Missed opportunity: if you’d done your simulations with a square sun and moon, you could have used the deviation from your actual data to calculate pi.
Actually, as a statistician, I can tell you this is done more than you think. If there are meaningless points like those, they can mess up the real, valid data. For example, this data appeared pretty nicely, but data usually needs to be smoothed. Outliers will really mess that up, so they need to be removed. It's not something we try to hide - it's just part of the process.
Doing the math on that title was fun. "0.000001 megapixel... so since a megapixel is a million pixels, that means... [counting] ... 1 pixel? Yeah, that sounds like a Matt Parker video."
11:45 This is why you paint the inside of the tube black. And install the light meter at the bottom of the tube rather than at the objective end. (Never change, Matt.)
And paint the inside black, and cut the end of the tube at a slant, kind of like the brim of a baseball cap. so that the lower part of the end of the tube is always in shadow.
It's possible that putting it too low in the tube might lower the available light level down to a place where this sensor wouldn't give as good of a result. But it also might resolve his dynamic range resizing spike event to keep it in a consistent range... ...feels a little too official for a "give it a go" tube
Putting it too low would make it measure just one patch of sky, which changes during the day. Averaging out a larger area does seem to make sense. It annoyed me though that they hadn't taped over the slit in the side...
21:35 "you don't need fancy equipment, you just give it a go" -- I think this sumarizes succintly why Matt is one of my favourite educational youtubers out there.
"...if you want to do something unique now, [unintelligible] you have to travel all the way around the world, make sure you're the ONLY RUclipsr here making a mathematics video..." LOL Clearly planned. Nicely done.
I guess that is why he has a career as a mathematician 😂. Love the channel by the way, only having a lend in the nicest Aussie way. Actually unless you track the sun with the tripod, a shorter tube will be less biased in the result.
Ah, *a true Parkerscope,* with the sensor near the _front_ (instead of at the back, where it should be), and the _outside_ painted black (but not the inside, which should be - or, ideally, flocked).
That 3B1B cameo really caught me off guard. I had just opened another RUclips tab and moused over another video, which started autoplaying, at the exact same time as the audio cut-in. I switch back to this video, see nothing on the screen, and wonder what the heck just happened lmao. Speaking thereof, You both are exemplary maths RUclipsrs which are the gold standard of your field. Time and time again, your completely different personalities offer the right combination of entertainment and teaching moments that brighten my day. Thank you.
Should have lined the inside of the tube with black velvet to absorb any "stray" (off axis) light. Shield the ParkerScope V2 sensor with a tube that is black, inside and out. Or set it way at the bottom of the tube (that has been lined with black velvet).
@@broncogrizz I was gonna comment on that too, but then I remembered that this is a Matt Parker video and I think that's a good enough answer to that question.
Also set it up on an equatorial mount, and rotate it to track the same spot in the sky relative to the sun. With a high precision instrument like the Parkerscope this could be done by just eyeballing the orientation of the tripod, and manually rotating it by 1 degree every 4 minutes.
@@donsample1002 An equatorial mount might not be the right choice in this case, as he is recording the ambient brightness of the atmosphere. The scattering and brightness changes between the horizon and the zenith. The best solution is probably to record the previous day with identical conditions.
Your viewer looks more professional than ours. My kids and I built a viewer with a 5000mm focal length (it projected about a 5cm image onto white card). When we first tested it in Exmouth we were worried about the two dark spots in our image, but after cleaning the lenses thoroughly and rotating the viewer and discovering the spots stayed the same way up we realised the spots weren’t on the lens they were on the sun.
14:10 yes, we can see big jumps an all the occasions where the sensor overstepped this range and only on those:😌 11.02 11.45(ish) 12.12(ish) on the position change it jumped from 12000 to 3000, so it also jumped over the 4000 mark between two measurement steps, and 12.24(ish)
6:45 I would like to point out, as a published scientist, that many experiments have significant tape and glue as integral parts to them. In my graduate work we often made a point that our apparatus had a C-clamp somewhere in it. Experimental science where you're only ever building one of something has a place for tape and glue.
I worked at a radio observatory... many things were still held in place with tape or zip ties that hadn't been changed in at least a decade. Like you said, experimental science and one-offs make for interesting choices and there's always something else to fix or chase. We used to joke that if you called something an "interim" solution, it was guaranteed to outlast the telescope. There was an 'interim' spectrometer that had been put in place decades ago and became such an integral part of the system that it was never taken out.
From making my compsci masters dissertation I took away that any experiment only has to be clean, precise, proper, and perfect at the thing you are experimenting with, everything else is hacks, glue, hammers, sweat and blood just to make sure that the proper bits are perfect.
The prototype LIGO at Caltech back in the eighties had dampeners composed of alternating lead blocks and... racecar-shaped erasers, bought from a local store.
don't worry, I used to drive 1.6Mm from work to home, and I'd never dream of thinking about it any other way (of course the speed limit of 0.11Mm/h didn't help)
I'm not 100 percent sure, but I think you were dealing with Binary Coded Decimal, which may seem ridiculous, but it was heavily used in early computers and embedded devices (e.g. 6502 processor) to make conversion to and from decimal easier. Processors had specific circuits to handle arithmetic with BCD and you could convert from a digit's ascii code by simple subtraction and back to an ascii code with addition. My guess is that a lot of lower power, and real time devices still use them to avoid expensive conversions.
It's definitely BCD, I thought that as soon as he said "base 10 but encoded in hex for some reason". I 've worked with low power embedded systems, that often don't even have the luxury of a BIOS to handle input/output for you, you have to roll your own IO stuff. BCD is a neat way to display base 10 numbers in this case, as each digit is stored in its own byte, so displaying them is trivial. The Z80 CPU even has a dedicated instruction to correct the A register after BCD addition.
That was my first thought heard it. I first learned of BCD when I started working on IBM mainframes some 35 years ago, they actually had instructions to work directly with it.
@@billr3053 Not sure what you mean by 'undocumented'. The z80 manual that I had waaaayyy back in the 80's explained that BCD math was performed by performing a regular 'add' with the accumulator register and then 'daa' to adjust the results. The 'daa' opcode was specifically created for BCD math.
Moments like this are why I love science! The odd times you're just screwing around with some idea or some equipment and all the sudden you realize you can, with the right data processing or novel use of existing tech, accomplish something in a way that hasn't been done before... it sends shivers down my spine, every time
7:50 some more terminology 1st to 2nd contract: penumbra 2nd to 3rd contract: umbra 3rd to 4th contract: penumbra 10:00 the tapering off is called limb darkening. 10:20 - there you go! 10:50 Since the sky is mostly cloudless, Rayleigh scattering is dominant. If it were overcast, Mie scattering would have been most obvious, and the Solar eclipse would be kinda boring😄
To go an extra step, since the gradient on the simulated Sun is based on a u value, you could now try to find the u value closest to the actual measured data. Which would provide further information into how thick the photosphere actually is.
Suggestion for next eclipse: put the sensor deeper in the tube, and paint the inside of the tube a matte black (Or Matt black). This should make the ambient light from people walking by less of an issue. Also, instead of doing a linear approximation, could you take a baseline brightness curve from the day before or after from the same location, in the same direction (assuming similar weather), and then subtract that from the day of the eclipse?
The compromise is that there's less light in total and you end up looking at a smaller part of the sky, meaning there'd be more noise and a bigger effect from the specific pointing direction. Still, I mostly agree that it's way too close to the opening of the tube. At first I was sure that he was going to put the sensor in through one side but have it pointing to the other side for exactly this reason.
Optics: nope... I'd have gone for "Optics NOT made by Zeiss" or something similar 😇 - It would probably work better if the INSIDE was painted black. I think the "walking by changing" is the reflected light from the inside of the tube...
I think the inside of the telescope should have been painted with Vantablack. Which would have make the Parkerscope one of the most expensive on the beach that day 😋
Vantablack can't really be exposed to air or touch. It won't last long. Its a forest of nanotubes and there can be a domino effect that causes nearby nanotubes to fall over and ruin the effect.
Hope you enjoyed your time in WA! I remember trying to look at the eclipse from the office down in Perth, best we managed was seeing little crescent shaped shadows through leaves and pinholes and stuff.
I hope you get to repeat this with all the learning from this. Would like to see a 1 pixel movie of the data. Just a visual idea of the change in light level over the whole eclipse.
Great video as always! Suggestions for the next time: Let the tube be longer or put the Luxmeter deeper in it to the closed end; the inside of the tube maybe painted black too and point the tube north in the northern hemisphere or point it south in the southern hemisphere all for preventing any spikes during measureing and/or shadows inside the tube. greetings from Germany :)
The proper thing would have been to have the camera in place the day before or the day after (hopefully with the same weather conditions), record the intensities for the time period an compare the differences, because you do see intensity changes from the sun moving in the sky, and the only way to compensate for that is to record the same time period the day before or after.
Might be fun to use that data and a couple of models to do some Bayesian analyses and Markov Chain Monte Carlo to find which model you suggested works best + find the ideal parameters for each. That way you can quote some Bayes factors on how much better one model is than another! It would be a fun project!
If you didn't remember that one from school, you probably also forgot decameters and hectameters, the 10x and 100x units. Don't mix up your decameters with your decimeters, though!
Spikes could also have been birds or insects winging by. Personally, I would have placed the meter much further back in the tube and painted the inside of the tube flat black. But it still worked out pretty well. 👍
I used a fancy light meter back in the day (late 1980's), before they were built-into cameras. I had two white covers for the sensor: a hemisphere, like the one you have, and a flat one. The hemisphere integrates the light from all directions. So, it's looking at the reflections on the side of the tube. You would be better off with the flat one, which looks _straight_ down the tube. In that case, you don't need much of a tube; just enough to prevent direct light from falling on it. In fact, like the baffles you see on normal camera lenses, or the bill of a baseball cap.
All the people suggesting improvements to the design don't realize the whole point was, probably, that science can be done even without access to state-of-the-art instruments...
If you live in the USA, in spring 2024 there will be an eclipse going from Texas to Indiana. If you get a chance, I highly suggest going. You'll need to start planning now, because hotels/etc can be hard to come by. But this is a once in a lifetime experience that it's absolutely worth taking some time off work for.
Matt, consider this an invitation to visit Waco TX and Baylor University for an all day festival centered around the full eclipse. For reals... Reach out to me and I'll help coordinate your visit and try to get you plugged in with the city and the university.
I saw the eclipse from Pebble Beach (near Exmouth) but I didn't think of making a pinhole camera or anything. I got a great shot at totality with the two cruise ships in silhouette, showing that the sky looked like it does at sunrise but all around the horizon, and not just in the east.
My family drove 13,000km from Brisbane and back over 3 weeks to see this. Totally worth it. I actually was in the same place you were and was hoping to see you there but clearly I am blind..... or you were hiding too well.
A Parkervid all the way! Right up to the final touch at 22m22s, "... and these days there are so many incredible maths RUclipsrs out there doing all sorts of videos ..." while Grant Sanderson walks by behind him, doing a YT video with his iPhone (?). Fred
I would say that the Parkerscope is a perfect-looking telescope. It's like what a five-year-old would draw when asked what a telescope looks like. I think whimsy is what really leads to scientific results.
Surely if you put the sensor a little further back and painted the inside vanta black or the darkest mat black you could find, you would have much less problems with people walking by etc.
Glad that the eclipse was a good experience for you Matt (and RUclips viewers). For me, it was a demonstration in corporate greed exploiting a natural event. After driving 5300km from Hobart to Exmouth, I found that Exmouth council, corporate interests and the police, were working together to extract ludicrous $$$ from this. For me to be allowed to stay within the area, it was going to cost me minimum $500 to book a 3m x 6m space for my campervan, in a gravel car park (minimum 4 nights at $125/nt). Road blocks were setup on the highway and local roads 2 days before the eclipse and anyone not having paid accommodation was evicted from the entire peninsula (150km) from the totality zone. Rather than participate and thus give my tacit approval of this exploitation, I went elsewhere. There will be other eclipses.
As a Western Australian, I feel your pain man. But believe me when I say that if the accommodation restrictions weren't in place, noone would have got any science done because of all the yobbo's that would have turned up.
"NASA?! Never heard of 'em..." - gotta love him xD Also thinking more about the eclipse as such, I've remembered that the most impressive eclipse I've ever seen "myself", was in Elite Dangerous. And it wasn't even intentionally: I was mucking around on the surface of a planet, gathering stuff and suddenly I see a literal wave of BLACK rushing towards me from the distance. My first thought was that the game's glitching out and was about to crash, but after the darkness was everywhere, it seemed "normal", like the same place just at night. Took a couple of seconds of gears turning in my head and I look up and bam: random eclipse. The literal wave of darkness rushing in was extremely likely because the game has game-light and not real light, so it has to cut corners. But none the less, it was impressive and pretty, even if just a "fake" one.
Probably would have had better data if you didn't use the cardboard tube. You don't really need to collimate for one pixel. Just put your detector over by the fence, away from where anyone's going to affect it Especially if you had collected during roughly the same time the day before and the day after so you could subtract out the effects caused by the Earth's normal movement around the sun so you could have subtracted that out and had a better defined inverted peak. Assuming it's normal Western Australian weather and are the same number of clouds on each day.
Also if the goal was to *not* have the sensor point directly at the sun, it would've been okay to just point it at the ground and measure the reflected light intensity. Though it would look even less like a telescope then :D
I loved it. It shows that massaging your data nad understanding what you are doing is so valuable :D Some minor points: 1. Why not have the parkerscope a deeper tube / the sensor farther down in the tube? It might help shutout noise from like people walking by and everything without having to resort to optics ;) 2. I would probably try pointing the parkerscope at astronomical south (or north on the other hemisphere). Then the point in the sky you are looking at might a bit easier to track, because it actually doesn't change with the rotation of the earth. Of course the angle of the sun in the atmoshpere still changes, so you might still have to pull some models to correct for this... Of course there are alternatives, but without having the parkerscope follow a point on the sky I dont really see a better option tbh. Looking forward to the next ecplise ;)
It was a real fun to watch the manufacturing process of the Parkerscope. And I admit, the "0.000001" Megapixel part on the title got my attention, cleverly done. Actually I'm kind of surprise you got a near zero reading during the totality. If you do not look exactly at the sun, you always partially see parts of the sky not covered by the shadow of the moon. Even in totality this part will not darken. As an extreme example, the sky near the horizon to the south (away from the sun) should not be affected by the eclipse at all.
It's amazing how much "science" you were able to deduce from your simple experiment. I've found analyzing my failures yields much more knowledge than analyzing my successes.😊
Could you have positioned the ParkerScope™ inside the tent? That would have smoothed the data a bit. Could you also have collected the same set of data at the same time a day earlier (a job for past Matt) and subtracted that from the eclipse set, thereby eliminating the effect of the motion of the sun in the sky (commonly called flat-field correction)? Your work is an excellent example of citizen sciencery (the act of doing science?).
Matt is in the southern hemisphere so he should have pointed the Parkerscope toward the celestial south pole as that part is never lit by the sun. It also means the scope would have never needed to be moved. Or maybe on the line between the zenith and the celestial south pole.
And just for good measure, add an opaque ring around it to keep the entrance to the tube in shadow, just in case it wasn't perfectly lined up with the pole.
Missed opportunity: if you’d done your simulations with a square sun and moon, you could have used the deviation from your actual data to calculate pi.
Oh this MUST happen.
IT'S EXACTLY 3!!!!
+
Well first we need to assume a spherical plasma in a vacuum.
.... 🙄🤔🤨🤔😳 ....
....Uh, yes. Very good. Moving on, step two....
Simulated with a ParkerSquare Sun I presume?
"I removed the outliers, I just straight up deleted them"
Matt used the one simple trick statisticians don't want you to know
CompSci's would have filtered that reading out of existence as well. Can't blame 'im lol.
But unlike some shady folks, Matt was fully honest about which data points, what he did, and admitted to it.
Actually, as a statistician, I can tell you this is done more than you think. If there are meaningless points like those, they can mess up the real, valid data. For example, this data appeared pretty nicely, but data usually needs to be smoothed. Outliers will really mess that up, so they need to be removed. It's not something we try to hide - it's just part of the process.
In fairness, he could do that because he already knew why they would exist.
spiders georg's downfall was swift and terrible
Doing the math on that title was fun. "0.000001 megapixel... so since a megapixel is a million pixels, that means... [counting] ... 1 pixel? Yeah, that sounds like a Matt Parker video."
The Parkerpixel
Just did exactly that 😂
@@squidward5110 How long does it take you to count to 1? 😉
You can also use microfortnights as your divisions on the time scale.
I like that it was also a 1x10^-9 gigapixel camera.
11:45 This is why you paint the inside of the tube black. And install the light meter at the bottom of the tube rather than at the objective end. (Never change, Matt.)
This was baffling to me. Why not paint the interior black and put the sensor at the bottom???
And paint the inside black, and cut the end of the tube at a slant, kind of like the brim of a baseball cap. so that the lower part of the end of the tube is always in shadow.
Dude's a mathematician, not a physicist.
It's possible that putting it too low in the tube might lower the available light level down to a place where this sensor wouldn't give as good of a result. But it also might resolve his dynamic range resizing spike event to keep it in a consistent range...
...feels a little too official for a "give it a go" tube
Putting it too low would make it measure just one patch of sky, which changes during the day. Averaging out a larger area does seem to make sense.
It annoyed me though that they hadn't taped over the slit in the side...
The lines went up, because your tube was filling up with photons. You should have drilled a hole in the lower end for them to fall out.
Next time bring a handheld vacuum cleaner!
21:35 "you don't need fancy equipment, you just give it a go" -- I think this sumarizes succintly why Matt is one of my favourite educational youtubers out there.
cool seeing you here, I love your videos!
@@zyansheep wow Thank you! Matt is definitely a massive inspiration for me even if our video style is pretty different haha.
One of my favorite Star Trek TNG memes, Picard quoted, "You may test your hypothesis at your earliest convenience." Translation, 'give it a go". :)
@@mikefochtman7164 more life lessons from TNG
This is actually amazing.
22:22 3Blue1Brown, what are you doing here smartereveryday, I think he needs to lengthen the tube, and he shouldn't cut the tube.
Seriously! Also, I love your channel, too! Thank you both for making us all 'smarter every day'!
Please don't beat your children 🙏
There is something fundamentally hilarious about Grant Sanderson just strolling through your shoot, just casually talking over yours 😂
I am sure that was planned but it was fantastic.
I need very badly for Grant Sanderson's video, when posted, to show him wandering through Matt's shot from the other POV.
genuinely cracked me up...
"...if you want to do something unique now, [unintelligible] you have to travel all the way around the world, make sure you're the ONLY RUclipsr here making a mathematics video..." LOL Clearly planned. Nicely done.
I'm not sure if Matt is stalking Grant or the other way around.
Matt is like a toddler next to his wife holding up his Parkerscope: "Look mommy, I am a real scientist just like you!"
The data is in!
is dark when e-clips
e hasn't clipped yet
he would call his wife WHAT?
@@proloycodesoh he would for sure... 😈
I feel sorry for the production crew who fancified the cardboard tube only to have it all hacked apart for transport and fitting out 😂
And then to have the experimental results affected by the shorter tube... 😢
That's showbiz baby
I don't understand why the other section wasn't slit so it could fit inside and then act as an extension when reassembled on site? What a waste
I don't understand why it wasn't painted black on the inside.
I guess that is why he has a career as a mathematician 😂. Love the channel by the way, only having a lend in the nicest Aussie way. Actually unless you track the sun with the tripod, a shorter tube will be less biased in the result.
The shot at 5:30 of him comparing it to the proper kit, and seeing it pointing in the opposite direction had me cracking up. 😂
It has me thinking of the derpy hydra meme.
Choosing to observe an empty patch of sky really added to his solar astronomer clout
The Parkerscope doesn't need to follow what the other telescopes are doing!
@@Septimus_ii There are no empty patches of sky. There are only patches where we cant see anything yet... because it's daytime and the sun is out.
Big Toph putting up posters vibe XD
Ah, *a true Parkerscope,* with the sensor near the _front_ (instead of at the back, where it should be), and the _outside_ painted black (but not the inside, which should be - or, ideally, flocked).
and despite those mistakes he got results, well almost.
@@ozradek1 - Well, technically, as long as the light meter was turned on he'd get results.
@@RFC-3514 meaningful results. almost, like the famous Parker Square
And set the scope against the fence in the background to eliminate the passers-by problem.
That 3B1B cameo really caught me off guard. I had just opened another RUclips tab and moused over another video, which started autoplaying, at the exact same time as the audio cut-in. I switch back to this video, see nothing on the screen, and wonder what the heck just happened lmao.
Speaking thereof, You both are exemplary maths RUclipsrs which are the gold standard of your field. Time and time again, your completely different personalities offer the right combination of entertainment and teaching moments that brighten my day. Thank you.
Should have lined the inside of the tube with black velvet to absorb any "stray" (off axis) light. Shield the ParkerScope V2 sensor with a tube that is black, inside and out. Or set it way at the bottom of the tube (that has been lined with black velvet).
Yeah, I don't know why he didn't just put it at the bottom of the tube.
@@broncogrizz I was gonna comment on that too, but then I remembered that this is a Matt Parker video and I think that's a good enough answer to that question.
Also set it up on an equatorial mount, and rotate it to track the same spot in the sky relative to the sun.
With a high precision instrument like the Parkerscope this could be done by just eyeballing the orientation of the tripod, and manually rotating it by 1 degree every 4 minutes.
but that would make him an almost physicist rather than a mathematician, so that would be wrong
@@donsample1002 An equatorial mount might not be the right choice in this case, as he is recording the ambient brightness of the atmosphere. The scattering and brightness changes between the horizon and the zenith. The best solution is probably to record the previous day with identical conditions.
😂The 3blue1brown cameo was peak parker
I wonder how many tries that took. Matt barely flinched.
Are Matt and Grant bffs now?
@@ChrisBreederveld I ship it.
@@ClaudiaCarranza1 Don't tell his wife
I was looking for another video and thought RUclips was doing some sort of glitch by playing two video’s audio together
Your viewer looks more professional than ours.
My kids and I built a viewer with a 5000mm focal length (it projected about a 5cm image onto white card). When we first tested it in Exmouth we were worried about the two dark spots in our image, but after cleaning the lenses thoroughly and rotating the viewer and discovering the spots stayed the same way up we realised the spots weren’t on the lens they were on the sun.
Sometimes, you see how the sun sets on some films and sometimes you actually see the spots on the sun.
The sun is a hell of a lot harder to wipe off than your lens.
the jump at 3999/4000 is a common spot for 4000 counts multi meters to switch ranges.
14:10 yes, we can see big jumps an all the occasions where the sensor overstepped this range and only on those:😌
11.02
11.45(ish)
12.12(ish) on the position change it jumped from 12000 to 3000, so it also jumped over the 4000 mark between two measurement steps, and
12.24(ish)
Probably same Intercil chip as in the cheap 4.5 digit meters
6:45 I would like to point out, as a published scientist, that many experiments have significant tape and glue as integral parts to them.
In my graduate work we often made a point that our apparatus had a C-clamp somewhere in it.
Experimental science where you're only ever building one of something has a place for tape and glue.
I worked at a radio observatory... many things were still held in place with tape or zip ties that hadn't been changed in at least a decade. Like you said, experimental science and one-offs make for interesting choices and there's always something else to fix or chase.
We used to joke that if you called something an "interim" solution, it was guaranteed to outlast the telescope. There was an 'interim' spectrometer that had been put in place decades ago and became such an integral part of the system that it was never taken out.
From making my compsci masters dissertation I took away that any experiment only has to be clean, precise, proper, and perfect at the thing you are experimenting with, everything else is hacks, glue, hammers, sweat and blood just to make sure that the proper bits are perfect.
@@ford9501 like that one famous quote:
"There's nothing more permanent than a temporary solution."
The prototype LIGO at Caltech back in the eighties had dampeners composed of alternating lead blocks and... racecar-shaped erasers, bought from a local store.
It ain't science without a twist-tie or cello-tape somewhere.
Loved the cameo from Grant at 22:24
Who's that?
Oh Grant from 2 blue one brown. Knew voice sounded familiar.
@@n27272 Please tell me that was a deliberate Parker approximation of Grant's channel - 3Blue1Brown.... 😄
That was AWESOME "only RUclipsr here" 🤣🤣 👏👏
you should've got 3 such telescopes pointing in 3 different directions and triangulate the sun's position & movement from the data
Great thing is you don't even need an eclipse for that :P
Next eclipse plan!
@@m.sierra5258 but where's the fun
So in total it'd be a... 0.000003 megapixel telescope? :P
It's a three-pixel interferometer!
13 seconds in and I already appreciate the use of megameters
300 Mm/s is a speed I can actually remember.
don't worry, I used to drive 1.6Mm from work to home, and I'd never dream of thinking about it any other way (of course the speed limit of 0.11Mm/h didn't help)
Matt: "We waited until the end of the eclipse to get all of that delicious data!"
Also Matt: "It's symmetric anyway, so let's just use the first half"
8:45 the fact that he counted every single pixel in these circles by himself and then calculated the percentage is some real dedication
i seriously do not know if this comment is serious or joking
@@NutchapolSal think its a joking comment
A wild Grant hass appeared! 22:22
I'm not 100 percent sure, but I think you were dealing with Binary Coded Decimal, which may seem ridiculous, but it was heavily used in early computers and embedded devices (e.g. 6502 processor) to make conversion to and from decimal easier. Processors had specific circuits to handle arithmetic with BCD and you could convert from a digit's ascii code by simple subtraction and back to an ascii code with addition. My guess is that a lot of lower power, and real time devices still use them to avoid expensive conversions.
It's definitely BCD, I thought that as soon as he said "base 10 but encoded in hex for some reason". I 've worked with low power embedded systems, that often don't even have the luxury of a BIOS to handle input/output for you, you have to roll your own IO stuff. BCD is a neat way to display base 10 numbers in this case, as each digit is stored in its own byte, so displaying them is trivial. The Z80 CPU even has a dedicated instruction to correct the A register after BCD addition.
@@alsmoviebarn Yes, the 4-bit add and half-carry instruction, undocumented in the code space table.
That was my first thought heard it. I first learned of BCD when I started working on IBM mainframes some 35 years ago, they actually had instructions to work directly with it.
@@billr3053 Not sure what you mean by 'undocumented'. The z80 manual that I had waaaayyy back in the 80's explained that BCD math was performed by performing a regular 'add' with the accumulator register and then 'daa' to adjust the results. The 'daa' opcode was specifically created for BCD math.
Being able to deduce something so complex and fundamental like this with such simple equipment is why I decided to study physics at university.
I love how chuffed Matt is with his research. It's that enthusiasm that inspires me to try things like this.
Moments like this are why I love science! The odd times you're just screwing around with some idea or some equipment and all the sudden you realize you can, with the right data processing or novel use of existing tech, accomplish something in a way that hasn't been done before... it sends shivers down my spine, every time
23:55 the excited jumps from Lucy warm my heart
What jump?
Timelapse at 11:05 , like a happy kid😊
Your team really Parkered it.
7:50 some more terminology
1st to 2nd contract: penumbra
2nd to 3rd contract: umbra
3rd to 4th contract: penumbra
10:00 the tapering off is called limb darkening. 10:20 - there you go!
10:50 Since the sky is mostly cloudless, Rayleigh scattering is dominant. If it were overcast, Mie scattering would have been most obvious, and the Solar eclipse would be kinda boring😄
Apparently it's all called "eclipse town."
To go an extra step, since the gradient on the simulated Sun is based on a u value, you could now try to find the u value closest to the actual measured data. Which would provide further information into how thick the photosphere actually is.
Suggestion for next eclipse: put the sensor deeper in the tube, and paint the inside of the tube a matte black (Or Matt black). This should make the ambient light from people walking by less of an issue.
Also, instead of doing a linear approximation, could you take a baseline brightness curve from the day before or after from the same location, in the same direction (assuming similar weather), and then subtract that from the day of the eclipse?
Matt black!😁
The compromise is that there's less light in total and you end up looking at a smaller part of the sky, meaning there'd be more noise and a bigger effect from the specific pointing direction. Still, I mostly agree that it's way too close to the opening of the tube. At first I was sure that he was going to put the sensor in through one side but have it pointing to the other side for exactly this reason.
I personally suggest using Stuart Semple's "Black 3.0", the blackest black. :) It'd guaranteedly absorb ALL of the stray photons.
I think he did alright the way it was, with the inside painted a nice approximate shade of Parker Black.
Optics: nope... I'd have gone for "Optics NOT made by Zeiss" or something similar 😇 - It would probably work better if the INSIDE was painted black. I think the "walking by changing" is the reflected light from the inside of the tube...
I think the inside of the telescope should have been painted with Vantablack.
Which would have make the Parkerscope one of the most expensive on the beach that day 😋
Vantablack can't really be exposed to air or touch. It won't last long. Its a forest of nanotubes and there can be a domino effect that causes nearby nanotubes to fall over and ruin the effect.
I'd go with Stuart Semple's "Black 3.0" instead. It's a darker black than Vantablack, and also isn't toxic like Vantablack is. :)
@@KizulEmeraldfire Toxic to health or toxic to art?
@@Aesculathehyena both. Vantablack is a bunch of vapor deposited carbon nanotubes. Funky carbon molecules arent healthy
Hope you enjoyed your time in WA! I remember trying to look at the eclipse from the office down in Perth, best we managed was seeing little crescent shaped shadows through leaves and pinholes and stuff.
Doesn't it feel weird to say "hope you enjoyed your time in WA" to someone who was born and grew up in WA? 😅
When was Matt in the US State of Washington?
@@carl11547 WA means Western Australia.
@@DerekHartley I know. It was a joke. I will be you $1000 (US or Australian) that Dr. Oliver got it.
I hope you get to repeat this with all the learning from this.
Would like to see a 1 pixel movie of the data. Just a visual idea of the change in light level over the whole eclipse.
Great video as always! Suggestions for the next time: Let the tube be longer or put the Luxmeter deeper in it to the closed end; the inside of the tube maybe painted black too and point the tube north in the northern hemisphere or point it south in the southern hemisphere all for preventing any spikes during measureing and/or shadows inside the tube. greetings from Germany :)
Love the Grant Sanderson cameo! I actually got to meet him recently! Definitely a peak moment as an ex- math teacher
Thank you for giving it a go and working it out.
The proper thing would have been to have the camera in place the day before or the day after (hopefully with the same weather conditions), record the intensities for the time period an compare the differences, because you do see intensity changes from the sun moving in the sky, and the only way to compensate for that is to record the same time period the day before or after.
Thank you patreons for enabling this silliness we all enjoy watching, and learning from!
LMAO at the "only math youtuber" segment
I love all the 3B1B cameos in your videos. It's great.
Might be fun to use that data and a couple of models to do some Bayesian analyses and Markov Chain Monte Carlo to find which model you suggested works best + find the ideal parameters for each. That way you can quote some Bayes factors on how much better one model is than another! It would be a fun project!
The grant fly by.. just absolutely brilliant. Take my thumbs up.
I've always wondered when doing long drives across Canada whether it was proper or not to use the term "megameters". I feel validated now! Thanks!
If you didn't remember that one from school, you probably also forgot decameters and hectameters, the 10x and 100x units. Don't mix up your decameters with your decimeters, though!
Ah, fine. You beat my 0.001 megapixel DIY digital camera sensor.
Finally some data proving it's darker during an eclipse than before or after it!
There has been hardly any research dedicated to answering that question.
Kudos to your production staff. Amazing work there!
That little teaser for 3blue1brown at the end, lol
Spikes could also have been birds or insects winging by.
Personally, I would have placed the meter much further back in the tube and painted the inside of the tube flat black. But it still worked out pretty well.
👍
I used a fancy light meter back in the day (late 1980's), before they were built-into cameras. I had two white covers for the sensor: a hemisphere, like the one you have, and a flat one.
The hemisphere integrates the light from all directions. So, it's looking at the reflections on the side of the tube. You would be better off with the flat one, which looks _straight_ down the tube. In that case, you don't need much of a tube; just enough to prevent direct light from falling on it. In fact, like the baffles you see on normal camera lenses, or the bill of a baseball cap.
All the people suggesting improvements to the design don't realize the whole point was, probably, that science can be done even without access to state-of-the-art instruments...
If you live in the USA, in spring 2024 there will be an eclipse going from Texas to Indiana. If you get a chance, I highly suggest going. You'll need to start planning now, because hotels/etc can be hard to come by. But this is a once in a lifetime experience that it's absolutely worth taking some time off work for.
If you do not live in the USA the eclipse is still going to be there , same place and time
Matt, consider this an invitation to visit Waco TX and Baylor University for an all day festival centered around the full eclipse. For reals... Reach out to me and I'll help coordinate your visit and try to get you plugged in with the city and the university.
Grant is now the Stan Lee of the stand-up maths channel with a cameo in the last two videos
See, now I want to build a Parker scope of my own for the 2024 eclipse
Okay I love the walkthrough at the end by 3B1B 😆
Is there any way to calculate pi with your scope since the sun is a circle?
Thank you for not fudging the data, that's good science :)
9:59 your LaTeX-fu is very disappointing, Matt. Should've used \left and
ight on those brackets!
10:53 "The Sun is moving during this." Pope Urban VIII appreciates.
I saw the eclipse from Pebble Beach (near Exmouth) but I didn't think of making a pinhole camera or anything. I got a great shot at totality with the two cruise ships in silhouette, showing that the sky looked like it does at sunrise but all around the horizon, and not just in the east.
I can’t believe someone hasn’t uploaded a 8k HDR footage of a total eclipse here on YT yet.
Sounds like prime real estate for you if you can get the equipment
Great video. As an aside, the green lines on your eclipse Matt T-shirt made my eyes go fizzy!
This whole video was to set up the Parker solar probe joke.
One of Matt's best goes to have ever been given. I'm actually surprised the shortness of the tube did not affect the results more.
My family drove 13,000km from Brisbane and back over 3 weeks to see this. Totally worth it. I actually was in the same place you were and was hoping to see you there but clearly I am blind..... or you were hiding too well.
Totality worth it*
Really enjoyed this thanks
Are you on tour with 3Blue1Brown or something?!
Wow, a million micropixel camera.
Matt: You need to calculate the limb darkening value from your data and then write a scientific paper with your results 😀👍
A Parkervid all the way! Right up to the final touch at 22m22s, "... and these days there are so many incredible maths RUclipsrs out there doing all sorts of videos ..." while Grant Sanderson walks by behind him, doing a YT video with his iPhone (?).
Fred
That graph match was so satisfying. I loved this video
I would say that the Parkerscope is a perfect-looking telescope. It's like what a five-year-old would draw when asked what a telescope looks like. I think whimsy is what really leads to scientific results.
Surely if you put the sensor a little further back and painted the inside vanta black or the darkest mat black you could find, you would have much less problems with people walking by etc.
one of your best jokes there at the end, Matt. Nicely done!
Glad to see Past Matt rocking the beard again. "How long did that take to film?"
Loved the Grant cameo at the end. This was a great "giving it a go"!
Glad that the eclipse was a good experience for you Matt (and RUclips viewers). For me, it was a demonstration in corporate greed exploiting a natural event. After driving 5300km from Hobart to Exmouth, I found that Exmouth council, corporate interests and the police, were working together to extract ludicrous $$$ from this. For me to be allowed to stay within the area, it was going to cost me minimum $500 to book a 3m x 6m space for my campervan, in a gravel car park (minimum 4 nights at $125/nt). Road blocks were setup on the highway and local roads 2 days before the eclipse and anyone not having paid accommodation was evicted from the entire peninsula (150km) from the totality zone. Rather than participate and thus give my tacit approval of this exploitation, I went elsewhere. There will be other eclipses.
As a Western Australian, I feel your pain man. But believe me when I say that if the accommodation restrictions weren't in place, noone would have got any science done because of all the yobbo's that would have turned up.
"NASA?! Never heard of 'em..." - gotta love him xD
Also thinking more about the eclipse as such, I've remembered that the most impressive eclipse I've ever seen "myself", was in Elite Dangerous.
And it wasn't even intentionally: I was mucking around on the surface of a planet, gathering stuff and suddenly I see a literal wave of BLACK rushing towards me from the distance. My first thought was that the game's glitching out and was about to crash, but after the darkness was everywhere, it seemed "normal", like the same place just at night.
Took a couple of seconds of gears turning in my head and I look up and bam: random eclipse.
The literal wave of darkness rushing in was extremely likely because the game has game-light and not real light, so it has to cut corners.
But none the less, it was impressive and pretty, even if just a "fake" one.
It's a real shame that someone cut that tube down so much. 😁
Yeah, might have been less vulnerable to the shift in the sun.
Don't worry Matt, you're my second favorite Math RUclipsr on this video. Second is very respectable.
Probably would have had better data if you didn't use the cardboard tube. You don't really need to collimate for one pixel. Just put your detector over by the fence, away from where anyone's going to affect it Especially if you had collected during roughly the same time the day before and the day after so you could subtract out the effects caused by the Earth's normal movement around the sun so you could have subtracted that out and had a better defined inverted peak. Assuming it's normal Western Australian weather and are the same number of clouds on each day.
Also if the goal was to *not* have the sensor point directly at the sun, it would've been okay to just point it at the ground and measure the reflected light intensity. Though it would look even less like a telescope then :D
@@Arnavion Make it look and point like a proper telescope, but leave the lower aperture open, the upper aperture closed and flip the sensor.
@@Arnavion Trouble with pointing at the ground is you're even more likely to pick up on passing shadows.
I loved it. It shows that massaging your data nad understanding what you are doing is so valuable :D
Some minor points:
1. Why not have the parkerscope a deeper tube / the sensor farther down in the tube? It might help shutout noise from like people walking by and everything without having to resort to optics ;)
2. I would probably try pointing the parkerscope at astronomical south (or north on the other hemisphere). Then the point in the sky you are looking at might a bit easier to track, because it actually doesn't change with the rotation of the earth. Of course the angle of the sun in the atmoshpere still changes, so you might still have to pull some models to correct for this... Of course there are alternatives, but without having the parkerscope follow a point on the sky I dont really see a better option tbh.
Looking forward to the next ecplise ;)
Wooo
1st!
It was a real fun to watch the manufacturing process of the Parkerscope. And I admit, the "0.000001" Megapixel part on the title got my attention, cleverly done.
Actually I'm kind of surprise you got a near zero reading during the totality.
If you do not look exactly at the sun, you always partially see parts of the sky not covered by the shadow of the moon. Even in totality this part will not darken.
As an extreme example, the sky near the horizon to the south (away from the sun) should not be affected by the eclipse at all.
How'd you know I'd be watching this in the FUTURE? Are you psychic???
Surprised there wasn't a visualisation using the pixel data to create a video of the eclipse
There is probably no Scope imaginable _more_ Parker than this
Park, Parker, Parkerer, Parkest.
That ending shot, along with the patreon supporter credits, was amazing.
patreon link points to this video
Time is circular
The first 0.2 seconds was enough for me to know with absolute certainty that you were in Australia. That red earth and intense sunlight is distinct.
do you like chicken
It's amazing how much "science" you were able to deduce from your simple experiment. I've found analyzing my failures yields much more knowledge than analyzing my successes.😊
Could you have positioned the ParkerScope™ inside the tent? That would have smoothed the data a bit. Could you also have collected the same set of data at the same time a day earlier (a job for past Matt) and subtracted that from the eclipse set, thereby eliminating the effect of the motion of the sun in the sky (commonly called flat-field correction)? Your work is an excellent example of citizen sciencery (the act of doing science?).
9th
No
So much fun you even got 3Brown1Blue into the video! 😂
Im not sure what the intentionally confusing representation of 1 pixel in the title was meant to do
Typically the resolution of cameras are written in megapixels
Be funny
To make us count the zeroes
Meant to spawn comments and improve engagement, mission accomplished 😎
@@claytoncoe838 why didn't I thought of that 🕳️
Matt is in the southern hemisphere so he should have pointed the Parkerscope toward the celestial south pole as that part is never lit by the sun. It also means the scope would have never needed to be moved. Or maybe on the line between the zenith and the celestial south pole.
And just for good measure, add an opaque ring around it to keep the entrance to the tube in shadow, just in case it wasn't perfectly lined up with the pole.
First
4th