Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[cameni]

Then maybe you don't need public demo at all, only techdemo for potential investors, which one you already have? I can wait for alpha version of the game.

I think a demo still has its place - it could be delivered sooner with less effort and lesser expectations on completeness, reaching farther to people that would not otherwise get the game, and maybe also appealing to more developers that will be able to contrive their ideas when experiencing the environment there.

At least I hope the ATI issues will be resolved before the alpha could come out.

[Stargun]

We won't release it before ATI issues are fixed. Nvidia users may not care, but we would receive enormous backlash from people who will not be able to understand why it doesn't work on ATI when all their games do.

Bad news...  

As I understand we're all waiting for AMD to fix their BETA driver, which will arrive to the Outerra team, and IF everything are perfect  then in best case after 2-3 months (!) this driver will became public. And after that point the Outerra demo will be released.

So waiting for AMD means the demo will not be released before May / June, right?  

It's not good. I think you SHOULD release the demo ASAP even if it's only working perfectly on nVidia cards: just name it as a closed BETA demo test version and provide download only for registered forum users who have nVidia cards. That's all.

In this case you'll get plenty of really useful feedback from the forum users to implement while waiting for AMD to fix the driver issue. It's a win-win solution.

Look at the ArmA team: they published ArmA II as a full game while having bugs in the engine, but continuously working on posting BETA patches. And people won't lose interest, there's quite a community at ArmA II forums.

[cameni]

So you are implying that we should urgently release a free technology demo because the fact that BI released their Arma II game prematurely is .. comparable?

I would understand that argument if we were talking about a game and not about a demo. For the game we already said that the model would be releasing it in alpha version etc. Plus we are interested to receive feedback for the game, but not so much for the demo - a technology demo is just a snapshot of game engine functionality and we are pretty aware of all its shortcomings and unfinished state.

You are right about the drivers, but it is the worst case. Currently we aren't very happy wasting time on looking for the workarounds, when the situation should be getting better with time during which we can continue working on engine features. But once the game should come out, we will have to focus on that.
Maybe you can see why we aren't rushing with the demo at all costs - we are rather working on the engine itself now, so that the final result is better. And even if the drivers were fixed tomorrow, we would have to prepare the demo into a demoable state.

[Stargun]

Sorry if I seemed too pushy, but it's not funny that AMD working on their bugfixes nearly a YEAR!
So to be honest currently there's no guarantee that they'll EVER make it right...    (sorry, bad joke).

I think they're too busy now with releasing new video card generations, and fine tuning the drivers for those needs (and of course mainly DirectX compatibility).

As I said the demo should be released ASAP - As Soon As POSSIBLE, not urgently! I think it's quite possible because the engine should run perfectly on nVidia cards based on the demo videos.

So no offense.    Take your time as you wish to polish this extraordinary engine to be even better.

The only thing that we (the users) can do, is to push AMD to work better on their drivers. I already made my push. Changed my AMD card to an nVidia.   I'm really tired of the driver issues.

By the way: there're specific nVidia PhysX demos, which run on nVidia hardware ONLY.
And no one complains from the AMD users group why those demos don't run on their card, because the answer is simple: currently the AMD card does not support all of the features needed.

It's the same with your engine: AMD currently does NOT support OpenGL 3.2 correctly. That's all.

[cameni]

I must say that their responsiveness got much better since ATI was acquired by AMD. We were previously considering to (painfully) switch to DirectX because of the stagnating state of OpenGL and ATI drivers, but seems that both these wheels got turning, although it still takes quite a lot of time with the drivers.

If we left the engine in the state when we first encountered the bugs, it would be all fixed by now. It's the new functionality that we are adding constantly, that somehow always manages to break their stuff.

It's also likely that a considerable portion of bugs roots in the stupid "undefined behavior" that is present everywhere in OpenGL standard. It's good from the performance point of view - not having to check for unsupported combinations - but it's hell to develop with. Especially when Nvidia is much more tolerant and everything works as expected, and there's no mechanism that would help to detect these kinds of errors.

But as I said, I'm being positive that it will all resolve soon because those guys are really interested in getting it fixed. Understandably, they have to focus on commercial support first. I don't think we would be generating big enough alarm to speed it up there if demo came out without support for AMD .. that would need something bigger.

[knackered]

no, i think if you just released a demo that states that it's unsupported on ATI hardware because of their driver bugs then ATI would get their shit together pretty quickly. If your videos are anything to go by, you'd certainly rapidly get huge exposure all over the I.T. blogs. It would be shameful only for ATI, not for you.
Incidentally, do you support Intel's crap?

[cameni]

Two things: we don't need worse relations with AMD, and we don't need a huge exposure at the moment, looking for example how Infinity fares. Moreover, general gaming public would moan about grass, trees and effects, so we would like to get these done before that. Meaning, it's not the demo that should get the attention, but the game.

[Abc94]

For now we are thinking that the demo could contain several islands around the world, also in order to keep the demo size down without having to deal with possible P2P download problems there. Runways and roads will be predefined, active modifications of terrain will be reserved for the game, that should be also more mod-friendly.

You're thinking about limiting the size of the demo world??   :/

That's one feature I really wouldn't want missing in the demo.  Heck, it's not just a feature, it's one of the main qualities of the engine!

What if you just released the demo with all the features available at the time and everything else that you implement after that would show up in your game?

[cameni]

We would like to release the demo as is and stop caring for it. Having to provide download service for 13GB dataset isn't exactly such. Besides the size of the world won't be limited, just the amount of land on the globe, to cut the data size down.

But don't worry, for qualified forum members it would be .. different

[Stargun]

Having to provide download service for 13GB dataset isn't exactly such.

Use torrent to solve this problem. I promise to seed it as long as I can on my 30/20 Mbit/s connection...  

(By the way 13 GB is smaller than i.e. a good quality HD movie. And I know some people who download and seed such amount of data on a regular basis - of course it's not me.   )

[Abc94]

Do you mean you will just be lowering the terrain resolution in order to cut down on the download size?

If that's the case, why don't you just provide both a 13GB dataset download and a larger one for people who would want and have the space for a higher quality dataset?  (I personally have around 1TB of space that is not being used)

Sorry if I misinterpreted your post wrong.

But don't worry, for qualified forum members it would be .. different

What are you hinting at?  

[knackered]

ah right, as Lou Reed says, I'm beginning to see the light. Things like your trees being collisionless screen aligned billboards (nicely directed videos by the way, you avoided showing this quite well), all the LOD popping etc. I wouldn't worry too much about that. Crysis was a mess of popping, and having recently played Just Cause 2 I can confirm that it's very obvious in even 2010 sandbox games. Personally I blend like a mad bugger to avoid single frame pop of any kind. I find it incredibly jarring - not a minor artefact at all.

[knackered]

one thing - I can't find any blog posts on how you tackle the shorelines with your water renderer. Could you give a brief outline on what you're doing?

[cameni]

Do you mean you will just be lowering the terrain resolution in order to cut down on the download size?

I said that demo could contain several islands around the world. I meant several real islands of the world, like .. Sicilia, Taiwan, New Zealand and such, residing in their proper locations. No reduced resolution.

It could be an interesting sight in itself, wouldn't it?

But don't worry, for qualified forum members it would be .. different

What are you hinting at?  

Me, hinting? I'd never  

[cameni]

one thing - I can't find any blog posts on how you tackle the shorelines with your water renderer. Could you give a brief outline on what you're doing?

There will be a separate blog about the water, when I get the damn thing right.
But currently it's driven only by how the terrain gets refined by fractal, there's no explicit shoreline correction from vector data yet. Though for certain game types it won't be necessary.