Really, you think the DEVS got to pick what system to use?? I work in Tech too, 9.9/10 times it is upper management/seniors that pick that crap to save a few bucks.
1 Like
Sirrian is the upper management in this, and yes I think he had the choices.
If the dev’s have no input on what database to use in an application, its already likely to fail. That is if you have competent devs.
1 Like
it’s not quite that simple. They likely afford to go bigger/better.
When you’re a small house organization, yea the devs have a say. When your organization has management that know their ass from a hole in the wall, the tech people have a say in things and the management listen to them.
1 Like
This. They’re a tiny company. It’s highly unlikely software choices were made by some bureaucratic process, considering they’re too small to even have a bureaucracy.
Since they are small, I’m sure there was budget constraints…hence the crap system.
The Pay-To-Win Rangers have given them plenty of cash.
Hey all this natter above your comment isn’t entirely fair. Our current problems stem from the development team actually trying to upscale to something bigger and better yes?
This isn’t fair. This is Sparta.
They are the creators of Puzzle Quest, no? Why in the world would they think their game wasn’t going to be successful?? 
1 Like
The only things wrong with Java and Adobe Air are the Memory leaks. They are resource hogs and always seem to have issues of freeing objects even when unloaded/unset properly.
Although, that is not the biggest issue with this. I would have to say MongoDB is the worst decision on this, I have not ever worked on 1 successful write intensive application that scaled well with Mongo. Mongo is great until the point where its not, and that point can be reached very quickly.
You can go from 25% server usage with 100 concurrent users to 100% usage with 105 users. Scaling of it is possible, but you lose so much from each scale that its like 40% increase per additional instance, it becomes a major headache. Lets also not forget that Mongo does not work well if the Entire database cannot be cached in Ram.
While I most likely wouldn’t have built this game this way, it is what they are stuck with. So try throwing 2x the resources (ram/cpu/disk io) at the database and see if that makes a difference. If it solves the issues, then you know where the problem is.
2 Likes
Our database is undergoing an upgrade at the moment. This should be finished in 1-2 hours. We apologise for the dropouts that continue in that time.
In the meantime, we believe we have come across an issue that has occurred because of the new servers. We will look to resolve that asap, but also want to be careful that we don’t corrupt or lose any data in the process.
10 Likes
@AeroCloud not to attack you or insult you, but you reminds me of an old computer friend who thought he knew everything. I must just tell you, it must be quite annoying to have people like you around acting like ‘knowing better’ for the devs, when they are working with fixing a problem.
4 Likes
Was I right in suggesting above that the backend was basically rented from the Parse service originally, before it got shut down (and possibly even before Facebook acquired Parse)? If so, a key question would be whether or not Parse told people that they were built on MongoDB, or if they told people that they’d solved Mongo’s scalability problems for all of their customers.
If Parse wouldn’t tell people, or they said “don’t worry, we’ve tamed Mongo”, or “you needn’t worry, we’ll handle all the internals on pain of breach of contract”, then that’s a little different.
Regardless, I would feel a little aggrieved if I hired a software service that appeared to be solid and scalable, and then later got told “oh, hey, we’re shutting down our service – but don’t worry, you can go run it yourself with all of the knowledge of our previously proprietary software that I’m sure you can easily get!” Even if I did conclude afterwards that, in hindsight, I should have reasonably foreseen that using their service was a risky proposition.
@Eika I appreciate your comments and I wont hold anything against you.
I do not know everything, however I have been working with Databases for 20 years, also worked on Social Networks, News Organizations and other Database Heavy systems.
I have even written my own games that supported 1000’s of concurrent users on much less powerful systems than we have today.
So while you may think I know it all, I don’t… but I do know quite a bit about this having been a programmer for almost 30 years and DBA for 20+ along with Linux Administrator and other IT hats including Networking.
2 Likes
No need to share your CV with me. I just told you what I thought of it, after reading your posts.
1 Like
For my Listintira account on the cell phone the game appears to be a band-width hog even in the best of times.
I remember when I started playing this game a year and a half ago, it would take about the same time for both accounts to start. Listintira might have been a little slower, but not seriously noticeable. Now while my Stormchasee accounts starts up pretty quickly (within 10-20 seconds) before I can do something, it takes the Listintira account 3-5 minutes to start. It seems like the game goes through several steps in the start up process. I see it loading the world, hero and assets. Why does that data have to be downloaded? Why can’t that be on the device and have a quick cross-check with the data on the server?
You’re absolutely right Eika. Trying to guess helps noone. I’m going to delete my comments as well, just because I feel like it’s disrespectful to even say, while I know they are working hard to fix this.
Even though my comments were nothing really. 
1 Like
If the game was working, I wouldn’t have the time to post here.
If I didn’t care about the game I also would not post here.
3 Likes
Guys it let me log in. Something’s wrong, I should leave it alone now! I mean, it hasn’t let me do that in hours, something’s up right?
1 Like