Official Discussion Thread for "Update 33 Preview- Account-Wide Achievements and More"

VaranisArano

This part of the Q&A seems really problematic for players who now have to care about which character gets credit for certain achievements.


Q: What happens with achievements that have some portions completed on one character and other portions completed on a different character? Will they combine?

A: Partially completed achievements will have their progress combined. If the combined progress unlocks the achievement, then the character that supplied the completing progress gets credit. For example, let’s say there is a Skyshard achievement with 18 Skyshards. If you first log into character A who has completed the first 15 Skyshards, then log into character B who has completed the last 3 Skyshards, character B will get credit for completing the achievement.


It seems very likely that my lesser-played alts will end up getting credit for certain achievements when they actually contributed only the final piece.

Example: Let's say my main character Varanis Arano has killed 95 Wardens and my EP Warden has killed 30 of her fellow Wardens in PVP. Depending on who I log onto first will determine which character gets credit.

Did the Devs consider giving credit based on which character contributed the most progress?

I'm sure this is all set in stone and won't be changed, but frankly, I think this change needed some more workshopping before it goes Live.

Cominfordatoothbrush

There has to be another way to improve performance. What about all of these server upgrades ZOS was talking about earlier? We can't at least see if that allows for character specific data to be kept? Where does all that subscription money go; you can't maybe use that to improve the game's database performance in a way that doesn't destroy the most important aspect of a role playing game?

Kesstryl

mystfit wrote: »

Can someone show what the map looks like for an alt character of a main who has fully finished a zone. I'm curious what WB's, locations, skyshards, etc look like in those cases. I have dsl, so downloading pts would take 3 days )))

Pics of this stuff is posted on PTS, look in both the Account Wide Achievemets bug thread and also feedback thread.

Kesstryl

ectoplasmicninja wrote: »

All right, so given that it's a storage and performance based change, it can't be opt-in. Fair. But how do alts work if we can't track our progress through quests and the game? It seems like we'll have to abandon any sense of chronologic achievement, because how do I know if I've done this delve yet if it turns white when I walk up because I have an Explorer character? Will Verandis never be on the balcony in Blackreach again to bequeath me with his name? That scene can only play out once ever?

Verandis will never be on the balcony again on alts. This was confirmed on PTS by someone who tested the Greymoor story arc.

RoseTheSnowElf

Kesstryl wrote: »

Verandis will never be on the balcony again on alts. This was confirmed on PTS by someone who tested the Greymoor story arc.

FrancisCrawford

Katheriah wrote: »

At least you finally admitted this is for database reasons, instead of just marketing it as god's gift to the community. This change is legit horrible.

Even though they're "admitting" it multiple times, it's not true.

More precisely, it's either an outright lie or -- more likely -- a confession of staggering technical ignorance or incompetence.

They mention "over 1 billion achivements" over the course of the game. Well, each of those achievements requires on average a few bytes of memory to track. So they might save perhaps 3 gigabytes of disk storage by making this massive change. That's not per player; that's 3 GB ACROSS THE ENTIRE PLAYER BASE. That's so trivial that even ZoS wouldn't care.

Clearly then, to the extent there's any truth to this claim, it's about in-memory data. But the amount of in-memory storage required will, to a first approximation, be identical; if there's any difference at all, the in-memory requirement will be slightly INCREASED.

OK -- maybe they're concerned with the effort to LOAD data into memory in advance of it being actually needed as you play your character -- i.e. at the time of a loading screen -- or similarly as you log off. I.e., if you switch alts, ZoS could leave the same achievement data in memory now rather than loading the character-specific stuff.

But even there, a few hundred data fields per character are a triviality to load in any halfway sensible database architecture.

There's just no way ZoS' technical claims make sense.

FrancisCrawford

By the way -- props to @ZOS_GinaBruno for the Q&A clearly explaining the extent of this catastrophe. It can be hard to get people to publicly confirm that level of disaster. Well done, Madam!

nightstrike

FrancisCrawford wrote: »

More precisely, it's either an outright lie or -- more likely -- a confession of staggering technical ignorance or incompetence.

I think it's both. They came up with "database performance" only at the 11th hour, and hoped that nobody with even a hint of CS background plays the game. The fact that they even thought that achievements could be related to DB performance shows their lack of experience. But even if there's remote (and stretched) truth to it (by deleting even one record, things are technically faster by an immeasurable, infinitesimal amount), the way you tackle problems like this is with DB design, not DB deletion! Anyone working in this field should know that!

FrancisCrawford

nightstrike wrote: »

FrancisCrawford wrote: »

More precisely, it's either an outright lie or -- more likely -- a confession of staggering technical ignorance or incompetence.

I think it's both. They came up with "database performance" only at the 11th hour, and hoped that nobody with even a hint of CS background plays the game. The fact that they even thought that achievements could be related to DB performance shows their lack of experience. But even if there's remote (and stretched) truth to it (by deleting even one record, things are technically faster by an immeasurable, infinitesimal amount), the way you tackle problems like this is with DB design, not DB deletion! Anyone working in this field should know that!

@DarcyMardin has a smart theory.

Assume ZoS is drastically rewriting the code. (This assumption makes sense for all sorts of reasons.) Maybe they just wanted to implement fewer features so as to save on their (re)development effort.

The specific choice to trash achievements still makes no sense unless they're database ignoramouses -- but game developers commonly are.

S0Z0H

nightstrike wrote: »

FrancisCrawford wrote: »

More precisely, it's either an outright lie or -- more likely -- a confession of staggering technical ignorance or incompetence.

I think it's both. They came up with "database performance" only at the 11th hour, and hoped that nobody with even a hint of CS background plays the game. The fact that they even thought that achievements could be related to DB performance shows their lack of experience. But even if there's remote (and stretched) truth to it (by deleting even one record, things are technically faster by an immeasurable, infinitesimal amount), the way you tackle problems like this is with DB design, not DB deletion! Anyone working in this field should know that!

My thoughts exactly , I knew when I heard this from ZOS , something didn't add up lol. These guys keep things very vague and non transparent for a reason. And from what I can see through my years of observing ZOS dev behavior, it's definitely not honest.
I really feel someone in charge is actually not competent in coding, and really doesn't know what they are doing. They are okay enough to work the job, but let's just say, we aren't working with world class talent here. No offense, but there needs to be a corporate eval of these people. This IP makes billions of dollars, yet it's ran like they hire sub contractors.
Maybe this is the problem ? Idk, but I appreciate a tech perspective on this. All this "helps game performance" talk being tossed around with zero evidence as to how on earth it would offer an noticable improvement.

DarcyMardin

I’m a huge ESO fan, here from the beginning, but I cancelled 2 of 3 ESO+ subscriptions today. The 3rd will in all likelihood go too, if the final AWA implementation that goes live is as flawed as it is on the PTS. I just can’t endure to see 8 years of my many characters’ very different journeys thru the game ruthlessly wiped away.

It’s heartbreaking to see how little ZOS cares about its loyal, long term customers and how clueless the folks making the decisions apparently are about how, and why, many of us have continued playing this game for so long.

Dovetail_Stormrider

Re-post from a guild member requesting that I ask about this since he seems unable to log into his ESO Forums account.

"if anyone is able to ask in that thread what effect it will have on the identical dual trophy lists on ps4 and ps5 that previously you popped by doing it all over on another character now that that will be impossible i would be very appreciative."

Araneae6537

Edit: Nevermind

zsitvaij

Araneae6537 wrote: »

I have a question, please: Will our character data remain separate for as long as we don’t log in with them? Could I, for instance, complete the Greymoor story arc on one character and then not log in with them after the update, and get the full story, scene with Verandis included, on another character?

if you really want to know, test on pts, but don't expect them to change anything regardless of what you find.

you won't get answers from those in charge, and those of us on pts don't feel like sticking around anymore.

Pyr0xyrecuprotite

FrancisCrawford wrote: »

More precisely, it's either an outright lie or -- more likely -- a confession of staggering technical ignorance or incompetence.

They mention "over 1 billion achivements" over the course of the game. Well, each of those achievements requires on average a few bytes of memory to track. So they might save perhaps 3 gigabytes of disk storage by making this massive change. That's not per player; that's 3 GB ACROSS THE ENTIRE PLAYER BASE. That's so trivial that even ZoS wouldn't care.

But even there, a few hundred data fields per character are a triviality to load in any halfway sensible database architecture.

I'm not convinced of a billion achievements, but this game is big enough that it's well over a thousand achievements by now, possibly heading for 10k? That said, the bigger issue is that each achievement requires tracking a LOT of data, in a system that had to be designed from the start to be able to add more achievements.

Consider skyshards alone - we're at around 40 zones, with a total of 501 skyshards. Since the skyshard achievements are not fixed (more can be added), this is not a simple binary number to track them. Instead, you need a database with (at minimum) fields like an index field, a key identifier for the account, a character identifier, a zone identifier/index, plus some field per skyshard (have you got it or not), plus at least one reference identifier/index for the database which tracks achievement names/types etc. So, since it's a 64-bit system, this is at minimum something like 501 records per character of 6 bytes (64 bits each), times an average of maybe 5 characters per account (guessing). So, 15kB times 20 million-ish accounts would be 300GB-ish? (much of this is split across different megaservers/platforms of course). Let's assume that the Skyshards achieve is unusually large and make the average total storage needed per achievement closer to 100GB instead. Multiply by 1000+ achievements, and we're into hundreds of TB already; heading into Petabyte range if there are more achievements (and there are always more achievements with each new DLC) or more characters on average per account. Being able to cut this amount of data by 75%-ish (would be 80% if there are an average of 5 characters per, but let's assume it's not entirely clean) seems appealing in terms of managing, indexing, searching, storing/retrieving and backing up that data etc.

I think you may be off with your estimate by a few orders of magnitude. I might be too, of course - I have no clue how ZoS actually implemented this feature.

nightstrike

Storage is cheap.

Databases for online games are more massive than you realize.

The problem is lack of skill (or lying).

Sarannah

Personally I am happy with the accountwide achievements, as long as they fix the auto-completed bug, and the gather quest bugs(museum/precursor/etc). However, I would still like to be able to receive achievements on other characters besides the first. Is it possible to have achievements pop up every time we meet the criteria for an achievement, regardless of if the account already has it? .. This way we can still see the achievements every time, and even multiple times per character. Which is incredibly fun.

About the new PvP motif requiring alliance rank 20, for me as a PvE'er this seems extremely high. My highest PvP-rank is only level 14, as I really dislike PvP.

Deconstruction merchant seems cool, does it incorporate the CP and skill passives to get the regular amount of deconstruction? ... As I do not want to miss out on any materials.

Thanks!

Kesstryl

Sarannah wrote: »

Personally I am happy with the accountwide achievements, as long as they fix the auto-completed bug, and the gather quest bugs(museum/precursor/etc). However, I would still like to be able to receive achievements on other characters besides the first. Is it possible to have achievements pop up every time we meet the criteria for an achievement, regardless of if the account already has it? .. This way we can still see the achievements every time, and even multiple times per character. Which is incredibly fun.

About the new PvP motif requiring alliance rank 20, for me as a PvE'er this seems extremely high. My highest PvP-rank is only level 14, as I really dislike PvP.

Deconstruction merchant seems cool, does it incorporate the CP and skill passives to get the regular amount of deconstruction? ... As I do not want to miss out on any materials.

Thanks!

No those achievement pops are forever gone for alts once done on one character, there is no achievement tracking per character anymore.

mystfit

I was remembering an incident in eq2 years ago when zone loading was getting longer and longer. The issue was related to the fact the older the characters got, the more recipes they got and this was making load times bog down. I don't recall what they did to fix it, if anything, but I can't imagine one crafting class of recipes would be larger than a character's achievement journey, so I'm willing to consider it as a performative change.
But this sure seems like an aggressive way to address the problem. I suppose in coding an older game, you work with the cards you are dealt but I can't shake the feelings this performative angle is just a hail mary to get us to hush. Why mention it so late? OTOH, we assume they aren't cruel jerks causing pain on purpose...so...sigh...I dunno what to think .
I feel capable of working with this change and continuing to enjoy my characters, but my heart aches for those who can't.

Kesstryl

mystfit wrote: »

I was remembering an incident in eq2 years ago when zone loading was getting longer and longer. The issue was related to the fact the older the characters got, the more recipes they got and this was making load times bog down. I don't recall what they did to fix it, if anything, but I can't imagine one crafting class of recipes would be larger than a character's achievement journey, so I'm willing to consider it as a performative change.
But this sure seems like an aggressive way to address the problem. I suppose in coding an older game, you work with the cards you are dealt but I can't shake the feelings this performative angle is just a hail mary to get us to hush. Why mention it so late? OTOH, we assume they aren't cruel jerks causing pain on purpose...so...sigh...I dunno what to think .
I feel capable of working with this change and continuing to enjoy my characters, but my heart aches for those who can't.

I might be capable of some of the loss if the multitude of immersion breaking bugs didn't come with it, bugs that are still yet unaddressed in both the Q&A and other sources. The game world stops treating my characters as individuals with NPC reactions, achievement quests are locked out on alts (besides the museum ones), and other things already mentioned ad nauseam, and these are all deal breakers and non-negotiable for me. Fix those, and I might find some semblance of begrudging acceptance to take a hit for game stability (if that's the real reason), but if those bugs remain and just pile on the multitude of other bugs that never get addressed for months or years, there's plenty of other games that treat my alts like individual characters in the game world and not simply avatars of me the player.


edited for typo

Araneae6537

mystfit wrote: »

I was remembering an incident in eq2 years ago when zone loading was getting longer and longer. The issue was related to the fact the older the characters got, the more recipes they got and this was making load times bog down. I don't recall what they did to fix it, if anything, but I can't imagine one crafting class of recipes would be larger than a character's achievement journey, so I'm willing to consider it as a performative change.
But this sure seems like an aggressive way to address the problem. I suppose in coding an older game, you work with the cards you are dealt but I can't shake the feelings this performative angle is just a hail mary to get us to hush. Why mention it so late? OTOH, we assume they aren't cruel jerks causing pain on purpose...so...sigh...I dunno what to think .
I feel capable of working with this change and continuing to enjoy my characters, but my heart aches for those who can't.

If it really is the case, then I wish ZOS had looked into removing things like that — there are so many low level recipes and some add nice flavor (pun intended) but nothing ever used by the player.

Or make that knowledge accountwide even as styles are. But I’m sure they won’t do that since some motifs are sold in the Crown Store.

I don’t know; it feels like there are many less used elements that could have been removed or condensed with far less pain. Or if AWA had been done properly and character progression in everything important (and not just profitable) had continued to be tracked by character.

And I can’t help but think this is a weakness of surveys — how many like me rated achievements of relatively low importance in playing the game? Because I was thinking about worrying about chasing them or completing them — can be a fun extra for me, sure, but very low on my priority list. But my characters are essential and their unique (previously) journeys in Tamriel. But I am repeating myself and others who have put it more succinctly or eloquently.

And when I look at what so many of us are losing and a card game is being added… It’s not right.

nightstrike

Araneae6537 wrote: »

And when I look at what so many of us are losing and a card game is being added… It’s not right.

A card game that nobody asked for, I might add

deleted221106-002999

It doesn't have anything to do with performance unless alienating sufficient players so they leave and free up server resources is intended.

We all know this won't make a difference to performance from prior experience of every time we've been told the same or similar in the past only to log in and discover things are actually worse.

I noticed absolutely ZERO difference in load screens times etc on PTS. None.

I fully expect this monumental performance improvement experience to be fully observable when this unwelcome mess goes live.

wenchmore420b14_ESO

FrancisCrawford wrote: »

Katheriah wrote: »

At least you finally admitted this is for database reasons, instead of just marketing it as god's gift to the community. This change is legit horrible.

Even though they're "admitting" it multiple times, it's not true.

More precisely, it's either an outright lie or -- more likely -- a confession of staggering technical ignorance or incompetence.

They mention "over 1 billion achivements" over the course of the game. Well, each of those achievements requires on average a few bytes of memory to track. So they might save perhaps 3 gigabytes of disk storage by making this massive change. That's not per player; that's 3 GB ACROSS THE ENTIRE PLAYER BASE. That's so trivial that even ZoS wouldn't care.

Clearly then, to the extent there's any truth to this claim, it's about in-memory data. But the amount of in-memory storage required will, to a first approximation, be identical; if there's any difference at all, the in-memory requirement will be slightly INCREASED.

OK -- maybe they're concerned with the effort to LOAD data into memory in advance of it being actually needed as you play your character -- i.e. at the time of a loading screen -- or similarly as you log off. I.e., if you switch alts, ZoS could leave the same achievement data in memory now rather than loading the character-specific stuff.

But even there, a few hundred data fields per character are a triviality to load in any halfway sensible database architecture.

There's just no way ZoS' technical claims make sense.

There were some responses in one of the other threads from people who actually work in Programing and Data Coding in RL who broke down the technical aspects of what ZoS was claiming on performance increase , but going back to find and quote them, I found they seem to have been removed.

I invite anyone who does coding and programing as a living to post your opinions about this. Thank You!
Just my 2 drakes.
Huzzah!

BigM

I want to thank you for this change to the game. You have finally done something that will make me walk away from the game.

Thank You.

Fennwitty

wenchmore420b14_ESO wrote: »

I invite anyone who does coding and programing as a living to post your opinions about this. Thank You!
Just my 2 drakes.
Huzzah!

Obviously I don't know exactly what's going on in the background at ZoS.

Only way I can make sense of this from a database perspective is if the current system is horribly denormalized and unordered in a single large table like:

Character A Achievement 1 Date X
Character A Achievement 2 Date X
Character A Achievement 3 Date X
Character B Achievement 2 Date X
Character C Achievement 3 Date X
Character B Achievement 1 Date X etc.

which is terribly redundant and would be slow. Instead you'd want:

Character A <reference to Achievement table for 1/2/3>
Character B <reference to Achievement table for 1/2>
Character C <reference to Achievement table for 3>

Then you don't have to search through 6 entries to figure out what Character C has earned. You just need to look through 3 entries (before you optimize even further).

But with a database, if you're pulling in more data to memory than you need -- stop doing that. Change the query or the triggering of that action.

If you're unable to seek through a giant denormalized table quickly -- stop doing that. Change the tables to be normalized and cut out redundancies. Or at least set up indexes or try to partition better.

Because this accountwide achievement is being handled based on when characters log in, it implies to me that they're selectively populating NEW tables with a more normalized structure based off the old table data which loads to memory at log-in. So for a short period, there will be both old and new tables in existence.

After the first time, the old tables stop being queried on log in for a given character and reference points to the new table(s).

But why not simply normalize the existing old tables in the background and change the query-on-character-load? Who can say.

Purely theorizing and I'm low on caffeine at the moment so apologies if my terms got mixed up typing this quickly.

VaranisArano

nightstrike wrote: »

Araneae6537 wrote: »

And when I look at what so many of us are losing and a card game is being added… It’s not right.

A card game that nobody asked for, I might add

That's simply not true.

Take a look back on these forums with the search function, and you'll find threads with people asking for card games, stuff like gwent, and other tavern and community type games.

It's probably true no one expected it to be the only new feature of the chapter, but people definitely asked for it.

Dojohoda

I'm not trying to be rude but that article is an optimistic "good news" spin on major changes that involves deleting character data. I would have preferred a eulogy of sorts for our alts.

Are you eager to get your hands on everything coming as part of the Update 33 base-game patch?

Not really, no.

Jim_Pipp

You are ripping the heart out of your game.

The "performative" improvements come with a huge experience cost, and I know most of this feedback is personal stories, but please listen to these feelings, because they drive the numbers you are trying to preserve.

I don't see myself as a roleplayer, but I have a main character... not my first character, I played since early access, levelled 3 characters to 50 in the first year, when that was a really hard thing to do. In 2015, after a year of daily play I chose my 4th character, my main, and set out to complete the game on one character.... you are changing what I thought were fundamentals of the game.

I've now got 18 maxxed out characters, I still do all content on my main, but each character has a place and a story... mostly they have a themed purpose (PVP, trading, race-specific quests, etc). Objectively I gain a lot from this update (all characters will gain some achievements), but subjectively, you've taken away the purpose of many of my characters, and my game.

I'm mostly a daily player (I'm on holiday abroad now, and still doing the event, and still speaking to you), but I really think this change will reduce my interest in the game. I play to complete projects on characters... I repeat content more often than I do new content, so the future content you justify this change with, cannot replace what you are taking away.

nightstrike

wenchmore420b14_ESO wrote: »

There were some responses in one of the other threads from people who actually work in Programing and Data Coding in RL who broke down the technical aspects of what ZoS was claiming on performance increase , but going back to find and quote them, I found they seem to have been removed.

It seems that censorship is in full effect. It's now ESOrwell.