Ashes Cricket 2013 cancelled AFTER realease, developer blames Unity for failure

Hmmm, they are using NGui, as can be seen from the logs.

 

If they had just chosen Unity for this game, I guess the developers probably had no considerable previous experience with Unity and just jumped to the project with whatever coding/asset workflow habits that they brought along from their previous engines. Im not surprised it didn't end well. For a big project like this, you need guys with proper hands-on experience with the engine.

 

You said it is a fact that the team has no ability to make a game. Not verbatim but the meaning is the same.

For reference here is the original quote.

 

Well, at least the plugin is there in unity's usual plugin folder after install (in the steamapps directory of course )

 

The meaning is not the same at all. Programming!=development. Team!=person. If you put ten top notch programers together in a team, they may never be able to create or even finish a decent game. And programming is only one aspect of the process, design, production, leadership, art and QA are all part of the development team as well.

 

Does anyone have any idea why the game is so large?



The above screenshot is worth almost 1 gig of data.

 

Re-attached screenshot

 

I don't think it's the game engine fault. I think the 3d and 2d artists did a hatchet job on the game while the developers were trying to control memory usage.


There was this strange 'Starting wall of nulls' and huge NULL exceptions generated from Unity for more than 5,000 lines of error messages. (See previous page for screenshot).


Do you know of any AAA game which have so huge resources for so poor model quality?
What kind of models shown in the various screenshots could cause 10+ mb multiple assets?

 

It's not a AAA game, it never was. It's not that surprising the game isn't optimized, no other aspect of it shown to be of professional quality. If they didn't have time to fix visuals and game play, it is unlikely that they would get around to optimization.

 

Give me 1 year and I can make a better game... with unity 2.6... seriously, what's the matter with those animation transitions?

 

It shouldn't matter if it's the engine's fault or not really, if it was a memory leak issue they should've fixed it in the source code. CD Projekt red built there own (may I say amazing) game engine and then released the Witcher 2 on the verge of bankruptcy, so if they can manage that then a team should be able to fiddle with a bit of C++ to find where the mem leaks are.

 

Not forgetting my earlier comment that the development house didn't deliver; the graphics, the models, and the art work look really good, but AI aspects not withstanding, it's clear they need professional animation help to complete the animations and that would go a long way towards making the AI be reasonable looking. I think it looks better than the other competing game which has worse graphics and doesn't appear to be much better in AI. It looks as if the animations was bought COTS which for a game like this, as the first of it's depth in type, doesn't cut it.

I would scrap nGUI too. It seems nGUI was more a hinderance than a help. When the new Unity GUI comes out they should try to fix that game and release it. They'll probably have to rebrand it though.

And I don't know anything about cricket to brand the AI completely broken or more as likely, the animations being incomplete enough to give an impression AI being more broke that it is.

What is clear is 505 as the publisher and the development house both should have known that that game wasn't releasable. They were counting on 'fanboy-ism' to make money and overlook problems and failed.

 

You do realise that the new GUI is pretty much the same as NGUI don't you?

 

Unless of course they don't have a source code license.

 

One would hope not. Otherwise it's a repetitive waste.

Having the nGUI developer doesn't mean that he is integrating nGUI into Unity. As a good developer with access to the engine he could create something much better then nGUI. That doesn't mean some of the same concepts can't reassert themselves. That's not the same thing as being pretty much like nGUI unless you consider attaching a UI script to any gameObject in Unity to be pretty much like nGUI.

But then I don't have access to betas alphas although you might.

 

Very true and for some reason I had it in my head that Unity gave modification access with Unity Pro. Well that would be problematic..

 

I think this is a valid point.. considering the news release which points at the engine by the publisher (based on the developer report), we could easily connect the dots that points back at the leadership team as being the failure I think.

 

just a note: seems the finger pointing at the engine is spreading. perhaps UT needs to say something.

http://www.gamasutra.com/view/news/...013_cancelled_following_poor_user_reviews.php

 

TrickstarGames.com takes so long to load in my browser.

Their years of experience making Cricket titles amounts to 1 previous cricket game. It mentions 2 founders but no employees. It lists a volume of output and quality of a sort 2 men could not have possibly produced in that time frame and the background of the company is indicative of a company with connections and a bit of management talent than that of a company with a boat load of talented developers and artists.

So a game publisher given big dollars goes to a couple of former game company managers that have formed their own company and gives those two a lot less money and so the contracted game developers, ahem, managers (TrickStar Games) with a long contact list of freelancers, proceed to give freelancers much, much less money, probably based on the Indian subcontinent (they like cricket too) or China, and those freelancers don't have enough money to develop all the resources they need to create the game, because that costs a lot more money then they were given. As a former defense contractor working the pressure was always to product what was promised at the original price no matter how unrealistic that price was. The big surprise is neither TrickStar Games or 505 seem to have even witnessed the playing of the game before release. They probably only saw very short videos and snapshots.

 

Not once in this article it is mentioned that Unity was used or even that they were blaming the engine. Am I missing something, here?

 

Well yeah, the article does cite "unexpected challenges the game engine threw up".

My limited understanding of PR suggests that saying nothing is the correct approach, though. Arguing with it just further associates yourself with the failure.

 

Well this is gossip and some developers are being catty is all...you know think M$ vs Apple vs Linux instead with Unity vs UDK vs Crysis. Yawn.

For those in this thread that said TrickStar Games are crummy developers one need only look at their about page and you can see they say they've 'established a solid Unity technology based production pipeline' which is management speak for hired freelancers cheap. Now freelancers and indie developers with LLCs will use the same language in an attempt to lure the type of contracts these guys get but these guy get contracts based on their past management experience not because of their development skills, which either they never had or haven't used in quite along time.

http://www.trickstargames.com/ -> [About]


So it seems 'Unity Free' + 'established a solid Unity technology based production pipeline' of cheap freelancers + disinterested management isn't enough to make a good game.

 

I saw that earlier and found it pretty funny. You are probably right that the "pipeline" they are talking about is freelancers/contractors.

 

Oooh. I LOVE management speak. Do they go to school for that?

 

Well in one form or another it has been around along time - "Officialese" or "Bureaucratese". If I may quote Milton Smith from the early 1950s

That was Smith defining a word he came up with "bafflegab" - which obviously now seems to have spread to private enterprise and of course politicians. I think the official staff job/employment description is ... er ... "spin doctor".

cheers, gryff

 

I think you might be looking at the videos differently to other people. For me the hud, models and art work aren't really that good, neither are the animations, but they aren't bad either and they have a good range of them. However the playback control over the animations is very broken, most likely down to poor AI, system, a broken state machine or both, this is where the problem lies, not in making better animations.

The end result of which is you have instances where 2 or 3 fielders running up to the ball and then just standing staring at it and the all too frequent instant transitions between animations (sometimes it seems to skip clips entirely), for example practically every time the ball goes pass the batter and is caught by the wicket keeper.I challenge anyone to actually see the ball being caught in those instances or to explain how the wicket keeper is able to teleport to where the ball is and back again.

These are the issues that people find so funny and so poorly coded, well at least it is for me and I assumed thats what everyone else was seeing? Unfortunately this makes other events that are actually natural to cricket to look even more comical, such as when 2 fielders lunge/leap to catch or block the same ball. In cricket you quite often see this near the stumps as the reaction time is so short, one fielder cannot wait to see if another has caught the ball. Alas due to teleporting balls/fielders and instant transitions or even skipped animation clips these events now look as though they are bugs or errors from the game.



Just to be clear whilst I find the visual quality lacking in some areas, the quality of the assets is not the biggest problem I have with this game. Sure I would expect the player models to be of a higher quality and resolution, they don't look on par with other cricket games or even older ones, but if this was a low budget title then you have to give them some slack (the assets that is).

What I am critical of and find funny from the video clips that have been posted is the poor AI at almost every level, general lack of polish and strange decisions that have ended up with the game in this state.

 

Nope, but one could guess its just part of the rest of the issues with the game and there was likely to be little over-sight on providing a streamline, efficient project. Its relatively easy to do, just bung in a load of high quality audio, perhaps use Unity's appalling ogg-theora video codec, use 4096x4096 texture everywhere without compression etc.

I would love to get my hands on this project and the assets. Firstly it would be fascinating to go through and do a post mortem on the contents and see what was going on, secondly to see what one could come up with using the same assets. I think some of the modelling is all right (though everything looks a bit low resolution and lacking polish), the animations are probably decent and varied for the most part (just badly played back), textures might need some work, but generally the visual quality just looking so flat in the game is more likely due to poor lighting and shaders.

Would be wonderful if it got open-sourced or perhaps if 505 own the rights to the assets for them to release them and have a competition to developer a better cricket game Obviously it can't use the licensed stuff any more, I can't imagine the ECB and Cricket Australia want anything more to do with either publisher or developer after the bad press this got, being picked up in every major media outlet in the UK since you know, its a game about the Ashes.

I wonder what the reporting of the events are like in Australia?

 

The bigger question is: why do people eat this S*** up? Or why do CEOs / politicians think people eat this S*** up?

 

Well this from the Herald Sun - a Melbourne/Victoria newspaper:

As you probably know News Corp (which owns the Herald Sun) is owned by Rupert Murdoch. So they are not exactly ignoring it "downunder"

And of course Melbourne is the HQ for Trickstar Games.

cheers, gryff

 

@noisecrime,
I have the game here.
It consumes 8gigs and soon going to be uninstalled. Thank goodness I'm getting a refund.


Does anyone have a credits list of who made the game?


Can you peek and find out what is contained in the .Unity3d file or .Asset file or DLL files inside the game?

 

I put it down to what I call the "Emporer's New Clothes" syndrome. Those who question the "spin doctors" must be stupid and ignorant. So nobody questions

The original story was written in early 1800s by Hans Christian Andersen

cheers, gryff

 

Or perhaps they're punished for questioning ...

 

I believe there are be methods, but reliability may be based on Unity version and wether they were using a beta version or even a custom version. Besides it involves going to 'dark places on the web' which are probably best avoided and obviously discussion of such tools would not be allowed on the forum.

You can probably inspect much of the code in the Assembly-CSharp.dll's if it was programmed in c# using an application such as ILSpy, which is an open-source .NET assembly browser and decompiler. Though its debatable as to how much use or interestingly it might be.

 

Reminds me of the show, 'Yes, Prime Minister' or if you've ever seen the episode of 'Absolutely Fabulous' with the round table meeting of the 'Fashionistas' Eddie works with, Eddie's boss and some man string together the most astounding stream of utterly fluff vocabulary and yet what they said makes sense, after you stop laughing.

 

lol. Thanks! I should give them a watch.

 

@nosecrime,

Besides that, do you think an average developer, average artist, average modeller would make the same kind of mistakes the Ashes developer would make?


Here is a possible theory of what went wrong:
My feeling is that they probably outsourced it to the most cheapest and less qualified developers. The developers probably colluded with the QA team to keep the bugs under wrap, the modellers made a polygon mess which made the game unstable. With huge sound files, huge model files, the developers ran out of memory very quickly.


Then they started to blame the engine... There is only 512mb of ram and there is only so much you can fit inside 512mb...

 

What is even more stunning are the developers, modellers and artists who made the game.


They are also on the Unity forums as well. I wonder why they won't comment, than 'just blame the game engine?'

 

A little quote from Sir Humphrey:

Translation: Someone's lying

cheers, gryff

 

A quick note to eskimojoe - thanks for all your enlightning contributions. The printouts are fascinating.

cheers, gryff

 

Why would they? What possible benefit is there to them either jumping on or trying to stop the bandwagon? What good could possibly come of them getting involved in this conversation?

I bet that they're way more displeased about this whole situation than anyone else. After all, they* spent some part of two years working on something, probably under less than ideal conditions in some way or another, only for it to get thrown in the can while they get ridiculed despite quite possibly having little or no control over some or all of what went wrong and probably trying their hardest to make this a success. How would you feel?

* And we don't even know who "they" are. Certainly the developers, but there are almost certainly others who also worked hard on the project.

 

schadenfreude

 

To be fair, the Unity Game engine was less capable and more buggy 2 years ago then now, that was even before 3.5, so what version of Unity did the game used when it was released? Is it fair to say they could have possibly upgraded that much code for that many versions of Unity, not to mention all the versions of the plugins they used and still had enough concentration and resource left to upgrade, test, and repair the important parts of the game like the AI, the animations and the state machine smoothly integrating it all together or were they constantly upgrading components as much as writing code?

 

Frankly I think anyone or any team can easily make the same mistakes, however I'm not sure I would simply classify their team as 'average'. The overall quality of assets doesn't look bad, not great, but certainly not bad, especially if you assume that this wasn't a huge budget title, which is likely as cricket doesn't quite have the mass market appeal as some other sports.

Now as to 'making mistakes' in terms of ending up with a huge project, I can easily see that happening with using any game engine, unless you've been a long time developer, or have the time and interest in ensuring that filesizes and memory use isn't abused. Heck it may just have been that no-one took responsibility for that area. I know personally that it is an area that I take control on and enjoy doing it, though probably to the opposite extreme - over cautious about memory use, keeping files small and the project clean. However I tend to pretty much be in charge of my client projects, with a couple of freelance artists coming on board for asset creation, which is clearly different to managing a whole team, even a small one.

I mean who would really be responsible for it on a team? I'd guess the project manager if they have the experience, otherwise you could claim the programmers as they are dealing with all the source assets, but then Unity is designed to be used by artists who may not be as clued into memory issues, so do you have to watch everything they do, so they don't go over budget?

It used to be that a programmer would develop an entire game, over time the skills needed have branched out, however you can't expect people from other disciplines to understand or even be interested in designing to a specification unless they've been given one. Plus coming up with a specification is no easy task these days and needs to be done by someone with experience, budgeting polygon counts, texture sizes, memory, load times etc. Its a whole aspect to game development that almost needs its own specialist.

Most likely explaination I can think of is simply in-experience with Unity and lack of time to fully get to grips with it.

 

Thats true, but i'm not sure it makes much sense. After all if they had appraised Unity 2 years ago and drawn up a design document you'd assume they felt capable of implementing it in the engine as it stood then. Its highly unlikely they would have known of things like mecanim were coming, so they would have had to code their own system, perhaps using available plugins. However even when mecanim came along, they would have had the option to prototype moving to it and should still of had their initial code base, so ultimately keeping up with new versions of Unity shouldn't have been a factor.

I'm not sure about claiming Unity was more buggy though, its only this last year that i've moved several client frameworks from 3.3 up to 3.5.7 which is pretty much my go to version and never really had any problems. Maybe I would for a larger project such as this, maybe not. Though thankfully i'm slowly getting to the point where i'm going to make the leap to 4.2 or 4.3 if no further major issues crop up with it. I can't wait as there are so many new features that I can leverage for better results.

I do agree completely with your last bit though. It is quite likely this was a small team, maybe 2-3 developers a few artists, it certainly seemed small from one of the video blogs I watched the other day from near the start of the Ashes project. Factoring in trying to target 2 consoles and PC's, whilst rebuilding your entire codebase (since they didn't own source to the old cricket games), learning to work with a game engine, possibly upgrading during development etc. Its all very easy to simply run out of people and time.

 

Why are you so insistent that the developers made mistakes? Sure, they might have, but on the other hand they may simply have been given too much work and too little time to do it in.

 

I think one of the big problems with Unity is that so many things from the Asset Store are unstable. If you just look at Unity's core product (now and 2 years ago), it's well-designed, flexible, and mostly stable. And you say to yourself:

"Hey! this is great tech! And look at all these coooool things from they Asset Store. There are so many things I WON'T have to make myself!"

And it is great tech, but as your project gets larger and larger, you keep making additions from the Asset Store. And that's where the "fun" begins, because you can get burned from the store in different ways:

1. Show stopping bug and no source code to work around it.
2. The addon makes a change that breaks your code, or is not backwards compatible.
3. The project looked good, but not enough people used it, so it was discontinued.
4. You thought the tech was good, and it got a lot of good reviews, but over time you really start to see its faults. There are too many 1 and 5 star reviews on the store, so it's hard to make an informed decision.

The more ambitious your project is, the higher the chance things will go wrong with store purchases. This should, of course, get better over time as more things from the Asset Store stabilize.

 

2 years ago mechanim was known about but not available and one of the 1st mechanim tutorials explains how to loop animations without getting glitches. I tried to do that tutorial, using tutorial animation material to repair an animation glitch and it kicked my tail. It's certainly not easy, in fact it is the hardest of making a game for a coders if you don't have prebuild animations to purchase COTS or similar. The other things a coder can be reasonable expected to learn but to obtain good animations you need equipment or an astounding ability to understand human and object physical animation mechanics and script animations that way.

A short search (http://www.ibtimes.co.in/articles/526140/20131129/why-ashes-cricket-2013-cancelled.htm) shows it was built upon a beta version of Unity. From the age of the development effect I'd guess that to be Unity Beta 3.5 from the fall of 2011. I don't think that using a Unity Beta was the cause of the problems but we know at least they didn't spend time constantly upgrading different components. So in this case I think upgrading at least to Unity 4 with Mechanim would have been a huge help as animating was even more difficult in the legacy animation system, particularly for animation amateurs (at best) which practically every game developer would have to admit to being at best. They probably were too far in to the AI to dump their legacy animations but at least one concurrent upgrade effort to Unity 4.0 must have been tried you'd think.

 

No, if you're making software professionally you know you can't just expect 3rd party stuff to work, and you know that every additional module of anything you have will impose a maintenance and integration cost on your project. That isn't the engine's fault.

 

Well. technically, it IS a problem. If S*** doesn't work, it doesn't work. It's as simple as that.

Do I blame Unity for this? No, I can't really. They made a flexible tech platform, and it's just going to take time for the store tech to catch up and stabilize better.

As I was saying, I think a lot of the problem arises from the quality of reviews on the store. There are too many glowing five star reviews, or negative 1 star reviews. There just isn't enough in the middle.

Edit: moreover, most people who use Unity use it for small mobile games. This means most reviews from the store are from people who make small games, not large ones. This presents a challenge for someone making a purchase decision for a larger game.

 

not necessarily, in many circles.. silence is acceptance, but I can understand UT not wanting to take an action that could be potentially liable.. like the one the publisher did. The blaming is hearsay... and could be considered as slander. If I was UT though, I would definitely be having my legal team check it out.

I for one, would definitely not be doing what they've done. And I certainly would not be thinking that the developers themselves are blaming.. but I definitely would be looking un-kindly upon their management team for this. This is definitely their mistake and they're accountable not only to 505 now, but also to whom the actual engine maker is.. which I believe most evidence presented here confirms what that is.