I sold my house 'bout to finance a game development

2 hours ago, AlanDontAsk said:

So when I meet some programmers, I have to tell them what to do. What technique is best for this? what coding language? What game engine? Is Unreal engine suitable for this?

You are not asking Game Design questions, so I'm moving this to the Production/Management forum.

A good project manager doesn't tell the programmers what to do. He gives them a detailed GDD and because they're experienced, they know what to do. They know what engine and language to use. You don't tell them what engine and language to use, if you aren't a programmer yourself.

you know what tom "Sloper", Did I ask you to tell me if my question is smart or not? Really?

If you're not going to help then don't talk to me. Ney sayers are the worst kind of people.

You know Tom, if I have 5 programmers telling me different things, how can i tell them apart? Are they supposably ALL good? Maybe you need to realize you're ignorant before judging peoples questions. BYE! You can lock my account with that attitude. I never said I'd be telling them what engine to use but I'd try to remotely understand what they are telling me and which one is right and which one is not.

BYE!

37 minutes ago, Tom Sloper said:

You are not asking Game Design questions, so I'm moving this to the Production/Management forum.

A good project manager doesn't tell the programmers what to do. He gives them a detailed GDD and because they're experienced, they know what to do. They know what engine and language to use. You don't tell them what engine and language to use, if you aren't a programmer yourself.

I HATE YOU! I literally HATE people like you. If I pay yes I tell them what engine to use. Some dumb programmer wants to use crappy game engine that he can easily program in and some java crap. Plus you move my thread to here? REally? THis is not a game design question??? Really???? Well if this isn't a game design question then i don't know what is. I am asking about designing the actual levels and THAT IS DESIGNING THE ACTUAL LEVELS literally! You feel competent to talk here and to judge??? Move my thread like that? Really? And then tell me to blindly listen to some scammer programmer?? Really?????? WOW thank you Tom... thanks a lot! So helpful

well, you could try to quickly get up to speed with [ this ] highly organized, to the point, gentle reads on those types of queries. On that also, I feel you may have not meet successful programmers yet judging from your reaction. I think I understand though, It's rough starting out with nothing. I'm wondering how detailed these 150 unique buildings would need to be? Instead of zoom in, might we have a vessel of sort to get around in and the ability to stroll on foot. Do you need artists/modelers also? besides programmers??? Don't forget the sound guy and the tax and paper work guy. Whatever those costs are, plus next to nothing for marketing when that time comes, how much development time does your current plan allow you before you'd be looking to replenish funding? Oh, and is multiplayer a must have or could do without kind of thing...

2 minutes ago, GoliathForge said:

well, you could try to quickly get up to speed with [ this ] highly organized, to the point, gentle reads on those types of queries. On that also, I feel you may have not meet successful programmers yet judging from your reaction. I think I understand though, It's rough starting out with nothing. I'm wondering how detailed these 150 unique buildings would need to be? Instead of zoom in, might we have a vessel of sort to get around in and the ability to stroll on foot. Do you need artists/modelers also? besides programmers??? Don't forget the sound guy and the tax and paper work guy. Whatever those costs are, plus next to nothing for marketing when that time comes, how much development time does your current plan allow you before you'd be looking to replenish funding? Oh, and is multiplayer a must have or could do without kind of thing...

wow thank you so much for this helpful... comment I really appreciate it.

Focused on the actual subject. I am driving back home will reply in about 20 minutes.

3 minutes ago, GoliathForge said:

well, you could try to quickly get up to speed with [ this ] highly organized, to the point, gentle reads on those types of queries. On that also, I feel you may have not meet successful programmers yet judging from your reaction. I think I understand though, It's rough starting out with nothing. I'm wondering how detailed these 150 unique buildings would need to be? Instead of zoom in, might we have a vessel of sort to get around in and the ability to stroll on foot. Do you need artists/modelers also? besides programmers??? Don't forget the sound guy and the tax and paper work guy. Whatever those costs are, plus next to nothing for marketing when that time comes, how much development time does your current plan allow you before you'd be looking to replenish funding? Oh, and is multiplayer a must have or could do without kind of thing...

One man has developed an incredibly good project and it takes my breath away:

Something like this, just with more characters and functions. For example interactive clicks on characters or cars etc...

When entering the building, would be able to do stuff inside of it interactively.

This kind of algorithm sounds dumb and pointless but the possibilities are huge what could be made

7 minutes ago, GoliathForge said:

Also my game wouldn't be a game with prupose of me getting rich or idk... The game would be practically usable in some scientific and engineering things. Also with only 10 % of it developed, it would already be usable and ready for sale. Eventual cities and stuff would just be an upgrade. However ensuring that proper algorithms and designs have implemented so that future adding of "levels" and "sublevels" is possible would be actually mandatory.

I'm working on something similar, not the Earth exactly but large world technology. The difference is with a procedural planet, what I'm doing, you aren't trying to duplicate something existing like the Earth. You are specifying parameters and letting the functions generate something.

That being said there is a mathematician named Michael Barnsley who did a lot of advanced work in Fractal compression (https://en.wikipedia.org/wiki/Fractal_compression) .

This is what you will probably need. You will also need a good meshing system to turn those fractals into actual terrain. This is the part I've been chiefly working on for my game.

As I'm sure you are aware, this is not going to be an easy thing to write. I've been working on my stuff on and off for a few years and only now am beginning to see a flicker of light at the end of tunnel. Whatever programmers you get, they will absolutely need to be well versed in multi-threading, or at least willing to research it. The free ride is over with computers. To get the power out of them now, you NEED multi-threading.

As an example I wrote my fractal voxel engine and I was only using about 13% of my computer's 4 cores. I had to scrap a fair chunk of it in the last few months and rewrote it for mili-threading. Now I'm up to about 70% and it runs a LOT faster. I actually had to do it more than once because until you try it and profile it, it's hard to tell where the hangups are. There are still things I can improve on too, but at least now I understand the data flow issues better . I advise C++ for maximum performance. I'm not anti C# or Java but this kind of thing needs all the power you can get.

Also you might have a problem with game engines, because most use 32 bit floating point CPU side (GPU side 32 bit is OK as long as you use view coordinates) . 64 bit float will make your life a LOT easier for planet sized stuff. The one I know of that uses 64 bit is Unigine game engine. Also I think Amazon Lumberyard is going that that direction, but I don't think they are there yet except for the Star Citizen version. And of course you can write you own, which is kind of what I'm dong.

In any case, good luck. You have a long slog ahead of you.......

26 minutes ago, Gnollrunner said:

I'm working on something similar, not the Earth exactly but large world technology. The difference is with a procedural planet, what I'm doing, you aren't trying to duplicate something existing like the Earth. You are specifying parameters and letting the functions generate something.

That being said there is a mathematician named Michael Barnsley who did a lot of advanced work in Fractal compression (https://en.wikipedia.org/wiki/Fractal_compression) .

This is what you will probably need. You will also need a good meshing system to turn those fractals into actual terrain. This is the part I've been chiefly working on for my game.

As I'm sure you are aware, this is not going to be an easy thing to write. I've been working on my stuff on and off for a few years and only now am beginning to see a flicker of light at the end of tunnel. Whatever programmers you get, they will absolutely need to be well versed in multi-threading, or at least willing to research it. The free ride is over with computers. To get the power out of them now, you NEED multi-threading.

As an example I wrote my fractal voxel engine and I was only using about 13% of my computer's 4 cores. I had to scrap a fair chunk of it in the last few months and rewrote it for mili-threading. Now I'm up to about 70% and it runs a LOT faster. I actually had to do it more than once because until you try it and profile it, it's hard to tell where the hangups are. There are still things I can improve on too, but at least now I understand the data flow issues better . I advise C++ for maximum performance. I'm not anti C# or Java but this kind of thing needs all the power you can get.

Also you might have a problem with game engines, because most use 32 bit floating point CPU side (GPU side 32 bit is OK as long as you use view coordinates) . 64 bit float will make your life a LOT easier for planet sized stuff. The one I know of that uses 64 bit is Unigine game engine. Also I think Amazon Lumberyard is going that that direction, but I don't think they are there yet except for the Star Citizen version. And of course you can write you own, which is kind of what I'm dong.

In any case, good luck. You have a long slog ahead of you.......

Thank you so much! That's a good read! I will look into this and research each and every topic... I am willing to do 1 -2 years of research and designing, testing and trials before actually going ahead with it.

26 minutes ago, Gnollrunner said:

I'm working on something similar, not the Earth exactly but large world technology. The difference is with a procedural planet, what I'm doing, you aren't trying to duplicate something existing like the Earth. You are specifying parameters and letting the functions generate something.

That being said there is a mathematician named Michael Barnsley who did a lot of advanced work in Fractal compression (https://en.wikipedia.org/wiki/Fractal_compression) .

This is what you will probably need. You will also need a good meshing system to turn those fractals into actual terrain. This is the part I've been chiefly working on for my game.

As I'm sure you are aware, this is not going to be an easy thing to write. I've been working on my stuff on and off for a few years and only now am beginning to see a flicker of light at the end of tunnel. Whatever programmers you get, they will absolutely need to be well versed in multi-threading, or at least willing to research it. The free ride is over with computers. To get the power out of them now, you NEED multi-threading.

As an example I wrote my fractal voxel engine and I was only using about 13% of my computer's 4 cores. I had to scrap a fair chunk of it in the last few months and rewrote it for mili-threading. Now I'm up to about 70% and it runs a LOT faster. I actually had to do it more than once because until you try it and profile it, it's hard to tell where the hangups are. There are still things I can improve on too, but at least now I understand the data flow issues better . I advise C++ for maximum performance. I'm not anti C# or Java but this kind of thing needs all the power you can get.

Also you might have a problem with game engines, because most use 32 bit floating point CPU side (GPU side 32 bit is OK as long as you use view coordinates) . 64 bit float will make your life a LOT easier for planet sized stuff. The one I know of that uses 64 bit is Unigine game engine. Also I think Amazon Lumberyard is going that that direction, but I don't think they are there yet except for the Star Citizen version. And of course you can write you own, which is kind of what I'm dong.

In any case, good luck. You have a long slog ahead of you.......

Do such services already exist and would such software be possible?

I have one question though...

I have seen 5G in Shenzhen China providing almost instant remote control over computers. So people used cloud computing to play heavy games like Crysis 3 on their very cheap laptops (basically instant streaming service). 5G is crazy... Is there any way to plan ahead and implement a platform where users would actually be getting a stream of the software, rather than the software itself and as a compensation some computing power could be provided from a third party?

THis is 100% definitely possible with 5G technology. 5G does not just mean higher bandwidth, that thing is fast as hell. I was still under impressions of what I've seen in China... The aim is to give that kind of strategic game on a small android phone for let's say 20 bucks a month.

30 minutes ago, AlanDontAsk said:

Thank you so much! That's a good read! I will look into this and research each and every topic... I am willing to do 1 -2 years of research and designing, testing and trials before actually going ahead with it.

I have one question though...

I have seen 5G in Shenzhen China providing almost instant remote control over computers. So people used cloud computing to play heavy games like Crysis 3 on their very cheap laptops (basically instant streaming service). 5G is crazy... Is there any way to plan ahead and implement a platform where users would actually be getting a stream of the software, rather than the software itself and as a compensation some computing power could be provided from a third party?

THis is 100% definitely possible with 5G technology. 5G does not just mean higher bandwidth, that thing is fast as hell. I was still under impressions of what I've seen in China... The aim is to give that kind of strategic game on a small android phone for let's say 20 bucks a month.

I'm not an expert on this but I imagine with a high enough band width it's possible. However now a player's phone basically just becomes a monitor and input device, and you have to provide all the computing power for players. I assume that's why you said 20 bucks a month. That's pretty steep for most games however.

There may be some compromise where you are only sending down polygons and still doing some rendering on the phone. In that case the task becomes to figure out the minimal data that needs to be sent.

Simulating the whole Earth with all the detail down to the toys inside the houses is a very ambitious goal. Rockstar's GTA series comes to mind and it doesn't even come close to what you're talking about and they have hundreds of people working on it for years. I'd start with one city and several small buildings to get a taste for the challenges ahead.

I'm curious, may i ask a personal question and enquire how much your house was worth?

How much do you intend to pay developers that work on your project and what is the timeframe? -- I'm genuinely interested in that figure.

A game of this kind of scope would require tens to hundreds of millions of dollars to develop, it is a combination of google earth and GTA, both of which have hundreds of millions of dollars of development and millions of man hours sunk into them.

Unless you're selling a mansion, i think you may need further financing to see this through to the end.

49 minutes ago, VoxycDev said:

Simulating the whole Earth with all the detail down to the toys inside the houses is a very ambitious goal. Rockstar's GTA series comes to mind and it doesn't even come close to what you're talking about and they have hundreds of people working on it for years. I'd start with one city and several small buildings to get a taste for the challenges ahead.

Oh absolutely not! GTA type of a game is not something imagine...

So let us say you would have around 100 cities and in each city you would only have 2 types of buildings per each city that would be specific for only certain parts of the map.

That is all... it would be a strategy game so bottom line is: around 2 types of buildings. Each building would have around 5 objects with their specific characteristics so that is around 1000 objects (objects would all be very similar with similar functions but all slightly different, So for example in one building youd find a car with golden doors that open horizontally, in the other youd find a car with doors that open vertically and bigger trunk)

I actually have the plan quite worked out so those numbers are not randomly given but indeed this is what it would be.

I am 100% convinced it is possible. The genere of the game would be very weird and idk what to call it...

13 minutes ago, Brain said:

I'm curious, may i ask a personal question and enquire how much your house was worth?

How much do you intend to pay developers that work on your project and what is the timeframe? -- I'm genuinely interested in that figure.

A game of this kind of scope would require tens to hundreds of millions of dollars to develop, it is a combination of google earth and GTA, both of which have hundreds of millions of dollars of development and millions of man hours sunk into them.

Unless you're selling a mansion, i think you may need further financing to see this through to the end.

The house was in Germany and it is/was my fathers house. Since I dont live there anymore and since I plan to move to America with my fiance, I decided it is best to sell it. The house was sold for 550k eur with taxes deducted. Time frame idk...

10 minutes ago, AlanDontAsk said:

1000 objects (objects would all be very similar with similar functions but all slightly different, So for example in one building youd find a c

This is a proof that it is possible:

It onlz depends on the simplicity and number of objects involved. Whether procedurally created or not.

This what I imagine would be 1000 objects at maximum that could be procedurally put into buildings and cities. And 200 types of buildings that would be proceduraly put into those cities. Everything else is not important. Grass, trees etc etc... Would be nice if something can be added, not mandatory for the gameplay.

Before diving into something big, best start with something small. I've published around 8 video games whose production I handled almost entirely myself... all of them "small" games. Puzzle games, boardgames, simple arcade shooters, etc. Gives some idea what you'd be up against with something as technologically challenging as you are proposing. The amount of hassle I had making even something as small as Cognizer is hard to believe.

As you've already got an adequate flat to live in, take just a few months to make a small game first, then move on to your real project, if you still think it is viable.

31 minutes ago, Almo said:

Before diving into something big, best start with something small. I've published around 8 video games whose production I handled almost entirely myself... all of them "small" games. Puzzle games, boardgames, simple arcade shooters, etc. Gives some idea what you'd be up against with something as technologically challenging as you are proposing. The amount of hassle I had making even something as small as Cognizer is hard to believe.

As you've already got an adequate flat to live in, take just a few months to make a small game first, then move on to your real project, if you still think it is viable.


Yes... you are right. I will be wise about this. So many dislikes so many ney sayers.. like people don't even tell me nice words like you to slow down and rething or check stuff and idk. No constructive advice,they just keep hitting that dislike button and telling me to pay programmers and shut my mouth.

I can't believe how much negativity on the forum. Thank you so much for being different and sharing your opinion instead of just hitting the dislike or telling me to pay a programmer and keep my mouth shut.

1 minute ago, AlanDontAsk said:

My post is not "negativity". It is "the reality of the challenge of video game development". I wrote it in very positive terms. It gives very specific advice about how to reach your goal. Negtivity would be, "You'll never reach this goal. Give up now." I said no such thing.

Example of the reality of game development: Two experienced developers from Harmonix (makers of Rock Band) set out to make a new rhythm game. The programmer decided to make his own engine. It took them seven years. If you watch the video below of Thumper, you will be able to tell that it is a narrow game. Yes, it's beautiful, well-polished, and super-fun. But it has so little of the complexity of your proposal.

I repeat my previous advice in full confidence that it is not "negative": make a small game first, then you will have a better idea about what your proposal will entail. That will better enable you to plan it properly and have a better chance of succeeding.

Here's what Thumper looks like. Remember, two experienced game developers, seven years.

Hi again. Just so you know......... When you quote someone you should respond below the quote box not inside it. Otherwise, it looks like he guy you were quoting wrote your reply.

Buy a few cheap buildings and cars on TurboSquid and start making cities out of them in your favorite engine. See how many buildings and cars you can cram in until you get an error or your computer starts chugging. Then make 2 cities, then 4. See how smoothly you can transition between them. Push the limits to find out where they are and where the challenges will be. I'm more familiar with the C++ side of things, so if I were you I would start working on dynamically loading and unloading cities in a separate thread and flying between them smoothly, but you can use existing solutions and see how far you can push them. Then get programmers involved.

7 minutes ago, Almo said:

My post is not "negativity". It is "the reality of the challenge of video game development". I wrote it in very positive terms. It gives very specific advice about how to reach your goal. Negtivity would be, "You'll never reach this goal. Give up now." I said no such thing.

Example of the reality of game development: Two experienced developers from Harmonix (makers of Rock Band) set out to make a new rhythm game. The programmer decided to make his own engine. It took them seven years. If you watch the video below of Thumper, you will be able to tell that it is a narrow game. Yes, it's beautiful, well-polished, and super-fun. But it has so little of the complexity of your proposal.

I repeat my previous advice in full confidence that it is not "negative": make a small game first, then you will have a better idea about what your proposal will entail. That will better enable you to plan it properly and have a better chance of succeeding.

Here's what Thumper looks like. Remember, two experienced game developers, seven years.

I understand... I mean I 100% agree with you. I will be smart and see what can be done

5 minutes ago, VoxycDev said:

Buy a few cheap buildings and cars on TurboSquid and start making cities out of them in your favorite engine. See how many buildings and cars you can cram in until you get an error or your computer starts chugging. Then make 2 cities, then 4. See how smoothly you can transition between them. Push the limits to find out where they are and where the challenges will be. I'm more familiar with the C++ side of things, so if I were you I would start working on dynamically loading and unloading cities in a separate thread and flying between them smoothly, but you can use existing solutions and see how far you can push them. Then get programmers involved.

That is very good idea! I will definitely be doing that. i actually made couple of things in unreal engine using the nodes and some tutorials so I did not code anything. But I know how the graphic system of the engine works and how to do what you have mentioned. I will definitely be doing that.

AS for the dynamic loading and unloading that is another thing I will do.

I will try to have a bunch of random objects with random functions doing the similar tasks Id need in a bunch of random buildings and see how well computer loads and unloads.

IF I manage to solve that properly and prove myself that it can work, I am willing to start making re objects with functions and 3d models for those objects.

Also i already have almost 1500 models I have made on my own. Many of these are buildings....

I will look into types of functions I need and apply random functions for such models with blueprints and even though it will make no sense and no gameplay or anything, it will be very crucia to prove that unloading and loading the amount of objects and functions that I need will actually work within the existing engine.

When I say earth... It doesnt have to be detailed but only scale-wise it should be just as huge as real Earth compared to a building. Between those generic cities, there is no need for any kinds of details...

A similar thing is also silent hunter series... They have real world sized maps with cities on the shores... Cities are quite details with realistic buildings and very unique looks.

Even better example is the xplane 11 series and yes i know xplane 11 is very complicated thing and many people worked on it, what i have imagined is nothing that sort of complicated and can be modular...

For example I could first create 10 cities with their 2-3 types of buildings and in each building around 5 characters (objects) that is roughly >>> 100 characters and objects and iit would be immediately playable ... oh MY GOD!!! That is it....


I should not first create all the cities but instead create 1 city 2 types of buildings and 5 objects in each building. Then just multiply it by 100 and see how engine can handle it!!!!!

And then if it works... just expand it...

The principle of the game can be played practically even if there are as little as just 5 or 10 cities very generic easy to make cities with around 100 objects unique objects in all of them.

As far as negativity goes, this forum tends to be rather specific practical solutions-oriented. When subjective things like ideas or feelings get involved, things tend to get heated and out of control rather quickly. I've been burned by this at first too. I've had better luck asking specific questions like "why is this faster than that" or "how to fix this" and providing code, screenshots or videos.