I’m not sure what part you don’t understand.
Printable View
That would be your definition of the word competent. I get it, the launch has had some serious problems, while it's not fun for anyone it's also not without precedent as many many game launches have similar issues.
I am not a server engineer however i have played video games since long before online gaming was even a thing, this is not my first rodeo. It's quite obvious to me that these problem happen because if a company provides a server infrastructure so big that everyone has a flawless experience when the servers are at their absolute maximum stress (launch week) then the company is losing serious money for every other week of the games life.
Your position seems to be that SE are simply incompetent, this is a position fundamentaly at odds with the reality of the fact that we are talking about a company that successsfully run not one but two MMO's for 20 years. Nobody should be taken seriously if they suggested that Rolls Royce don't know what they are doing if an engine goes bang, that would be patently absurd.
Your position is clearly so extreme, agenda driven and bias that at no point should it be taken seriously.
I hope that clears up my confusion.
Cease this inane bickering and let Square Enix handle their own server infrastructure. The Final Fantasy series is no place for cloud strife!
Cloud services turn online games into laggy trash. XIV is far more stable than, say, PSO 2 and Azure is the BEST option.
Everyone is an expert, peacocking.. eyeroll.
Funny you posted this and then all of the Pacific coast AWS servers went out one day later lol
Why would Amazon host their MMO's competition btw?
They could feasibly move SOME of their infrastructure to the cloud (like lobby servers and other non critical stuff) so it could scale and at least let people login and queue. The game servers could come later. Doesn’t have to be done all at once. :)
Can they? How would this affect the actual queues if the connection now has to go from their authentication databases which likely are in the same building or at least campus as the other servers, out to a remote lobby to wait, then handshaking back to the data center and world servers. The way the servers work with the login and lobby servers would have to be changed. Those changes would take time to write, test and inevitably implement.
This would introduce new errors, new problems and new calibrations necessary to tolerances like timeouts and error correction. Security concerns are also something to consider. Rather than authentication being mostly local, it'd now have possible vulnerabilities that weren't present before.
So while eventually POSSIBLE.. the question of feasible is still up to SE, and they have said no.
The difference here though is not that this is a launch specific thing, they had the exact same issue this spring when there was a mass influx of players; at least in europe there has always been a queue since around april/may this year (despite small about 200-300 people over all, which practically meant around 5-10 minutes queue time).
Even at that point it was known the server capacity was too low, to mitigate this SE stopped selling final fantasy. Now they obviously sold a shit load of endwalker licenses and this includes not only a influx of new players but a huge load of previous players that been on hiatus awaiting the new expansion. At the same time they've been driving advertisments and made the "trail/free" version a lot more generous in hope of getting additional people hooked. While at the sametime making little to no efforts in expanding the server infrastructure, now the official reason is the pandemic; yet YoshiP just recently said they _will_ expand the current datacenters with the hardware they _already_ secured, and you white knights still believe this company.
I’ve no idea, but in order for them to do this stuff requires ALOT of investment in terms of time and money (POCing stuff, re-architecting code, building cloud resources, etc etc) so it’s not a drag and drop your stuff on cloud and boom it works.
So given the age of their code and game engine and all that stuff, it’s probably not financially viable to even ever do it for 14 - hoping that people finish the msq and go play something else is the much cheaper option.
The time to do it would be for a brand new MMO. So they may do it for 7.0 or most likely won’t ever.
Among all the technical ignorance on display lately, I'm just waiting for someone to take the next step and suggest blockchain.
So.. using your own word: Feasibly.. they cannot and would not. The restrictive factors being cost and compatibility (likely). That's all I am suggesting.
It's not that they don't want to improve service (improving service accessibility increases profits) or that they are trying to cut corners... just that they've likely weighed and measured cost, pros and cons and came up with it not being an option.
Meanwhile OP stands there and acts like they don't even know about the option and it's a viable option to fix things RIGHT NOW.
Feasibly they could look at doing it yes
Yuck, my post broke.
I mean it’s feasible they could do it, but it would take time and it depends on what their budget is and whether they think it’s worth it. It’s not like they can drop their servers straight onto google and say “yay it works” just like that :)
.. to fix things immediately? And not spend excessive amounts of money on something that won't be necessary in a few weeks? And not possibly cause trouble later or increase the amounts of things that could go wrong that they won't have control over or have direct access to?
I think you and me think of the word feasibly differently.
I can technically get on a plane tomorrow and cart myself to the other side of my country, find a new job and residence and live comfortably.
It is not however, feasible.
Sorry, my post broke halfway through. I have a better answer above. :)
Nope, that is not the case. It was a spaghetti-coded mess in 1.0. They remade the entire game to get rid of the spaghetti code. Of course, the game has technical debt (as any large software in existence) but that is very different from spaghetti code. Even Yoshida said himself in a pre-EW interview that this spaghetti code argument is false.
I do understand what you're saying.. and I agree that it's possible, I just cannot agree with your choice of words. Language is incredibly important to me to convey what one means exactly and feasible doesn't apply here as far as I can measure.
Practically and functionally, cloud servers seem to be a large undertaking both financially and labor resource wise. Something SE has looked into and decided against. As I cannot say I am privy to any of their reasons I have to accept that a company of their resources, talent, and experience combined with the passion that the FFXIV team has shown to try and deliver a certain level of experience, has not dismissed them as an option lightly.
I like constructive criticism, it's helped the game a lot. Things got changed due to aging GUI and such (remember the Coke problem or token turn ins, etc...)
But I'm not gonna lie - it is kinda annoying that so many think now with their Comp-TwitterIA certs are coming in and thinking they know way more than the guys who actually have to run the game servers.
Some things may be better defaulted to those actually dealing with the logistical nightmare than thinking they have the heads up after reading a reddit post or running a home server.
AFAIK, they cannot block SE from hosting, at least not on the EU, if they did, they would have the EU Comission on then with a huge fine.
They recently got a fine from Italy of €1.1 billion, from not allowing all sellers into a program (dont know all the details), now immagine if somehow they dont allow someone from using their services because of competition.
That's the thing, they didn't completely "remake" the game. The game is built on the corpse of 1.0, and to do that quickly, they lifted whatever seemed to be functional and kept what they could. The inventory system is a mess because it's rooted so deep. Character creation has limited expandability.
My previous statement might be hyperbole, but it's far from false.
remember when the top scientists told us to do this with C and then NOT TO DO THIS.... and then they said it was fine to do this and then NO DON;T do that and then 3 months later oh no you should do that... i won't go into details but as someone who works for the NHS and with c-19 I can tell you the TOP experts don't know jack most of the time.
Assuming they actually run their code on CISC-based CPUs. If not (and the lack of available hardware seems to suggest this), all of the Cloud Architecture in the world isn't going to help one bit. Well, aside from running Stromasys on top of a bare-metal cloud server.
[This one has worked with physical RISC-based systems for decades, and Cloud-based options for the last 5 years. Want to purchase something from Sun/Oracle (or a cloud solution with any of the major vendors)? Come see me.]
As pointed out, they said they looked into it and found it to be unreliable. I haven't seen any superior evidence against that in this thread.