Path: chuka.playstation.co.uk!news From: Andrew Partington Newsgroups: scee.yaroze.freetalk.english Subject: Re: What platform Date: Mon, 15 Mar 2004 00:46:36 +0000 Organization: PlayStation Net Yaroze (SCEE) Lines: 220 Message-ID: References: NNTP-Posting-Host: spr1-ward1-6-0-cust140.manc.broadband.ntl.com Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624 X-Accept-Language: en-us, en In-Reply-To: Just out of interest, how are people likely to react to a bit of homebrew if you were to, say, whip out your GBA/other supposedly closed platform in an interview and show them a game you've made? On one hand i've heard that some people take a dim view of people doing their own thing on closed platforms, and on the other hand i've heard people saying (about some platforms at least) that they don't care where they've learned to code on it. Anyway, here's my entry for J2ME (MIDP 2.0), i've been giving it a quick going-over recently to see whether it's worth getting into. I can't say that i'm a fan of mobile phones myself, but some people are, so it may be worth putting something out there. Folks: 90% Easy enough if you have a MIDP 2.0 phone... Group: 95% Plenty of info at the manufacturer's sites. Public and private forums, and all sorts of other support. Std: 70% Some current MIDP 2.0 implementations are meant to be a bit dodgy, but will hopefully improve in the future. Many different screen sizes and processor speeds. Appears much better than the MIDP 1.0 situation is currently though. Skill: 60% Probably not as much demand commercially for MIDP 2.0 yet as there are millions of MIDP 1.0 phones still out there, but will probably get better. Public: 50% and probably set to rise as people upgrade their phones - you can get MIDP 2.0 phones so-called-free (if you do the contract thing) Cost: 80-95% If you just go with the emulators from the manufacturers and the Sun Wireless Toolkit, free! If you had to get a real device (or range of devices) to test on, or get the manufacturers to test for you, that will cost £/$. QStart: 95% Dead easy, sprites/scrollable tiled backgrounds, and built-in collision detection at block or pixel level. More involved if you want to start playing with network stuff obviously, but real easy for 2D games. Plenty of help, examples and info available too. Seems good? Not quite, there are a few things that are crap, namely the input method. Plus I can't see myself splashing out on a real phone just to do some J2ME on it. Considering frame drawing, it's just *wrong" to have a device with no vSync interrupt... (frame rates can get a bit variable, but whether that's me or Java is yet to be decided. Probably Java... :) ) And of course some programs will run on the emulator but fail on a real device. And while you get an alpha channel for transparencies, you don't get any hardware zoom/rotation functionality (spoiled by Yaroze!). Sound is fairly primitive too. But there's potential I suppose, and the network thing could be interesting if you tamed the latency. So overall, 75%. Are you making a start-up developer sim on the quiet and you need some stats for your engine? :) Andy Jon Prestidge wrote: > Just as a point of discussion I thought I'd post-up this analysis I was > doing to try and make-up my mind what platform to do future projects on, and > possibly to help others trying to choose a platform to write their own games > on. > At the Habbo Hotel meet I got the impression a lot of other people were also > thinking about platforms available... and Shawn's recent post was in a > similar vain. > > I thought of it because there are so many different considerations it's hard > to come to a conclusion. > > I thought of quite a few criteria... but I've already thought of a couple > more like: > "is it fun to program on?" and... > "how many more years will there be a following for it / how long will > it remain popular" like Shawn was wondering in his post. > > I tried to quantify each criterion so that diffent platforms can be compared > quickly.... but of course they're bound to be a bit subjective. > > I just very quickly did figures for 3 platforms, but I want to do it for > loads of platforms like PDAs, the Xgamestation, MAC ... anything in the > running. Also I think I'll have to split the PC as a platform and list it > separately as "PC Direct X", "PC SDL", "PC openGL" etc. > > If anyone is interested I could do with help in rating all the platforms I > can, and where possible come to a consensous for each... any help in > making the criteria better or thinking of new ones would also be good.... > > then we could put it on a web-site somewhere to help others --- people can > look at the criteria that matter most to them and choose wisely. > > Or failing that it could just be something to have a good old argument > about!! :-) > > > > > Choosing a Hobby Development Platform. Jon Prestidge, 13 Mar 2004 > ====================================== > > > > > FOLKS GROUP STD JOBINT SKILL PUBLIC COST QSTART > ===== ===== ===== ===== ===== ===== ===== ===== > homebrew > GBA 95% ? 95% 100% 70% ? 75% ? > > > Yaroze 5% 80% 99% 30% 40% - 50% 50% > > > PC 80% 0% 0% 80% 90% 100% 10% 0% > > > > > > FOLKS) You should to beable to easily show your games and demos to > your friends and family. > > Rating high if a lot of people are likely to own the platform too, > or if the platform is very portable to make it easy for you to take > your machine with you to demo your games. > > Rating low if few people own the platform, and the hardware > is heavy or bulky making it difficult to take with you > ... or for that matter if it takes a long time to set-up > when taking it somewhere (e.g. setting up their PC to download > the softawre). > > > GROUP) to have a news group (or similar) associated with the > platform where you can exchange ideas and programming tips and > try-out each others games and demos. The news group should mainly > be geared for games programming hobbiests (who may also be > professionals, or would-be professionals) -- this is > so that the newsgroup is not swamped by messages from people who, for > example, want to patch the operating system or try and put Linux on it > and use it as a server. > > Rating high if it is a members' only newsgroup for game programmers > also with a public newsgroup or web-site to: attract public interest, > attract new members, and show-case members' work. > > > STD) to have a platform which is standard, so when you write a game > on your own machine, you know for sure that it will work and perform > the same on everyone elses' machine. > > For example: the PC would score 0% for this because it's unlikely > any two people would have the same set-up. > The Yaroze or GBA would score 100%. The Amiga would score highly > (but not 100%) because although you could extend the memory and > increasingly fast versions were released, if you stuck to Commadore's > guidelines and stuck to the lower memory size, your software would > run on just about any Amiga. > > > JOBINT) you should beable to show your games and demos at job interviews > to prospective employers. > > Rated in similar ways to the "FOLKS" criterion, i.e.: prospective > employers must be likely to have a compatible machine or it should > be easy to take your machine to them. > > > SKILL) if you can program games or demos on the platform in a language > or set of APIs that that would be seen by empolyers as a commercially > useful skill then that gives a high rating for this criterion. > > > PUBLIC) to have access to a public audience for your games. If 50 or more > people can access your games on a regular basis it would give you extra > drive and enthusiasm for finishing projects and developing more. > Of course hundreds or thousands of players whould be better but I've > arrived at the figure of 50 as an 'OK' amount to keep a decent > momentum and excitment to the development process. > > An audience of about 50 would get 50%. > An audience of a thousand or more would get 100% > > > COST) The cost including all the development software and cables > (but not including the cost of a PC or MAC if that is needed) > > Rating 100% if it is 100 pounds or under > Rating 0% if it is over 500 pounds. > > > QSTART) The libraries for graphics, sound, I/O etc and the operation > of the system generally should be easy to pick-up, > or there should be plenty of examples, frame-works and tools to > help the learning process. > > Rating high if the system is easy to learn by someone with no prior > games programming experience ( but who has reasonable programming > skills) > > > > > >