Seperate Stable and Testing Builds...

[duddbudda 33]

DayZ is the critic's darling and steam's bestseller

there are a quarter million people playing DayZ

the model employed by the dev expects a small, committed community of mature individuals willing to work with the dev

but DayZ is a phenomenon that tens of thousands want to play

and when they can't play it they really don't want to help

SUGGESTION:

Rocket releases stable builds for public consumption

These do not have to be weekly or monthly or at all regular

Stable updates would be released whenever a stable build is achieved in testing

in conjunction, Rocket selects a handful of the most committed and capable community members to assemble and oversee a testing detatchment of a few thousand

these testers are given access to development builds released according to the dev's progress

when playing the dev build these testers are required to write up bug reports with clarity and concision

the testing hive would be completely seperated from the public hive, allowing testers to maintain a character hopefully protected from the worst bugs (eg: hatchets)

this shouldn't be necessary, but we're not all mature enough or in possession of time enough to make the long walk to gold

the alternative is to get a studio on board (professionals who won't follow insane coding practices like 'I have to release game destroying melee to fix other bugs' dolan pls) and work like World of Mass Development

---

I have to agree right now' date=' most bad posts/threads are coming from not fully fledged idea or bugs that a spotted right on the bat. Dev builds dont have to be kept dev for long, but at least 2 days will make sure the biggest frustrations, like the melee bug causing server crashes from 1.7.2, are eliminated.

Rocket, you are a good man and very enthusiastic, and i like that, but right now the versions are getting to everybody to quickly, to everyone's frustration.

[/quote']

---

ah, so the friend I was chatting to about this started the conversation over in the 1.7.1.4 announcement thread

here's their thoughts so far

Hey Rocket' date='

I am thinking it might be a good time to split your code into channels, one called stable and one called dev, this lets you put out a stable release for people that want to just play, and the rest that want to report bugs and test can run on dev. The whole "it's alpha" thing doesn't work when you have such a large userbase that just wants to play. From the sounds of it you are struggling with code management too, having to release big fixes with new features because you can't separate the new code. Proper version control systems such as git make this stuff pretty easy to manage.

I know this might seem like an early step to take at this point, but making the leap to stable/dev channels will make everyones lives easier.

[/quote']

I like this idea, players that just want to play can do so while people who want to help can test. Complaints should minimize and testing should be faster and more concise.

Sry, but: no. This is an alpha. This might be okay during a beta, but seriously it's an alpha version. We are supposed to test it!

Because finding bugs with a server full of 50 people is somehow harder than foisting obviously untested changes onto 50,000? If rocket actually wanted to get the bugs fixed he could just get 10 servers running off the master server list with easily accessible weapon/vehicle/loot storage and spawns. Nothing transfers over so you can bugtest away. Hell, it's what Eve Online has been doing for half a decade.

[ChaoticRambo 4]

Well considering this is the ALPHA TEST of the mod, I don't think this is necessary. You are talking about something that would be done once the mod is officially released.

Plus, a lot of the bugs being found now can only be found when large amounts of people are playing (either due to the possibility of encountering the bug or the server load).

[duddbudda 33]

Well considering this is the ALPHA TEST of the mod' date=' I don't think this is necessary[/quote']

neither do I

see:

'this shouldn't be necessary, but we're not all mature enough or in possession of time enough to make the long walk to gold'

in a community of 250000 randoms 'we're not all' means a hundred thousand people

re: hurr durr rocket needs 250000 fanatical voices or he can't fix it: 5000 dedicated testers would be more than most AAA games

EVE develops by giving a few hundred players access to dev builds whilst the wider community eats stable pie, so don't tell me the model isn't feasible

[Arkaddy 4]

Indeed, a pre-alpha maybe it's too much.

But it's true that some people even refuse to enter testing until they have confirmation that the testing has done, and the game goes smooth. If we all refuse to try every kind of things for the fear of losing our belongings, then only things in we focused on are: if the zombies sees too far, if they are too much aggressive, if other players want to kill us and how to prevent it, and if our inventory works fine (even between versions). The only thing that's being tested is if it's secure for our character/stuff to stay online.

I think other option maybe: every 24 hours, or week, or every update perhaps (we need to find what fit) all the characters are reset, die, and respawn. At least during the first days of stabilization of the hotfixes.

Yeah, bet it's popular idea. :D

[kriegar69@gmail.com 8]

I hate it when people look at a criticism and just start shouting "ITS AN ALPHA TEST", all i have to say to you people is stop being fucking morons, if we didnt criticise then that defeats the point of an alpha, if they didnt want criticism then they wouldnt have released it to the public till it was finnished...

on a side note i completely agree, they should have two seperate downloads for stable and testing builds and that way people can choose what they want to play instead of grinding for the best guns and vehicles for a new patch to come out a couple of days later and mess shit up.

... and yes there will be some whining kid that will reply to this saying "what do you expect when you play an alpha", well fuck you very much because people dont play this like an alpha and some of the patches were relativley stable untill they started adding new content that made it more buggy instead of fixing the existing bugs. This is why they need a stable version in which they fix bugs on and a testing version in which they add the latest features on to see how they work, then when they are stable it can be added to the stable version and so on.

[duddbudda 33]

some whining kid that will reply to this saying "what do you expect when you play an alpha"

that would be me' date=' OP :p

well fuck you very much

♥ you're too kind ♥

[DrRageQuit 17]

This is something i've mentioned a couple of times now. Every time i've been shot down with the "ITS ALPHA HURRR DURRR" bullshit response and it does get a bit tiring.

A dedicated 'team' of testers is exactly whats needed to get the best of the mod for people on the whole. Theres the people who want to play, and the people who want to help progress the mod to its full potential. At the moment we're all testers, and its quite clearly making some people unhappy when X/Y/Z bug affects them. This *shouldnt* be a problem as in an ideal world we'd all be mature adult like people who can handle it when their digital character gets busted up a bit. But its not, and we're not, and they cant.

At the moment, these people are defeating the entire object of what is being attempted, by going back to old builds or simply not playing. So it really wont hurt to just give them their toys and let them play. Out of sight, out of mind.

[duddbudda 33]

I'm not so concerned with what build people use

what does disgust is that rocket has to lead every update post with a size 72 request not to post abuse or support

and worse, I worry that parts of the community's attitude might deflate the creator's enthusiasm

[Vigil Vindex 8]

"It's an pre-alpha/alpha/pre-beta/whatever" is no excuse for poor practice. I'm a web developer, I know good coding practices, I also know a lot of it just creates overhead that you don't need when working on small projects. However, there are some good practices that benefit your work through out your work flow, even on small projects, regardless of alpha, beta release stage.

Separation of stable and development code is a no brainer, if only for the sake of having a backup. If you are using git you can get away with not doing this stage as you can merge in whatever feature or fix branch you are working on. Obviously git is not in use here, but I hope they at least have subversion running. Basically getting to the point where you can't do something with your code because you have added too much is a real big sign you are doing something very wrong.

[ImaffoI 22]

I have to agree right now, most bad posts/threads are coming from not fully fledged idea or bugs that a spotted right on the bat. Dev builds dont have to be kept dev for long, but at least 2 days will make sure the biggest frustrations, like the melee bug causing server crashes from 1.7.2, are eliminated.

Rocket, you are a good man and very enthusiastic, and i like that, but right now the versions are getting to everybody to quickly, to everyone's frustration.

[duddbudda 33]

Imaffo, if you don't mind, could i add your post to the OP?

[DrRageQuit 17]

Had AS50' date=' m9 silenced, gps,nv goggle, bandages, blood, come back after patches today with a makrov with 2 clips

fuck u rocket

lost 1 week char

cause this gay ass shit patch

fucking shit ass fucking retard patch

fucking dumb jew patch and dont test shit

shit ass dev[/quote']

I take it all back. Lets keep the current system. Moments like this are just priceless.

[ImaffoI 22]

@op sure my good sire, you can add my post to yours.

[duddbudda 33]

ty imaffo

@covering fire

moments like this cost the inevitable DayZ standalone a sale and deprive me of my victims

[DrRageQuit 17]

ty imaffo

@covering fire

moments like this cost the inevitable DayZ standalone a sale and deprive me of my victims

I know. Read my previous post on Pg.1.. That guy was just the perfect example of WHY this needs to be done as a priority. Kids raging then telling all their friends not to get the game.

On the other hand, the tears do help lubricate the servers.

[duddbudda 33]

apologies I got yourself and chaotic rambo mixed up

[kloust! 0]

Solution to your problems - http://dayzmod.com/forum/showthread.php?tid=16173

[Vigil Vindex 8]

Solution to your problems - http://dayzmod.com/forum/showthread.php?tid=16173

Actually that does nothing to help the issue, it in fact makes it worse cause you are now forcing every server to update to the latest version, potentially exposing more players to more bugs than necessary.

The whole point is that the benefit from having your code in two repositories, stable and dev, is two fold, you protect players from game breaking bugs, and your coding practices improve also as you are more robust to handle things like merging in new code from dev to stable.

To be honest, if no version control is actually in place then we can look forward to more nightmare hotfix scenarios like this recent one.

[kriegar69@gmail.com 8]

"It's an pre-alpha/alpha/pre-beta/whatever" is no excuse for poor practice. I'm a web developer' date=' I know good coding practices, I also know a lot of it just creates overhead that you don't need when working on small projects. However, there are some good practices that benefit your work through out your work flow, even on small projects, regardless of alpha, beta release stage.

Separation of stable and development code is a no brainer, if only for the sake of having a backup. If you are using git you can get away with not doing this stage as you can merge in whatever feature or fix branch you are working on. Obviously git is not in use here, but I hope they at least have subversion running. Basically getting to the point where you can't do something with your code because you have added too much is a real big sign you are doing something very wrong.

[/quote']

Yeah im a web developer (php and ruby on rails) /linux server administrator and on top of that im a project manager so I agree with you on every point, I know just as well as any good developer how important good coding practices and how important it is to get your priorities straight, dont get me wrong rocket and his team are doing a great job and I apreciate the work they put in but they really need to get a development plan in place so they can get their priorities straight instead of just going in to this saying "Lets see how people react if we do this" then ending up breaking the damn game lol