Maintenance for the week of December 15:
• PC/Mac: NA and EU megaservers for maintenance – December 15, 4:00AM EST (9:00 UTC) - 12:00PM EST (17:00 UTC)
• Xbox: NA and EU megaservers for maintenance – December 15, 4:00AM EST (9:00 UTC) - 12:00PM EST (17:00 UTC)
• PlayStation®: NA and EU megaservers for maintenance – December 15, 4:00AM EST (9:00 UTC) - 12:00PM EST (17:00 UTC)

Should Zenimax take a proactive approach not wait for chapters/dlcs to add performance improvements?

  • ZOS_FalcoYamaoka
    Greetings,

    Some posts have been removed due to bashing commentary and conspiracy theories. Please take a moment to review our community rules to understand what is expected from members of our community.
    Staff Post
  • Universe
    Universe
    ✭✭✭✭✭
    ✭✭
    Yes they need to do the proactive approach and not hold stuff back for dlc/chapter release updates. Releasing during non dlc/chapter Patches.
    Universe wrote: »
    ZOS should implement performance fixes much more frequently.
    Maybe not every week since that will be too much and too quickly(not enough testing and coding time), but once per month or three weeks will be good.

    One game I played the developers simply hard rebooted the server each morning at 7am. This helped keep the game performance somewhat in check. I know other MMOs do it weekly and some do it monthly.

    If a daily reset would improve performance I would be all for that if the shut down only lasted an hour or so.

    Daily reset will be an ineffective.
    The servers down time of an ESO maintenance is usually around 6-7 hours on average so the 1 hour is not possible.

    Moreover, it takes time to diagnose the issues, finding the solutions and confirming the effectiveness of the fixes internally so 3 weeks to 1 month is more reasonable.

    So doing a performance update once every 3 weeks to 1 month seems reasonable to me(that is if ZOS manages to write the code and test everything fast enough for such a release).
    Edited by Universe on March 19, 2020 9:42PM
    Some videos I recorded for fun: Main character:
    PC EU main: Universe - AD magicka Sorcerer, Former Emperor, Grand Overlord, The Merciless, Trial Bosses Solo Champion
    Top alts: Genius(stamina/sagicka Dragonknight) The Force(stamina Nightblade) and other chars.
    PC NA main: The Magic - AD magicka Sorcerer
    Started playing ESO in beta & early access
    User_ID: Daedric_Prince
  • danthemann5
    danthemann5
    ✭✭✭✭✭
    Other
    I'm not sure how many more "performance improvements" this game can take.
    ZeniMax has no obligation to correct any errors or defects in the Services.

    Greetings! We've closed this thread due to its non-constructive nature.

    "You know you don't have to be here right?" - ZOS_RichLambert
  • Auroan
    Auroan
    ✭✭✭
    Yes they need to do the proactive approach and not hold stuff back for dlc/chapter release updates. Releasing during non dlc/chapter Patches.
    I said "Yes", though I could qualify as "Other" as well. In my opinion, ZOS screwed themselves when they made the major announcement that they'll always release a DLC every Quarter. That instantly put them on a time hack to ensure they push content out when they said they would, and unfortunately, in doing so, they're constantly struggling to always catch up with the performance issues.

    I'm pretty sure some of the Class Reps and others have asked ZOS to consider revamping the base game since they're heavily out of date in comparison to DLC zones and dungeons, and I think there was mention about skipping a Quarterly update to strictly focus on maintenance and fixes. However, ZOS themselves said that they wouldn't do that because it's not "moving forward". I interpret that as saying, "we lose money for skipping out on a Quarterly DLC and gain nothing material in return by just having an update that strictly addresses a majority of bugs". Again, Idk how much truth there is to it, but I wouldn't doubt it actually happened. ZOS is profitable now. Back in the day, they came out with free updates, such as the initial release of Crypt of Hearts 2, City of Ash 2, etc., as well as Craglorn. All free. They knew they royally screwed up when ESO launched, and instead of scrapping the game, or doing what Star Wars: The Old Republic did and make it a 110% money grab, they just came out with free updates and tried to fix the game (which I respect). Once it became profitable and was fixed up though, there was no more need to make free updates like that anymore, as their game was finally in good standing with a good player base.

    I like the organization. I like the steady flow of content at the guaranteed times so players can always look forward to something new and exciting around that time. However, it's beginning to turn into the opposite now, and that kills me saying that. Now, instead of looking at all the new and exciting things that might be coming out, I'm instead thinking about how each update comes out with more and more bugs and game breaking issues to gameplay. If no one's watched Nefas' video rating this update, I'll provide a link. He's being as kind about it as he can, but what he's ultimately saying was, for an update that was suppose to fix a lot of issues, it turned out to be absolutely garbage and made things even worse, and as sad as it is to say, if I were to recommend this game to a friend, I'd tell them now is a horrible time to join and to wait until the game is in better shape.

    In my opinion, I think ZOS should limit themselves to 3 DLC's per year instead of 4. This will allow them to still come out with a reasonable amount of content per year that people can enjoy, but also allow them time to focus on fixing their game. Alternatively, they can keep 4 DLC's (or "Updates") per year, and make 1 of those 4 Updates a Quality-Of-Life Update, where bugs and gameplay are also addressed. Will that happen? Probably not. Zenimax Online Studios is just a subsidiary of Zenimax Media. I'm not sure who owns the most shares in the company (probably a Chinese company like most big gaming companies), but regardless, you know they're looking at their wallets first. The people working at ZOS probably genuinely wish they could address and do things different. However, like every other worker out there, they probably complain and groan about how things are run and how their hands are tied, and just have to ultimately do what they're told because their parent company/bosses told them to.

    Still though, I do have to give them lots of credit for continuing to try and fix the game. Comparing ESO to what it was on launch, to what it is now, is a complete night and day difference.

    Nefas' rating video: https://www.youtube.com/watch?v=dzQmwJubaFE
    "And the Scrolls have foretold, of black wings in the cold,
    That when brothers wage war come unfurled!
    Alduin, Bane of Kings, ancient shadow unbound,
    With a hunger to swallow the world!"
    60k Achievement Point Club
  • Commancho
    Commancho
    ✭✭✭✭✭
    Greetings,

    Some posts have been removed due to bashing commentary and conspiracy theories. Please take a moment to review our community rules to understand what is expected from members of our community.

    Uuuuu and what is expected from our developers?
  • Kalante
    Kalante
    ✭✭✭✭✭
    Yes they need to do the proactive approach and not hold stuff back for dlc/chapter release updates. Releasing during non dlc/chapter Patches.
    Sadly its not what we want, its what the investors want and that it is..

    CrOwN StOrE TiMEEEEEE YEEEEHAAAAWWWWW.
  • Aikar
    Aikar
    ✭✭
    this is just how things work in our field of work.

    It's not so easy to do as you said in most cases.

    Large changes refactoring logic are done in "branches" in source code. Then the branch becomes unstable due to large refactoring's, and then multiple features can be developed in parallel on this branch. Each feature then starts to depend on the changes in the branch.

    You then complete the features to MVP, and then start testing and hardening/polish stages.
    Once a branch is in this polished state, you restrict new features that could risk the stability of the branch, and then once QA is happy with the state of the branch, it is merged and prepared for release candidacy.

    But say you have 3 features built on top of 2 core refactoring's.

    Refactor 1 is 80%
    Refactor 2 is 100% but uses part of Refactor 1 that are done

    Feature 1 uses only ref 2 and is considered 100%
    Feature 2 uses mostly ref 2, but depends on that part of ref 1 that is good from 2. This Feature is considered 100% done
    Feature 3 uses mostly ref 1 and is considered 80%


    Features 1 and 2 are considered complete, past testing and no new work needed.
    However, you can't release them.... You can't merge "just the good parts" without extreme effort and that had to be the goal even before development started.
    It's typically not worth the cost to develop that way.

    You take the rest of the work to completion, then release it as a single unit.
    You also don't wan to over burden your players with constant maintenance cycles, and releasing stuff before it's ready (psst.... they just did that! So instead of you asking them to release faster, ask them to release slower when it's actually ready)

    Also, release cycles are complex time periods in software development. once you are nearing the end/past code freeze, what you're able to commit becomes stricter. Only targeted bug fixes.

    So during this time, what can be merged into that refactoring branch now becomes restricted, so that it can be merged into a release branch. There is so much politics involved in all this.

    Once that branch is merged, you can then start the next cycle of more large refactoring's and instability for the next round.

    The only way to even get close to what you are suggesting is to target your refactorings in incremental segments, and merge often.
    This then has the cost of increased risk to introducing bugs that then hit us players, as well as limiting the scope of the refactoring, making it harder to achieve some of the major goals you might want to do.

    It has an extreme cost. It's not worth it.

    The community needs to cherish the fact that ZoS is even doing this. They are doing a decent job, now let them continue it.


  • IndianaJames7
    IndianaJames7
    ✭✭✭✭✭
    When you only start to develop a plan after years of complaints and performance degradation, you can’t call it proactive any more no matter how they approach it.
  • Aikar
    Aikar
    ✭✭
    Universe wrote: »
    Daily reset will be an ineffective.
    The servers down time of an ESO maintenance is usually around 6-7 hours on average so the 1 hour is not possible.

    Maintenance for configuration changes and deploying new versions of code is vastly different from simply restarting.

    A simple restart is no different than the context of recovering from a crash.

    Ideally, ZoS would have the system with redundancy so if one processes crashes, it doesn't take out the entire system (just should reduce capacity and at worse, cause more lag)

    If restarting servers is light weight, it shouldn't impact the server and wouldn't even need a maintenance window.
    If it does need a maintenance window due to cost, the maintenance window should look like:

    shutdown front facing servers to players. - Seconds to minutes
    shutdown services when 'must complete' work is finished. - If this isn't measured in seconds to minutes... ZoS has an architecture or capacity issue.
    start services back up. - Again I hope this is seconds to minutes.
    open servers to players. - Seconds to minutes

    Script/Automate this in parallel, and a restart window should be 5 to 15 minutes at worse.
    But... If the architecture is reasonable, it should be possible to restart a lot of core systems without shutting down servers.

    But ultimately... Restarting of software should NEVER improve performance UNLESS.... you have a memory leak.
    That is the only reason a restart should improve performance. Fix the memory leak then and it won't be needed.
    Otherwise the only reason you would see any improvement is because you kicked players who just simply didn't reconnect.

  • Curious_Death
    Curious_Death
    ✭✭✭✭✭
    Yes they need to do the proactive approach and not hold stuff back for dlc/chapter release updates. Releasing during non dlc/chapter Patches.
    Maybe they should give this project to more competetive studio? Just saying... each month each patch game looks and runs like crap.
Sign In or Register to comment.