> OSes like Linux which allow natively compiled applications > will have a huge speed advantage.
Why? In my experiance a well designed meta-machine imposes very little penalty, and .NET has the advantage of M$ having the JVM as a prototype (as well as the fact that they only have to satisfy their own requirements) . I read an interesting article on Hard disks recently that pointed out that they can seek in less than 1 revolution so OS's that 'optimise' disk performance by elevator seeking (like NT) now actually slow the OS down, This may appear unrelated but the point is that as hardware changes assumptions about how to get 'performance' can become invalid And if .NET takes over the world (which at this point I can't see why not, though I hope mono performs) then the .NET spec will become what the hardware designers tune their chips for. This is not to suggest that the Linux/Unix method will die but that this could grossly improve M$ OS offering IMHO Neven ----- Original Message ----- From: "David Brennan" <[EMAIL PROTECTED]> To: "Multiple recipients of list delphi" <[EMAIL PROTECTED]> Sent: Thursday, October 09, 2003 9:38 AM Subject: RE: [DUG]: octane > I don't think he was comparing .NET and Java. Java was just an example of > another slow language. The point was that if .NET is the only Microsoft > option then other > > David. > > > -----Original Message----- > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > > Behalf Of Corey Murtagh > > Sent: Thursday, 9 October 2003 7:29 AM > > To: Multiple recipients of list delphi > > Subject: Re: [DUG]: octane > > > > > > Phil Scadden wrote: > > > > >>Short answer: YES! > > >> > > >>My take on this situation is as follows: > > >> > > >>1. Microsoft has already been bitten once having to support multiple > > >>architectures (see Alpha and MIPS), and to avoid this problem > > in the future > > >>(see x86-64 and IA-64) they hace created .NET as a platform neutral > > >>environment. > > > > > > Unless .NET can run a great deal faster than Java, then this is a > > > very risky strategy. A port of Linux to x86-64/IA-64 would > > allow developers > > > to create native applications (open office) that would run rings around > > > .NET applications compiled to CLR. > > > > There's no reason to expect that CIL code /can't/ run at the same speed > > as Java bytecode. They're the same basic concept, just a different > > implementation. > > > > Where it might fall down is in the implementation of the VM and the crap > > added to CIL to support various "look how much better than Java we are" > > features. > > > > Oh, and Java isn't 'native' at all, just in case you were confused about > > that. Where Linux wins is that a huge amount of source code is > > available to allow you to recompile your favorite applications on a new > > platform, without having to wait for the author to release a > > platform-specific variant. If the source /isn't/ available, you still > > have to wait. > > > > -- > > Corey Murtagh > > The Electric Monk > > "Quidquid latine dictum sit, altum viditur!" > > > > ------------------------------------------------------------------ > > --------- > > New Zealand Delphi Users group - Delphi List - [EMAIL PROTECTED] > > Website: http://www.delphi.org.nz > > To UnSub, send email to: [EMAIL PROTECTED] > > with body of "unsubscribe delphi" > > Web Archive at: http://www.mail-archive.com/delphi%40delphi.org.nz/ > > -------------------------------------------------------------------------- - > New Zealand Delphi Users group - Delphi List - [EMAIL PROTECTED] > Website: http://www.delphi.org.nz > To UnSub, send email to: [EMAIL PROTECTED] > with body of "unsubscribe delphi" > Web Archive at: http://www.mail-archive.com/delphi%40delphi.org.nz/ > > --------------------------------------------------------------------------- New Zealand Delphi Users group - Delphi List - [EMAIL PROTECTED] Website: http://www.delphi.org.nz To UnSub, send email to: [EMAIL PROTECTED] with body of "unsubscribe delphi" Web Archive at: http://www.mail-archive.com/delphi%40delphi.org.nz/
