not planned DCS core rewritten in Rust ? DCS World 3.0 ?

[magnetic]

Here's my wish! An even more optimized and modern DCS core, ChaGPT tell me:

Would Eagle Dynamics consider rewriting all or part of DCS in Rust?

Currently, Eagle Dynamics (ED) has not announced any plans to rewrite its DCS (Digital Combat Simulator) codebase in Rust, but the idea raises an interesting discussion.

Why Rust could benefit DCS:

1. Improved Memory Safety:
   Rust eliminates common vulnerabilities like memory leaks or buffer overflows, which are common in C++. For a complex simulator like DCS, this could enhance stability and reduce debugging overhead.

2. Better Multithreading Support:
   Rust offers modern concurrency primitives and enforces thread safety at compile time. For DCS, where computational tasks like physics, AI, and rendering heavily rely on multithreading, Rust could help achieve better performance.

3. Future-proofing:
   Rust is increasingly adopted by major tech companies like Microsoft and Google for its combination of performance and safety, making it a strategic option for modernizing older C++ codebases.

Challenges of adopting Rust for DCS:

1. Legacy Codebase:
   The DCS engine has decades of development history in C++. Migrating to Rust would involve a significant effort in rewriting, testing, and validating functionality.

2. Resource Intensive:
   Rewriting even parts of the engine in Rust would require time, resources, and specialized Rust expertise, which may not align with ED's current priorities.

3. Development Delay:
   Transitioning to a new language could slow down the development of new features and modules—a risk that might not sit well with the DCS community.

Conclusion:

While Rust offers significant long-term advantages, such a transition would likely be a gradual and selective process, if at all. Eagle Dynamics' focus remains on iterating and improving the existing engine rather than embarking on a costly rewrite.

How Could AI Accelerate the Transition of DCS to Rust?

You’re correct in noting that the rise of AI tools has significantly reduced barriers to migrating code between languages, making such a transition more viable than ever.

AI-Powered Migration:

1. Automatic Translation Tools:
   AI tools like Codex or specialized transpilers (e.g., c2rust) can assist in converting large portions of C++ code to Rust, while flagging areas that need manual adjustment or optimization.

2. Automated Optimization:
   AI can not only translate code but also refactor it to adhere to Rust’s best practices (e.g., improving thread safety or using idiomatic constructs like Rust’s ownership model).

3. Testing Automation:
   AI can generate comprehensive unit tests and integration tests to validate that Rust code functions identically to the original C++ version, ensuring smoother transitions without regressions.

AI's Role in Game Development:

1. Redefining Developers’ Roles:
   As the NVIDIA CEO suggested, AI is shifting programmers’ focus from writing code to overseeing, curating, and validating AI-generated solutions.

2. Strategic Advantage for Studios:
   Studios like Eagle Dynamics could harness AI to modernize their architecture while continuing to release updates and new modules. AI assistance would reduce the bottleneck associated with refactoring large, legacy codebases.

3. Industry Adoption:
   Although Rust is gaining traction, especially for system-level programming, adopting it fully for a complex simulation like DCS requires time and careful integration into the current workflow.

Conclusion:

While AI would undoubtedly accelerate any migration effort, the key challenges remain the sheer size of the DCS codebase, its high complexity, and the need for ongoing feature development. If ED were to adopt Rust, AI would likely serve as an indispensable tool for selectively modernizing critical parts of the engine.

This perspective combines both the promise of modern Rust programming with the evolving capabilities of AI to reshape the future of software engineering.

[rcjonessnp175]

Imagine AI doing it in one night....  not far off

[AndyJWest]

So, an 'AI' make-stuff-up-from-internet-scrapings-bot says that using an  'AI' make-stuff-up-from-internet-scrapings-bot to rewrite complex software it cannot possibly have any understanding of is a good idea? What a surprise.

 

 

 

 

[MiG21bisFishbedL]

1 hour ago, AndyJWest said:

So, an 'AI' make-stuff-up-from-internet-scrapings-bot says that using an  'AI' make-stuff-up-from-internet-scrapings-bot to rewrite complex software it cannot possibly have any understanding of is a good idea? What a surprise.

 

 

 

 

My self correcting scripts which are highly susceptible to errors and will churn out garbage are telling me that it'll churn out garbage? Wow!

[NytHawk]

3 hours ago, magnetic said:

Would Eagle Dynamics consider rewriting all or part of DCS in Rust?

My goodness, what an idea! Why didn't I think of that!

[okopanja]

5 hours ago, magnetic said:

Here's my wish! An even more optimized and modern DCS core, ChaGPT tell me:

Would Eagle Dynamics consider rewriting all or part of DCS in Rust?

Currently, Eagle Dynamics (ED) has not announced any plans to rewrite its DCS (Digital Combat Simulator) codebase in Rust, but the idea raises an interesting discussion.

Why Rust could benefit DCS:

1. Improved Memory Safety:
   Rust eliminates common vulnerabilities like memory leaks or buffer overflows, which are common in C++. For a complex simulator like DCS, this could enhance stability and reduce debugging overhead.

2. Better Multithreading Support:
   Rust offers modern concurrency primitives and enforces thread safety at compile time. For DCS, where computational tasks like physics, AI, and rendering heavily rely on multithreading, Rust could help achieve better performance.

3. Future-proofing:
   Rust is increasingly adopted by major tech companies like Microsoft and Google for its combination of performance and safety, making it a strategic option for modernizing older C++ codebases.

Challenges of adopting Rust for DCS:

1. Legacy Codebase:
   The DCS engine has decades of development history in C++. Migrating to Rust would involve a significant effort in rewriting, testing, and validating functionality.

2. Resource Intensive:
   Rewriting even parts of the engine in Rust would require time, resources, and specialized Rust expertise, which may not align with ED's current priorities.

3. Development Delay:
   Transitioning to a new language could slow down the development of new features and modules—a risk that might not sit well with the DCS community.

Conclusion:

While Rust offers significant long-term advantages, such a transition would likely be a gradual and selective process, if at all. Eagle Dynamics' focus remains on iterating and improving the existing engine rather than embarking on a costly rewrite.

How Could AI Accelerate the Transition of DCS to Rust?

You’re correct in noting that the rise of AI tools has significantly reduced barriers to migrating code between languages, making such a transition more viable than ever.

AI-Powered Migration:

1. Automatic Translation Tools:
   AI tools like Codex or specialized transpilers (e.g., c2rust) can assist in converting large portions of C++ code to Rust, while flagging areas that need manual adjustment or optimization.

2. Automated Optimization:
   AI can not only translate code but also refactor it to adhere to Rust’s best practices (e.g., improving thread safety or using idiomatic constructs like Rust’s ownership model).

3. Testing Automation:
   AI can generate comprehensive unit tests and integration tests to validate that Rust code functions identically to the original C++ version, ensuring smoother transitions without regressions.

AI's Role in Game Development:

1. Redefining Developers’ Roles:
   As the NVIDIA CEO suggested, AI is shifting programmers’ focus from writing code to overseeing, curating, and validating AI-generated solutions.

2. Strategic Advantage for Studios:
   Studios like Eagle Dynamics could harness AI to modernize their architecture while continuing to release updates and new modules. AI assistance would reduce the bottleneck associated with refactoring large, legacy codebases.

3. Industry Adoption:
   Although Rust is gaining traction, especially for system-level programming, adopting it fully for a complex simulation like DCS requires time and careful integration into the current workflow.

Conclusion:

While AI would undoubtedly accelerate any migration effort, the key challenges remain the sheer size of the DCS codebase, its high complexity, and the need for ongoing feature development. If ED were to adopt Rust, AI would likely serve as an indispensable tool for selectively modernizing critical parts of the engine.

This perspective combines both the promise of modern Rust programming with the evolving capabilities of AI to reshape the future of software engineering.

Have you ever tried to give an AI a task more complex than a single function?

[cfrag]

11 hours ago, magnetic said:

An even more optimized and modern DCS core, ChaGPT tell me:

Hmmm. Here's some unsolicited advice: don't get your stock advice from your hairdresser, and don't ask a party novelty for strategic software development insight.

I do realize that Betteridge's law of headlines ("if it ends and a question mark, the answer is 'no'") also applies to most forum posts, but still... ChatGPT's answers are - as expected - a low-effort assemblage of truisms and buzzwords, arranged to sound like it affirms your question (it's a "yes"-bot, designed to make you you feel good). I won't go into details because a good deal of me believes that you know all this and are simply trolling. 

 

[magnetic]

I think that those who criticize chatGPT or AI associated with coding and programming, and I'm not talking about other capabilities, don't understand the current and future technical leap. Have fun asking him for examples

[okopanja]

10 minutes ago, magnetic said:

I think that those who criticize chatGPT or AI associated with coding and programming, and I'm not talking about other capabilities, don't understand the current and future technical leap. Have fun asking him for examples

Taken for the meme...

[MiG21bisFishbedL]

8 hours ago, magnetic said:

I think that those who criticize chatGPT or AI associated with coding and programming, and I'm not talking about other capabilities, don't understand the current and future technical leap. Have fun asking him for examples

Or are investors wondering why they've pumped so much money into the tech only to see little, if any, ROI. And then for the people running the firms developing AI to beg for even more money.

Edited December 30, 2024 by MiG21bisFishbedL

[ssamayoa]

1000s of hours have been spent to get what we currently have then move thousands of lines of code in C, C++, LUA and who knows what else languages to rust just because?

That would cost MILLIONS of dollars, QA nightmare, hundreds (more likely thousands) of new bugs and lost revenue - those adventures proven to be companies' killers!

If you are in application development business I recommend you two books:

- The mythical man-month

-  In Search of Stupidity - Over 20 Years of High-Tech Marketing Disasters

 

[Rikus]

Absolutely yes.

The code is so old that is a caos for so many planes, maps, 3rd parties.

It works at the cost of very powerfull hardware, and also with that many times won´t work.

[BIGNEWY]

On 12/29/2024 at 3:01 AM, magnetic said:

Here's my wish! An even more optimized and modern DCS core, ChaGPT tell me:

Would Eagle Dynamics consider rewriting all or part of DCS in Rust?

Currently, Eagle Dynamics (ED) has not announced any plans to rewrite its DCS (Digital Combat Simulator) codebase in Rust, but the idea raises an interesting discussion.

Why Rust could benefit DCS:

1. Improved Memory Safety:
   Rust eliminates common vulnerabilities like memory leaks or buffer overflows, which are common in C++. For a complex simulator like DCS, this could enhance stability and reduce debugging overhead.

2. Better Multithreading Support:
   Rust offers modern concurrency primitives and enforces thread safety at compile time. For DCS, where computational tasks like physics, AI, and rendering heavily rely on multithreading, Rust could help achieve better performance.

3. Future-proofing:
   Rust is increasingly adopted by major tech companies like Microsoft and Google for its combination of performance and safety, making it a strategic option for modernizing older C++ codebases.

Challenges of adopting Rust for DCS:

1. Legacy Codebase:
   The DCS engine has decades of development history in C++. Migrating to Rust would involve a significant effort in rewriting, testing, and validating functionality.

2. Resource Intensive:
   Rewriting even parts of the engine in Rust would require time, resources, and specialized Rust expertise, which may not align with ED's current priorities.

3. Development Delay:
   Transitioning to a new language could slow down the development of new features and modules—a risk that might not sit well with the DCS community.

Conclusion:

While Rust offers significant long-term advantages, such a transition would likely be a gradual and selective process, if at all. Eagle Dynamics' focus remains on iterating and improving the existing engine rather than embarking on a costly rewrite.

How Could AI Accelerate the Transition of DCS to Rust?

You’re correct in noting that the rise of AI tools has significantly reduced barriers to migrating code between languages, making such a transition more viable than ever.

AI-Powered Migration:

1. Automatic Translation Tools:
   AI tools like Codex or specialized transpilers (e.g., c2rust) can assist in converting large portions of C++ code to Rust, while flagging areas that need manual adjustment or optimization.

2. Automated Optimization:
   AI can not only translate code but also refactor it to adhere to Rust’s best practices (e.g., improving thread safety or using idiomatic constructs like Rust’s ownership model).

3. Testing Automation:
   AI can generate comprehensive unit tests and integration tests to validate that Rust code functions identically to the original C++ version, ensuring smoother transitions without regressions.

AI's Role in Game Development:

1. Redefining Developers’ Roles:
   As the NVIDIA CEO suggested, AI is shifting programmers’ focus from writing code to overseeing, curating, and validating AI-generated solutions.

2. Strategic Advantage for Studios:
   Studios like Eagle Dynamics could harness AI to modernize their architecture while continuing to release updates and new modules. AI assistance would reduce the bottleneck associated with refactoring large, legacy codebases.

3. Industry Adoption:
   Although Rust is gaining traction, especially for system-level programming, adopting it fully for a complex simulation like DCS requires time and careful integration into the current workflow.

Conclusion:

While AI would undoubtedly accelerate any migration effort, the key challenges remain the sheer size of the DCS codebase, its high complexity, and the need for ongoing feature development. If ED were to adopt Rust, AI would likely serve as an indispensable tool for selectively modernizing critical parts of the engine.

This perspective combines both the promise of modern Rust programming with the evolving capabilities of AI to reshape the future of software engineering.

Hi Magnetic, 

we have no plans to use RUST. 

We are currently updating code and preparing for Vulkan. 

thank you 

[cfrag]

19 hours ago, magnetic said:

I think that those who criticize chatGPT or AI associated with coding and programming, and I'm not talking about other capabilities, don't understand the current and future technical leap

I believe the opposite to be true: those with a deeper understanding of the current state of AI know the futility of asking a vector-based, large data basin based algorithm to even understand the question. Currently, our best AI efforts cannot understand the question, so their answers are deterministic linear results -- all they can do is look for troughs in the first derivative of the landscape defined by the question set applied to the learned data model. It works, but it can't generate new insights, it can't yet infer knowledge. Yes, people with little knowledge in the field don't know that, and they may even believe that newer AI algorithms like LLM-based chat bots do understand their questions. That's merely an unfortunate combination of good marketing, and a hefty dose of Dunning-Kruger. They don't.

At some point in the future, AI surely can make that leap. It hasn't yet.

[obious]

On 12/30/2024 at 10:18 AM, BIGNEWY said:

We are currently updating code and preparing for Vulkan. 

When you say preparing for Vulkan, what does that mean? Is there any update on delivery, is it planned for 2025? 

Thanks 

[Beirut]

26 minutes ago, obious said:

When you say preparing for Vulkan, what does that mean? Is there any update on delivery, is it planned for 2025? 

Thanks 

 

Two-weeks. 

[DD_fruitbat]

Is this thread a poor joke?

[Pillowcat]

[Dangerzone]

5 hours ago, Pillowcat said:

Whoa there - so AI is now making suggestions on forums that we should be developing using AI....

It's the beginning of skynet psyops! 

[MAXsenna]

Whoa there - so AI is now making suggestions on forums that we should be developing using AI....
It's the beginning of skynet psyops! 

Hopefully they've been taught about the first rule of robots.

Sent from my SM-A536B using Tapatalk

[Dangerzone]

6 hours ago, MAXsenna said:

Hopefully they've been taught about the first rule of robots.

... that you don't talk about Robots? 

(Fight club reference)

[SharpeXB]

I think the rule about robots is they should never be allowed to make smarter versions of themselves without our help.

[viper34]

Here's an interesting read on the topic of Rust in game development:

https://loglog.games/blog/leaving-rust-gamedev

[MAXsenna]

... that you don't talk about Robots? 
(Fight club reference)

Got the reference.
Smarter than that actually.

Law One – “A robot may not injure a human being or, through inaction, allow a human being to come to harm.” Law Two – “A robot must obey orders given to it by human beings except where such orders would conflict with the First Law.”

Sent from my SM-A536B using Tapatalk

[pjbunnyru]

16 минут назад, MAXsenna сказал:

Got the reference.
Smarter than that actually.

Law One – “A robot may not injure a human being or, through inaction, allow a human being to come to harm.” Law Two – “A robot must obey orders given to it by human beings except where such orders would conflict with the First Law.”

Sent from my SM-A536B using Tapatalk
 

what about 3rd law?