Posted 11/02/2009 at 01:22:41pm

The Mac's record rise in market share to 5.27% is still waaay down from the 25% market share it enjoyed in the early 1990's.  Apple views itself as a hardware company, not a software company, so seeing Apple make an official release of the Mac OS for the Intel PC platform is not going to happen.

At the 1993 Apple Developers Conference, I remember attending a secret, invitation-only demonstration of the Mac OS running on the Intel platform.  They even had multiple monitors running for the desktop (something Windows did not support).  We were really excited about this since it would open up our applications to a whole new audience.  Unfortunately, nothing came of it and Apple gave us their reasoning: "We're a hardware company, not a software company.  Releasing the Mac OS on the intel platform would hurt our sales."

Things are still the same over at 1 Infinite Loop.

Posted 10/30/2009 at 01:47:15pm

...Anything available out there yet to plug into this?

Posted 10/23/2009 at 11:39:21am

64 and 32 bits refer to the size of the primary registers in the processor (not counting any vector processing registers).  32-bit programs handle atomic data types between 1 and 32 bits wide.  64-bit programs handle atomic data types between 1 and 64 bits wide.  A 64-bit programming model is a superset of the 32-bit model.  Both models, for example, support 16-bit data types.  But it doesn’t follow that 32 and 64 bit models aren't true 32 or 64 bit because they also support 16-bits.

Please understand something: converting any 32-bit program to 64-bit is a lot more than just "recompiling a program...to get it using 64-bit data types."  Under ALL operating systems--even Linux-- you need to perform "significant replumbing" to convert a program.  The only way a program can be built on the fly using either 64-bit or 32-bit memory models is to have its source code carefully constructed with conditional compilation blocks and carefully chosen identifier definitions that build according to what memory model is chosen.  And even then it isn’t completely fool proof.  It is no easier on Linux than it is on Windows or any other OS for that matter.

The unix.org page you cite is only talking about abstract simple data type identifiers under a C compiler and how they can use the choice of identifiers to help portability between 32 and 64 bit compilations.   Whether you want to call a 64-bit data type “longlong” or “Int64” is completely arbitrary. I have done lots of cross-platform software development in my time; LP64 helps, but it is no magic bullet.

“This is all because Microsoft chose to cater to a bunch of sloppy programmers that couldn't be bothered to figure out how to use typedef to avoid making hardcoded assumptions about data types.”

This is a patently absurd and asinine statement from someone who knows nothing about professional software development.  It takes a lot of time, and therefore a lot of money, to convert any legacy application from 32 to 64 bits.  It’s a hell of a lot more complex than knowing how to use “typedef.”  Most of the time it is just not worth the expense: how many 4GB+ Word files have you edited lately?

*crickets chirping *

It has nothing to do with sloppy programming.  If you have an application with literally millions of lines of legacy code, regardless of the operating system, you just can’t change it to a new programming model by re-defining a few compiler identifiers.  Anyone who believes that is seriously deluded.  If it ain’t broke, don’t fix it, fool.

Over the years, Microsoft has done a remarkable job of preserving backwards compatibility in their support of legacy software.  Windows has hundreds of times the number of applications available for it than any other operating system.  I have 12-year-old, 32-bit, Windows applications I can still run under Windows 7 x64.   Try running a Mac OS 7 application under Mac OS X sometime…lots of luck with that, pal.  And if you want to talk about sloppy programming, get under the hood of Mac OS X and look at the mess they have there with NextStep and Carbon.  Since the UNIX hobbyist geeks don't have millions of users depending on their software, they can afford to give their puny user base the finger and not support legacy applications. 

Stop sounding like an ignoramus and go troll somewhere else, please.

Posted 10/22/2009 at 04:51:30pm

Xp x64, Vista x64 and 7 x64 are all true 64-bit operating systems. WoW64 is the emulation layer that enables 32-bit Windows-based applications to run seamlessly on 64-bit Windows. The data typing is backwardly compatible so we can run most all of our old software.

damicatz, you are just as clueless about everything else you posted as well, so I won't waste my time. Yeesh!

Posted 09/23/2009 at 04:12:17pm

Most people confuse the persistence of vision with the perception of motion.  You can drop a frame rate down to around 16 fps before any flickering becomes too noticeable  That is the persistence of vision.  24 fps was chosen for the theatre because all flickering disappeared at that frame rate.   Additionally, 24fps has common factors with 60Hz electricity so making synchronous motors for film drives for 24fps was relatively easy.

However, the human eye can perceive motion up to around 60 fps.  That's why real-time rendered games look better at the higher frame rates.  The motion blur in Crysis and other games is just a gimmick to make the video look more like 24 fps film.  This detracts from the video quality and defeats the purpose of having higher frame rates, so I always disable any motion blur.

Posted 09/08/2009 at 03:18:16pm

...Maybe marketers can start calling products "Special Edition Xtreme" or "SEX".  That combines every marketing cliche of the past 20 years and sex at the same time!

Posted 09/08/2009 at 02:57:22pm

In the 1990's, the label "Special Edition" or "SE" was beaten to death by marketers looking for product names.  "Xtreme" is today's marketing cliche.  When everything is "Xtreme" then nothing is.

Posted 08/09/2009 at 05:39:53am

The solid copper version of the TRUE is just a gimmick.  It weighs a ton, so many vendors post warnings to not install it on an upright-standing motherboard.  In the end, the copper version only drops the temps around 2C lower than the standard version anyway. It does look nice though!

Posted 08/09/2009 at 05:39:12am

I'll support Thermalright, the inventors of the TRUE, and pass on the knockoffs.

Posted 08/09/2009 at 05:28:42am

Heh, I'm still saving up for my Mac G3.  ;)