GRLevelX Trial Support Forums

Any chance that you are planning to port your software for Linux? I am considering purchasing, but I primarily use Linux. The only Windows machine I have is old running win98 and I do not plan to purchase any Microsoft software in the future.

If no Linux port is planned, do you know if anyone has had success running your software on Wine or Crossover Linux?

Thanks!

There are no plans for Linux versions of the apps. I don't recall hearing of anyone successfully running the apps under any of the Windows emulators. They're closer to the hardware than most Windows apps due to the use of Direct3D.

Mike

Guys --

The obvious reason why GR only works on windows is because its very heavily dependant on the DirectX Graphics Libraries, namely (correct me if i'm wrong here) Direct3d and DirectDraw.


Personally, i wish you would port this program over to OpenGL (which is NOT hard), which would make it insanely easy to put GR on almost anything.... imagine having that 3d view on your iphone! *drool*


Basically, Mike's the developer and he can do what he wants... but IMHO with this stubbornly closed philosophy and with its astronomical price-tag, i think its only a matter of time before some open source project knocks GR out of the water.

-J

Direct3D, threading model, file systems, APIs, etc, etc.


Which version of OpenGL? Which extensions?

The screen is too small on phones to do anything useful with the Volume Explorer.


It's not a "stubbornly closed" philosophy. It's the development time required to make a quality program on all the different systems out there.

If an open source project of similar capabilities emerges, great! More choices for users.

Mike

Okay, maybe I was a bit strong.

From what it looks like (and by that i mean as much as i can tell without the source), the whole thing is written in C++ and uses the .net framework. Cool. Makes it very nice to craft a good-looking windows application. Additionally, i think i hit it right on the noggin with Direct3D, so we got that.

So can we port this app to other platforms without a total rewrite? I think so.

Basically you would have 2 tasks: Convert the Program to use a different GUI API (GTK for linux, Cocoa for Apple) for the different operating systems. Honestly, i think this is completely doable. From what i hear, once you have an app that uses GTK, its pretty doable to build a cocoa-shell around it thats tailored just for macs. As the code is written in .net, you would have to sanitize the entire thing to use a different framework, but once thats done it should be smooth sailing, and luckily since you wrote it in C++ you probably wouldn't have to do a complete rewrite (yikes!)

Of course, you could do the whole thing in java (which is specifically designed to combat this kind of problem), but that just screams *BAD IDEA!* in my head.

Your other tasks would be to port the renderer, which i maintain is not as hard as it sounds. OpenGL3 is a really great platform, and to me, the code just looks so much more cleaner. Since you aren't doing any high-demand rendering, i don't really think you *NEED* the bleeding-edge advantages of DirectX (which aren't really advantages so much as it is that most Vid Cards are optimized for DirectX). Laborious? Yes. Mind-Bogglingly impossible? Far from.


As for the open source deal, i think honestly the reason why you don't really have an open source competitor is due to your (pretty damn impressive) secret sauce; namely the 3d Interpolation of 2d radar slices. Its actually a pretty fascinating idea to generate useful weather information, but it seems to me that it took quite a bit of code-fu to get that feature working. As i'm not an experienced 3D Developer with Experience in modeling CAT scans and MRI's, i don't think i would be able to code something like that.


I'm not knocking your work mike, in fact I really love your program and the technical accomplishment is very impressive. I just wish there would be some expansion, thats all.

-ND

PS: It looks like theres actually a project that is pretty much GTK+ --> Cocoa, so if you just port it to GTK it looks like you might be done there.

If there's one thing that I'm "stubbornly closed" on, it's using generic frameworks. What you end up with is a generic, bloated, non-optimized app (like Java apps). In fact, the GR programs do not use .NET or MFC or any of the other middleware layers. They're written to the bare Windows and Direct3D APIs.

Also, don't forget that I don't have an army of programmers at my disposal. GRS is a one man shop. I have to focus on one, optimized platform and Windows is the most widely used OS (90+%).

The volumetric renderer was the single most difficult programming task I've ever done. And not just the actual rendering of the data, it involved optimizing the volume sampling, table generation, etc. Doing it a second time wouldn't be that difficult but the first time was a doozy:)

I guess the bottom line is this: I could spend a long time developing generic, somewhat platform-independent versions of the programs that would only directly benefit a relatively small number of potential users while penalizing the majority -- or -- I could spend that development time adding new features that will benefit the vast majority of users. There are many hardcore analysis and display features needed in GR2Analyst. For example, I want it to analyze the volumes and indicate areas of concern (beyond the current derived products), I want to be able to tell it what to look for and how to look for it, etc. I want better, optimized GIS handling, etc. If GR2Analyst was a static, best-as-it-can-ever-be program, then yes, I could focus on platform independence. We have a long way to go to get to that point.

Mike