They had no problems taking everyone’s money. Maybe companies should limit the number of sales when deploying a product tied to services they operate and need to scale.
They had no problems taking everyone’s money. Maybe companies should limit the number of sales when deploying a product tied to services they operate and need to scale.
The peak concurrent users for the game thus far has been less than 1/10th of that (EDIT: slightly more than 1/10th of that). They were well within the bounds of what they simulated. They just screwed up.
Where are you getting that number?
Yeah from what I hear, they kept the shitty self-downloader of the first one that is obscenely slow and counts as “playtime” because the game is open while downloading.
This is on Microsoft, they can make as many excuses as they want, but they fucked it up.
It’s ok tho they’re making server complexes out of wood now!
/s
Didn’t account for the initial interest peak.
No, that’s what I’m saying. That peak is well under what they claimed they simulated.
Peak concurrent was marred by the game not being playable. It’s not really a good indicator here.
It is, because everyone trying to download assets from the game servers were doing so from the executable that Steam tracks as running.
You don’t have to buy it from steam, might’ve been more users from gamepass and xbox than steam?
It is a flight simulator, somehow I doubt it has more console than PC users. Consoles are just too limited to satisfy that particular demographic. Can’t even connect half a dozen different input devices to a console.
Dunno what the sales for the new one are like, but the 2020 ms flight sim had a lot more mass appeal than any older flight sims, but maybe a lot of xbox players were put off by the pre download requirements?
Also when it’s on game pass the barrier to entry is a lot lower, as a lot of people might install it with the intention of only playing it for like 5 minutes to see if they can see thier house or something, which they probably wouldn’t pay full price on steam to do 😅
But yeah it definitely sounds like thier test for 200k players probably had bigger instances provisioned than what they ended up using in the production release 😅
You’d think they’d be able to do some sort of auto scaling when demand goes up though… 😅
Perhaps, but 10x as many? I suppose it’s possible, but I don’t think it’s likely.
It’s not, because downloading assets was not working, and with it not working eventually the user will shut it down thus having an effect on peak concurrent. It’s not even that complex. If people can play your game they will stay on it increasing the count, if people cannot play your game they will not, thus decreasing the count.
Exactly. The peak number of people trying to download it simultaneously was about 24k. They didn’t all stick around because it wasn’t working, even though that’s about 1/10th of what the devs expected and prepared for the demand to be. They didn’t get anywhere close to 200k people all hitting that server at once.