Beyond C++

It’s a fact that C++ has been the most successful and probably the most widely used programming language till date in the gaming industry. Even if you were to look at the software industry as a whole, a near countless software projects owe their existence to the language and probably many more will eventually be made using C++. Universities around the world still churn out thousands of CS grads with C++ knowledge and there will be no shortage of programmers who know C++, at least for some time to come. So why is C++ so popular? The reasons may be many (, I am sure there are other near infinite blogs which touch on this,) but at the end of the day it just boils down to the simple fact, “it gets the job done”! While it made good business sense to employ C++ (some years ago), it doesn’t make all that much sense when we consider scalability looking into the future. C++ in it’s current form is — well, simply inadequate. Most people will agree with me that C++ has probably outlived it’s time, and it was time the programming community as a whole moved away from the language. Easier said than done though, but the question to ask is, what real options do we have which provide radical new changes to the way C++ operates? Very few approaches I would say. Before you raise your hand and give out the name of <insert your favourite language here>, let’s look at some of the challenges facing future game development and/or rather future software development as a whole.

Lets first look at the C++ language itself. It’s well known that C++ has it’s faults. It’s not an easy language to learn and an even more difficult language to master (, compared to other languages). It takes a substantial amount of time and experience to understand the intricacies of the language and takes even more time and effort to fully grasp the quirks and subtleties involved in software creation using C++. Typically it takes quite a lot of time before an average programmer can become truly productive with C++. The learning curve for the language is high and not for the faint hearted. Besides this the language has increasing come under flak for allowing seemingly undefined behavior without complaining too much. The language does very little to deter flawed assumptions regarding some (very) basic constructs that contradict how things actually work. Even with proper planning, strict development cycles and stringent coding practises, software development using C++ is difficult. Memory management is a bane and can cause unexpected and unwarranted catastrophes which are known all too well in the industry.

There are a growing number of languages that address the shortcomings of C++. Be it Java, C#, Python or most new(er) languages, all try to fill in the gaps left out in C++. As a matter of fact most languages do a good job at it. However with all it’s faults, C++ still stands out as a viable game development choice. That’s because of 2 primary reasons; a) it has vastly more libraries, code dumps (, I am talking about game development only), engines, examples and everything really, that it simply wins over the argument right there. True many libraries have bindings to other languages, but most of them seem rather inadequate or poorly maintained. Besides there are a lot more examples on cutting edge technologies (,especially graphics) written in C++ than there are in all other languages put together. and b) It’s easier to get programmers for C++ than any other programming language, Java being the exception. Things are changing though and there are some concerted efforts being made to promote other languages and platforms (XNA, PyGame) as viable game development alternatives. However all those remain some distance away from challenging C++ for the number one position.

The above mentioned points in support of C++ are non trivial. They go a long way in weighing out the demerits of building a game using the language. So the question really is, do we really have any viable options beyond C++? The answer is somewhere in between a complete YES and a total NO. As we stand today the best scenario is probably building the core engine using C++ and then having a scripting system on top of it. Be it Lua, Squirrel, Python, or whatever. That way you can always find a middle ground between having to reuse existing code and at the same time allow rapid development and prototyping capabilities. Many engines/games take this route and there is little doubt that such a process proves to be advantageous in the game building process. There are already a lot of games out there that use scripting language for rapid prototyping and in some cases building large sections of the game. Having a scripting language on top of the engine core is clearly a step in the right direction.

Scripting languages solve but some of the problems. They do a part of the job and they do it pretty well. However, there are issues related to game development which require newer and more radical approaches. The challenge facing game development in the future is building an engine which can effectively and efficiently use parallel programming/computing techniques (Invasion of the multi-core machines). Current generation programming techniques fall short of addressing the issue effectively. More importantly most newer approaches to effectively address multi-core problem are just way too complicated to be implemented effectively in C++. Efforts are on to find radical new solutions (C++ STM), but thus far they look good only on paper and still seem too cryptic to be put in production use. The issue of effectively using multiple cores of a CPU will  probably be the biggest challenge for the next generation engine designer. The more natural choice for addressing the multi-core and parallel programming issue is the the use of functional programming languages. Since functional programming approaches are typically side effects free, parallelizing functional programming code is easier than imperative programming. However mixing functional and imperative styles can be an equally daunting task. As my argument in the above paragraphs suggest, there will still be a lot of code in C++ that will need, someway, of interacting with any functional language that may be used.

It’s debatable if going “strictly functional” will solve all the challenges the future will throw at us. A more plausible scenario would be to have specific portions of the engine/game parallelized either by using a functional language or by having a subsystem level parallelism. This obviously would allow existing (C/C++) code to be reused, however there are still challenges to overcome even with such approaches. Sub-system parallelism means having each subsystem (physics, renderer, logic, AI, sound, network…) run in a separate thread/s. This however is a very optimistic approach since sub-systems tend to overlap and in some cases critically depend on each other. Such a system could be achived with existing code also, however I remain very skeptical whether such an approach will actually work on the ground. Another approach is to have job based parallelism. Divide your most CPU intensive tasks into jobs and then have a kernel system to marshal them based on priority. This is something similar to what an OS does and seems the best way to shoot for parallelism with existing mechanisms. This approach however requires you to split your design into a job based approach and that could prove challenging. Having a functional language as a script system (on top of a C/C++ based engine) is another idea to think about. I am not really sure how helpful this would be and can’t really comment on this (, since I myself haven’t tried such a radical approach, maybe I ought to give it a shot). But it seems very much possible to have a functional language as a scripting system, and could in theory be used to parallelize sections of the game/engine.

So it would seem C++ might just survive in the game engine of tomorrow, although in a less prominent form compared to it’s position today. It may happen that over the years C++ may eventually fade out, however it’s part can’t be total ruled out. Transition from C++ to any other language will be slow and may be tumultuous, with teams opting for a hybrid approach than just downright building existing functionality from scratch. As new technologies progress and CPUs with 100s of cores start being commonplace, we sould see the popularity of C++ waning and been replaced by some other (maybe functional) language. As time progresses C++ might well become increasingly irrelevant as more and more libraries get promoted to newer and more modern languages or newer more efficient ones take their place.

Two great books from NVIDIA

NVIDIA released 2 great books (free) very recently which are a must for anyone who tinkers with graphics and specifically with shaders. First one is the GPU Gems book, which I happen to have as a h/c (, long before it was released). It’s such an invaluable resource of tricks that are still very much valid to this day. I would recommend this to anyone and everyone who wants to get their hands dirty with graphics. Then yesterday  The Cg tutorial was released. I haven’t got a h/c version of the book as I pretty much have a good hand at HLSL and both (HLSL and Cg) are essentially the same. I read thorugh the book and was equally impressed with it. So if you haven’t already read them, I would strongly recommend taking a good hard look at both.

EDIT [13th May, 2008]: Another one released today. GPU Gems 2.

STL is not slow.

Recently I was having a conversation with some former colleagues of mine and I got a feeling that most of them were of the opinion that STL was slow and/or inefficient . If you think there is truth to this then let me assure you it’s not the case, not at all. STL is used by so many people and so many libraries that it, in fact, is probably the most optimized piece of code there is. This misconception is actually a result of inappropriate use of STL library and not because STL is inefficient. It maybe true that different versions of STL may have different speeds and I have heard that MS STL is a little slower than others like STLport, but I have no data to either prove or disprove this. I never use MS STL even while programming under Visual Studio (find out why), so I can’t really say.

As I said earlier, the perception that STL is somehow slow and/or inefficient stems from the fact that programmers generally tend to abuse STL containers by not using correct ones. STL has different containers and each is specifically designed to address a particular problem. I don’t want to get into which container to select when, I think Scott Myeres has done a far better job that I ever can. If you haven’t read his book then you better get down to it right now. It clearly outlines how one should go about using STL and addresses several subtitles involved in correct container selection.

Too busy to write?

Sorry, but I have been a little busy for the past two weeks. Too busy to blog I guess. I have been aiming for a code freeze on the Doofus game and it’s been hard work getting all the bugs and issues in. I’m going for the final push this time around to get at least the coding issues out of the way. The good thing is there isn’t too much left on the coding side, so I may be able to push out another beta by next week. Hopefully it will be the last and final beta before (; wait, don’t get your hopes up) at least one release candidate before Doofus goes gold. I guess there are still a sizable amount to levels to be completed.

Unlike release cycles of other software, Doofus game release cycles are a little bit different. I devised a new method after we initially found the old method to be rather monolithic for this particular project, and because of obvious constraints we have as a small team (unavailability of testers at specific times and things of that nature). Traditionally, you have a set of alpha releases of a product where each alpha release is tested in-house by both the developers and/or testers. Bugs are filed for specific releases and fixed during bug-fixing stage, whenever that maybe (, generally differs from project to project). Beta releases are pushed out when alpha releases get stable enough for “general consumption”. Beta releases are generally widely accepted as “almost complete” versions of the product. So a beta release often signifies a “feature-freeze” of the product. A bug fixed beta release can become a Release Candidate if the dev team feels confident enough which eventually turns Gold when everyone is confident enough.

In the case of the Doofus game, things are a little bit different. A beta release signifies a “feature freeze” for “a particular set of” features. Let me explain. When we started developing the game, the O2 Engine was the first to come (, before we started on the actual game code). The name “O2 engine” comes from the repository branch of my older game that was never released because it had too many flaws! I guess a lot was carried over including primitive libraries and some design decisions and implementations. Anyways, since the new project was a bit complex and our testing team small and working part-time, we decided to have specific release milestones having only limited set of completely complete features. When I say completely complete I mean “feature frozen”. Each beta release addressed different features. The first was for engine integration with geometry. The level structure was finalized and resource management was put in place. The first release looked really ugly because the renderer was partially finished.

The second beta addressed collision systems, basic gameplay things like triggers activators and integrations with third-party modules and libraries. The third was for rendering sub-systems, when those screenshots were posted. This release, the fourth, will be for AI (NPC) and Physics and that marks the end of the game features. The beta still has to go through a bug fixing stage before I am confident enough to even look at a RC, but it does mark and end to any major code modifications to the game code. Many would say the betas are actually alphas, but there are 2 reasons I call them beta releases. a) They are feature freeze releases. No features are added or removed to the already tested features. b) Our testers are no full-fledged project member so white-box testing responsibility falls on the dev team, mostly. That said, the beta testers are not just kids beating at the keyboard and have been instrumental in testing the product.

I guess this release has got me a bit exited, and I am working on the website/s at the same time. I have actually started on quite a few blog posts in the past week, but haven’t had the time to polish and/or finish them yet. Maybe this week will see more posts on the blog, I hope.

No more T-Junctions.

I must confess, my original post on optimization of game levels was, well, incomplete and inaccurate. Optimizations were not fully complete. There were a lot of T-Junctions that were left behind after optimization (, Sandeep was probably the only person to catch that). However, I managed to remove those too. They were causing a lot of problems with A* navigation and I am glad they are gone! So here are the updated screens. Some extra level geometry has been added so the screens might not look exactly the same as the earlier post.


The updated scene (Doofus 3D).


T-Junctions removed.

Optimizations on game levels.

Just an update on the Doofus game and on what I have been working on for the past couple of weeks. The past couple of weeks have seen me seriously working at getting the triangle count down in the game levels. The tri count had been increasing steadily for the last few levels and it just started hitting on the FPS value real bad. That is why I had no option but to go for Triangle decimation. The amount of triangles for even moderately complex levels started turning out to be surprisingly high and most triangles were all but useless. The reason; Doofus 3D levels use brush based geometry and the tris are a result of successive CSG (Constructive Solid Geometry) splits. The more detail I added to the levels, the more redundant splits occurred with the brushes. Meaning the FPS started falling like a rock for arbitrarily complex levels.

The optimization technique I was working on reduces the number of triangles by a) Removing redundant vertices and b) Collapsing unwanted edges. Simple right, not quite. Triangle decimation turned out to be somewhat more complex than I had anticipated. Fortunately and after some real hard brainstorming I managed to get it working just as I wanted it to. Now in some situations the triangle count reduces to as much as 4%. But an average value of around 10 to 20 % is what I usually get. That is also quite significant to say the least. Thank God my effort has not been in vain after all. It was a real pain to get that working correctly. Check out the images below to actually see the optimizations at work.

Original scene.
A smaple Doofus 3D scene

Triangles in the unoptimized version (click to enlarge)
Triangles in the scene before optimization.

Triangles in the optimized version (click to enlarge)
Triangles in the same scene after optimization.

I have also been working on completing the AI. Sorry but I don’t have screens for those, maybe the next time. The AI still needs some amount of tweaking to get things working perfectly. I am not saying too much at this point in time; maybe in one of my next posts I will get into more details. Hopefully I can finish this last pending thing in the game soon.

Code::Blocks 8.02 has been released.

The much awaited release of Code::Blocks has finally happened. I was waiting for this for like forever! It’s time to go get the new IDE. From the initial looks of it, it’s been worth the wait. More later!

Into the mind of a casual gamer.

I haven’t said or ranted about gaming in quite a while now. Quite unlike me. I often have to say and rant a lot, especially about gaming. If you read the last 10 posts you would never guess I was a game developer. So I guess it’s time to put the Blog back on track! The absence of any game related post have been for a couple of reasons. For one, I haven’t installed any new games as of now. Too busy with finishing off Doofus 3D. Neither have I upgraded my old aging NVIDIA 6200 and 6600 cards. So anything new like Bioshock or Crysis is out! I have promised myself a brand new 8800 GTS or something newer as soon as I complete and ship the game, and that’s been now like, forever! On the contrary, for the past 6-8 months I have had the opportunity of actually watching others playing games and it’s been a learning experience as a game designer. You wouldn’t believe it, one of them happens to be my mother. She has been doing all the playing while I have been working my ass off to finish my game, though I must say her taste is very (very) different from mine. No Half Life-s or Oblivion-s for her, she kinda enjoys the Zuma-s and the Insaniquarium-s; casual games in general.

Just watching and being a spectator can teach you volumes about how people enjoy computer games. Casual gamers in many ways can very different from hardcore counterparts and can be very similar in other respects. A casual gamer is generally not interested in complex gameplay. They are the ones who enjoy simple yet immersive games. Simple puzzles and not overly complex challenges are ones that are favored over long story lines and dark alleys. Also appreciated are short levels. For instance, I tried selling a game called Caesar III to my mother but nope, didn’t work. For those who don’t know, Caesar III was a city building game, kinda like Sim City and I loved that game when I played it. The game has no gore, so I was pretty sure she would love it. However, the comments I got from her were, “Too complex”, “Very difficult to manage”, “Can’t understand it”. It just goes to show casual gamers tend to favour easier set of “game-rules”. Anything that is more complex than a few set of keystrokes is a big NO.

Over the past few months I observed and chatted with a few other casual gamers too. Mostly friends and relatives who often play casual games. It’s always good to understand the market mentality, especially if you are developing a product catering to them. Most casual gamers I chatted with have played 2D games, but quite a few I must say, have played 3D games. Interesting, since I am developing a 3D casual game myself. When asked if they would be interested in such a game, the overwhelming response was, Yes, to obviously a simple 3D game. Most of them said they would be very interested, but were unsure if they would like it. Some even told me they wouldn’t mind learning a few extra keys to play something like Marble Blast. So it means Doofus 3D does have a market after all, but it remains to be seen if it would indeed translates into sales.

In general I have found casual gamers dislike having to learn a lot of rules, period! Games that involve significant learning effort quickly become unpopular. So I guess games like Oblivion are out. Also not so much appreciated is, gore. There is an interesting paradox here, unexpected I must say; many casual players admitted to playing arena games like Quake III. The reason; such games was easy and fast paced. Most enjoyed it, some said the games were violent but they generally didn’t mind. None of the casual gamers I talked to knew exactly what game genres were, none cared. Most said they looked at the game cover or screenshots to decide if they would indeed play a game. 90% said they played flash games. About half said they don’t understand things like DirectX or OpenGL or h/w acceleration. Nearly all said they had only on-board graphic cards. More than half said they didn’t even know you could install separate graphics cards. More than 75% said they had and do play games on laptops.

OK, my survey is far from being completely accurate. All I have done is maybe asked a few questions to people who I have met in the past year or so. I have in no way used a proper methodical approach for the survey. However, it does give you an insight into the mind of a casual gamer. As a game developer I generally tend to forget these important lessons and sometimes do go overboard. Then, I usually drag myself back to reality!

Invasion of the multi-core machines.

They are here. They were already here, but now they are really here and can no longer be ignored. Developers and programmers, especially game developers, can no longer afford to sit back and just watch the invasion of machines with multi-core processors. While hardware manufacturers have scaled their processes to bring them to us, software and compilers haven’t scaled equally well. Ironically with current programming methodologies, programming practices and compilers, programmers can’t yet take full advantage of the CPU power thrown at them. It’s not entirely the programmer’s fault, and neither are they limited by their intelligence. The fact is current generation programming models fall short of addressing the multi-core issue in a reliable way. Yes currently there are workarounds and you can definitely use them to get some advantage on a multi-core machine. Applications using even simple threads can benefit from multiple cores, but merely having multiple threads in an application doesn’t mean that application or the threads will run at twice the speed for a dual core CPU. The performance boost for a “normal” multi-threaded application running on a multi-core system will be rather minuscule compared to the computing power of what a multi-core system provides. To be frank all those cores are of little use from the “average joe” programmer’s perspective. That’s just because, if not programmed with care, the advantages provided by those cores is useless, at least in one given application.

Lets look it from a more technical perspective. Most multi threaded applications are not written to use multiple concurrent threads aggressively. A typical multi-threaded application delegates only a minuscule portion of it’s program code to a thread, often called a worker thread. Generally the most CPU intensive or blocking operations in a program are done inside this thread. Consider an example of a web-downloader application. You have one worker thread doing all the downloading while the UI thread waits and processes user input like “Cancel” or “Quit”. Here the program was explicitly designed this way so that the UI can respond to user input, while at the same time the task of downloading a file can go on. In other situations threads may be used for a different purpose. Take the case of the Doofus game. In the game shadow calculations are the most CPU intensive operations, but are not required per frame (, or cycle). So the shadow calculations are done inside a thread at lower priority, typically the calculations are distributed so that the results are obtained per 2 or 3 frames. Whatever the case maybe, the fact remains, such designs are not the most optimal way to program for multi-core machines. In the case of the web-downloader application, one thread waits while one thread does all the work. In the case of the game the situation is a little bit better, but still the task is not optimally distributed between threads. The ideal case would be to have multiple threads share the entire workload of the application so that all the threads are busy all the time. If that were indeed the case and if these threads were to run on separate cores, you would then be able to harness the true power of a multi core machine.

Programming multiple concurrent threads in an application is difficult. Thread synchronization is not for the faint hearted and having a lot of threads in an application can create bugs that are difficult to debug. Traditional multi-threading is synchronized using lock-based methods and lock-based synchronization is prone to problems like deadlocks, livelocks and race conditions. Experienced programmers will try to avoid multi-threading if and when they can and try for simpler and often single threaded solutions. This is not what is advocated by concurrent computing and parallel programming, which clearly can take advantage of multiple cores very effectively. It is true that even with current multi-threaded designs you could benefit from multi-core architecture. Even if such programs internally wouldn’t be able to use the power of multiple cores, the OS can still make full use of the architecture for multi-tasking. To put it in simpler language, multiple applications running at one time will run faster (, note the subtle difference there). For a typical system, applications like anti-virus programs and background processes along with other user applications, running all at once, will definitely benefit from additional cores. This however isn’t very helpful from a game development perspective, since obviously most games are single applications. Games typically take up most of the system resources while running and the advantages of multi-tasking are all but useless while running a game. Games therefore must be able to harness the true power of multiple cores internally. What does that mean? Does it mean a paradigm shift in how games are built? Do we need a special language to do so? Should we shun the current programming practices? Drop C/C++ and look at maybe Erlang or Haskell? Use parallel programming concepts? The questions are many and are increasingly asked by a lot of people. The truth and the solution however, is not quite so simple.

Continue reading

An Unreal Crysis.

If you are graphics geek and love to see those so called next-gen effects, then recently released games like Crysis, UT3 and to some extent Bioshock will give you lot to cheer about. Crysis for one has shown that modern top line cards can push extraordinary amounts of detail. However, raw figures show that Crysis and UT3 sales have been anything but extraordinary. They have in fact fallen flat! Interesting figures there, and to some extent I am a bit surprised by what the figures show. As the articles point out both games were pretty hyped out before the release and they should have made flat out more sales than what the did. True Crysis has some crazy hardware requirements, but still the game can be played with older and less powerful graphics cards, so can UT3. Maybe not with all the graphics effects and resolution maxed out, but they can be played nevertheless. Besides both games have *huge* fan bases so the figures are very surprising indeed.

Well I can’t speak for everyone but, my personal take on the whole thing is the fact that vanilla FPS genre is kinda getting old. After so many games that churn out the same mundane gameplay, it has pretty much lost it’s charm. True the graphics have improved but not the gameplay in general. Games like Bioshock stand apart from the crowd because they give that little bit more to the overall game and it is exactly why they sell more. I can tell you from my experience over that years of playing games is the fact that (, and I have pretty much repeated this a lot of times on this blog,) FPS games are getting kinda boring. As a gamer I want more interesting stuff in there. That is exactly the reason I spent nearly 6 months playing Oblivion. The game gave me so much more to do than just run kill, run kill, collect ammo, run kill, collect health, run kill …..

I myself haven’t played UT3 and for that matter only watch someone else play Crysis, but from what I have heard people say about the games makes me wonder if they are nothing more than tech demos. Maybe we should look at it from a different perspective; it’s a fact Epic markets it’s engines via the UTx games, and I think to some extent Crytek does that too. So maybe that is exactly why those game are here for, to show off what their respective engines can achieve. The graphic brilliance achieved by both games/engines is amazing, there is little doubt to that, and the hardware requirements for the games is equally demanding. But that is for now. The same hardware will become mainstream in another 6 to 8 months and the same engines can be used/licensed to make other games. I therefore wouldn’t count them as outright failures.

Different people have different tastes and different points of view, so naturally have different tastes for game genres. However the feeling I get is, in general, game genres are beginning to overlap. This I think that is because of necessity. Game designers that strive to make their games “immersive” have started incorporating ideas and methods from other game genres to make gameplay more interesting and challenging. However having an equally good engine is a must. Case and point to Oblivion. The game looks great because it uses Gamebryo, which is another good engine. I am pretty sure we will see more and better games using both the engines in the future.