183 replies to this topic

[Mycrus]

Troutmonkey, on 26 February 2016 - 10:28 PM, said:

Moving to Unreal would solve lots of problems and make me very happy. Better performance for sure, as CryEngine sucks up far too much juice

Unreal sucks for triple screen and/or 3d

Be careful what you wish for

[Mystere]

[Oderint dum Metuant]

Never going to happen the best you can hope for is the upgrade to cryengine

[Lord Perversor]

MrMadguy, on 27 February 2016 - 02:08 AM, said:

This won't happen. PGI has curved hands - this can't be fixed by any engine. Game engine - is just a tool, like a hammer for nails hammering. It's still up to you to fill your game with content and code it properly. If you hold it for wrong side - then no matter, how decent it is - you won't succeed no matter what.

So your issues it's not about what can be achieved with a whole new engine or upgraded one as people is discussing, but your personal perception about PGI just screwing it and doing nothing because they are incompetent in your eyes?

[Sjorpha]

RedDevil, on 26 February 2016 - 11:21 PM, said:

How many of their art assets would they have to dump and redo from scratch? I can't imagine the amount of work that would have to be done if they had to redo maps, models, etc.

There shouldn't be any reason to redo any assets, except to make them prettier of course.

Depending on if the engine handles any element of the assets differently they might have to adjust some stuff, but that should basically be the same for all assets and on the coding side rather than the art side.

As an aside note, I would love to see some eye candy renderings of high poly mechs models with high res textures before they are optimized and exported for the game. Just for mechporn purposes.

[Troutmonkey]

Mycrus, on 27 February 2016 - 03:03 AM, said:

Unreal sucks for triple screen and/or 3d

Be careful what you wish for

Hmm. 3D I don't care for, but MWO is the reason I spent lots of money get a triple monitor setup. That would be a deal breaker for me

[TELEFORCE]

I just hope an engine upgrade gets us working inverse kinematics, better collisions (death from above!), and better looking graphics!

[TKSax]

Nik Reaper, on 26 February 2016 - 11:32 PM, said:

Hm, he's response to a tweeter showing him the new now in "beta" amazons open source lumberyard engine was: " interesting " so foreboding anyone X) ?

Amazon bought a Version of Cry Engine 3 (for a large sum of cash, which help keep cry tesk solvent) that they modified and release as "free". Pgi should know all about it because I suspect that is why Karl Berg went to work at Amazon....

Edited by TKSax, 27 February 2016 - 04:54 AM.

[Narcissistic Martyr]

Sandpit, on 26 February 2016 - 09:36 PM, said:

Russ says MWO could be moving to a completely new engine and hasn't decided between that or the cryengine update

Oh please great merciful goddess Eris avert your attention from this matter lest your chaotic nature make this a delightful disaster.

Also... please Russ choose an engine with great server authenticated net code. Your net code has been the biggest thing letting mwo down (yes biggest) since the beginning so if you're gonna make the effort please please please choose an engine that fixes the single biggest problem with mwo.

[C E Dwyer]

I'm kinda nervous about this, I mean its a very big project and well, three years to this state, just when PGI are finally developing the game at an acceptable rate, they want to change the Engine, then bug hunting, P.G.I's quality control isn't the best in the world.

MrMadguy, on 26 February 2016 - 10:45 PM, said:

If it will be even less optimized, then current one - then I'll have to quit this game. Yeah, when I was a kid, I was playing CS with 20FPS and didn't even notice, that something was wrong. But now I used to play this game smoothly with 60-75FPS on ultra settings. I won't play with lags, sorry.

don't let the door hit you on the way out

[oldradagast]

Chimera11, on 26 February 2016 - 10:11 PM, said:

The difference between this announcement and the previous town hall's is that the previous discussion was on the basis of simply upgrading to a new version of the current engine. Now, Russ is talking about changing to an entirely different engine, as in no longer cryengine. Apparently, it would take a similar amount of work to update the current engine as it would to get a whole new one.

Being involved in system-level software implementation, though in the realm of engineering tools, I find that part particularly interesting if accurate. Normally, going to a new tool is an order of magnitude more complicated than going to the next version of the tool you are using... but, again, that's in the realm of corporate level engineering tools. I wonder what the difference is here?

[sycocys]

While it's sort of a necessity for them to update the game engine at some point, they don't really have the coding staff to make it happen any time in the next year or maybe even two. That means by the time they actually move on the next engine will be out of date as well.

It's also an excuse for them to not actually develop the current game for the foreseeable future.

Edited by sycocys, 27 February 2016 - 06:27 AM.

[Wintersdark]

MrMadguy, on 26 February 2016 - 11:14 PM, said:

Some players, who don't understand anything about game development, assume, that engine update will magically bring some cool features. For example many players assume, that engine update will dramatically improve graphics in old games. The real truth - engine is just a tool to render art, that is created by artists. New engine won't improve graphics in outdated games, like Wow, just because almost 0 models in this game have bump/displacement maps on them and without them models will stay as flat, at they are now. Artist would have to create them first and it's too hard work to invest resources into it at this moment. Higher speeds will be allowed? Dream more. Speed limit depends on server framerate - not on game engine. Most of you don't even understand, that server side has nothing to do with CryEngine.

So I really doubt, that new engine will improve anything. But what I know for sure - is that DX11 is much less optimal in current engine and causes sudden drops of FPS and that newer versions of engine don't have DX9 support, that works smoothly.

An engine change can lead to things that aren't possible (or practical) to implement now becoming possible/practical in the future, however. There are a lot of things that are the way there are because of engine limitations.

That doesn't mean those things will all change with a new engine, but it means they'll be able to fix those things in the future. That's great news.

Graphics is basically the same here. A new engine will (probably) not bring better graphics out of the gate (though more recent engines have support for things that don't exist in the current one, so it's not impossible) but what it does do is allow a higher "ceiling" to how sexy they can make the game.

As well, a lot of optimization issues with the current engine arise from years of fudging in solutions to problems that cropped up. Whether they go to a wholly new engine or port to the new CryEngine (either way, they're basically doing the same work), they can avoid a lot of those hacky solutions and implement them properly from the ground up, leading to better optimization.

[pyrocomp]

Troutmonkey, on 27 February 2016 - 04:43 AM, said:

Hmm. 3D I don't care for, but MWO is the reason I spent lots of money get a triple monitor setup. That would be a deal breaker for me

Using game engine to run triple monitors sounds just like software emulation of graphic cards a decade ot two back. It's better handled at driver level, which is a different matter.

[Wintersdark]

DV McKenna, on 27 February 2016 - 04:07 AM, said:

Never going to happen the best you can hope for is the upgrade to cryengine

Russ, directly, said it turned out that upgrading to the latest CryEngine (it's being announced at GDC soon) will be basically the same work as using a different engine entirely.

He didn't specify which engine, but very strongly implied they had one in particular in mind, and that the decision would be made pretty soon which way they'll go.

[Wintersdark]

oldradagast, on 27 February 2016 - 06:14 AM, said:

Being involved in system-level software implementation, though in the realm of engineering tools, I find that part particularly interesting if accurate. Normally, going to a new tool is an order of magnitude more complicated than going to the next version of the tool you are using... but, again, that's in the realm of corporate level engineering tools. I wonder what the difference is here?

Presumably, because they're running CryEngine 2, and Crytek is about to announce CryEngine 4 at GDE - so this isn't a single version bump. Further, because the version of CryEngine2 they're running is massively customized - this is why they've stayed with it for so long. They can't port their changes easily, because the underlying structure of CE3 and 4 are different.

pyrocomp, on 27 February 2016 - 06:48 AM, said:

Using game engine to run triple monitors sounds just like software emulation of graphic cards a decade ot two back. It's better handled at driver level, which is a different matter.

Yeah, a triple monitor setup should be identical to a game as running one ultra-wide monitor - just a different resolution, the mechanics should be covered completely by your display drivers.

[Oderint dum Metuant]

Wintersdark, on 27 February 2016 - 06:49 AM, said:

Russ, directly, said it turned out that upgrading to the latest CryEngine (it's being announced at GDC soon) will be basically the same work as using a different engine entirely.

He didn't specify which engine, but very strongly implied they had one in particular in mind, and that the decision would be made pretty soon which way they'll go.

Even more reason why it won't happen. Look how long it's taken to get this far!
To re do it all from scratch in a new engine would be unfeasible for this team.

Not to mention that moving a live product to an entirely new engine is almost unheard of

[Appogee]

I think it would be wise for PGI to move to an engine more in keeping with their coding capabilities.

Edited by Appogee, 27 February 2016 - 08:52 AM.

[Goose]

Does nothing ever done by licensees get back to CryTek? Is their support as bad as their .docs?

… We are going 2.0, and jack-all of our stuff is transferring over, isn't it?

[GrimRiver]

Too bad they didn't put it on the unreal engine in the first place, then it would've been a simple matter of updating(well not "simple" but you know what I mean)

Unreal has always been the better engine with a better physics engine and it's much easier to work with(just what I've heard) hence why more games use unreal over cryengine.