Zaero wrote:
Terminus1066 wrote:
sevion wrote:
Zaero wrote:
silverfoxdmt73 wrote:
Pete17331 wrote:
IT guy here. Yes, it is perfectly possible to prepare for this kind of workload. For example, companies like EA Sports does it every time they get ready to release a new game. They outsource and get enough temporary processing power (in the cloud) to handle the predicted workload.
I imagine companies like EA Sports have more than 50 staff across their IT department, never mind the whole company and a few $billion to throw at outsourcing IT.
Fellow IT guy here. You can scale, sure, but if you don't have the hardware or the ability to invest, that scalability doesn't amount to a hill of beans.
EA has (I would argue) a far bigger resource pool to work with. But hey, obviously everyone has infinite servers to deploy at any time.
Thats why for a company like this they should be using Amazon WS. Sure owning hardware is cheaper in the long run but now instead of letting the service spawn a few more servers for you or you do it yourself they now have angry customers who may or may not have been future clients at full price or monthly subscribers to other services offered.
Agreed - building an infrastructure that doesn't scale may be easier in the short term, but eventually it's going to bite you in the ass, which is what has happened here.
It's all dependent on need. Sure it's easy to say "just use X" or "get more servers", but you ain't running the company. It may make more sense for them to host in house. You don't know that.
Playing Backseat Architect doesn't quite work. I mean, even Steam tips over when a new sale starts, and they're MASSIVE. No online service is perfectly scalable to meet demand. Yea, people are mad they can';t get their books NOW, but it's not like Paizo's going to run out of copies. This ain't TVs on Black Friday.
I'm being patient. Would I like my books now? Sure. But I'd also like a giant novelty...
Sure its dependent on needs and a company of this size really isnt saving any money in the long run housing its own servers. A company this size will actually find it cheaper and more effective to use Amazon WS so it may scale with its usage. A single hardware failure will set them back double what they are currently saving using in house dedicated servers vs using Amazon WS and not having any of these issues or having to worry about a unknown costs of hardware failure that are just a matter of when not if.
I do this for a living and I have to admit when I was first designing games and setting up servers I made these same mistakes. Hell I continued to make them for a good year and a half after cloud computing began to get big. But just as I needed to adjust scripts and functions to provide a better service on an archaic system so should they adjust to current issues. And much faster than they are, this fiasco instead of finding more users for life could find themselves with a angry group of paying customers who will now share their issues and may lead to others not moving to your product at all at a later date via word of mouth.
I really hope they drop this whole personalization watermark thing. Or personalize it with a generic mark. Its really useless (I dont know if they know how easy it is to remove those things but it is very) and its killing the experience for fresh new players.