HP-UX has always been clunky, trying to splice in some DNA from a totally unrelated (and more technically advanced) version of Unix was a pretty tall order. They probably would have had better luck porting Tru64 over to PA-RISC and trying to merge in the bits they really wanted from HP-UX.
all cpus will have peaked at some--obscenely high--MHz limit
Speaking as someone who started out with a 1.774 MHz processor, current CPU speeds are already obscenely high. Hell, my disk drive has more memory (2MB vs 16K) than my first computer...
Re:What the hell is this crap?!
on
RAD with Ruby
·
· Score: 2, Insightful
Actually, this *is* more like the stories from the elder days. It's news for nerds, not news for webmonkeys. BITD, this was the kind of thing I came to/. for -- what's in the new kernal, is there going to be a new filesystem, is there support for >512M of memory yet? Articles like these help keep this site true to its roots.
all I see is the manufacturer lowering the maximum specs to any tests will show it 'overachieving'
Look closer and you'll see Marketing ratchet the specs back up to just a bit higher than the competition. After all, when's the last time anyone asked Manufacturing (or Development, for that matter) for a spec?
Yes, as a fellow minor platform user myself, I've often wished I could just grab the source and build it myself. It's possible to run Java apps on my system, but they run in Linux compatability mode, which means the classpath is all wacky (paths all relative to root of emulated Linux system, not to real root, etc.)
HTTP is supposed to be stateless. Cookies violate that
<pedantic> HTTP is stateless. Cookies just make it possible for an application to store information on the client box. The actual state is held on the server, the cookie is just a key telling the server the index into its collection of saved states. HTTP itself doesn't track any state. </pedantic>
He said free as in beer. JREs for all major platforms are available for the downloading (either from Sun or the OS vendor).
OTOH, if you really want to add features to the JRE, there's no stopping you from extracting a class from rt.jar and replacing it with one you wrote. If you're really clever, you can even get a bytecode editor, change the name of the original class, then writing a new class with the original name that extends the original class, and only adds the new features you want...
The recompilation done in Java is no different than profiling in C or any other language
Actually, it is. A good JVM will actually keep different versions of the same method (optimized for different combinations of inputs) around. Profiling a C program will permit you to optimize it for only one type of usage. And should usage patterns change (slashdotting, anyone?), you'd have to pull the code, re-optimize and re-deploy (hopefully re-testing as well). Not necessary with the JVM.
Good points, but I think they underscore the need for a solutions that fits within the current system, as opposed to one that supplants it. I (well, my wife) have a station wagon and I *still* wind up renting the little truck from Home Depot to haul sheets of MDF or 12' 2x6es.
That said, the majority of my time in the car is spent commuting, and I could certainly make some concessions to use public transit. Unfortunately, there's no public transit between Lincoln and Omaha (~60 miles), so I'm stuck making the drive. However, when I lived in Chicago I took the El everywhere, and I actually had to make excuses for taking the car, as the battery would go flat if I didn't drive it once a month or so.
One of the things I found interesting about the original system is that it was designed by a company who had done a similar (but much smaller) system in Europe. They actually refused to bid on the project initially, because they believed the DIA planners' objectives were unachievable. I'm not sure how they were eventually coerced into building it.
Actually, the one near me has a refreshingly large selection of "actual" electronic stuff (resistors, capacitors, soldering irons, perf and copper-clad board, etc). In fact, I just bought a rectifier there to fix the power supply of my Ten-Tec Corsair.
Back when I lived in Portland, though, I would have agreed with you. It was all cell phones, DVD players and toys. The only radio-type things they had were scanners and those Grundig hand-cranked shortwave sets.
Yepper, I can vouch for that. My folks used to live close to Sandia Labs, and they had a Cray-I listed at one of their "garage sales" for $1000, my dad talked them down to $500 (Dad: "Would you take $500?" Sandia: "Yeah, sure"), then found out he'd need 208V 400Hz three-phase power (they weren't selling the MG set the Cray came with) to the tune of $900K/month. "That's why we're selling it." the Sandia guys said. Someone else eventually picked it up, don't know how much they paid, but it had been crossed off the list by the time we left.
Not necessarily. If you've got "many, many" heads, then you have several options. One would be to have some fixed heads above several tracks, eliminating seek time (great for swap space). Another would be to partition the disk into platter groups, with a separate R/W head serving each group. The separate actuator arms could use the same pivot point and magnet assembly. I don't think you'd need a special controller to prevent head crashes, the head assembly only sweeps ~1/4 of the disk surface anyway, adding another arm at the opposite corner wouldn't interfere. It could probably be done within an existing 5.25" form factor, too.
Most of the MP3s I have were recorded at pretty low bit rates, so they've got the quality of an 8-track. That said, I'd still expect to see the use of alternate formats (Ogg et. al.) rise before I'd declare MP3s on the decline. Or was it all just a fad, after all?
Absolutely. I bought a "starter" digital camera (Panasonic Lumix DMC-20) with only 2.1MP a couple of years ago, strictly as an interim unit until I determined what I really wanted. I never did buy another, because the Panasonic does really everything I need it to do. If I want to get all "artsy", I have an old Nikon 35mm with a variety of lenses and filters I can play with. The digital is just fine for snapshots and the like. Of course, the Lumix series all have Leica DC lenses, which contributes a *lot* to the picture quality.
If it's a library, doubtless they'd release it under the LGPL, which would mean that anyone using it would only have to provide the source to the library (or a pointer thereto) if anyone asked.
The first Superman movie was OK in a campy sort of way, but to flog a limp franchise through three sequels, each worse than the last and none as good as the first, should be criminal.
Of course, if it was, half of Hollywood would be hanging around the corner of Pico and Alvarado holding "Will write/direct/star for food" signs...
a simpler assembler language (for the 5% of the coding that takes 50% of the time:-)
Actually, one of the things that makes that 5% of the code so difficult is often because you're trying to calculate 32-bit values with an 8-bit accumulator. On the fly. While handling interrupts...
I think a lot of it has to do with the fact that the interconnects that come with most consumer-grade components are crap. I replaced the interconnects that came with my CD player (Denon DCD-620) with a $15 set and could actually hear the difference. I then tried with a (rich!) friend's $100 interconnects and didn't notice a difference.
I think that interconnect performance is an asymptotic curve, and it rises pretty steeply at the low end.
Slashdot Mirror
I used to play this a lot in college, but I don't have enough room in my refrigerator for all that Jell-O anymore.
Or did you mean the kind you play with your clothes on?
HP-UX has always been clunky, trying to splice in some DNA from a totally unrelated (and more technically advanced) version of Unix was a pretty tall order. They probably would have had better luck porting Tru64 over to PA-RISC and trying to merge in the bits they really wanted from HP-UX.
all cpus will have peaked at some--obscenely high--MHz limit
Speaking as someone who started out with a 1.774 MHz processor, current CPU speeds are already obscenely high. Hell, my disk drive has more memory (2MB vs 16K) than my first computer...
Actually, this *is* more like the stories from the elder days. It's news for nerds, not news for webmonkeys. BITD, this was the kind of thing I came to /. for -- what's in the new kernal, is there going to be a new filesystem, is there support for >512M of memory yet? Articles like these help keep this site true to its roots.
all I see is the manufacturer lowering the maximum specs to any tests will show it 'overachieving'
Look closer and you'll see Marketing ratchet the specs back up to just a bit higher than the competition. After all, when's the last time anyone asked Manufacturing (or Development, for that matter) for a spec?
Not only did throwing a hat on my plunger never neutralize me, but it seemed to encourage the ladies...
Wait, what were we talking about here?
The cache which can be cached is not the true cache.
drag is proportional to the cross-section area of the plane
Wouldn't that be frontal area? That's how it works for cars, anyway, I'll happily accept that planes are different.
Not that I'm an expert on drag, but I have been to New Orleans...
Yes, as a fellow minor platform user myself, I've often wished I could just grab the source and build it myself. It's possible to run Java apps on my system, but they run in Linux compatability mode, which means the classpath is all wacky (paths all relative to root of emulated Linux system, not to real root, etc.)
HTTP is supposed to be stateless. Cookies violate that
<pedantic>
HTTP is stateless. Cookies just make it possible for an application to store information on the client box. The actual state is held on the server, the cookie is just a key telling the server the index into its collection of saved states. HTTP itself doesn't track any state.
</pedantic>
He said free as in beer. JREs for all major platforms are available for the downloading (either from Sun or the OS vendor).
OTOH, if you really want to add features to the JRE, there's no stopping you from extracting a class from rt.jar and replacing it with one you wrote. If you're really clever, you can even get a bytecode editor, change the name of the original class, then writing a new class with the original name that extends the original class, and only adds the new features you want...
The recompilation done in Java is no different than profiling in C or any other language
Actually, it is. A good JVM will actually keep different versions of the same method (optimized for different combinations of inputs) around. Profiling a C program will permit you to optimize it for only one type of usage. And should usage patterns change (slashdotting, anyone?), you'd have to pull the code, re-optimize and re-deploy (hopefully re-testing as well). Not necessary with the JVM.
Good points, but I think they underscore the need for a solutions that fits within the current system, as opposed to one that supplants it. I (well, my wife) have a station wagon and I *still* wind up renting the little truck from Home Depot to haul sheets of MDF or 12' 2x6es.
That said, the majority of my time in the car is spent commuting, and I could certainly make some concessions to use public transit. Unfortunately, there's no public transit between Lincoln and Omaha (~60 miles), so I'm stuck making the drive. However, when I lived in Chicago I took the El everywhere, and I actually had to make excuses for taking the car, as the battery would go flat if I didn't drive it once a month or so.
One of the things I found interesting about the original system is that it was designed by a company who had done a similar (but much smaller) system in Europe. They actually refused to bid on the project initially, because they believed the DIA planners' objectives were unachievable. I'm not sure how they were eventually coerced into building it.
Actually, the one near me has a refreshingly large selection of "actual" electronic stuff (resistors, capacitors, soldering irons, perf and copper-clad board, etc). In fact, I just bought a rectifier there to fix the power supply of my Ten-Tec Corsair.
Back when I lived in Portland, though, I would have agreed with you. It was all cell phones, DVD players and toys. The only radio-type things they had were scanners and those Grundig hand-cranked shortwave sets.
Yepper, I can vouch for that. My folks used to live close to Sandia Labs, and they had a Cray-I listed at one of their "garage sales" for $1000, my dad talked them down to $500 (Dad: "Would you take $500?" Sandia: "Yeah, sure"), then found out he'd need 208V 400Hz three-phase power (they weren't selling the MG set the Cray came with) to the tune of $900K/month. "That's why we're selling it." the Sandia guys said. Someone else eventually picked it up, don't know how much they paid, but it had been crossed off the list by the time we left.
Not necessarily. If you've got "many, many" heads, then you have several options. One would be to have some fixed heads above several tracks, eliminating seek time (great for swap space). Another would be to partition the disk into platter groups, with a separate R/W head serving each group. The separate actuator arms could use the same pivot point and magnet assembly. I don't think you'd need a special controller to prevent head crashes, the head assembly only sweeps ~1/4 of the disk surface anyway, adding another arm at the opposite corner wouldn't interfere. It could probably be done within an existing 5.25" form factor, too.
I suspect it may be the passwd itself but I am not sure since it is not clear.
Of course it's not clear, it's been hashed -- haven't you been following along?
Most of the MP3s I have were recorded at pretty low bit rates, so they've got the quality of an 8-track. That said, I'd still expect to see the use of alternate formats (Ogg et. al.) rise before I'd declare MP3s on the decline. Or was it all just a fad, after all?
Absolutely. I bought a "starter" digital camera (Panasonic Lumix DMC-20) with only 2.1MP a couple of years ago, strictly as an interim unit until I determined what I really wanted. I never did buy another, because the Panasonic does really everything I need it to do. If I want to get all "artsy", I have an old Nikon 35mm with a variety of lenses and filters I can play with. The digital is just fine for snapshots and the like. Of course, the Lumix series all have Leica DC lenses, which contributes a *lot* to the picture quality.
If it's a library, doubtless they'd release it under the LGPL, which would mean that anyone using it would only have to provide the source to the library (or a pointer thereto) if anyone asked.
that the Salkinds won't be involved.
The first Superman movie was OK in a campy sort of way, but to flog a limp franchise through three sequels, each worse than the last and none as good as the first, should be criminal.
Of course, if it was, half of Hollywood would be hanging around the corner of Pico and Alvarado holding "Will write/direct/star for food" signs...
a simpler assembler language (for the 5% of the coding that takes 50% of the time :-)
Actually, one of the things that makes that 5% of the code so difficult is often because you're trying to calculate 32-bit values with an 8-bit accumulator. On the fly. While handling interrupts...
Heh. Go lemmings, go!
I think a lot of it has to do with the fact that the interconnects that come with most consumer-grade components are crap. I replaced the interconnects that came with my CD player (Denon DCD-620) with a $15 set and could actually hear the difference. I then tried with a (rich!) friend's $100 interconnects and didn't notice a difference.
I think that interconnect performance is an asymptotic curve, and it rises pretty steeply at the low end.