Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
For what you consider "real-time destruction physics on a scale never seen before " was true only back around 2008-2009. This kind of destruction is now something simple to produce as it, BUT it's also quite limited and require a lot more works (time, so money) for every assets and phase in the development of the game.
It's also unfriendly with many new technologies accessible world-wide. You might think "then they just have to not uses those new tech", but that's like saying that to build a computer, you should always start with scraps of plastic, iron, copper and aluminium and build everything from raw material.
To give you an idea of how limited the destruction in this game is :
- Destructions is highly limited to static assets. If it's animated/skinned, it doesn't get destroy the same way and, so, require actually 2 systems for the destruction which, if not calculated correctly, can easily become a bottleneck.
- There are many now common features of engines which would not work correctly with this without being a pain in the butt for the PC. For example, substance materials which is a great technology to not only reduce the required HDD space, but also make uses of the RAM instead of the VRAM for the texture generation. While it's possible to uses a basic substance in this system, you wouldn't be able to update the inner data of the material which remove 80% of the possible effects and customisation capacity of the subtance.
- Even if you think that this game have huge maps, which is true, what you ignore is that this game is actually really cheap on amount of content. There's a lot of recycling and you can see that they were really limited on some of the environment assets as they couldn't add more content (textures, material & assets) and had to mix things around. Every scenes are contain very little unique assets in ratio to their size.
- The asset themselves looks, at best, medium. The texture artist(s) did a great job, but you can still see that something is always fake and looks strange.
To give you an idea, someone made a really similar destruction system for the free Unity 4.X (now 5.0) engine and it's being sold for 60$. (I bought it for 15$ when it was 75% off to test it out) What you consider as something awesome is considered as a system that worth 60$ in the indie community. I'm not saying that you're wrong to find this system cool, but the reason why it's not used more often is because its limitation are making it unwanted. (With every system, there's a good side and a bad side which you must work with.)
it may very well be simple and easy to reproduce, which is partly my point
forget geo mod, why arent devs looking into better, more efficient ways of allowing similar or better destruction
be it lack of demand but as a creative medium should always seek to innovate
not exactly sure by what you mean with new technologies, but the hardware we have now is more powerful and people are getting better at programming as the industry grows
im no expert, and i admit that putting geo mod in the title is probably a bit misleading, geo mod or not, this proves that some form of destruction is more than possible with the hardware we have now, and that it is also possible to incorporate it in games
with that in mind, why arent more people giving it a go?
again, im no expert on this but back to talking about hardware, it can already handle the destruction in this game very well, we have more powerful hardware now, and im not really sure about what you mean by "not calculated correctly", which is a question of optimization
even so, if non-static assets can't deal with the destruction (although im not sure if the wind turbines count as "animated", but also that physx flex cant have sticky liquids stick to a moving wall) any destruction at all would be nice
limited in this instance, yes, but as far as games go, most dont even bother
many developers already take liberties with realism and world interaction, as developers continue to work towards building more and more realistic games, even destruction of static objects is a welcome sight to see; it shows progression towards, overall, a more realistic gaming experience and is also a huge milestone considering how many large studios still only use basic destruction systems
games are getting larger and so is our storage; most modern games released are upwards of 40gb, and im in no way saying this is a good thing or that it's okay simply because every other developer is doing it
at the risk of sounding repetitive, im sorry i didnt make it clear enough that im only using geo mod as an example of what was possible in 2009
if im not wrong in assuming what youre talking about in terms of substances, physx flex simulates materials but it is possible to have multiple different kinds of objects interact with each other (cloth and liquid, liquids and solid destruction) and that it is also possible to adjust the wind, surface tension, viscosity, and many other things to do with these materials
whilst physx flex isnt exactly easy on the gpu, namely fluid simulation, others seem to be much much less intensive, i dont see it being completely out of the realm of possibility to incorporate a destruction system in a videogame that is somewhere between what is doable in physx flex and what is doable with geo mod
that i can agree with, the map and some npcs is what actually seems to be the only things that arent duplicated in this game, even some entire bases are copied and pasted in different regions, and i dont just mean buildings, even the outer walls and courtyard layours appear to be completely identical in some places
i give it the benefit of the doubt seeing as the game is fairly old, and like i said, the hardware now is capable of bigger and better physics
this is still the case with many modern games, and forgive me for saying it again, hardware now is capable of much better things
few games are really well polished now, and you can't find a single thing out of place if you look hard; personally, i would prefer the presence of a destruction system over a near-flawless presentation of a indestructable environment, and besides, it's developers jobs to make things look better; lighting and environments in games nowadays look fantastic, but its no use if they cant get it to look good with a destruction system that might very well become much more common in the future
i dont want to edit what ive already said but i think i realized what youre trying to say now is specifically about geo mod and similar destruction, i might have just wasted my breath about everything above, at least as a response to you
either way my point is that i believe this is something developers should work towards games having more of, and RFG is an example of it working adequately with satisfactory results inside of a game environment
So, I'd say that the Op has it right. And that the reply regarding these companies being, essentially, too cheap, even though something said to be comparable, but not the same, is $60 and was bought for $15, if it's never taken to task and sold as this system here in Guerilla, then it doesn't matter what the cost was, nothing came to market, and maybe because this other system isn't comparable and wasn't able to be used with CAD as what they did in Guerilla. There is no balance of negatives and positives, they both exist, but the fact is if this system impresses to the point it has here for another game for a no name indie developer, that pays for it in spades. As a development it's an investment, and to view it otherwise is to wear the player buying a game hat and not the developer hat, at least not a quality one, not a developer trying to develop a game that sells itself by being very impressive, but a developer failing by applying principles that have nothing to do with development of a game as a business.
Hope more do what they did here with Guerrilla , with the quality this shows, as that is good for the gaming industry as a whole. Being cheap only leads to people passing over the game and not buying it, even when the game is on a Steam 80% off sale hahah
Where did you get all this info? GeoMod runs on quite different platforms in the end, PS3's Cell and Windows with x86 architecture for PC so it seems to be quite portable.
precisely, it's unique, much like crysis back then in all its glory was the crown jewel of visual achievement, and that was what sold the game back in the day
now, every other AAA title boasts similar if not better graphics, the crysis series not being able to outshine the competitors as much as it did back when it was the gold standard of visuals that every game thereon after wanted to have
if we put destruction, even on the level of RFG in more games today, who knows what kind of games will show up later on? such a thing in any game is very nice to see, and even better when its implemented well
to add to this point, yes, every developer has their own intentions, and behind those intentions were ideas.
im afraid though, it might just not be developers being cheap, but simply unaware or that it never occurred to anyone that something like this could be done, or re-done
RFG is just a whisper in the many titles that come to mind when people think of the word "physics", even games such as battlefield or angry birds would come to mind before before geo mod came anywhere close to the conversation
at the very least, RFG is proof that it can be done and done well within a playable game environment
it really is a mystery why this isnt as common as it could be
To help you understand, the problem with destructive environment, dependless of the system or engine used is the fact that it must be run at real-time. Sure, we got better components in our PCs and we can render more stuff on screen, but to be honest things are still limited.
To understand, first, we need to divide the type of destruction into the 2 main categories :
- Pre-made destruction :
This is the destruction created within the 3D assets. You might not notice it, but Red Faction makes use of this system a lot more than the other system. Simply put, every assets are already broken into pieces even as you see them unbroken. You can notice small seems (lines) around where the pieces gets broken prior to destroying it when the light generation is around 45 degrees and the shadow slightly appear for 1 pixel. When you hit a pre-broken part, only then its physique gets activated and gravity do the works.
The Battlefield series (since Bad Company 1) uses this system in its destructible environment.
The quality of this destruction is the possibility to every type of shaders and maps since the pieces are actually like any static assets with only a slightly more complex behavior.
The bad side of this is the time it takes to be made. Not only must the dev build the outside, but also the inside of all things that can get broken. It also force either more textures to be made or cheaper environment (if you star using the same set of textures everywhere in the building). Let's say that making a static undestructible building takes 2 days. The same building being fully destructible takes around 2-4 weeks instead. (This is why many assets are used multiple times in Red Faction)
Another bad side is that, since it require the assets to be premade in pieces, it also constantly use more vertex memories. To save on this, there is a small optimisation we uses that consist of having a initial model without destruction but as soon as it's hit once, it gets switched with a destructible model which disapear after a while (cleaning some memory). Remember that, while the vertex themselves are not something big and can be managed, the issues come from the generation of shadows and lights effects as well as possible drawcall bottleneck issues.
- Real-time generated destruction :
This is the destruction that is generated in real time. For example, the destruction of cars in some games where the car deform when it hit somethings. This type of destruction include things like vertex based scenes (like the Space Engineer and, recently, Medieval Engineer games), but also interactive assets like the "cut" enemies in some Action games like Metal Gear Rising.
A good side of this system is how imprevisible and how exclusive any action can change the world. Also, prior to the destruction, things seems less "imperfect".
The bad sides is that it's limited on the textures usages, require a lot more memory for all the physiques. The more "realist" is the behavior, the more ressource it takes.
Something you need to know is that even with a PC with some of the best components (like having 64Go of RAM and 2x 16Go of VRAM graphic card), most of the time the game engine will still be limited to the same old 32 bits of data. (Note that we're not speaking of the OS with 32 or 64 bits, but about the floats range which is the type of data used in positions, collisions, etc.)
This limitation bring a serious wall which we call "float precision". There are many ways of countering the issues, but still it means that, for example, any float values in a game can't exceed 7 numbers in its value by the end. From that 7 numbered value, the float precision means that the last 1-3 numbers can be slightly off the mark. (So, let's say that an object is situated at 79874.23 on X in the world space, at every frame, its position will varies within the 79874.XX and 79875.XX. This is why you can get a ragdoll stuck in a wall or the ground as its movement are calculated with a higher error range (due to the constant movement and restruction, which reduce the 32 bits float data to half of its capacity)
Where I go with all this? Remember that destructive environment require constant modification toware those floats values' for a LOT more assets at every frames so there come a reason why destruction is not always wanted.
What I'm trying to do with all those explaination is to explain the main question the OS asked which is "Why doesn't we see more destruction like what we find in Red faction in other game".
The most simple answer is : Because, in most cases, it would cost too much to do because of all the work related to what I explained above. (Everything is about money) This kind of system involve a crapload of additionnal hours for both the Q&A team and the 3d artists. Publisher (or selfpublishing devs companies) doesn't have such budget most of the time.
That's the reason.
For being funded, such game would need to have a serious fanbase and the devs should have prior experience with such system, otherwise publishers wouldn't dare to fund a project that would be "simplier" than other competitors' project just for the sake of its destructions.
Just take the previously said Bad Company side-series of the Battlefield series. While the trailer was popular, the game itself didn't interested as much player as it have hoped. The maps were too much of a "wreck" it made the tactics used in previous Battlefield 2 (which was hugely and is still rather popular) reduced to hit & run without any more secure hiding spots (unless the uses of glitches). The old players either liked or hated it, but it brought to light that a full destructive environment was not wanted by most of the players. (Hence why Battlefield 3 and 4 doesn't include as much destruction as in Bad Company 1 and 2)
but it could be done, right?
at the most basic level, we have RFGs destruction as it is
now we have better, stronger hardware
it's not completely farfetched to think that it would be possible to make a newer game with something similar, right
Reply follows, too many characters when quoting as well.
For 2) You're need to bifurcate for sake of your explanation is to eliminate all other discussion. That's wonderful in a vacuum of thought, but since you're points are being made aloud to others, you do not get to dictate what's possible, what is able to be considered in contemplation of something, especially an enthusiastic fan's question about a situation, that I can only frame for me, which is that it seems unbelievably stupid that these companies haven't built more games using this method.
For 3) After all your supposed effort to, you claim, help others understand, you generically say "slightly more complex behavior" without defining what slightly is at all. Thus, that what you have to say is entirely a belief is exposed for all to see, because you have qualifiers for what you had to qualify for yourself but you have no qualifiers for what you want others to trust you on their values, as clearly should be explainable beyond a term of absolutely no precision, "slightly more complex behavior," which, since this isn't an AI situation, there is no "behavior" involed.
For 4) "Bad side" according to you. If I am to take you to be claiming some representation as a developer, not necessarily of RFG but of any game, I'll also assume your game is pretty worthless, and I can arrive at that conclusion when you're suggesting the "bad side" is the very thing that a player of RFG made a post regarding, as to why such a great system isn't being used elsewhere more prevalently. You haven't presented a bad side, and instead, presented the market blindness of lousy developers. To help define this, this ist he developer who keeps making a game based on their own views and ideas, they fix bugs they find, they add content they want, and then get upset their game isn't selling. Their trouble is they failed to look at their forums, failed to ascertain who is genuine and who isnt, failed to post an email for ideas so people don't have to be subjected to posting somethng and having others try to "help them understand" something that says what they originally posted about is really not such a big deal, and undermine the very enthusiasm of that player. In fact, usually these developer "understand" so much that they have an excuse for why every player complaint is "just a carebear" or "just a whiner" while still that deverloper is complaining no one is buying their game, or playing their game for more than a few days after purchase, and that no one is buying in the in game store. This is a rather old tale of many many many developers, who love Early Access as a way to take the money and run, without ever getting the game finished, or ever having achieved anything the customers wanted, many doing a rinse and repeat and ripping people off so much that Early Access is becoming a bad thing to the very market that they are appealing to. These are the developers who are destroying gaming as an industry, wittingly or not, effect remains the same.
And I'll add your "bad side" is what would make me buy a game, that a company went to the trouble to do it well, I might even pay the full price for it (though I might wait a month or two to see if it goes on sale, I'd eventually cave because the quality put into the game deserves a reward, for everyone involved, they deserve to feed their families, buy new homes and cars, pay for kids to go to college, and the company deserves to payback their investors, who may have been so short-sighted as to want the game faster and have been the sole reason for short-cuts the diminished the quality of the game from A+ to B-).
[5] I can tell you that I see the seams on each block in Space Engineers which I have 1400 hours playing. I can tell you that when they recently decided to upgrade the voxels, that chaos has ensued for about a month now, to which some players stopped playing. I see no mention of how easily they can screw up the collision side of the game, nor how easily they can trigger, through the floating debris, a mass simultaneous calculation of triangles that doesn't resolve due to the factoring, and trying to make floating values work, when the reality is that the entirer map is integer, for binary knows no floating values. All floating values are fabricated period.
As to your 32 bit point, that only makes my point above, that these floating values are fabricated. Our cards do parallel 64 to 512 bit transfers of data, depending on the card's capabilities (most are 128 bit and their processors are 128 to 256 bit respectively). Those values are far in excess of anything you're talking about, but those floating values have to be converted to binary values for representation in RAM, and often these values are then just "tokens," equivalent to a "JUMP" or "JTS" (Jump To Subroutine) that executes the actual drawing codes. Mapping of an overlay system that creates an invisible "field" or "bounding box" around the item would also be triggered at some point, but may only actually excute upon the actual collision, depending on the coder's abilities and talent, as well as the Developer's sense of desire, their interest in quality in the player experience as we see it on screen.
So now back to Steve Wozniak's sweet 16 method for 8 bit processors, to double their capacity and precision in floating point, that was installed in the Apple II computers of the 1980's, this method remains a viable option to this day, and I dare say that what we have from game to game are varieties of that to the extent a coder can do their part in coding a binary to floating point conversion to include both negative and positive numeric values, but this is for sake of display to us since memory maps to 1s and 0s. I am sure Sweet 16 was extended to Windows through the DOS environment via Microsoft who was, at that time, the developer for Apple's Apple III operating system, their DOS (Disk Operating System) on the Apple II, and Apple's ROM based BASIC, as well a the CPM Z80 Processor "BASICA" that was the forerunner to BASIC in IBM's DOS once the 8088 chip based computers ("IBM PCs) started to show up.
You have no point to what you said unless you're going to tell me that in 30 years of computing that no one copied anyone else's processes and coding from one point to the other modifying it accordingly to the new and emerging technologies, so that we could use older proven methods that overcame the obstacles of the past to overcome similar obstacles in the future. I already know there is no truth to such a suggestion, which is what most of what you wrote is suggesting in fact.
For 6) Since your explanations are wonderful opines of beliefs you hold, and nothing more, I say: My turn. You are apparently unaware of the use of prime numbers and encryption methods, that these are how we overcome much of this, "tokenizing" and having layers of mathematical translation that provide the values displayed, whether as text or graphics, and that these cause the number of assets to be a meaningless amount, unless in excess of 128 bits in total, as that's the root for a 256 bit calculation to have everything from -1 to 1 in between, and also ignores that this can be doubled and is able to be "stacked" via pointers and stack values inside the CPU, which can mean another 4 per core, in many computers today, and that's not including the GPU. So I'll just figure you're in Computer Science 101 somewhere and are articulating what you understand up to this point in your computing education.
In summary, Donoghu....
I am a former assembly language coder and am ashamed that you present such a limited comprehension, and thereby appreciation, of what's before you and what it can do, how to get around and through the operating system, including what Direct X allows, should be something you'd be posing, and exceeding the limits claimed with pride. No, instead you are articulating and parroting everything that some controlling and manipulative operating system seller named Microsoft has told you. There's knowing your craft, and knowing what you've been told about your craft, and then there is using the latter as a stick to beat people into submission to the limits you won't challenge. You likely will never reach the point of appreciating the customer, and knowing that it's a business, and thereby requires such appreciation to sell its product, to then thrive, and further flourish, so that business can hire more people and help them get experience that supersedes their collegiate theoreticals, often presented by "scholars" who have distinguished resumes, but opted to be a "tenured professor" because they couldn't actually make anything they believe work in the real world. Please, I urge you to step back and appreciate that the OP posted something that represents a statement of the failing of the gaming industry. It is a criticism to be thanked for receiving, to then want to learn from this criticism how to actually, not just by appearances, deliver to the customer what they expect, and would qualify what you produce as a candidate for them to enthusiastically spend their money to buy. Anything less and, wittingly or not, you're headed in the direction of the poor developers described above, and as a gamer and computer enthusiast, as well as my "past" pre-retired life, indicates, I'd prefer not to see you go down that road, but the choice is yours entirely.
P.S. Apologize for typos in advance, old computer screen eyes, bugging out from light text on dark background contrast (not sure if it's cones or rods, but something aged poorly hahah).
Jackie, you couldn't be more right :)
That just might be the reason why we don't see destruction like this a whole lot, like the OP stated. It alters the game from a visual standpoint, sure, but also gameplay is affected. Have this destruction engine in every game where you're supposed to go up a flight of stairs, for example, to reach an objective. Maybe the stairs get destroyed and there's no way for you to continue, the developers have to put in other ways to get to your objective, and so on, and it becomes a work-around rather than telling the story they wanted to tell, which maybe only had one set of stairs and that's it.