> Outlook isn't an e-mail server, its a client. Get a clue.
The original poster was rather obviously going for a +5, Funny.
Re:Mining Asteroids and other economical tasks in
on
Collecting Stardust
·
· Score: 1
> Your post makes an excellent argument for the > banning of font weight variations in body text. > Of course, no one expects good typography on > Slashdot, home of the all-italics front page.
Heh. I do that for a reason. I've found that I tend to drone on when I write stuff, so many readers get bored before they can get to the meat of what I'm saying. So I try to highlight (well, make bold) words that represent the important points of the paragraphs. Some readers (granted, not all) find it easy to quickly scan a long post if they can pick out important key words.
For instance, when I said "So you start up a pipeline of automated mining probes to Ceres. Much like on modern microprocessors, the initial hardware cost would be greater, but you'd be able to transport mining materials at a much faster and more reliable rate than you otherwise would", a speed reader would see the boldfaced words " pipeline... automated mining probes... Ceres... microprocessors... faster... more reliable rate", and they might be able to make out the general gist of what I was talking about. I probably should have added a boldfaced term like "high initial investment" to balance out what I was saying there, but I was writing a lot of stuff on the fly. This isn't a term paper or a news article, you know.:p
Granted, I stole this technique from Mad and Cracked magazines, so it might not be the best approach.:)
> But what kind of server processor is the Itanium? Given the > unimpressive specint scores, it doesn't look like a processor > for those mass volume servers (web, mail, file and print etc). > Does it even look like a cpu for larger RDBMS servers?
Well, specint isn't necessarily representative of all those types of servers. A lot of these servers are data transfer intensive and essentially best on cpus with insane amounts of cache and a really high bandwidth cpu to chipset data path. Itanium II seems to have this.
Of course, that doesn't mean that there aren't better options. I could probably put together a multiple machine array of cheapish x86 Linux or BSD servers (that is, if I had a little more experience with clustering and similar methodologies) for the same price, and I'd be ending up with a more reliable setup (more machines mean better redundancy, and overall performance probably would be faster due to greater parallelism in every component, not just the cpu -- for example, the faster Itanium chip might be able to write to two memory modules at once, but it probably can't write to and read from as many hard drives as a multiple setup of slower processors could).
> To me it doesn't look like a processor for corporate servers, > maybe scientific clusters?
That is a toughie. Itanium in scientific applications goes in three categories: A. Really shitty (if it's an x86 app) B. Kinda slowish to kinda fastish (if you just compile it straight) C. Faster than a drug dealer who accidentally walked into a DEA New Year's Bash carrying a wad of dimebags (if you very very carefully rewrite the code and apply smart compiler options over several test builds).
Obviously, the software-optimization requirements of the Itanium family (remember, the nature of the chip's architecture means that it's an in-order processor; out of order processors, which started appearing in x86 with the Pentium Pro and earlier in other architectures, optimize and rearrange code on the fly, so you can get reasonably optimal performance even with only basically competent code) will turn off many people. But those who are willing to do the work will see the benefit they're looking for.
> If I currently have x86 servers, and I'm looking at the next step, > the Opteron would be the most attractive option (IF AMD doesn't > screw up;) ). The PowerPC looks just about as attractive as > the Itanium - for practical purposes they're both incompatible > with x86.
Yeah, x86-64 is really neat for the more mainstream server market. You get excellent performance in already existing apps, and recompilation for higher performance is trivial (well, comparatively).
> Would be interesting to know what's happening in Dell internally. > The Opteron really looks like a better match for them than the > Itanium;).
Don't delude yourself. Dell uses AMD as a bargaining chip against Intel. If Mike Dell every says something like "We are evaluating AMD products", he means "Hey, Intel, I need you to give me a better discount on your new chips". Dell/AMD rumours tend to pop up lik clockwork every three months. But I guess you never know....
> While Microsoft could be kingmaker/decider for the Athlon64, > Linux will be enough for Opteron.
Well, Athlon 64 should be okay either way. I mean, the 64-bit mode isn't even the best feature, as I've noted before. It's the low latency, on-die memory controller and possibly the HyperTransport connections that interest me, and you'd get benefits from those even in conventional 32-bit Windows systems.
-JC
PS: I apologize if my comments about Itanium II are exaggerated. I haven't seen recent scores outside of stuff like spec, and I was disappointed particularly with Merced/Itanium, so the situation may be not as intense nowadays.
Mining Asteroids and other economical tasks in 0g
on
Collecting Stardust
·
· Score: 2, Interesting
> I found it interesting that commercial mining of asteroids was mentioned > in the third article. Sure, raw materials are plentiful in asteroids, > but wouldn't the cost of getting there far outweigh the benefits > of the plentiful resources? I guess this would be practical if/when > we run out of certain ores, or as an "While we're here, we might > as well" measure, but I can't see it going anywhere otherwise > until the price of space travel drops dramatically.
Yeah, it's an "economy of scale" thing. Once we have regular (and by regular I mean many companies each making daily launches) access to space, the incremental cost of space travel will be far, far below what we pay now in terms of cost per unit mass.
And, thing is, it doesn't cost much in terms of energy to get to any given asteroid once you're outside the gravity well of Earth. The only cost would really be time, since momentum keeps you going until you want to brake. So you start up a pipeline of automated mining probes to Ceres. Much like on modern microprocessors, the initial hardware cost would be greater, but you'd be able to transport mining materials at a much faster and more reliable rate than you otherwise would.
And mining in a low gravity environment (Ceres, the largest asteroid, has less than a thirtieth of Earth's surface gravity) should be far less energy intensive than mining on Earth, where you have to expend unbelievable amounts of energy to merely move mined elements from the mines to the transports, and where you're severely limited in machine travel range due to friction related to that nasty 9.8m/s^2 that we have to put up with all the time here.
And some asteroids will probably have minable minerals in abundance, and remote spectral analysis will be able to allow us to identify mineral compositions from millions of miles away. That does away with a huge portion of the work involved in mining on Earth, where you have to often indirectly figure out what minerals are hidden beneath the surface, wasting time by drilling all over the place. That costs money continuously, and you could save on that with asteroid mining.
There are probably some other things in asteroid mining that I'm not thinking of. I'm not miner, and my "engineering" knowledge is rather amateurish (I'm more of a theoretical guy, being a programmer and all). Maybe wear and tear of machines would be less (due to both that lesser gravity we discussed above and the total lack of atmosphere or microorganisms to break down the machine parts). Maybe there are social benefits, like higher morale for the folks who like to work in larger and more open environments (this century, maybe "getting away from it all" will mean accepting a job that lets you work fifty million miles away for a year or two).
Economies of scale, my man. We have to force ourselves to pay the high costs to continue the proliferation of new technology. I we don't do this, we go stagnant. Space launches these days are often more expensive than they were in the old days, even when you take inflation into account. It's because we stopped doing them so often. It saves a few billion dollars a year in the short term. But we might have made back those billions -- and far more -- by now from the direct and indirect benefits of building up an economy-of-scale framework of space travel. I mean, heck, we landed on Luna six years before I was born. We have space probes ten light hours away. We also have advanced manufacturing facilities on Earth now. Had we continued our push into space instead of borking it all in the seventies through the nineties, I imagine that we would have simple factories producing complex objects on the cheap (at least with products that require tons of work on a small amount of raw material -- no friction and no gravity equals huge cost reductions!) now. Chip fabbing companies would be looking at near to medium term options for building astro-fabs. Do you realize how much more precision in the sub-nanometer range you'd get in building chips when you don't have to worry about gravity and vibrations from the local landmass and atmospheric variables? I mean, crap, we probably would be able to further accelerate process shrinks. 3.0GHz on a 130nm process? What a backwards technology for a space-enabled 2003!
Economies of scale, my man. Startup costs are [almost] always prohibitive. That's how science and technology works. Live with it. But don't work to inhibit it.
I dunno about that. On Star Trek last year, they had a comet with earthlike gravity. Now that's damned impressive, and it must be true since its on TV.
Maybe there are other unique comets out there, like ones with bizarre technobabble-inhibiting EM fields and occasional spaceberry orchards. ^_^
> > the BSA... logged into the... FTP... with the address "guest@nowhere.com"
> Ok, I agree, but let's think here. I generally use the > address "-a@" when logging into anonymous FTP. This is > because (shock) I believe that anonymous FTP should be, > well, anonymous. If someone really wants to track me > down, he can get my IP address easily. > Is it "OK" for normal folks to do this, yet "not OK" for > the BSA? Or is it "not OK" for anyone?
A few things: First, your address ("-a@") would just give an error in any normal email client. The email address that the BSA provided was owned by someone else.
Secondly, the BSA was using some sort of automated crawler, so they could have possibly posted this email address to hundreds or thousands of places, which means that the person who owns nowhere.com will be getting a bunch of erroneous hate mail and a huge load of extra spam in the future. This could affect this person from both a financial (bandwidth abuse, loss of productive time, etc.) and an emotional (I don't know about you, but I get absolutely thrown into a breakdown when I get large amounts of flames and spam) standpoint.
And then there's the fact that the BSA is using this person's email address for commercial gain. They stand to make money indirectly for doing this service from their client organizations. There's a lot of collateral damage involved, but the end result will likely be a temporary decrease in intellectual property infringement of software products (it's kind of like if we napalmed all of the Middle East: yeah, other countries would get messed up, but the primary target of removing Saddam Hussein from power would be achieved). So it's an ethics thing. Are you okay with making millions of dollars by knowingly hurting millions of innocents? If you are, then it's all good.
Incidentally, it never occurred to me that these domains are owned by somebody, and I now regret using them in my anonymized passwords. I think that I will use a method like the one you use when I want to be anonymous.
> You could put a P266 in your socket > 5 mb when you upgraded from a P100?
You might not have been able to do that, but the Centaur (the cpu design team that nowadays makes VIA's C3 chips) 240MHz WinChip could make that jump, if I recall correctly.
I am not certain if the WinChip 2 or 2A could work on Socket 5. They were basically just WinChips with 3DNow! tacked on, but they might have been out of spec for Socket 5, ever so slightly.
Oh, and there were probably "upgrade" chips, which were newer processors with adaptor thingies between the cpu and the socket.
> The following is a simplistic view of things, but we are talking > about a 64bit processor. Remember the Itaniums Intel is selling > are running at around 1GHz - 1.5GHz I believe and they run > circles around the 3Ghz P4.
That's overgeneralized. The 3GHz P4 is very much faster at most tasks than the 1.0GHz Itanium II, which is the fastest instance of the chip that has been given entries at spec.org. The reason why the Itanium II appears much faster is that you only see benchmarks that relate to its very narrow field of marketing. It's a server processor. You won't see it tested in areas more suited to general purpose computing (games, office suites, etcetera). And, hell, the Itanium sucks in specint, one half of the single processor version of the most prolific server benchmark suite in the world. The 900MHz (fastest speed submitted -- for some reason, they only gave specfp scores for 1000MHz, unless I missed an entry or few) Itanium II gets 674, compared to scores above 1100 for the 3.06GHz P4. That's a whopping 63% difference! The fastest Itanium II is almost 40% slower than the top of the line non-Xeon Pentium 4!
The Itanium II does fantastically in specfp -- bested, I believe, only by the DEC Alpha, which is sadly being pushed under the carpet for reasons more political than I'd like (Alpha IP is owned by HP and Intel, the companies that created the Itanium's core architecture) -- and many other benchmarks. But you can't simply ascribe a single, simple feature to the performance advantages of the processor. Yeah, the processor can address a 64-bit memory space and, yeah, the processor has 64-bit GP registers. But you're ignoring many other features piled on top. Itanium II is a server processor, so it can afford to have some extra doodads added to it, doodads that would be considered financially unfeasible on mainstream processors.
Hmmm, I did a quick google search, so I apologize if I pulled incorrect info on the following:
The Itanium II has a more than a megabyte and a half of cache memory on the die of the processor. It seems to optionally go up to 3MB on-die L3 cache. In comparison, the Pentium 4 has 512KB cache (there's some more cache, the L1, but that's inclusive), and the Athlon XP has either 384KB or 640KB cache (depending on whether you're counting the older Thoroughbred or the newer Barton). So the Itanium gets about three times as much cache memory on the processor die!
Itanium II has a 400MHz, 128-bit data path to the chipset. Pentium 4 is 533MHz, 64-bit. So the P4 gets a chipset that can send it 4.27 GB per second while the Itanium II gets a chipset sending it data at 6.4GB/s.
The Itanium II has more functional/execution units. The Itanium II gets predication, which is a very expensive (in terms of how much bulk it adds to the die) feature that effectively gets rid of a lot of the penalty associated with branch misprediction (a problem which is rather huge with the trillion-stage netburst microarchitecture of the Pentium 4, though I'm told that the multithreading implementation of the P4 can help alleviate some of that).
The number of bits in the processor don't matter *that* much, not after the 32-bit level. Yeah, it helps, but you have to take the whole package into account. A 64-bit scalar one-stage processor with a ten-byte, off-die cache would get its ass kicked mercilessly by an 80486DX-50.
To take another tack: I'm somewhat interested in possibly purchasing an Athlon 64 late this year or early next year. But if the Athlon 64 was just an Athlon with 64-bit extensions, I wouldn't give it the time of day. I'm interested because the Athlon 64 will have an on-die memory controller. I'm interested because the Athlon 64 will support twice as many registers as a typical x86 chip (which may decrease the need for cache accesses, which could increase performance on my recompiled linux apps). Either of these two advantages promise a far greater advantage for me than the simple increase in register size and memory addressability.
> For most people, security is not important. Top performance is not > important. Optimum configuration is not important. Control is not > important. Not having to power toggle is not important.
> Being able to put the CD in the CD drive, press a button a couple > of times, reboot, > and get what you want is VERY IMPORTANT. NOT THINKING is > VERY IMPORTANT.
> Users want things that work like coffee machines. You plug it in > and it works. If you want a different coffee machine, you get > a different coffee machine and plug it in and it works. Windows > makes computers a lot more like coffee machines than Linux does. > Having to turn your computer on an off to get a new feature is much > less of a problem than having to know what to type to get a new > feature. Linux wants you to figure stuff out. Microsoft wants > your money.
I'm not sure that I can entirely agree with you. I don't think that Windows is inherently easier to deal with. I think that the fact that Windows is by far the dominant operating system means that all hardware and nearly all software developers pay specific attention to how their products work with Windows. This isn't due to any particular feature of the operating system. It's simply because they have to make it easy on the popular platform.
For most applications, installations go like this: * In Windows 2000, I'd open up my web browser of choice. Then I'd go to one of the download sites or perhaps to google.com, and I'd search for an application online. I'd download the application (usually going through a few screens of ad-laden BS and choosing various mirror that are closest to me. I'd go to the location of the file (either through Start->Run or using a file manager like PowerDesk or simply by using Opera's excellent download manager). Then I'd double-click on the file. A "wizard" application opens and asks me a series of successive questions about where it wants me to install the program, whether I agree to a fifty-page application-specific legal document, where in the Start Menu I want this thing to go, and whether I want shortcuts placed in various locations. Then the program (sometimes) tells me that it needs to reboot, and I hit "OK". It reboots, occasionally does some further installation, and then I'm set. I would do all this every time for each program.
* In Mandrake Linux 9.0, I go to the package managing program (By clicking on "K->Configuration->Packaging->Install Software", or by hitting ALT-F2, typing "rpmdrake" and hitting ENTER). I change the little radio button group from "Mandrake choices" to "All Packages". I type a program name into the Search bar and hit the Search button (or I'd just look through the efficiently categorized list of programs). I check the checkboxes of any and all programs that I want to install, and I hit the Install button. Then I sit back as the installer automatically downloads, installs and configures all the applications I selected, grabbing any prerequisite programs from the servers automatically. In the time it took to search for, download and begin the installation of a program in Windows, I've finished installing the Linux app. Before I've finished mucking with the Next->Next->Next->Finish screen of that installation, another Linux app has finished installing (without me needing to click on anything more). By the time my computer has rebooted into Windows 2000 from that one install, the Mandrake Linux package manager has installed six or seven different apps (and I only had to click the "Install" button once). And you know what? Everything is installed into logical, well thought-out places. Instead of going into "Programs" and having to scroll down a clumsy list of company names to find the app you've installed (difficult especially if you've forgotten the company name!) like I see on Windows 2000, the Mandrake installer puts everything into intuitive, user-friendly subcategories. Stuff that uses the network is in "Networking". All my email programs are in "Networking->Mail". My news (usenet) readers (Pan for binary downloading, Mozilla Messenger for general reading) are in "Networking->News". Card games are specifically in "Amusement->Cards". Know what I have to do to find all my card games in Windows? I have to look in "Programs->Accessories->Games" and figure out which ones are card games. Then I have to look in every subgroup in Programs (the aforementioned company names) to check and see which ones have card games. I have to *memorize* this stuff in Windows. In Linux, I just go to "Amusement->Cards". Holy crap, you can't get any more obvious than that! Oh, I need to watch TV? "Multimedia->Video". I have to put this 800MB SVCD onto my 700MB CD without data loss? "Applications->Archiving->Cd Burning". I want my kid to learn stuff? "Applications->Edutainment". I'll never accidentally click on the "Hot Boobies" interactive porno game when I intended to show my female colleague my PG-rated "Hard Bodies" fitness management program. It just won't happen, because one of them would be in "Amusement->Sex" while the other would be found in... well, I dunno, maybe "Office->Time Management" or "Applications->Sciences->Health" (probably the latter). In Windows, it's a crapshoot. Yeah, real user-friendly.
It's better than that. If I want to be lazy in windows, I can set up links on the taskbar, Office Bar (if I spend the untold hundreds for their Office product) or desktop to the programs. I could also (with a non-native third-party extension program) map programs to a Win-key combination. Currently, in Windows, I use Win-O to open Opera, Win-M to open my Mail program (Eudora), Win-X to open eXcel, Win-W to open Word, and so forth. Mandrake natively supports key combinations to open programs, and I believe you can differentiate between the two Win keys if you had the desire (LeftWin+W goes to OpenOffice Writer, RightWin+W goes to KWord, for example). I don't use it much, for the following reasons: Mandrake (and, by the way, I'm using KDE to manage my gui, so ymmv if you use other programs) allows me to put links to programs on the desktop. It also allows me to put links to programs on my taskbar. But it lets me configure these links in interesting ways (and we're not talking about difficult configuration; we're talking about Right-click-on-panel->Size->Large and similarly easy means). I can have (and I do) two levels of bars with these links. I have a big taskbar with my extra-lazy application links. These are full-sized icons, so they're easy to click on when I'm too slothful to competently use the mouse. On the bar right above it, I have (among other things) medium-sized icons for a whole bunch of programs that I tend to frequently use, like my text editor and my web browser. Incidentally, that bar also has a dictionary bar, an ascii character picker (I could paste odd characters into any program instead of having to rely on some arcane, application specific "Insert->Character" features that don't work universally), a web news scroller, an advanced clipboard manager (you know how more recent versions of Microsoft Office allow for multiple clipboard levels? Well, KDE's Klipper application does this for *every app*) and quick shortcuts to lock the computer or to logout. But I don't every really use those icons very often. Why not? Well, I have session management turned on. Whenever I turn on my computer, the system reloads active programs so that I can continue from where I left off. And most of my programs (Opera, Konsole, Konqueror, Kate, Pan) have their own internal session management, so I don't have to click on bookmarks or whatever to get to where I was before. The other thing that makes it easy to not have to move my mouse to hit those "shortcut" icons is the nature of linux pathing. Remember when I installed those programs above? Well, the executables are automatically put into a place that's in the system path. Most of the programs have pretty short filenames for the binaries. Most of the time, if I want to run the program and happen to remember the program's executable name, I hit F2, type in the program name and hit ENTER. F2,pan,ENTER. F2,mozilla,ENTER. F2,kate,ENTER. Heck, even those programs that I installed through other means than the Mandrake Package Manager (sometimes, you can install the very latest versions of programs before they get packaged) will work with this. F2,gmplayer,ENTER runs the GUI version of MPlayer, the only multimedia program that can play just about every format out there, from mpeg to avi to asf to quicktime to rm to ogg to DVDs and Mode 2 SVCDs (which I *almost* have working in windows, with some occasional bizarre inconsistencies). I have to have three or four different players installed on Windows 2000 to get that sort of compatibility, and that's ignoring the easier interface and hotkeys in MPlayer.
The hardware side is sometimes easier in Windows, though my experience doesn't exactly completely agree with that. I have a somewhat generic 5.1 sound card with no discernable markings on it. It took me *forever* to find the drivers for Windows 2000, and it was actually Linux (and its "harddrake" hardware manager) that gave me enough clues about the main chips on this soundcard to find out that it was from some C-Media company or something like that. Some time after, I found the Windows drivers and everything went swimmingly. Know how much I had to look for the Linux drivers? They were already there. They. Were. Already. There. When I installed Mandrake 9.0, the sound card was autodetected and autoconfigured. I'll give you that an earlier version of Mandrake (8.1 or 8.2) didn't properly detect the card when I first installed it, but the drivers for it were in there and it was comparatively trivial to tell the computer this (I put the name of the sound card driver module, something like cmpci.so, into some configuration startup text file) compared to the herculean effort to get it running in Win2k.
My TV card used to be an outdated Hauppauge that didn't support scaling past 640x480. I had to guess which drivers it used from Hauppauge.com, and I was eventually successful in Windows 2000. The scaling thing was annoying, but it worked, except that the video capture seemed problematic. A few months later, Win2k went crappy on me, and I had to reinstall it. For the life of me, I could not remember which drivers and in which order I needed to install, and I couldn't get the TV program to work, no matter how hard I tried. So I did it in Linux. The Mandrake 8.x install autodetected, autoinstalled, autoconfigured. And it installed a whole wad of different TV programs that could use the TV card. One of them (xawtv) could inexplicably scale the TV screen to whatever dimension I wanted. I still use that program.
Heck, this past Christmas, my parents bought me an All-in-Wonder RADEON 8500. The installs worked fine on both systems. Unfortunately, I only have a choice of one TV program on Windows, and that program makes the system crash after I try to shut it down. I still have the exact same vareity of TV programs on Linux, and if I wanted to use my brain, I could probably pretty easily figure out how to broadcast the TV image onto my local network.
The USB CD burner that I recently gave up was fun. When I installed it in Windows, the Windows Media Player tried to autoinstall a "CD Burning Plugin" which caused all my CD drives (even the CD-ROM) to disappear (until I got all technical and figured out how to remove the stupid plugin). Mandrake 9 (and 8.2, I think) just installed it. No fuss. It worked on installation.
I think that my newly installed ATAPI burner is easier to install in Windows, but that's because (as I mentioned above) Windows gets the third-party support. I did have to change two or three text files (though I didn't need any installation program) to get this burner working in Linux. I haven't really tested the burner in Win2k, primarily because process management in 2k is sloppy. If I wanted to burn at the maximum speed, I'd have to close *everything* to avoid buffer underruns in Windows. In Linux, I'm simultaneously downloading from usenet, unRARing 800MB mpegs, viewing SVCDs from my CD-ROM drive and browsing the internet. It gives me a little trouble if I -- on top of all that -- run a particularly intensive parity checking program, but I think that this is on the whole better than having to avoid *breathing* lest Windows get cranky and reduce my CD-R media to useless silvery powder.
Granted, linux does have some usability drawbacks. Moving drives around is a big no-no unless you're a learned user. My system forgot where that ATAPI burner went when I recently rearranged some devices. I fixed that in under ten minutes, but it felt like the end of the world before I figured out what was going on. I can't get my Gyration gyroscopic mouse working properly in Linux (I've gotten it to the point that it takes *some* input from this device, but said input is completely incoherent and unmouselike). That one is due to the third-party effect, but it's still tremendously annoying. Games aren't as developed, of course, but that's not really a usability issue.
Damnit, doing things that I *need* to do, system-wise, is totally trivial in Linux. My Millennium II and Marvel G200 half a decade ago could zoom in with a hotkey in Windows, but I haven't been able to do that for years now since that feature is driver-dependent in Windows. Linux does this no matter what video card you have. This is set up intuitively in Mandrake's "please select the screen sizes you want available" install. These aren't things that you should have to reconfigure every time you get a new piece of hardware.
Aaargh, I'm sorry. This has turned into just a straight rant. I know that different people have different habits, but my own personal experience is that Linux is *easier* and requires *less thinking* unless you *want* to be an advanced user, in which case it seems happy to give you the power to be advanced. I can finally do things that maximize my personal productivity. In most cases, hardware and software just works, instead of just works until the blue screen appears. That may be overly mean to Microsoft, and I readily admit that several of their programs are top notch (I've loved products from them going all the way back to Decathlon, a game that Microsoft made for non-Microsoft operating systems!). Excel is great. Access seems strong. I'm told that the "Ages of..." series is phenomenal. MS-DOS Edit was a fantastic MDI editor (and boy was I disappointed when they downgraded to Notepad and Wordpad!). Media Player (well, using the Classic skin since the more recent interfaces have been very clunky) is usually fantastic. But it's still my opinion that Windows isn't inherently easier. The thing that is easier for users is the fact that nearly every company in the universe tailors their hardware and software to work best with Windows. I mean, wouldn't Dodge look unbelievably superior if 95% of body shops only did work on Dodge vehicles?
Eh, I'm done complaining. I may have made it sound like Linux is infinitely superior to Windows, but I was mostly overreacting to what I consider an equally extremist (but opposite) viewpoint. Windows 2000 is "good enough" for me. If there were no Linux, I could probably be comfortable using Windows 2000 for the rest of my life. Unlike the nightmare that was Windows 98, I can usually get Win2k to listen to me in a reasonably reliable manner. I use Linux largely because it prevents the need for me, a poor guy, to steal computer programs from P2P networks. I use free (and Free) software on Linux, and the very knowledge that people do these things to benefit others and not just to win a buck both makes me feel warm and fuzzy inside and compels me to be similarly generous with my gradually expanding coding knowledge.
BTW, there is one area where I will be stubborn: Qt beats any OS-specific class/widget programming package ever. I love, love, LOVE being able to develop and compile applications for Windows, Linux, OS/X, various unix variants and a couple PDAs using OS-native widget sets on a single codebase. So pbbbbllllt!;)
Note: Hey, neat, I just discovered that I can drag copied text to my desktop background and it'll automatically paste it into a new text file. That's rather useful.
> Linux is dead to me. Long live FreeBSD. (But seriously... I don't like desktop Linux. > It rocks for servers, though. It just doesn't float my boat... I mean... why would I use > Linux/X11/KDE|GNome as my desktop, when I can just use Windows XP?)
With Windows 2000 or XP on my 800MHz computer, I would not be able to simultaneously write a Mode-2 SVCD to CD-R while watching a Mode-2 SVCD on my DVD/CD-ROM while mass downloading from usenet while printing stuff while web browsing while unraring the next movie to burn. Heck, I can compile code while doing all that, and I'll read some usenet posts while I'm at it, and the chance of buffer underrun is still refreshingly low. On my NT5 box, I have to close *everything*, except maybe a browser window or my newgroup client, and I have to run the burner at half the speed (compared to how fast I burn it in Linux).
Considering that I'm continually burning my favourite television programs and such, I would be in pretty sad shape if I were running Microsoft more often than I currently do. As it is, I go into Windows typically only to play Civilization III.
I'm sure that many people have a lot of ease with Windows 2000/XP. YMMV, as they say. I just happen to be one of the quirky folk in those situations where [Mandrake] Linux installs happen straightforward without incident, setting up the proper drivers and including really flexible software and interfaces automatically, and Windows just gives me driver and application problems (I had to hunt forever for win32 drivers for my 5.1 sound card, my AIW RADEON 8500 is crashy if I try to close the TV app, etc..).
Windows does have two big advantages over Linux, at least in my house: Eudora, and easy compatibility with the Gyration mouse.
Like I said, YMMW. I don't complain if you made your OS decision based on your personal preferences. I only get miffy if your choices are based on ignorance ("I can't run this lennox thing because it's missing the internet explorer button"). So hah!;P
Oh, and I have no qualms with FreeBSD. My company's web server runs on it, and the daemon dude is cute. Any group whose mascot is Satan gets my vote. Plus, it's POSIX, so I can compile most of the useful apps for it. POSIX is my new best friend.:)
> I think the maximize window functionality found in all of > todays GUIs does this excellently, wouldn't you agree?
To address the original poster's requirement, you'd need an OS setting that forces all windows to open maximized. I don't mind having a windowed system, but I do happen to run almost all my windows maximized, and I have to frequently re-maximize them when I open them. That's an annoying inconsistancy, and it'd just be nice if I could check a box somewhere that made these program windows open up consistently.
> I fail to see what tabs accomplish that the Windows task bar or OS X > dock doesn't. Tabs are used for switching between windows of a specific > application, nothing more, nothing less. In fact, since tabs are an > alternative way of switching between windows, the combination of the two > (tabs and taskbar/dock) complicates matters even more.
Tabs aren't an alternative to taskbars. They're the same thing. Think about it: They're (usually) a single column or single row of generally rectangular widgets that, when clicked, replace a large widget (typically a full, resizable window) in another location. You right click for a context menu that allows you to close and sometimes resize the larger widget.
Heck, in several MDI applications the "tabs" are presented straight out as a taskbar. They look exactly the same. This is the case with Opera, Eudora and EditPad, three programs that I cannot live without. Right now, I have thirty-three web pages, eleven email windows (boxes, messages and my filter report) and twenty-seven text files open. Do you realize what a horrible joke the taskbar would be if I were using Internet Explorer, Outlook and Notepad in this manner? In any Windows before XP (and maybe ME?), I'd have a bunch of incoherent little icons and a stupid up/down widget just to the left of the taskbar. XP is a little nicer, using the KDE-like feature of grouping similar windows, but I think that interface is still a bit awkward.
The way I use it, generally, the OS taskbar contains a list of Applications, and the Application taskbar contains a list of Documents for that Application. This is the neatest, simplest way of representing and navigating through my documents that I've yet found, and it's all but eliminated the need to hover the mouse pointer over a taskbar item to figure out what it is (from the "mouse hint" popup).
I fully understand that this methodology won't work for everybody. But I do get pissed off when I don't even get the option for stuff like MDI (Multiple Document Interface - one taskbar entry for the entire application). I still use Microsoft Office 97 at work, largely because it doesn't clutter up my desktop. For Office 2000 and XP, they started to shift to an SDI (Single Document Interface -- one taskbar entry for each and every document of the application), which annoys me. Personally, I don't see why more companies (especially Microsoft, an organization with more resources than a city with uranium, oil, horses and grapes in its two square radius) don't simply allow the user to choose between the two. Opera asks you to choose between SDI and MDI when you first run it. In mozilla, you have the choice of hitting CTRL-T to open a new document in the MDI (tabbed browsing) or CTRL-N to open a new window with the SDI (windowed browsing). I can run as many single window instances of EditPad if I wanted, and each EditPad window can have as many documents as I want it to have. I don't know how flexible Eudora would be, though. I suspect that it'd somehow manage to overheat a VIA C3 if I got it to run more than two simultaneous instances (it's an incredibly awesome program, but I can surely see its memory bulk).
What was my point? Oh, yeah.... Cluttering the taskbar is bad. It's the number one no-no of GUI design, imho. Using a subtaskbar for each application is one way to get around this. Another way is to have a bi-level system taskbar: The bottom bar is like a regular application taskbar, and the top taskbar changes to a bar containing the documents of whatever application you have selected in the bottom taskbar. Task Grouping (like, as mentioned before, in KDE and WinXP) is another way (that I happen to find uncomfortable).
I agree with much of what you say about consistency, but I think that each person has a different idea on how it should be consistent. For example, I don't know why, precisely, but there's something wrong with the current "File, Edit,...." menu bar, something wrong with the basic idea of how it works (maybe I'm against turing machines or something), and I think that menus should really get some re-thinking before we really standardize on something like that.
> Yeah, and furthermore the dems want to tax the holy f*ck out of me and give > my hard earned pay to welfare slackers and then have the govt be my nanny and > protect me from myself in everything I want to do. The reps don't want to tax > me so much, but lean toward fascism/police state-ism and take away my > freedoms and let big corps run ramshod all over me and my rights too.
Hmmm. My issue is that the current administration is trying to grab the perks of both party philosophies while igoring all the repurcussions.
Imagine this, from the perspective of a businessman: Your company posted yearly revenues of about two billion dollars for 2002. But on the same note, they spent roughly 2.1 billion dollars in the same year for various business-related purposes, including several expensive instances of sending money and aid to competing companies. Your company has been operating in the red since World War II, upwards of half a century now, requiring you to borrow money every year, often by increasing degrees. Your company's debt is now $6.4 billion, more than three times your revenue for a year.
You became CEO in 2001. Even though the previous CEO did not back your promotion, he convinced the board of directors to pay you twice as much as he earned per year.
Every CEO's first year or two can be problematic, especially when the board of directors differ from you philosophically. But this year, you were able to get all your friends onto the board, so they'll unilaterally agree with whatever you decide.
So what do you do to fix your company's financial woes in 2003?
In your 2003 Annual Report in January, you declare that the company will drop prices, lowering revenue. And you announce that spending will increase substantially this year and likely sequentially over the next decade as well. As a result of that, the company will have to borrow more money this year than it ever has borrowed in any year before. You announce new policies which allow for increased surveillance of both employees and customers. You almost unilaterally decline membership in IEEE and ISO standards and any other standards that weren't specifically set in motion by your company.
But you end your Annual Report with the phrase "God Bless My Company", which means that you're obviously doing the right thing, eh?
Multiply everything by a thousand, and you have the US's current state. Yay, let's decrease taxes while increasing spending. Brilliant. Who's the Secretary of the Treasury these days? Groucho Marx?
> You can;t win for losing. I think we need a new political party, all composed of > professional wrestlers, to take over and set things straight again.
Aye... WHUT'CHA GONNA DO, BROTHER, WHEN POLITICS RUN WILD ON YOUUUUUUUUUU!!
Hmmm, on second thought, the problem is that all the popular wrestlers are owned by Vince McMahon, who's basically the GWB of pro wrestling. Now, I'd totally vote for an AJ Styles/Low-Ki presidential/veep bid, but that's not gonna happen.
> $159k over 3+ years isn't exactly a killing, > especially for something he was spending a > considerable amount of his time on (read: full > time job).
Anything over a hundred thousand dollars in three years is fucking incredible. As a netadmin (and programmer, and web master, and ISO 900x form designer, and part time journalist, and data enterer, and end user support dude, and machine setup guy) that could bring his company down with a simple push of a button (I'm in a room with all the servers, and that UPS switch looks mighty inviting at times), half that amount would be a pretty sweet raise. Heck, you can also figure in that he wouldn't have to spend four thousand dollars over those three years to pay for the gas for the commute. Heck, all that commuting has cost me a good six or seven thousand dollars on my car if you also count wear and tear (and that replacement engine) over the last three years.
Wow. This guy has a dream job. Work from home, exploit stupid people, make fifty thousand dollars a year with far fewer expenses. Shit, if only I had less in the way of ethics and morals....
> Intel's big problem is the binary compatibility they've stuck with since the > 80x86 (more or less). Binary compatibility was important because so much > programming was necessary at the assembler level that changing the chipset > was prohibitive. This has kept a bad chipset in commission long, long > after it should have died.
I think you mean "instruction set". Intel changes their chipset like they do their underwear (that is, frequently, though perhaps not as frequently as the analogy implies).
> But then, if you can successfully market clock speed as the sole > measure of performance, why bother offering something better?
Yeah, that's annoying. I always hated how the clock frequency is always called the clock "speed". I mean, it's not in physical motion. You don't call the cycling of your car engine its "piston speed", or whatever. That is perhaps a trivial sore point for me.:)
Still, the current version of the P4 is not bad at all. It is arguably an equal or superior microarchitecture to AMD's K7 family, though it's difficult to really make that determination solidly, since Intel has a six to eight month process technology advantage over AMD, and that gives them a frequency advantage somewhat independent of the base microarchitecture.
The Windows user in me is torn between getting an SMT P4 or a K8 ("Athlon 64", I think they're thinking of calling it) at the end of this year. The K8's on-die memory controller should give a boost to some of my operations, but the P4's SMT functionality would likely benefit me, as I have a tendency to run lots of apps at once (I make most power users look like AOL newbies in some respects, heh). Certain cpu intensive programs (like SmartPAR) that eat up all my time might run better on the AMD setup, while other programs (like WinRAR) will likely enjoy the benefit of the Intel box's higher raw memory bandwidth and cpu frequency. I guess that's a "wait-and-see" type of thing.
The Linux user in me is a steadfast AMD supporter. This has nothing to do with any "WinTel Conspiracy(TM)" or whatever; it simply appears to be the case that any AMD chip is substantially faster than an equivalently rated Intel chip in most Linux-based benchmarks. I am a little interested in seeing how much of a benefit the SMT gives to gcc, but it would take a lot to convince my Linux side to move over. The Athlon XP 2700+ seems faster in Linux than the 2.8GHz Pentium 4, and that's without the added benefits that the Barton brings to AMD's K7 core. Heck, that little Linux daemon (hrm, or maybe it's the FreeBSD dude) inside my head keeps telling me to drool about how much faster than Barton the K8 will be given its advantage of far lower memory latency (due to that on-die memory controller), 64-bit registers, doubled GP register space and those HyperTransport connections. I keep telling myself that only the memory latency and extra registers would make a difference, and even with that compiling probably won't be that much better per clock than with the K7, but even with minimal improvements, K8 should be faster than K7 in compiling, and since K7 is much faster than the Northwood/P4 in compiling, the K8 should be substantially faster at the task. Except for that unknown variable of SMT. I'm going to have too look and see how much it can add to the fray. In addition, it is not unlikely that the P4 will simply scale in frequency by a greater degree than AMD in the next ten months.
Oh, new data: www.uwsg.iu.edu/hypermail/linux/kernel/0212 .2/0107.html Allegedly, you get something like a 15-20% increase. Not bad.
www.uwsg.iu.edu/hypermail/linux/kernel/0212.2/01 98.html But this guy is getting some sort of substantial decrease in performance due cache problems between threads (I guess that there's more cache misses since twice as many threads needs twice as much data).
Interesting.
Damnit. Why can't companies give me this stuff for free so that I can test it all for myself? I'm a coder, and I have to know what hardware can render my code AFAP!
But it's all fun, anyway, this talking about microprocessor technology.:)
> I believe the code is up-up-down-down-left-right-left-right-B-A. Hitting > start was just to start the game. Your code would probably work because > the extra B-A-select-start wouldn't interfere with the code.
Konami put that code into a ton of games, and I vaguely recall both variations being used. Perhaps the select-start was done in the earlier games and they dropped it afterward.
> After spending 20 years as one of the lowest paid (yet consistently employeed) > network/sys admins on the planet, yes, I will get a pension, benefits, etc.
Interesting. I was under the impression that government jobs pay astronomically higher than private sector jobs in the computer field. I, for instance, make $24K a year in an occupation which involves user support, programming, computer setup/repair, website maintenance, ISO 9002 complaint form generation and (whenever there's a lull in that other stuff) heavy data entry. A friend of mine, who graduated with a similar degree at a lesser institution (as evidence, I present that after graduating he didn't understand the basic concepts of TCP/IP, C++, superscalar pipelining or modem handshaking), makes $56K a year for the FAA doing the user support and computer repair (I don't think that he does the setup).
Granted, he gets more stress from his job. The folks there make fun of him, and he has to take some sort of anti-depressant medication that "zonks" him out (his word), and he has to visit a psychiatrist once or twice a week, but on the top side he also gets to visit other states and take computer classes, all paid by the government (they even pay for his food when he does these things!).
I think that I would give up a little piece of mind for an extra two or three thousand dollars a month, free vacations to Pennsylvania and upstate New York and free learning classes. Hell, I went through a nearly complete mental breakdown last year and had to spend nearly the entire year working part time instead of full time in order to avoid going completely bonkers. So I don't think I have any real palpable benefit over this fellow with the government job.
The funny thing about your comment is that while the former WWF company (now known as WWE) has often been symbolized for chaos and violence, it is actually the most conservative and least violent company in its entire genre. I've recently gotten somewhat caught up in all this -- my friends are huge pro wrestling fans, and I've been watching it on and off since last May or so. I like a lot of the actors, and I have a great deal of respect for the performances they give, though I prefer to focus more on independent productions, since the writing quality of the mainstream product is crap, and that particular company has a lot of contempts for its fans.
Anyway, the former WWF is the least chaotic of this form of performance art (compared to, say, NWA or XPW). But the irony that I was initially trying to put forth here (I have ADD, and I sort of go off on tangents a lot) is that the *current* WWF (the World Wildlife Fund) offers a rather interestin analogous situation. Last year, they sued the similarly named wrestling company for using the same three letters in a totally unrelated industry and, in the face of common sense that would say otherwise, and in the manner in which RIAA totally and illogically decimated Napster, utterly won.
> How's that? Did someone actually find some good music to steal?
There are people who use Kazaa without ever downloading mp3 files. I don't download music, but Kazaa is useful to me, and I hope it becomes an important stepping stone towards easy distribution of free (whether beer or coffee or whatever kind of free you care about) software in the future.
My life will be so much better when I can download those operating system ISO images without having to wrestle an ftp server into the ground.
-JC
PS: I do admit that I may have helped my sister download mp3 files in the past. But her unhealthy addiction, much like someone's addiction to an illegal drug, must be weaned out of her gradually; I'm certainly not going to call the police on her, as that would obviously ruin her life without causing benefit to anyone.
> For most home (and, indeed, server) applications, I would have thought that having a > dual core processor won't make much of a difference, just as processor speed doesn't - > rather, what is important is the speed you can get data in and out of the processor.
> Overall CPU speed doesn't seem to make much of a difference when the bus speed is the same, > certainly not in the systems I've tested. However, up the CPU bus speed, and you'll find > your performance greatly improved, because you're getting data to the processor quicker.
> Some years ago, I tested this theory with a couple of old 686 chips - one 200, one 233. I > benchmarked the 200 and 233 both at 75MHz bus - virtually identical results. Then I ran them at > the same CPU speed, but 83MHz bus, and the benchmark results improved by exactly 83/75. > What does this tell you?:-)
It tells me that the benchmarks you use are not the same as the benchmarks I use. Here's my rundown:
Civilization III: The complexity of this algorithm increases exponentially based on number of cities. I'm running a 16 civ game at the moment, and the game literally takes more than twenty minutes to cycle a turn, and it's only the nineteenth century! Granted, my 800MHz Duron isn't state of the art, but it's not state of the fart, either, and even the top of the line processors would buckle under this stress.
PAR: I download very large binaries off usenet that are separated into multiple files. PAR is a tightly coded system that makes extra parity files that you can use to build missing files of a download set. It takes a *long* time to verify the parity on 800MB files.
RAR: This probably would be helped somewhat by faster memory access, but I suspect that extracting an 800MB multipart RAR set would also be strongly enhanced by a faster processor. I mean, it'd be nice to not have to wait five to ten minutes to extract this stuff before viewing it.:)
Qt: Compiling scales almost directly with processor speed, at least on some types of code. And I do a lot of compiling on Linux. And I do a fair amount of compiling on win32, as well. Compile of a large program can take many hours. When I type './configure && make' or 'qmake -project && qmake && make', I want to get up, prepare some coffee or munchies, come back with the yummes, and immediately test the newly created binary.
There are other programs that I use that depend on the processor, but those are some of the biggies. There are also programs that would benefit from faster data access, of course. But memory isn't really getting faster. It's just getting higher bandwidth. When you request data from PC2100 or PC800 memory, that data doesn't start coming back any sooner than it did with PC133 memory. It's just really expensive to increase the frequency of the entire northbridge and corresponding devices. That's why the microprocessor has been doing the majority of the speed boosts. It offers the most increase for the least expenditure.
> Wow - way to make up stuff. > The Itanium II is certainly not a dud
Agreed. From an engineering standpoint, it's quite a nice chip. I don't agree with some philosophical stuff in the ISA (I'm not that much of a VLIW-for-general-purpose fan, but hey), but the microarchitecture and implementation seems very nice. I do wish that it was easier to implement OOE on IPF, though.:(
> x86 compatibility is worthless ina high end 64-bit machine, somethin AMD doesn't seem to grasp. > They're marketing a high end technology (consumers and normal business users don't need > 64-bit technology and won't for a while) to the mainstream market. Morons.
Feh. A big "screw you" on that. AMD isn't catering to the high end server group. They obviously can't just teleport into that market. Their catering to the smaller business that uses Xeon servers. Backwards compatibility with x86 is of the utmost importance in this market. Basically, they're marketing x86 workstations and x86 servers that happen to allow you to enhance performance of some types of programs with simple recompilations. There is a good chance that I might get the lower end version of this product when it comes out, as I use Linux, which may strongly benefit from those extra registers in x86-64, on my home machine. We'll have to see, of course, before I pull out the green.
> And you seem to be ignoring the numbers (remember that 'reality' the rest of us > live in matters to us, if not to you). AMD is going broke. Intel isn't.
That's a bad measure to use. You don't have any controls in this analysis. There are a lot of reasons why AMD is losing money (poor management a la Hector Ruiz, inability for a relatively small company to handle a very harsh recession, etc..), and there are a lot of reasons why Intel is still doing phenomenally (people buy Intel no matter what, currently excellent execution, they can afford to strongly diversify). Many of these reasons have nothing to do with the technical/engineering side of the equation. IMHO, both AMD and Intel have incredible engineers, and frankly AMD especially warrants respect for being able to ramp technology at *approximately* the same rate as Intel despite having a very, very miniscule fraction of their resources. That is why I was a big AMD fan a couple years ago, at around the time when the company was dominated by the excellent triumverate of Sanders, Raza, Meyer as well as a couple critical folks like Norbert Juffa and Paul Hsieh. At this point in time, AMD was a quantum of a company that somehow managed to produce a piece of engineering that allowed them to, for a brief time, outdo the capabilities of a company fifty times their size. I am somewhat dismayed that AMD turned into a more traditional company over the last two years or so.
> The Itanium is unusable. Have you ever seen > one production system with it? I have not.
I know somebody whose workplace (a research place of some sort) got a cluster of five hundred of them.
> Itanium II is now out and is said to be OK. For the price of an Itanium II > system you could buy a car/house/small country.
Um. High end servers are supposed to be that expensive. Ever try shopping for a high end UltraSPARC or Power4 machine? Didn't think so.
IPF isn't supposed to be a replacement for x86 (well, it was originally eventually supposed to, but that's when various Intel execs were drunk on monopolistic stupidity). I do not agree with the means by which Intel has penetrated the market (eg, they coaxed several prominent chipmakers, such as HP, DEC, SGI, and so on, to dump or devalue their existing lines and support the Merced far before it reached A0 stage), but on the engineering side, the McKinley (or "Merced II", if you like fugly names) seems an excellent implementation of what was perhaps a not too well thought out ISA. Just my opinion, of course, and I'm merely an armchair designer (no, no, I don't design armchairs... you know what I mean!).
> and the relative lack of success of AMD (I use an XP1800+ and think its great, > the company just doesn't seem to do too well.)
Well, part of that is crappy management, but a large portion of their troubles are simple due to the fact that Intel is given the benefit of the doubt by the OEMs and the consumers. Even during the year or two when AMD consistently had faster chips with fewer bugs than Intel had, Intel made tons of money and AMD merely made enough to recoup past debts. People buy Intel because they're Intel. This will happen whether Intel is doing a good job or a bad job. Thankfully, they're doing a good, honest job and earning those buyers now, but from 1998 to 2001, they were not doing their customers honour.
> Intel doesn't need to keep spending money researching new chips > if it's current generation are so far ahead of its competitors.
They aren't. Intel's Pentium 4 is pretty much on par with AMD's Athlon. But Intel has five or so x86 plants that they can leverage to test different ways to most optimally ramp their chip frequencies. You don't just throw a design and a fab process into a bucket, shake it, and come up with the resultant chip speed. You have to devote a substantial part of your manufacturing resources to the research needed to optimally match your current chip design to your current manufacturing technology.
In addition to this, Intel happens to be something like a year ahead in base process technology. They moved to 130nm six months before AMD did their equivalent move. This means they're very much ahead in that respect. So if their chips were a generation *behind*, Intel would be competitive in chip performance (this is was almost happened with the Pentium III and the early implementation of the Pentium 4). As it is, the current P4 is a competitive design coupled with a slightly more advanced manufacturing process, so Intel is a couple speed grades ahead.
Intel has to keep researching constantly. AMD does a surprisingly good job at ramping technology at approximately the same rate as Intel, despite having about a twentieth of their capital resources. If Intel stopped researching for just a few weeks, they'd lose the leverage they have to stay superior in the current climate. And that's not counting on the outside possibility that K8/Hammer might exceed performance expectations and outperform the top Pentium 4 upon release.
-JC
Regarding K8/Hammer performance features
on
AMD's Fab 30 Revealed
·
· Score: 5, Interesting
> the Hammer/x86-64 chips have ondie memory controllers AND more registers than i386++ type > chips, combined they'll give a speed increase of not inconsiderable proportions.
It is notable to, er, note that the former advantage helps (possibly considerably) towards both recoded AND legacy (eg, normal) programs, whereas the additional physical registers would require recompilation in order to show a benefit, which means that everybody but Windows users will get an immediate use outta that.
Other advantages of the Hammer? Well, not counting the 64-bit yunk (that WILL provide benefits, but I want to cover benefits that will help legacy programs, like Civilization III, The Sims, Unreal II and other antiproductivity applications):
Hyper Transport. That's not much on its own, but it essentially equates to a reduction in loss of bandwidth to the chipset and between processors when you add an additional processor. On the Intel setups, the processors share a set amount of bandwidth to the chipset, so putting eight chips on a 2.4GB/s bus means that you have each chip getting 0.3GB/s. The AMD setup theoretically lets each processor get that 2.4GB/s. Of course, that's in a perfect world, chip-level, but it probably amounts to some benefit. AMD's K7 family has similar advantages, which probably assists in explaining why they get higher performance at each given clock in mainstream applications (which at least somewhat depend on the memory subsystem) even though the AMD cpu to memory bandwidth was usually 2.13GB/s (now it's 2.67GB/s, unless you count stuff like the nForce, which has some extra memory bandwidth, but the extra benefit there is eaten up by the onboard video), whereas the Intel cpu to memory bandwidth was usually 3.2GB/s. Anyway, the idea is that HyperTransport will (on a hypothetical level) make it much easier to make n-way systems without either a tremendous performance impact or an expensive crossbar workaround setup thingy.
SoI. Silicon-on-Insulator. This is one of those things that'll help with the process technology. In the end, it'll probably offer a little bit of a frequency boost by making the chip a little cooler or something like that. I forgot precisely what SoI's primary benefits were. It's been months since I've even thought about it.:)
Stages: As detailed here, the K8 adds two stages to the decoding part of the instruction pipe. The decoding part of the pipe is probably rather complex, so you may see a pretty neat frequency boost over the K7 family without the problem of a huge branch mispredict penalty. The number of cycles that a cpu wastes when it makes an incorrect guess on a low level "if/then" statement is somewhat proportionate to the number of pipeline stages. The AMD K6 and (iirc) Cyrix 6x86 were the mack daddies of branch prediction, since their pipes had only five stages or so, so they only had to wait a few cycles when they zigged instead of zagged. The PIII and K7 had over ten stages, so they had to wait a lot longer, but other advantages (such as the larger and sometimes faster caches and more accurate predictors) in those processors over their predecessors did their best to overcome this disadvantage. The Pentium 4 has a crippling 20 to 28 (depending on the situation, and depending how the trace cache handles the situation, and whether or not you want to count it) stages. This means that it can hit amazing clock frequencies, but it'll get cranky and drowsy for twice as long when it makes a predictive mistake. How does it get away with this? Well, the trace cache does its best to assist, but it didn't really help as much as I think the designers were hoping. But for multithreaded programs and OSes, the SMT implementation on the more recent members of the P4 family, an implementation known as "Hyper-Threading", probably pretty neatly alleviates much of this problem by putting operations from other threads into the cpu whenever the currently running thread stalls on a branch mispredict. The K8/Hammer approach is just to add stages where they hopefully will have the most balanced, beneficial effect to frequency boosting while only minimally increasing the branch penalty. SMT would be nice, but it isn't nearly as critical a need as it is on the P4.
Wider memory access. On the Sledge Hammer, if AMD's plans are still the same as when I wrote this, the memory controller (which is embedded onto the cpu) will access PC2700 memory in a 128-bit configuration (ignore the "126-bit" typo on the linked page -- I can't believe I didn't notice that when I typed it nearly a year ago!), which leads to a 5.3GB/s path to memory. That's damned good, though I really think AMD should have focused on 366/183MHz (equiv to "PC2933") or 400/200MHz (equiv to "PC3200") memory instead of the 333/166MHz PC2700 that came out over a year ago. Still, servers often use memory that's lower than bleeding edge clock in order to maintain reliability, so bleh. Still, 5.3GB/s isn't bad for a setup that isn't based on a shared bus.
Enhanced branch predictor. Well, that's if my notes from a year ago are accurate. If true, this'll probably overcome any mispredict penaly performance disadvantage from those abovementioned added stages.
Larger TLBs, TLB flush filter, etc.. This stuff will have itty bitty advantages on a per-clock performance basis, but every little bit helps.
Larger caches. Hey, I should look this up to see what they're planning on. Is it just 512KB on-die L2, or is AMD planning on bringing it up to 1MB L2? The interesting thing about AMD's designs is that the die is really small on each processor. Remember how AMD has gotten occasional fire for processors overheating? Well, aside from a stupid lack of shutdown diodes in the past, the real cause wasn't that the AMD processors used more heat than the Intel processors. They usually generated about the same amount of heat, often less, but their processor surface area was substantially smaller, which made the chips less expensive to produce and less likely to have defects. But when you try to push an equal amount of heat through half the surface area, you end up with a higher amount of heat per area, which equates to a higher running temperature. The funky thing about this is that you could just added a whopping huge amount of on-die cache. That'd increase performance while also increasing the surface area. But the heat production would not be substantially affected. So you'd end up with a lower temperature processor. So the Hammer will have a higher ratio of cache to processor units in the cpu, so it won't be as much of a fire hazard. Frankly, they should have put 1MB on-die L2 onto the Thoroughbred/AthlonXP.;)
Crap. I need to research more on the K8. It probably changed a lot since I went into hibernation. The interesting thing is that in the last half year, I've largely moved from being a Windows 2000 power user to a Linux coder (I still use both operating systems for several different purposes, but I'm talking primary usage). I stand to be in the group that benefits most from the Hammer when it comes out, since I'll be able to './configure && make' or 'qmake -project && qmake && make' most of the programs I use and/or develop. So I'll instantly see the benefit of those extra registers. ^_^
Slashdot Mirror
> Just enough to beat Gf Fx 5800
Nonsense. My old GF2MX performs better, just by way of it existing and the 5800 not.
> Outlook isn't an e-mail server, its a client. Get a clue.
The original poster was rather obviously going for a +5, Funny.
> Your post makes an excellent argument for the
... automated mining probes ... Ceres ... microprocessors ... faster ... more reliable rate", and they might be able to make out the general gist of what I was talking about. I probably should have added a boldfaced term like "high initial investment" to balance out what I was saying there, but I was writing a lot of stuff on the fly. This isn't a term paper or a news article, you know. :p
:)
> banning of font weight variations in body text.
> Of course, no one expects good typography on
> Slashdot, home of the all-italics front page.
Heh. I do that for a reason. I've found that I tend to drone on when I write stuff, so many readers get bored before they can get to the meat of what I'm saying. So I try to highlight (well, make bold) words that represent the important points of the paragraphs. Some readers (granted, not all) find it easy to quickly scan a long post if they can pick out important key words.
For instance, when I said "So you start up a pipeline of automated mining probes to Ceres. Much like on modern microprocessors, the initial hardware cost would be greater, but you'd be able to transport mining materials at a much faster and more reliable rate than you otherwise would", a speed reader would see the boldfaced words " pipeline
Granted, I stole this technique from Mad and Cracked magazines, so it might not be the best approach.
-JC
> But what kind of server processor is the Itanium? Given the
;) ). The PowerPC looks just about as attractive as
;).
> unimpressive specint scores, it doesn't look like a processor
> for those mass volume servers (web, mail, file and print etc).
> Does it even look like a cpu for larger RDBMS servers?
Well, specint isn't necessarily representative of all those types of servers. A lot of these servers are data transfer intensive and essentially best on cpus with insane amounts of cache and a really high bandwidth cpu to chipset data path. Itanium II seems to have this.
Of course, that doesn't mean that there aren't better options. I could probably put together a multiple machine array of cheapish x86 Linux or BSD servers (that is, if I had a little more experience with clustering and similar methodologies) for the same price, and I'd be ending up with a more reliable setup (more machines mean better redundancy, and overall performance probably would be faster due to greater parallelism in every component, not just the cpu -- for example, the faster Itanium chip might be able to write to two memory modules at once, but it probably can't write to and read from as many hard drives as a multiple setup of slower processors could).
> To me it doesn't look like a processor for corporate servers,
> maybe scientific clusters?
That is a toughie. Itanium in scientific applications goes in three categories:
A. Really shitty (if it's an x86 app)
B. Kinda slowish to kinda fastish (if you just compile it straight)
C. Faster than a drug dealer who accidentally walked into a DEA New Year's Bash carrying a wad of dimebags (if you very very carefully rewrite the code and apply smart compiler options over several test builds).
Obviously, the software-optimization requirements of the Itanium family (remember, the nature of the chip's architecture means that it's an in-order processor; out of order processors, which started appearing in x86 with the Pentium Pro and earlier in other architectures, optimize and rearrange code on the fly, so you can get reasonably optimal performance even with only basically competent code) will turn off many people. But those who are willing to do the work will see the benefit they're looking for.
> If I currently have x86 servers, and I'm looking at the next step,
> the Opteron would be the most attractive option (IF AMD doesn't
> screw up
> the Itanium - for practical purposes they're both incompatible
> with x86.
Yeah, x86-64 is really neat for the more mainstream server market. You get excellent performance in already existing apps, and recompilation for higher performance is trivial (well, comparatively).
> Would be interesting to know what's happening in Dell internally.
> The Opteron really looks like a better match for them than the
> Itanium
Don't delude yourself. Dell uses AMD as a bargaining chip against Intel. If Mike Dell every says something like "We are evaluating AMD products", he means "Hey, Intel, I need you to give me a better discount on your new chips". Dell/AMD rumours tend to pop up lik clockwork every three months. But I guess you never know....
> While Microsoft could be kingmaker/decider for the Athlon64,
> Linux will be enough for Opteron.
Well, Athlon 64 should be okay either way. I mean, the 64-bit mode isn't even the best feature, as I've noted before. It's the low latency, on-die memory controller and possibly the HyperTransport connections that interest me, and you'd get benefits from those even in conventional 32-bit Windows systems.
-JC
PS: I apologize if my comments about Itanium II are exaggerated. I haven't seen recent scores outside of stuff like spec, and I was disappointed particularly with Merced/Itanium, so the situation may be not as intense nowadays.
> I found it interesting that commercial mining of asteroids was mentioned
> in the third article. Sure, raw materials are plentiful in asteroids,
> but wouldn't the cost of getting there far outweigh the benefits
> of the plentiful resources? I guess this would be practical if/when
> we run out of certain ores, or as an "While we're here, we might
> as well" measure, but I can't see it going anywhere otherwise
> until the price of space travel drops dramatically.
Yeah, it's an "economy of scale" thing. Once we have regular (and by regular I mean many companies each making daily launches) access to space, the incremental cost of space travel will be far, far below what we pay now in terms of cost per unit mass.
And, thing is, it doesn't cost much in terms of energy to get to any given asteroid once you're outside the gravity well of Earth. The only cost would really be time, since momentum keeps you going until you want to brake. So you start up a pipeline of automated mining probes to Ceres. Much like on modern microprocessors, the initial hardware cost would be greater, but you'd be able to transport mining materials at a much faster and more reliable rate than you otherwise would.
And mining in a low gravity environment (Ceres, the largest asteroid, has less than a thirtieth of Earth's surface gravity) should be far less energy intensive than mining on Earth, where you have to expend unbelievable amounts of energy to merely move mined elements from the mines to the transports, and where you're severely limited in machine travel range due to friction related to that nasty 9.8m/s^2 that we have to put up with all the time here.
And some asteroids will probably have minable minerals in abundance, and remote spectral analysis will be able to allow us to identify mineral compositions from millions of miles away. That does away with a huge portion of the work involved in mining on Earth, where you have to often indirectly figure out what minerals are hidden beneath the surface, wasting time by drilling all over the place. That costs money continuously, and you could save on that with asteroid mining.
There are probably some other things in asteroid mining that I'm not thinking of. I'm not miner, and my "engineering" knowledge is rather amateurish (I'm more of a theoretical guy, being a programmer and all). Maybe wear and tear of machines would be less (due to both that lesser gravity we discussed above and the total lack of atmosphere or microorganisms to break down the machine parts). Maybe there are social benefits, like higher morale for the folks who like to work in larger and more open environments (this century, maybe "getting away from it all" will mean accepting a job that lets you work fifty million miles away for a year or two).
Economies of scale, my man. We have to force ourselves to pay the high costs to continue the proliferation of new technology. I we don't do this, we go stagnant. Space launches these days are often more expensive than they were in the old days, even when you take inflation into account. It's because we stopped doing them so often. It saves a few billion dollars a year in the short term. But we might have made back those billions -- and far more -- by now from the direct and indirect benefits of building up an economy-of-scale framework of space travel. I mean, heck, we landed on Luna six years before I was born. We have space probes ten light hours away. We also have advanced manufacturing facilities on Earth now. Had we continued our push into space instead of borking it all in the seventies through the nineties, I imagine that we would have simple factories producing complex objects on the cheap (at least with products that require tons of work on a small amount of raw material -- no friction and no gravity equals huge cost reductions!) now. Chip fabbing companies would be looking at near to medium term options for building astro-fabs. Do you realize how much more precision in the sub-nanometer range you'd get in building chips when you don't have to worry about gravity and vibrations from the local landmass and atmospheric variables? I mean, crap, we probably would be able to further accelerate process shrinks. 3.0GHz on a 130nm process? What a backwards technology for a space-enabled 2003!
Economies of scale, my man. Startup costs are [almost] always prohibitive. That's how science and technology works. Live with it. But don't work to inhibit it.
-JC
> Once you've seen one comet, you've seen em all.
I dunno about that. On Star Trek last year, they had a comet with earthlike gravity. Now that's damned impressive, and it must be true since its on TV.
Maybe there are other unique comets out there, like ones with bizarre technobabble-inhibiting EM fields and occasional spaceberry orchards. ^_^
-JC
> > the BSA ... logged into the ... FTP ... with the address "guest@nowhere.com"
> Ok, I agree, but let's think here. I generally use the
> address "-a@" when logging into anonymous FTP. This is
> because (shock) I believe that anonymous FTP should be,
> well, anonymous. If someone really wants to track me
> down, he can get my IP address easily.
> Is it "OK" for normal folks to do this, yet "not OK" for
> the BSA? Or is it "not OK" for anyone?
A few things: First, your address ("-a@") would just give an error in any normal email client. The email address that the BSA provided was owned by someone else.
Secondly, the BSA was using some sort of automated crawler, so they could have possibly posted this email address to hundreds or thousands of places, which means that the person who owns nowhere.com will be getting a bunch of erroneous hate mail and a huge load of extra spam in the future. This could affect this person from both a financial (bandwidth abuse, loss of productive time, etc.) and an emotional (I don't know about you, but I get absolutely thrown into a breakdown when I get large amounts of flames and spam) standpoint.
And then there's the fact that the BSA is using this person's email address for commercial gain. They stand to make money indirectly for doing this service from their client organizations. There's a lot of collateral damage involved, but the end result will likely be a temporary decrease in intellectual property infringement of software products (it's kind of like if we napalmed all of the Middle East: yeah, other countries would get messed up, but the primary target of removing Saddam Hussein from power would be achieved). So it's an ethics thing. Are you okay with making millions of dollars by knowingly hurting millions of innocents? If you are, then it's all good.
Incidentally, it never occurred to me that these domains are owned by somebody, and I now regret using them in my anonymized passwords. I think that I will use a method like the one you use when I want to be anonymous.
-JC
> You could put a P266 in your socket
> 5 mb when you upgraded from a P100?
You might not have been able to do that, but the Centaur (the cpu design team that nowadays makes VIA's C3 chips) 240MHz WinChip could make that jump, if I recall correctly.
I am not certain if the WinChip 2 or 2A could work on Socket 5. They were basically just WinChips with 3DNow! tacked on, but they might have been out of spec for Socket 5, ever so slightly.
Oh, and there were probably "upgrade" chips, which were newer processors with adaptor thingies between the cpu and the socket.
-JC
> The following is a simplistic view of things, but we are talking
> about a 64bit processor. Remember the Itaniums Intel is selling
> are running at around 1GHz - 1.5GHz I believe and they run
> circles around the 3Ghz P4.
That's overgeneralized. The 3GHz P4 is very much faster at most tasks than the 1.0GHz Itanium II, which is the fastest instance of the chip that has been given entries at spec.org. The reason why the Itanium II appears much faster is that you only see benchmarks that relate to its very narrow field of marketing. It's a server processor. You won't see it tested in areas more suited to general purpose computing (games, office suites, etcetera). And, hell, the Itanium sucks in specint, one half of the single processor version of the most prolific server benchmark suite in the world. The 900MHz (fastest speed submitted -- for some reason, they only gave specfp scores for 1000MHz, unless I missed an entry or few) Itanium II gets 674, compared to scores above 1100 for the 3.06GHz P4. That's a whopping 63% difference! The fastest Itanium II is almost 40% slower than the top of the line non-Xeon Pentium 4!
The Itanium II does fantastically in specfp -- bested, I believe, only by the DEC Alpha, which is sadly being pushed under the carpet for reasons more political than I'd like (Alpha IP is owned by HP and Intel, the companies that created the Itanium's core architecture) -- and many other benchmarks. But you can't simply ascribe a single, simple feature to the performance advantages of the processor. Yeah, the processor can address a 64-bit memory space and, yeah, the processor has 64-bit GP registers. But you're ignoring many other features piled on top. Itanium II is a server processor, so it can afford to have some extra doodads added to it, doodads that would be considered financially unfeasible on mainstream processors.
Hmmm, I did a quick google search, so I apologize if I pulled incorrect info on the following:
The Itanium II has a more than a megabyte and a half of cache memory on the die of the processor. It seems to optionally go up to 3MB on-die L3 cache. In comparison, the Pentium 4 has 512KB cache (there's some more cache, the L1, but that's inclusive), and the Athlon XP has either 384KB or 640KB cache (depending on whether you're counting the older Thoroughbred or the newer Barton). So the Itanium gets about three times as much cache memory on the processor die!
Itanium II has a 400MHz, 128-bit data path to the chipset. Pentium 4 is 533MHz, 64-bit. So the P4 gets a chipset that can send it 4.27 GB per second while the Itanium II gets a chipset sending it data at 6.4GB/s.
The Itanium II has more functional/execution units. The Itanium II gets predication, which is a very expensive (in terms of how much bulk it adds to the die) feature that effectively gets rid of a lot of the penalty associated with branch misprediction (a problem which is rather huge with the trillion-stage netburst microarchitecture of the Pentium 4, though I'm told that the multithreading implementation of the P4 can help alleviate some of that).
The number of bits in the processor don't matter *that* much, not after the 32-bit level. Yeah, it helps, but you have to take the whole package into account. A 64-bit scalar one-stage processor with a ten-byte, off-die cache would get its ass kicked mercilessly by an 80486DX-50.
To take another tack: I'm somewhat interested in possibly purchasing an Athlon 64 late this year or early next year. But if the Athlon 64 was just an Athlon with 64-bit extensions, I wouldn't give it the time of day. I'm interested because the Athlon 64 will have an on-die memory controller. I'm interested because the Athlon 64 will support twice as many registers as a typical x86 chip (which may decrease the need for cache accesses, which could increase performance on my recompiled linux apps). Either of these two advantages promise a far greater advantage for me than the simple increase in register size and memory addressability.
-JC
> For most people, security is not important. Top performance is not
... well, I dunno, maybe "Office->Time Management" or "Applications->Sciences->Health" (probably the latter). In Windows, it's a crapshoot. Yeah, real user-friendly.
;)
> important. Optimum configuration is not important. Control is not
> important. Not having to power toggle is not important.
> Being able to put the CD in the CD drive, press a button a couple
> of times, reboot,
> and get what you want is VERY IMPORTANT. NOT THINKING is
> VERY IMPORTANT.
> Users want things that work like coffee machines. You plug it in
> and it works. If you want a different coffee machine, you get
> a different coffee machine and plug it in and it works. Windows
> makes computers a lot more like coffee machines than Linux does.
> Having to turn your computer on an off to get a new feature is much
> less of a problem than having to know what to type to get a new
> feature. Linux wants you to figure stuff out. Microsoft wants
> your money.
I'm not sure that I can entirely agree with you. I don't think that Windows is inherently easier to deal with. I think that the fact that Windows is by far the dominant operating system means that all hardware and nearly all software developers pay specific attention to how their products work with Windows. This isn't due to any particular feature of the operating system. It's simply because they have to make it easy on the popular platform.
For most applications, installations go like this:
* In Windows 2000, I'd open up my web browser of choice. Then I'd go to one of the download sites or perhaps to google.com, and I'd search for an application online. I'd download the application (usually going through a few screens of ad-laden BS and choosing various mirror that are closest to me. I'd go to the location of the file (either through Start->Run or using a file manager like PowerDesk or simply by using Opera's excellent download manager). Then I'd double-click on the file. A "wizard" application opens and asks me a series of successive questions about where it wants me to install the program, whether I agree to a fifty-page application-specific legal document, where in the Start Menu I want this thing to go, and whether I want shortcuts placed in various locations. Then the program (sometimes) tells me that it needs to reboot, and I hit "OK". It reboots, occasionally does some further installation, and then I'm set. I would do all this every time for each program.
* In Mandrake Linux 9.0, I go to the package managing program (By clicking on "K->Configuration->Packaging->Install Software", or by hitting ALT-F2, typing "rpmdrake" and hitting ENTER). I change the little radio button group from "Mandrake choices" to "All Packages". I type a program name into the Search bar and hit the Search button (or I'd just look through the efficiently categorized list of programs). I check the checkboxes of any and all programs that I want to install, and I hit the Install button. Then I sit back as the installer automatically downloads, installs and configures all the applications I selected, grabbing any prerequisite programs from the servers automatically. In the time it took to search for, download and begin the installation of a program in Windows, I've finished installing the Linux app. Before I've finished mucking with the Next->Next->Next->Finish screen of that installation, another Linux app has finished installing (without me needing to click on anything more). By the time my computer has rebooted into Windows 2000 from that one install, the Mandrake Linux package manager has installed six or seven different apps (and I only had to click the "Install" button once). And you know what? Everything is installed into logical, well thought-out places. Instead of going into "Programs" and having to scroll down a clumsy list of company names to find the app you've installed (difficult especially if you've forgotten the company name!) like I see on Windows 2000, the Mandrake installer puts everything into intuitive, user-friendly subcategories. Stuff that uses the network is in "Networking". All my email programs are in "Networking->Mail". My news (usenet) readers (Pan for binary downloading, Mozilla Messenger for general reading) are in "Networking->News". Card games are specifically in "Amusement->Cards". Know what I have to do to find all my card games in Windows? I have to look in "Programs->Accessories->Games" and figure out which ones are card games. Then I have to look in every subgroup in Programs (the aforementioned company names) to check and see which ones have card games. I have to *memorize* this stuff in Windows. In Linux, I just go to "Amusement->Cards". Holy crap, you can't get any more obvious than that! Oh, I need to watch TV? "Multimedia->Video". I have to put this 800MB SVCD onto my 700MB CD without data loss? "Applications->Archiving->Cd Burning". I want my kid to learn stuff? "Applications->Edutainment". I'll never accidentally click on the "Hot Boobies" interactive porno game when I intended to show my female colleague my PG-rated "Hard Bodies" fitness management program. It just won't happen, because one of them would be in "Amusement->Sex" while the other would be found in
It's better than that. If I want to be lazy in windows, I can set up links on the taskbar, Office Bar (if I spend the untold hundreds for their Office product) or desktop to the programs. I could also (with a non-native third-party extension program) map programs to a Win-key combination. Currently, in Windows, I use Win-O to open Opera, Win-M to open my Mail program (Eudora), Win-X to open eXcel, Win-W to open Word, and so forth. Mandrake natively supports key combinations to open programs, and I believe you can differentiate between the two Win keys if you had the desire (LeftWin+W goes to OpenOffice Writer, RightWin+W goes to KWord, for example). I don't use it much, for the following reasons: Mandrake (and, by the way, I'm using KDE to manage my gui, so ymmv if you use other programs) allows me to put links to programs on the desktop. It also allows me to put links to programs on my taskbar. But it lets me configure these links in interesting ways (and we're not talking about difficult configuration; we're talking about Right-click-on-panel->Size->Large and similarly easy means). I can have (and I do) two levels of bars with these links. I have a big taskbar with my extra-lazy application links. These are full-sized icons, so they're easy to click on when I'm too slothful to competently use the mouse. On the bar right above it, I have (among other things) medium-sized icons for a whole bunch of programs that I tend to frequently use, like my text editor and my web browser. Incidentally, that bar also has a dictionary bar, an ascii character picker (I could paste odd characters into any program instead of having to rely on some arcane, application specific "Insert->Character" features that don't work universally), a web news scroller, an advanced clipboard manager (you know how more recent versions of Microsoft Office allow for multiple clipboard levels? Well, KDE's Klipper application does this for *every app*) and quick shortcuts to lock the computer or to logout. But I don't every really use those icons very often. Why not? Well, I have session management turned on. Whenever I turn on my computer, the system reloads active programs so that I can continue from where I left off. And most of my programs (Opera, Konsole, Konqueror, Kate, Pan) have their own internal session management, so I don't have to click on bookmarks or whatever to get to where I was before. The other thing that makes it easy to not have to move my mouse to hit those "shortcut" icons is the nature of linux pathing. Remember when I installed those programs above? Well, the executables are automatically put into a place that's in the system path. Most of the programs have pretty short filenames for the binaries. Most of the time, if I want to run the program and happen to remember the program's executable name, I hit F2, type in the program name and hit ENTER. F2,pan,ENTER. F2,mozilla,ENTER. F2,kate,ENTER. Heck, even those programs that I installed through other means than the Mandrake Package Manager (sometimes, you can install the very latest versions of programs before they get packaged) will work with this. F2,gmplayer,ENTER runs the GUI version of MPlayer, the only multimedia program that can play just about every format out there, from mpeg to avi to asf to quicktime to rm to ogg to DVDs and Mode 2 SVCDs (which I *almost* have working in windows, with some occasional bizarre inconsistencies). I have to have three or four different players installed on Windows 2000 to get that sort of compatibility, and that's ignoring the easier interface and hotkeys in MPlayer.
The hardware side is sometimes easier in Windows, though my experience doesn't exactly completely agree with that. I have a somewhat generic 5.1 sound card with no discernable markings on it. It took me *forever* to find the drivers for Windows 2000, and it was actually Linux (and its "harddrake" hardware manager) that gave me enough clues about the main chips on this soundcard to find out that it was from some C-Media company or something like that. Some time after, I found the Windows drivers and everything went swimmingly. Know how much I had to look for the Linux drivers? They were already there. They. Were. Already. There. When I installed Mandrake 9.0, the sound card was autodetected and autoconfigured. I'll give you that an earlier version of Mandrake (8.1 or 8.2) didn't properly detect the card when I first installed it, but the drivers for it were in there and it was comparatively trivial to tell the computer this (I put the name of the sound card driver module, something like cmpci.so, into some configuration startup text file) compared to the herculean effort to get it running in Win2k.
My TV card used to be an outdated Hauppauge that didn't support scaling past 640x480. I had to guess which drivers it used from Hauppauge.com, and I was eventually successful in Windows 2000. The scaling thing was annoying, but it worked, except that the video capture seemed problematic. A few months later, Win2k went crappy on me, and I had to reinstall it. For the life of me, I could not remember which drivers and in which order I needed to install, and I couldn't get the TV program to work, no matter how hard I tried. So I did it in Linux. The Mandrake 8.x install autodetected, autoinstalled, autoconfigured. And it installed a whole wad of different TV programs that could use the TV card. One of them (xawtv) could inexplicably scale the TV screen to whatever dimension I wanted. I still use that program.
Heck, this past Christmas, my parents bought me an All-in-Wonder RADEON 8500. The installs worked fine on both systems. Unfortunately, I only have a choice of one TV program on Windows, and that program makes the system crash after I try to shut it down. I still have the exact same vareity of TV programs on Linux, and if I wanted to use my brain, I could probably pretty easily figure out how to broadcast the TV image onto my local network.
The USB CD burner that I recently gave up was fun. When I installed it in Windows, the Windows Media Player tried to autoinstall a "CD Burning Plugin" which caused all my CD drives (even the CD-ROM) to disappear (until I got all technical and figured out how to remove the stupid plugin). Mandrake 9 (and 8.2, I think) just installed it. No fuss. It worked on installation.
I think that my newly installed ATAPI burner is easier to install in Windows, but that's because (as I mentioned above) Windows gets the third-party support. I did have to change two or three text files (though I didn't need any installation program) to get this burner working in Linux. I haven't really tested the burner in Win2k, primarily because process management in 2k is sloppy. If I wanted to burn at the maximum speed, I'd have to close *everything* to avoid buffer underruns in Windows. In Linux, I'm simultaneously downloading from usenet, unRARing 800MB mpegs, viewing SVCDs from my CD-ROM drive and browsing the internet. It gives me a little trouble if I -- on top of all that -- run a particularly intensive parity checking program, but I think that this is on the whole better than having to avoid *breathing* lest Windows get cranky and reduce my CD-R media to useless silvery powder.
Granted, linux does have some usability drawbacks. Moving drives around is a big no-no unless you're a learned user. My system forgot where that ATAPI burner went when I recently rearranged some devices. I fixed that in under ten minutes, but it felt like the end of the world before I figured out what was going on. I can't get my Gyration gyroscopic mouse working properly in Linux (I've gotten it to the point that it takes *some* input from this device, but said input is completely incoherent and unmouselike). That one is due to the third-party effect, but it's still tremendously annoying. Games aren't as developed, of course, but that's not really a usability issue.
Damnit, doing things that I *need* to do, system-wise, is totally trivial in Linux. My Millennium II and Marvel G200 half a decade ago could zoom in with a hotkey in Windows, but I haven't been able to do that for years now since that feature is driver-dependent in Windows. Linux does this no matter what video card you have. This is set up intuitively in Mandrake's "please select the screen sizes you want available" install. These aren't things that you should have to reconfigure every time you get a new piece of hardware.
Aaargh, I'm sorry. This has turned into just a straight rant. I know that different people have different habits, but my own personal experience is that Linux is *easier* and requires *less thinking* unless you *want* to be an advanced user, in which case it seems happy to give you the power to be advanced. I can finally do things that maximize my personal productivity. In most cases, hardware and software just works, instead of just works until the blue screen appears. That may be overly mean to Microsoft, and I readily admit that several of their programs are top notch (I've loved products from them going all the way back to Decathlon, a game that Microsoft made for non-Microsoft operating systems!). Excel is great. Access seems strong. I'm told that the "Ages of..." series is phenomenal. MS-DOS Edit was a fantastic MDI editor (and boy was I disappointed when they downgraded to Notepad and Wordpad!). Media Player (well, using the Classic skin since the more recent interfaces have been very clunky) is usually fantastic. But it's still my opinion that Windows isn't inherently easier. The thing that is easier for users is the fact that nearly every company in the universe tailors their hardware and software to work best with Windows. I mean, wouldn't Dodge look unbelievably superior if 95% of body shops only did work on Dodge vehicles?
Eh, I'm done complaining. I may have made it sound like Linux is infinitely superior to Windows, but I was mostly overreacting to what I consider an equally extremist (but opposite) viewpoint. Windows 2000 is "good enough" for me. If there were no Linux, I could probably be comfortable using Windows 2000 for the rest of my life. Unlike the nightmare that was Windows 98, I can usually get Win2k to listen to me in a reasonably reliable manner. I use Linux largely because it prevents the need for me, a poor guy, to steal computer programs from P2P networks. I use free (and Free) software on Linux, and the very knowledge that people do these things to benefit others and not just to win a buck both makes me feel warm and fuzzy inside and compels me to be similarly generous with my gradually expanding coding knowledge.
BTW, there is one area where I will be stubborn: Qt beats any OS-specific class/widget programming package ever. I love, love, LOVE being able to develop and compile applications for Windows, Linux, OS/X, various unix variants and a couple PDAs using OS-native widget sets on a single codebase. So pbbbbllllt!
Note: Hey, neat, I just discovered that I can drag
copied text to my desktop background and it'll
automatically paste it into a new text file. That's rather useful.
-JC
> Linux is dead to me. Long live FreeBSD. (But seriously ... I don't like desktop Linux. ... I mean ... why would I use
;P
:)
> It rocks for servers, though. It just doesn't float my boat
> Linux/X11/KDE|GNome as my desktop, when I can just use Windows XP?)
With Windows 2000 or XP on my 800MHz computer, I would not be able to simultaneously write a Mode-2 SVCD to CD-R while watching a Mode-2 SVCD on my DVD/CD-ROM while mass downloading from usenet while printing stuff while web browsing while unraring the next movie to burn. Heck, I can compile code while doing all that, and I'll read some usenet posts while I'm at it, and the chance of buffer underrun is still refreshingly low. On my NT5 box, I have to close *everything*, except maybe a browser window or my newgroup client, and I have to run the burner at half the speed (compared to how fast I burn it in Linux).
Considering that I'm continually burning my favourite television programs and such, I would be in pretty sad shape if I were running Microsoft more often than I currently do. As it is, I go into Windows typically only to play Civilization III.
I'm sure that many people have a lot of ease with Windows 2000/XP. YMMV, as they say. I just happen to be one of the quirky folk in those situations where [Mandrake] Linux installs happen straightforward without incident, setting up the proper drivers and including really flexible software and interfaces automatically, and Windows just gives me driver and application problems (I had to hunt forever for win32 drivers for my 5.1 sound card, my AIW RADEON 8500 is crashy if I try to close the TV app, etc..).
Windows does have two big advantages over Linux, at least in my house: Eudora, and easy compatibility with the Gyration mouse.
Like I said, YMMW. I don't complain if you made your OS decision based on your personal preferences. I only get miffy if your choices are based on ignorance ("I can't run this lennox thing because it's missing the internet explorer button"). So hah!
Oh, and I have no qualms with FreeBSD. My company's web server runs on it, and the daemon dude is cute. Any group whose mascot is Satan gets my vote. Plus, it's POSIX, so I can compile most of the useful apps for it. POSIX is my new best friend.
-JC
> I think the maximize window functionality found in all of
...." menu bar, something wrong with the basic idea of how it works (maybe I'm against turing machines or something), and I think that menus should really get some re-thinking before we really standardize on something like that.
;)
> todays GUIs does this excellently, wouldn't you agree?
To address the original poster's requirement, you'd need an OS setting that forces all windows to open maximized. I don't mind having a windowed system, but I do happen to run almost all my windows maximized, and I have to frequently re-maximize them when I open them. That's an annoying inconsistancy, and it'd just be nice if I could check a box somewhere that made these program windows open up consistently.
> I fail to see what tabs accomplish that the Windows task bar or OS X
> dock doesn't. Tabs are used for switching between windows of a specific
> application, nothing more, nothing less. In fact, since tabs are an
> alternative way of switching between windows, the combination of the two
> (tabs and taskbar/dock) complicates matters even more.
Tabs aren't an alternative to taskbars. They're the same thing. Think about it: They're (usually) a single column or single row of generally rectangular widgets that, when clicked, replace a large widget (typically a full, resizable window) in another location. You right click for a context menu that allows you to close and sometimes resize the larger widget.
Heck, in several MDI applications the "tabs" are presented straight out as a taskbar. They look exactly the same. This is the case with Opera, Eudora and EditPad, three programs that I cannot live without. Right now, I have thirty-three web pages, eleven email windows (boxes, messages and my filter report) and twenty-seven text files open. Do you realize what a horrible joke the taskbar would be if I were using Internet Explorer, Outlook and Notepad in this manner? In any Windows before XP (and maybe ME?), I'd have a bunch of incoherent little icons and a stupid up/down widget just to the left of the taskbar. XP is a little nicer, using the KDE-like feature of grouping similar windows, but I think that interface is still a bit awkward.
The way I use it, generally, the OS taskbar contains a list of Applications, and the Application taskbar contains a list of Documents for that Application. This is the neatest, simplest way of representing and navigating through my documents that I've yet found, and it's all but eliminated the need to hover the mouse pointer over a taskbar item to figure out what it is (from the "mouse hint" popup).
I fully understand that this methodology won't work for everybody. But I do get pissed off when I don't even get the option for stuff like MDI (Multiple Document Interface - one taskbar entry for the entire application). I still use Microsoft Office 97 at work, largely because it doesn't clutter up my desktop. For Office 2000 and XP, they started to shift to an SDI (Single Document Interface -- one taskbar entry for each and every document of the application), which annoys me. Personally, I don't see why more companies (especially Microsoft, an organization with more resources than a city with uranium, oil, horses and grapes in its two square radius) don't simply allow the user to choose between the two. Opera asks you to choose between SDI and MDI when you first run it. In mozilla, you have the choice of hitting CTRL-T to open a new document in the MDI (tabbed browsing) or CTRL-N to open a new window with the SDI (windowed browsing). I can run as many single window instances of EditPad if I wanted, and each EditPad window can have as many documents as I want it to have. I don't know how flexible Eudora would be, though. I suspect that it'd somehow manage to overheat a VIA C3 if I got it to run more than two simultaneous instances (it's an incredibly awesome program, but I can surely see its memory bulk).
What was my point? Oh, yeah.... Cluttering the taskbar is bad. It's the number one no-no of GUI design, imho. Using a subtaskbar for each application is one way to get around this. Another way is to have a bi-level system taskbar: The bottom bar is like a regular application taskbar, and the top taskbar changes to a bar containing the documents of whatever application you have selected in the bottom taskbar. Task Grouping (like, as mentioned before, in KDE and WinXP) is another way (that I happen to find uncomfortable).
I agree with much of what you say about consistency, but I think that each person has a different idea on how it should be consistent. For example, I don't know why, precisely, but there's something wrong with the current "File, Edit,
Eh, but I'm just odd like that.
-JC
http://www.jc-news.com/
> Yeah, and furthermore the dems want to tax the holy f*ck out of me and give
... WHUT'CHA GONNA DO, BROTHER, WHEN POLITICS RUN WILD ON YOUUUUUUUUUU!!
> my hard earned pay to welfare slackers and then have the govt be my nanny and
> protect me from myself in everything I want to do. The reps don't want to tax
> me so much, but lean toward fascism/police state-ism and take away my
> freedoms and let big corps run ramshod all over me and my rights too.
Hmmm. My issue is that the current administration is trying to grab the perks of both party philosophies while igoring all the repurcussions.
Imagine this, from the perspective of a businessman: Your company posted yearly revenues of about two billion dollars for 2002. But on the same note, they spent roughly 2.1 billion dollars in the same year for various business-related purposes, including several expensive instances of sending money and aid to competing companies. Your company has been operating in the red since World War II, upwards of half a century now, requiring you to borrow money every year, often by increasing degrees. Your company's debt is now $6.4 billion, more than three times your revenue for a year.
You became CEO in 2001. Even though the previous CEO did not back your promotion, he convinced the board of directors to pay you twice as much as he earned per year.
Every CEO's first year or two can be problematic, especially when the board of directors differ from you philosophically. But this year, you were able to get all your friends onto the board, so they'll unilaterally agree with whatever you decide.
So what do you do to fix your company's financial woes in 2003?
In your 2003 Annual Report in January, you declare that the company will drop prices, lowering revenue. And you announce that spending will increase substantially this year and likely sequentially over the next decade as well. As a result of that, the company will have to borrow more money this year than it ever has borrowed in any year before. You announce new policies which allow for increased surveillance of both employees and customers. You almost unilaterally decline membership in IEEE and ISO standards and any other standards that weren't specifically set in motion by your company.
But you end your Annual Report with the phrase "God Bless My Company", which means that you're obviously doing the right thing, eh?
Multiply everything by a thousand, and you have the US's current state. Yay, let's decrease taxes while increasing spending. Brilliant. Who's the Secretary of the Treasury these days? Groucho Marx?
> You can;t win for losing. I think we need a new political party, all composed of
> professional wrestlers, to take over and set things straight again.
Aye
Hmmm, on second thought, the problem is that all the popular wrestlers are owned by Vince McMahon, who's basically the GWB of pro wrestling. Now, I'd totally vote for an AJ Styles/Low-Ki presidential/veep bid, but that's not gonna happen.
-JC
> $159k over 3+ years isn't exactly a killing,
> especially for something he was spending a
> considerable amount of his time on (read: full
> time job).
Anything over a hundred thousand dollars in three years is fucking incredible. As a netadmin (and programmer, and web master, and ISO 900x form designer, and part time journalist, and data enterer, and end user support dude, and machine setup guy) that could bring his company down with a simple push of a button (I'm in a room with all the servers, and that UPS switch looks mighty inviting at times), half that amount would be a pretty sweet raise. Heck, you can also figure in that he wouldn't have to spend four thousand dollars over those three years to pay for the gas for the commute. Heck, all that commuting has cost me a good six or seven thousand dollars on my car if you also count wear and tear (and that replacement engine) over the last three years.
Wow. This guy has a dream job. Work from home, exploit stupid people, make fifty thousand dollars a year with far fewer expenses. Shit, if only I had less in the way of ethics and morals....
-JC
http://www.jc-news.com/
> Intel's big problem is the binary compatibility they've stuck with since the
:)
2 .2/0107 .html
1 98 .html
:)
> 80x86 (more or less). Binary compatibility was important because so much
> programming was necessary at the assembler level that changing the chipset
> was prohibitive. This has kept a bad chipset in commission long, long
> after it should have died.
I think you mean "instruction set". Intel changes their chipset like they do their underwear (that is, frequently, though perhaps not as frequently as the analogy implies).
> But then, if you can successfully market clock speed as the sole
> measure of performance, why bother offering something better?
Yeah, that's annoying. I always hated how the clock frequency is always called the clock "speed". I mean, it's not in physical motion. You don't call the cycling of your car engine its "piston speed", or whatever. That is perhaps a trivial sore point for me.
Still, the current version of the P4 is not bad at all. It is arguably an equal or superior microarchitecture to AMD's K7 family, though it's difficult to really make that determination solidly, since Intel has a six to eight month process technology advantage over AMD, and that gives them a frequency advantage somewhat independent of the base microarchitecture.
The Windows user in me is torn between getting an SMT P4 or a K8 ("Athlon 64", I think they're thinking of calling it) at the end of this year. The K8's on-die memory controller should give a boost to some of my operations, but the P4's SMT functionality would likely benefit me, as I have a tendency to run lots of apps at once (I make most power users look like AOL newbies in some respects, heh). Certain cpu intensive programs (like SmartPAR) that eat up all my time might run better on the AMD setup, while other programs (like WinRAR) will likely enjoy the benefit of the Intel box's higher raw memory bandwidth and cpu frequency. I guess that's a "wait-and-see" type of thing.
The Linux user in me is a steadfast AMD supporter. This has nothing to do with any "WinTel Conspiracy(TM)" or whatever; it simply appears to be the case that any AMD chip is substantially faster than an equivalently rated Intel chip in most Linux-based benchmarks. I am a little interested in seeing how much of a benefit the SMT gives to gcc, but it would take a lot to convince my Linux side to move over. The Athlon XP 2700+ seems faster in Linux than the 2.8GHz Pentium 4, and that's without the added benefits that the Barton brings to AMD's K7 core. Heck, that little Linux daemon (hrm, or maybe it's the FreeBSD dude) inside my head keeps telling me to drool about how much faster than Barton the K8 will be given its advantage of far lower memory latency (due to that on-die memory controller), 64-bit registers, doubled GP register space and those HyperTransport connections. I keep telling myself that only the memory latency and extra registers would make a difference, and even with that compiling probably won't be that much better per clock than with the K7, but even with minimal improvements, K8 should be faster than K7 in compiling, and since K7 is much faster than the Northwood/P4 in compiling, the K8 should be substantially faster at the task. Except for that unknown variable of SMT. I'm going to have too look and see how much it can add to the fray. In addition, it is not unlikely that the P4 will simply scale in frequency by a greater degree than AMD in the next ten months.
Oh, new data:
www.uwsg.iu.edu/hypermail/linux/kernel/021
Allegedly, you get something like a 15-20% increase. Not bad.
www.uwsg.iu.edu/hypermail/linux/kernel/0212.2/0
But this guy is getting some sort of substantial decrease in performance due cache problems between threads (I guess that there's more cache misses since twice as many threads needs twice as much data).
Interesting.
Damnit. Why can't companies give me this stuff for free so that I can test it all for myself? I'm a coder, and I have to know what hardware can render my code AFAP!
But it's all fun, anyway, this talking about microprocessor technology.
-JC
> I believe the code is up-up-down-down-left-right-left-right-B-A. Hitting
> start was just to start the game. Your code would probably work because
> the extra B-A-select-start wouldn't interfere with the code.
Konami put that code into a ton of games, and I vaguely recall both variations being used. Perhaps the select-start was done in the earlier games and they dropped it afterward.
-JC
http://www.jc-news.com/
> After spending 20 years as one of the lowest paid (yet consistently employeed)
> network/sys admins on the planet, yes, I will get a pension, benefits, etc.
Interesting. I was under the impression that government jobs pay astronomically higher than private sector jobs in the computer field. I, for instance, make $24K a year in an occupation which involves user support, programming, computer setup/repair, website maintenance, ISO 9002 complaint form generation and (whenever there's a lull in that other stuff) heavy data entry. A friend of mine, who graduated with a similar degree at a lesser institution (as evidence, I present that after graduating he didn't understand the basic concepts of TCP/IP, C++, superscalar pipelining or modem handshaking), makes $56K a year for the FAA doing the user support and computer repair (I don't think that he does the setup).
Granted, he gets more stress from his job. The folks there make fun of him, and he has to take some sort of anti-depressant medication that "zonks" him out (his word), and he has to visit a psychiatrist once or twice a week, but on the top side he also gets to visit other states and take computer classes, all paid by the government (they even pay for his food when he does these things!).
I think that I would give up a little piece of mind for an extra two or three thousand dollars a month, free vacations to Pennsylvania and upstate New York and free learning classes. Hell, I went through a nearly complete mental breakdown last year and had to spend nearly the entire year working part time instead of full time in order to avoid going completely bonkers. So I don't think I have any real palpable benefit over this fellow with the government job.
-JC
...who loves to bitch about his job
[off topic]
:p
The funny thing about your comment is that while the former WWF company (now known as WWE) has often been symbolized for chaos and violence, it is actually the most conservative and least violent company in its entire genre. I've recently gotten somewhat caught up in all this -- my friends are huge pro wrestling fans, and I've been watching it on and off since last May or so. I like a lot of the actors, and I have a great deal of respect for the performances they give, though I prefer to focus more on independent productions, since the writing quality of the mainstream product is crap, and that particular company has a lot of contempts for its fans.
Anyway, the former WWF is the least chaotic of this form of performance art (compared to, say, NWA or XPW). But the irony that I was initially trying to put forth here (I have ADD, and I sort of go off on tangents a lot) is that the *current* WWF (the World Wildlife Fund) offers a rather interestin analogous situation. Last year, they sued the similarly named wrestling company for using the same three letters in a totally unrelated industry and, in the face of common sense that would say otherwise, and in the manner in which RIAA totally and illogically decimated Napster, utterly won.
It's funny, these things. Well, it is to me.
-JC
> How's that? Did someone actually find some good music to steal?
There are people who use Kazaa without ever downloading mp3 files. I don't download music, but Kazaa is useful to me, and I hope it becomes an important stepping stone towards easy distribution of free (whether beer or coffee or whatever kind of free you care about) software in the future.
My life will be so much better when I can download those operating system ISO images without having to wrestle an ftp server into the ground.
-JC
PS: I do admit that I may have helped my sister download mp3 files in the past. But her unhealthy addiction, much like someone's addiction to an illegal drug, must be weaned out of her gradually; I'm certainly not going to call the police on her, as that would obviously ruin her life without causing benefit to anyone.
> AICN nerds make Slashdot nerds look like Puff Daddy.
You mean like exceedingly wealthy stuck-up, self-important assholes? I dunno if I want that label. Except for maybe the wealthy part.
-JC
> For most home (and, indeed, server) applications, I would have thought that having a
:-)
:)
> dual core processor won't make much of a difference, just as processor speed doesn't -
> rather, what is important is the speed you can get data in and out of the processor.
> Overall CPU speed doesn't seem to make much of a difference when the bus speed is the same,
> certainly not in the systems I've tested. However, up the CPU bus speed, and you'll find
> your performance greatly improved, because you're getting data to the processor quicker.
> Some years ago, I tested this theory with a couple of old 686 chips - one 200, one 233. I
> benchmarked the 200 and 233 both at 75MHz bus - virtually identical results. Then I ran them at
> the same CPU speed, but 83MHz bus, and the benchmark results improved by exactly 83/75.
> What does this tell you?
It tells me that the benchmarks you use are not the same as the benchmarks I use. Here's my rundown:
Civilization III: The complexity of this algorithm increases exponentially based on number of cities. I'm running a 16 civ game at the moment, and the game literally takes more than twenty minutes to cycle a turn, and it's only the nineteenth century! Granted, my 800MHz Duron isn't state of the art, but it's not state of the fart, either, and even the top of the line processors would buckle under this stress.
PAR: I download very large binaries off usenet that are separated into multiple files. PAR is a tightly coded system that makes extra parity files that you can use to build missing files of a download set. It takes a *long* time to verify the parity on 800MB files.
RAR: This probably would be helped somewhat by faster memory access, but I suspect that extracting an 800MB multipart RAR set would also be strongly enhanced by a faster processor. I mean, it'd be nice to not have to wait five to ten minutes to extract this stuff before viewing it.
Qt: Compiling scales almost directly with processor speed, at least on some types of code. And I do a lot of compiling on Linux. And I do a fair amount of compiling on win32, as well. Compile of a large program can take many hours. When I type './configure && make' or 'qmake -project && qmake && make', I want to get up, prepare some coffee or munchies, come back with the yummes, and immediately test the newly created binary.
There are other programs that I use that depend on the processor, but those are some of the biggies. There are also programs that would benefit from faster data access, of course. But memory isn't really getting faster. It's just getting higher bandwidth. When you request data from PC2100 or PC800 memory, that data doesn't start coming back any sooner than it did with PC133 memory. It's just really expensive to increase the frequency of the entire northbridge and corresponding devices. That's why the microprocessor has been doing the majority of the speed boosts. It offers the most increase for the least expenditure.
-JC
http://www.jc-news.com/
> Wow - way to make up stuff.
:(
> The Itanium II is certainly not a dud
Agreed. From an engineering standpoint, it's quite a nice chip. I don't agree with some philosophical stuff in the ISA (I'm not that much of a VLIW-for-general-purpose fan, but hey), but the microarchitecture and implementation seems very nice. I do wish that it was easier to implement OOE on IPF, though.
> x86 compatibility is worthless ina high end 64-bit machine, somethin AMD doesn't seem to grasp.
> They're marketing a high end technology (consumers and normal business users don't need
> 64-bit technology and won't for a while) to the mainstream market. Morons.
Feh. A big "screw you" on that. AMD isn't catering to the high end server group. They obviously can't just teleport into that market. Their catering to the smaller business that uses Xeon servers. Backwards compatibility with x86 is of the utmost importance in this market. Basically, they're marketing x86 workstations and x86 servers that happen to allow you to enhance performance of some types of programs with simple recompilations. There is a good chance that I might get the lower end version of this product when it comes out, as I use Linux, which may strongly benefit from those extra registers in x86-64, on my home machine. We'll have to see, of course, before I pull out the green.
> And you seem to be ignoring the numbers (remember that 'reality' the rest of us
> live in matters to us, if not to you). AMD is going broke. Intel isn't.
That's a bad measure to use. You don't have any controls in this analysis. There are a lot of reasons why AMD is losing money (poor management a la Hector Ruiz, inability for a relatively small company to handle a very harsh recession, etc..), and there are a lot of reasons why Intel is still doing phenomenally (people buy Intel no matter what, currently excellent execution, they can afford to strongly diversify). Many of these reasons have nothing to do with the technical/engineering side of the equation. IMHO, both AMD and Intel have incredible engineers, and frankly AMD especially warrants respect for being able to ramp technology at *approximately* the same rate as Intel despite having a very, very miniscule fraction of their resources. That is why I was a big AMD fan a couple years ago, at around the time when the company was dominated by the excellent triumverate of Sanders, Raza, Meyer as well as a couple critical folks like Norbert Juffa and Paul Hsieh. At this point in time, AMD was a quantum of a company that somehow managed to produce a piece of engineering that allowed them to, for a brief time, outdo the capabilities of a company fifty times their size. I am somewhat dismayed that AMD turned into a more traditional company over the last two years or so.
-JC
> The Itanium is unusable. Have you ever seen
... you know what I mean!).
> one production system with it? I have not.
I know somebody whose workplace (a research place of some sort) got a cluster of five hundred of them.
> Itanium II is now out and is said to be OK. For the price of an Itanium II
> system you could buy a car/house/small country.
Um. High end servers are supposed to be that expensive. Ever try shopping for a high end UltraSPARC or Power4 machine? Didn't think so.
IPF isn't supposed to be a replacement for x86 (well, it was originally eventually supposed to, but that's when various Intel execs were drunk on monopolistic stupidity). I do not agree with the means by which Intel has penetrated the market (eg, they coaxed several prominent chipmakers, such as HP, DEC, SGI, and so on, to dump or devalue their existing lines and support the Merced far before it reached A0 stage), but on the engineering side, the McKinley (or "Merced II", if you like fugly names) seems an excellent implementation of what was perhaps a not too well thought out ISA. Just my opinion, of course, and I'm merely an armchair designer (no, no, I don't design armchairs
-JC
> and the relative lack of success of AMD (I use an XP1800+ and think its great,
> the company just doesn't seem to do too well.)
Well, part of that is crappy management, but a large portion of their troubles are simple due to the fact that Intel is given the benefit of the doubt by the OEMs and the consumers. Even during the year or two when AMD consistently had faster chips with fewer bugs than Intel had, Intel made tons of money and AMD merely made enough to recoup past debts. People buy Intel because they're Intel. This will happen whether Intel is doing a good job or a bad job. Thankfully, they're doing a good, honest job and earning those buyers now, but from 1998 to 2001, they were not doing their customers honour.
> Intel doesn't need to keep spending money researching new chips
> if it's current generation are so far ahead of its competitors.
They aren't. Intel's Pentium 4 is pretty much on par with AMD's Athlon. But Intel has five or so x86 plants that they can leverage to test different ways to most optimally ramp their chip frequencies. You don't just throw a design and a fab process into a bucket, shake it, and come up with the resultant chip speed. You have to devote a substantial part of your manufacturing resources to the research needed to optimally match your current chip design to your current manufacturing technology.
In addition to this, Intel happens to be something like a year ahead in base process technology. They moved to 130nm six months before AMD did their equivalent move. This means they're very much ahead in that respect. So if their chips were a generation *behind*, Intel would be competitive in chip performance (this is was almost happened with the Pentium III and the early implementation of the Pentium 4). As it is, the current P4 is a competitive design coupled with a slightly more advanced manufacturing process, so Intel is a couple speed grades ahead.
Intel has to keep researching constantly. AMD does a surprisingly good job at ramping technology at approximately the same rate as Intel, despite having about a twentieth of their capital resources. If Intel stopped researching for just a few weeks, they'd lose the leverage they have to stay superior in the current climate. And that's not counting on the outside possibility that K8/Hammer might exceed performance expectations and outperform the top Pentium 4 upon release.
-JC
> the Hammer/x86-64 chips have ondie memory controllers AND more registers than i386++ type
:)
;)
> chips, combined they'll give a speed increase of not inconsiderable proportions.
It is notable to, er, note that the former advantage helps (possibly considerably) towards both recoded AND legacy (eg, normal) programs, whereas the additional physical registers would require recompilation in order to show a benefit, which means that everybody but Windows users will get an immediate use outta that.
Other advantages of the Hammer? Well, not counting the 64-bit yunk (that WILL provide benefits, but I want to cover benefits that will help legacy programs, like Civilization III, The Sims, Unreal II and other antiproductivity applications):
Hyper Transport. That's not much on its own, but it essentially equates to a reduction in loss of bandwidth to the chipset and between processors when you add an additional processor. On the Intel setups, the processors share a set amount of bandwidth to the chipset, so putting eight chips on a 2.4GB/s bus means that you have each chip getting 0.3GB/s. The AMD setup theoretically lets each processor get that 2.4GB/s. Of course, that's in a perfect world, chip-level, but it probably amounts to some benefit. AMD's K7 family has similar advantages, which probably assists in explaining why they get higher performance at each given clock in mainstream applications (which at least somewhat depend on the memory subsystem) even though the AMD cpu to memory bandwidth was usually 2.13GB/s (now it's 2.67GB/s, unless you count stuff like the nForce, which has some extra memory bandwidth, but the extra benefit there is eaten up by the onboard video), whereas the Intel cpu to memory bandwidth was usually 3.2GB/s. Anyway, the idea is that HyperTransport will (on a hypothetical level) make it much easier to make n-way systems without either a tremendous performance impact or an expensive crossbar workaround setup thingy.
SoI. Silicon-on-Insulator. This is one of those things that'll help with the process technology. In the end, it'll probably offer a little bit of a frequency boost by making the chip a little cooler or something like that. I forgot precisely what SoI's primary benefits were. It's been months since I've even thought about it.
Stages: As detailed here, the K8 adds two stages to the decoding part of the instruction pipe. The decoding part of the pipe is probably rather complex, so you may see a pretty neat frequency boost over the K7 family without the problem of a huge branch mispredict penalty. The number of cycles that a cpu wastes when it makes an incorrect guess on a low level "if/then" statement is somewhat proportionate to the number of pipeline stages. The AMD K6 and (iirc) Cyrix 6x86 were the mack daddies of branch prediction, since their pipes had only five stages or so, so they only had to wait a few cycles when they zigged instead of zagged. The PIII and K7 had over ten stages, so they had to wait a lot longer, but other advantages (such as the larger and sometimes faster caches and more accurate predictors) in those processors over their predecessors did their best to overcome this disadvantage. The Pentium 4 has a crippling 20 to 28 (depending on the situation, and depending how the trace cache handles the situation, and whether or not you want to count it) stages. This means that it can hit amazing clock frequencies, but it'll get cranky and drowsy for twice as long when it makes a predictive mistake. How does it get away with this? Well, the trace cache does its best to assist, but it didn't really help as much as I think the designers were hoping. But for multithreaded programs and OSes, the SMT implementation on the more recent members of the P4 family, an implementation known as "Hyper-Threading", probably pretty neatly alleviates much of this problem by putting operations from other threads into the cpu whenever the currently running thread stalls on a branch mispredict. The K8/Hammer approach is just to add stages where they hopefully will have the most balanced, beneficial effect to frequency boosting while only minimally increasing the branch penalty. SMT would be nice, but it isn't nearly as critical a need as it is on the P4.
Wider memory access. On the Sledge Hammer, if AMD's plans are still the same as when I wrote this, the memory controller (which is embedded onto the cpu) will access PC2700 memory in a 128-bit configuration (ignore the "126-bit" typo on the linked page -- I can't believe I didn't notice that when I typed it nearly a year ago!), which leads to a 5.3GB/s path to memory. That's damned good, though I really think AMD should have focused on 366/183MHz (equiv to "PC2933") or 400/200MHz (equiv to "PC3200") memory instead of the 333/166MHz PC2700 that came out over a year ago. Still, servers often use memory that's lower than bleeding edge clock in order to maintain reliability, so bleh. Still, 5.3GB/s isn't bad for a setup that isn't based on a shared bus.
Enhanced branch predictor. Well, that's if my notes from a year ago are accurate. If true, this'll probably overcome any mispredict penaly performance disadvantage from those abovementioned added stages.
Larger TLBs, TLB flush filter, etc.. This stuff will have itty bitty advantages on a per-clock performance basis, but every little bit helps.
Larger caches. Hey, I should look this up to see what they're planning on. Is it just 512KB on-die L2, or is AMD planning on bringing it up to 1MB L2? The interesting thing about AMD's designs is that the die is really small on each processor. Remember how AMD has gotten occasional fire for processors overheating? Well, aside from a stupid lack of shutdown diodes in the past, the real cause wasn't that the AMD processors used more heat than the Intel processors. They usually generated about the same amount of heat, often less, but their processor surface area was substantially smaller, which made the chips less expensive to produce and less likely to have defects. But when you try to push an equal amount of heat through half the surface area, you end up with a higher amount of heat per area, which equates to a higher running temperature. The funky thing about this is that you could just added a whopping huge amount of on-die cache. That'd increase performance while also increasing the surface area. But the heat production would not be substantially affected. So you'd end up with a lower temperature processor. So the Hammer will have a higher ratio of cache to processor units in the cpu, so it won't be as much of a fire hazard. Frankly, they should have put 1MB on-die L2 onto the Thoroughbred/AthlonXP.
Crap. I need to research more on the K8. It probably changed a lot since I went into hibernation. The interesting thing is that in the last half year, I've largely moved from being a Windows 2000 power user to a Linux coder (I still use both operating systems for several different purposes, but I'm talking primary usage). I stand to be in the group that benefits most from the Hammer when it comes out, since I'll be able to './configure && make' or 'qmake -project && qmake && make' most of the programs I use and/or develop. So I'll instantly see the benefit of those extra registers. ^_^
-JC
http://www.jc-news.com/