Hello forumsmotri people, we saw your discussion about us at 4080-pomogite-skachat-vidos#com3539 and actually tried to join you there, but noticed you have a closed registration, so, we will talk to you like this ;)
First of all, thanks to Ivan Falafel for getting the facts straight.
These are translated quotes from the user Ivan Falafel (if you don't want us to quote you, please contact us and we will remove them)

the traffic that was a year ago, and that they have gone now (even for the same queries in this topic can be seen), disproportionately higher. Moreover, I began to notice them already in search engines, i.e. began to be indexed.
We are having a tremendously hard time getting crawlers to actually index us, simply due to the huge volumes of page additions and removals, that they seem to think something strange is going on.

On the bare enthusiasm, it is unlikely that anyone will want to support all this, it was possible to understand both the introduction of mining and paid subscriptions
The project primary goal is: change the game of how webcam models perceive recorded shows of them, not as something negative rather than something that could help them instead of being everywhere, yet completely useless to them as they got nothing from it.
Because this would also lead to the fact the we could become the one stop source for every user seeking for any show from any model.
So it would be a win-win-win situation.

But apparently they can not optimize the DB for them, so they turn the nuts. mb will be returned as it was, if the solution is not effective.
To be honest, the translation doesn't make much sense, but about the optimization part: it's not an optimization problem, we build our own back-end systems which includes also our own database engine based on PostgreSQL and that it a huge amount of work and since we are also the only user of that engine, if we don't detect a bug before putting it in production, well, bad things happen.

given that they have for the last 2 months already three times database drizzled, apparently they do not maintain the load at first screenshots generation auto-update was changed to 5 minutes, now and decided to hide at all
As mentioned before, that is not a load or optimization problem, rather than critical bugs, we haven't caught before they went into the production system.
The video page refresh rate was an optimization, we run the screencap servers in an interval and batch requests to run them more efficiently, by expanding the time, we can utilize the servers better and don't have you think its only going to take a minute to generate screencaps for a 4 gig video.
We decided to hide everything because we saw a huge amount of scrapers start to take advantage of our huge dataset and this could have ended in a lot sites with bad intentions that only came to life because of us, which is not something we wanted to accept.

This might be just a minimal help, but if you get a volume account and don't drain it to 0, you can use it forever to view images (except if you use it to scrape us too, then you are going to get banned)

Apparently, they could not optimize. I do not think that there is a question of money and lust for money. Just not enough knowledge and skills, I think.
As mentioned at the start, we do this because we want to change things, if it was about the money, we wouldn't have started this project in the first place as it is quite expensive to store so vast amounts of data and we do not expect the project to become profitable anytime soon for us.
I would like to argue that we are skilled, maybe not the best skilled ones, but we do have a background in what we do, but we took a very radical approach when starting this project and that is developing every system we need on our own instead of trying to get 3rd party software to work together also because to run this project more cost effective.

We like to discuss things, so if you would like to ask us anything, feel free to contact us.

We hope you have fun and find what you are looking for.
Cheers