On Monday, 28 January 2013 at 12:04:42 UTC, Szymon wrote:
> On Saturday, 26 January 2013 at 21:30:22 UTC, mist wrote:
>> On Saturday, 26 January 2013 at 20:54:30 UTC, FG wrote:
>>> On 2013-01-26 20:55, SomeDude wrote:
> We actually are video game company ;) That is why D seems like perfect choice for future language. Native code + superb programmer productivity + fast compile times. If D was bit more mature decision would be dead simple.

Well, let's hope it becomes mature enough. My impression is that it needs to clean up itself a bit, to remove several ambiguities (that @property discussion...) that lasted far too long to bring any good, and to work heavily on tools: the compiler.

Otherwise, I think it is OK to start with it. Custom runtimes/libraries could provide GC-free implementation where needed (BSPs, games).

Not wanting to sadden anyone, but the main effort in tools should be, my feeling, directed at gdc/gcc. It will help a lot to having D a good-quality "gcc-standard" language.

Then, Eclipse/DDT with debugging and what else it is missing.

Then... happy coding.

On Friday, 25 January 2013 at 21:39:58 UTC, Namespace wrote:
> On Friday, 25 January 2013 at 20:45:22 UTC, Szymon wrote:
>> Hi,
>>
>> I would really like to start using D in our small company as a
>> C++ replacement. With that in mind I do have few questions:
>>
>> 1) Is D2 really ready for production code?
>
> Not really. A big pain in the ass is the missing rvalue ref that C++ has. So using structs is a big disaster. But there are plenty of other missing features.

Java has far less features than D, including what you complain about, yet it is considered a pretty productive programming language...

On Monday, 28 January 2013 at 15:15:38 UTC, Dejan Lekic wrote:
> On Friday, 25 January 2013 at 21:39:58 UTC, Namespace wrote:
>> On Friday, 25 January 2013 at 20:45:22 UTC, Szymon wrote:
>>> Hi,
>>>
>>> I would really like to start using D in our small company as a
>>> C++ replacement. With that in mind I do have few questions:
>>>
>>> 1) Is D2 really ready for production code?
>>
>> Not really. A big pain in the ass is the missing rvalue ref that C++ has. So using structs is a big disaster. But there are plenty of other missing features.
>
> Java has far less features than D, including what you complain about, yet it is considered a pretty productive programming language...

I hate to programming in Java because of the many many missing features.
And because _all_ is stored on the heap. Java is IMO only good (if any) for learning programming but nothing else.
But don't let us start a flame war about programming languages. ;)

On Monday, 28 January 2013 at 15:15:38 UTC, Dejan Lekic wrote:
> On Friday, 25 January 2013 at 21:39:58 UTC, Namespace wrote:
>> On Friday, 25 January 2013 at 20:45:22 UTC, Szymon wrote:
> Java has far less features than D, including what you complain about, yet it is considered a pretty productive programming language...

I think it is no more about bringing in new features. It is just about ironing out and clarifying existing ones.

People started complaining less and less about features (well, exceptions exist), but more and more about the quality of the implementation.

Two month of crash-session in clarification & implementation would make D ready for the summer.

On Monday, 28 January 2013 at 12:04:42 UTC, Szymon wrote:
> On Saturday, 26 January 2013 at 21:30:22 UTC, mist wrote:
>> On Saturday, 26 January 2013 at 20:54:30 UTC, FG wrote:
>>> On 2013-01-26 20:55, SomeDude wrote:
>>>> AFAWK, at least one major videogames studio is writing its new title in D.
>>>
>>> For real? You must be joking. :)
>>
>> Tsss, it is a secret ;)
>
> We actually are video game company ;) That is why D seems like perfect choice for future language. Native code + superb programmer productivity + fast compile times. If D was bit more mature decision would be dead simple.

I believe that if you stick with a particular version of the compiler + druntime + phobos then you could use D now.

If you treat it as a nicer c++ and don't branch out in to the more exotic features, I'd expect a pretty smooth ride.

Am 28.01.2013 16:22, schrieb Namespace:
> On Monday, 28 January 2013 at 15:15:38 UTC, Dejan Lekic wrote:
>> On Friday, 25 January 2013 at 21:39:58 UTC, Namespace wrote:
>>> On Friday, 25 January 2013 at 20:45:22 UTC, Szymon wrote:
>>>> Hi,
>>>>
>>>> I would really like to start using D in our small company as a
>>>> C++ replacement. With that in mind I do have few questions:
>>>>
>>>> 1) Is D2 really ready for production code?
>>>
>>> Not really. A big pain in the ass is the missing rvalue ref that C++
>>> has. So using structs is a big disaster. But there are plenty of
>>> other missing features.
>>
>> Java has far less features than D, including what you complain about,
>> yet it is considered a pretty productive programming language...
>
> I hate to programming in Java because of the many many missing features.
> And because _all_ is stored on the heap. ...

This is not true for current native compilers and JVMs.

If escape analysis can prove the object does not leave scope, new actually allocates on the stack.

--
Paulo

Am 28.01.2013 16:11, schrieb eles:
> On Monday, 28 January 2013 at 12:04:42 UTC, Szymon wrote:
>> On Saturday, 26 January 2013 at 21:30:22 UTC, mist wrote:
>>> On Saturday, 26 January 2013 at 20:54:30 UTC, FG wrote:
>>>> On 2013-01-26 20:55, SomeDude wrote:
>> We actually are video game company ;) That is why D seems like perfect
>> choice for future language. Native code + superb programmer
>> productivity + fast compile times. If D was bit more mature decision
>> would be dead simple.
>
> Well, let's hope it becomes mature enough. My impression is that it
> needs to clean up itself a bit, to remove several ambiguities (that
> @property discussion...) that lasted far too long to bring any good, and
> to work heavily on tools: the compiler.
>
> Otherwise, I think it is OK to start with it. Custom runtimes/libraries
> could provide GC-free implementation where needed (BSPs, games).
>

I have a post-mortem of Witcher 2 for the XBox 360 sitting on my desk explaining how they made use of a GC for the game...

On Monday, 28 January 2013 at 16:26:29 UTC, Paulo Pinto wrote:
> This is not true for current native compilers and JVMs.
>
> If escape analysis can prove the object does not leave scope, new actually allocates on the stack.
>

Modern can even speculate that an object will not leave scope, allocate it on stack, and move it on heap if it escape !

On Monday, 28 January 2013 at 16:28:51 UTC, Paulo Pinto wrote:
>
> I have a post-mortem of Witcher 2 for the XBox 360 sitting on my desk explaining how they made use of a GC for the game...

That is something confidential I assume...?

Am 28.01.2013 18:03, schrieb Szymon:
> On Monday, 28 January 2013 at 16:28:51 UTC, Paulo Pinto wrote:
>>
>> I have a post-mortem of Witcher 2 for the XBox 360 sitting on my desk
>> explaining how they made use of a GC for the game...
>
> That is something confidential I assume...?

No, quite public.

An article on the "Making Games Magazine". The Magazine of the
German games development community.

http://tinyurl.com/bjxxp5f

--
Paulo