Especially when you consider a "just hit reload" development cycle as the competition.
Since the Typescript compiler itself is > 2MB of minified js, it doesn't surprise me that compiling any TS has at least a second overhead.
What's more important to me is that:
- The compiler performance scales as the codebase increases in size.
- Incremental builds are fast. (ie the -watch option).
So far I have found both to be true. Incremental builds take < .5 sec on my laptop. That means I get a cute red-squiggly in Visual Studio in a convenient amount of time. Ultimately, that's half the reason for using TS in the first place; if it didn't improve our productivity, we wouldn't be using it.
I found that converting 10k of C# was incredibly quick and the overall build time was still totally dominated by the slow TypeScript compiler.
You are right that the development time is spent unit-testing C# and very rarely invoking the TS compiler so rather than making the workflow worse, it improves it immeasurably. The alternative to recompile TS each time and use Karma and Chrome dev tools is unbearably slow and clunky in comparison.
The real win I was after was that the quality of the test tools and debugger for .Net. Using a tool like TestDriven.Net from visual studio so from a r-click I have test coverage, performance, edit-and-continue and all the benefits of the Visual Studio debugger.
The generation to typescript is more of a publish step than part of the development cycle. I also convert the NUnit tests to Jasmine so I run the same test suite across the generated javascript. This does catch issues because there are differences, like lexical scoping, that my cs2ts tool did not try to solve.
I think the point is that you have some existing libraries that would be useful in a web application, you could go ahead and compile those down to TS and use them within your TS code.