Meanwhile, Robocop: Rogue City came out, it uses both Nanite and Lumen… and it looks and performs great.
First person means easy motion vector re-projection for temporal upscalers.
That game performs terribly on modern GPU’s requiring blurry upscalers to offer the basic standard of 60fps and doesn’t even look much better than plenty of PS4 games. Thanks for making my point more valid.
man do you even think and check before you write something! where did you see the robocap working well? if you mean tests like rtx4070ti and up to 4090, you can throw it in outher space! i have no doubt that aaa+ studios that make games like robocap (mostly singleplayer) can do much better at 60fps and ps5. do you have any idea, and have you looked how robocap works on rtx3060 and below that(older gpu). it can’t even push 30fps on lowest settings (and dont count on fsr and dlss please). what do you think most indie developers are hoping to sell? to someone paying more than $300 for a gpu, our market is primarily for the sub $300 gpu crowd. now ask yourself if the same can be done in the UE5 for such a market. the answer is simple, you can make puzzle games or 2d games or something like that but in 3d there is no way. so let’s take for example a cassual multiplayer game like unreal tournament!!! let for example count a maximum of 32 players on a server with average graphics. on the UE5 platform with the rtx3060 you got average 60fps on medium settings. the same version of the game on the UE4.25 on medium settings coolly pushes 150fps. means no lumens no nanite in both cases.the only promising example is actually the finals. but keep in mind that it was done by a top aaa team and that it is not a shooter where you have an arena with 15+ players in your field of vision, fights usually take place with half as that maximum. and be sure to keep in mind that i calculate all this without dls and fsr. when the finals start on e.g. rtx2070, you can’t go over 60fps on the lovest settings, but on the other hand, you have a squad game that is god knows how old, and not made by aaa studio (not even optimized well), which coolly pushes 150fps on the same configuration in a similar (buildings everywhere) environment. the point of everything is that currently in the UE5 no one can make a massive multiplayer shooter game like unreal tournament or battlefield (fortnite is special case, aint use realistic graphic and also aint scenes with 15+ players figthing arround), not even the strongest aaa developers, and as far as our indie developers are concerned, there is nothing of a multiplayer, in 3d possibly only a single player and that without larger dynamic and interactive scenes and that’s it.
The Finals performs well since it doesn’t use lumen nor chaos. It’s running on a heavily customized ue 5.0 that uses Physx 3.x and Nvidia RTX GI, both of which are more performant than their standard ue5 counterparts.
Even still the game performs worse than for example Battlefield V, which in my opinion looks generally better and features more players/ bigger maps.
It does perform well if you optimized the settings, still very blurry though since they forced TAA/upscalers.
This is such a disgusting trend.
This quote feels right at home. Not only do I also have an accessibility issue with blur present from video games (they make my eyes tear up and I cannot endure them for long)
I also do not like the visual representation it provides, often is far from good to not notice at all, and it ruins the visuals greatly to be represented as unsharp outside of standing still shots.
The console targeting also doesn’t make sense.
This post has my vote.
Yeah I’ve noticed lately a bunch of games look real blurry with default settings. TAA and upscaling disabled they look so much more crisp. Unfortunately that turns hair into pixel soup on a few games. I don’t understand why anyone would make hair dependent on a blurring algorithm.
Trend is not relegated to UE5 either. Rogue Trader, a Unity game, is super blurry with default settings. Have to tweak AA and turn off upscaling to make it look decent.
I dunno if its people used to gaming on large screen TVs 10 feet away, but the blurry graphics trend is awful imho. 
I wonder if you guys would have griped the same about LODs and mipmaps when devs started using them decades ago…
i’d rather see pixels than feel like i have -500 myopia not a fair comparison…
LOD’s still have to be done properly or they stick out prominently. Mipmaps have been serving a purpose still without moderate downsides.
I can’t really say the same about the current state of things and the blatant loss of information currently. It’s not something I’m ever going to be happy with but have to deal with because of the growing dependency on it. I’m never going to be sold on filmic anything. Chromatic aberration alone prevents my friend from being able to play a lot of games without migraines.
It makes me think of all the pining over AV1. The output is still going to be butchered with 4:2:0 chroma subsampling and/or limited range at the end of the day for most places.
Temporal and amortized effects also severely fall apart at lower frame rates.
There is a fine line between Technological Progress, and Consumer Affordability. We all must, at some point, face the self-realization of our personal economic status. This realization should not be reflected on the status of the progress. But rather, be used for self-reflection of personal goals. If technology would have allowed me to, i would still be using my 1981 IBM PC-XT IBM-prototype.
But my personal feelings in this regard, should have no impacts on the technological advancements available to others, or those of the industry as a whole.
It is Topics, such as this one, that make us all take a step back and look at the dynamics of the problem, rather than casting blame.
Hows Mover 2.0 ? I want to see how performance is looking compared to the old Mover. I had this 200 player project were you could notice how hard it was hitting performance. Would be good to know if mover 2.0 is somewhat lighter or actually even more expensive
LODs and mipmaps don’t break down and look horrible if you change a basic AA setting. Not really comparable. 
Some game hair rendering turns into a dithered mess on anything but TAA. With TAA off it looks like a rendering bug. That’s why it seems questionable to make a game dependent on it.
Some game hair rendering turns into a dithered mess
Add glass, foliage and shadows to that.
Heh wow that’s crazy.
i was recompile newest ue5 (its even ue5.5 on splash) to try mover 2.0, just brief short insight in example maps included in mover. first, i only succ to test in net emulation mode in pie listen server, and that aint work nice, lot of jitterring and rubber banding but its cant be taken relevant (barely tested, to many crashing). on first i cant see any serious performance impact over vanila cmc its even look there is not any. there is also example of moving platforms witch is very nice considering that is one of largest problem in multiplayer. also there is no any example of root motion, and its look it will be simillar problem trough blueprints as is in cmc, at least in this moment. but must to mention there is no any sign of change in physics performance.
interesting observations, do you notice in ue feature update, they have mass even tons of different features anouncements, updates and improvements but none literaly none adress core chaos solver, and any potential improvements about!!!
I just looked: https://portal.productboard.com/epicgames/1-unreal-engine-public-roadmap/c/1372-medio-io-minimize-tsr-taa-artefacts
103 votes, we are not too far away from 1#.
So far, we are getting paralyzation optimizations and bindless resources which can both benefit real-time performance. But unfortunately the Temporal focus will remain which is no surprise.
Here is an interesting test done in 5.0 with Nanite and a quote from the post regarding the results:
To Epic: Stop trying to force something that has worse performance as a bandaid to your workflow issues and deficiencies towards LODs.
I will be replicating the test in 5.3 when I’m done doing some more test in 5.0.
Someone remind me - how do you DOWNVOTE a thread?
This is such nonsense. ‘the unethical standpoint…’ seriously? wtf is the matter with you?
The reason why games consoles are a positive thing is because they become a development standard for the period they exist. This means developers target them (logically) because IF they want to release onto another platform like PC, which is dynamic, they can increase render perf to suit. tbh if you want to have a go at someone, have a go at teevee manufacturers, as they are the ones making UHD4K teevees, not Epic.
Nanite, or more accurately ‘mesh shaders’ are inside of the API which Epic Games did not develop. OpenGL, Vulkan, DirectX all have some variety of mesh shaders and were NOT developed by Epic Games. In addition to this, you do NOT have to use nanite and can use whatever you want, a little bit like y’know, you do NOT have to use Unreal Engine?
‘Lumen has a great future’ that is such a ‘lol’ comment to make. It is NOT the future, it’s here and it’s on your desktop/laptop running just fine, and, if you really want awesome speed in an fps game, just lightmap the **** out of the level and get yourself 9,000 fps at UHD8K.
Also, I like the way you side-stepped path tracing like ne0…
Third person (adventure) games have always avoided using pre-calculated lighting, because the amount of texture memory required would be fubar for storage of textures. Real-time lighting is important, in spite of what you say.
LOD’s are a pain in the neck, because they offer nothing of value other than mary-poppins, and, they need to store more geometry than you’d want. Again, in an FPS arena-style game, like valorant, or, counter strike; LOD’s and lightmaps make sense, but in an adventure game? no thanks. And again, you do NOT have to use them, you can use LOD’s if you want to, so I am (again, again, again) totally confused as to what you are complaining about.
If you want to work for Epic, you need only apply.
If you want to invest money in another company, you need only speak with your financial advisor or stock broker.
If you want to develop your own game engine, you need only begin today. And, bare in mind what the source code of Unreal Engine was originally, way back in '98.
If you want to use another game engine, you need only buy a licence for it.
Which brings me onto another question: have you actually paid Epic Games any money for a licence?
I thought not.
Those that know how to get their game looking and performing they way they want it to, would never, ever make such an insidious and downright disrespectful forum post to any company in the games industry.
There are a lot of very intelligent people here, and I think that if you had the slightest modicum of respect for your peers, you NEVER would have posted such a tirade against the very company who OWE YOU NOTHING.
Now then, sir, back to business:
I’d like a mocha latte and a shot of caramel syrup, barista boy.
Welcome to my block list.
Your completely missing the point.
This engine is built around an accessibility option(TAA/Temporal blending)
The problem with you is you think this is about my project, this is about the semi-dynamic games being brought to UE in the future. Semi-dynamic, I’m talking about DOORS, windows, basic things that UE’s baked systems can barely handle with good quality. They LACK systems and workflows that promote reasonable performance for players.
They are admitting to wanting to focus on replacing meshes with Nanite, They are admitting to trying to create MORE bullcrap in TSR even though more temporal frame=more chance to smear. There are ways to make this better for everyone and they are investing in poor directions.
I’m asking for basic standard and so are the #104 who voted in agreement.
GameDev needs more priority right now becuase the games that have used it perform terribly and they are mostly static.
Basic systems like volumetric lighting still have stupid amounts of leaking etc.
They are abandoning all reasonable systems and studios are just going with it because of OUT OF CONTEXT information.