Getting 100KBytes/s instead of 200KBs... (connecting to Sunsite NE normally max's our internet connection)
The link is probably being rather saturated.
Can anyone remember the international connection speed of SuperJanet? Last I remember it was 8Mbits, but I have this vague memory of it being upped to 30-40Mb, or something.
I've translated over 2100 pages of manga, but I'm still not that used to this kind of stuff. Didn't have my character dictionary around either, AND it was in SJIS...
Anyway, in the 3rd paragraph it says something like "one clear fact about the next generation playstation is that the development OS is Linux. The name of the hardware is still not clear."
Okay, the development OS is Linux. I believe that's also the case with "Project X". However, the OS for the shipping product will probably be a RTOS. Probably some development of the original Playstation OS, whatever that was...
Well, compared to recent CPUs, not that good. It does have about as good integer as a same speed P2, but pretty poor FP - having no FPU. The one in the netwinder runs at 275Mhz. However, it is cheap ($30), and very low power (1W). It's also been around for two years at about that level - development hasn't been good since the first version.
Intel are now designing the StrongARM 2, which'll come out in about a year, running up to 600Mhz, and with hardware FP. It'll still be cheap, and consume very little power though. Here's some info at The Register. Now, if that group doing multiple StrongARMs on a PCI card would do something, then we could have some fun...
PS I've had a StrongARM in one of my computers for over 2 years... overclocked too. (even overclocked, and with no fan or even heat-sink it barely gets warm...)
Yeah, Sun's margins on hardware in general, are about 50%, while their margins on software is about 30%. They probably don't care too much because 90% of their revenure is from hardware. To some extent, Solaris is just there to sell the hardware.
With regards to the 3.2GBps system bus/interconnect, what were you referring to? The main memory bus on a processor board is somewhere around that level. However, the main data bus is nearly 13GByte/s and about 15GB on the newer, slightly revised ones. It also has a 6GByte/s IO bus.
I think looking at the physical chip design is a bit extreeme, for trying to get a bit more performance...
Anyway, ARM is the current champion of chip IP. They don't make, or get made, any chips themselves. Instead anyone can get a liscence to make them, or to make designs with ARMs. About 50 million ARM 'chips' were made last year, up from 10m the year before. ARM are expected to capture 70% of the digital mobile phone market this year... 3Com are going to start using them in network cards, and they're already appearing in hard-discs and stuff.
It's not particularly sexy, but ARM is worth nearly $2bn, a 300% increase since they floated a year ago. They're doing pretty damn well ^-^
However, ARM do charge an upfront fee. This cost for the ARM 9 is rumoured to be around $5-10million. The cost per chip is about $0.10 to $0.20 though. Incidentally, liscencing Sun's Jini costs nothing up front, and has an individual cost of about $0.20, with a max of $250,000 I think.
There may be other exceptions, but Sun is definately not 'failing'. Of the big hardware companies, the only ones who got any decent growth last year where Dell, Compaq and Sun. However, Dell and Compaq aren't expected to do so well this year, according to analysists, while Sun is expected to do pretty well.
See this report at Forbes. A couple of months ago, Sun's stock was $40-$50, now it's at $100. There's also expected to be an announcement from HP today that it'll be splitting it's business up, so that it can try and compete better - IBM and Sun were named as two companies giving them trouble... Looks like Sun's "unix-only" strategy is doing pretty well.
I do agree with the general point that some of these 'open source' innitiatives could be considered "desperate", or "last ditch".
The last time we had a report of a GPL/open source code 'violation' it turned up, pretty much to be crying wolf - it was beta code. If you look at the web site, it says
MOSIX for Linux 2.2.2 is powering our 100 node scalable cluster. A 6 processor version, called MO6 is available for
Beta testing.
(Italics mine) Let's not have another witch hunt, unless they deserve it, okay? I wouldn't count myself as being able to give a good answer on the subject. btw, this is a university - though I'm not sure if this is a commercial operation or not. (doesn't appear to be commercial though)
I'm pretty sure this isn't a world first. I'm pretty sure I read people getting over 1Tb over a year ago...
Whatever. Most people won't see this stuff for real for ages. Though I happen to know that SGI are investigating using fibre instead of PCB for ludicrous speed main memory connections - for high end SMP machines of course. Apparantly they have stuff running in the lab, running at several 100 GBytes/sec. Most of todays PCs have 0.8GByte/sec main memory.
This is not in any way to an attempt to disagree about the accuracy of the excellent rebuttal article. Anyway, according to this Sun white paper Solaris 7 has 400,000 lines of code in the kernel, and 12 million for 'Solaris'.
Look for "lines of code" in the text - it'll be the 2nd match. the white paper also suggests NT might have some trouble with it's 40 million lines of code, most of which is very recent.
Fellow eternal MS-basher Scott McNealy (CEO of Sun) said not to break them up. (see news.com article).
"The structural remedies that people are talking about--separating applications from operating systems--is like one of those horror movies where you cut the monster in half, and now you have two monsters."
It's hard to say what would happen. Much would depend on the implimentation.
Personally, I'd say that if MS was forced to make ALL its APIs for all versions of Windows (inc W2K) publicly available and free to use for any purpose would be the best thing to go for from the DoJ's point of view. It's very hard to argue against it, and would also be very effective in giving MS some competition. Basically, anyone could then do a complete 'Wine' and make a system where you can run Windows apps without Windows.
The other thing to do would be to prevent all the software tying that makes it so hard for OEMs to build PCs without Windows pre-installed. MS also ties lots of its software together in certain ways, which rather forces people to run 'pure MS'.
I'm not saying that splitting MS could have it's uses, but at the moment, I would recommend the DoJ to do the above. When, one way or another, the case gets to the Supreme Court, then go for splitting.
It'll be another 1-3 months before the trial ends, and the judge'll probably take 1-2 months making a descision. So I guess we'll see something by mid-year. (about the same time the K7 comes out)
I'm sure there will be people who disagree, but theregister.co.uk has had good trial coverage, IMHO. They often mention things I've never seen elsewhere and they have good analysis.
Personally, I think it's better to keep this sort of thing off Slashdot, on the day-to-day side. I'd prefer to see more technical stuff.
As a "wouldn't that be interesting" endnote, I'd say it's pretty likely that a) the DoJ will win (nahh!) and b) that they will ask for the judge to order MS to make public all their APIs. (the chance of making all the APIs freely useable is another matter though). Nice for Wine...
Am I missing something? Else what's the big deal of 128Mbs? If it was 128Gbs, then that would be cool... though I thought the world record, in the lab, for a single bit of fibre was well over 1Tbit/s.
Or does he mean he's improved the efficiency? 100Mb ethernet runs at 200Mhz - it uses 2state changes to indicate 1 bit for reasons of reliability. (can't remember the technical term for this...)
I've read most of the Advanced Perl Programming book, but I didn't find that much useful info. I'd already done most of what was mentioned...
btw, I've been programming Perl professionally for well over 3 years... guess it kinda makes a difference. I would say that most people would find APP quite useful.
However, I will check out the perl cookbook. I'm particularly interested in the bit on daemons.
The bit on efficiency and programming style near the end of the Camel Book is quite good too. The bit on how to style programming was interesting - it is almost identical to what I do! (I was already doing things that way before I read it...)
I read this article back when it was originally published and through it was kinda weird, especially for Sun World. I've noticed that most of their recent articles have been kinda strange.
Just as a side note, while the (commercial) Unix workstation market isn't moving much, Sun themselves are doing great - they're getting 20-50% growth in workstation/server shipments. Co-incidentaly, or not, they're also the only ones 100% committed to Unix.
Take them with a bit of a pinch of salt, but here's some recent figures from Sun about workstation and server growth.
Sometimes called a 'log' file system, though they aren't the same. AFAIK the difference is that a log file system just writes the meta-data super-safely (so you don't need to fsck), while a full jounaling writes all data out as a 'log' - it just appends to where it previously wrote. This does mean you need a background garbage collection processes.
Doing writes in this way makes writes go MUCH faster. I read a review by one journalist (no pun intended) who didn't believe Sun's claims that it made long sequential writes go 3x faster or more... It did. Unfortunately, Sun haven't (yet) put full journaled FS support into standard Solaris, though there is an option to put "UFS logging" on - it can also be done on the fly. Still, deleting files and creating lots of small ones goes about 5-10x faster when you put logging on.
Was talking to a games developer on Saturday, who had access to both a P3 and a K7. He wouldn't say how much faster the K7 was, but left the impression that there is no contest, even comparing with games optimised for P3.
It seems likely to me that Intel will only catch up with the K7 when they bring out Merced, but I wouldn't even bet on that! (yup - it would not surprise me at all if the top end K7s that are being produced when the Merced comes out will be faster than the Merced)
Slashdot Mirror
The link is probably being rather saturated.
Can anyone remember the international connection speed of SuperJanet? Last I remember it was 8Mbits, but I have this vague memory of it being upped to 30-40Mb, or something.
- I'm a cynical optimist [laughs].
Wow... that's me that is. First time I've seen someone else describe themselves that way. Must have happened somewhere though...I'm a cynic who has hope for the human race, although I look at human folly and sometimes I get very frustrated.
Anyway, in the 3rd paragraph it says something like "one clear fact about the next generation playstation is that the development OS is Linux. The name of the hardware is still not clear."
Okay, the development OS is Linux. I believe that's also the case with "Project X". However, the OS for the shipping product will probably be a RTOS. Probably some development of the original Playstation OS, whatever that was...
Well, compared to recent CPUs, not that good. It does have about as good integer as a same speed P2, but pretty poor FP - having no FPU. The one in the netwinder runs at 275Mhz. However, it is cheap ($30), and very low power (1W). It's also been around for two years at about that level - development hasn't been good since the first version.
Intel are now designing the StrongARM 2, which'll come out in about a year, running up to 600Mhz, and with hardware FP. It'll still be cheap, and consume very little power though. Here's some info at The Register. Now, if that group doing multiple StrongARMs on a PCI card would do something, then we could have some fun...
PS I've had a StrongARM in one of my computers for over 2 years... overclocked too. (even overclocked, and with no fan or even heat-sink it barely gets warm...)
With regards to the 3.2GBps system bus/interconnect, what were you referring to? The main memory bus on a processor board is somewhere around that level. However, the main data bus is nearly 13GByte/s and about 15GB on the newer, slightly revised ones. It also has a 6GByte/s IO bus.
Anyway, ARM is the current champion of chip IP. They don't make, or get made, any chips themselves. Instead anyone can get a liscence to make them, or to make designs with ARMs. About 50 million ARM 'chips' were made last year, up from 10m the year before. ARM are expected to capture 70% of the digital mobile phone market this year... 3Com are going to start using them in network cards, and they're already appearing in hard-discs and stuff.
It's not particularly sexy, but ARM is worth nearly $2bn, a 300% increase since they floated a year ago. They're doing pretty damn well ^-^
However, ARM do charge an upfront fee. This cost for the ARM 9 is rumoured to be around $5-10million. The cost per chip is about $0.10 to $0.20 though. Incidentally, liscencing Sun's Jini costs nothing up front, and has an individual cost of about $0.20, with a max of $250,000 I think.
The release hasn't appeared on Sun's web site yet - their site normally gets updated late in the day.
Personal gripe - I submitted this story 5 hours ago with some extra info - not some 'paid subscription' junk, but did it get posted? No.
Btw, at the register, they also have this article titled Intel, Sun schmoozing for a chip cruising. The info is rather basic. Apparantly Microsoft is pissed off...
See this report at Forbes. A couple of months ago, Sun's stock was $40-$50, now it's at $100. There's also expected to be an announcement from HP today that it'll be splitting it's business up, so that it can try and compete better - IBM and Sun were named as two companies giving them trouble... Looks like Sun's "unix-only" strategy is doing pretty well.
I do agree with the general point that some of these 'open source' innitiatives could be considered "desperate", or "last ditch".
(Italics mine)
Let's not have another witch hunt, unless they deserve it, okay? I wouldn't count myself as being able to give a good answer on the subject. btw, this is a university - though I'm not sure if this is a commercial operation or not. (doesn't appear to be commercial though)
Whatever. Most people won't see this stuff for real for ages. Though I happen to know that SGI are investigating using fibre instead of PCB for ludicrous speed main memory connections - for high end SMP machines of course. Apparantly they have stuff running in the lab, running at several 100 GBytes/sec. Most of todays PCs have 0.8GByte/sec main memory.
From CPU info center
450 Mhz Pentium II 17.2 - 12.9
450 Mhz Xeon 18.9 - 14.7
400 Mhz PowerPC G3 17.6 - 12.2
200 Mhz IBM POWER3 13.2 - 30.1
450 Mhz UltraSparcII 19.6 - 27.1
600 Mhz UltraSparcIII 35+ - 60+ (est)
575 Mhz Alpha 21264 30.3 - 47.7
667 Mhz Alpha 21264 44 - 66
1000 Mhz Alpha 21364 ~70 - ~120 (est)
250 Mhz MIPS R10000 14.7 - 24.5
Look for "lines of code" in the text - it'll be the 2nd match. the white paper also suggests NT might have some trouble with it's 40 million lines of code, most of which is very recent.
It's hard to say what would happen. Much would depend on the implimentation.
Personally, I'd say that if MS was forced to make ALL its APIs for all versions of Windows (inc W2K) publicly available and free to use for any purpose would be the best thing to go for from the DoJ's point of view. It's very hard to argue against it, and would also be very effective in giving MS some competition. Basically, anyone could then do a complete 'Wine' and make a system where you can run Windows apps without Windows.
The other thing to do would be to prevent all the software tying that makes it so hard for OEMs to build PCs without Windows pre-installed. MS also ties lots of its software together in certain ways, which rather forces people to run 'pure MS'.
I'm not saying that splitting MS could have it's uses, but at the moment, I would recommend the DoJ to do the above. When, one way or another, the case gets to the Supreme Court, then go for splitting.
It'll be another 1-3 months before the trial ends, and the judge'll probably take 1-2 months making a descision. So I guess we'll see something by mid-year. (about the same time the K7 comes out)
Their trial home page is here
Personally, I think it's better to keep this sort of thing off Slashdot, on the day-to-day side. I'd prefer to see more technical stuff.
As a "wouldn't that be interesting" endnote, I'd say it's pretty likely that a) the DoJ will win (nahh!) and b) that they will ask for the judge to order MS to make public all their APIs. (the chance of making all the APIs freely useable is another matter though). Nice for Wine...
Or does he mean he's improved the efficiency? 100Mb ethernet runs at 200Mhz - it uses 2state changes to indicate 1 bit for reasons of reliability. (can't remember the technical term for this...)
btw, I've been programming Perl professionally for well over 3 years... guess it kinda makes a difference. I would say that most people would find APP quite useful.
However, I will check out the perl cookbook. I'm particularly interested in the bit on daemons.
The bit on efficiency and programming style near the end of the Camel Book is quite good too. The bit on how to style programming was interesting - it is almost identical to what I do! (I was already doing things that way before I read it...)
Just as a side note, while the (commercial) Unix workstation market isn't moving much, Sun themselves are doing great - they're getting 20-50% growth in workstation/server shipments. Co-incidentaly, or not, they're also the only ones 100% committed to Unix.
Take them with a bit of a pinch of salt, but here's some recent figures from Sun about workstation and server growth.
Doing writes in this way makes writes go MUCH faster. I read a review by one journalist (no pun intended) who didn't believe Sun's claims that it made long sequential writes go 3x faster or more... It did. Unfortunately, Sun haven't (yet) put full journaled FS support into standard Solaris, though there is an option to put "UFS logging" on - it can also be done on the fly. Still, deleting files and creating lots of small ones goes about 5-10x faster when you put logging on.
It seems likely to me that Intel will only catch up with the K7 when they bring out Merced, but I wouldn't even bet on that! (yup - it would not surprise me at all if the top end K7s that are being produced when the Merced comes out will be faster than the Merced)
Here's Sun's press release
Not exactly old news.
Anyway, I doubt the new parts would overclock
much better than the 300A's.