I've recently studied a bit of OpenGL and DirectX and I now want to develop something.

That book (Programming an RTS Game with Direct3D) has good reviews and an RTS would be so nice to build

It's 10 years old, so I would have to update codes and use D3D11/GL3.3 at least, but is it worth?

I've recently studied a bit of OpenGL and DirectX and I now want to develop something.
That book (Programming an RTS Game with Direct3D) has good reviews and an RTS would be so nice to build
It's 10 years old, so I would have to update codes and use D3D11/GL3.3 at least, but is it worth?

That's a very unspecific, generic question. Sure it's worth it if you think it is. If you're looking to just get experience developing a game then yes it's worth it. If your sole purpose is to develop the next big-RTS, you'll need to have a solid plan with an idea on what would separate your RTS from every other RTS. If you don't have this plan and if you don't think that your RTS will be the next big thing, then maybe it wouldn't be worth it. On top of that you'll need to decide whether or not you have the knowledge to develop such a game. You need to determine what your purpose is to develop the game and from there you can determine whether it will be worth it.

I have a very dog-eared copy of that book sitting on my shelf. I thought it was fantastic when I got it, almost ten years ago, and I still think it's pretty good.

It's one of the few game development books I've found that works through the process of building out a non-trivial game, piece by piece. Granted, at the end of the book, you're going to have an unpolished, simple, early-2000s 3D Warcraft clone, but I think it fills in a lot of gaps between what you learn in the straight OpenGL/Direct3D graphics books, and the fairly abstract game design, Game Programming Gems-style topics books.

If nothing else, I think the chapter on random terrain generation is pretty nifty.

That being said, trying to update it to DirectX 11 is not a trivial endeavor. It makes use of a lot of the D3DX stuff that no longer exists or has no direct analog in D3D11, which can be problematic, particularly for the model animation.

It's 10 years old, so I would have to update codes and use D3D11/GL3.3 at least, but is it worth?

That being said, trying to update it to DirectX 11 is not a trivial endeavor. It makes use of a lot of the D3DX stuff that no longer exists or has no direct analog in D3D11, which can be problematic, particularly for the model animation.

...and older versions of DirectX in fact work just fine on newer versions of Windows. Going through the effort of updating old code to work with the newest version of DirectX may not be worth the effort, and being able to work on Windows XP (which IS still commonly used, almost as common as Windows 10) might be a good reason not to.
And if you're going to go through the effort of updating the code to DirectX11, you might as well just work with DirectX 12 instead, since low level GPU programming seems to be the future of 3D graphics.

Only thing you have to worry about with old books is faded text. Any book that teaches you something is always worth reading and working out of. The only time I would recommend being picky is for a book trying to teach a language, otherwise books, both good and bad, can teach you something about game development.

So do I buy it or not?

This question is too short, I understand, but if I buy it then I wanna write some code and that promised game. But if you say "don't update the code with newer APIs", what's left?

Left would be just the design; I could instead buy a game design book and code everything from scratch

I generally have a really hard time finding a way to recommend technical books like this, ones that cover very specific topics or focus on specific implementations or examples. They tend to get outdated very quickly, and so tend not to make a very good investment.

I don't know much about this book specifically, but when I follow that Amazon link it's still setting for $40+ and for a book from 2006 that strikes me as a really bad deal. The book doesn't have the "see inside" feature enabled, so unless you can find a copy at a local bookstore to browse through and see if it's useful, I would recommend against buying it. It's true that the state-of-the-art in RTS design hasn't advanced much since 2006 since the genre is in a lull right now, but I'm fairly certain you can make one without having to sink $40 it. Put that money towards something else Maybe stick it in the bank and save it for buying art assets.

No way I can have any chance to have a quick look at it; it seems only ericrrichards22 has a copy of it of all the above posters

No way I can have any chance to have a quick look at it; it seems only ericrrichards22 has a copy of it of all the above posters

There's one in a city not far from mine, but they have no books on compute science