now that you read off the ChatGPT Hallucinations in your video about battling bugs next time It writes in the style of Code Bullet it will add more bug fighting. Which you will then read on camera and solidify it more as something Code Bullet would say until scripts by Code Bullet from ChatGPT is a full-on stories about the "BUG WARS". haha.
challenge: your next video on human benchmark cannot use AI programming tools, like chatGPT (basically no chatGPT or alternatives for programming, but you can use public/free libraries)
Very rarely it goes the other way, just write some code in two minutes and it just works. Then you spend an hour looking for the bug that are definitely there anyway only to not find any and wasting the time anyway.
Reminds me of when I tried to program a Sudoku bot. That piece of crap library had no problem deciphering a handwritten text, but you give it a single, perfectly captured number it completely craps its pants.
Sometimes watching Code Bullet reminds me of something my high school basic programming teacher said: "A computer is only as smart as the person using it." Then he told us that there were some dumb computers in here.
I love how this could have been done so much easier by just using a library that reads the contents of the number's HTML element right out of your browser, like selenium for python.
I came to the comments to make the same point. I do fear that in a world where people are thinking of resorting to ChatGPT to solve programming problems, we are going to have more cases where people don’t stop to think about how to make a solution /easy/ and instead say “f’ck it, it’s not my problem as long as it works” I seriously worry how code is going to be maintained in the future when this becomes the norm. The only way code will be able to be maintained is by getting more and more advanced AI to debug the things previous AI had generated and humans will have no idea what the hell is actually making things work anymore.
Even with screenshot, just using the characters that are used instead of tesseract would probably be faster. You just need a key value pair table, and sort by X then Y value. (So the numbers at the top all come first, and the ones at the bottom come later.)
Reading directly from the DOM would surely be easier. But in this kinda of content I feel like that would be more cheating than what's happening here. Because then you'll literally be reading the number from HTML, setting a value field and .click()'ing buttons
Absolutely love how when CB gets so excited the chair becomes an enemy. That’s true accomplishment. When you’re feeling so good.. you just NEED to prove it by beating on an inanimate object.
EasyOCR is another option for Image Text Extraction. Based on my experience, EasyOCR demonstrates higher accuracy in recognizing numbers/digits. However, it comes with a notable drawback, as it lacks support for a wide range of languages.
tessaract works for numbers fine if you change the config, for example "pytesseract.image_to_string(text_in_tile, config='--psm 6')" config 6 if for single number, without it it couldnt read it for me but with it it can easily with close to 100% accuracy.
really both options are overcomplicated. just use selenium and extract the number from html. Or if you're not married to the option of using python, just use JS
You inspired me to try to do the ones you hadn't done in the previous video. When I tried doing this one I ran into all the same problems at the start. The issue is that tesseract is optimised for text at about 25 pixels tall. The screenshot being taken makes it huge and it has a hard time reading it, thats why zooming out and taking a screenshot of more of the screen made it better. I did run into issues such as it recognising 7 as / or 5 as s but using psm --6 reduces the frequency of this. I had the same problem at level 18 and kinda gave up after a couple of hours because im not that good at coding. I am very happy that you posted this video and i hope you stick to uploading once a week.
You can do it with find a picture in another, then find all the numbers and sort them by X then Y value, so the numbers on the first line are all before the 2nd line.
@@satibel I ended up zooming out more and using isdecimal() to get rid of errors, spaces, and enters. It works really well now and I got to level 36. I am using the method you described to do the chimp test though.
gotta say, your voice us the No.1 reason why i watch every video of you that youtube suggest to me, doesn't matter what the video is about, i just love it. p.s. your humour is a close second place ;)
@@Connor-yn8pz The bell curve displayed is skewed (to the left). In order to be in the 50th percentile, you need to be at the point where the left half area = the right half area. Without a y-axis to see how large those small data points (both left and right of the large peak) actually are, we can't really tell if the ones on the right are the same height as the ones on the left. I suspect the displayed graph has a smoothing function applied to it to make it look nicer.
@@zecuse i wasnt arguing with the 50th percentile. Its still a normal distribution if a skewed one. And by your own wording hed be in the higher %tile because of skew. The smoothing function is definitely there if the number is righr
@@Connor-yn8pz He's still not guaranteed to be above 50% though. We don't know how high the right half of the graph actually is and therefore how much area it actually has. That's also what I was talking about regarding the smoothing. At best, each point over there seems to be the same height and the smoothing shrunk all of them down. We don't know if the valleys in the left side are the same size as the entire right side. We don't know how any of the valleys compare to the outlier peak on the left. The smoothing function has hidden this.
I gotta say... Code Bullet goes above and beyond in his videos! Who needs to use debugging consoles to scrape the page's layout and directly set the input value to the text you copied before, when you can just install and use four libraries to take a screenshot, apply a little post-processing to it, spend an hour to try your program, pray that it works without changing anything (aka "debugging"), and ultimately recognize the text from the pixel values so that after a small delay you send key press events corresponding to the characters in the text you extracted to complete the original task at hand? (I'd be lying if I said I didn't reread that to make sure it was actually supposed to end in a question mark. I'm actually still not 100% sure it is supposed to end in a question mark. Oh well.) In all seriousness though, this was an entertaining video and I enjoy seeing Code Bullet go through his projects, even if at times he is screaming in pain internally. I both 100% feel you and also can't help but laugh XD
~ 6:00 This is the exact excitement I get when my code works too. And just like me, outta context we're both excited over the silliest thing to other ppl: "FOUR! WE GOT A FOUR!" (relatable 100)
asking chat gpt to write a script to get a number for a screenshot when you could just get that number straight from the html. i love programmers!! i love that these are the people that i have to work with!!
😂 I like that analogy, because of how his code started controlling the mouse and was almost out of control as I picture the first person making fire having that same “oh f’ck!” moment straight after the jubilation ;)
Wow, this shows the dangers of using a sledgehammer for every problem. I looked at the page this test is on and it shows the numbers in plaintext on the page itself. So the whole thing could have just been a simple copy/paste job instead of all this complex screencapping to read images to get text from and such.
Eh, I could see it done easily in 3 or less lines of javascript if someone really wanted to minimize number of lines of code. But I resist the notion that the journey of getting to even a less concise working answer wouldn't have been as funny. There's plenty of room to stumble, especially for someone who hasn't done something like it before. Though the biggest difference would have been the absence of trying to bend the blackbox of whatever library was trying to read text from images to his will, which, admittedly, was a large portion of the video.
Hey bullet, glad to see you alive! This is one of the best promotions I've seen. Usually I only watch them to support the channel I'm watching, but this time I watched it while enjoying it.
I was born in Poughkeepsie, so weird hearing about it in the wild. I didn't have "Code Bullet trying to pronounce 'Poughkeepsie' properly" on my list of things I thought I would hear today.
The issue in the beginning is probably that the OCR is trained on text that's black on a white background. So though processing the image was right, you probably should've inverted it, lol
10:58 an easy trick to remember a number like this is to delete the 0s in your subconcious so that you only are thinking of 5896221 then you can create a separate stream of thought telling you where to put the missing 0s
Even though pi doesn't change, I bet people that memorize it would have a good chance at using their method to do well at this. I was working at a private school several years ago and one of the middle grade classes had a contest. The winner was a girl who memorized a few pages of ~16pt handwritten font. That's impressive as hell to me. Another student had about a full page, and the others were pretty far behind at a half page or less. I have no idea how long they had to memorize though.
You need to tell tesseract you are looking for numbers only, then it works much better. It is expecting text and seems to prefer a letter over a number with default settings. Also it seems that it prefers text a certain distance from the top and bottom of the screenshot if it's a single line.
Thank you for your patriotism sir. Good on ya mate. As much as I have hated all of Australia since the Cricket semi in 1999 - you do have a lot to be proud of you bastards.
Hey everyone, we made a game exactly like the chimp's memory game. We now have a DEMO on Steam called CHIMP MEMORY. In the full version, we'll have a leaderboard for the fastest times, allowing players to compete with the chimp that has already recorded memory times as fast as 210 milliseconds. If I'm allowed to share the link here, please let me know :) Thanks!
i guess he means in the sense of this challenge, for pi you have all the numbers for an infinite time, but these numbers are there for a couple seconds and are random.
Code Bullet: programs AI to take screenshot and read the number so he can copy paste it to beat Number Memory test. Me: makes manual screenshot and type the number into the textbox. We are not the same.
I think your issue with Tesseract being inconsistent is because the default page segmentation mode is set to pages of text. If you configure pytesseract to use page segmentation modes 6-9, it'll probably be more consistent.
Soooo....this one finally motivated me to have a shot at it myself. After some back and forth I actually got a good result. First tried it purely in JS, which didnt work because the input field needs an actual keyboard input to enable the submit button. Then I built a websocketserver in C# to receive the current number from JS and type it in. Worked fine, but a bit slow and I wanted it to be in one language, so I tried the screenshot+OCR method in C#. Got it to somewhat reliably work up until a single digit would show up on a new line. Back to the JS+server, optimized that a little and left it running when I left for a couple of hours. Last thing I might try again at some point is skipping the annoying and increasingly long timer. Too much for me as a non-web developer. The result: the time you get to memorize 216 digits is about 174 seconds and it takes about 5 hours to get there (typing the numbers not included). 😄
A lot easier way to do this would be to use something like the playwright or the selenium library to just grab the number from the HTML and then input it right away. It would also not be limited by the size of the window and could pretty much keep going until the browser crashes or something.
Secure your privacy with Surfshark! Enter coupon code CBDO for an extra 3 months free at surfshark.deals/cbdo
Ok
Looking sexy today
Nah
now that you read off the ChatGPT Hallucinations in your video about battling bugs next time It writes in the style of Code Bullet it will add more bug fighting. Which you will then read on camera and solidify it more as something Code Bullet would say until scripts by Code Bullet from ChatGPT is a full-on stories about the "BUG WARS". haha.
challenge: your next video on human benchmark cannot use AI programming tools, like chatGPT (basically no chatGPT or alternatives for programming, but you can use public/free libraries)
i love how each video he becomes more and more lazy with the coding, truly a Programmer.
now hes just missing programmer socks
Hey man, not all programmers are like that.
Some age like fine wine.
@@anonymoususer8517 Femboy CB arc 😲
@@FranXiTwe can only hope
Now all he needs to do is hire someone else to do it for him
“That took an hour, I thought it’d take, like, two minutes.”
This is basically a description of all coding.
E
Searching for this comment
He had the code in two minute and spent an hour debugging
Thanks ChatGPT
@@gabrieleymat6332coding
Very rarely it goes the other way, just write some code in two minutes and it just works. Then you spend an hour looking for the bug that are definitely there anyway only to not find any and wasting the time anyway.
using an AI to make a bot to beat human benchmark this is very human
✅️ I'm not a robot.
Y yall calling any program an AI now
@@methatis3013they just called chatgpt an AI. Are you stupid?
@@methatis3013chat gpt is AI tho
@@methatis3013 He used chat gpt to write the code
Reminds me of when I tried to program a Sudoku bot. That piece of crap library had no problem deciphering a handwritten text, but you give it a single, perfectly captured number it completely craps its pants.
E
It cant be, the number, its to perfect to be true. :D
@@EEEEEEEE Bowwow! This reply had me going for a giggle! 😂🤣🤣 I oughta' show this to my book club! 🤣😂📖🧑
@@EEEEEEEE Bowwow! This reply had me going for a giggle! I oughta' show this to my book club!
holy shit ive got the same fucking problem 😭😭
Sometimes watching Code Bullet reminds me of something my high school basic programming teacher said: "A computer is only as smart as the person using it." Then he told us that there were some dumb computers in here.
E
Only as smart means it's at maximum as smart as the user, it doesn't prevent it from being worse than that
lmao thats funny man
This human avatar has become seriously committed into having us believe he’s the real Code Bullet. Props to him!
E
@@EEEEEEEE STOP
i MISS THE OLD cODE bULLET
@@Scratchydoesmusic Do not interact with bots. Only report them.
"Fuck this, fuck you, fuck off, i'm done" - Sponsored by Surfshark
NGL that one sent me
E
Sent you where?
@@jc_art_ brazil
Ah yes, I must have heard wrong he's obviously not saying "I'm back".
@@mikejameson7678 Obviously, no way i could have miswritten that while laughing my ass off.
"I am not mentally prepared to have to think" - same man, same.
99% of a programmer's day is just "WHY DOESN'T THIS WORK??!?!!"
Oh I thought for sure we wouldn't see Code Bullet for another year. This video is a surprise to be sure, but a welcome one.
E
The song at 0:00 to 0:13 is called Men At Work - Down Under (Kryzon Remix) for anyone wondering
THANK YOU
your excitement when you finally got the program to recognize the number is truly infectious
I love how this could have been done so much easier by just using a library that reads the contents of the number's HTML element right out of your browser, like selenium for python.
Or make your life even easier by using javascript in the browser. Python is not the right tool for this.
I came to the comments to make the same point.
I do fear that in a world where people are thinking of resorting to ChatGPT to solve programming problems, we are going to have more cases where people don’t stop to think about how to make a solution /easy/ and instead say “f’ck it, it’s not my problem as long as it works”
I seriously worry how code is going to be maintained in the future when this becomes the norm. The only way code will be able to be maintained is by getting more and more advanced AI to debug the things previous AI had generated and humans will have no idea what the hell is actually making things work anymore.
Even with screenshot, just using the characters that are used instead of tesseract would probably be faster.
You just need a key value pair table, and sort by X then Y value. (So the numbers at the top all come first, and the ones at the bottom come later.)
Reading directly from the DOM would surely be easier. But in this kinda of content I feel like that would be more cheating than what's happening here. Because then you'll literally be reading the number from HTML, setting a value field and .click()'ing buttons
Absolutely love how when CB gets so excited the chair becomes an enemy.
That’s true accomplishment. When you’re feeling so good.. you just NEED to prove it by beating on an inanimate object.
12:40: 976 - Code Bullet
13:01: 970 - Code Bullet
976
976
976
976
976
8:20 I just thought of Bender's nightmare last night. My wife was looking for a '2' on a project she was doing and I said, "There is no 2".
EasyOCR is another option for Image Text Extraction. Based on my experience, EasyOCR demonstrates higher accuracy in recognizing numbers/digits. However, it comes with a notable drawback, as it lacks support for a wide range of languages.
tessaract works for numbers fine if you change the config, for example "pytesseract.image_to_string(text_in_tile, config='--psm 6')" config 6 if for single number, without it it couldnt read it for me but with it it can easily with close to 100% accuracy.
really both options are overcomplicated. just use selenium and extract the number from html. Or if you're not married to the option of using python, just use JS
AI is usually better with python and me in first year at uni doesn't know anything else
@@nmills3 This is a far better approach then relying on ANYTHING else. (I used to rely on clicking stuff manually until I discovered selenium lol)
Just using selenium or beautiful soup is both easier and faster and it just takes some imports and some string manipulation to get it running
You inspired me to try to do the ones you hadn't done in the previous video. When I tried doing this one I ran into all the same problems at the start. The issue is that tesseract is optimised for text at about 25 pixels tall. The screenshot being taken makes it huge and it has a hard time reading it, thats why zooming out and taking a screenshot of more of the screen made it better. I did run into issues such as it recognising 7 as / or 5 as s but using psm --6 reduces the frequency of this. I had the same problem at level 18 and kinda gave up after a couple of hours because im not that good at coding. I am very happy that you posted this video and i hope you stick to uploading once a week.
You can do it with find a picture in another, then find all the numbers and sort them by X then Y value, so the numbers on the first line are all before the 2nd line.
@@satibel I ended up zooming out more and using isdecimal() to get rid of errors, spaces, and enters. It works really well now and I got to level 36. I am using the method you described to do the chimp test though.
You are a legend making chat gpt do your programs
Never saw your face in a video before so all I can say is that you might be the first person I’ve seen on any platform who’s voice matches their face
I can only assume someone at surfshark had to approve this ad read and that, honestly makes me want to get a sub just for pure meme power
They typically do have to be approved. Look at what TomSka has gotten away with.
gotta say, your voice us the No.1 reason why i watch every video of you that youtube suggest to me, doesn't matter what the video is about, i just love it.
p.s. your humour is a close second place ;)
Dude, that Surfshark sponsor message xD hahaha well done dude :') That got me to laugh so hard AND made me actually watch a sponsor message
At this point it's worth watching every video for the banger stock music
I love how he tried to make a Python script with screenshots instead of just using Javascript to extract the number like a sane person.
Or pressing print screen
human benchmark is smart and uses usec refreshes to prevent persistent scripts. the number is also render in a canvas
That "dead on average" would be 50th percentile if it was a normal distribution, which it isn't.
To be fair, the graph shows a bell curve, with a very very small peak to the lower end. So that should be a normal distribution
@@Connor-yn8pz The bell curve displayed is skewed (to the left). In order to be in the 50th percentile, you need to be at the point where the left half area = the right half area. Without a y-axis to see how large those small data points (both left and right of the large peak) actually are, we can't really tell if the ones on the right are the same height as the ones on the left. I suspect the displayed graph has a smoothing function applied to it to make it look nicer.
@@zecuse i wasnt arguing with the 50th percentile. Its still a normal distribution if a skewed one. And by your own wording hed be in the higher %tile because of skew. The smoothing function is definitely there if the number is righr
@@Connor-yn8pz He's still not guaranteed to be above 50% though. We don't know how high the right half of the graph actually is and therefore how much area it actually has. That's also what I was talking about regarding the smoothing. At best, each point over there seems to be the same height and the smoothing shrunk all of them down.
We don't know if the valleys in the left side are the same size as the entire right side. We don't know how any of the valleys compare to the outlier peak on the left. The smoothing function has hidden this.
One of the few adreads I have watched in a while…wunderbar.
Good thing Code Bullet can memorize up to 9 digits, it's certainly handy for memorizing coordinates!
I normally stop the video at sponsored as time. But this reading was funny as hell. 😅
I like how he is always looking at himself when he is talking with the audience instead of in the camera. 😂
0:18 “new year new me… I’m going to try to keep uploading” just a couple of months and we have the new new code bullet😂
That is the best ad segment I’ve seen this far!
I do miss when CB actually coded and had to struggle with everything rather then have chatGPT do the programming
Me too man
“Every f’kin time…”
This is Code Bullets day off though
It happens on the main channel to though@@seansims4518
Why you tryna make my man work on his day off?
Fr chat gpt is cheating
CB’s personality is so magnetic i love when he gets excited
i love how ur making a bot to play *"human"* benchmark
the bot is very human
The design is very human
??????? yeah bro thats the point
thats what i thought too xD@@Nae_Ayy
@@Nae_Ayy you don't get it
that intro was something else
12:48 ngl CB has very advanced way of memorizing number
I didn't even know he had a second channel, or expected this is what he'd look like. I absolutely love both of these discoveries 💛
"i didn't know this was detachable"
"how do i put that back"
"alright so this is just separate now"
using any datagrips software in a nutshell lol
In the eleventh hour as I am rewatching old videos…HE RETURNS 🎉🎉🎉
My greatest fear now is Code Bullet teaming up with I did a thing. That much Australian power cannot be contained.
3:05 😆 The Tesseract path bug patch spotted in the wild 😆 Good times, man!
I love that in memorizing the pixel positions of the buttons for his AI to beat the game he's unintentionally playing it..
Heckin dang it CB, stop making such hilarious content because then I have to watch it
17 minutes of code bullet looking at a computer writing numbers down. Peak content.
This channel is called Code Bullets Day Off, so it makes sense that ChatGPT does all the work for him.
I gotta say... Code Bullet goes above and beyond in his videos! Who needs to use debugging consoles to scrape the page's layout and directly set the input value to the text you copied before, when you can just install and use four libraries to take a screenshot, apply a little post-processing to it, spend an hour to try your program, pray that it works without changing anything (aka "debugging"), and ultimately recognize the text from the pixel values so that after a small delay you send key press events corresponding to the characters in the text you extracted to complete the original task at hand?
(I'd be lying if I said I didn't reread that to make sure it was actually supposed to end in a question mark. I'm actually still not 100% sure it is supposed to end in a question mark. Oh well.)
In all seriousness though, this was an entertaining video and I enjoy seeing Code Bullet go through his projects, even if at times he is screaming in pain internally. I both 100% feel you and also can't help but laugh XD
Your energy is why I come here, I don't care about the programming!
Says he’ll try to keep up and he instantly stopped
~ 6:00 This is the exact excitement I get when my code works too.
And just like me, outta context we're both excited over the silliest thing to other ppl: "FOUR! WE GOT A FOUR!"
(relatable 100)
you're my favorite youtuber honestly
asking chat gpt to write a script to get a number for a screenshot when you could just get that number straight from the html. i love programmers!! i love that these are the people that i have to work with!!
The program finally working was the same energy as the first men had when they created fire. Straight visceral joy haha
😂 I like that analogy, because of how his code started controlling the mouse and was almost out of control as I picture the first person making fire having that same “oh f’ck!” moment straight after the jubilation ;)
Wow, this shows the dangers of using a sledgehammer for every problem. I looked at the page this test is on and it shows the numbers in plaintext on the page itself. So the whole thing could have just been a simple copy/paste job instead of all this complex screencapping to read images to get text from and such.
Yep, JavaScript to read the number direct from the DCOM and voila - probably 10 lines of code and 10 mins…but it wouldn’t have been so funny then ;)
Eh, I could see it done easily in 3 or less lines of javascript if someone really wanted to minimize number of lines of code.
But I resist the notion that the journey of getting to even a less concise working answer wouldn't have been as funny. There's plenty of room to stumble, especially for someone who hasn't done something like it before. Though the biggest difference would have been the absence of trying to bend the blackbox of whatever library was trying to read text from images to his will, which, admittedly, was a large portion of the video.
Hey bullet, glad to see you alive!
This is one of the best promotions I've seen. Usually I only watch them to support the channel I'm watching, but this time I watched it while enjoying it.
I learned everything I know about machine learning from this man...
I was born in Poughkeepsie, so weird hearing about it in the wild. I didn't have "Code Bullet trying to pronounce 'Poughkeepsie' properly" on my list of things I thought I would hear today.
a couple years ago you didnt even need a bot to beat everybody, you could just select the number and copy paste it.
The issue in the beginning is probably that the OCR is trained on text that's black on a white background. So though processing the image was right, you probably should've inverted it, lol
Only a human would be so lazy to replace itself with ai by will. Its more human than a human naturally doing the test.
7:36 i feel that “i changed nothing” 😂
Yep - and sometimes understanding why that happens is the key to unlocking the real answer.
That was hilarious with having Chat GPT make the words for your sponser SurfShark.
10:58 an easy trick to remember a number like this is to delete the 0s in your subconcious so that you only are thinking of 5896221 then you can create a separate stream of thought telling you where to put the missing 0s
its not always a 0 though, i usually subconsciously delete repetitive numbers with a pattern so its easier for me to add them back in
Even though pi doesn't change, I bet people that memorize it would have a good chance at using their method to do well at this.
I was working at a private school several years ago and one of the middle grade classes had a contest. The winner was a girl who memorized a few pages of ~16pt handwritten font. That's impressive as hell to me. Another student had about a full page, and the others were pretty far behind at a half page or less. I have no idea how long they had to memorize though.
THAT is what he looks like, my guy is a model
Congrats on winning the bet !!!
Homie knows programming but keeps looking at Camera 2 when the recording is being done on Camera 1
i am never gonna get used to seeing cbs face omgggg ❤
I read the last 2 word of the title as number *theory* and yell "oh yeah, something computer is really good at!"
Oh, you won the bet? Awesome! I worried about that because you stopped uploading for a while
PLEASE keep the weekly uploads I need more codebullet content
Did code bullet have his cybernetic implants removed? He looks like a totally different person now. I was not ready for that.
i love how you play a recording of you backspacing your prompt backwards lol
this guy entrusts jarvis with his computer... cramzy!!!
Lol I grew up near Poughkeepsie. You had it close the first time
WTF, did not know that you had a second channel, got to say it, your handsome as hell bro
The unexpected return!
12:54 cant wait for a full remix!!
Using AI for surfshark ads is the best use of AI for ever
I am so jealous of this man’s voice and ability to grow facial hair
2:12 Code bullet not knowing the difference between median and mode for skewed data
13:30 Laurence Kim Peek enters the room! its in capable hands bullet.. enjoy ur movie 😂👊
'it works! it works! i changed nothing! and it works now!' - everyday life of a programmer
You need to tell tesseract you are looking for numbers only, then it works much better. It is expecting text and seems to prefer a letter over a number with default settings. Also it seems that it prefers text a certain distance from the top and bottom of the screenshot if it's a single line.
Thank you for your patriotism sir. Good on ya mate.
As much as I have hated all of Australia since the Cricket semi in 1999 - you do have a lot to be proud of you bastards.
I mean you just could take a picture and do it this way but instead spending 2 hours pasting code from chatgpt I just love programming
Such a good ad read
9 was the Mode of the distribution, not the median. The head of the distribution only goes back to 0 while the tail was pushing in the 30s
title should be chat gtp vs the human benchmark test with how much you used it lmao
Hey everyone, we made a game exactly like the chimp's memory game. We now have a DEMO on Steam called CHIMP MEMORY. In the full version, we'll have a leaderboard for the fastest times, allowing players to compete with the chimp that has already recorded memory times as fast as 210 milliseconds. If I'm allowed to share the link here, please let me know :) Thanks!
When he said it would be a problem for a human to remember 16 digits just remember some people have memorized 7000 digits of pi
i guess he means in the sense of this challenge, for pi you have all the numbers for an infinite time, but these numbers are there for a couple seconds and are random.
Code Bullet: programs AI to take screenshot and read the number so he can copy paste it to beat Number Memory test.
Me: makes manual screenshot and type the number into the textbox.
We are not the same.
I think your issue with Tesseract being inconsistent is because the default page segmentation mode is set to pages of text. If you configure pytesseract to use page segmentation modes 6-9, it'll probably be more consistent.
haha 69 nice
New Title: Code Bullet Discovers He Has Discalcula
Soooo....this one finally motivated me to have a shot at it myself. After some back and forth I actually got a good result. First tried it purely in JS, which didnt work because the input field needs an actual keyboard input to enable the submit button. Then I built a websocketserver in C# to receive the current number from JS and type it in. Worked fine, but a bit slow and I wanted it to be in one language, so I tried the screenshot+OCR method in C#. Got it to somewhat reliably work up until a single digit would show up on a new line. Back to the JS+server, optimized that a little and left it running when I left for a couple of hours.
Last thing I might try again at some point is skipping the annoying and increasingly long timer. Too much for me as a non-web developer.
The result: the time you get to memorize 216 digits is about 174 seconds and it takes about 5 hours to get there (typing the numbers not included). 😄
7:35 is the most relatable programming moment .. ever.
do you guys remember weekly uploads? haha me neither
A lot easier way to do this would be to use something like the playwright or the selenium library to just grab the number from the HTML and then input it right away. It would also not be limited by the size of the window and could pretty much keep going until the browser crashes or something.
finally another video mr code bullet where were you?