fordlemon:
domokun:
I thought I'd pick up on this Autocad theme, and domokun's inquiry about 3DStudio Max in particular. It has pertinence to all things Windows-to-Mac
3DStudio is heavily depending on Microsoft's compiler, in C++, and on the Windows operating system. They didn't consider platform independence early on, and I think now it's quite late to consider it. I'm sure they could, but that depends on the nature of market forces. It will, as others replied, run on Windows in the x86 MAC, but probably not natively under OSX. I think the VPC will fail to represent a multiple processor machine to a guest operating system, so MAX would suffer rendering performance greatly under the imposition. It's not the kind of application one should want to run under significant constraints of RAM, CPU or disk. I understand Maya's port to MAC OS X has been difficult, and quite buggy in early releases, but then Maya has been known to be a bit buggy all along, and extremely particular about OpenGL compliance.
Now, AutoCAD is an interesting subject. I know a great many architects, and I've written a number of plugins for AutoCAD. That product does have some platform independent roots, but now it is more dependent on Windows. It was in DOS for years (through version 11, if memory serves). It's interface is quite strange, compared to other GUI applications, in that it inherits a command driven method as the primary means of controlling it. As fordlemon said, it's not pretty, but that's not the point. It's an industry standard, and it's probably never going away.
In 85% (or higher) of all architectural firms in the U.S., AutoCAD files (the DWG file) is the workpiece of the architectural firm. For good or bad, if they loose a DWG file, they've lost dozens or hundreds of hours of work. Architecture was once based on paper as the workpiece, but that's no longer the case. There are other CAD packages, and some get the attention (even near-religious devotion) of supporters, but they are few and far between. Many architects have a decade of familiarity (and some 15 years or more) with AutoCAD, it's use, it's kinks, and it's problems. I know many firms that still insist on using AutoCAD 14, which is from 1995/96 (and I think their first 32 bit version).
It is quite true that the product is a sprawling mess, from a developer's viewpoint. It's huge, requiring mind boggling memory and file space footprints for what's going on. What's even more interesting to me is that well over half (and possibly 75%) of the architects that depend on it, use recent versions but only access features that were available in AutoCAD 14.
It's probably not going to be ported to the MAC. The architectural customer is their primary market, with nods to engineers (which are more oriented to ProE or more industry specific tools). Architects are notoriously single minded about hardware, computers and AutoCAD. They're just not MAC like people, generally. I know, my wife is an architect
Ultimate response to the specific points in domokun's question are instrumental in understanding the porting question of any complex application.
Isn't that a good reason for Mac to port it? Don't know if that is the right term? To make it run better, faster and improve the look too? I don't understand exactly what the the Microsoft .net frame work entails but won't apple have something similar somewhere down the road?
What kind of frame work does 3dStudio run on? Is that the same story?
Working backwards, 3DStudio is based, essentially, on MFC and custom controls. That's what ties it to Windows, though with some painful work that could be switched to another framework that's platform independent, like wxWidgets.
Microsoft's .Net is a confusing thing to me. It's a bunch of different things rolled under one named umbrella. At the heart of it is a run time engine, a little like Java's run time engine, called the CLR (common language runtime). Unlike Java's runtime, which was one language for many platforms, the CLR is an engine for one platform for many languages. Although there's a port for Linux underway, .Net appears to me to be a means of locking applications into Windows (again), and of developing something new for everyone to buy. I don't see any great benefits, though mindless followers of Microsoft seem to think otherwise.
The idea in the CLR is that without regard to language choice (which are generally Microsoft tools, be it C#, J#, F# (not yet released I think),or C++), there is a certainly level of interoperability between code. The underlying glue is the CLR, which then has the power to execute code conditionally, something like the way a firewall limits IP traffic. The CLR could be used to implement security strategies at the code execution level.
It's generally slower than native compiled code, though speed is considerably better than most Java implementations. What's most important to me is that the new graphics engine in XP (longhorn) will be limited to .Net code, which I think sucks. I'm not fond of the CLR, nor do I have any reason to use it myself. I don't know of any client that's asking for it (I'm a developer), and since it has at least as much implication for corporate use, and corporations are increasingly interested in Linux instead, I'm not sure it matters to anyone except Microsoft, or their hypnotized followers.
If this doesn't serve to explain .Net, then you're on the right track it just about defies explanation. Oh, and no, Apple has little in mind that functions in a similar way, and I don't think it matters. I think it's a direction about as strange, and unproductive and meaningful as the countless hacks Intel put into the Netburst architecture. Only, Microsoft is putting all their newer things in .Net, so we'll have to follow along if we want to use them.
.Net, among other things, is a way to provide garbage collection, like in Java, but I have my own arguments with any real need for garbage collection 3DS Max, for example, seems to run quite well without it. Several SQL engines, server's like Apache, and a whole generation of fine, large, complex, ambitious applications do well without it. Java fans will argue it's merit, and so does B. Gates, but frankly, I'm not entirely convinced.
And to the last point (your first, domokun) porting AutoCAD to the MAC is something like painting lipstick on a pig. It won't change the underlying nature of it's appearance or the origin of it's stench. It would need a bath and a genetic makeover to really change all that. It would serve to show, however, that applications for the MAC aren't automagically wonderful they have to be compliant to be that.
