doctorclot

Free Play Account
  • Content count

    17
  • Joined

  • Last visited

Posts posted by doctorclot


  1. That does sound really great, so please don't take this next bit the wrong way, but are you creating it with the intention of working alongside CRS or as a standalone project that will, I can only imagine, result in being a competitor?

    I ask purely because if you intend to present this to CRS as a proposal to port the existing game over to the suggested Unreal engine, then CRS should probably be made aware of the existence of your idea incase they turn out to be entirely uninterested in it, resulting in your time being wasted.

    On the other hand, if you're aiming to release basically a new version of WW2 Online as a competitor, you probably shouldn't be talking about it on the rats forum as it would give a bad impression to your new potential customer base :P

    Like I mentioned above, I sincerely hope you don't take offense by this since i'm genuinely interested in the project, but I believe, given CRS's history with accepting ideas openly from the forums etc, that it needs to be excecuted and presented to them in a way that will do anything but pee them off or feel threatened by a potentially superior (modern) product :)

    Good luck regardless though from me, I look forward to reading your next progress report on this! S! :)

    No offense taken. :)

    Here is what I am planning on doing. Hopefully this will shed some light on this.

    1. Make a prototype and "shop" it around to various smaller studios and publishers including CRS, Paradox Interactive, etc.

    2. If Plan 1 fails - I will attempt to crowdfund the project so I can develop it further with others help.

    3. If plan 2 fails - Open Source everything thus far (Assets, Game Code, Levels, etc.) so people can learn from it as well as help develop it further if they so desire.

    Hope that I cleared things up. I am currently re-writing the Game design Document. Which will be part of Update 1.

    - DoctorClot


  2. Ok, this is the third time for me typing up this message. So I am going to keep it short.

    That said - Here is what I have been up to.

    1. I have evaluated Unreal 4. It will work for what I have planned for the massively large world. However some compromises will have to be made in the name of performance of lower end PCs.

    * I have been modeling buildings. The buildings are untextured right now. No vehicles or characters yet due to my work load.

    * I am going to be using this network stack for the game.

    * I need to find a network programmer to integrate Game Machine into the engine.

    2. I have been working on some freelance projects. I am still under NDA but the money I have gotten should allow me to develop the game for a few months more. Some of the contracts still need to be finished up. So If I go silent that is why.

    3. I am overhauling the game design document for this game with what I have learned with these other projects as well as some personal stuff. I do not want to loose interest in this project. So I am going to customize it a bit. It will also help out performance.

    4. I want to make this game really deep in terms of game mechanics and gameplay. However - Some compromises will need to be made. I typically do not like Compromises however I like these. I am going to run these by you guys as soon as I have overhauled the Game design document.

    5. Here is the middle ware that I am planning on using for the weather and clouds. LINK

    I should have something to show off hopefully very soon. :)


  3. So I thought I would give an update on this.

    Currently I am putting together a series of art packs to fund development of this project.* I am planning to sell these art packs via the Unity Asset store and eventually the UE4 Marketplace.

    My Development computer is still out of commission but that is OK - I recently got a hold of a less powerful Intel computer. Just needs a heat sink. :)

    As Paper work / Planning goes - I recently put together a "Open Source" Game Design Document Template**. I will be putting an actual game design document together soon. :)

    Question time - What does everyone think of the Oculus Rift? I would like to integrate this into the game. For the Air Combat, Tank Combat and Ship Combat portion I want Oculus Rift integration. I am not entirely sure about Infantry yet. For those of you not in the know go

    .

    *Costs will be going towards software licenses, upfront Hardware costs and eventually mission critical people needed to get certain aspects in the game running. (Networking, Damage model, Database, etc.)

    ** Link to the Template if you are interested :)


  4. FMOD, now it starts to make sense. Still miss a few things, like convoluted DSP, but that will possibly one day be made as plugin for the mixer.

    I saw native support for Unity and plugin support for several other engines, but no mention of Unreal?

    Epic has an Integrated partners program which I think Firelight technologies (Makers of FMOD) are a part of. However I am now sure 100% about that.

    I might be able to get a programmer when my mainboard / PSU gets back in from RMA to make a plugin for FMOD. :)


  5. I run Linux Mint 14. I've been running WWIIOL off and on via WINE since early 2008. Over time, the support has been better, then worse again, back and forth. I even wrote a few patches in WINE to fix support over time, and documented process for getting WWIIOL working in the forums here. I once even rolled a .deb that contained a self-contained WINE install, WWIIOL, and a script to make them run well together on Linux.

    So yeah, you could say it's important to me.

    System Specs:

    Phenom X4 955

    Geforce 560Ti

    8gb ram

    I am not sure if Linux Mint will be supported by Unreal 4. The Only version of Linux that is really talked about is SteamOS - a modified version of Debian (Wheezy).

    I will ask on the unreal forums what other versions of Linux are going to be supported if any and report back here.


  6. I am really interested to see if this pans out.

    My guess is that this all could be done in a modern engine, whether UE4 or another. However, given that the Rats don't have the resources or funds to put towards a full-scale re-development, someone would have to come in the door with a very good business proposal to make something like this happen.

    FWIW, I would really push to include a Linux release as part of any port to UE4. Especially since the engine supports it out of the box. For me, Linux support (even via WINE) is the difference between me paying to sub, and wishing I was able to.

    What distro of linux do you run?

    Ubuntu, Debian, SteamOS, Other?

    I can get an environment port of the game and send it to you when ready for testing of course. :)

    To make sure that it runs. I will agree that supporting Linux is a big one at-least for me. It is a market that is literally untapped.

    Shoot me a PM with you contact details and we will work something out. It is not often I come across a Linux user. I want to make the experience for Linux users as smooth as possible :)

    - Ben


  7. I could do the audio engine design and knock the socks off the rest of the industry. Audio is in general the most under-prioritized aspect of the gaming experience today' date=' with huge gains in immersion and control of player behaviour waiting to be harvested. Just read through the entire Unreal Engine 4 audio specifications and was a bit underwhelmed, but the basics are there and would be enough to catapult this game's audio experience forward.[/quote']

    What do you have in mind? I am thinking of Using FMOD as a plugin to aid in your quest to make the game stand out audio wise. :)

    FMOD Studio Link

    Have you put your thoughts in a email to Jim Mesteller doctorclot and if so did you get a reply ?

    At this moment in time no I have not. I would rather have something to show than nothing at all. That and I want to be sure that Unreal 4 can pull this off 100% before going to him. I have PMed XOOM but no reply yet.

    My only concern at this point is map of Europe and the shear amount of objects that are placed around the maps (Trees, Buildings, etc.) not to mention player spawned objects (Aircraft, Tanks, Infantry, etc.) would overwhelm the render but as stated prior there are ways around this.

    EDIT: On a note about the actual in-game map I have managed to finally finish downloading the necessary DEM data to start work. My Dev Computer is out for repairs and I should be getting back by next week. I should however be able to convert the DEM data to a necessary height map format stitch them together and see how performance is and go from there optimizing it.

    But progress is being made. :)


  8. If you put a very concrete proposal in to CRS, it could be feasible that they would be willing to cooperate. I have no idea if they will or not, but it could be that they don't have a lot to lose with RA. But you'd need to have a super-convincing and very concrete plan with very concrete people and reasonable time plan. And one of the first things you need to really figure out is what would prevent you from taking their code and running away with it. Figure out what everything CRS could lose by cooperating first and make sure any negatives are mitigated.

    Quite frankly, I don't see it as realistic, but I don't want to discourage you too much.

    I guess I should have been more specific with the Rapid Assault to WWII online bit. I am thinking about making the game in modules. The game would start off as small scale infantry combat. With each module (major iteration) the game world would get bigger in terms of playable area and would get allot more features. :)

    For example -

    Infantry Module - Focuses on the initial net code, FOB's, and Infantry combat.

    Tank Module - Focuses on Tank Combat, Initial Armor damage model, rendering larger environments, and Improving the net code.

    Air Module - Focuses on Air Combat, Performance Improvements client side, adding more vehicles to the database, and a bigger environment. :)

    etc.

    But Yeah - I see where you are coming from Tigger6

    No Time Table = No Deal

    No Solid Plan = No Deal

    Right now it is just talk. Guess I should start putting my plan into action :)


  9. The main point is that someone has to do it' date=' and there is no someone. Ask someone that has a game-related management position to do some napkin calculations on how much it would cost..[/quote']

    So - I am looking for more experience to get into game development as a full time job. I have done some freelance work but mostly Pro Bono for friends.

    I know people in the same position who want to do this full time.

    Here is what I am thinking -

    CRS has Rapid assault - Why not use that as a test case to see if the full World War II online game could at least run on Unreal 4? I am NOT saying turn WWII online into Rapid Assault but turn Rapid Assault into World War II Online. Start out basic and then get more and more advanced.

    Start off with some smaller (Omaha Beach for Example) environments with larger player counts then slowly integrate the more advanced stuff that I listed before?

    Now here comes the interesting bits - I know allot of people from over the years who are experienced with Unreal and other engines. With Unreal 4 Coming out they want to get on a project - Now keep in mind that these are experienced individuals. Tools programmers, Network Programmers, Graphics Artists, Sound Engineers, etc. But not experienced enough to be hired by an AAA Studio like Valve or Bethesda Softworks.

    Basically they need more experience to be hired by the companies they want to work for.

    With that said - We make money off the launch of the game (Which will be negotiated with CRS). So basically royalty.

    If needs be we could go to kickstarter and raise funds for this undertaking for hardware costs and software licenses.

    But if there is a will there is a way. :)

    - DoctorClot


  10. OK. I suppose that you have a clue about programming.

    Wwiionline, unlike majority of multiplayer games, has many (more than 128) players and very large terrain. In principle, you might need to track a 1000 objects on the battlefield- players, bombs, mortar rounds, tank shells, hulks, dead bodies, etc. In addition, there are large network latencies. If you are playing a shoebox shooter, it's common for the latency of all players to be <150ms, wwiiol has a single server that people connect to, so 500ms latency is not that rare. That means that you need a completely different network stack and back-end, as you need to prioritize different information on the fly. You are fine with receiving updates about airplanes 100 miles away once every 10 seconds, but the infantry that is running towards your bunker should be updated way more often. Now, you will say that that is no problem, since it's the front-end that you want to switch. But it seems likely (I haven't seen the code but I have observed the symptoms for a long time), that the code that does this is a huge ball of 12 year old hacks and spaghetti code that spreads between the server and the client. So by changing a large part of the front-end, you need to rebuild the backend as well. Note that doing this properly is actually a difficult problem. You need someone that can design efficient algorithms and understands nearest neighbor lookups.

    Now, front-end wise. I expect (although I don't know for sure), that unreal engine won't be happy if you load quarter a million square kilometers of terrain into it. You need a way to prioritize loading so that a infantry player doesn't need to load a cell 10km away, but a fighter pilot does, but doesn't load every trench until he is reasonably close. A diving airplane travels at 500mph, possibly skimming over ground. The last thing you want is loading stutters while you do that.

    And here comes the actual problem of converting the whole world into data that unreal engine can understand. Clasically, models have independent model (various LODs) and collision model and often a very simple damage model. WWIIOnline models use some different strange format that binds everything together. There are presumably no tools that facilitate easy generation of models and easy conversion to usual 3d data.

    That brings us to a damage model, that is built-in and needs some extractor, but mostly, it is completely different to anything that any other engine supports! You don't deal damage to components, you calculate actual penetration probability if you have armoured plate in the way. This is something that you do on the front-end, and this would thus need completely new code. Worst of all, there is nobody that understands how the old code works, so good luck writing the new one.

    And that is just something that I, who has never seen the code, can come up with in 5 minutes. Reality is, as usual, more complex than that.

    Alright so - I am going to try and make sense of this as much as possible using links to explain things primarily for my own sake. :)

    Lets start with the continent as that is what the players will be fighting over.

    World Space -

    So the unreal engine 4 uses a thing called Floating Origin which was first used in the game dungeon siege by gas powered games. You can find the details there but what it is basically - the world has no absolute center of origin at-least for its landscape system. The unreal engine render can render up to 500k this is both in the X,Y, and Z directions (possibly more?) without it getting choppy, glitchy or things breaking apart. This is not a hard limit though if you want to go past the 500k (or whatever) mark you can just tick a box in-editor and you can go on wards.

    As for the tracking objects on the battlefield - Render the "Visual Stuff" on the client side and do the actual calculations on the server side. (I will talk more about this at the end.) To prevent Cheating make the server authoritative. Meaning the client connects to the server and the server makes the "decisions" so to speak. This will cut down on cheating and crap like that.

    Network Stack / Networking -

    Now network stack - I am not 100% positive but Raknet provides a pretty solid network framework / stack for online games MMOs ( Planetside 2 and Everquest Next) and the like. Planetside 2 is having at-least 1000+ (Give or take) Players a server. They Modified it quite heavily I assume but Raknet can handle it. :)

    Skipping on the Vehicle component damage - Some sacrifices may need to be made in this area workflow wise. You will see what I mean in a moment.

    Database / Component Damage -

    So there will be a giant database of vehicles - each vehicle entry will have a series of penetration values, Gun Velocities, etc. under each database entry. Each of these data base entries will refer to a mesh component inside or on the vehicle. This could in theory act as both the collision mesh for a object and the damage component mesh for the object with a bit of tweaking to the engine.

    As for getting that working on the network with an authoritative model here is some (very bad) pseudo code.

    If it does penetrate and produces spall - Example

    If turret hit -> Send to Server -> Retrieve turret damage table -> Send to Client -> Penetrate? -> If Yes send to Client -> Get angle, Velocity, slope -> Send to server -> Bounce? -> No - Send to client -> Bounce Shell? No -> Spall? - Send to server -> Yes -> send to client -> Decapitate driver -> Send to server - Dead? -> Yes? - Send to Client -> You are immobilized.

    If it does not penetrate but bounces and produces spall - Example

    If turret hit -> Send to Server -> Retrieve turret damage table -> Send to Client -> Penetrate? -> If No send to Client -> Get angle, Velocity, slope -> Send to server -> Bounce? -> Yes - Send to client -> Bounce Shell? No -> Spall? - Send to server -> Yes -> send to client -> Kill Commander, Driver and Ingnite ammo rack -> Send to server - Dead? -> Yes? - Send to Client -> You are dead.

    It may not be the most efficient way but it is a way. Obviously you would not want me doing the server code and there might be a better way of doing this.

    Prioritization of objects in the Scene -

    You do not need to render a spitfire that is 5 miles up in the air if you want to use a "render bubble" like maxios said and make it rather large and ignore things such as Occlusion culling volumes and objects within them to optimize your scene.

    As for mortars, bombs, and other explosive projectiles. You could just cull them out as needed. For example Someone fires a rocket from a spitfire or hurricane and banks to the left. You could in use "vision arcs" to determine the field of view as well as determine what is being culled for the players camera. Unless the player is looking at that rocket 100% of the time then you will see the rocket and its final explosion.

    You could just use raycasts for bullets and stuff.

    But yeah... If I missed something let me know. :)

    Your concerns are very valid and I can see why.

    Doctorclot

    EDIT: TLDR - A good network stack like raknet with some modification and clever tricks like the ones I mentioned to keep the game running fast. Is all you really need. Oh a programmer might help as well :P


  11. Currently there is no active development on Rapid Assault. We still maintain ownership and the potential to develop it.

    Priority is to keep WWII Online going and to build up the community and our resources so we can get onto Steam opening a new door to the market.

    I am going to shoot you a PM XOOM about something that may or may not deal with Rapid Assault. :)

    EDIT: PM SENT.


  12. I belive CRS have already checked on most of the engines, and usally the engines seem to fall short on supporting 128 visible players within your "rendering bubble" as the current engine can handle. Combine that with all of the calculating of projectiles flying all over the place in a busy AO, and a prio system that is constantly calculating what it should render for your computer, based on what kind of unit you are currently playing.

    The above is a guesstimate from my part, im not 100% sure of the above but it's the little pieces i've gathered from here and there of DOC and rest of CRS guys replying to questions in the forums :).

    P.S. Ofc it would be awsome to get a new graphics engine for the game, i would love it if they could make it work with WW2OL!

    Hey Maxois - Thanks for the reply :)

    Honestly they (Epic Games) provide full source code for 20.00 USD (as stated above) so you could change out the rendering engine and network stack to what ever you wanted to if it does fall short.

    My main concern in a game like WWII Online would not be the rendering but physics. Which they could do just port over the current physics implementation (assuming it would work.) or use one of the many open source ones or Licensed out (see Bullet or PhysX) not to mention allot of performance improvements come with the engine such as multi-threading and the like. :)

    P.S. I really want to see a WWII online game that looks modern in terms of graphics aswell :)


  13. S! CRS,

    I am just curious about something - I know that you guys have been developing your own in house engine for about 13+ years now. I am just curious if you guys thought of porting WWII online to Unreal Engine 4? You know let some one else do the heavy lifting. :)

    I know that this sounds kinda absurd but here is why I say it.

    1. Engine development can take allot of time and money. You guys have been doing this for 13+ years. According to Wikipedia the average game engine programmer salary is 95,300 USD annually as of 2010. It can be higher or it can be lower.

    2. Unreal Engine 4 is 20.00 USD per month / user + It includes full source code access via Github. LINK

    3. The game looks dated - I will admit the game no matter how old it is awesome and will hold a place in my heart. But Graphics do play a major role in attracting new customers.

    4. Multi-Platform - The Unreal Engine 4 can be run on allot of different platforms ranging from Android and iOS Devices to Windows, MacOSX, Linux, SteamOS and HTML5 Enabled Browsers :).

    5. The Future - According the Steam hardware survey most people are running at least a dual core system with 8 GB of system ram. Why not take advantage of that?

    What ever you decision may be I will support it. Keep in mind that this was just a suggestion NOT me being a rude arrogant prick. I want CRS to succeed like many other developers out there that I actively support.

    Thank you for your time,

    doctorclot


  14. Hey CRS,

    I am curious what ever happened to Rapid Assault?

    I backed it via Kickstarter about a year ago or two ago and It did not go through but I feel that there is room for a game like this in the current market.

    As much as I love WW2OL I feel overwhelmed at times when playing it. To Much going on sort of thing. I like smaller more focused experiences.

    P.S - If this has already been asked forgive me.

    Also - If you guys do not mind I would like to tell you what you guys did right and what you did wrong with your kickstarter campaign. Just so that you can learn from your mistakes. I really want to see Rapid Assault come to fruition. :)