> Why oh why are they testing IA32 code on the Itanium? That is hardly likely to show the performance of the processor in a good light.
Because they didn't have the option to. This isn't an Intel sponsored benchmark as they actually "borrowed" the machine for a while and did their stuff.
Installing Linux would be the best idea because they could've compiled benchmark apps but they were afraid to do so. They didn't want to alter the benchmark machine much and probably had a strict deadline.
>> Early benchmarks with IA-64 code on Itanium systems have shown that the new architecture is certainly capable of blowing the P4 out of the water with half the clockspeed.
> Uh, shouldn't it be at least capable of blowing the P4 out of the water at a quarter of the clockspeed?
It should be neither, really. Some 32 bit instructions can be executed 2 at a time with a 64 bit register but some can't. You also have the overhead of condensing 2 32-bit instructions into one 64-bit and vice-versa. If execution were absolutely linear, IA-64 wouldn't even be twice as fast as IA-32. It's not, so results will vary.
We should compare running 64-bit math on an Itanium and 64-bit math on a P4. Running legacy code is as illogical as running 16-bit code on a Pentium so I also question Intel's hardware emulation decision.
Flavio
Re:Good Fnarg! that article is so full of shit.
on
2.2 vs 2.4
·
· Score: 1
How clueless can this guy be ? If someone went to such great lengths to defeat the 2gb limit, then I'm pretty sure it's because it's been a problem for a while. Uncompressed video comes to mind, where a reasonable clip bucket can contain well over 10gb of data. Databases also get pretty huge when you start collection data from the web (search engines perhaps, or DoubleClick stats). Next.
Yeah, tell me about it.
When I capture video at 640x480 I stick to MPEG1 at 3.5MBPS. Believe me, your hard drive fills up FAST.
Flavio
Re:A Primer on History from an Erudite Source
on
100 Years of Radio
·
· Score: 1
Amen, brother.
It's great to see someone who also enjoys the music of Tesla and respects the great man who gives the band its name.
The basic utility of neural networks is basically the same as "genetic programming".. they avoid the necessity of having to describe a solution to a complex (or merely subtle) problem.
I know, but I believe the hassles one goes through in the learning process aren't worth the final result, specially in a very deterministic environment like a house. As another reader pointed out, you wouldn't like the net to deal with events like waking up at 4 am for some random reason, a fact that isn't likely to repeat itself.
Neural nets usually work well in areas where conventional binary logic fails. Text, image and speech recognition are examples.
A neural net in your home would, in my opinion, be too complex to bother with. Neural nets are self programmed through experience to perform choices that aren't easy to define in computer code.
For example, some time ago there was this slasdot post about a program which supposedly recognizes "innapropriate" images (i.e., porn). It can be used in web proxies, for example. The program didn't perform very well (in my opinion, it failed miserably, as the task is extremely difficult).
A very complex neural net could theoretically learn from experience to recognize the difference between a naked baby and two people having sex. You can't program such a net by hand because there are so many neurons involved and influencing other neurons in ways so complex you can't even imagine.
Now why would you want to use that to switch/dim lights in a house? The net could learn your behavior through experience at first, for example. It could have this "learn mode", much like smart network switches do. The net itself would be useless at this point, and it would record what you do at particular times of the day and perhaps make notes on your moods and act accordingly. After this learning period it would do things on itself and be corrected if the choice was incorrect. But hey, I really don't think you need the black magic that neurons give you to notice that.
The bottom line is that people are usually very predictable. Neural nets are great when nuances from an objective POV are fundamental (like position/density of beige colors in an image as an indicator of porn) and this doesn't seem to be the case.
With everyone hacking their Dreamcasts to run BSD, and noone buying games, and Sega losing money on each piece of hardware sold, won't this drive Sega into further money problems? They already lost something like $300 million this past year (and this was a good year for them).
1. It's not "everyone". Let's be generous and state that's 0.05% of all Dreamcast owners.
2. That's Sega's problem.
Re:Will GTK become Yet Another X?
on
GTK+ without X!
·
· Score: 2
The X architecture is great in theory for remote administration, but you can't, for example, display a skinned gtk window from an app that's running remotely over a 10baseT network with acceptable performance. It should be the server that does all the drawing of widgets, the client should just run the main functionality of the app and send commands to the server to draw a widget, instead of creating a bitmap of a widget and sending it over the network.
Notice you're talking 10baseT. That's 1.2 MB/s. I haven't run skinned GTK apps, but I occasionally run adobe acrobat and netscape navigator without any issues (it's a 100baseT, but network usage never goes as high as 1MB/s).
And in any case, you can use compression on the link (either with SSH or with LBX (lbx is included with practically all distros out there)).
I partially agree with you that the client should draw the stuff or at least leave that as an option.
The problems are that:
1. That would make things much more complicated than they already are. Each lib would have proprietary commands for each set of widgets.
2. That forces the client to have the widget drawing library of the exact same version. That may strike you as simple, but it'd be a headache to sysadmins out there. Just imagine not being able to run some program because you don't have its library. Or worse, the library's proprietary AND commercial.
3. The client would use more CPU and RAM than without normal bitmap drawing. This isn't much of an issue today, but at the time X was designed this was absolutely crucial.
Has anyone here compared the performance and/or functionality of these two distros? I just installed Yellow Dog 1.2.1 on a G3 laptop the other day, and was sorely disappointed...
the weird thing is that my experience was the opposite. My burn of LinuxPPC2000 wouldn't install on my iMac. The RPMS were in the *wrong directory* in the CD, and I couldn't use the graphical installation because at the time my iMac (classic) had only 32 MB of ram and the installation obviously doesn't init any swap.
After copying the CD's contents to another machine that's networked with the iMac I actually setup an FTP install and that also didn't work because some packages had different names than the packages the install program searched for. I ended up renaming about 30-40 packages until I realised about 40% of the RPMs had wrong names. By this time I gave up on the install and installed YellowDog. I now use lots of packages from LinuxPPC, but to this day I haven't figured out what's wrong with that CD. This coming from an experienced Linux user.
Anywho, to your questions:
* Would lock or power-down on me if I booted straight "linux"; could only sucessfully do things booting to "linux-novideo". This may be a somewhat known issue, but it keeps me from using 256 colours.
I also had several problems in the yaboot install process. This was the first time I did a LinuxPPC install so I had to learn some stuff the hard way. The OpenFirmware setup went along fine, but some bugs like the
image = hd:,\\\\vmlinux
line in yaboot.conf (now documented in their support site; there was a typo in the official version) gave me some headaches =)
In any case, read the yaboot docs and you should be running in at most half an hour because you'll already know about the yaboot.conf bug.
* Gnome was crap on my machine. The task bar (gnome-session?) cored every time I tried to load up X. the mouse cursor would "float" as it approached max/min/etc buttons on the window.
I also had some issues with Gnome, but not as bad. Either recompile it or use KDE. I personally use Window Maker for the speed and memory usage so I didn't have the problem.
* KDE was better, but only somewhat. If screen blanked, palette never shifted back to normal colours.
Weird, KDE ran fine with me.
* Since there doesn't appear to be any real "text" mode, minicom had everything shoved into an 80x25 corner of my display, leaving 2 inches of blank black space to the bottom and right.
recompile the kernel with larger fonts.
* Could not alt-Fx between terminals. Could alt-F7 to the blank one reserved for X -- but could not get back to any other terminals.
it's actually command-Fx on the mac. If you did try command-Fx and it didn't work... well, you've got to enable the terms in/etc/inittab. Perhaps I did that myself and can't remember.
* Other little, nagging issues.
That's Linux for you. I always customize my installs, because anything I consider not to my liking is a nagging issue (like the lack of --color=yes in RedHat's 'ls' command).
I haven't been able to easily find information on most of these issues, either.
Google should simplify your life. Use lists.linuxppc.org as well.
I guess what I'm trying to say is, when Jason tells us "You can run on a PPC box as well as an x86 box", I just certainly hope this is true,
It is VERY true. My iMac now has 96 MB of RAM and it runs like a charm. I love using that box now (it's actually my Mom's -- she couldn't stand MacOS's sluggishness and crashes; I couldn't stand MacOS). It has _NEVER_ crashed on me, runs as fast as my 400 MHz Celeron, looks and feels just like Linux on an i386.
and it certainly won't be easy for joe user to find ways to fix these issues.
Agreed. Some of the stuff's well above joe user's head. In any case, I love LinuxPPC and I'd like to give a big thanks for Haas and all developers out there.
Flavio
Re:Will GTK become Yet Another X?
on
GTK+ without X!
·
· Score: 4
I'd love to see the architectural disaster that is X go
Why do people keep saying that X is an architectural disaster?
I think this is one more example of false ideas that somewhat became mainstream and keep being repeated in a drone fashion.
Of course X isn't the most optimized application around, but I seriously don't see how it's an "architectural disaster". Most people don't remember that the whole point of X is to run applications in a network environment, even if they're used locally.
It's fine with me if you don't use even half the functions X has to offer you, but don't call it a disaster because of that.
Looking Glass Studios' games weren't even mentioned.
These people mentioned Half-Life, a 1998 game but didn't mention System Shock I, published much earlier (1994?), with similar graphics (even though it lacked OpenGL/Direct3D support and was a DOS game) and a MUCH better plot/gameplay/engine.
I agree that perhaps Counter-Strike should've been mentioned, as its counterpart, Rainbow 6, is pretty weak, but there is _no way_ that Half-Life is better than System Shock I.
And there's also Flight Unlimited, which defined the standard regarding photographic landscapes and non-military flight models.
I've actually seen something extremely similar to this on the Discovery Channel about 4 years ago. It was about this biker who pedaled across the US with a computer on the back "seat".
Solar cells (which I believe he could extend when we wasn't on the road for larger power generation) powered the computer and he actually typed while riding because he had a similar type of keyboard in each hand.
I've been wanting to see such a thing for a long time, and this could be the same company.
The rad-hard CPU of choice for spaceborne equipment is the 1802. Remember the RCA COSMAC personal computer of long ago? I think they have this in silicon-on-sapphire. There are a small number of satellite hackers who still practice 1802 assembler at this late date. It should suffice to say that nobody uses this CPU for anything else any longer. So, an experiment with a modern CPU was very desirable. It looks like the problem might not be in the CPU.
Well, unfortunately I don't remember the RCA COSMAC PC because I wasn't even alive back then:)
I do remember that this exact same CPU is one exotic baby, starting with its silicon-on-sapphire CMOS fabrication. It's in the heart of the Viking, Voyager and Galileo!
As I couldn't remember anything else, I had to consult the "Great Microprocessors of the past and present" page.
Here's the text for those who actually don't wanna visit the page:
RCA 1802, weirdness at its best (1974)
The RCA 1802 was an odd beast, extremely simple and fabricated in CMOS, which allowed it to run at 6.4 MHz (at 10V, but very fast for 1974) or suspended with the clock stopped. It was an 8 bit processor, with 16 bit addressing, but the major features were it's extreme simplicity, and the flexibility of it's large register set. Simplicity was the primary design goal, and in that sense it was one of the first RISC chips.
It had sixteen 16-bit registers, which could be accessed as thirty-two 8 bit registers, and an accumulator D used for arithmetic and memory access - memory to D, then D to registers, and vice versa, using one 16-bit register as an address. This led to one person describing the 1802 as having 32 bytes of RAM and 65535 I/O ports. A 4-bit control register P selected any one general register as the program counter, while control registers X and N selected registers for I/O Index, and the operand for current instruction. All instructions were 8 bits - a 4-bit op code (total of 16 operations) and 4-bit operand register stored in N.
There was no real conditional branching (there were conditional skips which could implement it, though), no subroutine support, and no actual stack, but clever use of the register set allowed these to be implemented - for example, changing P to another register allowed jump to a subroutine. Similarly, on an interrupt P and X were saved, then R1 and R2 were selected for P and X until an RTI restored them.
A later version, the 1805, was enhanced, adding several Forth language primitives. Forth was commonly used in control applications.
Apart from the COSMAC microcomputer kit, the 1802 saw action in some video games from RCA and Radio Shack, and the chip is the heart of the Voyager, Viking and Galileo (along with some AMD 29000 bit slice processors) probes. One reason for this is that a version of the 1802 used silicon on sapphire (SOS) technology, which leads to radiation and static resistance, ideal for space operation.
A backup flight computer, itself an experiment to see if the CPU would be radiation-hard enough to survive, has its RAM corrupted every 1 or 2 days in orbit as it crosses radiation belts and currently is not set up to reboot automaticaly.
I don't want to sound arrogant here, but couldn't they either:
1. Put a radiation hardened processor instead just like everybody does or...
2. Ask NASA, other space agency or people in university labs that research this stuff about proper radiation shielding for processors? [I understand this option may be more complicated than it sounds]
Similarly, you can't run Windows 98 on a PowerPC or Alpha. Does it matter?
Yup, because the only OS you can run on all that crap is Linux.
It remains to be seen if the Itanium is really where personal computing is headed. (...) Moving to an entirely new processor *family*, not just the next generation of what's currently available, is not to be taken lightly.
I believe the future is totally different, precisely because of technologies like Java, Amiga's VM,.NET, emulation and also because of Linux.
The best architecture for a single task will be chosen and software shall be ported for that. The choice won't be tied to software compatibility or portability issues because these will be minimal.
One can understand exactly what I mean when one runs Linux in different architectures in the same network. All these machines behave exactly alike and you could only tell them apart because of some specific compatibility issues that still exist.
The AmigaOS promises to remove architectural barriers that still exist, mainly regarding 3D graphics and high quality sound playback. One could have a DirectX-like API on every main architecture and porting software would be to compile it, providing there weren't any architecture-specific optimizations.
I don't think the Itanium is out of the league just because it doesn't follow the x86 hierarchy. That will probably be its biggest strength.
We should be discussing what the Itanium will actually be used for and how it will sell (oh, God, I feel like a marketing/business person) considering its powerful (and expensive) capabilities, but not compatibility issues.
Agreed. Since we're on the ATI subject, I've got an All In Wonder 128 card (the 16 MB version, not the Pro/32MB one).
OpenGL support does exist by default in XFree86 4.0, but it's quite slow. I get 50-60 fps with Windows on my Celeron 400, but only 35-40 fps in Linux. When explosions come in Quake 3 I actually see what I call the "webcam effect".
Video capture is out of the question, of course, since there's no software for it. There's a program called xatitv that works well, but it hasn't been updated in about 5 months, is quite beta and doesn't enable bilinear filtering on the TV image, giving you the 'line effect' when the image moves sideways. (Remember that PAL/NTSC are interlaced)
2D support is pretty good, but what card can't handle that nowadays?
In any case, stay away from ATI if you want to play 3D games in Linux.
[before someone comments on it: XFree 4.0.2 gave me a 5% speed increase in exchange for a very unstable server]
I don't think they were stupid either, just arrogant; in my opinion, their big problem was that they started believing their own marketing.
That, for some reason, looks exceedingly simple to be the complete answer.
I believe 3dfx claimed "we don't need 32-bit textures", "we don't need large textures/AGP" and "we don't need T&L" precisely because they were having trouble with new processor design. Hence, they had to stick to the old ones and you see the result.
I have never heard about 3dfx and design difficulties, of course. But I haven't heard anything denying that either, so it could very well be the reason.
IMHO is that they went for raw speed over features.
Definitely. They not only neglected 32 bit color like you said, but also neglected AGP in favor of PCI and started using double/quad processor/card configurations instead of actually developing ONE card with all these features built into one, non-redundant chipset.
Of course there are features like motion blur and the T-Buffer, but I believe they are (a) too novel for their time; (b) slow; (c) restricted to 3dfx cards, making designers careful when to implement them; (d) a late and improper solution to their problems...
But then why did this happen? They aren't stupid.
The XBox deal with NVIDIA was as lethal to them as EIDOS' investment in Daikatana was lethal to Looking Glass Studios. 3dfx and Looking Glass were already in trouble when these events happened; they were only the nail in the coffin.
3dfx probably had some serious design issues, kind of like what Intel had with RAMBUS. 3dfx just doesn't have the leverage that Intel does to stay in control.
Slashdot Mirror
> if you'll remember correctly, the pentium was released in 1992. windows 95 wasn't released until 1995.
Of course I remember.
In the real world, illogical and stupid things are done all the time.
Oh, crap. It looks like so.
I got to read the whole article without any issues.
> Why oh why are they testing IA32 code on the Itanium? That is hardly likely to show the performance of the processor in a good light.
Because they didn't have the option to. This isn't an Intel sponsored benchmark as they actually "borrowed" the machine for a while and did their stuff.
Installing Linux would be the best idea because they could've compiled benchmark apps but they were afraid to do so. They didn't want to alter the benchmark machine much and probably had a strict deadline.
Flavio
>> Early benchmarks with IA-64 code on Itanium systems have shown that the new architecture is certainly capable of blowing the P4 out of the water with half the clockspeed.
> Uh, shouldn't it be at least capable of blowing the P4 out of the water at a quarter of the clockspeed?
It should be neither, really. Some 32 bit instructions can be executed 2 at a time with a 64 bit register but some can't. You also have the overhead of condensing 2 32-bit instructions into one 64-bit and vice-versa. If execution were absolutely linear, IA-64 wouldn't even be twice as fast as IA-32. It's not, so results will vary.
We should compare running 64-bit math on an Itanium and 64-bit math on a P4. Running legacy code is as illogical as running 16-bit code on a Pentium so I also question Intel's hardware emulation decision.
Flavio
How clueless can this guy be ? If someone went to such great lengths to defeat the 2gb limit, then I'm pretty sure it's because it's been a problem for a while. Uncompressed video comes to mind, where a reasonable clip bucket can contain well over 10gb of data. Databases also get pretty huge when you start collection data from the web (search engines perhaps, or DoubleClick stats). Next.
Yeah, tell me about it.
When I capture video at 640x480 I stick to MPEG1 at 3.5MBPS. Believe me, your hard drive fills up FAST.
Flavio
Amen, brother.
It's great to see someone who also enjoys the music of Tesla and respects the great man who gives the band its name.
Flavio
_A note to the editor_
Holy shit, Taco.
In 3 years of Slashdot I've grudgingly tolerated your atrocious spelling, but you just don't change.
You're illiterate. You can't even spell.
Mathamatical? Damn, aren't you ashamed of this?
That, added to your regular trolling and clueless comments pisses me off.
I'm outta here.
Flavio
Wavetable MIDI card (something made by Orchid, I can't recall which model).
Gamewave 32? Soundwave 32?
Flavio
The basic utility of neural networks is basically the same as "genetic programming" .. they avoid the necessity of having to describe a solution to a complex (or merely subtle) problem.
I know, but I believe the hassles one goes through in the learning process aren't worth the final result, specially in a very deterministic environment like a house. As another reader pointed out, you wouldn't like the net to deal with events like waking up at 4 am for some random reason, a fact that isn't likely to repeat itself.
Flavio
Dude, that's sick.
Neural nets usually work well in areas where conventional binary logic fails. Text, image and speech recognition are examples.
A neural net in your home would, in my opinion, be too complex to bother with. Neural nets are self programmed through experience to perform choices that aren't easy to define in computer code.
For example, some time ago there was this slasdot post about a program which supposedly recognizes "innapropriate" images (i.e., porn). It can be used in web proxies, for example. The program didn't perform very well (in my opinion, it failed miserably, as the task is extremely difficult).
A very complex neural net could theoretically learn from experience to recognize the difference between a naked baby and two people having sex. You can't program such a net by hand because there are so many neurons involved and influencing other neurons in ways so complex you can't even imagine.
Now why would you want to use that to switch/dim lights in a house? The net could learn your behavior through experience at first, for example. It could have this "learn mode", much like smart network switches do. The net itself would be useless at this point, and it would record what you do at particular times of the day and perhaps make notes on your moods and act accordingly. After this learning period it would do things on itself and be corrected if the choice was incorrect. But hey, I really don't think you need the black magic that neurons give you to notice that.
The bottom line is that people are usually very predictable. Neural nets are great when nuances from an objective POV are fundamental (like position/density of beige colors in an image as an indicator of porn) and this doesn't seem to be the case.
Flavio
With everyone hacking their Dreamcasts to run BSD, and noone buying games, and Sega losing money on each piece of hardware sold, won't this drive Sega into further money problems? They already lost something like $300 million this past year (and this was a good year for them).
1. It's not "everyone". Let's be generous and state that's 0.05% of all Dreamcast owners.
2. That's Sega's problem.
The X architecture is great in theory for remote administration, but you can't, for example, display a skinned gtk window from an app that's running remotely over a 10baseT network with acceptable performance. It should be the server that does all the drawing of widgets, the client should just run the main functionality of the app and send commands to the server to draw a widget, instead of creating a bitmap of a widget and sending it over the network.
Notice you're talking 10baseT. That's 1.2 MB/s. I haven't run skinned GTK apps, but I occasionally run adobe acrobat and netscape navigator without any issues (it's a 100baseT, but network usage never goes as high as 1MB/s).
And in any case, you can use compression on the link (either with SSH or with LBX (lbx is included with practically all distros out there)).
I partially agree with you that the client should draw the stuff or at least leave that as an option.
The problems are that:
1. That would make things much more complicated than they already are. Each lib would have proprietary commands for each set of widgets.
2. That forces the client to have the widget drawing library of the exact same version. That may strike you as simple, but it'd be a headache to sysadmins out there. Just imagine not being able to run some program because you don't have its library. Or worse, the library's proprietary AND commercial.
3. The client would use more CPU and RAM than without normal bitmap drawing. This isn't much of an issue today, but at the time X was designed this was absolutely crucial.
Flavio
Has anyone here compared the performance and/or functionality of these two distros? I just installed Yellow Dog 1.2.1 on a G3 laptop the other day, and was sorely disappointed...
/etc/inittab. Perhaps I did that myself and can't remember.
the weird thing is that my experience was the opposite. My burn of LinuxPPC2000 wouldn't install on my iMac. The RPMS were in the *wrong directory* in the CD, and I couldn't use the graphical installation because at the time my iMac (classic) had only 32 MB of ram and the installation obviously doesn't init any swap.
After copying the CD's contents to another machine that's networked with the iMac I actually setup an FTP install and that also didn't work because some packages had different names than the packages the install program searched for. I ended up renaming about 30-40 packages until I realised about 40% of the RPMs had wrong names. By this time I gave up on the install and installed YellowDog. I now use lots of packages from LinuxPPC, but to this day I haven't figured out what's wrong with that CD. This coming from an experienced Linux user.
Anywho, to your questions:
* Would lock or power-down on me if I booted straight "linux"; could only sucessfully do things booting to "linux-novideo". This may be a somewhat known issue, but it keeps me from using 256 colours.
I also had several problems in the yaboot install process. This was the first time I did a LinuxPPC install so I had to learn some stuff the hard way. The OpenFirmware setup went along fine, but some bugs like the
image = hd:,\\\\vmlinux
line in yaboot.conf (now documented in their support site; there was a typo in the official version) gave me some headaches =)
In any case, read the yaboot docs and you should be running in at most half an hour because you'll already know about the yaboot.conf bug.
* Gnome was crap on my machine. The task bar (gnome-session?) cored every time I tried to load up X. the mouse cursor would "float" as it approached max/min/etc buttons on the window.
I also had some issues with Gnome, but not as bad. Either recompile it or use KDE. I personally use Window Maker for the speed and memory usage so I didn't have the problem.
* KDE was better, but only somewhat. If screen blanked, palette never shifted back to normal colours.
Weird, KDE ran fine with me.
* Since there doesn't appear to be any real "text" mode, minicom had everything shoved into an 80x25 corner of my display, leaving 2 inches of blank black space to the bottom and right.
recompile the kernel with larger fonts.
* Could not alt-Fx between terminals. Could alt-F7 to the blank one reserved for X -- but could not get back to any other terminals.
it's actually command-Fx on the mac. If you did try command-Fx and it didn't work... well, you've got to enable the terms in
* Other little, nagging issues.
That's Linux for you. I always customize my installs, because anything I consider not to my liking is a nagging issue (like the lack of --color=yes in RedHat's 'ls' command).
I haven't been able to easily find information on most of these issues, either.
Google should simplify your life. Use lists.linuxppc.org as well.
I guess what I'm trying to say is, when Jason tells us "You can run on a PPC box as well as an x86 box", I just certainly hope this is true,
It is VERY true. My iMac now has 96 MB of RAM and it runs like a charm. I love using that box now (it's actually my Mom's -- she couldn't stand MacOS's sluggishness and crashes; I couldn't stand MacOS). It has _NEVER_ crashed on me, runs as fast as my 400 MHz Celeron, looks and feels just like Linux on an i386.
and it certainly won't be easy for joe user to find ways to fix these issues.
Agreed. Some of the stuff's well above joe user's head. In any case, I love LinuxPPC and I'd like to give a big thanks for Haas and all developers out there.
Flavio
I'd love to see the architectural disaster that is X go
Why do people keep saying that X is an architectural disaster?
I think this is one more example of false ideas that somewhat became mainstream and keep being repeated in a drone fashion.
Of course X isn't the most optimized application around, but I seriously don't see how it's an "architectural disaster". Most people don't remember that the whole point of X is to run applications in a network environment, even if they're used locally.
It's fine with me if you don't use even half the functions X has to offer you, but don't call it a disaster because of that.
Flavio
Looking Glass Studios' games weren't even mentioned.
These people mentioned Half-Life, a 1998 game but didn't mention System Shock I, published much earlier (1994?), with similar graphics (even though it lacked OpenGL/Direct3D support and was a DOS game) and a MUCH better plot/gameplay/engine.
I agree that perhaps Counter-Strike should've been mentioned, as its counterpart, Rainbow 6, is pretty weak, but there is _no way_ that Half-Life is better than System Shock I.
And there's also Flight Unlimited, which defined the standard regarding photographic landscapes and non-military flight models.
And what about Thief?
I'm outraged.
Flavio
I've actually seen something extremely similar to this on the Discovery Channel about 4 years ago. It was about this biker who pedaled across the US with a computer on the back "seat".
Solar cells (which I believe he could extend when we wasn't on the road for larger power generation) powered the computer and he actually typed while riding because he had a similar type of keyboard in each hand.
I've been wanting to see such a thing for a long time, and this could be the same company.
Flavio
Argh. I probably made a typo and Slashdot ate my link. Here it goes:
Great Microprocessors of the Past and Present (V 12.0.0)
The RCA 1802's in there.
Flavio
Well, unfortunately I don't remember the RCA COSMAC PC because I wasn't even alive back then
I do remember that this exact same CPU is one exotic baby, starting with its silicon-on-sapphire CMOS fabrication. It's in the heart of the Viking, Voyager and Galileo!
As I couldn't remember anything else, I had to consult the "Great Microprocessors of the past and present" page.
Here's the text for those who actually don't wanna visit the page:
Thanks for the info, Bruce!
Flavio
A backup flight computer, itself an experiment to see if the CPU would be radiation-hard enough to survive, has its RAM corrupted every 1 or 2 days in orbit as it crosses radiation belts and currently is not set up to reboot automaticaly.
I don't want to sound arrogant here, but couldn't they either:
1. Put a radiation hardened processor instead just like everybody does or...
2. Ask NASA, other space agency or people in university labs that research this stuff about proper radiation shielding for processors? [I understand this option may be more complicated than it sounds]
In any case, I wish them luck.
Flavio
Similarly, you can't run Windows 98 on a PowerPC or Alpha. Does it matter?
.NET, emulation and also because of Linux.
Yup, because the only OS you can run on all that crap is Linux.
It remains to be seen if the Itanium is really where personal computing is headed. (...) Moving to an entirely new processor *family*, not just the next generation of what's currently available, is not to be taken lightly.
I believe the future is totally different, precisely because of technologies like Java, Amiga's VM,
The best architecture for a single task will be chosen and software shall be ported for that. The choice won't be tied to software compatibility or portability issues because these will be minimal.
One can understand exactly what I mean when one runs Linux in different architectures in the same network. All these machines behave exactly alike and you could only tell them apart because of some specific compatibility issues that still exist.
The AmigaOS promises to remove architectural barriers that still exist, mainly regarding 3D graphics and high quality sound playback. One could have a DirectX-like API on every main architecture and porting software would be to compile it, providing there weren't any architecture-specific optimizations.
I don't think the Itanium is out of the league just because it doesn't follow the x86 hierarchy. That will probably be its biggest strength.
We should be discussing what the Itanium will actually be used for and how it will sell (oh, God, I feel like a marketing/business person) considering its powerful (and expensive) capabilities, but not compatibility issues.
Flavio
Agreed. Since we're on the ATI subject, I've got an All In Wonder 128 card (the 16 MB version, not the Pro/32MB one).
OpenGL support does exist by default in XFree86 4.0, but it's quite slow. I get 50-60 fps with Windows on my Celeron 400, but only 35-40 fps in Linux. When explosions come in Quake 3 I actually see what I call the "webcam effect".
Video capture is out of the question, of course, since there's no software for it. There's a program called xatitv that works well, but it hasn't been updated in about 5 months, is quite beta and doesn't enable bilinear filtering on the TV image, giving you the 'line effect' when the image moves sideways. (Remember that PAL/NTSC are interlaced)
2D support is pretty good, but what card can't handle that nowadays?
In any case, stay away from ATI if you want to play 3D games in Linux.
[before someone comments on it: XFree 4.0.2 gave me a 5% speed increase in exchange for a very unstable server]
Flavio
Katz, I've seen you rephrase the exact same ideas at least 5 times.
Don't you think you've made your point?
Flavio
I don't think they were stupid either, just arrogant; in my opinion, their big problem was that they started believing their own marketing.
That, for some reason, looks exceedingly simple to be the complete answer.
I believe 3dfx claimed "we don't need 32-bit textures", "we don't need large textures/AGP" and "we don't need T&L" precisely because they were having trouble with new processor design. Hence, they had to stick to the old ones and you see the result.
I have never heard about 3dfx and design difficulties, of course. But I haven't heard anything denying that either, so it could very well be the reason.
Flavio
IMHO is that they went for raw speed over features.
Definitely. They not only neglected 32 bit color like you said, but also neglected AGP in favor of PCI and started using double/quad processor/card configurations instead of actually developing ONE card with all these features built into one, non-redundant chipset.
Of course there are features like motion blur and the T-Buffer, but I believe they are (a) too novel for their time; (b) slow; (c) restricted to 3dfx cards, making designers careful when to implement them; (d) a late and improper solution to their problems...
But then why did this happen? They aren't stupid.
The XBox deal with NVIDIA was as lethal to them as EIDOS' investment in Daikatana was lethal to Looking Glass Studios. 3dfx and Looking Glass were already in trouble when these events happened; they were only the nail in the coffin.
3dfx probably had some serious design issues, kind of like what Intel had with RAMBUS. 3dfx just doesn't have the leverage that Intel does to stay in control.
Flavio