useful like engineering.
He should just use the Chinese method of problem solving. Step one: Find a company or person who already found a solution to this problem. Step two: Hack said company or person and steal their solution. Step three: Use the stolen solution and pass it off as your own. Step four: Deny that you stole any solution or that you're killing off an ethnic minority living in your country. This is similar to the Russian method of problem solving. Except you poison all of your political rivals instead of killing off ethnic minorities. Secretly you're doing that too but no one yet.
@@Jonathan.D Lol you are such a weirdo wtf. A video from funny engineering man doing a ridiculous and complex project and you decide to turn it into a tirade about China that even Mike Pompeo would think is a bit too much. Go outside or something lmao
@@shepardice3775 It's called satirical comedy. I'm sorry don't get it and can't understand humor. There was no tirade in it. However, it seems that you might be a triggered snowflake. Take a chill pill snowflake.
@@shepardice3775 I think most will laugh at what he wrote because it's funny. It's not a tirade. What you wrote is a tirade. Why don't you go outside and find a sense of humor! Don't call people weird when it's you who doesn't understand what is being said. Do you always criticize what people say when you don't have a clue about what they're talking about? Either you are a troll or like the other guy said you are a snowflake. Do us all a favor and keep your comments to yourself.
What I love is how every seemingly impossible project ends with him saying how much he enjoyed a process made-up largely of false-starts, dead-ends and frustration. In a world where everyone wants easy success, instant wealth and a life free from problems, perhaps that’s a lesson we could all benefit from. It’s all about finding pleasure in what you do, overcoming adversity and enjoying the journey.
Adversity is most rewarding when it is chosen. We're still very far from being able to grant that to everyone. He enjoys the work precisely because there's no real stakes to cause anxiety and desperation.
It’s always been a dream of mine to one day own a house where I could drill holes into the floor when I need to make things work. Never mind all the other amazing stuff you did over this project… just based on the drilling, you’re my new hero. Awesome work 👌
Once i was changing tires ant the bolt was VERY stuck,we pulled out a 2m bar and still couldnt get it out,because we were lifting the car 😂😂@@jamesMcroy12
I watched this video 7 months ago, but when it popped up again I couldn’t resist. You are clearly a genius on many levels, but the most impressive thing to me is how humble you are. Your content is very entertaining and I always do a little happy dance when I see you uploaded a new video!
I've spent a good amount of time chainsawing, and I can say that that little chain on that little saw would get dull very quickly, especially if it's always going in tip first, even with softwood. I'd guess that's why it got so hot. A longer chain with a built in cutting oil squirter on the arm seems like it would be crucial. That being said, awesome video as usual. The sheer perseverance your projects take is nuts.
I am doing a really hard programming assignment right now and the quote "There are only a finite number of ways you can screw up" has really motivated me :)
You’ll probably never see this, but I think you’re a really talented teacher. The way you explain things is brilliantly understandable and enjoyable to learn from you. It’s makes you videos better than 99.9% of what’s in this digital universe 😌
I'm a 30-year mechanical and electrical/software engineer and I continue to be amazed by your breadth of knowledge and ability. Thank you for showing the world what a knowledge of math and science and fun can bring to our lives.
@Peter Evans If all the men asked only the old ladies about the expected results of possible ideas, then we would all still live in caves, because we would know that trying to make an ax out of the stone we are holding in our hands is nonsense. The old lady said so.
@Peter Evans Sorry. Most new ideas are by "old ladies" understood as "non-working" paths. So it is better to toss coins than ask your "old lady". Better results.
@Peter Evans Are you being deliberately foolish? Any old lady reading this, should StuffMadeHere have used a Metabo or a chainsaw to make this? Hardwood or Styrofoam? Stepper motors or servos? Should he have used an armature or field windings? I can only trust them to know the difference between Chick Peas and Garbanzo beans.
Yes, so many things combined with a true sense of sculpting - certain elements of that dog's face are subtle (for a chainsaw). Would people think he's a nerd? Not with that cute sweetie of a wife standing next to him.
Yeah I'm just continuously baffled by the insane shit he builds and comes up with. Studying to become and engineer and I just hope that I will be able to understand and learn even just one part of all he does
He's also kind of stupid tho. He never tested whether he's actually able to cut wood before going through the trouble of writing software that automatically computes a tool path. But yes, he's an impressive polymath and his CNC parts look pretty neat.
Dude... remember he is doing this full time and it doesn't matter how long it takes him and NO ONE is yelling at him when he gets it wrong. Those two things make his work ... nothing to do with real world engineering. You have been trained to just be proficient not creative. Creativity is a multiple disciplinarian practice.
This is mind boggling. Do you get any sleep at all? It’s hard to imagine the amount of work on the engineering side of things, plus there’s the whole youtuber/editor side of it on top of it. Simply amazing.
I'm a trained mechanical engineer and I am compensated handsomely. Having said that, in order to sleep better at night i tell myself that you are not just one guy who has this enormous breadth of knowledge but that there is a collective of at least 7 different people behind this channel doing all the work and you just pretend that it was all you. Please don't destroy that illusion for me. And i hope my boss never stumbles across this channel and suddenly expects me to code and weld and video edit and built robots in a month
I know that feeling!!, I'm a software engineer, and he just says "I would have to write a lot more code" just after saying that he can't implement the previous algorithm, as if he would be stoping that plan due to his inability to write it fast, correct and optimized... which is what he ended doing. I really tell myself, those 7 people MUST be real.
Im a safety engineer. Even tho my job dont need me to build stuff or coding anything, but it need me to understand what kind of machine is this and WHY YOU SHOULDN'T PLAY WITH IT.
Can I just give Tormach a additional shoutout for supporting you on this and enabling you to test this new piece of machinery? I've gained so much respect for them as a result of this ploy, even though I know it's marketing and user testing, it just speaks heavily to their company.
I swear the engineers at Tormach are fans of Shane and got the marketing department hella drunk so they'd do this :D But tbh Shane is the best kind of beta tester. His projects are so far outside the norm that if the robot has any weaknesses they will come out. I've done beta testing for video games, and a part of it is trying really stupid ideas or really weird speedrunning tricks that might screw with the code. Basically if a speedrunner searching for new tricks can't glitch the game even by manipulating code, a normal gamer is really unlikely to ever fall through the floor. Same principle applies here: if the robot arm can survive Shane's project, a normal customer will never run into a problem.
The most valuable part of SMH videos is by far the fact that he shows us is approach to the problem, fails and then takes a new approach. Great content!
I think it's the only way to trust he didn't faked it. Because you could also spend an entire month describe hard coded all the instructions to carve this exact sculpture. That would be easier, just the program would be stuck with one specific shape, for one specific machine, in very specific conditions. He proved that it is from an imported file, which means once he debugs, he can carve anything. Which is absolutly mind blowing.
I'm going to leave a reply actually relevant to the comment, cuz nobody else is. Yeah, A coworker of mine once snapped a socket wrenches' ratchet doing that.
Smart != being able to do everything He’s clearly very smart. But some other smart person still knows more about this particular software problem than he does.
Best part is that the solution is to just take multiple guesses at the initial parameters and run the ascent simulation simultaneously; and for best results, having the simulations and their results linked and iterative. Its called a MCMC (Markov-Chain Monte Carlo) algorithm, and usually doesn't take too much time.
because in this channel he tried doing many different projects instead of just one, he realized that there are people who are specialized for certain project, who knows way better
I’m an engineering student, and I watch these videos imagining the day I can make projects this good, but the further I get in my degree the more I realize this guy is just a wizard.
He does this as a job so he has way more time and money to do this stuff, and every tool under the moon so don't feel too bad. But he still is very impressive.
@@willsk3122 The most important thing that he has is experience. You could have all the tools, money and time under the sun, if you don't know what you're doing and how to troubleshoot when something goes wrong, you're pretty much screwed.
@@chrism45 I think he was in 3D printer product development, so very familiar with cnc coding, manufacturing and the product lifecycle. That plus a passion for DIY and craft will get you 75% of the way there. The rest is, I believe, dark magic
The set of ways to screw it up might be finite, but there's no guarantee that it's small enough to assure that you can correctly identify when you're screwing it up the same way multiple times.
Yep. There's also another popular corollary to this statement: An expert is someone who has made all of the possible mistakes in a particular, very narrow field.
I'm already worried for the next video he going to post. "I figured that if my robotic chainsaw arm could carve a dog sculpture, why couldn't it give me a haircut?"
He is freaking amazing. So glad he shares all this with the world. Not only can he make a living, but he can share his wealth of knowledge with the younger generations. A lot of his projects are stupid and have no practically BUT damn it really shows what goes into design and build
I dont work in a similar field to him, but i also feel like i learn by watching him slove problems. Hes got a very intelligent process and a great way of explaining thing.
You have some of the best content on the platform, and are one of the few creators that I can just start the video and be absolutely enraptured for the full run of the video.
As a software engineer I’m amazed how the author is converting “I have spent a week on this algorithm with no success” into the coolest gradient descent “animation” I have seen! Amazing video!:)
As an engineer and hobby youtuber I know how many hours it took you to make this video, but I still can't believe you actually did it :D That is some serious coding right there!
Ok how they make this is really insane you need to now what you are doing its not just cut you need to put proper speed it is not a saw were you can rip it fast as you want and thats not a proper saw there is no gaurd you need a guard a gaurd is these teeth in the back of the chain the chain will pull foward sow it blocks its wood gaurd you need a person to run that saw not a robot that why it smoked
Check out how to sharpen a chainsaw. The “anti-kickback” tabs that you referred to are depth gauges. You need to file them down a little in order for the teeth to cut. Should work great after a proper sharpen!
"There is a finite number of ways to screw things up so you will eventually finish" is possibly the best and most truthful inspirational quote I've heard so far
It sounds good but it's not true at all. Just because you did something wrong 10 time before you did it right diesnt mean there aren't 10 more ways to do it wrong. You just quit looking.
My first thought: That looks like a good book to read. Stopped here to comment, looked up the book, and haven't even watched the rest, yet (I promise I will as soon as I'm done looking)
honestly man what i love about you and your channel is that you make me and a lot of people actually interested in engineering, you make younger audiance into engineering once again you're one of the few rare people who make engineering fun dreamy and intresting the way you mix AI, Robotics and software engineering all together to build so much intresting things is amazing you're one of the few who make kids dream about becoming engineers not tiktokers and "influencers" once again anyways man i love your content keep it up ❤
"I'm almost out of things I can screw up, so I'm feeling pretty good about this one" - Engineer, right before screwing up about 500 things he never could've even imagined
i used to be a part of a team that built wall painting robots. the robot had a "suicide" button that when flipped, it just came down from the wall like a rock. the guys working with the robot complained all the time about it.
Oh damn, you mentioned a programming issue that I think I could actually help with for once! For the "false summit" issue you're having, you can add "momentum" to the next location guess by summing the vertical distances (or just the normalized gradient if you're working in a space where you can calculate that) of a set of the previous moves and then adding that (as a fraction between 0-1 * a multiplier/'step size' constant) to the next movement. This way, once you reach a peak, you will go past it and then your height change will go negative, so it will cause it to go back towards the top. This will make it oscillate around the max/vertex (or min if you want to go concave), then eventually settle at the point. HOWEVER, it will allow the tool to move past peaks and instead of oscillate back, if it hits another region with a positive gradient before the momentum runs out, it will start climbing back up that. It's like rolling a ball down a bumpy hill (but inverted in our case), the momentum allows it to keep rolling a certain distance (depending on the step size) back up a hill once it rolls down into a valley. if there's another valley on the opposite side of this hill, the momentum might allow the ball to roll through the first valley (the 'local' minimum), up the next hill, and down into a lower valley (eventually the 'global' minimum). So we can just flip gravity by looking for increases in gradient rather than decreases as you would in that example and we can find the highest peak this way. Of course we don't really want the chainsaw having to oscillate around so much so all of these checks can be done in advance and then used to instruct the next move. If you or anyone want to look into this, it's a common way of improving artificial neutral networks where the error surface is mapped in a 3D vector space and the weight change calculation (which we use to minimize the error) can be made by using "momentum" to avoid finding local minimum errors (i.e. not the most accurate way to train the neural net) and find global minima. So just search up something like "ANN momentum for finding minimum MSE" (where ANN is artificial neural network and MSE is mean-squared error, fyi) or just message me and I can share my notes :)
MuchADrewAboutNothin it’s just slower in the long run. If you have to sample from a large surface, picking random points in a region and testing will find the peak much slower than making a guess, then using the gradient of points around it to “walk” towards the peak
Hope you enjoy this robot as much as I do 🙂. A big thanks to everyone who helps to make these projects possible through their patreon support. If you'd like to help me make more projects like this (and get more behind the scenes content, an AWESOME discord, livestreams, etc) check out patreon.com/stuffmadehere
❤️
Where lpl
Reeeeee
Hi
I want to meet you how
You know you've made a good robot when it has a "kill radius"
You know you've almost created the perfect robot when it has a "Kill Mode"
It's weird that I started watching Stuff Made Here, then I found your channel, started watching you, and now that I come back to Stuff Made Here you're here, too.
"Stay back. I will kill you." - so relatable, my robotic friend.
Bender would 100% love this
Glad a poltergeist can’t get into that,
Would be bad
Best video yet. Great job buddy.
I feel like you guys could be best buds
I’m surprised I’m the second person to comment :o
also didn’t expect to see so many checkmarks here
Yes it's a great one.
Big brain
The two nicest and smartest guys on the internet!
Thats the way: " Why buy one when I can do a much worse one for free."
Yo integza keep up the great vedios
Tomato lord junior is behind you
Not for free, for thousands dollars more!
@@treasuretotrash2067 lol
Ain't that the truth!
2 tips for wood:
1. Use a carbide tipped chain
2. File the guards down. Lots of people that don’t do it for a living don’t realize you need to file the guards every 3 or so filings of the teeth otherwise they block the cut. The more you file, the more the teeth bite. Since you have a robot with lots of strength holding the saw, you can file them down way more than a human could hold back and have it cut well at all angles.
filing down the rakers allows the teeth to remove more material at a time which is dependent on the power of the chainsaw.
Use the tip of the saw
@@gadgetg9479 If you paid attention and/or watched the whole video, you'd know that the guards are precisely why he can't cut with the tip.
@@Oroborus710 I got that bit but the tool parth might need some adjustment after looking at quite a few freehand chainsaw milling videos over the past 3 years Me and my father have completely changed our technique with the saw.
Less is more to optimise a saws cutting ability you want to cut with as few teeth as possible this let’s each tooth do the most work possible. Not slow the saw by making lots of teeth all work at once!!! This can be quiet dangerous if the saw is held by a man but (safety first always watch for kickback) but I don’t see that problem with a robot!
You are SEVERELY overestimating how strong a robot arm is lol, that particular robot can only lift 13 lbs. If you look closely you don't see any massive hulking motors sticking off that robot or anything like that because the motors are so small they fit inside the arms and they don't even really have insane gearing ratios either because robots are quite fast, usually a 2:1 or 3:1 reduction with a small belt and sprockets is very standard. Its very unlikely he will ever get a robot that small to be able to cut wood with a human chainsaw.
I love how his third screen always has something fun to show off
But why ARE wiener dogs so long?
With the amount of work you put into these videos, I can't believe you're able to do more than one a year! I'm a full time UA-camr and I'm in awe of effort that goes into these. 3D modeling, CNC, 3D Printing, Robotics engineering, more math than I've ever even heard of, welding, electrical, software engineering, etc., etc. You combine about 40 full time professions into every video! And I love the self deprecating humor too :) Keep up the amazing work!
I agree, it's unbalivable, the knowledge neeeded for this and the time in which he does this stuff, i mean he posts a video every month with a compltly crazy ideia that would take an expert like an year to reproduce!
Don't forget the editing and cinematography!
Well said. This channel is so underrated. He has a brain big enough for 10 adults. LOL
You could add standup, teaching & video editing to the list of skills as well
@@frankierzucekjr I think its less do with the brain and more with perseverance and the ability to stick to a project from start to finish
"I optimized it so hard that it didn't actually compute anything" legend
Everything about his journeys is so relatable as an engineer.
Except that he can write a 3D path planning algorithm in his spare time with a baby in a week. Friggin genius.
Been there, done that, burnt the t-shirts.
Lazy Man's Optimization - if you're working hard it's not optimized hard enough yet!
Its faster! no one said it had to do something...
thats a quote for the ages
This channel is the epitome of "I could spend 2 weeks manually doing this, but I think I will spend 2 months automating it to showcase my brilliance."
With 59 days of failing tho. I think he is amazing to never give up
@@Etrehumain123 100% of human existence is trial and error....
One man doing the work of an entire team of engineers... As an engineering student, this guy is my inspiration and role model
If this guy teaches you ONE thing, it is to keep at it and not give up easily. Bravo!
Inspired to make enough money to be like this guy one man team
Idk mann this guy is either blessed or he worked hard sooo muchhhhhh !!😭 He deserves much respect + he's good at memeing and sarcasm ! God damn it
@@ameenurrahmankhan6933 both obviously
Anyday.
"Fortunately I know a guy with a tree" God I wish I had these kinds of connections
That might be too narrow, start with looking for a guy with wood and go from there :P
I know a tree
Man that’s funny 😆
aaand now he doesn't
Have you tried having a bear instead i heard their good with their knowledge about trees
“There’s a finite amount of ways to fail so eventually you’ll get it” best quote ever
I feel underestimated
there's actually an infinite amount of failure types, tho. that's just wishful thinking.
Its actually quite motivating lmfao
Time stamp?
Seriously gave me hope for life
Bro I love this man. He made a useless wood part just to explain a concept to us to make sure we were on the same page.
This guy just made an entire slicer software for a highly advanced robot system. That is incredible. Well done
@Umbrella Corporation whoop whoop wins all round
it's not a slicer software but cam
Yep. He used to work at Formlabs writing many, many slicing algorithms for 3D printers.
@@mgalese what doesn't mean, that this is also a slicer. my understanding is, that a slicer works layer for layer and has an relative simple strategy, while a cam has a way more complexe approach((as long as it should give a good result)
Then rates it as a fun project
"there is a finite way to screw things up, so you will eventually finish" is great life advice, honestly.
Mr. Stuff: "There is a finite way to screw things up, so you will eventually finish."
My dumb, incompetent ass: *Hold my beer, son*
You underestimate the big O notation of my capacity for failure
"there is a finite way to screw things up". Ever seen re;zero episode 15? They may be finite, but not any less painful
But not necessarily true... in general there are uncountably infinite ways to screw things up!
@@ryanmccampbell7 many, but not infinite, if you put monkeys in front of typewriters they'll eventually produce shakespeare, or whatever.
I'm a PhD student. This guy could honestly write academic papers in several different fields if he wanted to. Just listening to him describe optimization algorithms and the trade-offs behind each really shows a level of intuition that inspires me to keep going whenever my code isn't working or my results aren't making sense. He doesn't stop trying until it works. That never giving up mindset is what inspires me more than anything else.
Same! I really want to have the same perseverance that he has with my project or my code
Me: Is slamming head against table every 30 mins
@@nikkiofthevalley have you tried slamming the table against your head?
@@shadowlord0162 lol
He has... He also has a bunch of patents to his name. He headed up/ran Formlabs entire R&D department for like 5 years.
Computer scientists explaining gradient ascend: "imagine a ball on a curved surface"
Mr.s stuff explaining gradient ascend: *actually have a ball on a curved surface*
Being able to physically represent your explanation is always great for teaching.
what? how does this make sense
@@BloodChampagne At 9:23 he built a physical demo of gradient ascend
Wow, this guy is crazy smart. I’m a developer myself, that doesn’t just take a couple of minutes to program. But he also knows how to metalwork, engineer, teach what he’s doing, make you laugh, and edit a very high quality video. 👏🏼
Yet I would not trust him to cut down a Christmas Tree. 🤣😂
Yeah, this video makes me feel really stupid 🙃
He’s straight up stupid smart
Hes a talented man
He is mostly very motivated.I am in engineering but i am so lazy . He makes me feel lazy atleast lol
As a Robotic Design Engineer, this project impresses, frightens, and concerns me. Great work!
does it inspire you to do something awesome? like.. evil genius awesome? 😈
We asking the good questions here😈
"This is one of those ideas that's so bad it wraps back around to good." Buddy, that's your entire channel. It's great.
+ most of my decisions
The Maryland point of robots
“I almost run out of way to screw it up” Is my life in a quote.
The amount of energy you're able to put inside these projects and videos is absolutely amazing. So many creative details in the video. And all the stuff that needs to happen off-camera to get the projects done. So far watched each of your videos at least two times, Just fantastic!
100% agree
Oh hi Marius!
I'm shocked, shocked I yell you, to learn someone like Marius enjoys SMH videos.
James is an
He is truly a monster tho.
I mean he has an ENORMOUS amount of skills just to make these things work
"There is a finite number of ways to screw it up!" The ultimate in optimism!
right - because there is more like an infinite number of ways to screw anything up :D
"And I'm going to be the one to invent all of them! :D"
my life: are you sure about that?
I work as an engineer in a well-respected global company. I solve many complex projects for employers or customers. But watching every each of your videos makes me feels like a school grader tinker.
And the storytelling of this video is at the NatGeo level (or even better).
You are the model I would like my kid looks up to.
Feeling the engineering to the core !
I work as a well respected global company in an engineer.
raise the bar then man. maybe a 4-6 motor dc brushless dronecraft with a chainsaw on the front of it. with radio or ir tracking that ominously stalks people
@@loginavoidence12 Ah yes, wish I'd thought of that
"If it's so hard, why does this dog have a crown on?"
You are a brave man, asking that question.
Because the wife said so. No other reason needed
"There is a finite number of ways to screw it up, so you will eventually finish"
That's oddly inspiring and optimistic
This dude is full of quotes, the entire comment section is people quoting him
Unfortunately, they have tried to calculate pi, also. =)
oddly enough, its the absolute truth.
hidden in the demoralization and pessimism of failures
That assumes that you learn from your mistakes... When has that been shown to be the case?
I'm just about finishing a year long dissertation that's working toward using a robotic arm on a rail for ship welding. I know of at least 2 PhD students and a professor working on the same project. You've covered a significant proportion of the work this team has done over a few years within a couple weeks. My mind is blown
"There is a finite ways you can screw it up, so you will eventually finish" is surprisingly motivational.
As a software developer I’m gonna start applying this mindset to my work, it’s actually a great way to think which will probably make you more resilient
As a mechanical design engineer with 10 years in product development, this guy is an absolute beast. Makes me realize I need to step my game up. He would make newton, tesla, franklin etc proud the way he applies the scientific method. Mad respect.
His ability to explain extremely difficult formulas and theories to children or just dumb adults like myself is incredible
That's a sign of intelligence. He's very intelligent.
Agreed. Makes me wish I didn't have dyscalculia. It really gets in the way of my ADHD hyper-focus on problem solving from being productive... By which I mean I have a BA in political science / sociology instead of something useful like engineering.
"Dude is smarter than me" -Destin from SmarterEveryDay
You guys understand him?
Right? He makes me feel a lot smarter than I actually am
"There's a finite number of ways to screw it up, so you'll eventually finish" is probably one of the best things I heard in my life. xD
That is the core of the R&D engineer's mind, right there.
He should just use the Chinese method of problem solving. Step one: Find a company or person who already found a solution to this problem. Step two: Hack said company or person and steal their solution. Step three: Use the stolen solution and pass it off as your own. Step four: Deny that you stole any solution or that you're killing off an ethnic minority living in your country.
This is similar to the Russian method of problem solving. Except you poison all of your political rivals instead of killing off ethnic minorities. Secretly you're doing that too but no one yet.
@@Jonathan.D Lol you are such a weirdo wtf. A video from funny engineering man doing a ridiculous and complex project and you decide to turn it into a tirade about China that even Mike Pompeo would think is a bit too much. Go outside or something lmao
@@shepardice3775 It's called satirical comedy. I'm sorry don't get it and can't understand humor. There was no tirade in it. However, it seems that you might be a triggered snowflake. Take a chill pill snowflake.
@@shepardice3775 I think most will laugh at what he wrote because it's funny. It's not a tirade. What you wrote is a tirade. Why don't you go outside and find a sense of humor! Don't call people weird when it's you who doesn't understand what is being said. Do you always criticize what people say when you don't have a clue about what they're talking about? Either you are a troll or like the other guy said you are a snowflake. Do us all a favor and keep your comments to yourself.
What I love is how every seemingly impossible project ends with him saying how much he enjoyed a process made-up largely of false-starts, dead-ends and frustration. In a world where everyone wants easy success, instant wealth and a life free from problems, perhaps that’s a lesson we could all benefit from. It’s all about finding pleasure in what you do, overcoming adversity and enjoying the journey.
Do not pray for easy lives. Pray to be stronger men.
Life is difficult so happiness must come from enjoying adversity!
@@AcMrPro @BravoNineThreeTwo hes trying to carve a bear, chill.
@@AcMrPro Praying is kind of an oxy-moron in that phrase that I never noticed
Adversity is most rewarding when it is chosen. We're still very far from being able to grant that to everyone. He enjoys the work precisely because there's no real stakes to cause anxiety and desperation.
It’s always been a dream of mine to one day own a house where I could drill holes into the floor when I need to make things work. Never mind all the other amazing stuff you did over this project… just based on the drilling, you’re my new hero. Awesome work 👌
"When you have a bolt that's really stuck, you get way out here and snap your wrench in half"
Had me rolling.
😂😂😂💀
After your wrench is snapped in half, you think to yourself "oh ya, lefty loosey" and finish your beer
I once tried it with an allen wrench and it literally broke in half
Once i was changing tires ant the bolt was VERY stuck,we pulled out a 2m bar and still couldnt get it out,because we were lifting the car 😂😂@@jamesMcroy12
"I optimized my code so hard that now it doesn't even do anything" sounds like a title from a post on StackOverflow
me
Or the title of a book
@@ncot_tech "I optimized my card so hard that it doesn't do anything - And other things to say to pretend you're a good programmer"
you can't waste cpu cycles if your program doesn't have any code **taps head**
1. All programs can be optimised, and 2. All programs have bugs. Therefore, you can optimise every program down to only one line which doesn't work.
Absolutely amazing. The time taken to produce that dog really was inspirational. How did she make it glow as well!!!
Haha you got me there, good one!
😂
😂😂
Hahahaha
Great; now I have to watch the video - no I don't.
I watched this video 7 months ago, but when it popped up again I couldn’t resist. You are clearly a genius on many levels, but the most impressive thing to me is how humble you are. Your content is very entertaining and I always do a little happy dance when I see you uploaded a new video!
No matter what I do in life, this channel will always make me feel like I'm gluing popsicle sticks together.
Just build a machine that would glue them together for you. Easy!
You can build a lot of cool things out of popsicle sticks.
That'll be true even if you become a hardcore CNC programmer, because this guy insists on doing everything the hard way.
I've spent a good amount of time chainsawing, and I can say that that little chain on that little saw would get dull very quickly, especially if it's always going in tip first, even with softwood. I'd guess that's why it got so hot. A longer chain with a built in cutting oil squirter on the arm seems like it would be crucial. That being said, awesome video as usual. The sheer perseverance your projects take is nuts.
i think a string saw would work too
Did his saw even have any bar oil?
haha.. tip first
I am doing a really hard programming assignment right now and the quote "There are only a finite number of ways you can screw up" has really motivated me :)
while (screwedUp) {
breakSomeOtherCode();
}
Actually they are
countable but infinite :(
You’ll probably never see this, but I think you’re a really talented teacher. The way you explain things is brilliantly understandable and enjoyable to learn from you.
It’s makes you videos better than 99.9% of what’s in this digital universe 😌
Agreed
I'm a 30-year mechanical and electrical/software engineer and I continue to be amazed by your breadth of knowledge and ability. Thank you for showing the world what a knowledge of math and science and fun can bring to our lives.
@Peter Evans If all the men asked only the old ladies about the expected results of possible ideas, then we would all still live in caves, because we would know that trying to make an ax out of the stone we are holding in our hands is nonsense. The old lady said so.
@Peter Evans Sorry. Most new ideas are by "old ladies" understood as "non-working" paths. So it is better to toss coins than ask your "old lady". Better results.
@Peter Evans Are you being deliberately foolish? Any old lady reading this, should StuffMadeHere have used a Metabo or a chainsaw to make this? Hardwood or Styrofoam? Stepper motors or servos? Should he have used an armature or field windings? I can only trust them to know the difference between Chick Peas and Garbanzo beans.
Guys he's trolling stop falling for it.
'30 years old'
"continue to be amazed"
Go check out a maker space and interact with the people you least expect to know shit.
Dry sense of humor gets me every time
The final dog was surprisingly elite
"snap the wrench in half" The timing was so unexpected and caught me off-guard, Hilarious.
Bro, i felt that one for sure
Exactly 🤣
Who hasn’t either snapped it or snapped the bolt off and gone flying lol.
@@SD-oi9gr I know I have Haha
Yea.. that's a ratchet tho...? Not a wrench😬
I really like how the foam dog turned out. It looks cool! The extra "error" cuts give it more character IMHO.
I'm an engineer. This guy is on a whole other level, I can't believe the scope of his expertise.
Fr. I’m in cnc and it’s wild how much he knows
Yes, so many things combined with a true sense of sculpting - certain elements of that dog's face are subtle (for a chainsaw). Would people think he's a nerd? Not with that cute sweetie of a wife standing next to him.
Yeah I'm just continuously baffled by the insane shit he builds and comes up with. Studying to become and engineer and I just hope that I will be able to understand and learn even just one part of all he does
He's also kind of stupid tho. He never tested whether he's actually able to cut wood before going through the trouble of writing software that automatically computes a tool path. But yes, he's an impressive polymath and his CNC parts look pretty neat.
Dude... remember he is doing this full time and it doesn't matter how long it takes him and NO ONE is yelling at him when he gets it wrong. Those two things make his work ... nothing to do with real world engineering. You have been trained to just be proficient not creative. Creativity is a multiple disciplinarian practice.
This is mind boggling. Do you get any sleep at all? It’s hard to imagine the amount of work on the engineering side of things, plus there’s the whole youtuber/editor side of it on top of it. Simply amazing.
My theory is that he's actually a team of quintuplets. I have no real evidence, other than the inhuman speed at which he works.
I'm a trained mechanical engineer and I am compensated handsomely. Having said that, in order to sleep better at night i tell myself that you are not just one guy who has this enormous breadth of knowledge but that there is a collective of at least 7 different people behind this channel doing all the work and you just pretend that it was all you. Please don't destroy that illusion for me. And i hope my boss never stumbles across this channel and suddenly expects me to code and weld and video edit and built robots in a month
I know that feeling!!, I'm a software engineer, and he just says "I would have to write a lot more code" just after saying that he can't implement the previous algorithm, as if he would be stoping that plan due to his inability to write it fast, correct and optimized... which is what he ended doing.
I really tell myself, those 7 people MUST be real.
Lol
We can't all be the most talented and motivated one in the room.
Im a safety engineer. Even tho my job dont need me to build stuff or coding anything, but it need me to understand what kind of machine is this and WHY YOU SHOULDN'T PLAY WITH IT.
@@soninh4357 He’s not playing with it he is doing experimental beta testing :)
Can I just give Tormach a additional shoutout for supporting you on this and enabling you to test this new piece of machinery? I've gained so much respect for them as a result of this ploy, even though I know it's marketing and user testing, it just speaks heavily to their company.
At least they care what users of their products think, have seen companies much worse
I swear the engineers at Tormach are fans of Shane and got the marketing department hella drunk so they'd do this :D
But tbh Shane is the best kind of beta tester. His projects are so far outside the norm that if the robot has any weaknesses they will come out. I've done beta testing for video games, and a part of it is trying really stupid ideas or really weird speedrunning tricks that might screw with the code. Basically if a speedrunner searching for new tricks can't glitch the game even by manipulating code, a normal gamer is really unlikely to ever fall through the floor. Same principle applies here: if the robot arm can survive Shane's project, a normal customer will never run into a problem.
@@nalinea18 Course, actual speedrunners will immediately glitch through the floor or go the speed of light
@@nikkiofthevalley sounds about right
The most valuable part of SMH videos is by far the fact that he shows us is approach to the problem, fails and then takes a new approach. Great content!
I love how he always includes his failures. Really shows how you just have to be persistent if you really want to succeed.
I think a lot of that is to fill time tbh. Like how Top Gear dramatises things.
I think it's the only way to trust he didn't faked it. Because you could also spend an entire month describe hard coded all the instructions to carve this exact sculpture. That would be easier, just the program would be stuck with one specific shape, for one specific machine, in very specific conditions. He proved that it is from an imported file, which means once he debugs, he can carve anything. Which is absolutly mind blowing.
i like how having a small child at home doesn't stop you from making something called a "kill box"
@@yeoj_ well, guns are predictable and therefore quite safe. What is however unpredictable and therefore quite unsafe are guns wielded by idiots.
On gradient descent: I think there are some approaches that start from a number of random points in order to more likely find a global optimum point.
“When you have a bolt that is really stuck you get way out here and snap your wrench.” Haha this line was great.
I have type 2 depression
@@OgSUBSCRIBO what is that?
i came to the comments looking for something about that
@@OgSUBSCRIBO joe
I'm going to leave a reply actually relevant to the comment, cuz nobody else is. Yeah, A coworker of mine once snapped a socket wrenches' ratchet doing that.
10:49
"I'm sure there's someone really smart out there who could make this work, but I cant."
- Someone really smart
Smart != being able to do everything
He’s clearly very smart. But some other smart person still knows more about this particular software problem than he does.
Knowing your limitations is probably the surest sign of intelligence there is.
Best part is that the solution is to just take multiple guesses at the initial parameters and run the ascent simulation simultaneously; and for best results, having the simulations and their results linked and iterative. Its called a MCMC (Markov-Chain Monte Carlo) algorithm, and usually doesn't take too much time.
LOL I was thinking the same thing. He is basically using DUM-E as a personal artist. Stark barely got the thing to put out a fire.
because in this channel he tried doing many different projects instead of just one, he realized that there are people who are specialized for certain project, who knows way better
"there is [sic] a finite number of ways to screw it up, so you will eventually finish" This is great advice for many things in life.
This guy is genius, and I say this to each and every single one of his video. How wide and deep his expertise is on different subjects is inhuman.
I’m an engineering student, and I watch these videos imagining the day I can make projects this good, but the further I get in my degree the more I realize this guy is just a wizard.
He does this as a job so he has way more time and money to do this stuff, and every tool under the moon so don't feel too bad. But he still is very impressive.
@@willsk3122 The most important thing that he has is experience. You could have all the tools, money and time under the sun, if you don't know what you're doing and how to troubleshoot when something goes wrong, you're pretty much screwed.
Another important side effect of this being his job:
Giving up on a project is a much worse option than usual
@@willsk3122 He only very recently started making videos. He had most of this equipment at his home already so I'm curious what his job was previously
@@chrism45 I think he was in 3D printer product development, so very familiar with cnc coding, manufacturing and the product lifecycle. That plus a passion for DIY and craft will get you 75% of the way there. The rest is, I believe, dark magic
20:57 "my wife is totally different, which is one of the reasons i love her so much" i was not expecting this amount of wholesomeness
I don't know how you can invest so much time, brain power and money in a single project, but I really appreciate it
Man is a computer engineer, electrical engineer, and mechanical engineer all at once. Respect.
"There is a finite way to screw it up"
A phrase to keep in mind during any creative process.
The set of ways to screw it up might be finite, but there's no guarantee that it's small enough to assure that you can correctly identify when you're screwing it up the same way multiple times.
Yep, that's me when coding anything whatsoever that's complex..
i want this on a Tee shirt, written backwards, so I can see it in a mirror.
Yep. There's also another popular corollary to this statement: An expert is someone who has made all of the possible mistakes in a particular, very narrow field.
@@vbscript2 Now, THAT is going on my business card. I'm a consultant, so I have really only seen the result of the mistakes, but close enough.
Love tuning in to see what kind of suicide booth you’ve made this week.
Lol
Kill all humans
@@MrKert86 yup
Best comment. Pretty sure this is coming on this channel at one point or another.
I'm already worried for the next video he going to post. "I figured that if my robotic chainsaw arm could carve a dog sculpture, why couldn't it give me a haircut?"
"there are no mistakes, just happy accidents"
Easily one of the best UA-cam channels. Both ingenious and WHOLESOME 100.
I can seriously learn from this guy. So glad the internet exists! 👍
Couldn't we all!
He is freaking amazing. So glad he shares all this with the world. Not only can he make a living, but he can share his wealth of knowledge with the younger generations. A lot of his projects are stupid and have no practically BUT damn it really shows what goes into design and build
I dont work in a similar field to him, but i also feel like i learn by watching him slove problems. Hes got a very intelligent process and a great way of explaining thing.
@@adamhoisington9761 exactly. Super valuable concepts in every video that nearly anyone can learn from.
You just want a heart from stuff made here
"You get way out here and you snap your wrench in half"
Can confirm from personal experience
Can second it for sure
I had to pause the video to laugh at the beautiful delivery of that joke.
@@lauraner99 Me too 😂
I love how it always ends with
Xs has a
him saying it's the greatest thing he has ever done and his wife pointing out that they're having a child?
@2025 Muhammad Zahir i’m incredibly sorry, but it’s a scam. You cannot possibly get robux free.
Same dude i like it how it always starts with
@@pakkunen8491 etherium mining
“If I tell it to stab itself in the face, it’ll do it” has to be one of the most terrifying sentences I’ve heard.
This is your channel's 1 year anniversary!
21 incredible videos in 52 weeks, with millions entertained and inspired!
Unbeliable what he's already done in a year. I've not seen anyone so talented on UA-cam (or real life). Amazing!
I just don’t know how one man and his wife could produce such amazing content so quickly
So true right!? Just look at that cute cuddly little human baby! Goochie goochie goo...
He only started mid last year, he got a million subscribers in like 2 months!!!!!! Dark magic
@@guy_in_ashopping_cart-sfs967 Thats what happens when you produce quality content
Lots of good food and love to increase productivity
the guy is a single man engineering team. like seriously. by far the most skilled and driven maker in the scene.
"This is one of those ideas that is so bad it wraps back round to good" - I will be remembering this for some time to come
🤌 Every super successful web business
You have some of the best content on the platform, and are one of the few creators that I can just start the video and be absolutely enraptured for the full run of the video.
"I optimized it so hard that it didn't actually compute anything" is the most relatable thing ever
I like how he dumbs down everything for us 😂😂
“Imagine this pretty shape here. I made it a fun color to hold your attention”
@@dave6012 😁lol
Is it possible to dumb down the dumbed down version even more? Iam asking for a friend! 👀
And I still can't understand it
"Dance till you're dead 10 hours" is the best music for coding.
And crying.
Thank you I was was wondering what he was watching.
I was hoping I wasn't the only one who noticed it playing in the background. xD
what’s the difference?
Code until you re dead
same thing
As a software engineer I’m amazed how the author is converting “I have spent a week on this algorithm with no success” into the coolest gradient descent “animation” I have seen! Amazing video!:)
As an engineer and hobby youtuber I know how many hours it took you to make this video, but I still can't believe you actually did it :D That is some serious coding right there!
As a Software QA i’m amazed by the work!
Remember the last
"There is a finite number of ways to screw it up, so you will eventually finish" hit a little too hard
''The true definition of insanity is doing the same thing over and over and expecting a different result'' is what i usually do.
The only problem is: There are actually an infinite number of ways of screwing it up.
Until there are multiple variables and sometimes doing the human representation of a rapidly expiring random tree is the best way to go.
Ok how they make this is really insane you need to now what you are doing its not just cut you need to put proper speed it is not a saw were you can rip it fast as you want and thats not a proper saw there is no gaurd you need a guard a gaurd is these teeth in the back of the chain the chain will pull foward sow it blocks its wood gaurd you need a person to run that saw not a robot that why it smoked
Anyone here watch Wintergatan, he knows that struggle, and has almost won.
This channel might just be the most underrated in all of UA-cam. This guy is nothing less than a genius.
This begs the question: have you subscribed?
I hope he is because I feel bad for myself whenever I watch him.
He's just an engineer
L H ....with a very broad skill set. None of the engineers I know have this many skills at this high a level.
@@loul7239 yeah he does the programming + all the machine work + engineering and math like i just dont understand where he learned all of this lol
14:30 The key to being a great robot maker is apparently 10 hours of dance till you're dead
Check out how to sharpen a chainsaw. The “anti-kickback” tabs that you referred to are depth gauges. You need to file them down a little in order for the teeth to cut. Should work great after a proper sharpen!
"There is a finite number of ways to screw things up so you will eventually finish" is possibly the best and most truthful inspirational quote I've heard so far
It sounds good but it's not true at all. Just because you did something wrong 10 time before you did it right diesnt mean there aren't 10 more ways to do it wrong. You just quit looking.
I don't think anyone but the man himself would realize all the pain that went into this... you sir, are a one man army of engineers.
He's basically doing a master's thesis work of work for a youtube video, very impressive stuff.
Greetings from Brazil! Just graduated as a Mechanical Engineer and have watched your videos for a while now! Thanks for the inspiration
This is what kids imagine when they hear that engineers design and build cool things.
Lol yeah I bet being a engineer on a jobsite isn’t very fun
Hey, I'm glad your wife seems to think my book is more interesting than carving logs with a robot! Hahaha. That made my day. :)
Thanks for making such an amazing book! I learned so much from it - one of my favorite textbooks :)
@@StuffMadeHere I appreciate the kind words, and many thanks to you for your brilliant work that shows why we are in love with robotics!
@Connor Towns it is, it’s great that two people who admire each other are able to communicate via the comments of a video platform.
Go Steve! Okay, I'm going to fall prey to such shameless marketing and buy the damn book. With my limited faculties, I need all the help I can get.
My first thought: That looks like a good book to read. Stopped here to comment, looked up the book, and haven't even watched the rest, yet (I promise I will as soon as I'm done looking)
“We both made dogs.”
Who else here thought the green wire glow thing was a cat???
I think that may have been the joke!
I've once taught of investing in Bitcoin but was always discouraged by my friends
me too
Bot go brrrrrr
honestly man what i love about you and your channel is that you make me and a lot of people actually interested in engineering, you make younger audiance into engineering once again you're one of the few rare people who make engineering fun dreamy and intresting the way you mix AI, Robotics and software engineering all together to build so much intresting things is amazing you're one of the few who make kids dream about becoming engineers not tiktokers and "influencers" once again anyways man i love your content keep it up ❤
"I'm almost out of things I can screw up, so I'm feeling pretty good about this one" - Engineer, right before screwing up about 500 things he never could've even imagined
The famous last words!
The project is 90% complete. Only 90% left to go! -- Engineering mantra.
@@MichaelSteeves that also feels like animation and music.
Can you invent a machine that will slap snacks out of my hand
Try a mirror
I think the chainsaw robot arm could easily do that.
Talk to @davie504 maybe
go to Michael Reeves channel, he does those things
so, integza's slappatron?
17:05 You need to add a Chain saw oiler to keep the blade lubed while cutting wood.
An automatic chain sharpener would be a cool project
“You get way out here, and snap your wrench in half” been there lmao 😂
Absolutely incredible 🙌🏼🔥⚡️
can u reply to my comment jimmydiresta so i can brag to my friends
@@ksi1275 looser
Penile Pudding Pudwacking Make it Come.
ksi nuuuuuurd
One thing I love about these videos is just how humor filled they are.
“Besides, I have a half-baked solution that has a chance of working”. That’s how you know this guy is a real engineer 🤟
And a real programmer!
@@gamemeister27 this guy is a modern day davicni. he is a master of many disciplines
16:44 I'm dying!!! That emotion is so relatable :)
I love how the robot being turned on is described as “being in kill mode”
I may have read this wrong and was wondering why or how the robot would ever be aroused, and how intense it must be for it to be called "kill mode".
i used to be a part of a team that built wall painting robots. the robot had a "suicide" button that when flipped, it just came down from the wall like a rock. the guys working with the robot complained all the time about it.
@@creative_leafeonpony_fan1545 you could try watching the video
"I caught the perpetrator red handed... And it was me." I'm studying Electrical engineering and this resonates too much
Oh diagnostics... why is it always my fault?
If you were REALLY smart you'd be studying mechanical engineering.
Oh damn, you mentioned a programming issue that I think I could actually help with for once!
For the "false summit" issue you're having, you can add "momentum" to the next location guess by summing the vertical distances (or just the normalized gradient if you're working in a space where you can calculate that) of a set of the previous moves and then adding that (as a fraction between 0-1 * a multiplier/'step size' constant) to the next movement. This way, once you reach a peak, you will go past it and then your height change will go negative, so it will cause it to go back towards the top. This will make it oscillate around the max/vertex (or min if you want to go concave), then eventually settle at the point. HOWEVER, it will allow the tool to move past peaks and instead of oscillate back, if it hits another region with a positive gradient before the momentum runs out, it will start climbing back up that. It's like rolling a ball down a bumpy hill (but inverted in our case), the momentum allows it to keep rolling a certain distance (depending on the step size) back up a hill once it rolls down into a valley. if there's another valley on the opposite side of this hill, the momentum might allow the ball to roll through the first valley (the 'local' minimum), up the next hill, and down into a lower valley (eventually the 'global' minimum). So we can just flip gravity by looking for increases in gradient rather than decreases as you would in that example and we can find the highest peak this way. Of course we don't really want the chainsaw having to oscillate around so much so all of these checks can be done in advance and then used to instruct the next move.
If you or anyone want to look into this, it's a common way of improving artificial neutral networks where the error surface is mapped in a 3D vector space and the weight change calculation (which we use to minimize the error) can be made by using "momentum" to avoid finding local minimum errors (i.e. not the most accurate way to train the neural net) and find global minima. So just search up something like "ANN momentum for finding minimum MSE" (where ANN is artificial neural network and MSE is mean-squared error, fyi) or just message me and I can share my notes :)
Not in computer science, but why not run multiple tests in a given area or settle on a "tests per square inch" target?
MuchADrewAboutNothin it’s just slower in the long run. If you have to sample from a large surface, picking random points in a region and testing will find the peak much slower than making a guess, then using the gradient of points around it to “walk” towards the peak