I’ve had several bad experiences with purchasing relatively expensive assets that turn out to be very buggy, unwieldy, poorly programmed (variable names like c, br, tm all over the place making modifications impossible to make), or just not work on our target platform. Reviews doesn’t say much as most users doesn’t take a look at the code, or use all features of an asset.
I’m not rich and need to spend my money on things that work. Now I’m in a chicken-and-egg scenario with several rather expensive assets I want to get, all lacking demos. I need to test the assets before I purchase them, but I’m expected to purchase assets so that I can test them. Often, I’ll message the authors, asking them to simply send me the asset so that I can evaluate if it’s what I need (in which case I promise to make a purchase), but the authors doesn’t seem willing to do this (I haven’t got any responses).
How can this be solved? Maybe demos should be mandatory on the asset store? Maybe Unity need some kind of SVN’ish test environment?
I never had this problem I spent time checking if the assets have a forum, read the review, then you know if it will work for you. when an asset have problems its quickly reported on their forum, if they do not have avoid the asset, it mean most of the time the dev is not fully available for support
My experience doesn’t confirm this. Reviewers rarely evaluate the code itself, even some of the most popular assets on the store are spaghetti monsters of horrible variable naming on the inside (which is problematic when changes are needed). Reviewers doesn’t test the asset on all platforms, especially consoles (I don’t expect everything to work on consoles, but I need to have the asset to find out how much effort is needed to make it work, which is why a demo would help so much). Reviewers usually don’t test every feature, make stress tests, or check how much garbage is being generated. I’m suspecting people write reviews very early, before they start to run into problems. All bad assets I’ve used have had very good reviews, and the problems has not been reported on the forums.
Any store that has minimal policing of content is going to suffer this problem. You’ve learned that most users don’t exhaustively test or look at the code of assets, so don’t trust their reviews. The only thing you can trust is a demo. Why not just ignore assets that don’t offer one? You are doing your due diligence by messaging authors and asking for a trial. If they don’t want your money, that is fine.
I’d rather force asset creators to make demos with our wallets than create some store policy that requires them. Too much policy and we risk shunning developers that Unity is designed to attract.
Yeah, maybe. Problem is, some of the assets would be incredibly useful to me if they turn out to be very nicely programmed. It’s hard to ignore an asset I really want when no similar asset offers a demo.
This is already making me spend way less money on assets in total just out of fear that they are bad, and maybe other users are in the same position. This must be bad for both sides, so I think it should be looked into. If not requiring demos, maybe there is some other ways to fix this?
I’ve found that the quality of assets vary wildly on the asset store. I’ve paid $5 for code that was absolutely beautiful, and I’ve paid hundreds for code / software I wouldn’t expect from a 1st year student. I’m still way up on cost / time ratio even with the frankly rare “dud” asset (so far really just 1 for me out of nearly 100 purchases). The vast majority of authors have gone way above and beyond for the pittance I’d paid for their work (although I’ve only contacted 2).
Expecting flawless code at the price points offered in the asset store is pretty absurd, and wanting it fully tested across unitys extensive platform bases equally so. I’m almost certain you’d know the cost of a professional code review, and full set of exhaustive testing (and we’re talking well over $10k+ for anything non trivial). It certainly is true there’s some “big name” assets I would have hoped for more quality from but even those are perfectly usable.
There isn’t really an elegant solution to this, it might be nice if Unity offered a “certified” program, or something but for it to be worth while the costs would be huge, and probably do the “market” more harm than good. Would I pay say $150 for an asset that would be $50 pre-certification? Actually yes, although I’d guess I’m in a minority - and the author would have made a LOT more profit at the smaller price. The market is also far too small (albeit growing massively) to just be able to soak up such costs without passing it on.
I’d personally like a more clear platform matrix on the assets (e.g. a list of platforms with ticks next to them if it has been runs/is tested on it). Mainly because at the moment you need to be fairly eagle eyed to spot assets with DLL’s, which instantly remove the web player platform etc. That way the smaller subset of console developers can spot when the asset has actually been tested on such platforms.
I guess in a perfect world the abstracted interfaces supplied by Unity would actually completely isolate platform specifics, but alas we won’t be there for a good few more major versions.
I suggest posting on here and requesting feedback on specific assets in regards to your target platform. Demos are too easy for people to rip off, and asking for evaluation versions will likely not get you far, as you’ve seen. If an asset does not work, you can get a refund if you request within 7 days.
I couldn’t afford to get 100 assets here, just getting one of the more expensive assets is a big deal to me. However, I didn’t suggest a solution that involves that every asset is flawless and tested on every platform, I’m fully aware how ridiculous that would be. What I’m mainly requesting is a way that I can test and find out myself before I make the purchase. However, maybe it’s just practically impossible
I have never been satisfied with any non model asset off the asset store. Even models you have to be careful because they are portrayed in a way that if they look x good into the screenshots they will look x/2 good in Unity (look for hazy screenshots with loads of image effects).
With non models, a big warning is when the author says “easy” in their presentation, thats actually a secret code word that means - incredibly obtuse, obscure and difficult to use.
Either never ever bother using an asset unless it has a demo or just spend what you can afford to lose without caring.
They must be plugins submitted a long while ago if they have no examples? My first plugin Master Audio was rejected the first time I submitted it for not having a good enough example scene. I still had one though…So what I am saying is: in most cases demos ARE required.
Also, I try to provide fully functional (but crippled in some way) free versions of our plugins. I wish more people would do this.
SmellyDogs, I can’t say nearly the same. I’ve bought over a dozen awesome non-model assets that are indispensible to us. Sure there were a few that had terrible support and we no longer use. Which ones did you buy?
I don’t really want to say specifically which assets for fear of starting arguements but lets just say I’ve brought for example:
Shader packs that crash and have continual glitches.
Road generation tools that simply don’t do what they promise.
GUI tools that are incredibly difficult to work with and don’t work nearly as well as promised.
Weather effects packs that stop working after updates.
I could go on here.
Often the attitude of the developer is a snarky response like “what do you expect for only $50?”.
In a sense that is true a lot of APIs cost 1000’s but even so I’ve brought some great programming tools for $50 or even freeware from Github that worked so well, the expectation is there that assets are going to be as good as that, but they are not.
To me its not worth the hassle anymore so I write my own, and I know I’d never sell my work as assets because theres a big big difference to something working on my machine, in my head making sense to me contrast to how it makes sense to someone else, getting it to work in their environment, workflow, keeping the updates etc. just to sell at asset at $50? I kinda see why asset devs are snarky.
Fair enough. Our two plugins we originally made because they didn’t exist (what we wanted to do) and later adapted them to be more generic. If the things you made are “holes” that currently exist in the Asset Store, I would urge you to pick the best one and release it as a plugin.
$50 is not worth it, but when you get hundreds of sales it may well be. Support can be an issue sometimes, but if you are having to offer a lot of support that generally means you have tons of users, which also means you’ve made some money off it.
Still (devs are snarky comment) to some degree you’ve been buying the wrong plugins. I can site some plugin developers who have killer support. Neatplug, Unikron, ArenMook and hopefully myself for starters.
I’d like to think most developers (myself included) want to help people enjoy and gain value out of what we produce. I’d like to see some better CRM tooling provided from the asset store to engage my customers and get feedback and deliver more fluid support.
Right now, I check every avenue I can think of a few times a day for any communication. Forums, PMs, emails, asset review posts (another huge issue… not enough sales channel pushing to do this from Unity). There is no customer portal for someone to log in and access support channels.
Forums are not CRM tools, yet they are being used as one. And to boot, its being coupled as a marketing channel as well. If I log into the Asset Store as a customer, I should have an account profile that I can goto… a unified control panel that then allows me to delve into each product I own and each product should be issued a support and issue repository (that can be exported to real tracking systems as well like JIRA, TFS, FogBugz etc).
Part of consumer confidence is allowing the potential customers visibility into this support. I wish potential customers could see how much effort goes into products to give expected levels of polish and usability… even for a single person’s request. If I could see that the developer was even willing to add in a single feature to make someone happy and I was already on the fence, I’d feel more comfortable purchasing. As opposed to seeing that there’s only been two releases and nothing’s been improved and there are lots of open issues against it.
Sorry Nifflas, I didn’t mean to be as harsh as that sentence seems in isolation. Demo’s, and light version are excellent ideas - and it is very much worth supporting those that provide them. Finger in the wind estimate would be the market already rewards such activities. I’ve read posts about how easy it is for people to steal unity assets from things like demo’s/webplayers etc. which I believe makes a few authors wary. If/when I publish an asset I’m certainly not going to let that aspect cause actual customers to suffer.
Dantas, I’m fairly sure pulling a sentence out of the middle of a paragraph and pulling it apart in a removed context seldom helps. My point on DLL’s is that if it’s not stated if a plug-in is web player compatible then it’s a rather solid warning sign (and I have no idea how to tell from an asset store listing if a DLL is fully managed or not). In most cases unless it’s something I really want I exclude it anyway since I like reading code / being able to fix stuff. Not that DLL’s implicitly mean full source isn’t there (before I’m called on that…).
Tstpierre, you know what. That’s a rock solid idea. Many times you see support requests for asset store stuff in the most bizarre places (my favourite is youtube comments). Trying to keep up with all those different points must be a complete mess. Even something simple like an RSS feed for asset comments would make life a long easier I’d guess.
Or my favorite place: seeing people who have never requested any support from our company at all, but go post a bad text review on the Asset Store instead with mostly incorrect points. I cannot understand that mentality, especially as a cursory read of the other comments will indicate we are extremely responsive and helpful on support issues. Making it worse, responding to Asset Store review comments does not notify the original reviewer, so they may never get the help they need if they don’t go back and read our response. It makes me sad.
I find it’s generally not feasible to offer demo versions, but I do have a 30-day guarantee if you buy assets from my site (I’d do it for the asset store as well, but I don’t have any control over that). So far I’ve had one person request a refund, which makes it a < .01% return rate. I make enough from them that I don’t respond to support requests with snarky comments. I know of people who’ve added their own functions, so I expect the code isn’t too hard to work with. I put a lot of effort into not having unnecessary allocations and so on, and I actually use all of my assets, so if I find anything difficult about them I say “Well, that’s annoying” and fix it, aside from taking user feedback into account.
Your sentence in the middle of the paragraph clearly stated:
You can write that with or without context, it’s wrong. I certainly understand now how you meant it. But there are many people who think that dlls don’t work in the web player at all. That’s why I wrote that. It was not my intention to offend you.
