Actually this topic is intended to be a black hole for such behavior. And, so far it seems to be working, but we’ll see when we get about 10 pages in.
Sadly, Trolls can only be contained; their regenerative powers (and whopping 3000 HP) make it hard to truly defeat them. However, they are weak to hammers…
I know the old argument, that it is up to the artist, but here is one of the best thing I have seen (of similar nature) in Unity
Though they are both very talented artists, the ones made with unreal just have that extra something, which I assume may have to do with the engine used.
On a slightly different node, this fan made pokemon on cryengine, is looking pretty impressive too (the render quality anyway).
I don’t see why Unity 5 shouldn’t be able to. Most of the looks in the UE4 demo come down to the assets, and Physically Based Shading (Unity 5 will have this as well), since that makes it easier to get materials to look “right”. That and good post-processing.
W
What you are describing is more level and environment design. Making a cool world (level, corner) to run through is nice, but that’s not a finished game.
The UE4 video is a staircase with reflections and rain. The Unity video is architectural visualization. I could most likely replicate the UE video in Unity, but I won’t bother.
The staircase is not something finished, it’s just a mood demo of a few square feet. The Unity video is a finished and most likely paid for project that can be used to actually visualize a space you can walk through. I understand it might seem more impressive, but the reality is that the projects and their goals are completely different, not to mention that the UE4 one isn’t even a project.
Now the CE one clearly has some problems:
It’s cool as a fan project, but if they think it will actually lead to anything, well… good luck.
Visually you can say that the other engines are superior, but the only project with value is the Unity one. I can’t even compare them: a staircase, a house and some models in Sandbox environment.
I went to meet these guys last week, in Helsinki. Only 2 Artists and its all done in Unity 4:
Oh, and other than a few obvious moments, thats in-engine and part of a full-game. 
Oh, and here is a Making-Of video, from ADN, who scanned someone, and imported them into Unity 5 to show off it’s rendering capabilities.
I think the best way to compare the visuals would be for an artist to recreate the same scene in both UE and U5. Not just use the same assets, but to try to achieve the best image quality. Until then we can’t really compare anything.
Well skipping past the fact that’s obvious beyond belief, the level design / artwork / animation portion is still a massive part of the process and one where the most amount of money get’s drained in larger 3D games.
Also were talking about engines not licences, it’s obvious Nintendo won’t license it’s prime IP out.
Fact of the matter is in terms of OMG GFX, there all based on a DX11 / GL renderers. The key component is how much work YOU have to do to make it look good and keep a game performant.
I’m not impressed with any of the showings, the arch viz demos from UE4 are a bunch of cubes and a showcase for lightmass / maybe the particle system for the rain, the arch viz from Unity suffers from poor post and lighting. The pollen trailer is clever, a cool showcase for two artists but as soon as the Bee moves out of scope it starts looking all “Unity again”.
The forest would of looked much better in CryEngine!. Not to say it’s priority number 1 as Skyrim wasn’t exactly a “gorgeous” looking game but sold very well, then again it’s three years old nearly.
Could you post some visual examples from different engines, that impress you? You seem very hard to please!
Witcher 3: Capable of rendering far distance geometry, great terrain / foliage shading / particles / lighting. It is the whole package:
Crysis 3, more of the same. If you use CryEngine and some decent artwork it’s pretty much how it comes out the box…
Unreal 4 more of the same, although I’ve not seen much come out of UE4 yet. Just shiny examples as opposed to games:
I had a really big post, but the fact is I decide I’m not going to go down this road.
The gist of it is this: are we comparing what can be made by a modest team, with a modest budget in these engines or what corporations that are from almost 100 to almost 1000 can throw together in the span of 3-4 years (with source access to their own tools and the teams dedicated to their development being in house)?
What you are expecting is out of your price range, even if you can get Unreal for $20 and CE for a tenner. I have seen your project in the past and for an indie team it was good, but what you see in the Witcher 3, is not something the engine will solve for you, it helps, but it won’t solve it.
You don’t know our “price range” and nothing has been shown bar a few concepts to acknowledge the project exists, what gives hope is not what AAA’s are doing but invdividuals in the WIP section. There are plenty of examples of small teams and one man bands doing good looking games in UE4, I fully believe you can create games like Arkham Knight / Witcher 3 in UE4. Maybe not to same scale / detail or complexity. But it comes with all the tools necessary to create whatever you want.
Even in Unity you have people like Manufactura 4K doing great level asset packs, so it’s not un-reasonable. I just doubt we’ll drop 20 - 50 million in marketing.
It’s apparent what you can do by just using it yourself, there’s no guesswork or opinions to change this. It’s still early days and projects take years, we’ll see…
I think what is really needed is a standard scene, complete with textures, which can then be taken into each engine (by whoever has the time) and tweaked to get the best possible look. A visual benchmark so to speak.
The scene could be something that shows off different aspects of the engine capabilities, so it includes:
I know this is like opening the door to a fireball storm… but why do some of you put so much emphasis on the visuals? I appreciate the talents of a good artist as much as anyone but at the end of the day this is what I think has become a big problem with games.
Too much focus on the “gee whiz look at the shiny objects and the shadows” and too little focus on the game-play. If a person wants to experience the ultimate in real world visuals then open the door and walk outside. Awesome frame-rate, very realistic lighting (one would hope anyway) and subtle shadows. All are there waiting to be discovered. Or put in a movie. I just don’t understand placing so much focus on how things “look”. Unless at least as much focus is being put on how it plays. And that is usually not the case.
Because the customers do, but I don’t believe it needs to be at the bleeding edge of graphical technology , just consistent. The issue stemming from what’s your competition going to look like in 2 - 4 years time when you release your game?
So you try and make it the best you can now so at the point of release it doesn’t look 5 - 10 years out of date before it’s lifted off the ground.
The problem with a standard scene is that even if you match it in every engine it’s not representative of anything really. ShadowK has used Manufaktura40K’s packs and still had issues. Who knows how big he wanted his environment to be for example.
Basically what he needs - I assume - might be a huge streaming world. That’s great and it’s nice that other engines can do that out of the box, but the vast majority of games don’t need that feature. The example scene won’t cover that, so matching the visual fidelity is a none issue, since you have to match other features first.
I understand his frustration, my game has passed Greenlight and it’s Unity based. Getting performance to be optimal on a dense terrain is difficult and a pain, but I wouldn’t switch engines just for that. For one, last year, Unreal was asking for 25% of revenue, their tools were crap compared to Unity and you had to conform to their workflow to get things working.
Cryengine was a joke with people waiting for licenses since 2012 and losing projects to forum admins…
As we have asserted in multiple threads in the past engines are tools, use the right tool for the job you want to do. The problem here is that sometimes you have to do a lot of stuff to understand which tool is better for the job you want to do. Even ShadowK says it looks promising based on forum threads, I guess when he done what needs to be done we will know if it was for the best and most likely will be, because Unity’s main problem area is huge open worlds. (this is mostly due to a terrain tool that hasn’t been battle tested and I think was the result of a ninja camp or external project brought through, it’s quite possible you can do better if you invest the time).
They might not absolutely need it, but they sure can use it to eliminate loading times or at least keep them very short.
My main interest is the visual right now, I want to make photo realistic scenes. Other features of game engines can have their own tests/benchmarks done too for the interested parties.
The only way to really compare all the engines, is to compare each one of their features one at a time and then have an overall summary. Based on that summary, people can better understand which engine suits their needs.
I don’t think there is any problem in having a benchmark type scene, it measures one aspect (in this case rendering), in a standard manner across all engines. How you load those scenes and any issues around that is another matter.