Try again and gcc is terminating with some obscure error.
You may have a broken cpu or mobo. I actually have this behaviour on one of my systems. Put one cpu in. Works fine. Put second cpu in. Breaks. Remove first cpu. Still broken. Switch second cpu to known-good first slot. Still broken. Thus we conclude that the second cpu is bad. Since cpus these days have cache on board, it's entirely possible that the cpu itself is fine but the cache is bad. This seems reasonable, since cache is still memory and bad memory is the known cause of random sig11s. HTH.
Well, whether anyone cares is another matter, but it almost certainly will be an option. Linux already runs on everything up to an E6500 (known) and theoretically supports the E10k (nobdy's ever gotten hardware to test on though). I would certainly consider running Linux on US3-based systems. Up to about 8 cpus it beats the shit out of solaris. Past there people assume it would lose, but it's hard to know for sure as the most we've ever seen tested is 14. (shrug) Maybe nobody talks about running linux on these things, but certainly it's possible.
but 2 Terabyte memory support under Linux is IMHO much more interesting than the latest rumormongering from Sun.
There's nothing special about 2TB support. First, you can already do that on UltraSPARC, and second, Alpha is a fairly obscure platform (right or wrong, it is). What would be real news is if
working for a company which has nearly completed the process of dumping Sun in favor of FreeBSD and Linux solutions[...]run an operating system which has no compiler included
Linux runs great on Sun systems. You imply that Sun and Linux are mutually exclusive; they are not. JMHO of course, but I find that Linux on UltraSPARC is far superior to Linux on peecees. If you like Linux on peecees, you'll like it on Sun hardware too, and in that realm the hardware isn't nearly as dodgy as peecees. If your budget makes Sun hardware impossible, fine. But don't imply that Linux on peecees can touch Linux on Suns. It can't even come close, and the hardware is clearly the limiting factor. It may meet your needs, and that's fine, but there's no reason to imply that a good, complete, low-cost OS and nice hardware are mutually exclusive. It sounds to me like you should have used your existing hardware and simply switched operating systems if the OS was giving you trouble. Oh well, more used Sun equipment for me to buy cheap.
Actually both SGI and sun pick kernels for a particular architecture (for example, IP27 or UltraSPARC II). They aren't actually identical. Sure, it comes from the same codebase and in most cases it's possible to run identical kernels on different machines, but usually it isn't done that way. The best example I've seen of actually doing it that way is UltraLinux which can do all sun4c, 4d, and 4m including SMP with a single kernel (Solaris _has to_ use separate kernels for 4c - not that anyone cares since 4c is essentially useless now). Even they can't do 4m and 4u from the same kernel though, nor can SunOS.:)
But I've been stuck working on our Sun Ultra5 worstations far too often, and they are TERRIBLE. Sooooo sloooooow. They're configured with 128M of RAM and, if I remember right, a 300 mHz sparc. My PII-350 with Linux and the same amount of RAM is much more responsive. Not to mention that they go down with alarming frequency, and they cost four times what I paid for my intel box around the same time period.
Yep. We have those too, and they are complete shit. As people say, "the only thing Sun about an Ultra 5 is the price." I'm sure the Sun technical people are sick of being hated over the U5/10. If you look at it, the decision to develop that kind of system could only have come from marketing. There's no way anyone at Sun really believes an U5 is worth having. If you want to judge Sun's workstations, get an Ultra 2 or an Ultra 80. These machines typify Sun's capabilities. Expensive but worth it.
Will SparcStations be able to survive the onslaught?
Well, the most recent machine to carry that name is obsolete, and only the fastest versions of it are still useful. So I'd say no.:) If you mean "will SPARC survive" then the question is more difficult. There will always be a market for something that isn't Intel (ie doesn't carry the baggage of the 4004 along with it). Whether that will be SPARC I don't know. Your question about McKinley destroying its competition is likewise unanswerable. Intel is betting a lot on what is really unproven technology while the traditional RISC makers are improving their technology one step at a time. By the time Intel finally ships their sooper-dooper new processors they may be well behind the "older" technology of other vendors like Decompaq and TI/Sun. If so, SGI would be foolish to ditch MIPS. It'll be interesting to see how this plays out.
Yes indeed. However, that's not the whole story. SGI's high-end systems (and the E10k for that matter) blur the line between cluster and single machine. As a low-end example from the SGI world, you can attach two Origin 200s using the Craylink and to the OS it appears as one machine. Is that a cluster, or one system? Well... Similarly, the Craylink technology is used with the O2k series to interlink systems as well as build composite systems. There even exist routers, hubs, and so forth. So it's really just an exceedingly fast networking technology, but with some clever hardware and OS support can also be used to make multiple systems tied together look like one. Very clever, very nice, and very fast. Expect to see more like this.
Yep. Sorry I sort of left the BSDs out of the discussion, mostly because Linux emulation is available. Certainly it would be nice to see native ELF BSD versions as well however (as well as versions for xBSD/y, where x != Free and/or y != Intel). Especially since there's absolutely no reason it can't be done. Just one flag is all it takes... That's what burns me about the whole thing. It takes far more effort to keep obsolete systems around for the purpose of doing builds than it does to just use up-to-date systems and support the software people are actually using. Your annoyance is justified.:)
This argument does not make sense. The fact that a codebase is old and nearing retirement does not mean that its builds have to be linked against old libraries or only made available for old operating systems. It's obviously possible to build communicator against glibc 2.0 on Linux 2.2/Intel. So what's stopping them from going to glibc 2.1? Or Linux/SPARC? Or Solaris 7? Or...? If the Mozilla codebase were intended for release as Netscape 5 within a few months, I probably wouldn't consider this a big deal. But seeing as how it's going to be NS 6 and is probably a year or more away from release in that form, I consider this a serious issue.
How much more ground is Netscape willing to give up here? Will NS5 work on modern platforms? Or will Microsoft take the opportunity to crush Netscape for good, before Mozilla has any chance to make an impact?
I know Mozilla won't die off completely, a nice benefit of Open Source and all that, but there are already pages that will only work in IE (just as there were mages that Navigator but not Mosaic could display, and this eventually killed off mosaic). How long until it's a choice between security fixes and a working web browser? I know which I'll be forced to choose if it comes to that, and the thought of losing like that is unpleasant.
1) Your distro manufacturer may have packaged netscape incorrectly. See their site for details or upgrades.
Well, I use the latest one from SGI with their own customizations. Neither it nor the standard Netscape-supported version seems to be immune to the typical crashes and hangs. On Linux, AKAIK RedHat just sets it up to use the libraries it was linked against if they're available. See my other post on Netscape not supporting current platforms properly/at all.
2) Turn off Cascading Style Sheets (Style Sheets) support in your preferences. It generally doesn't work well at all and really isn't all that necessary. And IME it makes NS crash. A lot.
Do not do this. CSS is a good idea. The separation of content from layout is both necessary and good. Now, it shouldn't be harmful to your ability to glean content from sites while browsing without CSS; however, without CSS it's assumed that you don't care what the pages look like and thus you might as well just use Lynx. CSS support in Mozilla actually looks good. Somebody there read the standard, which I can't really say about Netscape. Anyway, turning off CSS just encourages webmasters to use other methods, like crufty HTML or "that other browser's" proprietary extensions. Much like the lack of CSS in the first place caused Netscape's extensions to become de facto, and later, real, standards. Bad, bad. If CSS causes crashes, file bug reports or complain to the webmaster in question.
5) Don't invoke mystery components like Messenger and Composer and all that crap unless you actually need to use them. They tend to suck a bit.
Agreed. If you don't use them, consider Navigator instead of Communicator. Quicker to download as well.
The bottom line is, whatever you do, Netscape is still a crufty, bloated piece of crap on any platform they bother to support. Part of the problem is probably library and kernel version mismatches; since they refuse to support the current versions of most platforms, it's no real surprise that random crashes happen. Fixing this would go a long way toward improving stability. Aside from that, most problems seem related to either Java or Javascript. I agree with this poster; disable them if you can.
Netscape/AOL's choice of platforms to support makes no sense to me. Figure this: Linux 2.2 is the current version; most people have stopped using 2.0 by now. Libc5 hasn't been supported by its authors in almost three years. So why is it that the Linux 2.2/glibc version is a) still unsupported, and b) linked against not-even-current versions of those libraries, especially when they still support 2.0/libc5???
As if that weren't sufficiently annoying, there's not been a Linux/SPARC version since 4.51, but there are five Solaris versions. Of course, none of those are for current versions of Solaris either. All this, of course, would be excusable if they didn't have any systems to build on or they weren't going to support oddball platforms (I will admit that Linux/SPARC is marginal, though it has multiple supported distributions and a growing user base), but then why oh why are there two SINIX builds??? Most people don't even know WTF SINIX _is_, much less have it.
The point? It doesn't seem like Netscape understands what people are actually using today. If the objective is to be compatible with as much as possible, then not having up-to-date Linux builds for at least Intel, SPARC, and Alpha makes no sense. For that matter, they don't even claim to support Solaris 7 or 8. 2.5.1 is itself archaic. If the objective is instead to support only the most popular platforms, then I certainly don't see the need for five Solaris builds, two AIX builds, anything related to SINIX, or a Linux 2.0/libc5 build. The obvious platforms to support would be Solaris 7, Linux 2.2-intel/glibc, the latest AIX, the latest Digital Unix, and IRIX 6.5.
So what can we do to increase their awareness of this problem? Numerous polite letters have garnered either no response or a polite "get lost" form letter. Ideas?
Alternate paranoid theory: AOL wants all the Unix builds to be against old systems so that people will switch to windoze and buy more aol service. Pretty paranoid, but aol is pure evil after all.
PS: Kudos to the mozilla team for recognizing the value of compatibility and multi-platform support. The Linux/SPARC build works as well as any other.
--TM, still using 4.51 on Ultralinux, the preferred platform of all major deities
Of course the downside is we'll probably wind up with enlightemnent 0.18's source code just being the words "NUMBER FOUR. NUMBER FOUR. NUMBER FOUR." over and over.. so maybe we should only let them have it during the design phase or something. i dunno.
No, this is Raster we're talking about. It'd have to be "nUBMER FRUO"
I noted kernels as a specific exception. In fact, I don't know of any common kernel that will build with multiple compilers, except possibly some odd Mach variants. For userspace applications, failure to compile with a compliant compiler is a bug. I fix these bugs, and report them. But I think you overestimate the number of packages that have these problems. Or maybe you just use a lot more crappy software than I do. Sure, going to Freshmeat and downloading every two-bitter's kewl h4ck and trying to build it without gcc is going to be difficult. But then, most of that stuff doesn't do anything useful even if it does compile, so I fail to see an issue there.
As for my projects, I make sure to test with as many different environments/compilers as possible. It's good practice, ensures portability, and helps find bugs that might otherwise have been missed. I'm not surprised to find that not everyone does this, but that doesn't excuse them. If these "Open Source advocates" want to talk about standards, then they need to follow them, and ANSI C is one of the most important. To do less is hypocrisy of the worst kind. Quite frankly, I would prefer that Linux could be compiled with any old compiler, but I understand the reasons it can't be. Kernels and their native compilers are just too closely tied. Other vendors do it as well. It's not ideal but it's workable.
So kudos to Borland for giving people with few resources to learn C++ on Windows if they want to.
Naturally I don't object to Borland doing this, but I wonder whether you're aware of efforts like the cygwin project that have not only given poeple that same opportunity, but have given the opportunity to learn C or C++ in a portable environment. Using gcc under dos (if that's your cup of tea) makes the inevitable transition to Unix much easier. As an added bonus, gcc is moderately Free Software. As I said, I don't have a problem with Borland giving away their tools, under whatever license, but it's important to note that they aren't giving anybody something they couldn't already get for nothing.
For the truely paranoid, though, Corel now has a way to make a proprietary Linux distribution. They still have to open the source to their mods, but it doesn't do you much good if it will only compile with Borland's proprietary compiler.
Indeed. While I doubt anyone would waste the effort to do this, I hope Corel does. That will further discourage people from using their awful distribution (I never thought anyone could make Red Hat look good; then Corel came along). With any luck it'll also discourage further clueless luser Linux adoption in general.
Well, buying software isn't exactly popular among Linux users, so if I were to release some Open Source software now which compiled only on the Portland compiler, it wouldn't be a big success, would it?
If you release something which only compiles with one compiler, I don't expect it to be much use anyway. There are rare exceptions (kernels come to mind) but for the most part I want programs that will compile using any standards-compliant compiler. Strange as it seems, most Free Software fits that definition; I've compiled a great deal of such software with SGI's MipsPro compiler. It works just fine, thanks.
No, but I expect it won't be long until such offers as this Borland C++ compiler will be available for Linux (yes, probably without source). And I'll consider that day an important point in the development of Linux, more so than KDE or Gnome.
Good for you. I consider an ant taking a shit more important than kde or gnome however.
Well, if you don't consider having a choice of compilers as important to anybody, I'm afraid you haven't grasped what the little word "open" is all about.
I don't care one way or the other. I want a compiler that works and complies with standards. I have one. If you feel it's necessary to have another, have at it. That was my whole point.
Yes, that's the response I expect when posting anything else but "everything's great" about Linux on Slashdot.
This has nothing to do with Linux. It has everything to do with having a choice of compilers. I believe that choice is already sufficiently available, and that the existing compilers are adequate. You obviously don't. Thus the onus is on you to correct that perceived problem rather than griping about it. Start a project. Announce it here and elsewhere. If others share your views, the project will be successful.
Quite frankly, I'm not even in the "Linux is great" camp. It's an OS. It works for me. It's not perfect and neither is gcc (the real subject of discussion here). Both are, in my opinion, sufficiently better than anything else out there, for my hardware, budget, and interests, that I choose to use them. You aren't required to make the same choices, and I really don't give a rat's ass whether you do or not. My point is that if you believe the existing products are inadequate for any reason, it is your responsibility to either correct the shortcomings, live with them, or change to a platform that offers choices you believe to be acceptable.
You have two choices: write your own (hard, but I'm sure that your obviously amazing programming skills are up to the job), or fork gcc and start making it whatever you think it should be. History shows that your changes may well even be integrated back in at some point. Actually, there are also other choices - the Portland Group sells a proprietary compiler for Linux, as does Decompaq for Alphas. So, "quit yer bitchin."
Do you really expect someone else, who finds gcc works just fine for his/her needs, to sit down and write a compiler for you in the name of open computing just so you can have additional compiler choices?
There are two types of projects in the Free Software world: those that are genuinely important to somebody, and those that never happen. It seems that Slashdot attracts a lot of posts about the latter type, probably because it's easier to post "I want a..." than it is to post "Here's my patch to..." Wankers.
--TM, thinking it's time to fork the community and let the lusers rot
Slash CAN be beaten. Post anonymously to your heart's content, then moderate away. Then don't post anymore, even anonymously. This has the advantage of being able to moderate your own (anonymous) posts up, though there are usually better things to do and that's rather unethical anyway.
And let's not forget the most important part of all. Has anyone here ever actually used this compiler? No? Well, I have, and I can tell you all that even if it they gave me money and the code I still wouldn't use it. Seriously, people, this is one of the worst compilers available to the free world. The cygwin32 environment runs circles around it. Even, yes, I'll say it, even visual studio is better. GCC in the unix environment, or any of the unix vendor compilers (even Sun's, barely) is in another league altogether. Borland/Inprise/Corel/whoever-the-fsck can only dream of competing at that level.
You aren't missing anything here. In fact, just to let them know how the world feels about their "tools" (appropriate, no?), don't download it. Prove that they can't move that crap out the door even if it's "free."
Funny, that's all I ever get, thanks to $CHEAPASS_UNI. And really, I fail to see how any sort of catastrophic failure could make the net any worse than it normally is. ISTM that it would be a lot more fun if everybody was on 64k leased lines and mail went through UUCP again. At least there'd be no ad banners, and.com stock valuations might return to rational levels.
Well, maybe a wee bit more. I don't blame Apple for the disk failure. That happens, oh well (blame Quantum if anyone). I used a screwdriver. Removed every single screw in the thing. Finally found some guy's online "how to disassemble a 6400" page - including the line "now pry the front panel off with two screwdrivers. You will need at least two people for this. You'll think you'll break it, but keep going. It can be made to come off." Fun, fun. Couldn't reinstall MacOS because we didn't have a copy, though ultimately we borrowed another department's copy, since the 6400's OF implementation is apparently not up to the job of booting anything other than a macos rom (another four hours wasted trying, deep inside Apple's "tech" docs - ugh - good thing the macos has no functionality; I'd hate to see these clods try to document it). I'm not stupid, really. But for something that's supposed to be friendly, I was really just appalled. I expected peecee-ish hell but instead got something much worse.
As background, we had this machine dumped in our lab by a guy who walked out the door. I certainly didn't own it or anything, and we had no other macs at all, so knowledge of them wouldn't have been easily found. It was something of a "hmm, what can we do with this?" project. The performa 6400 mightn't be representative, but it turned my disinterest in Apple into active dislike. Fair or not, that's how it worked. Sound a bit better than "I hate macs"?
You know, I'm supposed to be pissed that you think I'm an idiot, but I really just don't care. Macs aren't my area of interest. I don't work on cars, either; does that make me stupid? But I really must respond to this one:
How many six-packs of malt liquor did you and your friends go through during your traumatic 6 hour ordeal?
Notice you didn't even address my first contention: PC hardware doesn't require stupid design and all the things you bitched about. Sure, there's poorly made PCs, but that doesn't mean other people can't put together a clean box. By claiming so and making such a wide generalization, you are trolling. No other interpretation. Explain how any statement containing "peecee luser" can be anything but a troll on a forum with many PC owners.
Look, I've been around for quite a while. I've seen, worked with, and used peecees from every major manufacturer, including clones, homebrews, and "other" ranging from $400 cheapo specials to $5000+ "servers." I've yet to see one such machine built in a reasonable, well-thought-out fashion. Some of the $10k [c|*|*|*|*|q] servers come close, but even those are really just overgrown peecees, and for that kind of price you can do better. I could buy all custom components and put together a nice system, but it would cost more than just buying a workstation from a decent manufacturer. Do peecees have to be built poorly? No, but for some reason all of them seem to be.
We're all supposed to pay huge prices for expensive non-i386 computers like you? You remind me of someone who has an expensive sportscar and bitches about all the econoboxes out there without realizing maybe others can't quite spend so much. Get off your high horse and realize some people make decisions based on performance per dollar.
I'm getting excellent performance per dollar. The important thing is in how you measure performance. I will agree that modern x86 processors can't be beat if you measure performance as "integer instructions executed from L1 cache per second." But I can funnel far, far more bits around with my workstation than any peecee can, and in fact, this machine probably has more disk-to-ethernet performance per dollar than anything that comes in a beige box ever has. As with most things, it's all in how you measure. I don't think well-built workstations are always as expensive as you assmue, either. A small lightweight network computer type thing is probably what most people would find most useful, and such a machine could easily be sub-$100. Think of things like the Sun Ray 1. It can be well-built and still be cheap. Sure, the performance isn't what today's monster peecees have, but the performance per dollar for things that matter most to most people would be unparalleled. Instead of laying out extra cash for the lastest 5000GB IDE disk, why not put in a small, fast 2GB SCA SCSI disk? Easy to manage, easy to replace without removing the cover, fast, and inexpensive. Most peecee buyers have no need of today's giant disks anyway. Similar arguments can be made about memory, processor, and so on. Instead of investing R&D money into finding ways to increase raw numbers (MHz, MB, and GB - the "big three" for naive peecee buyers), why not find ways to make systems that are simple, lightweight, long-lasting, and manageable; a pleasure to work with and own? A well-designed system with a small, low-power hybrid processor that provides decent performance, a high-speed built-in graphics subsystem, a small, high-speed disk, and built-in networking, with simple, universal connectors. As the label says, "no user-serviceable parts inside" - everything could be swapped out externally through plugin modules, much like the SGI O2 uses.
Now tell me who's going to make that - Intel? Dell? IBM? Maybe, but I doubt it. Certainly not any of the clone makers. But it's easy to envision Sun, SGI, or even - yes, even Apple making a go of this. Nobody's done it seriously yet - the Sun Ray 1, the Netwinder, and some other efforts are starts but not really serious - but someday it WILL happen. But a necessary prerequisite is the end of the peecee DSWs over MHz, MB, and GB and a recongition of how most people really use their computers. Getting back to the original discussion, I don't think cheap junk like [serial] ATA has any place in this sort of environment. The idea is to trade some performance for quality and manageability. Serial ATA is tangent to this concept.
I'm not hosting E-Bay or something on my box
Nor am I. Graphics work, development, and network services.
Considering that the majority of people use their computers for tasks more like this than whatever you're doing, I'd say making blanket statements about the non-usefullness of a platform is a common characteristic of trolls and idiots.
See above description of what most people would probably prefer if they knew enough to realize it. I don't pretend that one solution is right for everyone, but the current one doesn't really seem right for anyone. The original and current designers of the peecee and the hardware in it hate them as much as I do. Why do you think Intel is working on so much "legacy-free" technology? Because the very people who got us here today recognize their mistakes and are trying to correct them. The peecee's time has come and gone.
even something like WinCE-stuff or a TI-82, can be useful.
Anything wince does, something else can do as well or better, at a lower cost. That makes it useless. TI-82s probably fall into the same category, but that's debatable. I've not kept up on TI's graphics calculator line. If you have one, it's useful. If you don't, it's probably not a good time to buy one. There's a fine line between obsolete and useless; most people lump them together which is wrong. Brand-new technology can be useless, and obsolete technology can be useful. The criteria depends on application, as always. But the peecee is both obsolete and useless. It's not a reflection of what today's technology can do for people.
It's your own damn fault. If you buy or deal with people who buy crap or don't know how the hell to put boxes together correctly - then suffer in your own self-made hell of shoddy boxes and knotted cables.
I don't. I keep hoping others will see the light and, as I said, that there will still be people around who know how to make things right.
I'm assuming you're a jackboot-licking Mac user
Heh. Not. Even. Close. I've used a mac exactly once - in order to install ppclinux on a performa 6400, I and three others spent two hours trying to take the thing apart and replace the failed disk in it, then another four hours trying to get bootx onto a machine with no network, (un)stuffit, or (de-)binhexer. I came away from that experience convinced that the only thing worse than a peecee is a mac, and the only os worse than windoze 3.0 is the macos, choose any version.
Glad you're willing to submit to exorbant[sic] prices and a sub-standard OS so you can avoid the chance of putting your box together.
Well, the box I'm posting from was somewhat expensive compared with peecees, but my OS certainly isn't substandard. I know of at least three reasonable (read: unix) ones that will run on it; it's linux currently. I don't fear opening the box. In fact, this one's so damn beautiful inside I'd run it without the cover except that, well, the cover is part of the cooling system (yes, directed airflow - won't find *that* in peecees). What I fear is what's inside most boxes sold today. Ugh.
Fscking trolling Mac lusers
I hate macs. I'm anything but a luser and this is not a troll.
Go intellectually wank on Mackido
I honestly have no idea what this is.
post something relevant like "Increases in ATA speed don't matter as long as the inherent flaws in the design exist - SCSI's ability to use multiple devices simultaneously properly is more important."
Increases in ATA speed don't matter as long as the inherent flaws in the design exist - SCSI's ability to use multiple devices simultaneously properly is more important.
Gee, that added a lot to this discussion, no? Of course, this might be a good opportunity to mention that Apple (whom I supposedly dearly love) sold out and started using ATA in their own, previously all-scsi, systems some time ago, which further worsened their overall quality. The last decent machines they made were in the LC III timeframe, which are disturbingly similar inside to sparcstation 1s, another fine machine of that era. Both all-scsi of course, though I have trouble calling Apple's implementation "SCSI." Or the 53C90 in the SS1, for that matter. (shudder) Bad SCSI implementations. Guess the goats weren't sad to see those go, eh?
I love it when people make wildly wrong assumptions about me and thusly make themselves look like total fools when attempting to flame. Of course, this seems a common characteristic of anyone who thinks peecees are useful for any purpose other than keeping oceangoing vessels in one place.
--TM, wondering which lose more, peecees or the people who defend them
Yet another low end "solution" for low end people. You know, the world would be a lot simpler if we all just ditched [serial] ATA, USB, and so on. It's really a simple thing: SCSI-SCA for drive-type devices, 1394 for everything else. Nice easy connectors, high-speed, high-quality interfaces, and some actual semblance of design. Show of hands, now, who thinks having twenty different standards with twenty different cables and connectors makes sense? Who likes installing drives where you get to horribly twist ribbon cables and jam them into connectors whose pins have a breaking force of 3 pN while cutting your hands to shreds on peecee metal? Who likes getting shitty performance out of it after all that hassle and mess?
Nobody? Wow, what a surprise. Congratulations, you've taken the first step to peecee recovery: admitting that peecees aren't really what you want. Now, in that context, is serial ata really what you want either? No? Baby steps, kid, baby steps. You'll get there eventually. Hopefully the companies that know how to make what you really want will still be around by the time you realize what that is.
Fscking peecee lusers and their super-duper "standards" from their savior Intel. I'm going to go retch now. You think about what I've said.
Really? I do. I think it was funny and probably amusing for all. Nobody was harmed. No business was lost. No data was destroyed. No accounts were compromised. No systems were 0wn3d. In fact, for all practical purposes nothing happened at all. An irc'r logged in with a nick of his choosing and the system let him use it. What's the big deal here? It's not like this guy broke into billyboy's computer and tried to electronically sign pending legislation. Jeez, people, lighten up.
Slashdot Mirror
You may have a broken cpu or mobo. I actually have this behaviour on one of my systems. Put one cpu in. Works fine. Put second cpu in. Breaks. Remove first cpu. Still broken. Switch second cpu to known-good first slot. Still broken. Thus we conclude that the second cpu is bad. Since cpus these days have cache on board, it's entirely possible that the cpu itself is fine but the cache is bad. This seems reasonable, since cache is still memory and bad memory is the known cause of random sig11s. HTH.
Well, whether anyone cares is another matter, but it almost certainly will be an option. Linux already runs on everything up to an E6500 (known) and theoretically supports the E10k (nobdy's ever gotten hardware to test on though). I would certainly consider running Linux on US3-based systems. Up to about 8 cpus it beats the shit out of solaris. Past there people assume it would lose, but it's hard to know for sure as the most we've ever seen tested is 14. (shrug) Maybe nobody talks about running linux on these things, but certainly it's possible.
There's nothing special about 2TB support. First, you can already do that on UltraSPARC, and second, Alpha is a fairly obscure platform (right or wrong, it is). What would be real news is if
working for a company which has nearly completed the process of dumping Sun in favor of FreeBSD and Linux solutions[...]run an operating system which has no compiler included
Linux runs great on Sun systems. You imply that Sun and Linux are mutually exclusive; they are not. JMHO of course, but I find that Linux on UltraSPARC is far superior to Linux on peecees. If you like Linux on peecees, you'll like it on Sun hardware too, and in that realm the hardware isn't nearly as dodgy as peecees. If your budget makes Sun hardware impossible, fine. But don't imply that Linux on peecees can touch Linux on Suns. It can't even come close, and the hardware is clearly the limiting factor. It may meet your needs, and that's fine, but there's no reason to imply that a good, complete, low-cost OS and nice hardware are mutually exclusive. It sounds to me like you should have used your existing hardware and simply switched operating systems if the OS was giving you trouble. Oh well, more used Sun equipment for me to buy cheap.
Actually both SGI and sun pick kernels for a particular architecture (for example, IP27 or UltraSPARC II). They aren't actually identical. Sure, it comes from the same codebase and in most cases it's possible to run identical kernels on different machines, but usually it isn't done that way. The best example I've seen of actually doing it that way is UltraLinux which can do all sun4c, 4d, and 4m including SMP with a single kernel (Solaris _has to_ use separate kernels for 4c - not that anyone cares since 4c is essentially useless now). Even they can't do 4m and 4u from the same kernel though, nor can SunOS. :)
Yep. We have those too, and they are complete shit. As people say, "the only thing Sun about an Ultra 5 is the price." I'm sure the Sun technical people are sick of being hated over the U5/10. If you look at it, the decision to develop that kind of system could only have come from marketing. There's no way anyone at Sun really believes an U5 is worth having. If you want to judge Sun's workstations, get an Ultra 2 or an Ultra 80. These machines typify Sun's capabilities. Expensive but worth it.
Will SparcStations be able to survive the onslaught?
Well, the most recent machine to carry that name is obsolete, and only the fastest versions of it are still useful. So I'd say no. :) If you mean "will SPARC survive" then the question is more difficult. There will always be a market for something that isn't Intel (ie doesn't carry the baggage of the 4004 along with it). Whether that will be SPARC I don't know. Your question about McKinley destroying its competition is likewise unanswerable. Intel is betting a lot on what is really unproven technology while the traditional RISC makers are improving their technology one step at a time. By the time Intel finally ships their sooper-dooper new processors they may be well behind the "older" technology of other vendors like Decompaq and TI/Sun. If so, SGI would be foolish to ditch MIPS. It'll be interesting to see how this plays out.
Yes indeed. However, that's not the whole story. SGI's high-end systems (and the E10k for that matter) blur the line between cluster and single machine. As a low-end example from the SGI world, you can attach two Origin 200s using the Craylink and to the OS it appears as one machine. Is that a cluster, or one system? Well... Similarly, the Craylink technology is used with the O2k series to interlink systems as well as build composite systems. There even exist routers, hubs, and so forth. So it's really just an exceedingly fast networking technology, but with some clever hardware and OS support can also be used to make multiple systems tied together look like one. Very clever, very nice, and very fast. Expect to see more like this.
Yep. Sorry I sort of left the BSDs out of the discussion, mostly because Linux emulation is available. Certainly it would be nice to see native ELF BSD versions as well however (as well as versions for xBSD/y, where x != Free and/or y != Intel). Especially since there's absolutely no reason it can't be done. Just one flag is all it takes... That's what burns me about the whole thing. It takes far more effort to keep obsolete systems around for the purpose of doing builds than it does to just use up-to-date systems and support the software people are actually using. Your annoyance is justified. :)
How much more ground is Netscape willing to give up here? Will NS5 work on modern platforms? Or will Microsoft take the opportunity to crush Netscape for good, before Mozilla has any chance to make an impact?
I know Mozilla won't die off completely, a nice benefit of Open Source and all that, but there are already pages that will only work in IE (just as there were mages that Navigator but not Mosaic could display, and this eventually killed off mosaic). How long until it's a choice between security fixes and a working web browser? I know which I'll be forced to choose if it comes to that, and the thought of losing like that is unpleasant.
Well, I use the latest one from SGI with their own customizations. Neither it nor the standard Netscape-supported version seems to be immune to the typical crashes and hangs. On Linux, AKAIK RedHat just sets it up to use the libraries it was linked against if they're available. See my other post on Netscape not supporting current platforms properly/at all.
2) Turn off Cascading Style Sheets (Style Sheets) support in your preferences. It generally doesn't work well at all and really isn't all that necessary. And IME it makes NS crash. A lot.
Do not do this. CSS is a good idea. The separation of content from layout is both necessary and good. Now, it shouldn't be harmful to your ability to glean content from sites while browsing without CSS; however, without CSS it's assumed that you don't care what the pages look like and thus you might as well just use Lynx. CSS support in Mozilla actually looks good. Somebody there read the standard, which I can't really say about Netscape. Anyway, turning off CSS just encourages webmasters to use other methods, like crufty HTML or "that other browser's" proprietary extensions. Much like the lack of CSS in the first place caused Netscape's extensions to become de facto, and later, real, standards. Bad, bad. If CSS causes crashes, file bug reports or complain to the webmaster in question.
5) Don't invoke mystery components like Messenger and Composer and all that crap unless you actually need to use them. They tend to suck a bit.
Agreed. If you don't use them, consider Navigator instead of Communicator. Quicker to download as well.
The bottom line is, whatever you do, Netscape is still a crufty, bloated piece of crap on any platform they bother to support. Part of the problem is probably library and kernel version mismatches; since they refuse to support the current versions of most platforms, it's no real surprise that random crashes happen. Fixing this would go a long way toward improving stability. Aside from that, most problems seem related to either Java or Javascript. I agree with this poster; disable them if you can.
As if that weren't sufficiently annoying, there's not been a Linux/SPARC version since 4.51, but there are five Solaris versions. Of course, none of those are for current versions of Solaris either. All this, of course, would be excusable if they didn't have any systems to build on or they weren't going to support oddball platforms (I will admit that Linux/SPARC is marginal, though it has multiple supported distributions and a growing user base), but then why oh why are there two SINIX builds??? Most people don't even know WTF SINIX _is_, much less have it.
The point? It doesn't seem like Netscape understands what people are actually using today. If the objective is to be compatible with as much as possible, then not having up-to-date Linux builds for at least Intel, SPARC, and Alpha makes no sense. For that matter, they don't even claim to support Solaris 7 or 8. 2.5.1 is itself archaic. If the objective is instead to support only the most popular platforms, then I certainly don't see the need for five Solaris builds, two AIX builds, anything related to SINIX, or a Linux 2.0/libc5 build. The obvious platforms to support would be Solaris 7, Linux 2.2-intel/glibc, the latest AIX, the latest Digital Unix, and IRIX 6.5.
So what can we do to increase their awareness of this problem? Numerous polite letters have garnered either no response or a polite "get lost" form letter. Ideas?
Alternate paranoid theory: AOL wants all the Unix builds to be against old systems so that people will switch to windoze and buy more aol service. Pretty paranoid, but aol is pure evil after all.
PS: Kudos to the mozilla team for recognizing the value of compatibility and multi-platform support. The Linux/SPARC build works as well as any other.
--TM, still using 4.51 on Ultralinux, the preferred platform of all major deities
No, this is Raster we're talking about. It'd have to be "nUBMER FRUO"
As for my projects, I make sure to test with as many different environments/compilers as possible. It's good practice, ensures portability, and helps find bugs that might otherwise have been missed. I'm not surprised to find that not everyone does this, but that doesn't excuse them. If these "Open Source advocates" want to talk about standards, then they need to follow them, and ANSI C is one of the most important. To do less is hypocrisy of the worst kind. Quite frankly, I would prefer that Linux could be compiled with any old compiler, but I understand the reasons it can't be. Kernels and their native compilers are just too closely tied. Other vendors do it as well. It's not ideal but it's workable.
Naturally I don't object to Borland doing this, but I wonder whether you're aware of efforts like the cygwin project that have not only given poeple that same opportunity, but have given the opportunity to learn C or C++ in a portable environment. Using gcc under dos (if that's your cup of tea) makes the inevitable transition to Unix much easier. As an added bonus, gcc is moderately Free Software. As I said, I don't have a problem with Borland giving away their tools, under whatever license, but it's important to note that they aren't giving anybody something they couldn't already get for nothing.
Indeed. While I doubt anyone would waste the effort to do this, I hope Corel does. That will further discourage people from using their awful distribution (I never thought anyone could make Red Hat look good; then Corel came along). With any luck it'll also discourage further clueless luser Linux adoption in general.
--TM, spiteful, maybe, but not paranoid
If you release something which only compiles with one compiler, I don't expect it to be much use anyway. There are rare exceptions (kernels come to mind) but for the most part I want programs that will compile using any standards-compliant compiler. Strange as it seems, most Free Software fits that definition; I've compiled a great deal of such software with SGI's MipsPro compiler. It works just fine, thanks.
No, but I expect it won't be long until such offers as this Borland C++ compiler will be available for Linux (yes, probably without source). And I'll consider that day an important point in the development of Linux, more so than KDE or Gnome.
Good for you. I consider an ant taking a shit more important than kde or gnome however.
Well, if you don't consider having a choice of compilers as important to anybody, I'm afraid you haven't grasped what the little word "open" is all about.
I don't care one way or the other. I want a compiler that works and complies with standards. I have one. If you feel it's necessary to have another, have at it. That was my whole point.
Yes, that's the response I expect when posting anything else but "everything's great" about Linux on Slashdot.
This has nothing to do with Linux. It has everything to do with having a choice of compilers. I believe that choice is already sufficiently available, and that the existing compilers are adequate. You obviously don't. Thus the onus is on you to correct that perceived problem rather than griping about it. Start a project. Announce it here and elsewhere. If others share your views, the project will be successful.
Quite frankly, I'm not even in the "Linux is great" camp. It's an OS. It works for me. It's not perfect and neither is gcc (the real subject of discussion here). Both are, in my opinion, sufficiently better than anything else out there, for my hardware, budget, and interests, that I choose to use them. You aren't required to make the same choices, and I really don't give a rat's ass whether you do or not. My point is that if you believe the existing products are inadequate for any reason, it is your responsibility to either correct the shortcomings, live with them, or change to a platform that offers choices you believe to be acceptable.
--TM
Do you really expect someone else, who finds gcc works just fine for his/her needs, to sit down and write a compiler for you in the name of open computing just so you can have additional compiler choices?
There are two types of projects in the Free Software world: those that are genuinely important to somebody, and those that never happen. It seems that Slashdot attracts a lot of posts about the latter type, probably because it's easier to post "I want a ..." than it is to post "Here's my patch to ..." Wankers.
--TM, thinking it's time to fork the community and let the lusers rot
Slash CAN be beaten. Post anonymously to your heart's content, then moderate away. Then don't post anymore, even anonymously. This has the advantage of being able to moderate your own (anonymous) posts up, though there are usually better things to do and that's rather unethical anyway.
You aren't missing anything here. In fact, just to let them know how the world feels about their "tools" (appropriate, no?), don't download it. Prove that they can't move that crap out the door even if it's "free."
goddamned fscking borland compiler...ugh!
--TM, having unpleasant flashbacks
--TM, ucbvax!unrvax!donner!theman
As background, we had this machine dumped in our lab by a guy who walked out the door. I certainly didn't own it or anything, and we had no other macs at all, so knowledge of them wouldn't have been easily found. It was something of a "hmm, what can we do with this?" project. The performa 6400 mightn't be representative, but it turned my disinterest in Apple into active dislike. Fair or not, that's how it worked. Sound a bit better than "I hate macs"?
How many six-packs of malt liquor did you and your friends go through during your traumatic 6 hour ordeal?
Clearly not enough.
Look, I've been around for quite a while. I've seen, worked with, and used peecees from every major manufacturer, including clones, homebrews, and "other" ranging from $400 cheapo specials to $5000+ "servers." I've yet to see one such machine built in a reasonable, well-thought-out fashion. Some of the $10k [c|*|*|*|*|q] servers come close, but even those are really just overgrown peecees, and for that kind of price you can do better. I could buy all custom components and put together a nice system, but it would cost more than just buying a workstation from a decent manufacturer. Do peecees have to be built poorly? No, but for some reason all of them seem to be.
We're all supposed to pay huge prices for expensive non-i386 computers like you? You remind me of someone who has an expensive sportscar and bitches about all the econoboxes out there without realizing maybe others can't quite spend so much. Get off your high horse and realize some people make decisions based on performance per dollar.
I'm getting excellent performance per dollar. The important thing is in how you measure performance. I will agree that modern x86 processors can't be beat if you measure performance as "integer instructions executed from L1 cache per second." But I can funnel far, far more bits around with my workstation than any peecee can, and in fact, this machine probably has more disk-to-ethernet performance per dollar than anything that comes in a beige box ever has. As with most things, it's all in how you measure. I don't think well-built workstations are always as expensive as you assmue, either. A small lightweight network computer type thing is probably what most people would find most useful, and such a machine could easily be sub-$100. Think of things like the Sun Ray 1. It can be well-built and still be cheap. Sure, the performance isn't what today's monster peecees have, but the performance per dollar for things that matter most to most people would be unparalleled. Instead of laying out extra cash for the lastest 5000GB IDE disk, why not put in a small, fast 2GB SCA SCSI disk? Easy to manage, easy to replace without removing the cover, fast, and inexpensive. Most peecee buyers have no need of today's giant disks anyway. Similar arguments can be made about memory, processor, and so on. Instead of investing R&D money into finding ways to increase raw numbers (MHz, MB, and GB - the "big three" for naive peecee buyers), why not find ways to make systems that are simple, lightweight, long-lasting, and manageable; a pleasure to work with and own? A well-designed system with a small, low-power hybrid processor that provides decent performance, a high-speed built-in graphics subsystem, a small, high-speed disk, and built-in networking, with simple, universal connectors. As the label says, "no user-serviceable parts inside" - everything could be swapped out externally through plugin modules, much like the SGI O2 uses.
Now tell me who's going to make that - Intel? Dell? IBM? Maybe, but I doubt it. Certainly not any of the clone makers. But it's easy to envision Sun, SGI, or even - yes, even Apple making a go of this. Nobody's done it seriously yet - the Sun Ray 1, the Netwinder, and some other efforts are starts but not really serious - but someday it WILL happen. But a necessary prerequisite is the end of the peecee DSWs over MHz, MB, and GB and a recongition of how most people really use their computers. Getting back to the original discussion, I don't think cheap junk like [serial] ATA has any place in this sort of environment. The idea is to trade some performance for quality and manageability. Serial ATA is tangent to this concept.
I'm not hosting E-Bay or something on my box
Nor am I. Graphics work, development, and network services.
Considering that the majority of people use their computers for tasks more like this than whatever you're doing, I'd say making blanket statements about the non-usefullness of a platform is a common characteristic of trolls and idiots.
See above description of what most people would probably prefer if they knew enough to realize it. I don't pretend that one solution is right for everyone, but the current one doesn't really seem right for anyone. The original and current designers of the peecee and the hardware in it hate them as much as I do. Why do you think Intel is working on so much "legacy-free" technology? Because the very people who got us here today recognize their mistakes and are trying to correct them. The peecee's time has come and gone.
even something like WinCE-stuff or a TI-82, can be useful.
Anything wince does, something else can do as well or better, at a lower cost. That makes it useless. TI-82s probably fall into the same category, but that's debatable. I've not kept up on TI's graphics calculator line. If you have one, it's useful. If you don't, it's probably not a good time to buy one. There's a fine line between obsolete and useless; most people lump them together which is wrong. Brand-new technology can be useless, and obsolete technology can be useful. The criteria depends on application, as always. But the peecee is both obsolete and useless. It's not a reflection of what today's technology can do for people.
I don't. I keep hoping others will see the light and, as I said, that there will still be people around who know how to make things right.
I'm assuming you're a jackboot-licking Mac user
Heh. Not. Even. Close. I've used a mac exactly once - in order to install ppclinux on a performa 6400, I and three others spent two hours trying to take the thing apart and replace the failed disk in it, then another four hours trying to get bootx onto a machine with no network, (un)stuffit, or (de-)binhexer. I came away from that experience convinced that the only thing worse than a peecee is a mac, and the only os worse than windoze 3.0 is the macos, choose any version.
Glad you're willing to submit to exorbant[sic] prices and a sub-standard OS so you can avoid the chance of putting your box together.
Well, the box I'm posting from was somewhat expensive compared with peecees, but my OS certainly isn't substandard. I know of at least three reasonable (read: unix) ones that will run on it; it's linux currently. I don't fear opening the box. In fact, this one's so damn beautiful inside I'd run it without the cover except that, well, the cover is part of the cooling system (yes, directed airflow - won't find *that* in peecees). What I fear is what's inside most boxes sold today. Ugh.
Fscking trolling Mac lusers
I hate macs. I'm anything but a luser and this is not a troll.
Go intellectually wank on Mackido
I honestly have no idea what this is.
post something relevant like "Increases in ATA speed don't matter as long as the inherent flaws in the design exist - SCSI's ability to use multiple devices simultaneously properly is more important."
Increases in ATA speed don't matter as long as the inherent flaws in the design exist - SCSI's ability to use multiple devices simultaneously properly is more important.
Gee, that added a lot to this discussion, no? Of course, this might be a good opportunity to mention that Apple (whom I supposedly dearly love) sold out and started using ATA in their own, previously all-scsi, systems some time ago, which further worsened their overall quality. The last decent machines they made were in the LC III timeframe, which are disturbingly similar inside to sparcstation 1s, another fine machine of that era. Both all-scsi of course, though I have trouble calling Apple's implementation "SCSI." Or the 53C90 in the SS1, for that matter. (shudder) Bad SCSI implementations. Guess the goats weren't sad to see those go, eh?
I love it when people make wildly wrong assumptions about me and thusly make themselves look like total fools when attempting to flame. Of course, this seems a common characteristic of anyone who thinks peecees are useful for any purpose other than keeping oceangoing vessels in one place.
--TM, wondering which lose more, peecees or the people who defend them
Nobody? Wow, what a surprise. Congratulations, you've taken the first step to peecee recovery: admitting that peecees aren't really what you want. Now, in that context, is serial ata really what you want either? No? Baby steps, kid, baby steps. You'll get there eventually. Hopefully the companies that know how to make what you really want will still be around by the time you realize what that is.
Fscking peecee lusers and their super-duper "standards" from their savior Intel. I'm going to go retch now. You think about what I've said.
Really? I do. I think it was funny and probably amusing for all. Nobody was harmed. No business was lost. No data was destroyed. No accounts were compromised. No systems were 0wn3d. In fact, for all practical purposes nothing happened at all. An irc'r logged in with a nick of his choosing and the system let him use it. What's the big deal here? It's not like this guy broke into billyboy's computer and tried to electronically sign pending legislation. Jeez, people, lighten up.