hit game idea micro mini multiverse cells a game about creating cells that produce micro mini multiverses at high speed semi popular youtube channels and an app this could be a big profit game and is a super interesting idea
I gotta admit, I'm not a Unity fan these days but those statistics, without ostensible evidence, are pretty freakin' dubious... especially with a $90 price tag and no free version. I'm taking most of this with a grain of salt, especially given the number of zero-cost and low-cost engines which can already do most of what they're claiming. Smells a bit scammy to me.
Poor decision making of segregating from free to Pro tier IMO; programming and languages is still skill based so even if C++ was in the free version, they still had to work to make something, they could had tried Pro version has VIP support for X months.
Scam, probably not. A cost effective product with all the features needed to bring a game product to market? Probably not. It being 10x faster by default because it isn't built on a decade of cruft and doesn't have a ton of design decisions built around usability and interoperability, etc sounds plausible enough. But Unity can also run fast enough for anything but projects too large for this either. Ultimately, this isn't going to be as easy to develop in as something fully featured like Unreal, CryEngine, etc. It's a ton cheaper, but ultimately you're making those kinds of big trade offs with any kind of 'small' game engine. I reckon it kind of suffers from the same issue stuff like 'small' video editing software has: It's too pricy for true hobbyists, but also not quite powerful enough for anyone dedicated enough to commit to $200+ software packages.
Not providing a free version makes it really hard to test. I can also see that they provided the benchmarks as executables but that makes it really hard. Some questions in my mind: - Why there is no Unreal Engine? - Why use built-in instead of URP? (for use of SRP batcher.) - Which version of each engine was used? Was Unity LTS or latest 2023 release? Making something faster then Unity shouldn't be too hard if you just use C++/Rust and not dive into the cross-language function call territory. Also I would love to see some feature comparisons with Unreal/Unity.
yes, instead of press harder with c++, i can't understand why people just using a sh*tty alternate language especially like python or java. they're like a Coward trying to avoid being hit, so they use a cushion language, and they often swear in defeatist like "we don't need performance that much" those coward might as well use Php or VBA for their game engine out of Unreal Engine, it was very hard to find a Game Engine that was Ballsy enough to use C++ as their main language,
@@jensenraylight8011 im aware of the shift. But still it doesnt mean that by your comment sounds with vexed emotions, if it was satirical it isnt applied correctly.
They haven't managed their documentation well in the past. They introduced breaking changes and took the docs offline for quite a long time. I can't recommend going with this engine, in case the shitshow continues
The dev seems to be afraid of free version as many people can go and test out the "10X Unity performance" which is obviously 10X over exaggerated statement.
He has all the project files (unity and ultra engine) on his github. So you could pretty easily try it out yourself. If it is far off reality, just refund it. But they are trustworthy. Never used the engine, but other tools from them. They exist since like two decades I guess. So they aren't bs'ing around. They just do things slightly different than everyone else. And I also think it's only one person doing this.
I hate to agree with this because I was a fan of Leadwerks for years (and even wrote small tools for the community) but it's true. Unsolved bugs on important features like character controller and navmesh and abandoning mobile support because it takes too much time to work on and not giving customers their money back (separate extra $100 purchases for Android and iPhone) are some of the issues. If you have the money and don't mind tinkering as a hobby, have fun. If you are serious about game development, use a serious engine.
Ultra Engine got its docs redacted while introducing breaking changes and at the same time the developer launched a "new" thing named "ultra app kit". I can't fathom why they thought this was a good idea and I won't trust the reworked Ultra Engine for a while.
Agreed. Leadworks was impressive enough as a small-team project but it left a lot to be desired just from a UI/UX standpoint. Most devs looking at tools like this are going to be indies. And above all else indies need effective workflows that allow them to develop as fast as possible. Selling this engine on performance alone (assuming it even is as good as it claims) is tough. Like, I can write an engine 10x faster than 'the next leading brand engine' too. Indeed I have. But it doesn't offer me any of the features that a major engine out there isn't already providing at 100x.
Not even source-available for the subscribers. Dev: "Ultra Engine is specifically designed to yield the fastest possible performance for VR and 3D and provides order-of-magnitude faster performance over other engines. It uses many trade secrets I spent years researching and developing, and I am not giving the source code out." Mhm i don't know that i'd get something from someone who is so full of themselves.
Instanced rendering is extremely efficient how the hell do you get 40 fps in unity using instanced rendering, meaning you're rendering millions of objects and somehow their game engine gets more than 1,000?? If they're not lying this is either a breakthrough in computer graphics or magic
If the numbers are real it'll be them using a worst-case scenario for one that the other is optimized for. It's why the idea of a 'perfect' tool doesn't exist, when you remove incompetency you're just left with different priorities.
Simple: flawed tests. Might be unintentional, but I would be hesitant to trust devs that introduce breaking changes and take their documentation offline to launch a new thing in the middle of development.
I briefly skimmed over their paper on what they're doing and saw that benchmark and it seems that what they are really mostly measuring in that benchmark is frustrum culling on the CPU. They created a 3D grid of 30K+ instanced cubes that are all visible. With frustrum culling enabled, Unity hits 40FPS while Ultra Engine hits 1206FPS. The writer claims the majority of this increase is that occlusion culling is being ran at 45hz rather than every frame and their "advanced threading architecture" (which as far as I'm aware is pretty much the same as Unreal's where there's a GameThread, RenderThread, and Occlusion / RHI thread. Don't know how it compares to Unity though) and that they don't send data every-frame to the GPU from the CPU as that's a bottleneck in applications (I'm assuming since the objects in that test don't move they don't send any data so there's no bottleneck unlike Unity perhaps? Also they compress their data when sending it over to the GPU so it doesn't take up as much bandwidth vs full precision floats / doubles).
The performance is mindblowing and it has a lot of potential to be great. Gonna be tough to compete against Godot, UE, and Unity though. Their performance isn't great, but they have support for all devices, have good licensing, have plenty of documentation, and are easy to use.
@lillybyte 3D is doable in godot, it is just not beginner friendly since it lacks the same helper functions unity and unreal. You have to have an extensive knowledge of shaders and math along prior game dev experience before tackling it.
@@rytifwhat do you mean? They've already implemented Vulkan renderer and it seems to be getting even better in upcoming versions. They already have C# support and that is pretty good too.
@@kloa4219 my dude, been using Godot for eight years, the 3D is absolutely trash, and worse the engine can't even load anything larger than a demo. You will be waiting for hours for it to load any significant game, if it loads at all. We had to move a 3D game to UE because it couldn't even load a quarter of the level we wanted.
This looks like something I would be very interested in but you are correct. Without a free tier I'll never try it. You gotta have a free tier if you want people to move over.
Even if it was free, I would not recommend it due to the documentation being handled so bad that it might as well be a clone of Unity. One does not simply introduce breaking changes and redact all the documentation.
@@TheExileFox I mean that's mainly my point is that unreal and unity are free and Much more established to think that they could come in and take any market share without a free option is unrealistic
Great review! I've been using Leadwerks for a few months now, and honestly really loving it. Not too fond of what I like to call "feature bloat" in certain other engines, and just want to get started coding and working on levels. In addition, because I used to mod for the original Half-Life way back when, the editor is very familiar to me. All in all, Leadwerks is kinda perfect for me in terms of my current projects, and Ultra Engine just seems like an upgraded version of that. In fact, I think I read that it should be able to load in Leadwerks projects, too. So I might migrate to Ultra Engine once it gets its full release.
They have been mismanaging the documentation in the past by introducing breaking changes and then redacting the docs entirely leaving you with next to nothing for a while. I'd be very hesitant to try this engine given the history of the project.
If this is the offspring of Leadwerks that would explane why there is no free tier, the creator is all about the money first. Also Leadwerks started off life as being made from Blitzmax from Blitz3D and could explane where the clunkiness comes from in Ultra Engine. The 10x better than Unity is the creators huge ego talking and probably arent based in real world numbers.
Their marketing is super bad. If you go to their website it will show that you can get 1000+ FPS compared to Unity. On top of it they don't even have a free version. Maybe the engine is really fast and optimized but oh boy the marketing doesn't do any good for them. It really sounds like a scam LOL (I am not saying it is a scam just saying it sounds like one.) Also they DESPERATELY need a free version or this is gonna be a big flop. Looking to hear more from it.
Yeah I cover the grave sin of not having a free tier/version/trial in the video. It is their biggest mistake IMHO (and was with their previous engine as well).
I understand they want to capitalize on the recent Unity blunder, but if they wanted to brag about performance, Unity is not a good comparison. Nobody chooses Unity because it's fast, they choose Unity for its platform support and convenience (C#, big community, assets store).
Hell yeah, I've been wondering why there isnt something like this already. Im doing a masters in games technology and just did a project in opengl after working with unity and this is what I feel like I want, just a 3d editor, with asset management and animation tools. The coding of game mechanics, physics and architecture is something I want to do, I basically want to type more and click and drag less :D Btw gamesfromscratch I want to say thank you. I've learned a lot from your videos - I'm doing really well in my course and that's partly cos of your channel thanks mate🙏
one thing that looks really intresting to me is the way the editor is layed out like the old hammer editor for valve games. with a 3d free fly view, a top down grid view, a front grid view, and a side grid view. could potentially be very usefull for retro styled games
They seem to be very inspired by Source and the like. Leadwerks was a lot like that too. Nice to see it making an appearance as other engines seldom use that style of level editing which is sad
I got that same impression lol. Totally missed the viewports being like Hammer's but I was reminded of their FGD definitions when Mike opened the properties sidepanel and then later on when he drew a brush on the viewport. I've heard Hammer Editor being well regarded because of how simple and fast it is to prototype a level, but it's surprising how this style of editing is left up to be implemented through plugins or absent at all...
@@Aidiakapi You can, but there is no 2D ease of use like in Hammer. In Hammer, the 2D viewports let you resize and drag brushes around like they're images in Photoshop. In Unreal and Unity, objects have a pivot point. Unity doesn't even have BSP brushes so whatever there, but Unreal's have pivots, but Source's don't, so Source's can be manipulated really easily, whereas Unreal's you find yourself pulling faces around only to find the pivot is way off and it's super annoying.
Just rendering 3D models very efficiently is nice, but it is lighting and reflections and all that jazz, where things tend to get more complicated and resource intensive.. It is forward renderer only, so I doubt that a scene with say 50 lights would render faster than the competition. They don't seem to have many lighting features yet, so I wish them good luck implementing those! I think a bigger game-scale scene as a demo is something they should strive for to actually test their performance..
Im checking out their news page right now and appararetly the background that lead to the claim is that inVR simulations you can not optimize the polygon count as engineers for example need to be able to inspect all the details. Which makes sense, but honestly now im wondering if this shouldnt be called game engine then if the focus is so heavily set on simulations.
Love the idea of a code-first game engine, especially with C++. I could see a lot of Raylib users liking this engine. It would be interesting to see if systems could be easily disabled, like the physics engine if a game doesn't need physics, for example, so you could get even better performance. That level of easy customizability would be a huge selling point to me. Free/trial version is definitely a must IMHO. Also, gotta see if mobile and maybe consoles will be supported at some point. I'll definitely be watching this project closely.
Reminds me of back when I was using Dark Basic Pro or Dark GDK. Not the best to work with, due to it's limitations, but was simple enough to learn. This to me, looks like a vastly better Dark GDK. And I love that it has VR support. To me, this might have been worth looking into, if not for the advancements currently made in UE5.
And looks like there's no upgrade path. Say I got Lua engine first and decided to upgrade to C++ paying the difference I cannot do that. Need to buy Pro version and then I'll end up having the Lua-only version redundant. "Pro" version should have been a DLC rather than a separate product.
My current project will take approximately 6,000 hours to complete in Unity. I am currently at 3,000 hours. Based on what we saw in this video, I estimate it would take me at least 10,000 hours to build the project with Ultra. I like the concept, I just don't have the time...
Saying you made a game engine faster than Unity is just saying you made a game engine. If they made a fully functional engine that was _slower_ than Unity then yeah that would be impressive. Rip whatever the name of this engine was
The alternative forms (and, or, not) are part of the C++ standard and are used in countries where there is no easy keyboard access to symbols like & or |, which make sense in an US keyboard but not so much in some international layouts. You used to have to #include "iso646.h" to use them.
Pretty excited about this engine, each copy doesn't cost that much and the royalty licensing is more than fair. But like any piece of software, it's only as good as its documentation.
"only as good as it's documentation" nail in the coffin already perhaps. Because they have previously been introducing breaking changes while taking the entire documentation offline for months. And in the middle of this, they launched "Ultra App Kit". It's like an insult to injury.
@@gamefromscratch np. Great video, anyways. It'll be a very interesting product to follow and see whether they manage to improve the UX when using the editor. Thx
this engine looks really promising, and not really bc of the performance aspect, if its even remotely as fast as it advertises itself, I am more than fine with it, even if its paid. bc the biggest problem I had with both unity and godot, is the lack of *in-engine* c++ support. because I really really like c++, so when I do get an actual job, this might one of the first things I purchase of off steam. just out of respect. (I really really like c++ and devling into the godot source code isnt hard at all, its just a little more fiddely than I'd like. (but thats expected, it is the source code after all))
Looks like you'd be happier using a framework instead on an engine, if you like C++ so much. SDL, SFML, Raylib or even OpenGL/Vulkan so you can do anything you want. Engines imho are needed to minimize the coding.
This is the great thing about the claims of 10x performance. You've already bought the software to find out the truth. How many are going to bother going through any return process, if there is one.
Yeah but 10x more performance than what? In Unity, you can use forward path, forward+, or deferred -- each will have different performance. You can use GPU instancing, static batching, dynamic batching, SRP batcher, or the new GPU Resident Drawers, or if you want to get some serious performance you can use Indirect Instancing or BatchRendererGroup. Also there will be a big performance difference depending on if you're using GameObject for object oriented programming or if you're using ECS with DOTS for data oriented programming. ECS is crazy fast. And I'm just barely touching the tip of the ice berg here.
I am so sorry the developer decided to focus on VR/Windows only and make a brand new VR focused engine from scratch instead of working on Leadwerks but sadly it makes sense from a business perspective. As Leadwerks being super simple to use and no royalties, many US Federal agencies including NASA and others decided to use Leadwerks for their VR simulations and they did hire the developer to work on their VR staff even. This is where the idea of this new VR engine come from. I wish Leadwerks was good enough for them.
My indie team and I moved from Unity to Unreal Engine 5.3 about a month ago, and I m SO satisfied with that decision. UE5 is so good it's literally "UNREAL". My only regret is that I spent over a decade on Unity before trading up. Not sure what Ultra hopes to gain by introducing another money-up-front game engine into the market when there's so many top-tier, free to use options already. Seems a bit suss.
I would be interested to see how does it fare against C++ based engines. (UE5, WickedEngine, Ogre etc). But seeing this vid, I'm not convinced. Doesn't it do basically same or less things than wicked engine?
Wicked engine doesn't really have the tooling side, but WE + A level editor would be very much a competitor to this. WE has partial tooling of course. I also think from an API perspective this might be more approachable
This engine already has a troubled history. Tldr: the predecessor Leadwerks is not friendly to use. Later the dev introduces breaking changes and takes the entire documentation offline for months.
It's weird that they went for Unity's throat by using another language. If it used C#, the comparisons would be more reliable and it would be easier to steal some users from Unity...
It's not new, it's been cooking for long time. But I'd be hesitant to trust this engine as the dev introduced breaking changes while taking the documentation offline for months. It's unacceptable and likely to happen again.
Is this a rebrand of Leadwerks? I doubt that this is 10x faster than using the proper tools to be more efficient in Godot or Unity, which will also provide better documentation, more support and a are free.
I dont think it would be hard to 10x unity performance. However this reminds me of leadwerks 2 which i tried once upon a time and couldnt get into. That being said, thanks for your videos on flax, I did start using it and its great. They just released version 1.7.2, might be worth covering.
I mean, you could say that about any engine. It all depends on how you do things. In Unity, you can use forward path, forward+, or deferred -- each will have different performance. You can use GPU instancing, static batching, dynamic batching, SRP batcher, or the new GPU Resident Drawers, or if you want to get some serious performance you can use Indirect Instancing or BatchRendererGroup, or even more performance if you do some custom runtime mesh combining, like I'm doing. Reduces draw calls down to virtually nothing, and causes culling overhead to be trivial. Also there will be a big performance difference depending on if you're using GameObject for object oriented programming or if you're using ECS with DOTS for data oriented programming. ECS is crazy fast. But here's an example of what you "can" do. This was made in Unity: 10 MILLION Characters On Screen?! - Ultimate Epic Battle Simulator 2 The Making Of Pt1 th-cam.com/video/kpojDPlIjdQ/w-d-xo.htmlsi=wDoF4k95VV2d6fNV&t=50 .
@@Churdington that's probably a fair point. Also, unity is way easier to get started with than most engines by the sheer amount of learning materials and ways for non-programmers to work with the engine.
Heya, i wanted to ask, i was following your every step (i am complete newb in any kind of game engines) and when you first time launch the scene from the visual studio for me it shows error message which is bound to an atlbase.h not being found. Am i doing something wrong or do i have to buy professional license for visual studio just to be able to render a scene or run an application i make in Ultra Engine?
@@gamefromscratch Thanks a lot man, i did one more google search and i found a guide on stackoverflow on how to add that component in the VS installation, you are a real champ.
If you go their web site and read it a bit you'll see that the engine is not developed for games in mind, it's for high performance simulations. Their customers are not us, governments and military.
[X] to doubt. I do like that it starts with a U, we could make that a meme... Looking at the gallery in their website, seems to be another engine that focus just on PBR graphics. It's a shame there doesn't sem to be an engine for anime/cartoon style games, everything has to be added in with shader scripts or some sort of shader math to recreate something that at the end should be extremely light on your system.
Given their history, I'd be skeptical. Leadwerks had some interesting features, like full FBX support (no other engine does this as far as I know) - but Leadwerks is not an engine you want to be using now, it's too old and obscure. This new project has a weird history of breaking changes and taking the entire documentation offline while rolling out "Ultra App Kit". I don't trust it.
I have my doubt. C# is object oriented and so is C++ mostly. Data oriented design should be faster with multi threading. Meaning Rust should beat C++ and C# always since it is data oriented by default.
@@techpriest4787 All of these languages are capable of data-oriented approach. C++ is basically C superset with support to other paradigms. You can use ECS with parallization the same way Unity DOTS is doing it and have great performance. Rust is safer and still fast, but I guess slightly (maybe 5-10%) slower in its idiomatic form.
First of all, nativly compiled C# isn't as fast as C++, because it's still has garbage collector, and in Unity for example it has null and array bounds checks by default unless you disable them with attributes for each class. But that's really not the point here. Ultra Engine has great rendering performance, but they do nothing special for user code and game logic. It's common GameObject-Component approach with OOP, so the speed won't be 10x times faster than other engines. It will be basically the same as Unreal, maybe 10-20% faster than Unity even if compiled to native code, but probably same or worse speed with Lua. DOTS will outperform all of them btw.
Are writing game engines that easy ? Because we have dozens of game engines right now, and new ones are produced every 5-6 per month. They are starting to resemble web frameworks.
I believe they should... make at least the standard version free, it kinda sucks not being able to test it to see if it's my cup of tea even though i've used Leadwerks before. Most paid engines like that don't fly off
Not sure how I missed this video, I usually follow this channel pretty closely. We have been looking for a good game engine for the past 2 years. And when looking up to see if they had a free trial for a small team I saw that this channel did a video on it answering my very question. Copper Cube, S2, and various other engines have already eroded my trust in "Pay for a game engine to evaluate it." for my Team of 25 . . .This is a HARD pass. Cross this one off the board.
It should be compared with Godot 3.5 on the GLES 2 render. And it seems to me that Godot will win in terms of quality and will be equal in terms of schedule
Yes. It is a flat rate, but there is a subscription option available if you buy from their store. $8 a month. www.ultraengine.com/community/store/category/1-software/?&_rid=25867
Very weird seeing stuff like window creation and the actual game loop in my own game source code. I would expect this to happen inside the engine code and out of my view.
I feel like they're fighting an uphill battle trying to sell an unproven engine with up front costs when you've got other engines that are more easily accessible, with some even being completely free/open source. At the very least, these guys need to prove that their engine is more powerful and easier to use than Godot, otherwise, what's the point? I feel like anyone who wants a real engine will lean towards Unreal, and if you want something simple/free, you'd go with Godot which has already been used for some indie projects. Anyway, best of luck to these guys, I hope they achieve the success they're looking for.
Although performance is still a thing for many games, in my mind using an engine is all about convenience and speed of development. Without C# support and a free tier to try it out, it's not a contender to replace Unity. It may be competition for Unreal, but that's hard to beat. It's seems to do a lot of things right, but also seems a little clunky. As the moments it looks to me as a very niche thing, but with a little polish it may change very fast. The devs are probably skilled enough to add C# support, but I doubt they have the customer focus to do that.
It competes with unity very well considering the fact that they have previously been introducing breaking changes and took the full documentation offline for months. This is almost as bad as the recent Unity scandals.
Probably. But then again, UE5 is a bit bloated because of all the features. Maybe having a stripped down but powerful engine has its advantages. Truth or bullshit, if I'm going to need to master c++, I might as well go for UE5 because it has a large community and this engine just came out.
I won't even look at a software that tries to deceive me before I've even installed it. All those metrics are useless when they are not in context. Every one of them can vary wildly depending on settings, engine features and resolution. Unreal can probably render 1000FPS too when you render Switch quality graphics like shown here.
Lighting/rendering increase? That's honestly not hard to achieve at all. Anyone who knows anything about writing a render pipeline can take a single look at legacy source/SRP and agree with me the devs who wrote that code should have been fired. SRP isn't nearly as bad aside from its bloaty, and to get the most out of it you gotta rewrite a lot. For legacy, you basically have to rewrite the whole thing.
Similar to Irrlicht "engine", looks like a framework:. Its fast, but indie devs won't use it and professionals use custom engine. Who will use it? However, I don't believe that in Lua this: 'while(true) { engine.Update() }' would be 10 times faster than Unity or Unreal Engine.
With all the free engine with an history, and also standard pro engine, this is gonna be a flop... I'm actually searching for the most open and free engine, who pay would probably go on unity or unreal, so where is their market ? Anyway thx for all your videos, they are very informative
I do think they can get away with selling the engine. There is a niche for (relatively) cheap, C++ and performance focused with tooling game engines that isn't really well filled. In some ways I would put the closest competitor to UltraEngine as something like OGRE, but even then it's not a really great apples to apples comparison. But asking people to pay first to find out, that I think is the biggest mistake here. As you said there are so many open and/or free game engines out there that this is a major mistake IMHO.
Nah... for indie developers if your tools are not free open source then you risk a lot. Wicked Engine, Bevy and Godot for engine, Blender for 3D, Krita for art and tons of other free open source tools.
Unity has no money and is carrying a heavy debt load from the Riccitiello years. They baught too many other companies on credit, when interest rates, were low. Expecting to cover it by selling scraped user data off UnityAds to advertisers. Then Apple put a blew a massive hole in their bottom line. And it only got worse from there.
Maybe I will stick with this, I havent found my game engine yet, I tried Unity first which sucks, then Gamemaker which is an engine for kids, now I'm with Unigine which is actually great but I struggle with performance problems, its slow and uses much GPU capacity (in edit mode!!!).
The devs of stuff like this do realize that UE5 is free right? Also that if u want it to run 10x faster than unity you just click the "Enable Nanite" button :)
This is one hell of a punt for the money. No guarantees they won't start charging for additional features, and it looks very much like an alpha product atm. Oh... And the frame rate looks terrible on your examples.
Very far from being production ready, easy to pick out one example and market insane speed ups by comparing the worst of an engine to the best simplified implementation in a custom environment…You can also leverage compute shaders and instancing to draw an insane amount of objects in Unity or Unreal (it’s also interesting they didn’t compare to Unreal). Looks very naive as an approach, not providing free tiers is imho intentional to get some cash to finance the whole thing. Not saying the whole thing is bad or anything, but I don’t see serious developpers jumping ship just based on the claims, so this marketing is clearly targetting people with less experience with gamedev, which feels wrong to me
I'm not surprised at the superior performance, the creator of Leadwerks (Josh) spent probably close to 20 years working with new and experimental rendering pipelines all the time. The problem with Leadwerks wasnt the graphics or performance, it was ease of use. If they can make it easy to use like Unity, then Unity and other such indie engines will be dead in no time.
@@MuhammadHosny0 Unity's only saving grace is ease of use. Graphically its not any better than other engines out there. Performance-wise its most definitely not any better than most. Ease of use and quick to prototype is its main draw. Its reliance on external assets to fill its gaps and subscription pricing model is its main downsides. Any other engine out there that could match Unity's ease of use would kill off Unity. Especially Unreal. If Unreal could revamp their toolkit to make it easier and more friendly, Unity would be dead.
@@Doofus171 C#, asset store, Plug and play services (multiplayer for example), IL2CPP, DOTS Burst HDRP URP You say the ease of use is the only thing it has going for it? 🥱 Did you actually work for companies using Unity and made that opinion based on what they're looking for or just chatting shit because you're a fanboy of some other engine? Yes unreal could take some Unity users if they revamped their toolkit and their eye burning editor workflow + their garbage C++ Now let me guess, you think Godot will also kill Unity if they get more C# support? 😱
@@Doofus171 Unity has probably the biggest community of game-developers, tons of free/paid assets, plugins and libraries. It's defacto a standard for mobile games, so no, it will be very hard to kill it. To kill it you should take it userbase, but in order to do that you should offer developers something more than they already have so they leave their current jobs and take time learning new technology.
What does it matter if your engine is fast? The engine can be as fast as it wants, but that doesnt mean anything if the rendering pmatform isnt great. Ill trust unreal engine anyday over these 3rd party engines
Links
gamefromscratch.com/ultra-engine-released-in-early-access/
Ultra Engine Store Affiliate Link:
www.ultraengine.com/community/store/category/1-software/?&_rid=25867
-----------------------------------------------------------------------------------------------------------
*Support* : www.patreon.com/gamefromscratch
*GameDev News* : gamefromscratch.com
*GameDev Tutorials* : devga.me
*Discord* : discord.com/invite/R7tUVbD
*Twitter* : twitter.com/gamefromscratch
-----------------------------------------------------------------------------------------------------------
hit game idea micro mini multiverse cells
a game about creating cells that
produce micro mini multiverses at high speed
semi popular youtube channels and an app
this could be a big profit game and is a super
interesting idea
Yay, one more U-Engine. Unreal, Unity, Unigine, Ultra...
dont forget UPBGE, and others lol
It’s why FLAX stands out ;)
They wish that one of those Sucker on the internet would download the wrong Engine
and just proceed to use it forever
@@owdoogamesthe devlogs are amazing
Honestly if the title doesn't start with a U, it doesn't deserve my time of day.
I gotta admit, I'm not a Unity fan these days but those statistics, without ostensible evidence, are pretty freakin' dubious... especially with a $90 price tag and no free version. I'm taking most of this with a grain of salt, especially given the number of zero-cost and low-cost engines which can already do most of what they're claiming.
Smells a bit scammy to me.
I spent $200 on the pro version of leadwerks 10 years ago and I can tell you it was a scam then and I wouldn't be surprised if it's still a scam now.
@@switted823 it probably still is...
Yep definitely fails the sniff test, I won't just be taking it with a grain of salt... I'll be taking a wide berth.
Poor decision making of segregating from free to Pro tier IMO; programming and languages is still skill based so even if C++ was in the free version, they still had to work to make something, they could had tried Pro version has VIP support for X months.
Scam, probably not. A cost effective product with all the features needed to bring a game product to market? Probably not. It being 10x faster by default because it isn't built on a decade of cruft and doesn't have a ton of design decisions built around usability and interoperability, etc sounds plausible enough. But Unity can also run fast enough for anything but projects too large for this either. Ultimately, this isn't going to be as easy to develop in as something fully featured like Unreal, CryEngine, etc. It's a ton cheaper, but ultimately you're making those kinds of big trade offs with any kind of 'small' game engine. I reckon it kind of suffers from the same issue stuff like 'small' video editing software has: It's too pricy for true hobbyists, but also not quite powerful enough for anyone dedicated enough to commit to $200+ software packages.
Finally a 10x faster engine, now I can finish my first game in 1 year instead of 10 years 🔥🔥
lol haha
If only that were what it really meant xD
LOL
SOunds like me, more than half the time I procrastinate and often search for tools(that does not infringes ©) to use in my workflow
Not providing a free version makes it really hard to test. I can also see that they provided the benchmarks as executables but that makes it really hard. Some questions in my mind:
- Why there is no Unreal Engine?
- Why use built-in instead of URP? (for use of SRP batcher.)
- Which version of each engine was used? Was Unity LTS or latest 2023 release?
Making something faster then Unity shouldn't be too hard if you just use C++/Rust and not dive into the cross-language function call territory. Also I would love to see some feature comparisons with Unreal/Unity.
yes, instead of press harder with c++, i can't understand why people just using a sh*tty alternate language especially
like python or java.
they're like a Coward trying to avoid being hit, so they use a cushion language,
and they often swear in defeatist like "we don't need performance that much"
those coward might as well use Php or VBA for their game engine
out of Unreal Engine, it was very hard to find a Game Engine that was Ballsy enough to use C++ as their main language,
@@jensenraylight8011 u ok buddy?, did someone hurt you?
@@MiriadCalibrumAstar it was just a social commentary of the current trend
@@jensenraylight8011 im aware of the shift.
But still it doesnt mean that by your comment sounds with vexed emotions, if it was satirical it isnt applied correctly.
They haven't managed their documentation well in the past. They introduced breaking changes and took the docs offline for quite a long time. I can't recommend going with this engine, in case the shitshow continues
The dev seems to be afraid of free version as many people can go and test out the "10X Unity performance" which is obviously 10X over exaggerated statement.
He has all the project files (unity and ultra engine) on his github. So you could pretty easily try it out yourself. If it is far off reality, just refund it.
But they are trustworthy. Never used the engine, but other tools from them. They exist since like two decades I guess. So they aren't bs'ing around. They just do things slightly different than everyone else. And I also think it's only one person doing this.
Unfortunately Leadwerks engine(predecessor) was pretty notorious, and not much expectations considering their past....
I hate to agree with this because I was a fan of Leadwerks for years (and even wrote small tools for the community) but it's true. Unsolved bugs on important features like character controller and navmesh and abandoning mobile support because it takes too much time to work on and not giving customers their money back (separate extra $100 purchases for Android and iPhone) are some of the issues. If you have the money and don't mind tinkering as a hobby, have fun. If you are serious about game development, use a serious engine.
Ultra Engine got its docs redacted while introducing breaking changes and at the same time the developer launched a "new" thing named "ultra app kit". I can't fathom why they thought this was a good idea and I won't trust the reworked Ultra Engine for a while.
Agreed. Leadworks was impressive enough as a small-team project but it left a lot to be desired just from a UI/UX standpoint. Most devs looking at tools like this are going to be indies. And above all else indies need effective workflows that allow them to develop as fast as possible. Selling this engine on performance alone (assuming it even is as good as it claims) is tough. Like, I can write an engine 10x faster than 'the next leading brand engine' too. Indeed I have. But it doesn't offer me any of the features that a major engine out there isn't already providing at 100x.
Not even source-available for the subscribers.
Dev: "Ultra Engine is specifically designed to yield the fastest possible performance for VR and 3D and provides order-of-magnitude faster performance over other engines. It uses many trade secrets I spent years researching and developing, and I am not giving the source code out."
Mhm i don't know that i'd get something from someone who is so full of themselves.
Leadwerks Pro costumer here who threw away $200. it was notoriously bad.
Three ways to become Unity competitor:
- Blue robot
- Undertale
- Name starts with U (bonus points for N after that)
Does mismanagement of documentation count? Because that's a significant part of the history for this project.
I'm talking unity levels of bad.
Instanced rendering is extremely efficient how the hell do you get 40 fps in unity using instanced rendering, meaning you're rendering millions of objects and somehow their game engine gets more than 1,000??
If they're not lying this is either a breakthrough in computer graphics or magic
If the numbers are real it'll be them using a worst-case scenario for one that the other is optimized for.
It's why the idea of a 'perfect' tool doesn't exist, when you remove incompetency you're just left with different priorities.
Or just bad testing. It's simple to miss some optimizations engines like unity/unreal get when You don't have experience of actively using them.
Simple: flawed tests. Might be unintentional, but I would be hesitant to trust devs that introduce breaking changes and take their documentation offline to launch a new thing in the middle of development.
I briefly skimmed over their paper on what they're doing and saw that benchmark and it seems that what they are really mostly measuring in that benchmark is frustrum culling on the CPU. They created a 3D grid of 30K+ instanced cubes that are all visible. With frustrum culling enabled, Unity hits 40FPS while Ultra Engine hits 1206FPS. The writer claims the majority of this increase is that occlusion culling is being ran at 45hz rather than every frame and their "advanced threading architecture" (which as far as I'm aware is pretty much the same as Unreal's where there's a GameThread, RenderThread, and Occlusion / RHI thread. Don't know how it compares to Unity though) and that they don't send data every-frame to the GPU from the CPU as that's a bottleneck in applications (I'm assuming since the objects in that test don't move they don't send any data so there's no bottleneck unlike Unity perhaps? Also they compress their data when sending it over to the GPU so it doesn't take up as much bandwidth vs full precision floats / doubles).
If they are getting that huge difference in fps then they aren't using instanced rendering in unity.
The performance is mindblowing and it has a lot of potential to be great. Gonna be tough to compete against Godot, UE, and Unity though. Their performance isn't great, but they have support for all devices, have good licensing, have plenty of documentation, and are easy to use.
That number might be from the stripped down pipeline with minimal code/pure C++ loop.
Not much expectations...
In 3D, Godot is easy to compete against, as if you actually use it for anything significant it is trash, lol. UE is hard to compete against though.
@lillybyte 3D is doable in godot, it is just not beginner friendly since it lacks the same helper functions unity and unreal. You have to have an extensive knowledge of shaders and math along prior game dev experience before tackling it.
@@rytifwhat do you mean? They've already implemented Vulkan renderer and it seems to be getting even better in upcoming versions. They already have C# support and that is pretty good too.
@@kloa4219 my dude, been using Godot for eight years, the 3D is absolutely trash, and worse the engine can't even load anything larger than a demo. You will be waiting for hours for it to load any significant game, if it loads at all. We had to move a 3D game to UE because it couldn't even load a quarter of the level we wanted.
This looks like something I would be very interested in but you are correct. Without a free tier I'll never try it. You gotta have a free tier if you want people to move over.
Even if it was free, I would not recommend it due to the documentation being handled so bad that it might as well be a clone of Unity. One does not simply introduce breaking changes and redact all the documentation.
@@TheExileFox I mean that's mainly my point is that unreal and unity are free and Much more established to think that they could come in and take any market share without a free option is unrealistic
Great review! I've been using Leadwerks for a few months now, and honestly really loving it. Not too fond of what I like to call "feature bloat" in certain other engines, and just want to get started coding and working on levels. In addition, because I used to mod for the original Half-Life way back when, the editor is very familiar to me. All in all, Leadwerks is kinda perfect for me in terms of my current projects, and Ultra Engine just seems like an upgraded version of that. In fact, I think I read that it should be able to load in Leadwerks projects, too. So I might migrate to Ultra Engine once it gets its full release.
Looks interesting. Did check a bit of the documentation and it looks very accessible and thought out. Looking forward to future improvements.
They have been mismanaging the documentation in the past by introducing breaking changes and then redacting the docs entirely leaving you with next to nothing for a while. I'd be very hesitant to try this engine given the history of the project.
If this is the offspring of Leadwerks that would explane why there is no free tier, the creator is all about the money first. Also Leadwerks started off life as being made from Blitzmax from Blitz3D and could explane where the clunkiness comes from in Ultra Engine. The 10x better than Unity is the creators huge ego talking and probably arent based in real world numbers.
Their marketing is super bad. If you go to their website it will show that you can get 1000+ FPS compared to Unity. On top of it they don't even have a free version. Maybe the engine is really fast and optimized but oh boy the marketing doesn't do any good for them. It really sounds like a scam LOL (I am not saying it is a scam just saying it sounds like one.)
Also they DESPERATELY need a free version or this is gonna be a big flop.
Looking to hear more from it.
Yeah I cover the grave sin of not having a free tier/version/trial in the video. It is their biggest mistake IMHO (and was with their previous engine as well).
How many products, games or otherwise, on Steam have a free trial? Just saying..
@@gamefromscratch now what is this "previous engine" ?
@@starby1243 he said it in like the first 2 minutes of the video
@@GaryChikebut unlike other products like games you get the fun playing it. But here you can have a free version where you cannot publish the game
I understand they want to capitalize on the recent Unity blunder, but if they wanted to brag about performance, Unity is not a good comparison. Nobody chooses Unity because it's fast, they choose Unity for its platform support and convenience (C#, big community, assets store).
Hell yeah, I've been wondering why there isnt something like this already. Im doing a masters in games technology and just did a project in opengl after working with unity and this is what I feel like I want, just a 3d editor, with asset management and animation tools. The coding of game mechanics, physics and architecture is something I want to do, I basically want to type more and click and drag less :D
Btw gamesfromscratch I want to say thank you. I've learned a lot from your videos - I'm doing really well in my course and that's partly cos of your channel thanks mate🙏
one thing that looks really intresting to me is the way the editor is layed out like the old hammer editor for valve games. with a 3d free fly view, a top down grid view, a front grid view, and a side grid view. could potentially be very usefull for retro styled games
They seem to be very inspired by Source and the like. Leadwerks was a lot like that too. Nice to see it making an appearance as other engines seldom use that style of level editing which is sad
I got that same impression lol. Totally missed the viewports being like Hammer's but I was reminded of their FGD definitions when Mike opened the properties sidepanel and then later on when he drew a brush on the viewport.
I've heard Hammer Editor being well regarded because of how simple and fast it is to prototype a level, but it's surprising how this style of editing is left up to be implemented through plugins or absent at all...
You can have those viewports in Unity or Unreal too.
@@Aidiakapi You can, but there is no 2D ease of use like in Hammer. In Hammer, the 2D viewports let you resize and drag brushes around like they're images in Photoshop. In Unreal and Unity, objects have a pivot point. Unity doesn't even have BSP brushes so whatever there, but Unreal's have pivots, but Source's don't, so Source's can be manipulated really easily, whereas Unreal's you find yourself pulling faces around only to find the pivot is way off and it's super annoying.
Just rendering 3D models very efficiently is nice, but it is lighting and reflections and all that jazz, where things tend to get more complicated and resource intensive.. It is forward renderer only, so I doubt that a scene with say 50 lights would render faster than the competition. They don't seem to have many lighting features yet, so I wish them good luck implementing those! I think a bigger game-scale scene as a demo is something they should strive for to actually test their performance..
Im checking out their news page right now and appararetly the background that lead to the claim is that inVR simulations you can not optimize the polygon count as engineers for example need to be able to inspect all the details.
Which makes sense, but honestly now im wondering if this shouldnt be called game engine then if the focus is so heavily set on simulations.
By forward renderer they probably mean forward+ / clustered, which scales similarly to deferred but without the problems of deferred
instead of the problems of deferred rendering you get new and different problems :^)
Love the idea of a code-first game engine, especially with C++. I could see a lot of Raylib users liking this engine. It would be interesting to see if systems could be easily disabled, like the physics engine if a game doesn't need physics, for example, so you could get even better performance. That level of easy customizability would be a huge selling point to me. Free/trial version is definitely a must IMHO. Also, gotta see if mobile and maybe consoles will be supported at some point. I'll definitely be watching this project closely.
Reminds me of back when I was using Dark Basic Pro or Dark GDK. Not the best to work with, due to it's limitations, but was simple enough to learn.
This to me, looks like a vastly better Dark GDK. And I love that it has VR support.
To me, this might have been worth looking into, if not for the advancements currently made in UE5.
And looks like there's no upgrade path. Say I got Lua engine first and decided to upgrade to C++ paying the difference I cannot do that. Need to buy Pro version and then I'll end up having the Lua-only version redundant. "Pro" version should have been a DLC rather than a separate product.
For an early access it's looking pretty good. People be creating crazy stuff like that, while i'm stuck writing python scripts...
My current project will take approximately 6,000 hours to complete in Unity. I am currently at 3,000 hours. Based on what we saw in this video, I estimate it would take me at least 10,000 hours to build the project with Ultra. I like the concept, I just don't have the time...
I bought and tested it today (pro version).
It is very nice for learning C++ game programming since the workflow / setup is very simple.
I can't think of a faster way for a game engine company to ruin it's reputation than to claim 1100 fps.
Saying you made a game engine faster than Unity is just saying you made a game engine. If they made a fully functional engine that was _slower_ than Unity then yeah that would be impressive. Rip whatever the name of this engine was
That usage of 'and' in the C++ code on line 33 at 6:30 is the most cursed thing I've ever seen 😮
The alternative forms (and, or, not) are part of the C++ standard and are used in countries where there is no easy keyboard access to symbols like & or |, which make sense in an US keyboard but not so much in some international layouts. You used to have to #include "iso646.h" to use them.
#define and &&
jep, very cursed indeed :))
The creator is a python addict
12:50 I can see that it performs smoothly and without hitches...at stunning 3fps.
the editor felt like -10x faster
Pretty excited about this engine, each copy doesn't cost that much and the royalty licensing is more than fair. But like any piece of software, it's only as good as its documentation.
"only as good as it's documentation"
nail in the coffin already perhaps.
Because they have previously been introducing breaking changes while taking the entire documentation offline for months. And in the middle of this, they launched "Ultra App Kit". It's like an insult to injury.
How about target-build platforms it currently supports? Only x86/x64 Pc? How about consoles? Or mobile devices?
Yeah, 100% an omission in the video, my bad.
@@gamefromscratch np. Great video, anyways. It'll be a very interesting product to follow and see whether they manage to improve the UX when using the editor. Thx
this engine looks really promising, and not really bc of the performance aspect, if its even remotely as fast as it advertises itself, I am more than fine with it, even if its paid.
bc the biggest problem I had with both unity and godot, is the lack of *in-engine* c++ support.
because I really really like c++, so when I do get an actual job, this might one of the first things I purchase of off steam.
just out of respect.
(I really really like c++ and devling into the godot source code isnt hard at all, its just a little more fiddely than I'd like. (but thats expected, it is the source code after all))
Looks like you'd be happier using a framework instead on an engine, if you like C++ so much. SDL, SFML, Raylib or even OpenGL/Vulkan so you can do anything you want. Engines imho are needed to minimize the coding.
that's where i'd tell you to look at flax.
This is the great thing about the claims of 10x performance. You've already bought the software to find out the truth. How many are going to bother going through any return process, if there is one.
Yeah but 10x more performance than what?
In Unity, you can use forward path, forward+, or deferred -- each will have different performance. You can use GPU instancing, static batching, dynamic batching, SRP batcher, or the new GPU Resident Drawers, or if you want to get some serious performance you can use Indirect Instancing or BatchRendererGroup.
Also there will be a big performance difference depending on if you're using GameObject for object oriented programming or if you're using ECS with DOTS for data oriented programming. ECS is crazy fast.
And I'm just barely touching the tip of the ice berg here.
I am so sorry the developer decided to focus on VR/Windows only and make a brand new VR focused engine from scratch instead of working on Leadwerks but sadly it makes sense from a business perspective. As Leadwerks being super simple to use and no royalties, many US Federal agencies including NASA and others decided to use Leadwerks for their VR simulations and they did hire the developer to work on their VR staff even. This is where the idea of this new VR engine come from. I wish Leadwerks was good enough for them.
I really liked Leadwerks, the only issues I had were with the Linux support, just hoping to have good support on UE soon
One year ago, buddy! Not sold on it yet.... where are the Drones Flying Today?
My indie team and I moved from Unity to Unreal Engine 5.3 about a month ago, and I m SO satisfied with that decision. UE5 is so good it's literally "UNREAL". My only regret is that I spent over a decade on Unity before trading up. Not sure what Ultra hopes to gain by introducing another money-up-front game engine into the market when there's so many top-tier, free to use options already. Seems a bit suss.
I would be interested to see how does it fare against C++ based engines. (UE5, WickedEngine, Ogre etc).
But seeing this vid, I'm not convinced. Doesn't it do basically same or less things than wicked engine?
Wicked engine doesn't really have the tooling side, but WE + A level editor would be very much a competitor to this. WE has partial tooling of course.
I also think from an API perspective this might be more approachable
@@gamefromscratch True, but wicked engine is multiplatform and free. Which are also important aspects
The engine itself is pretty interesting. I'd have to wait to use it until their editor is in more of a finished state.
This engine already has a troubled history. Tldr: the predecessor Leadwerks is not friendly to use. Later the dev introduces breaking changes and takes the entire documentation offline for months.
It's weird that they went for Unity's throat by using another language. If it used C#, the comparisons would be more reliable and it would be easier to steal some users from Unity...
Actually would be more difficult because Unity has IL2CPP burst compilation, DOTS (ew) and much more than just simple C# scripting
@@MuhammadHosny0 Iam now doing stuff in DOTS and with 1.2 it is kinda working finally
Can we compile to other platforms other than windows?
How is it that a brand new game engine in early access has documentation that's lightyears ahead of Unreal
It's not new, it's been cooking for long time. But I'd be hesitant to trust this engine as the dev introduced breaking changes while taking the documentation offline for months. It's unacceptable and likely to happen again.
Is this a rebrand of Leadwerks? I doubt that this is 10x faster than using the proper tools to be more efficient in Godot or Unity, which will also provide better documentation, more support and a are free.
I am waiting for someone to make the giga engine
10x less features too
Easily.
For many though, that isn't necessarily a bad thing.
@@gamefromscratch Yeah, I don't care for additional features if I can program them myself when I need them keeping the code clean.
@@MikAlexanderstart moving your game to Ultra Engine then
wow! big, big claims!! thanks 4 showing!
I dont think it would be hard to 10x unity performance. However this reminds me of leadwerks 2 which i tried once upon a time and couldnt get into. That being said, thanks for your videos on flax, I did start using it and its great. They just released version 1.7.2, might be worth covering.
Thanks. I'll take a look. I like Flax too.
I mean, you could say that about any engine. It all depends on how you do things.
In Unity, you can use forward path, forward+, or deferred -- each will have different performance. You can use GPU instancing, static batching, dynamic batching, SRP batcher, or the new GPU Resident Drawers, or if you want to get some serious performance you can use Indirect Instancing or BatchRendererGroup, or even more performance if you do some custom runtime mesh combining, like I'm doing. Reduces draw calls down to virtually nothing, and causes culling overhead to be trivial.
Also there will be a big performance difference depending on if you're using GameObject for object oriented programming or if you're using ECS with DOTS for data oriented programming. ECS is crazy fast.
But here's an example of what you "can" do. This was made in Unity:
10 MILLION Characters On Screen?! - Ultimate Epic Battle Simulator 2 The Making Of Pt1
th-cam.com/video/kpojDPlIjdQ/w-d-xo.htmlsi=wDoF4k95VV2d6fNV&t=50
.
@@Churdington that's probably a fair point. Also, unity is way easier to get started with than most engines by the sheer amount of learning materials and ways for non-programmers to work with the engine.
Heya, i wanted to ask, i was following your every step (i am complete newb in any kind of game engines) and when you first time launch the scene from the visual studio for me it shows error message which is bound to an atlbase.h not being found. Am i doing something wrong or do i have to buy professional license for visual studio just to be able to render a scene or run an application i make in Ultra Engine?
It sounds like your visual studio install is missing a required feature
@@gamefromscratch Thanks a lot man, i did one more google search and i found a guide on stackoverflow on how to add that component in the VS installation, you are a real champ.
Does it support edit and continue?
If you go their web site and read it a bit you'll see that the engine is not developed for games in mind, it's for high performance simulations.
Their customers are not us, governments and military.
[X] to doubt. I do like that it starts with a U, we could make that a meme...
Looking at the gallery in their website, seems to be another engine that focus just on PBR graphics. It's a shame there doesn't sem to be an engine for anime/cartoon style games, everything has to be added in with shader scripts or some sort of shader math to recreate something that at the end should be extremely light on your system.
Interesting. This will definitely be one to follow
Given their history, I'd be skeptical.
Leadwerks had some interesting features, like full FBX support (no other engine does this as far as I know) - but Leadwerks is not an engine you want to be using now, it's too old and obscure. This new project has a weird history of breaking changes and taking the entire documentation offline while rolling out "Ultra App Kit". I don't trust it.
Are there any open sourced game engines that teach you how to use them ?
now officially u can rename the channel to ADSfromScratch since are no more tutorials ??
I love code first engines, that is why I like Stride so much.
..but now dotnet 8 can be compiled to native binary that runs almost as fast as real c++ program. That 10x faster than Unity won't stand for long.
I have my doubt. C# is object oriented and so is C++ mostly. Data oriented design should be faster with multi threading. Meaning Rust should beat C++ and C# always since it is data oriented by default.
Unity already compiles to native C++ since 2017 using IL2CPP
@@techpriest4787 All of these languages are capable of data-oriented approach. C++ is basically C superset with support to other paradigms. You can use ECS with parallization the same way Unity DOTS is doing it and have great performance. Rust is safer and still fast, but I guess slightly (maybe 5-10%) slower in its idiomatic form.
First of all, nativly compiled C# isn't as fast as C++, because it's still has garbage collector, and in Unity for example it has null and array bounds checks by default unless you disable them with attributes for each class.
But that's really not the point here.
Ultra Engine has great rendering performance, but they do nothing special for user code and game logic. It's common GameObject-Component approach with OOP, so the speed won't be 10x times faster than other engines.
It will be basically the same as Unreal, maybe 10-20% faster than Unity even if compiled to native code, but probably same or worse speed with Lua. DOTS will outperform all of them btw.
Advantage of buying it from them instead of Steam: the ability to launch the engine without launching steam
I guess im the only one wanting to make or use a engine more akin to source,idtech,etc.
For one specific game.
what's the performance hit using lua?
this guy never gets tired
I can't see why you'd use this instead of Unreal since you have to use C++ ?
Are writing game engines that easy ? Because we have dozens of game engines right now, and new ones are produced every 5-6 per month. They are starting to resemble web frameworks.
Writing an engine of that level isn't easy but it's still kinda easy to some people
I believe they should... make at least the standard version free, it kinda sucks not being able to test it to see if it's my cup of tea even though i've used Leadwerks before. Most paid engines like that don't fly off
problem with the marketing is nobody is using unity for performance in the first place.
I dont understand how they think they will compete if the charge 120$ unless they get big studios buying it
If you ever find yourself excited by lofty performance claims, just remember: the less there is, the better it performs. Something is sacrificed.
i think that fps comprassions is a lie. Where is screenshots? in what situation in 2fps ? never saw this.
Not sure how I missed this video, I usually follow this channel pretty closely. We have been looking for a good game engine for the past 2 years. And when looking up to see if they had a free trial for a small team I saw that this channel did a video on it answering my very question. Copper Cube, S2, and various other engines have already eroded my trust in "Pay for a game engine to evaluate it." for my Team of 25 . . .This is a HARD pass. Cross this one off the board.
It should be compared with Godot 3.5 on the GLES 2 render. And it seems to me that Godot will win in terms of quality and will be equal in terms of schedule
Is the price a subscription or flat rate?
Yes.
It is a flat rate, but there is a subscription option available if you buy from their store. $8 a month.
www.ultraengine.com/community/store/category/1-software/?&_rid=25867
It looks like a one off payment, I see nothing to indicate a subscription.
This might be good for better handling of multiplayer server side rewind. 🤔
Very weird seeing stuff like window creation and the actual game loop in my own game source code. I would expect this to happen inside the engine code and out of my view.
I feel like they're fighting an uphill battle trying to sell an unproven engine with up front costs when you've got other engines that are more easily accessible, with some even being completely free/open source. At the very least, these guys need to prove that their engine is more powerful and easier to use than Godot, otherwise, what's the point? I feel like anyone who wants a real engine will lean towards Unreal, and if you want something simple/free, you'd go with Godot which has already been used for some indie projects. Anyway, best of luck to these guys, I hope they achieve the success they're looking for.
Don't make the video title a question if you're gonna spend 16 minutes wasting my time not answering it.
ULTRAKILL run Unity, now we have ULTRAENGINE
Although performance is still a thing for many games, in my mind using an engine is all about convenience and speed of development. Without C# support and a free tier to try it out, it's not a contender to replace Unity. It may be competition for Unreal, but that's hard to beat. It's seems to do a lot of things right, but also seems a little clunky. As the moments it looks to me as a very niche thing, but with a little polish it may change very fast. The devs are probably skilled enough to add C# support, but I doubt they have the customer focus to do that.
I was reading the news page and some comments there and apparently C# suppoert was originally intended. Weird that it isnt the case anymore.
It competes with unity very well considering the fact that they have previously been introducing breaking changes and took the full documentation offline for months. This is almost as bad as the recent Unity scandals.
Do I hear godot 4.3 snapshot is out? And it has D3D ?
Do I hear crickets?
Yes, Godot 4.3 have DirectX 12
No way, they lying. That pretty much means its better than UE5.
I hope they will.
Probably. But then again, UE5 is a bit bloated because of all the features. Maybe having a stripped down but powerful engine has its advantages.
Truth or bullshit, if I'm going to need to master c++, I might as well go for UE5 because it has a large community and this engine just came out.
Not hard to be faster than UE5 either 😂 Lumen is a hog
Minimal loop with a bare skeleton pipeline can beat almost every engine by "number".
Anyway 'a game development appropriate' is a different thing.
"better"
I won't even look at a software that tries to deceive me before I've even installed it. All those metrics are useless when they are not in context. Every one of them can vary wildly depending on settings, engine features and resolution. Unreal can probably render 1000FPS too when you render Switch quality graphics like shown here.
anything will be faster than unity
Why are so many game engines U.E... Unreal Engine, Unity Engine, Unigen Engine, Ultra Engine
Very interested in the code first approach
in my opinion this rather competes with raylib than unity, ue, etc.
Lighting/rendering increase? That's honestly not hard to achieve at all.
Anyone who knows anything about writing a render pipeline can take a single look at legacy source/SRP and agree with me the devs who wrote that code should have been fired.
SRP isn't nearly as bad aside from its bloaty, and to get the most out of it you gotta rewrite a lot.
For legacy, you basically have to rewrite the whole thing.
Similar to Irrlicht "engine", looks like a framework:. Its fast, but indie devs won't use it and professionals use custom engine. Who will use it? However, I don't believe that in Lua this: 'while(true) { engine.Update() }' would be 10 times faster than Unity or Unreal Engine.
With all the free engine with an history, and also standard pro engine, this is gonna be a flop...
I'm actually searching for the most open and free engine, who pay would probably go on unity or unreal, so where is their market ?
Anyway thx for all your videos, they are very informative
I do think they can get away with selling the engine. There is a niche for (relatively) cheap, C++ and performance focused with tooling game engines that isn't really well filled. In some ways I would put the closest competitor to UltraEngine as something like OGRE, but even then it's not a really great apples to apples comparison.
But asking people to pay first to find out, that I think is the biggest mistake here. As you said there are so many open and/or free game engines out there that this is a major mistake IMHO.
You do have Steam's 2-hour return policy but that's probably not enough to really test the engine
Flax is probably the best option after Godot.
This is clearly a marketing lie
Even if that is the case, you are comparing a tank to a children's bike
Nah... for indie developers if your tools are not free open source then you risk a lot.
Wicked Engine, Bevy and Godot for engine, Blender for 3D, Krita for art and tons of other free open source tools.
Sounds a bit as a unity clone, I think something oriented to non-coder artists, similar to scirra construct but in 3d would be a good idea.
So that’s why they never finished Leadwerks.
They only really support nVidia GPUs currently.
Why dont Unity purchase it and include ? :)
Unity has no money and is carrying a heavy debt load from the Riccitiello years. They baught too many other companies on credit, when interest rates, were low. Expecting to cover it by selling scraped user data off UnityAds to advertisers. Then Apple put a blew a massive hole in their bottom line. And it only got worse from there.
Maybe I will stick with this, I havent found my game engine yet, I tried Unity first which sucks, then Gamemaker which is an engine for kids, now I'm with Unigine which is actually great but I struggle with performance problems, its slow and uses much GPU capacity (in edit mode!!!).
The devs of stuff like this do realize that UE5 is free right? Also that if u want it to run 10x faster than unity you just click the "Enable Nanite" button :)
Kinda suspicious when a product advertised as better version it's competitor
This is one hell of a punt for the money. No guarantees they won't start charging for additional features, and it looks very much like an alpha product atm.
Oh... And the frame rate looks terrible on your examples.
Very far from being production ready, easy to pick out one example and market insane speed ups by comparing the worst of an engine to the best simplified implementation in a custom environment…You can also leverage compute shaders and instancing to draw an insane amount of objects in Unity or Unreal (it’s also interesting they didn’t compare to Unreal). Looks very naive as an approach, not providing free tiers is imho intentional to get some cash to finance the whole thing. Not saying the whole thing is bad or anything, but I don’t see serious developpers jumping ship just based on the claims, so this marketing is clearly targetting people with less experience with gamedev, which feels wrong to me
Nah gotta wait for unity to buy this engine to improve the performance ngl
I'm not surprised at the superior performance, the creator of Leadwerks (Josh) spent probably close to 20 years working with new and experimental rendering pipelines all the time. The problem with Leadwerks wasnt the graphics or performance, it was ease of use. If they can make it easy to use like Unity, then Unity and other such indie engines will be dead in no time.
Yes Unity the indie engine will be dead in no time 😂
@@MuhammadHosny0 Unity's only saving grace is ease of use. Graphically its not any better than other engines out there. Performance-wise its most definitely not any better than most. Ease of use and quick to prototype is its main draw. Its reliance on external assets to fill its gaps and subscription pricing model is its main downsides. Any other engine out there that could match Unity's ease of use would kill off Unity. Especially Unreal. If Unreal could revamp their toolkit to make it easier and more friendly, Unity would be dead.
@@Doofus171 C#, asset store, Plug and play services (multiplayer for example), IL2CPP, DOTS Burst HDRP URP
You say the ease of use is the only thing it has going for it? 🥱
Did you actually work for companies using Unity and made that opinion based on what they're looking for or just chatting shit because you're a fanboy of some other engine?
Yes unreal could take some Unity users if they revamped their toolkit and their eye burning editor workflow + their garbage C++
Now let me guess, you think Godot will also kill Unity if they get more C# support? 😱
@@Doofus171 Unity has probably the biggest community of game-developers, tons of free/paid assets, plugins and libraries. It's defacto a standard for mobile games, so no, it will be very hard to kill it. To kill it you should take it userbase, but in order to do that you should offer developers something more than they already have so they leave their current jobs and take time learning new technology.
What does it matter if your engine is fast? The engine can be as fast as it wants, but that doesnt mean anything if the rendering pmatform isnt great. Ill trust unreal engine anyday over these 3rd party engines