You could modify the plot a bit - how about a transsexual from the planet Transylvania? It probably wouldn't survive as a series - maybe turn it into a movie musical. You could cast Susan Sarandon as Janet and Barry Bostwick as Brad... Hmm... and how about Frank-N-Furter instead of Who?
yeah - a lot of companies are discontinuing IRIX support for software, and I think it's because IRIX has been running a bit behind the times the last couple of years, probably due to SGI's financial woes. For instance, the available build of Mozilla is 1.6, but some work stuff required a fix that came in 1.7 (javascript bug fix, I think), and I found I had to build it myself. It wasn't easy to get it to run, even after I got it to compile. Lots of lib updates - had to bump our minimum (sub)version of IRIX supported up about 4 notches, which our customer that used IRIX didn't like much (customers really don't like patching production machines, I've found), though they didn't complain at all about having to build Moz 1.7 themselves.
Anyhow, I can understand Alias dumping IRIX, as it really appears to be a dying platform. Reminds me of the death crawl of Digital UNIX a few years back.
It's not that the OSI model is bad, but OSI was designed by committee and like many things designed by committee, features with questionable practical usage got into the spec on the insistence of one or more committee members. The problem is, when you try to be everything to everyone, you get a bloated, slow, difficult to implement piece of software, and that's essentially what happened to OSI.
Try to design a car that way - it needs to be big enough that a person, or in some cases, more than one person, needs to fit both 3' midgets and 9' giants and still give them a perfect view of the road, needs to have great styling that appeals to everyone, etc. It may not be impossible, but to design a car for the average person and customize for the exception is much more practical.
Also of note is that OSI people were quick to point out faults in TCP/IP as a reason to switch to their model, but before OSI could even be implemented, TCP/IP added some of the important missing capabilities and declared others as unnecessary or impossible. For instance, I recall there was a part of one layer where language needed to be translated to a neutral format - I think the Transport layer - and the TCP/IP said that was unnecessary and should be handled by the Application. The OSI people insisted that it needed to be in their layer, then struggled for two years (or more) to implement it. The last thing I heard was that the developer that tried to implement it gave up, and I don't think anyone ever successfully implemented that part of that layer (last I heard, it was decided that the TCP/IP people were correct and it should be part of the Application).
A spec can be good, especially when working with a group and trying to coordinate diverse groups (e.g., developers, publications people, and QA), but in some cases they're useless (like if you're the only developer and know exactly what you want), and in some other cases they're unnecessary. For instance, a programmer writes a self-documented interface to handle GUI actions. Other programmers then use the self-documented interface to write platform specific code that handles the functionality described in the self-documented interface, essentially making the interface documentation the spec (I see that a LOT, but I also mostly write cross platform code).
SATA itself is fairly risky on Linux due to the newness of the drivers, although I haven't had any problems with it using Mandrake 11 (one of the first distributions that contained a non-beta driver). I also have a Promise SATA (model 20376, I think) controller on my MSI K7N2 Delta board.
Debian values stability, but in the same respect, they also don't adopt new hardware as quickly, so you can't have a mobo with bleeding edge tech without adding unstable branch drivers.
The article left me with a lot of hardware questions, especially since a blue screen on Linux or Windows NT based OS's usually indicates hardware or hardware interface (driver) problems. My first question would be whether they were running SATA, since the drivers were in beta for most distributions early this year. It doesn't sound like a BIOS problem, at least not if the same system ran Windows fine, but little things like improperly configured memory timings can cause the BSOD, as well (I know this firsthand from having to use a stick of CAS 3 memory while waiting for some CAS 2.5 stuff to get off back order - the K7N2 mobo defaults to CAS 2.5).
The other thing that confounds me is Redhat contains autorpm, so if they really wanted auto-update functionality, they already have it. You can even set it up with a test system and then distribute it to other machines by putting it in an official directory that the other machines check when they run autorpm, which is exactly how most Windows production environments are run. The auto-update rationale just doesn't make sense to me - sounds more like a Windows loving admin's rationale. As for SAP, I've heard from one of my company's admins that it's easy to admin on Windows (tho that guy hero-worships Windows, so take it with a grain of salt), so if it's hard to admin on Linux, I can understand the switch from that angle.
I can vouch that they're out there - my Brother-in-law had his internet connection shut off because of a rootkit that was spamming. No run command in regedit, no visible service. I've since blocked the port it was contacting the spamming service through and since it's no longer sending, they have Internet back, but have yet to find a way to clean off the rootkit (yes, I've tried about a dozen rootkit removers, none successfully). I'm reading this thread for ideas, honestly, 'cause I'm out of them. I have not tried a bootable CD yet, but if that's unsuccessful, I'll probably just reformat their drive.
all DSL is "Up to," so it's really redundant if you know the technology. Typically, you get something near what you pay for, however, which is in contrast to cable where you can get wildly various speeds depending on time of day.
I lived on over-saturated cable for 2 years before switching to Speakeasy for sanity's sake (I was getting 56-100k at peak hours with massive lag), but I pay double what I did for cable. The fastest offered service in my state is 6M/768, but the cheapest I can find that speed in my area is $140US. Even 6M/256 with a dynamic IP can't be had for less than $65.
I don't expect much to change, as the phone companies now have a firm leash on the pricing structure by a recent FCC ruling. Want to set up ADSL2+? How about $99/month for a leased line for that? Maybe it won't be that way, but I can't see anyone taking that gamble, so the US will continue to be years behind the rest of the world. With the slow moving (and in some cases, nearly bankrupt) phone and cable companies in control of growth, I don't expect speeds above 6Mbps anytime soon - maybe in 10 years we'll get ADSL2+.
yeah, and it's not exactly like they can treat men as whores (gigolos yes...)
Sillyness aside, I completely agree - heck, extend that to all media, though the fat and/or ugly guy/girl could make a career in radio if they have a sexy voice (for some reason the 90s song "I might like you better if we slept together" comes to mind, but I've also seen DJs that fall into that category).
What exactly is not demeaning to women, a tiny breasted lesbian woman with a butch hairdo and pant suit that has a man hating attitude in-game? To sell any copies you'd better have great gameplay, have a non-offensive or tongue-in-cheek attitude and a lot of word-of-mouth from the GLBT crowd. How about modeling the character after Betty Crocker? Which would sell to anyone better, the Betty Crocker beach volleyball or DoA Extreme Beach Volleyball? Even if Betty Crocker gameplay is WAAAY better, I doubt it'd sell more, even though Betty is a composite of a generic woman made to appeal to all women of all ages. Think of it this way - why does one watch beach volleyball? It's probably not for the game, or regular volleyball would fill stadiums.
What about a dress or skirt? Where do you draw the line at being too high cut for a skirt or low cut (in the bosom) for a dress? A similar sort of criteria could be applied to guys (pulling straight from Duke Nukem, muscles?, shirt?, sunglasses? pant bulge?).
How about skintight leather? Is that demeaning or empowering (think Catwoman... er, make that Underworld... er, Bloodrayne... er, just picture it in your mind)? Skintight black leather on women is traditionally associated with S&M, often with a woman in charge (at least in US media). So in one respect you have the empowered woman in control, yet you also have the duality that most men find skintight leather sexy, whether they're into S&M or not, which is demeaning.
Dual core and dual processor aren't really all that different. The main difference is that dual core has two CPUs on the same die and therefore has shorter distances to travel to the shared memory cache(s). Yes, there's additional glue and such (stuff like shared registers and even some pipeline chunks), but nothing that's a huge fundamental difference.
For that matter, dual core processors often report as two separate processors, which potentially would cause Windows user license violations if you used a dual, dual-core setup. I believe Microsoft has found a way to identify dual-core CPUs before this even became an issue, so a 2x2 should be fine, but it was an issue.
As for the performance numbers for games, I'm not entirely surprised... maybe if games threaded themselves better, but most of the time they don't and that is a waste of an extra CPU (well, ok, some compilers may optimize the code for two CPU use, but I don't think they help much, yet). I would expect performance to be optimal in stuff with lots of independent threads like Web Servers (which usually have many processes running many threads) that don't depend on other hardware (like games, which also are throttled by the GPU).
before welfare reform in the US, there was the smart person attitude of "I'm too poor to have more kids" and dumb people attitude of "I gets me sum more welfare money if I has dem kids"
now all those big families before welfare reform want big families of their own just because it doesn't feel like a family without a swarm of kids around.
Unfortunately, these days, I think you're right, it's not stupid people that are breeding, but far too often it's religious people (and I think many of these people lack free thinking - like that silly Catholic belief that dogs don't dream and the devil makes them twitch in REM sleep... shah, right - that's as inane as the Jehovah's Witnesses age of the earth and dinosaur bones were placed there by satan - who makes up this stuff?).
No, it's a Windows feature - Microsoft considers IE an integral part of Windows;)
Unfortunately, changing brosers isn't an option - we have a script heavy tool that uses a number of Microsoft only technologies, so we're pretty much tied to IE. I'd love to blame management or the developer for only supporting IE, but I think the real reason is bad timing - JavaScript died developmentally with Netscape 4.75 and JScript progressed and added the features the developer needed with IE 5 and IE 6. By the time Mozilla emerged, the tool was entrenched in JScript code.
The developer has started working on making it firefox compatible, though it's a spare time project. Maybe next year...
The one that erases my diatribes is the Windows feature where you click a link in an e-mail or Word doc and it opens in your current open browser rather than starting a new browser (or as I'd prefer, Tab, but Tabs are evil from an interface perpective). It's not just slashdot - I've lost hours of documentation that has to be entered in a browser when I clicked on a link in a corporate mandated Word doc (and the program only runs in IE - yes, we're pretty much under Bill's thumb here, even though we mainly sell to UNIX customers).
Incidentally, Caps Lock is a legacy name from manual typewriters and the functionality changed with early electronic typewriters.
but nothing beats bugs from lack of product testing... I can probably write a phone book on XCode bugs on mac (and have submitted a bunch).
Sharky Extreme has created guides like this for several years. They used to update each one monthly, but have since switched to updating each one every four months (staggering the extreme, mid-range, and value).
They also give options between AMD and Intel and among video card manufacturers, as well as advice when shopping (e.g. check the dead pixel policy for LCDs).
I don't always agree with them (for intstance, I would spend the extra $10-$20 to get CAS 2.5 memory instead of the CAS 3 value select memory, even if it only runs at CAS-3 until/unless you have 2 chips) but in general they give pretty good buying guides.
80kg average for a fat person? Who do you think reads this, hefty 5'1" Cambodian girls?
(I apologize in advance for generalizing you like that, our chubby Cambodian goddess readers)
Ok, if you take average male height worldwide into account (which I think is 5'3" or 5'4", mainly due to lots of shorter men in eastern and SE Asia), but the average white male in the US is 5'10" (1.77m), and 80kg * 2.2lbs/kg = 176lbs is between the 50th and 75 percentile. 176 lbs is barely overweight for that height... I think you probably meant between 140 and 180kg (3-4 bills, as the idiom goes) - now there's some rolls of chub. 12 liters of joe may put one of those girthwide giants at a 50% death rate, but it'd almost certainly kill me, if they or the Cambodian goddesses didn't get to me first for this post.
did you ever notice that ingredient #3 or 4 in Mountain Dew is Orange Juice? While OJ is probably something like 1/15 as good as coffee in anti-oxidants, you're still getting some, which is better than most beverages.
Most Dew drinkers I've known drink about 10 cans a day (one programmer I work with drinks a 2 liter bottle a day), so compared to the 1.64 cups of coffee the average person drinks, the numbers are pretty close.
I had read this article, actually - I also read somewhere else that the cassettes stored all the data in decles, which makes sense with the prototyping on 2 8 bit EPROMs and only using the upper 2 bits of the high byte EPROM, but it's entirely possible I'm reading between the lines by just assuming the CPU ran in 10 bit mode all the time.
According to Wikipedia Windows added it with the release of Windows 3.1 in 1992, not Windows 3.0. Heck, I'd pretty much forgotten about it being an option in Windows 3.1, but after looking this up, I recall it all too clearly. I really didn't use virtual much until I got heavy into backgrounded applications using multifinding (something Windows prior to 95 didn't do well, though it worked sorta-ok on Windows 3.1), which was available as an extension to system 6 on mac (I recall thinking it was the coolest thing to background POV-Ray while playing games).
From a quick bit of research (such as this article, 386 enhanced mode in Windows 3.0.x automatically turned on virtual memory. I've also read that the 386 handled virtual paging on chip, so it's possible that a hardware implementation existed for Windows 3.0 in Enhanced mode and a software controlled implementation was added to Windows 3.1 for Standard mode.
From the gist of what I've read, it's possible we're both correct - Intel hardware version (built into the 386) automatically switched to disk paging when it ran out of physical memory, so technically, Windows had Virtual Memory in 3.0 when running in Enhanced Mode even though it may not have had OS handled and toggle-able virtual memory until Windows 3.1.
heh - I actually went from DOS 3.2 to DOS 3.3, but on an entirely different operating system (Apple ][)
I don't remember the first version of PC DOS I used... whatever it was, it was on a PC JR my mom borrowed from work. I remember using DOS 5.0, 6.1, 6.22 DR-DOS, as well on the PC before GUIs.
Windows 2.0 was awful from the tiny bit I used it - I much preferred GEM on DR-DOS.
Windows 3.0 was entirely too unstable, but 3.1 & 3.11 were a HUGE improvement over its predecessors. I even started preferring 3.11 to GEM, but was using it a lot at work (and Compaq's Tabworks, if anyone remembers that).
Win95 was a nice jump from 3.11 - I still consider it the second best of the 95 based OS's. 98 had stability issues and didn't add many useful features IMO. 98SE is the best of the 95 based OS's, adding USB and fixing a lot of the stability problems with 98. ME was the worst release, I've never seen an OS so bug ridden and unstable.
Of the NT based releases, all are pretty stable. 3.5.1 and 4.0 had some application crash problems, but for the most part, 2000, XP, and 2003 server have few problems that aren't application or library specific. XP without built in SP1 has some flaky disk handling issues.
I've used Debian, GenToo, SuSE, Mandrake, RedHat, and Slackware linux's, and wouldn't want to see my mom try to install any of them (she can install Windows). Linux has to do something about simplifying the partitioning. As for usability, I'm mixed on both Gnome and KDE - both have issues, and neither really identifies what an app does in the GUI, which is a big problem with open source, in general - names chosen to be cute or funny, but not identifying what it does. I'll use Eclipse as an example - looking at the name or icon, do you have any idea what it is? I'd probably expect this to be under programming tools, which would help identify that it's a programming tool, but still nothing else. Anyhow, just a peeve.
I've also used various other UNIX (based) flavors: Solaris, IRIX, SINIX/RUNIX, HP-UX, AIX, NeXT, VAX, a dinosaur 6 1MHz processor Mini that used it's own variant of UNIX (I don't know the name). Only IRIX and NeXT had usable GUIs - the others I spend in the command line.
In the other world, MacOS 3, 5-X.4. Of them, 7 and X.2 offered the most improvements. I found most of the stuff in X.3 that I liked better than X.2 were library improvements (like wchar support). Still, I didn't feel compelled to upgrade my home mac until X.4.
I've also used the OS's on the Pet, Vic 20, and C64, but have no idea what they were called. Never used AmigaOS, though I had friends that swore by it.
Macs had Virtual Memory before Windows, as well as no 640k HIMEM/LOMEM boundary, though they did have 16k page alignment issues that were a pain in the ass until they were rolled into the compiler (or maybe it was just the Symantec compiler didn't handle them automatically, but Codewarrior did - that stuff was a long time ago).
What I think they're probably referring to is memory handles, however. MacOS's memory manager used a pointer-to-a-pointer memory allocation structure called a Handle that registered the memory allocation with the memory manager. The memory manager would then periodically move the pointers, but since the Handle never changed, the user would not lose the memory.
example (Pointer tells us the address where the memory is allocated, and Handle is a pointer to that pointer): Handle-->Pointer-->[heap memory] 0xbc00 0x4000 1 2 3 4 5
memory manager finds an empty space lower in the heap and decides to move the memory there. It then updates the pointer with the new location of the memory: Handle-->Pointer-->[heap memory] 0xbc00 0x3000 1 2 3 4 5
since the handle doesn't move, the user can always be sure that dereferencing the handle always gives them their data, even if the data moves.
Prior to having a memory manager, heap fragmentation was handled (or not) by the programmer, which sometimes resulted in programs slowing down the longer they were run.
Strange that they'd use it for the Intellivision 25 in 1 system - the famicom was an 8 bit system and the Intellivision was a base 10 system (yes, you heard right - 10 bit words). I suppose you could just use 2 byte words and ignore the top 6 bits like the PC emulator does if the famcom could handle it (some old consoles couldn't handle anything larger than a word, but I may be thinking 1970s tech like the 2600).
I don't know about the UK, but in the US, federal law actually allows kids to drink booze, but most states have restrictions on it. US Federal law only restricts purchase and public possession, so in some states, if a kid steals the booze from the liquor cabinet and drinks it in a private residence, he or she's ok, while in others a kid can only drink in church and still in others the kid can't drink at all (heck, if you go to Utah, it's practically banned for adults, too). Here's a bit on the US law
Not that a kid drinking is a good idea, but my point is that there may be some religious or rite-of-passage ceremony that involves alcohol or other restricted yet legal substance (read: tobacco). Completely banning them or having age only restrictions steps on freedom of religion and likely will cause cultural rifts.
In reference to the parent, movies rated R (age 17) or lower can by definition be seen by children with parent or legal guardian supervision. Anything rated NC17 or X (on the old system) requires the attendee to be at least 17. I've heard there are liability concerns with theaters letting in underage viewers, so there is some incentive to be self-policing. The same goes for stores that sell video games.
As the subject stated, the poster was referring to is Windows Vista (nee Longhorn), which will have an internal compositor similar to OSX's Quartz layer. This compositor will take OpenGL calls and convert them into DirectX, but will only support OpenGL 1.3 or earlier and it's Microsoft's intention that that will be the last OpenGL support in OS (so it will never support shaders). This means that OpenGL in Windowed mode (composited) will be less feature rich and slower than OpenGL in fullscreen mode and as time passes, will be far inferior to using DirectX. What's not apparently clear to me is if Vista will continue to support OpenGL extensions that add the additional functionality to fullscreen mode, or if fullscreen mode will just be treated as a big window and depend on the DirectX context.
Apple currently writes and updates the OpenGL drivers on mac (both hardware and software) and doesn't support the function pointer extension method you can get in Windows. If Microsoft blocks the function pointer extension support route, then fullscreen OpenGL on Windows will be restricted to OpenGL 1.3, as well, in Vista. This brings up one of the problems with mac games - if you want the latest and greatest graphics features, you need to wait on Apple.
As for Direct3D and Id, yes, if they're writing XBox360, they're using DirectX, but it's entirely possible the PC game will support both DirectX and OpenGL through an abstraction layer.
Consumer level machines on Windows are hardly ever upgraded, either, but they are extremely cheap. Most gamers will spend more that $500 to get a non-consumer level box with non-onboard graphics and hardware sound. Apple lacks a mid-tier machine with expandability (they basically sell low-tier machines with better stuff inside). This is where the majority of gamers fall, so basically, Apple is forcing themselves out of the gaming market - Apple needs a $700-1300 machine with graphics card swapability to break into that market and they don't have it. It's my opinion that this is Steve Jobs' fault - he thinks consumers want toasters (everything included, no worries!) and professionals want Indy cars (buy custom parts to continuously get best performance and add upgraded stock parts as needed), but he misses out on the gamers want street racers (buy a stock car, but replace parts to increase performance).
The full study (pdf file) is objective and does a nice job of quoting sources, but that press release reeks of yellow journalism - quoting inside sources (consumer union, CFA and freepress, which are essentially the same, as far as I can tell), giving a biased opinion, and sensationalizing the problem.
Still, I can understand a press release bulleting facts from the study, but quoting (exclusively) inside sources? Did the writer of that thing have even one high school journalism course?
Yeah - I liked the original Tomb Raider for the same reason I liked the Prince of Persia games it was based on - part action, part puzzle solving, but usually not too much of both at the same time and had a nice difficulty ramping. I hated the Tomb Raider sequels, which seemed like they were written to appeal to hardcore shooter gamers and not many of the more casual gamers like the first one. Incidentally, I finished the first game, but gave up on the second after about 5 or 6 areas. The third I played at a friends house and seemed incredibly hard from the beginning, so I never bothered to play if further than about 10 minutes into the first level.
This is the problem with sequels - most of them mutate into something only hardcore players can appreciate and destroy their casual gamer fanbase. I played the latest Madden a couple of days ago, and that game has gone from being a fairly easy to pick up game to one that is nearly impossible, though I'm sure some practice would help (but games that have a learning curve of more than a couple of hours to be proficient are very frustrating to casual gamers, and this is one of them).
although a technicality, OpenGL is not Open Source, but the API is. If you want to be OpenGL, you need to buy a license from SGI and submit to their compatibility tests. Libraries like Mesa are OpenGL compliant, but can't say they're OpenGL because they're not certified.
The main advantage of OGL is it's a cross platform graphics library, much like C is a cross platform programming language. The disadvantage of OpenGL is that it is somewhat slow at adopting new technologies and has a lot of infighting, which are somewhat inter-related. On a mixed note, the hardware companies like ATI and nVidia basically control the direction of OpenGL, while Microsoft controls the direction of DirectX. I say that's mixed because a single company in control results in a much tighter API (look at the mass of extensions on OGL if you think their API is loose), but can stifle innovation since they tell the hardware vendor what goes in.
As far as XBox goes, the division has pulled a profit before (like when Halo 2 was released) but expects to drop back into the red until 2007. May seem like a big deal that that division has been losing money, but on the whole the worst loss of their two losing divisions (entertainment and mobile and embedded) is still covered by their least profitable division, and most of the other divisions dwarf that loss by 4-5x.
Slashdot Mirror
You could modify the plot a bit - how about a transsexual from the planet Transylvania? It probably wouldn't survive as a series - maybe turn it into a movie musical. You could cast Susan Sarandon as Janet and Barry Bostwick as Brad... Hmm... and how about Frank-N-Furter instead of Who?
yeah - a lot of companies are discontinuing IRIX support for software, and I think it's because IRIX has been running a bit behind the times the last couple of years, probably due to SGI's financial woes. For instance, the available build of Mozilla is 1.6, but some work stuff required a fix that came in 1.7 (javascript bug fix, I think), and I found I had to build it myself. It wasn't easy to get it to run, even after I got it to compile. Lots of lib updates - had to bump our minimum (sub)version of IRIX supported up about 4 notches, which our customer that used IRIX didn't like much (customers really don't like patching production machines, I've found), though they didn't complain at all about having to build Moz 1.7 themselves.
Anyhow, I can understand Alias dumping IRIX, as it really appears to be a dying platform. Reminds me of the death crawl of Digital UNIX a few years back.
It's not that the OSI model is bad, but OSI was designed by committee and like many things designed by committee, features with questionable practical usage got into the spec on the insistence of one or more committee members. The problem is, when you try to be everything to everyone, you get a bloated, slow, difficult to implement piece of software, and that's essentially what happened to OSI.
Try to design a car that way - it needs to be big enough that a person, or in some cases, more than one person, needs to fit both 3' midgets and 9' giants and still give them a perfect view of the road, needs to have great styling that appeals to everyone, etc. It may not be impossible, but to design a car for the average person and customize for the exception is much more practical.
Also of note is that OSI people were quick to point out faults in TCP/IP as a reason to switch to their model, but before OSI could even be implemented, TCP/IP added some of the important missing capabilities and declared others as unnecessary or impossible. For instance, I recall there was a part of one layer where language needed to be translated to a neutral format - I think the Transport layer - and the TCP/IP said that was unnecessary and should be handled by the Application. The OSI people insisted that it needed to be in their layer, then struggled for two years (or more) to implement it. The last thing I heard was that the developer that tried to implement it gave up, and I don't think anyone ever successfully implemented that part of that layer (last I heard, it was decided that the TCP/IP people were correct and it should be part of the Application).
A spec can be good, especially when working with a group and trying to coordinate diverse groups (e.g., developers, publications people, and QA), but in some cases they're useless (like if you're the only developer and know exactly what you want), and in some other cases they're unnecessary. For instance, a programmer writes a self-documented interface to handle GUI actions. Other programmers then use the self-documented interface to write platform specific code that handles the functionality described in the self-documented interface, essentially making the interface documentation the spec (I see that a LOT, but I also mostly write cross platform code).
SATA itself is fairly risky on Linux due to the newness of the drivers, although I haven't had any problems with it using Mandrake 11 (one of the first distributions that contained a non-beta driver). I also have a Promise SATA (model 20376, I think) controller on my MSI K7N2 Delta board.
Debian values stability, but in the same respect, they also don't adopt new hardware as quickly, so you can't have a mobo with bleeding edge tech without adding unstable branch drivers.
The article left me with a lot of hardware questions, especially since a blue screen on Linux or Windows NT based OS's usually indicates hardware or hardware interface (driver) problems. My first question would be whether they were running SATA, since the drivers were in beta for most distributions early this year. It doesn't sound like a BIOS problem, at least not if the same system ran Windows fine, but little things like improperly configured memory timings can cause the BSOD, as well (I know this firsthand from having to use a stick of CAS 3 memory while waiting for some CAS 2.5 stuff to get off back order - the K7N2 mobo defaults to CAS 2.5).
The other thing that confounds me is Redhat contains autorpm, so if they really wanted auto-update functionality, they already have it. You can even set it up with a test system and then distribute it to other machines by putting it in an official directory that the other machines check when they run autorpm, which is exactly how most Windows production environments are run. The auto-update rationale just doesn't make sense to me - sounds more like a Windows loving admin's rationale. As for SAP, I've heard from one of my company's admins that it's easy to admin on Windows (tho that guy hero-worships Windows, so take it with a grain of salt), so if it's hard to admin on Linux, I can understand the switch from that angle.
I can vouch that they're out there - my Brother-in-law had his internet connection shut off because of a rootkit that was spamming. No run command in regedit, no visible service. I've since blocked the port it was contacting the spamming service through and since it's no longer sending, they have Internet back, but have yet to find a way to clean off the rootkit (yes, I've tried about a dozen rootkit removers, none successfully). I'm reading this thread for ideas, honestly, 'cause I'm out of them. I have not tried a bootable CD yet, but if that's unsuccessful, I'll probably just reformat their drive.
all DSL is "Up to," so it's really redundant if you know the technology. Typically, you get something near what you pay for, however, which is in contrast to cable where you can get wildly various speeds depending on time of day.
I lived on over-saturated cable for 2 years before switching to Speakeasy for sanity's sake (I was getting 56-100k at peak hours with massive lag), but I pay double what I did for cable. The fastest offered service in my state is 6M/768, but the cheapest I can find that speed in my area is $140US. Even 6M/256 with a dynamic IP can't be had for less than $65.
I don't expect much to change, as the phone companies now have a firm leash on the pricing structure by a recent FCC ruling. Want to set up ADSL2+? How about $99/month for a leased line for that? Maybe it won't be that way, but I can't see anyone taking that gamble, so the US will continue to be years behind the rest of the world. With the slow moving (and in some cases, nearly bankrupt) phone and cable companies in control of growth, I don't expect speeds above 6Mbps anytime soon - maybe in 10 years we'll get ADSL2+.
yeah, and it's not exactly like they can treat men as whores (gigolos yes...)
Sillyness aside, I completely agree - heck, extend that to all media, though the fat and/or ugly guy/girl could make a career in radio if they have a sexy voice (for some reason the 90s song "I might like you better if we slept together" comes to mind, but I've also seen DJs that fall into that category).
What exactly is not demeaning to women, a tiny breasted lesbian woman with a butch hairdo and pant suit that has a man hating attitude in-game? To sell any copies you'd better have great gameplay, have a non-offensive or tongue-in-cheek attitude and a lot of word-of-mouth from the GLBT crowd. How about modeling the character after Betty Crocker? Which would sell to anyone better, the Betty Crocker beach volleyball or DoA Extreme Beach Volleyball? Even if Betty Crocker gameplay is WAAAY better, I doubt it'd sell more, even though Betty is a composite of a generic woman made to appeal to all women of all ages. Think of it this way - why does one watch beach volleyball? It's probably not for the game, or regular volleyball would fill stadiums.
What about a dress or skirt? Where do you draw the line at being too high cut for a skirt or low cut (in the bosom) for a dress? A similar sort of criteria could be applied to guys (pulling straight from Duke Nukem, muscles?, shirt?, sunglasses? pant bulge?).
How about skintight leather? Is that demeaning or empowering (think Catwoman... er, make that Underworld... er, Bloodrayne... er, just picture it in your mind)? Skintight black leather on women is traditionally associated with S&M, often with a woman in charge (at least in US media). So in one respect you have the empowered woman in control, yet you also have the duality that most men find skintight leather sexy, whether they're into S&M or not, which is demeaning.
Dual core and dual processor aren't really all that different. The main difference is that dual core has two CPUs on the same die and therefore has shorter distances to travel to the shared memory cache(s). Yes, there's additional glue and such (stuff like shared registers and even some pipeline chunks), but nothing that's a huge fundamental difference.
For that matter, dual core processors often report as two separate processors, which potentially would cause Windows user license violations if you used a dual, dual-core setup. I believe Microsoft has found a way to identify dual-core CPUs before this even became an issue, so a 2x2 should be fine, but it was an issue.
As for the performance numbers for games, I'm not entirely surprised... maybe if games threaded themselves better, but most of the time they don't and that is a waste of an extra CPU (well, ok, some compilers may optimize the code for two CPU use, but I don't think they help much, yet). I would expect performance to be optimal in stuff with lots of independent threads like Web Servers (which usually have many processes running many threads) that don't depend on other hardware (like games, which also are throttled by the GPU).
part of it is attitude
before welfare reform in the US, there was the smart person attitude of "I'm too poor to have more kids" and dumb people attitude of "I gets me sum more welfare money if I has dem kids"
now all those big families before welfare reform want big families of their own just because it doesn't feel like a family without a swarm of kids around.
Unfortunately, these days, I think you're right, it's not stupid people that are breeding, but far too often it's religious people (and I think many of these people lack free thinking - like that silly Catholic belief that dogs don't dream and the devil makes them twitch in REM sleep... shah, right - that's as inane as the Jehovah's Witnesses age of the earth and dinosaur bones were placed there by satan - who makes up this stuff?).
No, it's a Windows feature - Microsoft considers IE an integral part of Windows ;)
Unfortunately, changing brosers isn't an option - we have a script heavy tool that uses a number of Microsoft only technologies, so we're pretty much tied to IE. I'd love to blame management or the developer for only supporting IE, but I think the real reason is bad timing - JavaScript died developmentally with Netscape 4.75 and JScript progressed and added the features the developer needed with IE 5 and IE 6. By the time Mozilla emerged, the tool was entrenched in JScript code.
The developer has started working on making it firefox compatible, though it's a spare time project. Maybe next year...
The one that erases my diatribes is the Windows feature where you click a link in an e-mail or Word doc and it opens in your current open browser rather than starting a new browser (or as I'd prefer, Tab, but Tabs are evil from an interface perpective). It's not just slashdot - I've lost hours of documentation that has to be entered in a browser when I clicked on a link in a corporate mandated Word doc (and the program only runs in IE - yes, we're pretty much under Bill's thumb here, even though we mainly sell to UNIX customers).
Incidentally, Caps Lock is a legacy name from manual typewriters and the functionality changed with early electronic typewriters.
but nothing beats bugs from lack of product testing... I can probably write a phone book on XCode bugs on mac (and have submitted a bunch).
Sharky Extreme has created guides like this for several years. They used to update each one monthly, but have since switched to updating each one every four months (staggering the extreme, mid-range, and value).
They also give options between AMD and Intel and among video card manufacturers, as well as advice when shopping (e.g. check the dead pixel policy for LCDs).
I don't always agree with them (for intstance, I would spend the extra $10-$20 to get CAS 2.5 memory instead of the CAS 3 value select memory, even if it only runs at CAS-3 until/unless you have 2 chips) but in general they give pretty good buying guides.
80kg average for a fat person? Who do you think reads this, hefty 5'1" Cambodian girls?
(I apologize in advance for generalizing you like that, our chubby Cambodian goddess readers)
Ok, if you take average male height worldwide into account (which I think is 5'3" or 5'4", mainly due to lots of shorter men in eastern and SE Asia), but the average white male in the US is 5'10" (1.77m), and 80kg * 2.2lbs/kg = 176lbs is between the 50th and 75 percentile. 176 lbs is barely overweight for that height... I think you probably meant between 140 and 180kg (3-4 bills, as the idiom goes) - now there's some rolls of chub. 12 liters of joe may put one of those girthwide giants at a 50% death rate, but it'd almost certainly kill me, if they or the Cambodian goddesses didn't get to me first for this post.
did you ever notice that ingredient #3 or 4 in Mountain Dew is Orange Juice? While OJ is probably something like 1/15 as good as coffee in anti-oxidants, you're still getting some, which is better than most beverages.
Most Dew drinkers I've known drink about 10 cans a day (one programmer I work with drinks a 2 liter bottle a day), so compared to the 1.64 cups of coffee the average person drinks, the numbers are pretty close.
I had read this article, actually - I also read somewhere else that the cassettes stored all the data in decles, which makes sense with the prototyping on 2 8 bit EPROMs and only using the upper 2 bits of the high byte EPROM, but it's entirely possible I'm reading between the lines by just assuming the CPU ran in 10 bit mode all the time.
According to Wikipedia Windows added it with the release of Windows 3.1 in 1992, not Windows 3.0. Heck, I'd pretty much forgotten about it being an option in Windows 3.1, but after looking this up, I recall it all too clearly. I really didn't use virtual much until I got heavy into backgrounded applications using multifinding (something Windows prior to 95 didn't do well, though it worked sorta-ok on Windows 3.1), which was available as an extension to system 6 on mac (I recall thinking it was the coolest thing to background POV-Ray while playing games).
From a quick bit of research (such as this article, 386 enhanced mode in Windows 3.0.x automatically turned on virtual memory. I've also read that the 386 handled virtual paging on chip, so it's possible that a hardware implementation existed for Windows 3.0 in Enhanced mode and a software controlled implementation was added to Windows 3.1 for Standard mode.
From the gist of what I've read, it's possible we're both correct - Intel hardware version (built into the 386) automatically switched to disk paging when it ran out of physical memory, so technically, Windows had Virtual Memory in 3.0 when running in Enhanced Mode even though it may not have had OS handled and toggle-able virtual memory until Windows 3.1.
heh - I actually went from DOS 3.2 to DOS 3.3, but on an entirely different operating system (Apple ][)
I don't remember the first version of PC DOS I used... whatever it was, it was on a PC JR my mom borrowed from work. I remember using DOS 5.0, 6.1, 6.22 DR-DOS, as well on the PC before GUIs.
Windows 2.0 was awful from the tiny bit I used it - I much preferred GEM on DR-DOS.
Windows 3.0 was entirely too unstable, but 3.1 & 3.11 were a HUGE improvement over its predecessors. I even started preferring 3.11 to GEM, but was using it a lot at work (and Compaq's Tabworks, if anyone remembers that).
Win95 was a nice jump from 3.11 - I still consider it the second best of the 95 based OS's. 98 had stability issues and didn't add many useful features IMO. 98SE is the best of the 95 based OS's, adding USB and fixing a lot of the stability problems with 98. ME was the worst release, I've never seen an OS so bug ridden and unstable.
Of the NT based releases, all are pretty stable. 3.5.1 and 4.0 had some application crash problems, but for the most part, 2000, XP, and 2003 server have few problems that aren't application or library specific. XP without built in SP1 has some flaky disk handling issues.
I've used Debian, GenToo, SuSE, Mandrake, RedHat, and Slackware linux's, and wouldn't want to see my mom try to install any of them (she can install Windows). Linux has to do something about simplifying the partitioning. As for usability, I'm mixed on both Gnome and KDE - both have issues, and neither really identifies what an app does in the GUI, which is a big problem with open source, in general - names chosen to be cute or funny, but not identifying what it does. I'll use Eclipse as an example - looking at the name or icon, do you have any idea what it is? I'd probably expect this to be under programming tools, which would help identify that it's a programming tool, but still nothing else. Anyhow, just a peeve.
I've also used various other UNIX (based) flavors: Solaris, IRIX, SINIX/RUNIX, HP-UX, AIX, NeXT, VAX, a dinosaur 6 1MHz processor Mini that used it's own variant of UNIX (I don't know the name). Only IRIX and NeXT had usable GUIs - the others I spend in the command line.
In the other world, MacOS 3, 5-X.4. Of them, 7 and X.2 offered the most improvements. I found most of the stuff in X.3 that I liked better than X.2 were library improvements (like wchar support). Still, I didn't feel compelled to upgrade my home mac until X.4.
I've also used the OS's on the Pet, Vic 20, and C64, but have no idea what they were called. Never used AmigaOS, though I had friends that swore by it.
Macs had Virtual Memory before Windows, as well as no 640k HIMEM/LOMEM boundary, though they did have 16k page alignment issues that were a pain in the ass until they were rolled into the compiler (or maybe it was just the Symantec compiler didn't handle them automatically, but Codewarrior did - that stuff was a long time ago).
What I think they're probably referring to is memory handles, however. MacOS's memory manager used a pointer-to-a-pointer memory allocation structure called a Handle that registered the memory allocation with the memory manager. The memory manager would then periodically move the pointers, but since the Handle never changed, the user would not lose the memory.
example (Pointer tells us the address where the memory is allocated, and Handle is a pointer to that pointer):
Handle-->Pointer-->[heap memory]
0xbc00 0x4000 1 2 3 4 5
memory manager finds an empty space lower in the heap and decides to move the memory there. It then updates the pointer with the new location of the memory:
Handle-->Pointer-->[heap memory]
0xbc00 0x3000 1 2 3 4 5
since the handle doesn't move, the user can always be sure that dereferencing the handle always gives them their data, even if the data moves.
Prior to having a memory manager, heap fragmentation was handled (or not) by the programmer, which sometimes resulted in programs slowing down the longer they were run.
Strange that they'd use it for the Intellivision 25 in 1 system - the famicom was an 8 bit system and the Intellivision was a base 10 system (yes, you heard right - 10 bit words). I suppose you could just use 2 byte words and ignore the top 6 bits like the PC emulator does if the famcom could handle it (some old consoles couldn't handle anything larger than a word, but I may be thinking 1970s tech like the 2600).
I don't know about the UK, but in the US, federal law actually allows kids to drink booze, but most states have restrictions on it. US Federal law only restricts purchase and public possession, so in some states, if a kid steals the booze from the liquor cabinet and drinks it in a private residence, he or she's ok, while in others a kid can only drink in church and still in others the kid can't drink at all (heck, if you go to Utah, it's practically banned for adults, too). Here's a bit on the US law
Not that a kid drinking is a good idea, but my point is that there may be some religious or rite-of-passage ceremony that involves alcohol or other restricted yet legal substance (read: tobacco). Completely banning them or having age only restrictions steps on freedom of religion and likely will cause cultural rifts.
In reference to the parent, movies rated R (age 17) or lower can by definition be seen by children with parent or legal guardian supervision. Anything rated NC17 or X (on the old system) requires the attendee to be at least 17. I've heard there are liability concerns with theaters letting in underage viewers, so there is some incentive to be self-policing. The same goes for stores that sell video games.
As the subject stated, the poster was referring to is Windows Vista (nee Longhorn), which will have an internal compositor similar to OSX's Quartz layer. This compositor will take OpenGL calls and convert them into DirectX, but will only support OpenGL 1.3 or earlier and it's Microsoft's intention that that will be the last OpenGL support in OS (so it will never support shaders). This means that OpenGL in Windowed mode (composited) will be less feature rich and slower than OpenGL in fullscreen mode and as time passes, will be far inferior to using DirectX. What's not apparently clear to me is if Vista will continue to support OpenGL extensions that add the additional functionality to fullscreen mode, or if fullscreen mode will just be treated as a big window and depend on the DirectX context.
Apple currently writes and updates the OpenGL drivers on mac (both hardware and software) and doesn't support the function pointer extension method you can get in Windows. If Microsoft blocks the function pointer extension support route, then fullscreen OpenGL on Windows will be restricted to OpenGL 1.3, as well, in Vista. This brings up one of the problems with mac games - if you want the latest and greatest graphics features, you need to wait on Apple.
As for Direct3D and Id, yes, if they're writing XBox360, they're using DirectX, but it's entirely possible the PC game will support both DirectX and OpenGL through an abstraction layer.
Consumer level machines on Windows are hardly ever upgraded, either, but they are extremely cheap. Most gamers will spend more that $500 to get a non-consumer level box with non-onboard graphics and hardware sound. Apple lacks a mid-tier machine with expandability (they basically sell low-tier machines with better stuff inside). This is where the majority of gamers fall, so basically, Apple is forcing themselves out of the gaming market - Apple needs a $700-1300 machine with graphics card swapability to break into that market and they don't have it. It's my opinion that this is Steve Jobs' fault - he thinks consumers want toasters (everything included, no worries!) and professionals want Indy cars (buy custom parts to continuously get best performance and add upgraded stock parts as needed), but he misses out on the gamers want street racers (buy a stock car, but replace parts to increase performance).
The full study (pdf file) is objective and does a nice job of quoting sources, but that press release reeks of yellow journalism - quoting inside sources (consumer union, CFA and freepress, which are essentially the same, as far as I can tell), giving a biased opinion, and sensationalizing the problem.
Still, I can understand a press release bulleting facts from the study, but quoting (exclusively) inside sources? Did the writer of that thing have even one high school journalism course?
Yeah - I liked the original Tomb Raider for the same reason I liked the Prince of Persia games it was based on - part action, part puzzle solving, but usually not too much of both at the same time and had a nice difficulty ramping. I hated the Tomb Raider sequels, which seemed like they were written to appeal to hardcore shooter gamers and not many of the more casual gamers like the first one. Incidentally, I finished the first game, but gave up on the second after about 5 or 6 areas. The third I played at a friends house and seemed incredibly hard from the beginning, so I never bothered to play if further than about 10 minutes into the first level.
This is the problem with sequels - most of them mutate into something only hardcore players can appreciate and destroy their casual gamer fanbase. I played the latest Madden a couple of days ago, and that game has gone from being a fairly easy to pick up game to one that is nearly impossible, though I'm sure some practice would help (but games that have a learning curve of more than a couple of hours to be proficient are very frustrating to casual gamers, and this is one of them).
doh - didn't proofread again... I meant the API is Open (as in openly available to read and use), not Open Source.
although a technicality, OpenGL is not Open Source, but the API is. If you want to be OpenGL, you need to buy a license from SGI and submit to their compatibility tests. Libraries like Mesa are OpenGL compliant, but can't say they're OpenGL because they're not certified.
The main advantage of OGL is it's a cross platform graphics library, much like C is a cross platform programming language. The disadvantage of OpenGL is that it is somewhat slow at adopting new technologies and has a lot of infighting, which are somewhat inter-related. On a mixed note, the hardware companies like ATI and nVidia basically control the direction of OpenGL, while Microsoft controls the direction of DirectX. I say that's mixed because a single company in control results in a much tighter API (look at the mass of extensions on OGL if you think their API is loose), but can stifle innovation since they tell the hardware vendor what goes in.
As far as XBox goes, the division has pulled a profit before (like when Halo 2 was released) but expects to drop back into the red until 2007. May seem like a big deal that that division has been losing money, but on the whole the worst loss of their two losing divisions (entertainment and mobile and embedded) is still covered by their least profitable division, and most of the other divisions dwarf that loss by 4-5x.