Should game devs have to support new hardware?

[mcgeezer]

I posted this on Social media due to a bit of frustration last night and it generated a bit of discussion so, I'd be interested to know what the thoughts of the EAB members are.

Even for experienced programmers I'd say making a good quality action game on Amiga is pretty difficult to make, more so say than the Megadrive or SNES where the hardware hasn't changed at all and it's really just a dedicated game console. On the Amiga, things are somewhat different and I have no end of frustration when my game crashes or glitches on Amiga's with expanded hardware yet runs fine on stock Amigas. When I say expanded I do not mean more memory or hard-drives, I'm talking mainly about accelerator boards.

My opinion is that if the Amiga scene wants to attract new developers then the devs should not have to test/support all of the different boards that people have in real Amiga's these days.

Any new developer looking at making an Amiga game will probably just say "bollocks to that, I'm off to the Megadrive/C64/ZX scene where the hardware is all standard" and I don't have to follow masses of standard ways of doing things.

So what are your thoughts?

[BippyM]

Guys,

I have spoken with mcgeezer before he created this thread. I expect that this thread could become heated so please understand the following:

• Be respectful of each other. Any attacks on other members in any capacity will result in a ban without warning
• Keep your personal feelings out of it. Opinions are fine, feelings are not
• Don't assume you are correct. Everyone has the right to have an opinion, doesn't make it right
• I will close this thread without warning if it de-rails

[DamienD]

I agree with you Graeme. Developers should not have to test/support all of the different boards that people have in real Amiga's these days.

Just create games for standard OCS / ECS or AGA.

Also, I find that when people say "Why don't you create games for high end Amiga's / Vampires?", this is a whole different kettle of fish, the games would be more complex to code and would require way more time...

[Matt_H]

The traditional primary causes of incompatibilities were usually
1) probing specific memory/kickstart addresses that didn’t exist on all systems
2) probing specific microprocessor instructions that didn’t exist on all systems
3) insane copy protection schemes

All were a result of non-recommended programming practices. I like to think programmers nowadays adhere to standards that don’t result in any of the above. That being said, it’s up to individual coders to decide how rigorously they want to test something. Ensuring compatibility with the standard m68k CPU lineup and AGA/ECS/OCS would be my preference/suggestion as a user, but it’s not really up to me.

[roondar]

My personal opinion is that developers should feel free to aim their games/demos at any specification they desire. I see no more problems in making an A500 only game than I see in making a Amiga OS 4.0 PPC / RTG only game. That said, I do think that keeping some aspects of at least the 'standard' issues with faster systems in mind is only fair - so I'd use proper Blitter waiting, dealing with more memory properly, not timing with the CPU, correct interrupt/VBR use etc.

But I must say, for me the only Amiga's that truly have my passion and interest are the stock low end machines. So I'll aim anything I make mostly at them. To each their own

[LoadWB]

Developers should be forced to support NTSC. Joking aside, developers should feel free to target whatever hardware they want. No one bitches about (the few) SuperCPU games not working on stock C64, nor games which require an REU or second SID, even if the hardware is niche or rare. You get to decide on what hardware you want your game to run.

I trust that our quality developers are sophisticated enough to know their target and the market. You guys understand the prevalence of original hardware in its various forms of OCS/ECS to AGA to RTG and whether accelerated or not*, the expanse of next-gen accelerated hardware or upgrades like the ACAs or the Vampire, operating systems from OS3, OS4, to MorphOS and AROS, and so on. No one is going to get rich developing games for any generation or level of Amiga, so why not have fun just making something?

tl;dr no developer should be required to support anything beyond what they want. Just do it for the love of the sport and someone will be there to play.

* Every accelerator I have ever owned has a way to be turned off, so if I have to downgrade my system to play a game I want to play I will do so.

[lesta_smsc]

My view is simple:

Game developers can choose to do what they like.

An Amiga isn't a fully loaded beast that is trying to be something it is not natively. What I refer to is the likes of Vampire boards making the Amiga have significant higher capabilities than would be expected. Just use a PC.

I would prefer, however, for games to be run on stock Amiga with maybe a 512KB RAM upgrade but nothing that requires a heavy financial investment. THAT is what an Amiga was for me when I grew up and for nostalgic purposes that's where I would like to see it.

In fact, I'd like it even more if a game was INCOMPATIBLE with my A600 and I had to use Relokick to use it lol!

To developers: thank you for keeping the scene alive... but, with the cost of hardware these days, I fear that the fruits of your hard work may only be appreciated by those able to afford the hobby if they were able to purchase all the bells and whistles expected to make their Amiga like a PC.

[Minuous]

No one bitches about (the few) SuperCPU games not working on stock C64, nor games which require an REU or second SID, even if the hardware is niche or rare.

A minimum requirement is generally in place because the capabilities of that hardware are required. A maximum requirement is quite different and is generally due to lazy coding.

Poorly written games were a very significant reason that users were/are reluctant to upgrade, hence the Amiga died. It was designed to be an expandable system but was often just treated by devs as though the stock A500 was the only Amiga that ever did or would exist, which became something of a self-fulfilling prophecy. Whereas an IBM could usually be upgraded painlessly, doing likewise on an Amiga would result in a very sizable chunk of programs no longer working. The official Commodore guidelines are clear and not particularly onerous, mostly it is just a matter of not making assumptions.

68060 came out in the mid-90s so it is hardly "new hardware" anymore ;-)

[chiark]

They should target whatever they like, the only polite request I’d make is that a working configuration is well documented .

Personally I find the stock ocs the most exciting platform to target, but I wouldn’t dream of saying people “have to” target that

[Antiriad_UK]

Originally Posted by roondar

My personal opinion is that developers should feel free to aim their games/demos at any specification they desire. I see no more problems in making an A500 only game than I see in making a Amiga OS 4.0 PPC / RTG only game. That said, I do think that keeping some aspects of at least the 'standard' issues with faster systems in mind is only fair - so I'd use proper Blitter waiting, dealing with more memory properly, not timing with the CPU, correct interrupt/VBR use etc.

But I must say, for me the only Amiga's that truly have my passion and interest are the stock low end machines. So I'll aim anything I make mostly at them. To each their own

This is how I feel also.

[manossg]

I am also of the opinion that the developers should aim whatever they like/feel comfortable with.

We have emulators today and, if something doesn't work on our Amigas, we can always run it in our emulator of choice.

[greedy] Having said these, I think most people today have more or less expanded Amigas, so giving them a chance to stretch their legs is more than welcome for me. A game that has extra features for beefier machines but can also work on stock hardware (like Black Dawn Rebirth) is a very nice combination of both worlds. [/greedy]

[ShenLong]

Pick the platform that works best for the coder then once he/she/it feels its made its mark OpenSource it to the other platform. Code the first version in 68k build a user base and once you feel its all good with the 68k user base get in touch with PPC coders and see if they'll port it with your help. Its a win win for both sides of the fence.

[Foebane]

The projects MUST run on stock hardware of A500 or A1200 AT LEAST, and be compatible with faster hardware.

[BSzili]

I agree with OP, the developers should be allowed to target whatever machine they like. If they develop for stock Amigas and there are issues with accelerators, these can be fixed with WHDLoad Slaves. With expanded machines you have to deal with a huge number of different hardware combinations.

[mcgeezer]

Originally Posted by Foebane

The projects MUST run on stock hardware of A500 or A1200 AT LEAST, and be compatible with faster hardware.

Why do they have to be compatible with faster hardware though? Please elaborate.

[dreadnought]

There is no "They should target whatever configuration they wish", so I didn't vote. Option #3 would be perhaps the closest, but that puts a strain on the hardware people, so...

In case of programming for stock Amigas, obvioulsy there is no obligation to make it work with anything else than that. This should depend only on the goodwill and time/hardware resources of the developer. It could also benefit from the help of other people.

[phx]

I would say the developer must define the minimum hardware requirements precisely. But then a user may expect that the product works on all Amigas which meet them. So, when I say my game works on A500/OCS/1MB, it should also work on A4000/AGA/CSPPC060 and "most" other superiour hardware. There can always be special cases with exotic boards, which you miss, but a good programmer should reach about 90% of Amiga hardware in the given requirements.

Personally I would exclude configurations from this rule which are in development themself, or no real Amiga hardware. Like an emulator or modern FPGA stuff, which is constant flow. I wouldn't care, as the error is not on my side as long as it works on real classic hardware.

Otherwise, the developer has to say "this product only works with that specific configuration". Which I can understand and which is better than saying nothing, but it looks a bit lazy to me. We are in the 21st century and most developers should know how to take over the hardware and remain compatible with accelerators, more RAM, CPU-caches, AGA, etc...

[MikeRichmond]

If it’s a game it should work on stock gaming machines: a500/600 and not break on an A1200 if it’s an ECS/OCS game. If it doesn’t work on anything other than that it’s not the developer’s problem. It’s reasonable to assume it would work with (or even require) RAM expansions that would have been common at the time, of course.

[mcgeezer]

Originally Posted by phx

We are in the 21st century and most developers should know how to take over the hardware and remain compatible with accelerators, more RAM, CPU-caches, AGA, etc...

This is easy for you to say though PHX because you're an experienced programmer, But would you expect a programmer new to the Amiga and who wants to make a game to go and understand all of that stuff up front?

Even after over 3 years there's parts of the system I still don't understand properly (CIA's spring to mind).

[fxgogo]

Yeah, the poll choices are kinda all over the place and multiple choices could be valid to a person.

Devs can do whatever they want IMO. Certain games could benefit from acceleration if it was there, but it should be optional.

I also would not have an issue with a dev targeting only an accelerated Amiga. Of course the target market becomes a lot smaller.

In the end, we should not care too much, and just enjoy what we do get. This is a small and passionate community. Let’s celebrate them all.