Open Letter from a Private Server Developer

#1 - Nov. 10, 2017, 8:34 p.m.

Yes, I am one of those people. Kind of. Worth noting, I am not an IT professional; I do this as a hobby, and I won't be applying for a position on the new team. The emulator I am working on is not public, and our small team works on it entirely at our own expense. For me, it's just a fun and interesting way to learn about programming and game design. However, my experience does give me some understanding of the technical and non-technical challenges ahead.

From a technical standpoint, the 1.12 client is a relic. It is entirely unsuitable for modern gaming, in the same way that Windows 95 is unsuited for modern gaming. Trying to push 12 years of functionality, accessibility and security updates on software designed for use on antiquated systems would be a waste of time. If they aren't using the original client, they cant use the original server code either, since the packets they send back and forth to each other have to match up, and the number and structure of those packets has changed dramatically over the years as new features have been added to the game. Not to mention how many changes to the server code/database structures have occurred, and the fact that the old code is optimized for use on ancient, legacy hardware.

The result of these incompatibilities is that the Blizzard team will likely either create a new standalone client, or else build a "Retro UI" functionality into the current client, and appropriate server-side code to effectively emulate a Vanilla server. In essence, remaking the Vanilla world using the current game engine, and restoring the functionalities that have been abstracted out, such as spellpower or spirit. None of that is particularly difficult for a professional developer, but it is a very large and very tedious job. They aren't kidding when they say they have a mammoth task ahead of them.

The main challenge with a venture like this isn't on the technical side, it's pleasing the community and convincing them to accept any changes you intend to make. There are a number of smaller QoL issues which are largely aesthetic and can be dealt with fairly easily; people will like them or not, but they aren't game breakers. Updated graphics, AoE looting, guild banks, etc. None of these things particularly matter to me one way or the other. In my experience in dealing with the Vanilla private server community, and we're talking about the most hardcore of Vanilla enthusiasts, the two big questions that come up, always, are regarding itemization and tuning (there are a couple other important ones, but I'll summarize those later).

Itemization is a big issue because of changes made in 1.9 through 1.11, where new items were added to loot tables, green items were turned into blues, and a new PvP set was added. Most of this stuff was largely irrelevant at the time, since most raiding groups had superior gear by then. My understanding is that these were intended as catch-up items for players late to the game, so to speak. However, the itemization on them is so vastly superior to the prior offerings that these level 50-55 blues actually surpass Tier 1 and even Tier 2 epics (here's looking at you, Ban'thok Sash). Some of the new PVP blues from 1.11 are genuinely absurd. Having them available from the beginning of the realm trivializes a lot of content, both from a difficulty and a reward perspective. I'm not going to make any suggestions as to how this should be addressed, but it is certainly one of the things the team will need to look at.

Secondly with tuning, the reality is that players already know how to defeat the encounters, and the end-of-expansion talent trees and itemization changes combine to help trivialize much of the content. The initial wonderment of figuring out strategies is long gone. A big one that typically gets overlooked as well is the change to taunt with the release of Naxxramas, prior to which it didn't actually cause an aggro swap, just a short duration fixate and a threat match. To compensate for all this, the encounters can be buffed to reduce margin of error, or altered to change the mechanics, blanket nerf auras on players entering specific zones, or left entirely as-is, among other options. Again, I'm not going to try to tell a professional designer what the best answer is, but it's something which merits consideration.

A couple other points.. A progressive content release in terms of talent changes, etc. is a tremendous amount of work, but content gating / itemization changes are trivial. Talent changes mean UI updates and client data updates, meaning a constant cycle of patches being pushed to players (the reason private servers don't do this is because it would require custom patches, which have been used in the past to spread viruses etc. Players don't trust them anymore, so servers don't do them anymore - not an issue for official realms). Updating mob stats, unlocking portals, or changing item stats only require a quick database update and can be done entirely server-side during a maintenance outage. I think it's worth investigation as to how far the team is willing to go with this, and whether/how often progression should reset if the realms are to last a long time. I'd think a "Path of Exile"-style approach with fresh realms starting and existing characters being pushed to a static 1.12 realm every so often might work best, but that's up to the team, assuming they even want to go progressive in the first place.

There's a lot of argument about rolling back talent specs, or altering them to buff underpreforming class/spec combinations. Personally I think rolling back talents to the original trees is not very likely to occur. Many of the original trees were objectively bad from a design and playability perspective, and I don't see a major game studio intentionally putting objectively bad content back into a game they hope to make money from. I could be wrong; it's just something the development team will have to look at.

As far as buffing underpreforming specs, by 1.12 things were in a pretty good place as far as balance. Not every spec was A++ for raiding, but people seem to forget that the game wasn't only about raiding, either. They also forget that there is plenty of flex room for offspecs in a 40 man raid. Boomkins were pretty bad, sure, but stuff one in a group with four fire mages and opinions start to shift a little. You certainly won't hear those mages complaining anyways. No, you couldn't just always swap out one class/spec for another, but they were never intended to be hot-swappable. Each build excelled at it's own thing, and sometimes that meant raiding, sometimes pvp, sometimes levelling, etc. Some builds didn't excel at anything but simply offered greater versatility. I could give my opinion on how I would handle it, but my opinion isn't worth any more than anyone else's, just more things for the team to consider.

Population caps make sense, a big part of Vanilla was seeing the same people and building both friendships, and rivalries. However, I worry about how the initial "tourist" rush will affect this.

The Honor System is kind of a mess, but in my experience nostalgia wins on this one. People are diehards for it and react violently to any proposed change. I don't expect to see a significant change here (though I would applaud it).

I don't think anyone will complain about bug fixes and exploit loopholes getting closed. Nobody should be able to solo group content by utilizing bad terrain geometry to infinitely kite bosses, for example. I could be wrong, though. In my experience nothing makes an exploiter angrier than trying to take away their exploit.

Lastly, consideration needs to be given to re-implementing the spell batch system. It's a huge part of Vanilla, basically defining raid healing and a myriad of PvP interactions. The inability to cancel a cast for fear that a tank might not get enough healing, or the ability for two players to simultaneously crowd control each other are hallmarks of Vanilla, even if it does make the game feel less responsive.

So yeah, I do wish the team the best of luck in satisfying the player base with this ambitious project. I'll be intrigued to see how things develop as time goes on. And thanks to anyone who took the time to read this ridiculously long post.

Ok, well, that post got a lot more legs than I ever thought it would, and I've seen the comments from a couple sources. First of all, I'm grateful to everyone who stuck through reading what was a pretty long and rambling letter, and for taking the time to post your thoughts and opinions. I figure it would be appropriate to make a short response to clarify a few points:

First, I'm still not a professional software engineer. That's true, but I have been working with Vanilla/TBC for several years, and I have a fairly strong understanding of how they work under the hood. I purposely avoided going into technical details because the letter was meant to be understood by the layman. I regret rephrasing Client and Server Opcodes as "packets", in fact I regret mentioning them at all. What I was trying to get at was that the old client and server code are really outdated. This is from both from a functionality standpoint (bnet, authenticator, smooth movement, etc.), and from a code efficiency standpoint (almost 20 years of improvements to C++). It makes sense that they will either create a new client or backport the current one. This will be a big job and a tedious one, but it's not incredibly difficult for a team of professionals. Several other people have done a much better job than I of explaining this point in considerable detail, so I'm grateful to them for that.

Secondly, I tried to avoid including my own personal opinions into the topic of QoL and tuning changes. My intention was to point out and clarify a few areas of contention, several common proposals for addressing them, and (hopefully) politely and impartially frame a conversation about those topics. Including my own opinions wouldn't have accomplished that. I do have thoughts on each of the points mentioned, among others, but I don't assume that I speak for the majority and didn't want to come off seeming like I did. In re-reading it, I see that I let a few things slip, and I apologize for that. (Except for the bit about the Honor System. That really is a torturous design.) The proposed solutions I included come from my experience and interaction with the rabidly pro-Vanilla private server community, and represent several (but not all) of their preferred methods for recreating the experience they want. Consensus on any of the issues is hard to reach.

I don't really have anything else to add to the topic without expressing personal beliefs, so again I thank everyone who took the time to read and reply. Here's hoping for a fun nostalgia trip when the realms go live.