Hi all! This is not really a devblog or techblog per se but I figure I'd do a little writeup on our core plans as we head into the holidays.
As a side note, both me (Vahr) and Bao (Altair) will be having some absences. I will be out from November 23 through December 3rd and Bao will be out from November 15 through December 10. The rest of the team will make sure y'all are behavin and take care of regular maints and patches and all that.
ETA: Early November
We will be modernizing all of our nests and raids to have behaviors and conveniences players expect out of modern DN, as some of our content is sourced from old versions of nests. Alistair and myself have been driving the majority of the development behind this (so shoutouts to him). Modernization encompasses:
- Ability to parry or iframe most stomps
- Ability to parry or iframe most basic attacks and some key mechanics
- Ability to iframe buff wipes (aside from a few key mechanics)
- Removing some excessive buff wipes
- DPS meters everywhere
- Skippable cutscenes regardless if you've cleared before
- Removing certain cutscenes that tend to cause players to disconnect (DDN clear, BDN clear)
- Gating boss barmechs behind longer cooldowns (e.g. VN/VTN Phoenix) so that you don't spend the entire fight doing barmech over and over and over
A full list of modernization changes will be published in the patch notes when they come.
ETA: Early November
Adding some solo versions of our existing nests for you Kiritos out there (except Kirito didn't have the luxury of solo versions of anything, so git gud like Kirito). Alistair is primarily driving this effort.
ETA: Early November
tl;dr the 70/80 cap trials suck. 95 cap trials introduced new or remixed trials that have been much better received by players. We'll be switching ATN and VTN over to the new set of trials and adding a few of our own.
- REMOVING: All current trials
- ADDING: Trial of Silence, Trial of Ice, Trial of Chaos, Trial of Weakness, Trial of Curse, Trial of Flame, Trial of Delay, Trial of Pain, Blessing of Haste, Blessing of Power, and Blessing of Vitality (notice that there's no Trial of Stagnation!)
Details about each of these trials (and blessings) will be hashed out and posted in the patch notes when they come.
Red Dragon Nest Hardcore
ETA: Mid November
First off, apologies that this is taking sooooo long. RDNVC was released back in August so its been almost 3 months now. We've had several higher priority issues come up coupled with the technical difficulty of extricating the old raid out of RDNVC. We should have this out mid-month so that you can actually get Ardent Dawn equipment without getting carried by a tiny portion of the population.
Red Dragon Nest Vahrcore
ETA: Early December
Making some adjustments and tweaks to the raid after RDN HC is released. I'll be implementing some of the features mentioned in the Ice Dragon Nest section as a pre-test (e.g. working out how checkpoints will work, testing new scripts, etc).
I'll also be posting a dev blog about the design and technical work that went into making RDNVC sometime over the holidays, as many of you have expressed interest in finding out more behind making the first VC raid.
Ice Dragon Nest
ETA: Q1 2020
The next raid in our bucket is Ice Dragon Nest. We're pulling out the big guns for this raid, as we'll be taking our learnings from RDNVC and we'll have new tools that will let us have unprecedented levels of customization. Plus I have a total soft spot for snow and ice themes so I'm gonna go hard on this.
The tentative plan is to have IDN 4 man, IDN HC, and IDN VC.
IDN 4 man will be nest difficulty and mostly unchanged from stock IDN, aimed at more casual players, average solo players, and yolo pug parties.
IDN HC will have several custom changes from IDN VC but will be oriented towards somewhat organized average player teams (so some wiggle room for pugging) and be the primary source of materials for the next set of gear.
IDN VC will essentially be an entire overhaul of the raid and as before is aimed at challenging veteran players. Expect brand new environments, new bosses, and new mechanics that will put you and your team to the test.
Player Fatigue and Frustration
One of the things that we learned from RDNVC was that we need to be more sensitive to player fatigue and frustration. While a VC raid needs to be extremely difficult, we need to ensure we aren't pushing players away because of it.
One major source of frustration (besides bugs, sorry ;_;) is the all-or-nothing fuck-up-you-fail design that pervades RDN in general. Lost two players to early Jasmine? Guess you can't do barmech. Oopsied and failed a mechanic? Party wipe. We'll be looking to have far fewer wipes and replace them with penalties that make things harder (silence, no mana, anti-CDR, % HP, etc) and toss in some rewards for cleanly clearing mechanics (e.g. not letting in a single Jasmine bomb resulting in a party damage buff, bigger DPS window, etc).
One trend that's been happening in official DN is that raids are allowing players to revive after each gate. This allows for more punishing and more difficult mechanics without causing as much frustration and time-wasting because you lost a player at a boss and now are unable to continue the raid. This also means we can keep the difficulty up throughout the raid instead of ramping it down as the raid goes on, since we can expect each boss to start with a full party, and make bigger difficult ramps within a given boss fight (so the final HP bar becomes more challenging because we can afford to kill off a few players).
So to that end, IDN HC and IDN VC will be using the revive after each gate system or some variant. Additionally, we may explore adding special mechanics that provide an opportunity to revive dead teammates mid-fight.
We are also exploring the idea of checkpoints - rather than having to restart the entire raid because of a wipe, you'll be able to go back to certain checkpoints in the raid and try again up to a certain number of times, similarly to how FFXIV does their raids.
Time Limit and Raid Duration
IDN HC and IDN VC will have some sort of time limit - most likely a per-gate time limit (shown on screen) or a "soft" time limit (not shown on screen) that causes the fight to get much more difficult or result in an outright fail. We've found that having time pressure in a raid is an important factor towards players optimizing and improving how they approach the raid.
We'll aim to have the raid last about 45 minutes assuming no potty or food breaks in between bosses.
Finally, there will be a per-character entry limit. With all the previous systems in play, this should be less of an issue than it originally was with RDNVC. We also want to make sure that players are not unintentionally burning themselves out by throwing themselves at the raid over and over again and force them to be more frugal about how they spend their time in and out of the raid - having no entry limit means that progress is made primarily via brute force rather than through strategization, study, and practice.
In RDNVC, Team Infernal and Team Barcode both ran the nest over 90 times (excluding runs that didn't make it past first barmech of first boss) before clearing. Assuming official DN's FDN entry pass limits, that's 7.5 months of runs that each team did in the span of about 3 weeks. That's insane.
Anyway, this is a pretty big writeup and overview of the plans for IDN. Thank you all so much for your feedback (and unexpected beta testing) on RDNVC, it really has been super valuable to me and I hope to make IDN the best raid you've ever experienced in DN (which is a pretty tall order considering the nest itself many consider to be one of the worst).
[PvE] Class Reworks
Note that this is NOT the list of classes we're reworking, just some classes that I'd like to rework personally (and some thoughts and writeups). This will smell a little bit like developer favoritism, but this is me purely speaking as a player and writing up some concepts I've had float around for the past few years.
Honestly, Guardian's 95 gameplay is SHIT. We tried to make it a little better with some adjustments (spearheaded by Darky/Salad) but in the end its not enough. Guardian needs a full overhaul.
During this break, I took some time to take one of my old tank class concepts (for a game that never actually went anywhere) and meld it into the DN gameplay system.
Here's a draft (warning - big image of text) that I wrote up. It is obviously a very initial rough draft and doesn't cover all the skills, but this is the general idea of where I'd like to go with Guardian.
Of course, the challenge (besides the technical challenge) would be to balance this concept within the frame of the game and with regards to the other classes (this would easily make Guardian the only tank). The other tanks would need their own reworks to hash out their own identity and match Guardian's power - frankly, none of the other tanks really have an identity in their own right in the first place, but I've also never seen anyone take a good hard look at Destroyer or Defensio, distill the class, and make something truely unique.
On a technical level, everything in that doc is feasible. We would need to change some boss mechanics to allow for things like Shield Charge to work properly, e.g. the megabreath at Red Dragon Pether would need to become blockable (although no class would be able to actually block the breath because of how many times it hits, but Shield Charge's self-extension would allow that to work).
Saint is in one of those weird places where it doesn't really have an identity. Sure, we did make Heal Relic relevant again, but the class overall is this weird hodgepodge of relics and everything else (especially awakening skills). I'd like to realize a relic-centric, relic-wielding Saint whose gameplay revolves around placing, throwing, striking, and working with relics. Something dynamic, rather than afk drop relics and push big heal when someone screams in voice chat. Again, I do have a concern about how such a rework would upset the balance between healers, especially as we took away LB's AFK float (which arguably is LB's "identity" but tbh a pretty crappy one).
I haven't really sat down and sketched anything out yet for Saint, but at least its on my radar.
Return of New PvP Maps
ETA: Late November
When Bao launched the server, we had some custom PvP maps (like Ice Dragon's Den) which eventually were removed due to issues with player spawn locations and play areas. I've made some progress on understanding DN's map formats and have found what should control spawn locations, so we'll be bringing those maps back so you can PvP in new environments. This also lays down the groundwork for more varied PvP maps and new gameplay gadgets, such as adding obstacles, platforms, hazards, perks, and more. After we restore the original set of new PvP maps, we'll start exploring how we can spice up individual and group PvP.
Other PvP Stuff
ETA: Whenever I have time to get to it
I'll be honest, PvP is a difficult part of the game to work with. I don't really have that much experience in this space and there's a bit of an entry wall to it. If you guys have ideas please hit us up and we'll see how feasible it is. We wanna make cool stuff for PvP too!
There's a lot of tech that powers our server to keep things running smoothly or your games running. Here's a peek at what we're looking at next as well as some recent stuff that I didn't really get to talk about publically.
Live in PROD
Turns out the server executables we use have an expiration date (happens to be October 11th). Oops. I created the server equivalent of Athena (called Artemis) to launch the server and inject modules (DLLs).
I then created a timewarp module that basically lies to the server executables as to what date it is until some condition passes, so the server starts and thinks its 2018, but after the expiration date check passes, the module then stops lying and the server warps back to the present day (so things like events, database time, etc aren't in the past as well).
Live in PROD
This one is a scary thing - DN has a SQL injection vulnerability involving in-game mail. I won't divulge the details behind the attack due to its sensitivity, however this is the method that a partner private server Celestial DN was being attacked by. The attacker used the vulnerability to muck with the database, causing all sorts of bugs, lag, and even dumping their (plaintext) passwords.
MailGuard sits on the server and monitors the packets flowing between the game executables, essentially inspecting and scrubbing mail. Using several detection methods, it detects attempts to execute SQL statements and nukes the packet(s).
ETA: Late December
The current DN patcher does its job but the method that ED uses for patching is slow and nondeterministic, meaning that it ends up creating entirely unique versions of the game for most installations. It also doesn't like to clean up after itself and leaves old versions of files laying around in the paks. It can take a long time for an out of date client to patch, especially with the number of patches we put out. The UI is essentially Internet Explorer with a funky skin (Eugh). There's also no way to repair your client if it gets corrupted.
There will be a full techblog written on the new patcher later on with the details, but we'll most likely be moving to a skiplist-delta hashed patching strategy that will minimize the amount of data downloaded to patch. We'll have a repair function as well, so if you somehow manage to bork your game you can just slap that repair button and the same patching tech that makes patches fast and light will also fix up your game as fast as it possibly can.
Expanding Player Profiles (Website)
ETA: Late November
Want to know how many nests you ran in the current week? The past month? All-time? Can't find where you stored a certain item? Want to check your mail to see if your in-game sweetheart actually sent you those Steve Coupons they promised they would? Want to share your stats to your friends but you aren't home?
We're planning on adding several features to the website to allow you to check all sorts of aspects about your account and characters, including their content history, inventory, mail, marketplace, and more.
This is a pretty difficult thing for us to do on the technical side, since we basically have to lie to a LOT of parts to get the client to go to the right server and not get yeeted out. We'll be working on this on-and-off between other projects for now, but we'll need some minor breakthroughs to get this to work. As it stands, we've got pretty great coverage with our single NA East server - a majority of players are within the 100ms raw ping range (which equates to about 180ms after all the overhead the game and game servers introduce). PvP due to its nature is still going to be laggier than PvE, but hopefully regional local servers can make that better.
Other servers (like Infinity Nest) have multiregion support, but they also have access to source code that we do not (trust me, if I had the source code of our current server and client, we'd have 144 FPS and much better netcode), so we have to do things the long and hard way with custom made programs, client/server injections, and other kinda difficult stuff.
ETA: Always happening
DN's performance sucks ASS, we all know that. It's an ongoing battle that's very tough to fight, given our lack of access to source code and all that. I'm making efforts to identify places where we can optimize and improve the experience.