No, Matrox was just always way out in front in terms of 2D image quality. That was their chief advantage for a while, after everyone else surpassed them in performance.
I have the impression that within the last 12 or 18 months, 2D image quality has become a priority (maybe a prerequisite) among enthusiasts again. In any case, I recently got an nVidia FX 5200 card (I think the vendor is eVGA) and the quality is superb on my 19" Sony Trinitron--better than the Matrox G400 I used to use.
I am by no means an expert on the subject, but I'm pretty sure language designers can and do take performance and "scalability" (whatever OP means by that) into account.
For example, Java uses a hybrid type system where most values (strings, array) are objects and have by-reference semantics, but primitive values (ints, longs, doubles, bools, etc.) do not behave as objects and have by-value semantics. It's less elegant for the programmer than some other languages (Smalltalk, Ruby, others?) where Everything Is An Object, but makes a big difference for performance and memory consumption (and memory usage patterns).
The Java language spec also goes into quite some effort to specify a threading model. IMO this is a great boon for people who want to build serious servers in Java... contrast this to many scripting languages, which (appropriately, perhaps) seem to treat concurrent programming as an afterthought.
Or if you take a look at Bjarne Stroustrup's Design and Evolution of C++, he mercilessly rejects adding any language feature that cannot be implemented without slowing down code that doesn't use the new feature. (I can't think of any examples right now, and I can't find my copy.)
>> A "supercomputer" might be good as a graphics or physics engine for gaming
Actually I believe today's 3D graphics cards have processors that look more like supercomputers than general purpose microprocessors--i.e. many pipelines optimized to handle streams of vectors instead of von Neumann style.
Re:Especially in the fog of marketese that is .NET
on
Advanced .NET Remoting
·
· Score: 2, Informative
It's basically a framework for remote procedure calls, over either TCP sockets or on top of HTTP. You can use it in a way similar to Java's RMI, or like SOAP (web services), or you can provide your own protocol.
If that's still too jargony for you, then think of it as a way for objects in different processes or machines to interact with each other (almost) as if they lived in the same process.
For example,
StockQuotes stockQuotes = new StockQuotes(); double ibmPrice = stockQuotes.GetLatest("IBM");
In this example, the StockQuotes object could actually be running on a different machine. In.NET, all you have to do is make the StockQuotes class extend a special system-provided class, and then add a configuration file to the server.exe and the client.exe. The rest is "magic"--you can even use ordinary constructors.
Note that unlike Java RMI, you don't have to declare an extra interface, run an extra compilation step, etc. It all just sorta happens behind the scenes...
Agreed that there is a big impedence mismatch here... but if you haven't already, take a look at ASP.NET Web Forms. It does a very surprisingly good job of mimicking traditional UI programming frameworks, and with a relative minimum of ugly hackery under the covers.
project home page
on
Grid Processing
·
· Score: 2, Informative
Recent versions of Red Hat include up2date (Red Hat Update Agent) which isn't just a notification service--it's an icon that sits in the corner of your screen and turns to an exclamation point when there are patches and updates to be had. Click on the icon, and it launches a wizard that will automatically download and install any patches you want. Correct me if I'm wrong, but I think it costs like $60/year for the basic service; I paid the fee, it's worth it to me to not have to go get patches by hand.
Windows XP has a Windows Update utility that does the same. I think you can also download the utility for Windows 2000. Until Windows Update, I never installed any patches on my Windows box--just the big service packs. Now I'm never more than a few days out of sync.
Most computer science majors I knew at MIT ('96-'00) didn't have a computer, and yet most business majors (myself included) had the latest ThinkPads. Though come to think of it, the underpowered Sun workstations in the clusters were plenty powerful enough to gcc comp-sci projects, while business majors needed powerful notebooks for their wizzy Powerpoint 2000 presentations.
I'd say the situation was "ironic" but I'm afraid someone will correct me...
I found this article on the same site quite interesting.
"By the mid-nineties the new breed of computers made from off-the-shelf commodity chips arrived on the market. Those funded from the US ASCI programme and consisting of several thousand CPUs grabbed the headlines, but because of communication and memory bandwidth limitations, they often deliver very little of their potential peak performance to the user."
Five years ago, there were numerous groups in academia investigating other navigational metaphors than the forward/back, history, and bookmarks mechanisms we have today. For example:
Recency-based, instead of stack-based, history. If you're on Page A, click to Page B, hit 'Back' to return to Page A, and then click on Page C, you've lost Page B from your history. Some had the idea of having 'Back' not move a pointer down the history stack, but copy or move a history entry to the top of list. That way, hitting 'Back' twice would always return you to the page that was on your screen two navigational gestures ago.
Tree-based history view. A lot of people have thought of, and tried, this one--show history as a 2D tree-like structure instead of a flat list.
Smarter bookmarks and history. For example, knowing how many times you've visited each page and sorting them accordingly. Or showing pages in some arrangement that reflects the "nearness" of pages to each other.
For some details, check out some of the papers by S. Greenberg. (There are tons of other links I had around but I can't find them right now.)
I think the heavy research into this kind of "browser innovation" may indeed have died five years ago. What researchers began finding out then is that people had become very conditioned to the Back/Forward/History/Bookmark behavior provided by Netscape/IE. Any deviation from that made users uncomfortable and confused.
Notice that while Opera, Firebird, and the like have provided some nice advancements, they have not changed the basic behavior of these buttons. Either they (Opera, Mozilla) didn't think about any alternatives, thought the accepted behavior is the best, or didn't think users would accept the alternatives.
It's really inevitable, isn't it? At some point a UI convention becomes so ingrained to so many people, that an alternative that provides 50% improvement is not enough. It would take an order of magnitude improvement to make the masses switch. Basic browser behavior seems to have hit that wall.
Not sure what your definition of "small form factor" is, but if a 17cm x 17cm motherboard qualifies, the mini-ITX boards from VIA seem like a genuinely good value. For $100-150 you get the motherboard and processor, with VGA, video-out, sound, LAN, etc.
Slashdot Mirror
I have the impression that within the last 12 or 18 months, 2D image quality has become a priority (maybe a prerequisite) among enthusiasts again. In any case, I recently got an nVidia FX 5200 card (I think the vendor is eVGA) and the quality is superb on my 19" Sony Trinitron--better than the Matrox G400 I used to use.
"Significant experience with Internet navigation."
That one went up on the company intranet's quote board!
For example, Java uses a hybrid type system where most values (strings, array) are objects and have by-reference semantics, but primitive values (ints, longs, doubles, bools, etc.) do not behave as objects and have by-value semantics. It's less elegant for the programmer than some other languages (Smalltalk, Ruby, others?) where Everything Is An Object, but makes a big difference for performance and memory consumption (and memory usage patterns).
The Java language spec also goes into quite some effort to specify a threading model. IMO this is a great boon for people who want to build serious servers in Java... contrast this to many scripting languages, which (appropriately, perhaps) seem to treat concurrent programming as an afterthought.
Or if you take a look at Bjarne Stroustrup's Design and Evolution of C++, he mercilessly rejects adding any language feature that cannot be implemented without slowing down code that doesn't use the new feature. (I can't think of any examples right now, and I can't find my copy.)
Intellectual property rights for all!
boo!
Intellectual property rights for none!
boo!
Intellectual property rights for some, miniature American flags for others!
yay!
Actually I believe today's 3D graphics cards have processors that look more like supercomputers than general purpose microprocessors--i.e. many pipelines optimized to handle streams of vectors instead of von Neumann style.
If that's still too jargony for you, then think of it as a way for objects in different processes or machines to interact with each other (almost) as if they lived in the same process. For example,
In this example, the StockQuotes object could actually be running on a different machine. In .NET, all you have to do is make the StockQuotes class extend a special system-provided class, and then add a configuration file to the server .exe and the client .exe. The rest is "magic"--you can even use ordinary constructors.
Note that unlike Java RMI, you don't have to declare an extra interface, run an extra compilation step, etc. It all just sorta happens behind the scenes...
"WARNING: In HTML 4.0, FONT is deprecated. It may become obsolete in future versions, consider using style sheets instead."
Agreed that there is a big impedence mismatch here... but if you haven't already, take a look at ASP.NET Web Forms. It does a very surprisingly good job of mimicking traditional UI programming frameworks, and with a relative minimum of ugly hackery under the covers.
They have some papers available there...
I also shelled out $60 for nothing... oh well, hopefully a few pennies of that were allocated to the anti-SCO legal fund.
Windows XP has a Windows Update utility that does the same. I think you can also download the utility for Windows 2000. Until Windows Update, I never installed any patches on my Windows box--just the big service packs. Now I'm never more than a few days out of sync.
I'd say the situation was "ironic" but I'm afraid someone will correct me...
"By the mid-nineties the new breed of computers made from off-the-shelf commodity chips arrived on the market. Those funded from the US ASCI programme and consisting of several thousand CPUs grabbed the headlines, but because of communication and memory bandwidth limitations, they often deliver very little of their potential peak performance to the user."
Oh man. SCO is soooo gonna sue your ass for saying that.
(Before anyone corrects me, I know the Open Group actually owns the trademark. But s/Unix/SCO/ just doesn't have the same ring.)
For some details, check out some of the papers by S. Greenberg. (There are tons of other links I had around but I can't find them right now.)
I think the heavy research into this kind of "browser innovation" may indeed have died five years ago. What researchers began finding out then is that people had become very conditioned to the Back/Forward/History/Bookmark behavior provided by Netscape/IE. Any deviation from that made users uncomfortable and confused.
Notice that while Opera, Firebird, and the like have provided some nice advancements, they have not changed the basic behavior of these buttons. Either they (Opera, Mozilla) didn't think about any alternatives, thought the accepted behavior is the best, or didn't think users would accept the alternatives.
It's really inevitable, isn't it? At some point a UI convention becomes so ingrained to so many people, that an alternative that provides 50% improvement is not enough. It would take an order of magnitude improvement to make the masses switch. Basic browser behavior seems to have hit that wall.
http://www.idot.com/TheStore/Peripheral/motherboar d/default_itx1.asp?Cate.id=5