24 hour patch! Probably the Solution.

Discussion in 'The Veterans' Lounge' started by Jimbob/Silvarfox, Apr 21, 2018.

  1. Jimbob/Silvarfox Elder

    A monthly 24 Hour patch would probably be something to consider. Would most likely be enough time to patch, reboot, And fix any other issues that may arise during. People would be able to plan around it and the peanut gallery comments would be minimal. I know that last part almost made my teeth explode from trying to hold in the laughter! But either way trying to throw a creative idea out there instead of begging for bonus exp as most of peanut gallery does. offering a thought for solution to the past problems is all i trying to accomplish. Let the peanut gallery commence below!
  2. Joules_Bianchi A certain gnome

    That would totally work because things always break on a schedule.

    Mmkk,
    Quatr likes this.
  3. EnchFWO Augur

    We just had our near-24 hour patch. It went splendidly as your post predicts!
  4. Stephsanity Augur

    Do I have to pay for that day?
  5. Blazay New Member

    To improve upon your idea, I think it should be 25 hours, occur the day after an announced 25% experience make up weekend and only to those servers with the most # of folks that can only play that particular day that week.

    There...fixed it!
  6. Zonawa New Member

    boiled, salted, shelled or un-shelled ... some with carmel, some with chocolate ... just name your flavor, we got: garlic, cheese, anchovy ...
  7. Blazay New Member

    w/o my glasses on it looked like you wrote Camel...was the first interesting thing i've seen during the downtime....
  8. Toruch Fleshrot Augur

    Peanut gallery? That's being awfully nice,lol.
  9. IblisTheMage Augur

    Yes, or change product. Plenty of other MMOs too choose from.
  10. IblisTheMage Augur

    Problem is not only time, but also conditions that only occur when servers are being played upon by real players.

    If EQ was being run on AWS, I would suggest a patch cycle like this: test —> one server (rotating, to share the load) —> the rest, so that one live/tlp server had 1 week of being pilot for the rest. Customer impact would greatly decrease.
  11. Waring_McMarrin Augur

    How would the patcher handle multiple versions of the client and determine which one the player needs to run?
  12. IblisTheMage Augur

    From the top of my head: dl both, run version based on server selection?
  13. Waring_McMarrin Augur


    Considering you launch the client before you pick the server that would take a bit of work to change that around
  14. IblisTheMage Augur

    Aye.

    A lot of hard work, including community work to make it a success.

    If you go crazy, you could go for splitting up changes to different servers, and kill test. Updates are then smaller, being pushed out every 14 days (or even better every week) to a single server, with the team ready to fix issues as they arise. There should be no need to actively patch the client, it could be evolved to patch in real time, and flip over on zoning. A change can be flipped in, tested, and flipped out.

    Even targeting smaller segments of players on a given server could be possible. Let the affected players chat directly with coders, and make sure they get 100% xp bonus and some kind of social recognition for testing, possibly creating a parallel “raid currency” for some super augs, only available if the player is actively testing the content. Hell, let the coders test stuff on a daily basis, with players queueing up to test content, in an opt-in system.

    Lets say a dev has a change to a quest. He requests 10 groups to test this, opt-in players apply for the job as a group. As they play it, the dev can follow them real time, logs what they do, maybe filming it. After they have done the effort, their clients are flipped back, and the dev can decide what to do next...

    It is a pipe dream right now, only meant as food for thought, but there are inspiration to find in the Spotify develoment model. The smaller chuncks of change you can make, the more automation of pushing change, the less impact and the more customer satisfaction can you drive. It could be a part of a long term vision.

    Add in game playtest scripting, so game designers can build scripted toons that test changes...

    The problem is of course the investment needed to get there.
  15. Aurastrider Augur

    Maybe a better solution would be to roll out patches on test for three weeks before they make it to live. Test patch would be the 4th Wednesday of every month and the live patch would remain on the 3rd Wednesday of the following month for example. This would allow for much more testing of a patch before it hits live servers. They would effectively have to skip the first patch on live servers for this to work but I would be in favor of no new changes for an extra month if it meant the process going forward allowed for more real testing and less bugs.
  16. I_Love_My_Bandwidth Mercslayer



    With fewer resources, Test has become Alpha testing and Live servers have become Beta testing. This is the way of things now. It is known. We are all testers. Embrace it or move on.
    Quatr and IblisTheMage like this.
  17. Aurastrider Augur



    How is this an embrace it or move on type of thing? Rarely does that kind of mindset ever lead to anything productive in life unless there literally is no alternative to a process that could make it better. Even if test does not have nearly as many people as live there are still people doing all of the same things players on live servers are doing. If they had more time to actually play or test new features before they hit live it could prevent a lot of things from ever making it to live. Sure there will still be things that are missed but having more time in a test environment should prevent less bugs from ever making it into the live patch.

    As it stands now there is a week to test things, two of those days fall on the weekend or a developers day off to actually try to fix potential bugs leaving four days and a couple of hours when its all said and done for any changes to actually be made and considering the devs don't work 24 hours a day it pretty much leaves them with around 32 hours of work time assuming they work 8 hour shifts to fix anything that comes across their desk, retest it and then push it out into the patch and hope it actually works.
  18. I_Love_My_Bandwidth Mercslayer



    Aura, you're one of the few people I have respect for on Forum Quest, so I'm responding honestly. DBG is running a skeleton crew. Limited resources mean cutting back on 'expensive' procedures to keep/regain profitability and increase efficiency. Testing is one area of production expense. Testing doesn't generate product or profit, only expense. So, as a skeleton crew there is little choice but to perform minimal amounts of testing.

    My point about accept or move on is that I get a sense from complainers that if they just complain enough that DBG will dedicate more resources to testing. To make this happen, they have to pull resources away from other things (expansions, bug fixes, quests, code upgrades, etc). My guess is the leads of each resource team are allocating resources as best they can. I would prefer to let them do their jobs rather than Monday morning quarterbacking every decision/error made. I'm mostly happy with the state of the game. Code upgrades on legacy systems NEVER go to plan. That's just the way it is. Even with unlimited resources (an impossibility) these types of upgrades reveal unforeseen bugs and incompatibilities which need to be dealt with.

    By all means, do voice your opinions and complaints. But also understand that DBG is doing the best they can given the subscriber base and resources at hand.
  19. Aurastrider Augur



    I still don't understand how leaving patches on test server longer is going to result in more developer time spent? I am not saying have the actual development team test things for another month. Maybe as a non programmer my knowledge is flawed here but it would seem to me that regardless of if a bug is encountered on live or test it will still take the same amount of developer time to fix it in the end. All I am proposing is a longer period of time for new patches be on test before they actually roll out on live. I honestly am not all that dissatisfied with the current process but I do think it could be better. There are times that a bug is found right after a new patch and we have to wait a whole month before the fix goes live. It just makes more sense to me for the majority of bugs to hopefully be found before the live patch so people don't come on here complaining about arx crashing or not being able to target xyz or what ever the issue is. Most bugs that impact me in a zone or mob setting I just ignore and move on with my day until they get fixed. Maybe its just a lack of knowledge on my part regarding the whole process which is entirely possible as my computer background is minimal at best. I studied how to fix people not machines.
  20. Kaenneth [You require Gold access to view this title]


    You're paying for the work done on that day.

    Do you complain that you can't eat while at the dentist; can't see the color of your bedsheets while asleep, or drive on the roads while they are being built?