TheSwamp
Code Red => .NET => Topic started by: It's Alive! on November 20, 2021, 10:48:40 PM
-
I wanted to use the priority_queue class, but it’s in .NET core, not framework.
Then I tried to use RBush: R-Tree Implementation for C# on GitHub. But this also requires features newer than what’s in framework.
Just disappointed that Microsoft depreciates stuff people have worked for years on :crazy2:
-
Framework is dead, not .NET :-)
I assume we move to .NET 5 in a near future AutoCAD version. Or we move to web programming when AutoCAD becomes web only.
-
I'm all for web based app's but CAD is a tricky one. I think they're getting closer with webgl and wasm etc but it's the data that's the bottle neck. Loading and saving gig's of of a drawing db over the net is loaded with problems to solve like speed and data races etc.
It can be done but can AutoDesk keep up? As Daniel noted, they're behind already.
Just disappointed that Microsoft depreciates stuff people have worked for years on :crazy2:
It's all to do with bring .Net to Linux/Unix for web/cloud services. It would be nice if they at least kept the framework as a wrapper into Core.
-
Just disappointed that Microsoft depreciates stuff people have worked for years on :crazy2:
Microsoft has always encouraged the vocal crowd to chase the shiny new things, presumably because it is what they make the most money from. There has also always been a silent majority (https://www.hanselman.com/blog/dark-matter-developers-the-unseen-99) who continue to use the stuff that works because it already works - Windows Forms is still a thing and it "ain't goin' nowhere".
.NET 3.5sp1 is scheduled for demolition in 2029, so 4.8 should last a while. See https://docs.microsoft.com/en-au/lifecycle/products/microsoft-net-framework
The ODA has released SWIG-generated .NET wrappers for for .NET core but not .NET classic which is the "legacy" source-code compatible SDK. Code for the SWIG version looks a lot more like C++ than the "legacy" source code.
But yah, it's annoying when the shiny new things are made from other shiny new things. Somethings you can head down the rabbit-hole and compile them from frameworks you do actually use, sometimes the rabbit-hole just keeps going.
People will abandon on-premise apps and storage to flock to the the cloud when it has no latency, data breaches, outages and bill-shock.
-
If the speed of NET inside Acad is too slow-of speed
--then in the future, Acad+NET may be dead, sure55%
:woow:=Anyone, the speed of DLL/s are very-important.
Maybe, the VLISP is dead.?
Anyone==The Visual-Lisp was rescued by BrisCAD,ZwCad and GstarCad.
:knuppel2:
I wanted to use the priority_queue class, but it’s in .NET core, not framework.
Then I tried to use RBush: R-Tree Implementation for C# on GitHub. But this also requires features newer than what’s in framework.
Just disappointed that Microsoft depreciates stuff people have worked for years on :crazy2:
-
d2010: The speed seems okay. Especially considering that most newer computers are overkill for AutoCAD. I could be wrong though as I'm not too experienced with .NET.
Microsoft absolutely had to improve .NET (and C#) to remain competitive. Short term pain for long term gain. Well... at least that's what I'm hoping for as I've committed a lot of time towards learning it. So hopefully the pain associated with upgrading from Framework is worth it.