Letter from the Producer Live Part IX (10/19/2013)

[Xanthe]

I think the main issue is that they have been QAing these additions with the 2.1 build, and don't want to risk breaking things by dumping them into the 2.0.x codebase. I don't have a link, but I remember a dev post mentioning something to that effect.

[Kirbyprime]

Except changing xp gain wouldn't even be a code change, it should just be a query to change the XP multiplier in the dungeon, or the base XP of the various mobs in the dungeon, both of which should be database changes not client or server.

[Deuce]

With this past week's maintenances I'm not so sure they can easily hotfix anything in this game or even understand when is a good time for patching the servers.

From this week alone we've seen that any patch they perform requires anywhere from 2-4 hours to perform, even one as simple as removing a special event. On top of that the servers are usually full of lag for hours after they're brought back up. I hate to imagine what a major content patch would be like, but I imagine there'll be some ridiculously long maintenance window which will end up breaking multiple things requiring several emergency maintenances.

So if you ask me, I'd prefer a mega patch versus multiple patches because if it means dealing with fewer latency issues and less server downtime then I'll wait. Until they can fix whatever clusterfuck that is their backend server architecture the fewer the minor patches the better. Hopefully with patch 2.1 they have some server ninja fixes planned because they seriously need them, but I won't count on it.

[Grey Jorildyn]

SE ought to devise a set patching schedule and just stick to that. WoW did this for years (still does?) by always doing maintenance or patching on Tuesdays iirc. I didn't play so I just know this colloquially but I think it's accurate.

[Obsidian]

Except changing xp gain wouldn't even be a code change, it should just be a query to change the XP multiplier in the dungeon, or the base XP of the various mobs in the dungeon, both of which should be database changes not client or server.

I think assuming anything with regards to how Square is communicating with whatever databases it uses is a dangerous game to be playing. Even then, who knows what fields need to be changed or if the query needs to be restructured. Just because it sounds easy or "should" be easy doesn't mean it is.

[Hyan]

Mega patch means more shit added than multiple patches. dunno why people still dont get it.

I'll always vouch for the schedule that means more shit added. I guess people just want to think it can be done differently from their armchairs.

[Niiro]

Game launches, people sign up for 3 month subs.

People get disappointed by lack of content.

"Eh, I'll see what the first major patch delivers before I decide whether to re-sub."

First major patch delayed until after the 3 month mark.

[Elcura]

Mega patch means more shit added than multiple patches. dunno why people still dont get it.

I'll always vouch for the schedule that means more shit added. I guess people just want to think it can be done differently from their armchairs.

Don't you ever get tired of saying dumb shit?

[kisada]

Game launches, people sign up for 3 month subs.

People get disappointed by lack of content.

"Eh, I'll see what the first major patch delivers before I decide whether to re-sub."

First major patch delayed until after the 3 month mark.

wtf, i have no idea why this isn't more obvious to some people

[Seravi Edalborez]

Remember the ~3 month wait before Jobs, Cutter's Cry, and Aurum Vale? When the only content was Ifrit and Moogle?

Yeah I didn't burn out over that at all. *rolls eyes*

[Lucavi]

wtf, i have no idea why this isn't more obvious

Because Hyan is being Hyan. Its perfectly obvious for people who weren't brainwashed by XI's or Vanilla WoW's long-winded update cycle.

[Hyan]

Don't you ever get tired of saying dumb shit?

lol so mayad always. c'mon dude. how hard can it be to ignore me? you're the only one having trouble here.

its easier to manage everything when you dont have multiple patches to prepare, just one. obvious as shit, but the denial is strong as always.

[Kirbyprime]

I think assuming anything with regards to how Square is communicating with whatever databases it uses is a dangerous game to be playing. Even then, who knows what fields need to be changed or if the query needs to be restructured. Just because it sounds easy or "should" be easy doesn't mean it is.

At the end of the day, it's useless since SE does what SE does, but for the sake of having something to type...

We already know that SE uses a form of base / multiplier for XP calculation. They gave a pretty good breakdown of that back when they adjusted DoL exp and gave out the old vs new xp values. It is a fairly safe assumption that such values would be stored in a DB of some sort and that these are easily changeable.

Few assumptions, given that DoH exp can go to 4000-5000xp per synth, if we were to base that DoW functions similarly, they should both have the same type so there shouldn't be any issues with incorrect types or overflow, that is also assuming that not all exp gain is calculated using the same functions since DoH can gain close to 200k per synth.

The real issue would be how the servers sync up this information. Depending on the architecture, there's a chance that the base exp is only grabbed once on server start-up and that unless it is restarted it won't grab any new values. Since we don't know what happens during maintenance, it's possible that the actual servers aren't shutdown during this time, or that only some servers are shutdown while others are perpetually on. If this is the case, especially since JP and NA data centers are separate, it could very well be that 2.1 will be the one real time where all the servers are down at the same time and thus changes could be implemented at the same time.

This could very well be worked around by having created a "refresh" broadcast call worked into the architecture beforehand, but judging that ARR was fast-tracked as a way to save FFXIV, it's highly possible that such things would be cut in order to push out the game asap.

All things considered, it "should" be a safe change. Although there is an issue of things that are linked to the exp calculations which to be honest is one of the critical downsides of programming inheritance as opposed to true encapsulation.

tl;dr I'm too bored at work...

[Obsidian]

If they're using linking tables or using joins, or even if they (are horrible and) have multiple fields with the same data, it could easily become a mess though. Specifically because they're adjusting EXP inside dungeons and not EXP overall, I don't think it's safe to assume that there's a separate EXP multiplier being used inside instances (which would be separate from the Rested EXP multiplier and from the normal level multiplier) so they may actually be adjusting the EXP value for every enemy inside every dungeon, individually (I tend to lean towards this incredibly gross option because all of their fixes so far have been specific changes as opposed to system-wide solutions, which is why Twintania was breakable in much the same way Titan was). That's a truckload of fields to change and if any one of them is missed or changed incorrectly everything could explode.

Obviously those hotfixes go through less testing than actual updates (see Turn 5 patches). I think it's the safer (and smarter) bet to implement these changes once they've had the chance to properly verify their efficacy, instead of applying untested patches and then having to roll the entire game back by three days because we were gaining one lever per mob. The other reason I can think of is that they just want this to arrive at the same time as their Duty Finder enhancements to provide additional incentive, as opposed to letting people get all their levelling done beforehand and letting that part of the new system go to waste.

[Lucavi]

lol so mayad always. c'mon dude. how hard can it be to ignore me? you're the only one having trouble here.

its easier to manage everything when you dont have multiple patches to prepare, just one. obvious as shit, but the denial is strong as always.

Yet other companies have managed to do it and reap the rewards.

Try again. SE choosing to be old fashioned because old fashioned got us 1.0. How'd that work out?

[Elcura]

lol so mayad always. c'mon dude. how hard can it be to ignore me? you're the only one having trouble here.

its easier to manage everything when you dont have multiple patches to prepare, just one. obvious as shit, but the denial is strong as always.

Do you even know what mad means? If anything, I pity you for not getting the hint already.

[Akiyama]

Are we allowed to predict when this game dies in this thread? D: Probably not, but I'm just going to guess if 2.1 doesn't please, it won't make it to 2.2.

Seriously though.. in agreement with everyone here. This whole 'patch cycle' of theirs of the archaic days of FFXI/WoW etc just..doesn't work anymore. It's not how you keep people wanting to play. Even those who slowly ground their way to 50 (talking a month+) that I know are bored with the game. :< There's.. just not much to really do. It stinks. I wanted it to succeed after 1.0 flopped.

[Lucavi]

Seriously though.. in agreement with everyone here. (

Everyone but Hyan. Corrected. Or have we stopped considering him a person?

[Kachiko]

GW 2 tried to pump out patch after patch, in an attempt to create their 'living' world. Pretty quickly, it just became the same type of event, just re-skinned.

Both have their benefits and dis-advantages. JP is old-school because that's all they know how to do. Yoshi can only pull it so much, but even he has his old-school dies hard moments.

[arus2001]

Much as some people want to bitch and moan about FATEs, they're pretty much potential EZ Mode implementation of new content in that they can create scripted encounters without having to create new environments like dungeons demand. In the end, I don't believe there are enough sequential FATEs right now, even branching depending on success or failure.

As always, though, I fear the game will become too dungeon/raid heavy and quality open world content will fall on the backburner, leaving out people who can't commit to raid schedules for find the perfect FC for them to even try on occasion without getting bitched at for not being +1 everything.