A Unity ID allows you to buy and/or subscribe to Unity products and services, shop in the Asset Store and participate
in the Unity community.
Discussion in 'General Discussion' started by animationcat, Nov 28, 2013.
I don't think AI is that bad. They play cricket better than me.
No-one is commenting about physics or rendering performance with the game, those are the very least of its problems.
None of which would really help you make a sport game, I mean locomotion? You're on a flat field! Sensory systems would also be a poor fit. As someone else said the biggest flaw with the game is the poor logic being applied to the various sub-systems
The technical difficulties were quite simply that it had memory issues on xbox and some tech stuff that they never revealed. Honestly very little of this is a problem with the engine and certinaly not for the PC release. Here is a quote from their community manager from Aug 2013
Only the memory issue on the 360 could be said to be an issue with Unity, though based on everything else I wouldn't rule out programmer error either.
As far as I've been able to discover they didn't make any of the other games exclusively and certainly don't own the code or engines for those previous games (CodeMasters owns at least one of them). Apparently a number of developers left the company during development and there is speculation (see the reddit link someone posted earlier) that they may not have the best working practices as Trickstar games is a re-incarnation of another failed company.
From what information i've been able to glean, this game was in development for at least two years maybe longer. It was due a release in the summer, but they got a 6 month extension. For the last few months their public face has been saying how wonderful the game is talking about polishing, yet its clear the game was fundamentally broken at various levels.
Clearly no-one but those who worked for Trickstar know what happened, but I seriously doubt anything will ever come to light.
I have a great deal of sympathy for the developers who worked hard on this, but none for the companies invovled. At best Trickstar perhaps burnt up too much time trying to get console versions fixed and neglected to actually finish the game at worse they were out of the depth.
What annoys me the most though is that the companies (both 5050 publishers and Trickstar) are putting the blame on Unity, even though nothing they have stated would have had any direct bearing on the PC release. Worse is that its now getting the attention of news sites like the BBC that are spreading the message of blaming the game engine. How 505 had the gall to sell the game I don't know and its yet another in a growing list of games where Valve did nothing to protect their customers.
I would love to read a developer postmortem of the issues they hit with this games development, have they written one yet?
On a side note what is the best team sport game written in Unity, just for comparison?
None on the gallery, any other 3D team sport games made with Unity?
noisecrime, that "fielder logic update" obviously went well - the chaos around the behaviour of fielders seems to be one of the dysfunctional highlights that is providing the merriment for thousands of people. Mind you, if what we are seeing is an improvement , what was it like before.
Personally, I find the wicketkeeper animations strange and jerky - kind of a personal pet peeve.
Yes you would expect professional journalists to be very familiar with the age old idiom : "a bad workman blames his tools" and based upon knowing this their reports should mention that the reason for failure as stated by the developers is "x" but this has not been established by any other outside sources..
The "a bad workman blames his tools" phrase can sometimes have validity. For instance, if they decided to make the game in say, flash, they would be justified in blaming the tool.
It's largely a matter of picking the right tool, and I think that's a little easier said than done. With regards to Unity, I think it has pretty good central design, but it is an evolving tech. If you look at the state of the Asset store today vs two years ago, you'll see a huge difference.
I'm talking about poor physics and rendering, so somebody is talking about it. There is a metric hail storm of issues in that game beyond extremely laughable AI. I wasn't talking about this game in regards to what an engine can specifically do, nothing is a poor fit if you know what to do with it. All anim will have been rigged in an external program, so if they didn't spot errors at that point then god help them.. Also what version of Unity were these guys using? It could've been an older buggier release, we don't know.
You see where they move and suddenly stop, jerk and move? That's a screwup in some sort of LMG blend, where it's resetting animation.. was Mecanim around at this time? I'm unsure. Then theres the navmesh bounds, obviously the guy chasing the ball wouldn't go out of the edges was a simple screwup and that's obviously not an engine issue.
They used Euphoria engine before hand, how could they all of a sudden go from half decent game to this? Somethings not quite right, if it was a game dev company with no previous experience then yeah.. It's more than likely going to be them, but with this sort of experience why the sudden complete screwup?
Well if they had any sense they would of left the lower level code out of the equation, memory leaks is obviously going to be something to do with the source code.
Anyone, we can't slag off the developers if no one else has managed to make a better team sports game with Unity?
I must be blind. I didn't see a single mention of Unity
Where does it says it was made with Unity?
I think people pieced it together from other places, such as company forums, reddit etc.
it could be a matter of too many projects at once and just not giving a f**k? It seems like its quantity over quality with them which should've been a huge red flag for someone like 505
It appears they definitely rushed it out with little to no testing. I guarantee that it was most likely a battle between the developers and the publisher.
But to call them untalented and unskilled is a cop out for you guys. Every engine will have a flaw or you will eventually hit a major road block if certain conditions are right. It doesnt matter if the engine is Unity, or Unreal, or CryEngine, all will have either one major flaw, or many little flaws that can be major show stoppers for a project. Some may encounter the flaws, and some may never encounter them.
No one knows for sure what the hell happened behind the scenes, and until the developers specifically say what happened or what the problems were, we'll never truly know.
Edit: Yeah, with this many released titles, I really doubt unskilled would be the word I'd use to describe their developers
As bad as it may be, if the developers were "unskilled" and not talented as a lot of the fanboys here claim, they wouldnt have been able to pull that amount of work off
505 was the publisher not the developer.
They'd clearly have contacts and connections to talented and skilled developers, and it wouldnt make sense why this one project would be the outlier of a long list of games.
Publishers like 505 have skin in the game, i.e. they are likley financing the developer, and having worked in a similar situation where Codemasters was the publisher, 505 will have had at least one Producer overlooking it all, and will have been well aware what was going on and would bring in their own people if needed. But again everyone is making assumptions on the lack of ability of the programmers involved which is amateurish behaviour. There are a load of factors that play into making a commercial game in a company. On Reddit claims were made that because the developer was supposedly a bit dubious they could not get the best staff. Total BS. People are making comments about poor AI, some of whom don't even know cricket, sometimes watching a real game of cricket the players look like they have wonky AI as well. I never understand these black and white discussions where people take sides as if they have some emotional connection with a game engine. What if Unity were largely responsible for the failure? One simply takes note and puts the thought on a back burner, because even if the engine was responsible the fact it failed to deliver on a supposed AAA title doesn't take away what the engine has done and can do and will be able to do in the future. I personally think Unity is the best game engine for indie developers and nothing else comes anywhere close, and knowing maybe it failed to deliver in this case doesn't change that opinion because I am never going to make AAA titles, and for me Unity can give an amazing polish to an Indie game. We have no idea what really went on. Did the company pay for Unity source code, were they trying to do it on the cheap and didn't have access to the code. All these comments about the engine failing should really have just been read as a developer playing the blame game to avoid some fallout, and filed away. Instead on various forums it's turned into an attack on the lack of ability of people who were just doing their jobs in a difficult situation and it's happened here to defend a game engine when really Unity can stand up for itself, and doesn't need our help at all.
What is being missed in all this discussion is the fact that there is another cricket video game to be released soon - "Don Bradman's Cricket 14". This game has been four years in development but is not licenced by any cricket board. However, while you start with a noname team you can infact import current teams. (And the fiasco that is the fielding AI in the above videos can apparently be controlled in this game at manual, semi-automatic and automatic - interesting AI control)
This game is being developed by Big Ant Studios - they claim to be "one of the largest game development studios in Australia"
Don Bradman's Cricket14 by Red Ant Studios
For those who do not know Don Bradman is a famous Aussie cricketer - possibly the greatest cricket batsmen ever and statistically the greatest team sportsman ever.
The release date is supposedly late 2013.
So what part did some kind of business/political oneupmanship - the need to be first - play in this farce? And of course Aussie rivalry?
Don Bradman's Cricket 14 - Youtube Video - a preview
No one is "making assumptions" about the lack of ability on the part of the development team. It is a fact. The game was so bad the publisher pulled the game after one day, citing it was poor quality. Are you assuming that the publishers are Unity fanbois? The game is simply bad. You can just look any of the videos (not just the buggy ones people are posting) and see crappy animation bugs. There is no debate or question that the development team failed to make a passable game. Regardless of any internal issues or challenges they may have faced (management, unrealistic timelines, skill, staffing, whatever), they built and shipped a poor quality game.
The issue in this community is that they blamed the Unity for their problems. That is nothing more that trying to shift the blame. It really doesn't matter what engine they used, the visible problems with game are very clearly developer related. If it were slow performance or memory leaks or crashing or other issues of that nature, then they might be able to make an argument (though probably still a weak one), but the glaring camera and animation issues are clearly in the realm of the developer. Random characters on the field spinning around in place have nothing to do with the engine.
The same the character AI, that is all in the hands of the developer. Unity (and most engines) are quite capable of handling these, there are plenty of examples of games that have much more complexity.
I don't think anyone is "defending" the Unity game engine. They are mostly just calling BS on them trying to shift blame. The development team designed, animated and coded that game. More importantly they built and shipped it. Lack of skill, knowledge, experience, or they simply didn't care. Who knows why they failed, but they did, and it takes multiple people failing to ship a large game that low of quality. Sure, there are probably some competent developers on the team (or maybe they good devs all saw where it was going and left a while ago), but probably not in the key/lead roles.
This is completely a guess, but one way that I could see this playing out would be if the publisher/studio decided they could reduce development costs would be to use an existing engine like Unity. They could have made a choice like this with little involvement from the dev team. They could have had an existing team/studio that may very well be skilled developers but never have used Unity, assuming they could pick it up on the fly. (which doesn't work well for big game to pivot on tech quickly). But that is just a guess. It is hard to believe skilled developers who know Unity well and are at the level to be put on a large title would build something that visibly bad. Something went horribly wrong in matching the tech and dev team.
Or maybe they all just hate cricket, and this was all just a big conspiracy to cast cricket in a bad light. (I am still a little surprised that it isn't made by an Indian team/studio).
Actually if they decided to make a game like this Flash, they would be the tools for thinking that was a good idea.
(Though, in fairness, Flash is/was a fantastic tool for what it does/did)
Well, lets be honest... regardless of how good an adaptation of cricket the game was... It was going to be horrible no matter what.
That is terribly flawed logic. Releasing a bad game (it's incomplete more than it is bad) is no indication of ability or lack thereof. For example id's games and engines are buggy and quite bad in alpha, does that mean John Carmack is a useless programmer? Using your logic, I can say that it is a fact that John Carmack is a terrible programmer. The only difference between id and this game is that id doesn't release an alpha as a finished game.
And they would of been the ones forcing the incomplete game out before the second ashes series ended.
Two years later for a project like this using the Unity engine?
@zombiegorilla - didn't one of the posts state that Unity was not named by the developers but it was pieced together from reddit etc. So if that is true it's not, as you suggest, Unity being accused which is why the community is being defensive. In fact, it's a defensive posture taken on the assumption it is Unity, so in the absence of a direct statement from the developers that is very revealing about the belief in the engine. However, this statement could be wrong as I haven't done an exhaustive search to find out every statement made not on this forum.so maybe the finger has been pointed at Unity by a rep of the developer or publisher.
Unity is not named in 505's press release, however it is pretty common knowledge for those interested in the game as Trickstar have mentioned it several times on forums. It may have been mentioned elsewhere, but the official parts of the game are being scrubbed from the net, such as its facebook page which was removed.
Granted consumers will not be that interested (though even they do pick on bad games and associate it wrongly to the engine i.e. dismissive comments about Flash, Shockwave or Unity 'Android' games) but anyone invovled in game development upon reading 505's press release is naturally going to be interested and it only takes a quick google to discover the name of the engine since Trickstar publicly announced it.
We finally have an equivalent of Crysis for Unity. Pushing the engine to it's limits.
Technical problems aside it's just very bland. But perhaps that is what developers were going for. When you release a game it is then de facto finished/complete and should be judged as such. Kinda makes me want to take a crack at making a cricket game, though. Cricket is competely foreign to me but the rules seems straight enough and there aren't that many moving parts.
It is here on wikipeadia:
I find that quote weird.
Our game has 100 times more NPCs per terrain (around 2,000 per terrain vs. 20),
Our game has 120 different NPC types on 20 different terrains (more than 40,000 NPCs total)
Our previous version of our game worked on an iPhone 2, then iPhone3G, then iPhone3GS when Unity 4.0 came out.
We still test on crappy Gingerbread Android 480 x 430 devices and our game still works.
I'll be curious to know what kind of memory leak problems they have?? I do know about Garbage collection, but memory leaks? seriously? in C#?
Actually I wrote a quick GC benchmarking app, basically throwing lots of balls about (via instantiation) and destroying them when they bounce 4 times. It triggers and times the GC, but what surprised me was the amount of memory the test used up and kept using up, in effect it should have hit a maximum memory usage level then plateaued but it just plowed on until it crashed.
Web version here - https://dl.dropboxusercontent.com/u/19148487/UnityGCBenchmark/WebApp/WebApp.html
PC version here - https://dl.dropboxusercontent.com/u/19148487/UnityGCBenchmark/App.zip
And this is on a PC where your 32bit app can run up a 4Gb ram space, compared to the Xbox 360's 512mb.
On a side note modern phones probably have way more memory than the Xbox 360.
This one's kind of a tough call, as most of the glaring issues are with things related to AI, and that's something you don't necessarily need Unity for. But Unity could have hurt the AI in more indirect ways:
1. If there were other problems, with rendering, physics etc, that could have taken up developer time, and left less time for AI.
2. When making AI, it's generally nice to have tools to help you along, such as an FSM or behavior tree editor. Today, there are pretty good addons for those in Unity, but a year ago they were a lot worse. You could always roll your own, but that takes time, especially if you're new to the tech.
Unity is just a tool , this is what happens when a dev hires no internal QA and then expects the publisher to catch everything .
To all devs , you need experienced QA testers , even if it's a guy hired on for a week , you need to pay someone to look though your game and find issues with it . As a 5 year QA vet I assure you that we are actually needed .No you can't have your devs do all your QA( for the little guy working for free/hopes of future profit you have an excuse ). No you can't expect interns to know what their doing.
Sports games have very particular requirements , you'll spend serious time getting a stadium to look just right( due to legal reasons , if so and so is a bit off every one gets all pissed ) . This means other things get neglected .
Considering 505 is a budget dev , I think the dev house under bid . They assumed Unity was magic which would let them develop the game with far less staff .
These are some of the error messages on the output.txt when the game stops.
The output.txt contains messages from the Ashes 2013 that was run 3 days ago. It just keeps crashing all the time (8gig RAM computer). It cannot run now as the game authenticates to Steam and closes itself instead. Can anyone else run this game or have output.txt error message they can share?
I genuinely laughed at
Oh man, that log output...
@eskimojoe - the funniest thing I've read on the internet in months... cheers!
Those were debug messages from output.txt.
There is even more because many of these error messages repeat themselves.
After it starts up ...
Even more errors...
Even more errors. (Eye rolls)
More error messages...
Even more error messages...
More error messages...
More error messages...
More error messages...
'Starting Wall of Null' error message
continues 8,000 lines later...
Same face you make when you see the interns code.
I don't get why they would even release the game in that state.
I watched the IGN review on ytube and.. that was a "far" cry comparison I'ld say. it was total rubbish.
After seeing some of the demos of Unity from Unite 2013, I hardly believe this game's problem was an 'engine' issue at all.
I watched a little bit of the review and most of the problems shouldn't of even made it out of prototyping stage.
If I were a developer (well, I am) who made this atrocity I would have jumped out the window! That is atrocious to blame the engine for the game's shortcomings... A carpenter can't blame his hammer for bending the nail...
Looking at the game some more, I noticed that most aspects of the game are actually pretty good. It doesn't look super amazing or anything, but the graphics and UI are fine. The tweens are good, The camera work is good. The way the ball rolls and hops is good. The animations are good, when they play correctly.
Yup, everything is pretty good, except for the AI and gameplay code. The problem with AI is, you can't just toss a bunch of people at it and expect it to improve quickly. Even an experienced team can screw up AI royally (see Rome, Total War 2 release).
So what probably happened was they didn't budget enough time for the AI, or maybe something unexpected happened with it, and it brought the entire project down. I'm curious what Unity tools they were using for AI, or if they made their own.
Would be interesting to see how other engines behave in exactly the same test situation.
There really is no plateauing after a few minutes (at the moment it's 40,000,000 bytes). Shouldn't a destroyed item be collected by the garbage collector?
Sorry if that's stupid question: I am not a programmer...
(and while writing this post the plugin crashed )
Then again I wonder if it really is a good idea to create a Project on all the major plattforms with an expensive license and sponsors behind it - all that on a relatively tight set timeframe - on a completely new tech to develop on. And I don't mean new tech in the meaning of unity being new tech - but rather how much experience you have with it. As much as many people love Flash - from quite a few teams i know just how many quirks that program has that jou simply have to either know about or have encountered yourself to develop a workaround. I imagine encountering and finding out about all these things during a large first prjoject could end just as devastating.
AFAIK they used PlayMaker, how this helps or hinders you in the long run on such a project i cannot judge. But i scratch my head why they'd would not build something for their own. IMHO That would be the best option in such a specialised scenario.
Hmmm, interesting. Playmaker might have hurt them in a couple ways:
1. PM is quite stable now, but it used to have some pretty nasty bugs. And it's closed source.
2. PM gives you some nice features for prototyping, such as global events and global variables, so you can avoid writing a lot of infrastructure. However, as your program gets larger, these features turn against you. Global vars obviously don't scale well. And lots of global events mean lots of crazy race-condition bugs, which might be what gave them problems. If you want to use PM on a larger project, write your own inter-object communication code. PM can, however, function as a nice state machine editor in a larger project.
That is the most probable explanation. Even though the developer's site claims they are experts in Unity.
Really? If true, that is just nuts.
Where exactly did I say anything about anyone's ability to program, or comment on any individual? You even quoted what I said and still didn't actually read it.
The logic is very simple, the development team is who develops a game. That is what they are there for. The publisher doesn't write code, the marketing department doesn't create the models. The developers created a bad game. Every video showcases the flaws in the animation. The fans think it is bad game, the publisher thinks it is bad game, the owners of the ip think it is bad game. Not just a bad game, but a game so bad it was newsworthy. Of course there some individuals that were more responsible than others for this mess, but ultimately it is a team. Doesn't matter if you are the UI designer, lead engineer, QA or any other role, if you aren't catching this stuff and bringing it up and actively working to improve the game, you take the hit just like everyone else. Everyone in this industry is judged by their body of work.
The only difference between between have a loved one die horribly in a car accident and having a loved survive a car accident, is just that someone you love is dead. … You can't simply dismiss a radical, key difference in two situations as one of semantics. ID made some of the most popular and influential games in the industry. They helped define gaming genres, set what would become standards and defined many tech approaches/advancements that have shaped our industry. Trickstar was founded a couple of years ago and has created a couple of unremarkable games (though some articles say that this game was actually subbed out), and managed to ship a game late that pissed off their fans, publisher and owner because of the terrible quality. And they did it by failing to execute using of the shelf tech that is designed for beginners as well as pros (a tool that includes a pretty specific feature for doing one of the big things they failed at). No, they have nothing in common, regardless of wether they call their game an alpha or not.
It is neither an incomplete or "alpha", beta. It shipped. Those words actually mean things. They all indicate a development stage. Shipped is shipped. Sure, they could have released it as a public beta or private beta or something like that, but they didn't. The said it was final, and took people's money for it. Every game, ever built, probably has a list of sitting around with all the things (big and small) the devs would want to improve on, add or change if we only had more time.
They lacked the ability to make it correctly, or worse lacked the understanding to recognize it was awful. Either way a failure. The game should have been scrapped much earlier or fired weak staff and/or switch the tech stack.
That is an assumption. Though, probably accurate. Even though this certainly wasn't a AAA game, it is still a professionally built game with a licensed property. Deadlines aren't an excusese, and management/publisher challenges aren't an excuse for the problems it has. Game design, ship dates and things of that nature can be blamed upstream, not flaws in the mechanics and actual functionality. The devs failed. Sure, the publisher and others in the chain did as well I am sure, but that doesn't get the devs off the hook. I have no doubt that is was messy situation, and they just weren't up the task.
The press release from the publisher says that the developer (Trickstar) faults the engine for problems. Trickstar hasn't said anything publically that I have seen, but that is the statement from the publisher. And it is known the game is made in Unity.