We found SCons to be very slow - we're talking 2-3 minutes of churning for it to decide that everything is up-to-date. Boost.Jam on the same set of files is around 5-10 secs.
The class wizard gets very unhappy if you add functions with telling it. It is possible if you spend the time to learn all the little comments it wants to embed with your code but it is very painful.
Apps written in this way tend to end up with pretty much all of the code in a FooDoc.cpp or FooView.cpp since it is generally easier to let the class wizard have its way than fight against it.
We use Accurev for our source control - the GUI is a little rough around the edges but the source control system scales brilliantly - it is definately worth checking out. It isn't free though.
Prior to that we were using CVS for source control under windows with no problems. The CVS server was hosted on a Linux machine - but then so is our Accurev server.
As for build systems, we were using cygwin/make with no real issues but have since switched over to Boost.Jam which has proved really successful. We have some rules that general.vcproj files that you can use with.NET since most of our developers work within that and like being able to use the code browser and stuff like that. Compilation is still all controlled by Jam though.
We used SCons for a while, but found it to be incredibly slow for big projects. This was about a year ago now so this may have been improved.
With Jam and SCons though, just like with make, you'll need to spend quite a bit of effort getting your build system setup. The results are well worth it - maintaining lots of.NET project files just wasn't feasible for us, especially since we need to compile for several different platforms.
Also, if you want to do GUI stuff I can't recommend Qt enough. I suppose MFC works, but I find it really alien, especially if you're used to more Unixy style development.
Yeah, but he's done that and made RPMs for it. But don't forget that anything where you mention PS2 and Linux seems to instantly create some kind of excitement on this site.
I agree totally - however, some people do seem to have been tempted by these sort of things in the past. Isn't this somewhat similar to what Lionhead were doing a few years back with their satellite studios?
I suppose it would free up development teams from having to drag around all the extra weight of the executive type people who have to deal with publishers, get funding etc.
I'd imagine that what carbon6 are hoping to do is to tempt entire teams away from whatever studio / publisher they are currently working for and setting up on their own.
Also there may be teams around who have great coders / artists but need a good solid design to work with - which is something I think carbon6 are offering. (I sure wouldn't like to work in that sort of environment though!)
Before getting too excited, check out their website - they are looking for development teams, not individuals. Sending in just your CV probably isn't going to get you very far!
However, there is actually in interesting story here - the whole idea of there being a third party connecting the money (publishers) with the developers, but controlling the design and management on the project seems to be quite different to how things currently work in the games industry.
I spent a while trying to convince Maya to support the particular type of higher-order surfaces we needed for our game. Due to bugs in the Maya API though this couldn't be done - so we decided to write our own modeller.
The/Zc:forscope option tells the Visual C++ compiler to follow the standard C++ definition for the scope of variables declared in a for loop.
Is this true? We've been getting desperately annoyed trying to write code that runs on both VC and gcc by this. Unfortunately I can find no reference to this in the docs anywhere. Can anyone confirm this?
Although I generally agree with you, I have to say that I laughed out loud at 'The programs are utterly immense, yet need an interface free of ad-hoc additions' when talking about Max. That's Max the poly-modeller, Max the broken-bezier-patch-modeller and Max the NURBS modeller.
As far as clean implementations go, I haven't heard of any better than Maya's.
However, there is room, and the possibility of, an opensource 3d MODELLER at some point. This is where packages get it wrong - the idea that a 3d package needs to be an texturer, animator and a renderer all at the same time. I'd do it myself, but I'm too busy writing Maya plugins to try and fix their broken UI.
I had a look at blender recently with the idea of suggesting that some of the artists at work evaluate it.
I didn't get much beyond the broken navigation. The view rotation doesn't keep a sense of what 'up' is -- thus making it virtually useless for modelling anything that is going to exist in an environment where there's gravity.
I'm starting to wonder whether or not there are actually two versions of Nascar. The first is the version that reviewers are sent resulting in comments like 'good looking racing sim with excellent physics' and 'the graphics are very clean, with good downfield visibility'.
The second version is the one that I have played. It is appalling! Probably no more than 2mb of textures per frame, with one small texture for track another for grass and another for fence. The sky looks airbrushed. The game switches to low-res mode if you turn on the rear-view mirror or more than one car is on screen.
The physics are laughable, playing more like a PSX game.
So, can anyone explain why a game that is so obviously bad is getting good reviews?
Umm....the X-box is going for the console games market not the PC games market. However, many console games are also ported to the PC. Think Wipeout, XG2, Pod Racer.
Since all console companies act as monopolies anyway, I think Microsoft can be forgiven for being another monopolistic company getting into that market.
The reality is that the X-Box is interesting to developers because it is PC based and supports DX8. This will make porting games from the PC very easy -- and this is what gives it a huge advantage over anything else.
Going slightly offtopic, but at ECTS most games on the-console-that-the-press-wants-to-bury were looking as good as, if not better than, their PS2 equivelants.
But this was because the company was trying to sell themselves to me and wanted to tell me which contracts they had etc. They couldn't have told me this if I wasn't under NDA and I'd have to sign one if I took the job so I was happy to sign it in the interview.
They were however perfectly happy for me not to sign it -- they just wouldn't have told me some of the stuff that they did.
Slashdot Mirror
Well if you're in NSW the weather today was lovely! Not all that wintery at all...
(dammit, I still spent most of saturday in front of my computer though)
We found SCons to be very slow - we're talking 2-3 minutes of churning for it to decide that everything is up-to-date. Boost.Jam on the same set of files is around 5-10 secs.
The class wizard gets very unhappy if you add functions with telling it. It is possible if you spend the time to learn all the little comments it wants to embed with your code but it is very painful.
Apps written in this way tend to end up with pretty much all of the code in a FooDoc.cpp or FooView.cpp since it is generally easier to let the class wizard have its way than fight against it.
We use Accurev for our source control - the GUI is a little rough around the edges but the source control system scales brilliantly - it is definately worth checking out. It isn't free though.
.vcproj files that you can use with .NET since most of our developers work within that and like being able to use the code browser and stuff like that. Compilation is still all controlled by Jam though.
.NET project files just wasn't feasible for us, especially since we need to compile for several different platforms.
Prior to that we were using CVS for source control under windows with no problems. The CVS server was hosted on a Linux machine - but then so is our Accurev server.
As for build systems, we were using cygwin/make with no real issues but have since switched over to Boost.Jam which has proved really successful. We have some rules that general
We used SCons for a while, but found it to be incredibly slow for big projects. This was about a year ago now so this may have been improved.
With Jam and SCons though, just like with make, you'll need to spend quite a bit of effort getting your build system setup. The results are well worth it - maintaining lots of
Also, if you want to do GUI stuff I can't recommend Qt enough. I suppose MFC works, but I find it really alien, especially if you're used to more Unixy style development.
They also get a mention in this article - which seems tailored towards a slightly different type of reader.
The view examples on the unicast site...it sat there downloading the commercial and got stuck on 99%.
Yeah, but he's done that and made RPMs for it. But don't forget that anything where you mention PS2 and Linux seems to instantly create some kind of excitement on this site.
Indeed it does!!!
However, it was so well hidden behind the flash interface it took me about two minutes to find it even after you'd pointed it out.
Oh well. I'll get me coat.
I agree totally - however, some people do seem to have been tempted by these sort of things in the past. Isn't this somewhat similar to what Lionhead were doing a few years back with their satellite studios?
I suppose it would free up development teams from having to drag around all the extra weight of the executive type people who have to deal with publishers, get funding etc.
I'd imagine that what carbon6 are hoping to do is to tempt entire teams away from whatever studio / publisher they are currently working for and setting up on their own.
Also there may be teams around who have great coders / artists but need a good solid design to work with - which is something I think carbon6 are offering. (I sure wouldn't like to work in that sort of environment though!)
Before getting too excited, check out their website - they are looking for development teams, not individuals. Sending in just your CV probably isn't going to get you very far!
However, there is actually in interesting story here - the whole idea of there being a third party connecting the money (publishers) with the developers, but controlling the design and management on the project seems to be quite different to how things currently work in the games industry.
See the Gamasutra postmortem for more details.
I spent a while trying to convince Maya to support the particular type of higher-order surfaces we needed for our game. Due to bugs in the Maya API though this couldn't be done - so we decided to write our own modeller.
...working out what the heck the image means is much harder.
:-)
(getting the image took me about 10 mins in emacs)
I understand the pictures
Looking at their paper, I guess the rest of it is explaining mathmatical principles and stuff like that.
The /Zc:forscope option tells the Visual C++ compiler to follow the standard C++ definition for the scope of variables declared in a for loop.
Is this true? We've been getting desperately annoyed trying to write code that runs on both VC and gcc by this. Unfortunately I can find no reference to this in the docs anywhere. Can anyone confirm this?
Although I generally agree with you, I have to say that I laughed out loud at 'The programs are utterly immense, yet need an interface free of ad-hoc additions' when talking about Max. That's Max the poly-modeller, Max the broken-bezier-patch-modeller and Max the NURBS modeller.
As far as clean implementations go, I haven't heard of any better than Maya's.
However, there is room, and the possibility of, an opensource 3d MODELLER at some point. This is where packages get it wrong - the idea that a 3d package needs to be an texturer, animator and a renderer all at the same time. I'd do it myself, but I'm too busy writing Maya plugins to try and fix their broken UI.
I had a look at blender recently with the idea of suggesting that some of the artists at work evaluate it.
I didn't get much beyond the broken navigation. The view rotation doesn't keep a sense of what 'up' is -- thus making it virtually useless for modelling anything that is going to exist in an environment where there's gravity.
I'm starting to wonder whether or not there are actually two versions of Nascar. The first is the version that reviewers are sent resulting in comments like 'good looking racing sim with excellent physics' and 'the graphics are very clean, with good downfield visibility'.
The second version is the one that I have played. It is appalling! Probably no more than 2mb of textures per frame, with one small texture for track another for grass and another for fence. The sky looks airbrushed. The game switches to low-res mode if you turn on the rear-view mirror or more than one car is on screen.
The physics are laughable, playing more like a PSX game.
So, can anyone explain why a game that is so obviously bad is getting good reviews?
Yup, that's right.
IAAPS2D (I Am A PS2 Developer)
Sounds like your monitor can't cope. Does it work under windoze?
Errr....I work for a company listed in the X-Box developers posted earlier. Where do you get your information from?
Umm....the X-box is going for the console games market not the PC games market. However, many console games are also ported to the PC. Think Wipeout, XG2, Pod Racer.
PS2: 3 extra processors. You forgot the IOP -- unless you don't want any CD or controller access!
Since all console companies act as monopolies anyway, I think Microsoft can be forgiven for being another monopolistic company getting into that market.
The reality is that the X-Box is interesting to developers because it is PC based and supports DX8. This will make porting games from the PC very easy -- and this is what gives it a huge advantage over anything else.
Going slightly offtopic, but at ECTS most games on the-console-that-the-press-wants-to-bury were looking as good as, if not better than, their PS2 equivelants.
But this was because the company was trying to sell themselves to me and wanted to tell me which contracts they had etc. They couldn't have told me this if I wasn't under NDA and I'd have to sign one if I took the job so I was happy to sign it in the interview.
They were however perfectly happy for me not to sign it -- they just wouldn't have told me some of the stuff that they did.