n01487477
Posts: 4779
Joined: 2/21/2006
Status: offline
|
I woke up and sat thinking this morning whether I would respond to your baiting for a full 30 secs and decided to go do other things like work and life and then as night followed I thought ... I have too much time on my hands today and I'd rather have a debate with a guy I don't know. On the Internet of course! Instead of going out an having a nice beer out.
I find it funny that we are in agreement that building an engine from scratch is not the way to go. We also agree in your second bait thread about steam and matrix pricing. Although I would probably take it further about steams tactics and have no desire to get into a price debate with other people here. So, let's let this sleeping dog ...
You are even kind enough to concede that even Unity is overkill. And yet you want to take this further because I dared to step on your toes and suggest that Unity was a unacceptable fit from a software development stand point?
Sir, you accuse me of "bullshit" and a "hypocrite" on the internet ... that is a laughable moment in my life. Sure, I'm full of it. I've never made a commercial game. Good call mate. My rhetorical knives are not as sharp as yours and I may have had to resort to "Have you?" if I thought I was boxed in but I'll withdraw asking you stupid questions like that.
I have evaluated (yes, I have a computer science degree - although I get your point over on Space Sector) and have bought 4 game engines over a period of time including Torque 3d & 2d ( 'tis was a mistake I tell ya. And yes, it is now free), Panda3D(free), c4 engine(licenced and the best I've bought & used), Unity3d - demo (and I'm downloading 4 now so that I can give it a whirl - you want me to make a game in a day? Ok - I'm being facetious), JME(Java monkey engine), Slick(2d engine java), lwjgl(ok not an engine but a framework for java graphics), Xna(now unsupported by M.S) and an assortment of others (Shiva, esenthal etc) that I have tried.
Why didn't I buy Unity you may ask? Cause I'm not made of money and the 30 day pro trial in the last version made me realise I was boxed into the Unity way As I stated before, no source code access is a turn off for me.
Ok, so all I've tried all these engines and no commercial game ... very true and something I must do something about if not just to clear my bullshit title.
I no longer do software development(non-game) as a career as I prefer teaching and my own programming life cycle had come to an end.(I no longer wanted to sit glued to a computer screen starting to resemble Neuman from Jurassic Park)
I also co-coded witptracker for WITP and witptracker-ae for witp-ae in Java on these very forums. Sure it doesn't come up to high professional standards but it was a hobby for the both of us while we were dabbling in Java. Probably a mistake in the end and we should have picked a better language. Oh, isn't that sort of my point?
So, lets start the rebuttal ... a tedious little job and something I'm loath to do cause you will reply and then I'll have to reply again and in the end we'll end up hugging and being best buds.
Furthermore, the whole debate is predicated on what we say actually matters to Elliot ... and as I'm sure it doesn't, this whole exercise in futility could have been stopped by my folly of not disagreeing with your obvious acumen in these and other matters.
quote:
ORIGINAL: Kayoz
quote:
ORIGINAL: n01487477
I've used the engine, I've used others...
Aah. And some of these engines had multi-threading security flaws?
I'm not saying that everything you said is BS - but nothing you said actually checked out. Not threading security - whatever the heck that is.
Firstly, I said quote:
IIRC it doesn't use 64bit as yet and its multi-threading is not complete (or secure) .
I also said "IIRC" as I hadn't dealt with Unity since the turn of the decade (until today) and things have changed since then such as the 64-bit support which seems to be now happening. So I will conceed that point to you.
I should have used the word "safe" rather than secure but that is probably Semantics to you.
But having a quick search on the unity forums from that time I'll cherry pick some snippets for you.
http://forum.unity3d.com/threads/168108-Discussion-Unity-and-Multi-Threading
What I glean from this discussion (mind you its old) is a number of things.
1. The Unity API is thread safe. The scripting engine is not.
2. Unity only uses one core for its API. Maybe that has changed by now ... But as stated by Mod's - It would take Unity a long time to do a re-write of this.
3. To make your threads safe you need to write a load of wrapped classes or come up with alternatives in managed code which can have a speed impact.
4. Find a M.S solution and make another wrapper cause you are at the whims of Unity as to what they will do or not.
Threads about no multi-core thread support (Okay most of them here are about physics but I said complete and in parenthesis secure)
http://feedback.unity3d.com/suggestions/multi-core--multi-thread-physic
http://forum.unity3d.com/threads/71357-Multi-core-physics
http://feedback.unity3d.com/suggestions/multithreaded-physx
http://feedback.unity3d.com/suggestions/make-unity-multithreading-suppor
http://answers.unity3d.com/questions/205435/Can-I-make-a-64bit-game-in-Unity.html
Furthermore, as we are talking about thread safety within the synchronisation of memory address space and as Unity's API is AFAIK limited to one core - then unless I am mistaken their multi-core support is not actually complete.
quote:
Not plugin support.
Where did I say you can't do plug-in support? I didn't. I said that you would have to use a load of plug-in support to get around the real failing of this engine. No source code, having to wait for the dev's to address an issue or provide the functionality. In fact I was recently watching a youtube interview with the dev of KSP was talking about how they had to find at least one large work arounds for one issue - I'm sure there were others too. http://www.youtube.com/watch?v=CupGRIL2h64&feature=c4-overview&list=UUxzC4EngIsMrPmbm6Nxvb-A
Having to write plug-in support for anything that is not addressed in the API is what a software developer has to do BUT with an overhead. But back to where we agree, it is better than writing and engine from scratch.
quote:
Not 32-bit memory addressing. Nothing. Not one of your objections was supported by fact.
And yes, I was writing about a lack of 64 bit support which seems to be working well for OSX and Linux and still encountering teething problems for win. But why quibble - you are the man and are right on this point. My lack of being a Unity fanboi has foiled me again.
Sure, you can make a game in 32-bit. Optimise your code and graphics to overcome memory address problems but simply put Unity has not addressed this barrier as yet... AFAIK.
If you want to get technical 32 bit aplications can address 2GB of memory on XP or 3GB with the /LargeAddressAware flag set.On 64 bit O.s's 32bit applications can go up to 4GB of memory. Whereas
64bit applications according to M.S "provide a huge leap in the amount of physical and virtual memory that can be addressed."
Anyway, I'm sure you'd like to learn more about 32v64 bit... (as you mention below)
http://msdn.microsoft.com/en-us/library/windows/desktop/ee418798%28v=vs.85%29.aspx
quote:
And for all your supposed experience with game engines, you seem oddly silent on what might be a better fit. I threw out one example, you say it isn't appropriate - yet while claiming knowledge of alternatives - completely fail to suggest one. I trust that you can understand how I see this as odd.
Oh - you wish me to suggest an alternative and then be shot down on that too...nice try. I really want to know why you advocate so strongly for Unity when one box doesn't fit all? Are you a user? Do you have shares in the company? Or are you as I think - just wanting to be confrontational about anything I write.
quote:
ORIGINAL: n01487477
I could go round to round with you, toe to toe ... but why bother ?
Indeed. I might exploit one of the multi-threading security flaws you mentioned and leave you hakzored. Heaven forbid!
But seriously, this isn't going "toe-to-toe". If you have something to say which I might learn from, I'd love to hear it. None of your complaints has actually been supported by fact, so I
think it's understandable that my current position is rather dubious of your claims to experience with game engines.
And yet ... (answered above)
quote:
ORIGINAL: Kayoz
quote:
ORIGINAL: n01487477
I kinda chuckled when I heard Zero (Stardrive) was going from XNA(sunburn) to Unity because he thinks it will solve all his coding problems.
Well, I had a little time, so I actually looked into this. Guess what I found? Some facts to support your claims? Guess again...
Here's his reasoning:
Reinvent the wheel:
"I wanted to use an IDE like Visual Studio and code everything from scratch. I didn’t understand these visual engines like UDK or Unity; I tried to understand them, but I lacked the foundational
knowledge to understand their advantages. The result is that I ended up spending a lot of time reinventing the wheel for StarDrive."
More efficient use of his time:
"But Unity is the way of the future. I can’t emphasize enough how much faster it is to work in Unity, how many things just work that would otherwise take days of development time in .NET. So
when I made the decision that I wanted to expand StarDrive, to bring in new content and game modes, it made so much more sense to me to just move to Unity now."
Cross-platform support:
"Future StarDrive games will be released not only for PC but for Mac and Linux as well, with the possibility of Android and iOS iterations just a short control-scheme change away."
Able to focus on developing the game, not minutae:
"My process is basically the same. There are artists scattered across the globe who are freelancing some top quality 2D and 3D art. I’m writing the code and integrating it all into a game. The only difference is that now I’m probably working three times faster with these new tools."
Full interview here: link
All that seems a far cry from your claim that "he thinks it will solve all his coding problems". He does not, in fact, seem to believe this.
Yes, I have read the space sector interview, I have the game and have read his posts on the stardrive forum. Actually, unless I'm mistaken and maybe you are not a native speaker the quotes you provided from him seem to suggest that he actually does infact believe that unity will solve his problems. That he no longer has to reinvent the wheel, that it will speed up development time, that at a click of a button he will be able to deploy various builds and unshackle himself from the minutae.
Anyway, this is a bit off point cause I was talking about his code optimisation and you only need to play stardrive into the mid-end game to see my point here. So is it the man or is it the engine?
I think from the little reading and his dev blog that his new SDU will be a cut down version of stardrive and maybe Unity is a perfect fit for the nature of that game. I stand by my initial thoughts that it isn't for DW though.
Why?
1. No source code. Without it, it doesn't allows you to do whatever you want without limitation. This point alone is enough to send most professionals running. Hence given the probably unfair moniker of being just a scripting engine.
2. Modifying the 3d engine for a 2d game. Yes, it is no more than just having a fixed focal length. But once again Unity is killing my argument
http://www.gamasutra.com/view/news/199109/Unity_introduces_new_2D_tools.php
3. API that only uses 1 core AFAIK. So, no true multi-threading.
4. Not totally thread safe - you cannot call objects from one to the other.i.e The API to managed code threads. AFAIK.
5. Unity is more a prototyping engine. Sure, you can knock together something quick in a cookie cutter way but then the pain begins. Well that is how I remember it back in 3.? I'll try the new 4.2 demo though - just to be sure.
quote:
You later stated, "Elliot has shown he knows more about making and using game engines than you or I." - which seems to be disingenuous since you clearly claim superior knowledge over Zero.
One assumption based on your own statements is that you believe Elliot to be far superior a game developer than Zero - and place yourself somewhere in-between. Or you're simply a hypocrite.
I am inclined to give you the benefit of the doubt and favour the second choice. But if you believe yourself to be a superior game developer to Zero, I would gladly see your portfolio of published games to refute this. What's that I hear? A deafening silence?
Oh well done... brilliant!
quote:
Well done. Business experts who can't tell the difference between profit and revenue. Marketing experts promising the land of milk and honey if Matrix takes DW to Steam. And now a guru developer
who knows more about game development than Zero or Elliot. Well done, Damian - you have company you can fit in marvellously with. Introduce yourself and take a load off. You have many compatriots in this thread (and others) who share your self-image.
I'm very much tempted to scream "BULLSH!T" at this point, but I'll await your rebuttal.
[edited] cause I'm being drawn into a level of argument I don't want. Sorry...
http://www.spacesector.com/blog/2012/11/limit-theory-an-open-boundless-space-sim-kickstarter/
I'm outta here and I bid you fair winds. I will no longer be drawn into a debate about this - you scored some blood.Let's leave it at that before both of us get a warning or are banned. And as I see you are still posting about this you must really should get up and move away from a computer sometime ...
Me - I'm out to a nice bar 
[edited for formatting]
< Message edited by n01487477 -- 10/25/2013 11:23:11 AM >
_____________________________
|
Printable Version
RE: Niche game or undermarketed? - 
Sure, would be nice if they'd cut the price point on their games I could afford to buy more of them but, ah well. Still lots of old classics to revisit one more time 
New Messages
No New Messages
Hot Topic w/ New Messages
Hot Topic w/o New Messages
Locked w/ New Messages
Locked w/o New Messages
Post New Thread