I have a project from last year where I made my own little pokemon game using processingjs. I thought I had some decent knowledge of developing but the code consists of a file with 1900+ lines of code. It's terrible.. Also because I had barely any knowledge of how I should structure code.

In your post you initialize your weights a certain way but you said there are better ways. Do you have any resources for better ways?

I'm curious too, how would he set up so many machines that do the same thing?

It's very easy to terraform 1000 nano machines on ec2, as he said it wouldn't be expensive at all. In fact, he estimates $150, but for 1000 nano machines for 10 minutes i believe it's under a dollar. Possibly nano machines aren't powerful enough to spam the verify endpoint enough

200 curls can be ran from any ec2 instance. there is literally no overhead.

personally i would spin up a million lambda functions and just see what happens. each one def wouldn't get its own ip but maybe you would have good coverage?

Wouldn't you be hitting EC2 limits pretty quickly though? I guess you could get them raised, but last time AWS wanted a valid reason.

> but last time AWS wanted a valid reason.

Yep - my reason is normally "I'm planning a project that needs more instances"

I doubt they even look at that field to be honest

"load testing my website INSERT_VALID_WEBSITE_LINK"

The author said in the post comments that he used 1000 EC2 instances.

He's probably lambda rather than EC2 or even docker but I'd be interested to know the details too.

Docker doesn't help distribute requests over IPs, lambda or ec2 most likely.

You can easily set up a launch configuration for EC2 that runs a script or program... But with a bit more work lambda will save a lot of money

> Docker doesn't help distribute requests over IPs

That depends on your orchestration. AWS provide several tools for running Docker without having to maintain your own EC2 hosts.

> You can easily set up a launch configuration for EC2 that runs a script or program... But with a bit more work lambda will save a lot of money

You could do that via a launch configuration but that would be a pretty naff way of doing it. Baked AMI would be easier but personally I'd prefer ECS (Docker) or lambda. Cheaper, quicker to deploy, lower ramp up times.

Ultimately though, there's no wrong way to do this - just personal preference.

If you want to test a company look at their 'Responsible disclosure'. For example: https://www.facebook.com/whitehat

From personal experience: this is absolutely true. There is this csgo gambling scene on the internet. It's basically betting with real money but masked as a value of weapon skins. I have made some idiotic decisions in the past regarding these websites and I still regret it a ton. The problem really is that these websites also allow people of any age to gamble with their weapon skins (aka money), and it's fairly easy to buy new skins with actual money.

What's interesting though is that it's easy to observe a ton of gamblers for a long time. And since there is usually also a chat, interacting is also possible. These sites usually have 14 numbers for roulette where 1-7 is red and 8-14 is black, 0 is green which gives 14x your bet. Red and black give you 2x.

If you observe enough gamblers long enough (which I did), you see the same pattern over and over and over again. It goes like this: 1. Gambler starts out with a random amount of money (could be high or low). 2. Gambler wins big (usually like 1-3 times on green) 3. Gambler feels lucky and gives out money to 'passive' beggars* in the chat. ( Sometimes I would feel bad for the gamblers and hint them in the chat that they should cash out, no one ever listened ) 4. Gambler continues and might win some more, or lose some but not all. 5. Gambler has been betting for such a long time that he/she is back at where he/she started. ( Beggars keep begging but gambler says he is down so doesn't give out money right now ) 6. Gambler suddenly wins big on green again (or went all in) and feels like a king again. 7. Now Gambler cashes out (by buying a skin) the money with which they started and continues with what his/her leftovers. 8. Gambler loses the leftover money on the website. 9. Gambler deposits the skin that was previously withdrawn again. 10. Gambler loses everything now and the beggars focus on a different big gambler. 11. Gambler usually comes back within a week or month and loses more money. 12. The cycle continues.

* People, usually kids, that lost their (usually small amount of) money earlier and our now cheering for some big gamblers so these gamblers might send them money on the website.

I personally also lost a (for me) quite big amount of money on these sites over a decent period of time. Later, I was able to sort of stop my addiction and I made a bot which predicted random results (red/black/green) in the chat. It got me like 30 dollars of tips from people that had probably used the bot but the website changed some things so my bot stopped working and I have been too lazy to fix it.

Last thing: I also kept some stats of the roulette because I had to know the timing of the roulette so my bot wouldn't predict if it was already rolling. I added an extra feature for myself so I could see how many times red/black and green were rolled and how many times my bot got the predictions right. Obviously this always ended up in a big loss for my bot (in terms of predictions, I did not bet money with my bot) in the long term.