One thing that doesn't seem to get addressed by ANY Linux 3D hardware review I've seen (One was also in LJ recently), is the issue of stability.
We run Linux because it's stable, even in cases where we know that we lose performance because of that stability.
Well, all of these reviews basically say, "If you can afford it, buy NVIDIA! They're fast!".
So what if your card is fast but your machine crashes in the middle of a game? NVIDIA's drivers just plain suck in the stability arena - My machine crashes on a regular basis in both Q3 and UT with those crap drivers. Even Windows looks rock-solid compared to Linux when using those accursed drivers! And all this because the only difference between the Quadro and GeForce is a PCI/AGP device ID and a few flags set in the driver because of said ID - it's the only plausible reason for being the only vendor not to release source/specs. (Check out http://www.geocities.com/tnaw_xtennis/)
Just to let you know - The K6 SMP support was OpenAPIC. Too bad no one made a mobo for it. Don't know if the Athlon is or a different protocol. (I think it may be different - Considering how different the bus/cache design is from the old Socket 7 designs in general.)
But unfortunately, the workaround made the K6 SMP protocol incompatible with that of the Pentium. And AMD wasn't big enough at the time for any chipset/mobo manufacturer to justify creating a new SMP mobo for a small market.
It's a known feature of the Athlon. In fact, one of the things AMD was proud of is that in SMP configurations, each CPU apparently gets its own frontside bus into the chipset IIRC.
Even the K6 (Maybe, definately the K6-2 and K6-3) supported SMP. Problem was, it was not the same SMP scheme as used by the classic Pentium, and there was never an SMP mobo made for it. I hope the same thing doesn't happen to the Athlon. (Although from what I remember, an SMP chipset was due relatively soon on AMD's chipset roadmap. As in by the end of the year, I think.)
Well, it seems that most of the people in this thread have had no problems whatsoever with test8 - I love it.
As to Microsoft products - Winblows won't recognize the second IDE channel on my new mobo (VIA KT133 chipset). Bye-bye DVD drive. The drivers that come with Winblows give a "Device is not present or is not working properly.", VIA's updated drivers crash the machine when I try to install them. Yeah, Windows is reeeeaaally stable. That's in addition to the standard BSODs.
Recent versions of gcc (2.95 and above, RH7 has 2.96) and kernels don't mix. It's a kernel bug that for some odd reason the kernel developers never accept the existence of. (The same went for egcs during the 2.0.x series...)
Because of this known issue, RH7 includes a "kgcc" package for compiling kernels. You will have to change the kernel makefile to force use of this compiler. (On my machine, 'export CC=kgcc' doesn't seem to do anything...)
While I don't know about HDs, USB audio devices and scanners are supported. (Your particular device may not be, but many are.) There is USB storage support in the kernel, but I think it's going to remain marked as experimental even in the final 2.4 release. (Of course, some of the "experimental" code in Linux works quite well...)
The kernel support for USB mice definately kicks ass. I finally moved my IntelliMouse Explorer from the PS/2 port to the USB port (Higher sampling rate with USB), and I'm 2-3 times as lethal with the sniper rifle in Q3Fortress and Unreal Tournament.:)
I have to agree... Yeah, 2.4 may be a bit late. But it's going to be rock-solid when it comes out...
I'm running 2.4.0-test8 (was running test7) on my box, and it's a champ. This pre-release already kicks the crap out of anything our friends in Redmond can put out...
The only time my box is unstable is when doing 3D work. But that's because NVidia sucks and won't release open-source drivers that people can debug.
In an offtopic note: Could NVidia's refusal to release source have to do with the fact that the GeForce and Quadro are identical chips, only differing in their device ID number? (i.e. from what I've read at http://www.geocities.com/tnaw_xtennis/, the driver is what makes the difference between the two "chips" - If the card reports itself as a GeForce, the driver disables some features.)
From what I recall, his project was, among other things, "Mobile Linux", essentially creating a version of Linux to take advantage of all of the Crusoe's power-management features. Admittedly, I'm sure he's up to other things, too.
"ESR is a proud supporter of the USCP" - Um, that's blatantly false. ESR has (intelligently) kept his personal politics completely seperate from his business life/advocacy. But anyone who has met him or has bothered to read his homepage (http://www.tuxedo.org/~esr/) will realize that he is NOT a communist, but is a strong supporter of the Libertarian party.
Yeah. For one thing, all of the information in speech is encoded within a 4 kHz bandwidth, while music is a 22 kHz bandwidth. In addition, there are many differences between the nature of voice signals and music signals.
If you're ONLY dealing with speech, it's much easier to get a good compression ratio than if you have to deal with music, and I'm not just talking about the bandwidth differences.
I wouldn't bother with Vorbis - it was designed for music, so it won't work for voice signals as well as codecs designed for voice.
I would look at codecs like the aforementioned GSM or G.whatever (G.711 is one speech codec, can't remember the others. I'd go to http://www.openh323.org/ for some more information on speech codecs among other things.
Note that G.whatever (and I think GSM) too, are at least somewhat encumbered by patents, but the licensing terms are relatively friendly from what I gather. And they are most definately standardized. (The only speech codec in wide use that I can think of off the top of my head is Qualcomm's PureVoice codec, used quite heavily in CDMA cell phones.)
Is it just me, or does that look EXACTLY like the Innovator WebSurfer Pro? (Remember, the one that you could get for $50 for a few days because CompUSA screwed up in their advertising?)
Innovator probably licensed one of AllWell's designs and added their own software.
AFAIK E-beam lithography was simply not practical to actually MAKE anything until recently. It wasn't until the past year or two that projects such as Lucent's SCALPEL went from the research to equipment design stage. E-beam equipment is on the horizon for commercial use, while the stuff in this article is still research and still gets beat by E-beam techniques.
The CueCat device is a physical entity. Hence covered under these regs.
Before, the loophole was that the software was the only thing covered by DC's EULA. Hence to be in the clear legally, all you had to do was throw away the software and obtain a driver/hack the device w/o looking at the software.
But DC is now trying to wrap usage of the device within the EULA. This conflicts with the postal regulations.
A lot of people don't seem to realize that he's not proposing created-by-the-industry-and-forced-down-your-throa t formats. He's saying that now it's feasible to create-your-own format.
In fact, it's already been done, although not for peer-to-peer stuff. Check out www.launch.com (I think.) Unfortunately, last I checked, it was Winblows-only. (Windows Media Audio for their codec.), but if you occasionally boot to Windows, launch.com is a good example of a broadcast version of what he's talking about. Basically, you pop in a list of radio stations and genres that you like. Then launch.com starts broadcasting music. If you think a song totally sucks, you tell the system never to play that song again. The system will remember that, and the chances of a similar song playing will be reduced. Rate a song high, and the system will try to increase the chances of similar songs being played.
Unfortunately, it takes an hour or two of learning before it starts putting out mostly good music. (I think my rather varied music taste confused it... Not fine-grained enough...) Note: This is an hour or two of time, even skipping songs that suck.
Check it out, play with it, and think how much cooler it would be if it had a Napster-sized music collection, including all the nifty esoteric stuff that Napster has and launch.com does not.
Another similar approach - Go to mp3.com. Go to the site of an artist you like. (Assuming that you know of an artist or two there that you like.) Then check out the "other artists we like" links.
BTW, if you like the Rocket Arena 3 soundtrack, most of those artists are on mp3.com. If you liked the stuff from Silent Warrior, Upbeat Depression, or Masada, there's a lot more good stuff on their pages.
Just the opposite - The DoS packets are spoofed, because they only need to go one way.
As has been pointed out numerous times in this article before, THIS DOES NOT AFFECT TCP STREAMS! If you have a TCP connection, YOUR IP IS ALREADY KNOWN! You cannot combine spoofing with the ability to recieve data. If you want to remain anonymous, use an anonymizer proxy, which itrace will not affect.
If itrace sent one traceback packet for each packet that passed through a router, it would far more than double the effectiveness of the DDoS - For every packet that went from source to destination, a new packet would be generated for EVERY HOP! Of course, this is a moot point, since it's only one out of every 20,000 packets that goes through a router. (Of course, this means that if you have 20 hops, a traceback message will come from somewhere in the route every 1000 packets or so...)
But last I heard, PowerPCs weren't nearly as hot as Apple made them out to be. Remember long ago when Slashdot covered a set of benchmarks that discovered that the whole "G3 is twice as fast as a PII" was utter bullshit except for one or two special circumstances?
PIIIs were an incremental increase over the PII, and last I checked, the G4 was only an incremental increase over the G3 if you didn't take into account AltiVec. So the speed comparison of the two probably still remains similar. Hence a 1 GHz PIII is going to beat a 500 MHz G4 by a significant amount. It's not going to be the 2x performance increase that the clock speed says, but it will be quite significant. Throw the Athlon in there, which has a tendency to kill the PIII at a given clock rate in floating-point performance, and those G4s start looking anemic even to someone who thinks seriously about the situation.
And next is where the "Processor MHz isn't everything" idea works against Apple. Given that the PowerMacs and PCs have the same memory bus width (Both use SDRAM and neither have a requirement of memory being installed as matched pairs), when it comes to memory bandwidth, FSB MHz means everything. When most PCs moved from 100 to 133 MHz FSB is when Apple finally started moving to 100 MHz. And the Athlon has 100 MHz DDR for an effective FSB clock of 200 MHz... Although unfortunately, that's limited by the RAM running at only 133. (I haven't seen on-motherboard cache since L2 was moved from the mobo to the CPU.)
I think Apple's G4 advertising campaign was pushing the fact that anything over 1 GFlop was considered a supercomputer (And hence, a non-exportable munition) by the US Government.
So that's the definition the guy is talking about - The government was saying that 1 GFlop was not exportable because it was a supercomputer, and now they've upped that number.
I'm pretty sure it's somewhere on www.digital-digest.com - There's a HUGE amount of DVD information there. If it's not linked to from there, I don't remember where I saw it...
Slashdot Mirror
Do you call crashing routinely within an hour (sometimes within 5 minutes) of starting a game good performance?
I don't. NVidia's drivers are unstable and crash my system all the time.
One thing that doesn't seem to get addressed by ANY Linux 3D hardware review I've seen (One was also in LJ recently), is the issue of stability.
We run Linux because it's stable, even in cases where we know that we lose performance because of that stability.
Well, all of these reviews basically say, "If you can afford it, buy NVIDIA! They're fast!".
So what if your card is fast but your machine crashes in the middle of a game? NVIDIA's drivers just plain suck in the stability arena - My machine crashes on a regular basis in both Q3 and UT with those crap drivers. Even Windows looks rock-solid compared to Linux when using those accursed drivers! And all this because the only difference between the Quadro and GeForce is a PCI/AGP device ID and a few flags set in the driver because of said ID - it's the only plausible reason for being the only vendor not to release source/specs. (Check out http://www.geocities.com/tnaw_xtennis/)
Just to let you know - The K6 SMP support was OpenAPIC. Too bad no one made a mobo for it. Don't know if the Athlon is or a different protocol. (I think it may be different - Considering how different the bus/cache design is from the old Socket 7 designs in general.)
That's why the K6 had SMP support.
But unfortunately, the workaround made the K6 SMP protocol incompatible with that of the Pentium. And AMD wasn't big enough at the time for any chipset/mobo manufacturer to justify creating a new SMP mobo for a small market.
It's a known feature of the Athlon. In fact, one of the things AMD was proud of is that in SMP configurations, each CPU apparently gets its own frontside bus into the chipset IIRC. Even the K6 (Maybe, definately the K6-2 and K6-3) supported SMP. Problem was, it was not the same SMP scheme as used by the classic Pentium, and there was never an SMP mobo made for it. I hope the same thing doesn't happen to the Athlon. (Although from what I remember, an SMP chipset was due relatively soon on AMD's chipset roadmap. As in by the end of the year, I think.)
Well, it seems that most of the people in this thread have had no problems whatsoever with test8 - I love it.
As to Microsoft products - Winblows won't recognize the second IDE channel on my new mobo (VIA KT133 chipset). Bye-bye DVD drive. The drivers that come with Winblows give a "Device is not present or is not working properly.", VIA's updated drivers crash the machine when I try to install them. Yeah, Windows is reeeeaaally stable. That's in addition to the standard BSODs.
Recent versions of gcc (2.95 and above, RH7 has 2.96) and kernels don't mix. It's a kernel bug that for some odd reason the kernel developers never accept the existence of. (The same went for egcs during the 2.0.x series...)
Because of this known issue, RH7 includes a "kgcc" package for compiling kernels. You will have to change the kernel makefile to force use of this compiler. (On my machine, 'export CC=kgcc' doesn't seem to do anything...)
While I don't know about HDs, USB audio devices and scanners are supported. (Your particular device may not be, but many are.) There is USB storage support in the kernel, but I think it's going to remain marked as experimental even in the final 2.4 release. (Of course, some of the "experimental" code in Linux works quite well...)
:)
The kernel support for USB mice definately kicks ass. I finally moved my IntelliMouse Explorer from the PS/2 port to the USB port (Higher sampling rate with USB), and I'm 2-3 times as lethal with the sniper rifle in Q3Fortress and Unreal Tournament.
I have to agree... Yeah, 2.4 may be a bit late. But it's going to be rock-solid when it comes out...
I'm running 2.4.0-test8 (was running test7) on my box, and it's a champ. This pre-release already kicks the crap out of anything our friends in Redmond can put out...
The only time my box is unstable is when doing 3D work. But that's because NVidia sucks and won't release open-source drivers that people can debug.
In an offtopic note: Could NVidia's refusal to release source have to do with the fact that the GeForce and Quadro are identical chips, only differing in their device ID number? (i.e. from what I've read at http://www.geocities.com/tnaw_xtennis/, the driver is what makes the difference between the two "chips" - If the card reports itself as a GeForce, the driver disables some features.)
From what I recall, his project was, among other things, "Mobile Linux", essentially creating a version of Linux to take advantage of all of the Crusoe's power-management features. Admittedly, I'm sure he's up to other things, too.
"ESR is a proud supporter of the USCP" - Um, that's blatantly false. ESR has (intelligently) kept his personal politics completely seperate from his business life/advocacy. But anyone who has met him or has bothered to read his homepage (http://www.tuxedo.org/~esr/) will realize that he is NOT a communist, but is a strong supporter of the Libertarian party.
Yeah. For one thing, all of the information in speech is encoded within a 4 kHz bandwidth, while music is a 22 kHz bandwidth. In addition, there are many differences between the nature of voice signals and music signals.
If you're ONLY dealing with speech, it's much easier to get a good compression ratio than if you have to deal with music, and I'm not just talking about the bandwidth differences.
I wouldn't bother with Vorbis - it was designed for music, so it won't work for voice signals as well as codecs designed for voice.
I would look at codecs like the aforementioned GSM or G.whatever (G.711 is one speech codec, can't remember the others. I'd go to http://www.openh323.org/ for some more information on speech codecs among other things.
Note that G.whatever (and I think GSM) too, are at least somewhat encumbered by patents, but the licensing terms are relatively friendly from what I gather. And they are most definately standardized. (The only speech codec in wide use that I can think of off the top of my head is Qualcomm's PureVoice codec, used quite heavily in CDMA cell phones.)
IIRC, one of the big reasons for MPEG-4's great performance is the fact that it uses wavelet compression.
I could be wrong, tho.
Yeah. I have one too, just haven't had time to hack it yet... my iOpener has priority, just need to make a cable...
Is it just me, or does that look EXACTLY like the Innovator WebSurfer Pro? (Remember, the one that you could get for $50 for a few days because CompUSA screwed up in their advertising?)
Innovator probably licensed one of AllWell's designs and added their own software.
AFAIK E-beam lithography was simply not practical to actually MAKE anything until recently. It wasn't until the past year or two that projects such as Lucent's SCALPEL went from the research to equipment design stage. E-beam equipment is on the horizon for commercial use, while the stuff in this article is still research and still gets beat by E-beam techniques.
FREE is just plain FREE.
FREE* has fine print attached somewhere. Just look for another appearance of that damned *.
The CueCat device is a physical entity. Hence covered under these regs.
Before, the loophole was that the software was the only thing covered by DC's EULA. Hence to be in the clear legally, all you had to do was throw away the software and obtain a driver/hack the device w/o looking at the software.
But DC is now trying to wrap usage of the device within the EULA. This conflicts with the postal regulations.
A lot of people don't seem to realize that he's not proposing created-by-the-industry-and-forced-down-your-throa t formats. He's saying that now it's feasible to create-your-own format.
In fact, it's already been done, although not for peer-to-peer stuff. Check out www.launch.com (I think.) Unfortunately, last I checked, it was Winblows-only. (Windows Media Audio for their codec.), but if you occasionally boot to Windows, launch.com is a good example of a broadcast version of what he's talking about. Basically, you pop in a list of radio stations and genres that you like. Then launch.com starts broadcasting music. If you think a song totally sucks, you tell the system never to play that song again. The system will remember that, and the chances of a similar song playing will be reduced. Rate a song high, and the system will try to increase the chances of similar songs being played.
Unfortunately, it takes an hour or two of learning before it starts putting out mostly good music. (I think my rather varied music taste confused it... Not fine-grained enough...) Note: This is an hour or two of time, even skipping songs that suck.
Check it out, play with it, and think how much cooler it would be if it had a Napster-sized music collection, including all the nifty esoteric stuff that Napster has and launch.com does not.
Another similar approach - Go to mp3.com. Go to the site of an artist you like. (Assuming that you know of an artist or two there that you like.) Then check out the "other artists we like" links.
BTW, if you like the Rocket Arena 3 soundtrack, most of those artists are on mp3.com. If you liked the stuff from Silent Warrior, Upbeat Depression, or Masada, there's a lot more good stuff on their pages.
Just the opposite - The DoS packets are spoofed, because they only need to go one way.
As has been pointed out numerous times in this article before, THIS DOES NOT AFFECT TCP STREAMS! If you have a TCP connection, YOUR IP IS ALREADY KNOWN! You cannot combine spoofing with the ability to recieve data. If you want to remain anonymous, use an anonymizer proxy, which itrace will not affect.
If itrace sent one traceback packet for each packet that passed through a router, it would far more than double the effectiveness of the DDoS - For every packet that went from source to destination, a new packet would be generated for EVERY HOP! Of course, this is a moot point, since it's only one out of every 20,000 packets that goes through a router. (Of course, this means that if you have 20 hops, a traceback message will come from somewhere in the route every 1000 packets or so...)
Ever heard of Medicare/Medicaid?
Yes, I know MHz can be meaningless.
But last I heard, PowerPCs weren't nearly as hot as Apple made them out to be. Remember long ago when Slashdot covered a set of benchmarks that discovered that the whole "G3 is twice as fast as a PII" was utter bullshit except for one or two special circumstances?
PIIIs were an incremental increase over the PII, and last I checked, the G4 was only an incremental increase over the G3 if you didn't take into account AltiVec. So the speed comparison of the two probably still remains similar. Hence a 1 GHz PIII is going to beat a 500 MHz G4 by a significant amount. It's not going to be the 2x performance increase that the clock speed says, but it will be quite significant. Throw the Athlon in there, which has a tendency to kill the PIII at a given clock rate in floating-point performance, and those G4s start looking anemic even to someone who thinks seriously about the situation.
And next is where the "Processor MHz isn't everything" idea works against Apple. Given that the PowerMacs and PCs have the same memory bus width (Both use SDRAM and neither have a requirement of memory being installed as matched pairs), when it comes to memory bandwidth, FSB MHz means everything. When most PCs moved from 100 to 133 MHz FSB is when Apple finally started moving to 100 MHz. And the Athlon has 100 MHz DDR for an effective FSB clock of 200 MHz... Although unfortunately, that's limited by the RAM running at only 133. (I haven't seen on-motherboard cache since L2 was moved from the mobo to the CPU.)
I think Apple's G4 advertising campaign was pushing the fact that anything over 1 GFlop was considered a supercomputer (And hence, a non-exportable munition) by the US Government.
So that's the definition the guy is talking about - The government was saying that 1 GFlop was not exportable because it was a supercomputer, and now they've upped that number.
I'm pretty sure it's somewhere on www.digital-digest.com - There's a HUGE amount of DVD information there. If it's not linked to from there, I don't remember where I saw it...