Learn OpenGL and Direct3D at the same time

Started by
9 comments, last by n00by2077 11 years, 3 months ago

It's easy to learn OpenGL and Direct3D simultaneously? I mean learning each way of make things, and see in which way there must to be used in each API, any book/tutorial out there use this kind of learning method?

Advertisement

I'd say it's not even easy learning one of them at a time for most people, at least in the beginning phase.

If you have some basic understanding of what you'll have to do, and maybe a bit programming experience with libraries that are similiar, it might be okay. But i guess the frustration level is much higher, not only because you have longer phases of coding without seeing a result.

Also, the setup for both APIs is (in my opinion) quite different, so i'd suggest you start with one (=OpenGL) and then, when you're feeling comfortable with it, try something in D3D (or stick to OpenGL :)).

I have background in programming and using libraries, but I have not in the graphics ones, I was just thinking this dual-learning as a way to save time in future, in place of learning the same things two times, learn them same time with a bit more of effort.

I have background in programming and using libraries, but I have not in the graphics ones, I was just thinking this dual-learning as a way to save time in future, in place of learning the same things two times, learn them same time with a bit more of effort.

You could, instead, learn to use another API's that performs the underlying 3D manipulation for you. Ogre3d and Irrlicht come immediately to mind.

My Gamedev Journal: 2D Game Making, the Easy Way

---(Old Blog, still has good info): 2dGameMaking
-----
"No one ever posts on that message board; it's too crowded." - Yoga Berra (sorta)

[quote name='n00by2077' timestamp='1358181353' post='5021440']
It's easy to learn OpenGL and Direct3D simultaneously?
[/quote]

No.

Well, okay, "easy" is subjective to the learner's intelligence, experience, and time constraints. But learning one will include learning the more abstract "programmable pipeline" concepts (as well as shader programming concepts), which will make learning the second less of a chore. Much like learning to program in one language shortens the time it'll take you to be competent in another language later on, because you're familiar with the common language-unbound programming concepts.

Hazard Pay :: FPS/RTS in SharpDX (gathering dust, retained for... historical purposes)
DeviantArt :: Because right-brain needs love too (also pretty neglected these days)

I would say it's a very good idea. Learning 3D API's are mostly about learning the 3D theory, and then finding the correct way to implement it. You will find out that OpenGL and DirectX have a different vision, but at the end there's an exact equivalent for each concept in both API. Of course it will take a bit more time, but it may help clarifying thing up as you go.

I find weeska point of view a bit limited, why suggest to stick to one API when the question is clearly about learning both. You will find OpenGL faster to start with, while DirectX require creating lot of object with lot of flags/parameters who seems useless at first, but at the long run DirectX is easier to use in a complete well-made graphic engine than OpenGL.

I think learning both simultaneously can become confusing in that you would have two (or sometimes several) different ways to do things at the lowest level and when they're still fresh in your mind, you might mix some of the conventions.

However, I first started 3D with OpenGL, gave up on trying the more complex things, and a while later gave DirectX a chance. I got farther with it than OpenGL, but doing some things still felt like a chore because it was still pretty low level. Next, I moved on to XNA and learned a lot from tutorials, was able to write more complex rendering systems while not getting too bogged down in the language's own obstacles.

So at least to me, I wasn't able to break down complex systems and concepts until I was first able to do them at a higher level. Now I would probably feel more comfortable going back to DirectX to replicate what I've already done.

New game in progress: Project SeedWorld

My development blog: Electronic Meteor

I find weeska point of view a bit limited, why suggest to stick to one API when the question is clearly about learning both.

That one was with a bit of irony. Sure the opener can learn both, but i think it's better to learn them one after another. In any order ;)

I know this feeling when you want to learn everything instantly, but i've made the experience that things get out of control very fast this way. It's important why one wants to learn both APIs at the same time.

If it's for game programming, i don't think it'll give any benefit in the short run.

If it's out of technical curiosity, it's worth a try.

If it's about learning graphical theory, there are more things to learn than an API, so i'd go with one of them.

Also, it's important, it's never useless to learn OpenGL/D3D in the first place and later switching entirely to the other one, because the concepts used are the things that matter most and require the most efford in learning.

I would recommend that if you intend to truly learn the underlying technologies of both of them you would want to do them separately. When learning programming languages and libraries "quicker" normally means you don't actually learn it correctly. That is to say that the faster you try to rush through the learning process the more likely you are to get an overall idea of what you are working on but not that you actually "learn" what is going on. If you take your time and learn each separately you will be exposed to many of the same techniques such as programmable pipelines and shaders twice. This effectively doubles your experience which in turn gives you a much greater understanding of what you are doing.

With that said it may not be all THAT imperative that you learn all of the technical's of either. If you are simply learning the basis of both API's for the purpose of knowing what your engine is doing behind the scenes than by all means skim through the both at the same time. I am always a big advocate of expanding ones knowledge and exposing oneself to many different technologies for the greater understanding of what the computer is actually doing. You would be amazed how much better you will be using other people's engines when you have a decent idea of what they are doing at the hardware API level.

I guess the real question you might want to ask is of yourself an is "Do I want to make engines or games?" Although this is a completely different topic in and of itself it's a good factor as to how much time you will want to invest in actually learning the API's. If your answer is "I want to make games" or "I want to make the engine to power this one game" I suggest learn both at the same time, get the basic idea down and use someone else's engine. If you have experience using other people's engines and have completed at least a couple demo's of game ideas that you have attempted and your more interested in working on a new engine either for personal use or public release than my suggestion would be to take your time. Learn each separately and maybe even go back a third time to study them both over again at the same time. To make a good engine you really want to have a very high understanding and knowledge of the underlying API's you don't want to rush it just so you can get to work.

Dan Mayor

Professional Programmer & Hobbyist Game Developer

Seeking team for indie development opportunities, see my classifieds post

Your ability to do much in either API depends on your domain knowledge (3D graphics) and your overall programming level. If you already know a lot about computers processing 3D graphics, and have used large and complex APIs before, then learning both APIs at the same time will probably not be a difficult task. If you're a beginner at either, and certainly if you're just beginning both, don't try to learn both at the same time, because learning one will be enough a challenge.

This topic is closed to new replies.

Advertisement