After the BIOS hands the control of the
machine to the OS, to what extent is the
BIOS used, if at all?
Some of the APM stuff is done by the BIOS in a way that is very hard for an OS to interfere with, or at least to alter other then preventing it. Some OSes use the "BIOS" (actually video card extensions to it) to change vid card modes. An OS could use it to do disk I/O, but then you can't do async I/O which is why no modern OS does it that way.
Sun uses their equivoent of the BIOS (OpenPROM I think) more, it actually does the console screen I/O (and use to suck because they read it out of 8 bit wide ROM with no caching, but that is over with now). Sun didn't make it palatable to do disk I/O or network I/O via OpenPROM though. "New World" Macs uses the same OpenPROM stuff Sun does, but I don't think they use any of it after the boot anyway.
if it tries,
the process gets killed by the OS.
Not exactly true. There isn't a way to try in most cases. You can't try to alter another processes memory because you can't express that except via an OS construct, and that ones doesn't kill you if it doesn't work, just returns an error code. Sometimes the response is more severe then an error code -- a signal. But almost never is your process just killed...not in any normal Unix (I think OpenBSD will panic for some security breaches, like a second chroot in the same process hierachy...but it might just do the equivoent of kill -9).
phones that do text messaging through the numeric pad and other stupid gimmicks
The text messaging on some phones is quite easy to use, and frequently costs less money then a short call...and is a lot less intrusave to recieve (and sometimes make). So for me, text messaging via the keypad is a win.
I don't need to friggin' surf the Internet on an 80x120 pixel screen!
I can't say I've really found web browsing on a phone all that useful. Except once. I had taken a walk in a show storm and managed to get lost on twisty little roads. Mapquest even on the tiny phone was quite useful. A GPS might have been better, but I didn't have one. That was 3 years ago or so, I havn't had great use for a web browser in my phone since.
I don't think this new addressing scheme will have anything to do with how we use cell phones though, just how we use computers to talk to cell phones...
Plus, isn't this just Carl Malamute's tpc.int all over again? Same thing, revers the digit order, put dots between them...
First one to start whining about Java's year-584544016 problem gets whacked with a wet noodle.
I remember seeing a Sun press relase about Java being Y2K complient, how long it would last, and that Sun promised to fix it at least 3000 years beofre it became a problem. Or something like that. It amused me greatly at the time.
It makes no real difference since the roots have to be scaled to be able to survive a sustained DDoS attack for at least as long as it takes remediation measures to kick in. Get rid off all the bozo queries and you still need the same size box because of the script kiddies.
Um, during a DDoS attack the servers have to handle the normal load which includes the bozo queries. If you eliminate the bozos you reduce the peak demand load (DDoS + other traffic).
That may well not matter much since the DDoS may be 100x the normal load, so getting rid of the bozo queries would only let you save 1% of the peak demand (and 98% of the normal demand).
As for complaints...why? Unless a device is regulating some biological function neccesary for your continued existance as a living being, surely you can live without it for the 15-20 minutes it takes to take off and land. Just sit back, relax, read a magazine, say hello/goodbye to your neighbor, or find something else to do that isn't battery powered.;)
The time around take off and touchdown are the most interesting to take photos (or I assume movies), and almost all modern cameras have eletronic autofocus and lightmeteres (unless you have a Nikon FM3a at least...and eve there you have to remember to turn off the lightmeter). Also unless there is a ton of light one probbably wants to use an image stabalised lens, so that is also eletronic...
Yeah, maybe not a huge deal to most people, but I really like the view, and would rather frame a few then keep buying plane tickets to get it:-)
Another way to look at this is that people are willing to spend a lot more to make sure that their bridges and airplanes are well-designed.
The money simply isn't there to do the same thing for software.
Money and time. The only software engenering project that I know of that has been aproached like this is the space shuttle's software. It has had zero bugs detected during a mission. It has had a small (like less then a dozen) bugs that were found in deployed code though.
The rather good article on the Space Shuttle code that I read years ago (in Fast Company??) one of the project members said "there is no reason we couldn't do with for MS Word...but the package would cost $36,000 a copy, and it would take 8 years to build even after throwing out most of the features -- who would buy that?"
I expect you can find similar defect rates in the embeded medical software world, and other places where a bug would kill people (or in the millatary where a bug might fail to kill the Bad People)... other places where the wait and cost for "bug free" code can't be managed...well, you won't see it.
You might come close in an airline reservation system though, or other things that have downtime worth milluions a minute... unless deploying the newer code can increse the $/min value enough to make the chance of having minutes without $ worth it... hmmmm..
Re:Today's diesel engines are WAY better
on
239 MPG Car
·
· Score: 1
While I do not doubt this at all, at least on the 2002 VW Golf TDI I have, one of the ways VW cut down on the clattering isusing a rubber/poly timing belt rather than a timing chain. It makes the engine much quieter, but presents the possibility of a maintenance horror story if that belt fails. It's supposed to last 60,000 miles but there's no way I'm going to let it go that long.
That isn't new for Golf-type bodies, I had a US built '79 VW Rabbit, and it had the rubber timing "chain". It broke somewhere around 100k miles. Supprisingly it didn't destroy the whole engine, but it was a bitch removing enough stuff to change the belt (the VW approved method involved removing the engine from the car, if I recall corectly we managed to get by with removing parts of it, and susspendign it on stout boards...).
However...what does a desal use a timing chain for? There aren't spark plugs with carefully timed sparks fireing! What is the deal with timing on a diesel?
None of the code involved should be special to Apple in any way
Mac OSX by default uses HFS+ rather then FFS, so there is a lot of Apple-specific code getting executed in there. Maybe they don't do namei cache invalidation correctly in their HFS+ file system code (for example).
Not a huge unforgivable bug to have, but one hopes they will try to fix it quickly. It would definitly re-enforces my opnion of OSX as very stable for a desktop OS, but not very stable as a server OS. Which is why I own an Apple laptop, but not an Apple rackmount computer;-)
However if they don't fix this kind of bug fast they are less likely to sell Xserve systems...
...not that Sun didn't have a bug where if you ftruncate'd/dev/audio you got a panic for something like five years! Sure that is a little less serious because you could deny users access to/dev/audio on a share machine and not suffer, but still... and I think it worked on any streams object that lived int he file system, so....
...but it would be nice if Apple proved themselves to be better then that.
Apple would have been better using someone else's kernel -- like the FreeBSD one -- verbatim instead of trying to hack up that Mach POS (which I've heard nothing but bad things about from OS people up at Carnegie Mellon, where Mach was developed).
Well MACH isn't exactly an OS, it is more of an OS for running OSes, and one of the OSes it can run is the "BSD Single Server" which is a BSD4.3+/4.4ish derieved OS that isn't in my opnion as good as some of the other BSD4.4ish derived OSes (like FreeBSD).
One of the other OSes that runs under MACH is a modifyed MacOS9. I havn't run OS9 (aka "Classic") on purpose for months, but other people find it rather indepsnsable, and wouldn't use OSX without it.
As you say they could plop Carbon and Quartz ontop of FreeBSD just as easally as onto MACH's BSD Single Server. However getting OS9 to "run under" FreeBSD would have been a much larger pain.
Of course, I certainly could be missing something -- could be that the FBSD kernel just didn't do something that Apple had to have it do, and that the FBSD people wouldn't have accepted. Could be SMP issues, I suppose...
I doubt it is SMP issues. I'm not even sure the FreeBSD people would reject the stuff needed to get OS9-under-FreeBSD working, after all it might not be that different from what WINE needs from the kernel...but it would have taken a whole lot more time then getting OS9 running under MACH more or less along side the BSD Single Server (kind of under it and off to one side I susspect...)
the device driver model is also different, and in a lot of ways better (and unfortunitly in a lot of ways worse) then FreeBSD.
I'd say it's a symptom of a missing feature. Obviously people want to be able to adjust their preferences, and don't have an interface for doing so. If the tvio had a slider for 'gay', and 'military' etc, which could be adjusted by the user, then that would give the user the ability to decrease the likelyhood of programs they don't want to be downloaded.
The older TiVo software had this as an undocumeted backdoor feature (you could assign thumps up/down to writers, actors, show types and all, not just shows). It made my suggestions a bit better, but as far as I can tell thumbing shows really does the same thing (gives a thumb to the actors and direcotrs and all). I have heard it is back (still undocumeted, and nobody has figured out how to get to it, but string'ing the binary shows bits of it's UI).
Still TiVo doesn't do too bad for me. Stuff pops up that I don't like, I hit 'em with some thumbs down. I think the big problem is something like I watched one or two cartoons so it thought maybe I would like cartoons (reasonable), but it seems like it only redoes the list o' possiable suggestions every once in a while, so it tried to give me cartoons for a few days before it got the message. Thumbs down seem to have less effect then thumbs up (well a thumbs down'ed show doesn't get recorded again, but a thumb up seems to get similar shows while a thumb down doesn't do so much to discurage -- not supprising, most people that like sitcoms don't like all sitcoms!).
It doesn't really bug me since the suggestions are "free" anyway, but it does seem to bother other people.
Your average DVD costs 15-20 dollars. So does the average premium CD. A movie is generally at least an hour and a half long, plus bonus materials. A CD is usually not much more than an hour long. And a movie is a much more engaging experience with video and surround sound. If consumers have limited dollars to spend on entertainment they will pick the better value.
Most people watch movies a small number of times, but listen to music a large number of times. I think I watch movies an unusually large number of times, I may have seen Apollo 13 20 times in the last 2 or so years. Most movies a lot fewer times. I have listened to a lot of the CD's I own (or MP3s of them) way way way more times. I'm positive I've listened to oh, say Tweekend more times then I have watch Apollo 13...and I'm not nearly as fond of Tweekend!
Why? Well few people will watch a movie and do much else at the same time. Many people will listen to music and do something else (drive, walk, run).
I don't know if that makes music "of more value", but I do know that kind of dashes the DVD's being more valuable strictly because they are a few hours vs. a bit under one hour.
I've personally never seen an airplane power port in coach class yet.
US Air has them on their A333's, British Midlands's does not. So like you say definitely not ubiquitous, but they ain't non-exsistant. I'm thinkin' of this battery because the BMI flights are cheaper and much faster...but I know a paper back and an iPod is a cheaper solution (since I already own the iPod).
More importantly, why would ANYBODY want to implement the x86 ISA (Instruction Set Architecture or smtn like that). It's the most horrid instruction set in use today.
Because that is where most of the desktop CPU money is going, some of the high end, and frighteningly enough a fair bit of embedded CPU money too.
In short if you can navigate the patent mine field, the brutal competition mine field, and deal with the instruction set making things a royal bitch doing an x86 CPU is a total no-brainer.
In Pentiums and Athlons, the instruction set isn't really emulated. It's translated to a smaller instruction set (uops, iops, pick whatever term you like and run with it). However, these smaller sets are still made pretty much specifically to cover the overlying ISA (x86 in this case).
Other then needing a whole new decoding front end, and being forced to use a trace cache because decoding multiple instructions in x86 land is very hard... the instruction thing isn't a big deal. Handling the odd-ball 80 bit FP format is. So is emulating all of the trap stuff and the other little odd bits close to the instructions set (like the MMU).
A big pain. But with much of the effort not being where folks think it is!
Speaking as an American who visited the UK recently...
Speed cameras everywhere
Apparently they have to put signs up for the cameras, which makes them way way more bearable. In fact I would rather deal with that then all the unmarked police cars, and the cop hiding nooks in interstates we have here!
Draconion speed limits (less than 35 km/h in some places)
On the other hand they have pretty reasonable speed limits on their major highways, like 70 to 75mph, as opposed to here where there are 55mph limits in most places (at least in VA and MD!), but traffic tends to move at 70MPH!
The UK also tends to enforce the "pass only on the right" laws while here "faster traffic stay left" is totally ignored.
Insane fuel tax, (costs around 75p/litre)
Every time I tried to work it out in My head it came to $4/gal or so. Pretty amazing. It makes the smaller cars thing more understandable. The Ford Fiesta is popular there. It was removed from the USA market because nobody wanted a 3cyl engine here.
High Taxes
That is a more generic thing then targeting autos. Their government provides more services, so it needs more money. Now I happen to beleve the government is the least efficient way to provide services, so that leads to higher costs then doing it privately. It may well lead to more universal service though. It also tends to lead to lower quality service for various economic reasons.
Speed bumps, chicanes, etc
What is/are "chicanes"?
Plans for congestion charges/road tolls in cities
Toll charges are actually a pretty good way to deal with congestion. Especially if the money from the tolls is applied to make underlying problem better (more roads, more clover leaves, fewer stop lights, or better public transportation). Of corse toll road money in most countries just goes into a general fund and ends up paying for totally unrelated things.
I will note that the UK has way better public transportation then at the very least my part of the USA!
Now, if Apple would have come out with such a beast in 1995 when I switched my desktop over to Linux then there is no question that I would have been thrilled.
Well Apple did. Except it was NeXT, and it was 1990, but it was close enough. Had real apps and stuff. A real desktop, and a real Unix even. Now 12 years later it's almost the same OS, but in color, with the ability to run old MacOS code, and people are all excited. (and, hey, I'm one o f 'em, but I like the NeXT too)
Don't tell me that the version of MS Word for Mac OS X is better than the Windows version either.
Well it is, even Microsoft says it is. However I expect they are only talking about the GUI, not the stuff under it. You never know though, if there is an Apple store close by go and check. They will let you load in the book and give it a shot (put it on CD). Somehow I think it'll blow goats though:-(
I don't think there is a WordPerfect. On the other hand there is Claris' MacWrite, and before you dis it because it is part of a $79 office suite, I'll tell you in 1986 I worked for a company that did two 400 page manuals in MacWrite, and it had lots of layout, and changes to custom fonts (it was an APL manual, and APL uses it's own charactor set...). It may have gotten worse in the 15 years since then...but you know what? The original version still runs. On OSX, on a different CPU.
Or you could give Quark Express a shot, I don't think it is normally used for tons of pages, but it is a serious layout tool, many magazines use it.
However, I think maybe Lyx would be best. I mean it's LaTeX with a GUI, right?
This article was about UNIX users switching to the Mac, and I just don't see the draw.
Well it's a Unix machine other people can use? Oh, wait, that just means other people will fight you over it. Not so good.
Hmmmm, well last time I tryed to run Unix (other then OSX) on a laptop it didn't do so good. I mean when I buy a desktop I make sure it only has good parts with drivers for the Unixlike OS I'm going to run. On a laptop it is way harder to control everything. Buy a Mac laptop, any of the current ones, and many of the old ones, and it all works. That's nice. To me it's worth paying the extra that a Mac laptop costs. Plus, well, the TiBook really does rock.
As far as the desktops go? Well they seem over priced to me. I mean, sure, being able to pop open the side and swing the motherboard down and not be in a maze of cables is nice. Being able to slide in a new drive is nice. Not cutting yourself on something every time you fiddle with the machine, that's just nice. But I don't do that real frequently (I mean it would get in the way of my uptime!), so I don't see payin' a whole bunch extra for that!
Now a whole bunch of the place I work for uses MS Office, and can't be dissuaded from it. It is nice being able to run Office and look at the crap they send. That's hard to do on most Unixes, but not OSX. Of corse that costs more, but...
Hmmmm, what else? Well it does worth with dirt cheap USB photo printers, and it has nice photo manipulation software. Er, what do you mean I'm the only geek with a digital SLR? Ah well, not compelling to the masses then:-)
To be honest while I love free software, and even write it frequently, sometimes being able to shell out some money and get a nicely polished chunk o' software is nice. I can do that in OSX, I can't do that very much in Linux or FreeBSD.
It's even nicer in OSX because not only can I pay out $700 of PhotoShop, I can use the GIMP. I can keep using it until I'm sure PhotoShop is $700 better. So far I'm still using it. If I had Linux on a Wintel box I wouldn't really get to change my mind! It wouldn't be $700 for PhotoShop unless I could stomach Windows to run it under. Stomach Windows and not have any services running on the Unix box other stuff depends on. Plus maybe have to buy Windows. Ick! Nope, I would rather overpay on the Mac then buy Windows!
Sure I write real servers in real (and painful!) C++. I hack together a ton of Perl scripts. But....sometimes...I just want to fire up Civ III, and should I have to boot windows for that? Sometimes I want a web browser that renders most of the web correctly, should I have to fire up windows for that? (er, and Mozilla didn't work good 'nuf when I bought the Mac...now it's the browser I use most...'cept when I use OmniWeb). Sometimes my cable ISP dies and I have to convince them to fix it. For that I need to run a supported OS. Guess what, "Unix" of all flavors is absent from that list, except...OSX. Those are all good reasons to have a Mac.
There are good reasons not too. My Mac is outnumbered 5 to 1 by non-Mac Unix boxes in my house. 3 to 1 if you ignore the Unix machines managing my TV. Maybe 15 to 1 if you count all the computers I own but am not using! (my C=64 is in a box somewhere...and I insist that it is a computer, even if it is so outclassed by not only my Palm, but even my cell phone that it's not funny!)
Any you know what? Those arguments don't have to convince you to buy a Mac, just get you to see that people that do buy one might not be nuts.
Oh, and your mom needs a Mac. Now. Go get her one. You do love her don't you? Come on now...
Er, sorry, according to Apple's EULA I have to let Steve Jobs control my body 5 minutes a year. Hey, it's still better then what Bill wants....
The hardware in the two boxes will be very comparable, with the edge probably going to Dell. After all, Apple uses commodity hard drives, video cards, sound cards, and memory just like Dell does.
My mother has a name brand PC, it cost about $1000 when it was new. My mother-in-law has an iMac. It cost about the same.
The monitor on the iMac is way way way sharper, and edges and corners can be used.
The built in speakers on the iMac while they suck suck less then the speakers on the PC. I expect the sound hardware on the iMac is better too, but I don't know 'cause the speakers on the PC hide it.
Other then that, I don't see a reason the hard drive on the Mac would work better, or the RAM.
But hey, who cares about all that crap. The absolute most important thing? I get next to zero help calls about the Mac. It Just Works. Really. Honest. When they buy hardware that has a Mac sticker on it and plug it in the it doesn't screw up all the existing settings. They don't seem to get a bizzilion little auto-start crap-lets every few months. They don't end up with some commercial software they buy overwriting half a dozen important system files with some other version of the files an having stuff no longer work.
In short the Mac does the most important thing possible: it doesn't screw up as much as a windows box.
To me it is worth the extra money to hear from my relatives less. Or in a less cynical mind, to hear them talk about interesting stuff when I hear from them, not about computer problems.
Now maybe you want the fastest CPU in Mhz, I just want the one that "does the job" the fastest. "Does the job" includes time for the user to figure out how to do the job, and the time lost if it crashes part way through. For me "does the job the fastest" is frequently a Unix box. I mean if I do it a lot, I probably already wrote a program to do it, and I've been using Unix forever, so that'll be a Unix program. I'm not most people though. Most people can (gasp!) get stuff done faster on a box that coddles them. So a Mac or a Wintel box. And of the two? It seems the Mac really does a better job way more offen then people think.
Don't beleve me? I tell you what, for the price difference between my in-law's iMac, and my mom's PC will you take her tech support calls?
I think you'll find that the variations aren't so much variations from Unix, but from Linux. Many of the differences the article outlines are simply "hiding" the Unix from newbies (i.e. dangerous directories) and can easily be over ridden.
That's more or less true. If you ignore the fact that pretty much every Unix system that has a GUI except Apple's uses X11 the differences from "Apple's Unix" to anyone else's isn't really any bigger then the differences between any two other Unix-like systems. Sure Apple uses NetInfo, but it really isn't any different from Sun's YP or NIS. Yes, Apple has a ton of GUI admin tools that whizz all over/etc, but what is IBM's SMIT? Or HP's...er...what does HP call their admin tools again?
If you are talking about command line tools, Mac OS X is "just another Unix", period. One of the less common ones, so you may not find as many things compiling out of the box, but that isn't because OSX is more different from whatever Unixish system the author used (most likely Linux these days) then, say NetBSD or SunOS is, but just that whatever 3 random things that always seem to trip people up when going to a new platform weren't already spotted and fixed.
I remember when SunOS was king, and it was a slight pain to port stuff to Ultrix (DEC's Unix). This is no harder. Straight down to programs sometimes forgetting a htonl or the like.
Once you get to GUI's then it's a whole different thing (unless you remember when Suns came with Sun Tools, DEC had X11, AT&T had the BLiT, and everyone else had their own thing too). OSX is way different from other Unix-like systems. You could install X11 on it, but X apps will never feel like native apps, and most apps that are written for OSX that you might want to modify won't be using X. Then again, it's nice to learn a new thing once in a while, isn't it?
Although Aliases are more powerful, but most Unix tools don't recognize them
Don't worry about satellite radio as the providers have some large pockets to draw upon - the auto makers.
Um....well, the auto makers may have big pockets, but they have a deal to put receivers in their cars, one assumes they got the year's programming for free, and the receivers at a cheaper rate...not a deal to give XM and all money.
Next year, GM will offer the service as standard equipment on some of the vehicles - with a free year of service. If only a small percentage renew, then the satellite providers will be listening to satellite disco. If a significant percentage renew, then you are looking at a threat to FM.
Yes, but there is no money in the short term, or even the medium term. Let's see, starting sometime next year XM gives free service so lots of people with new GM cars. That doesn't cost XM very much money (they might have product support costs, but they should be low, right?). For a year after that they have no extra money from those people. Then, after over a year they can start getting an idea about how many people keep the subscription.
Think about that.
We are still over a year from them getting money from the GM deal, and more importantly over a year from finding out how much money they will get from the GM deal. Do they have money to make it that long without going into bankruptcy?
On that note, with all these satellite head units running around in the new autos, it would make financial sense to provide some publically funded stations free-of-charge. That would be the largest opportunity.
How does that makes sense (to XM, not the listening public!)? It wouldn't make money for XM, and it would make the XM subscription slightly less valuable in as much as people that like the publicly funded stations would then have something they could listen to "whenever" without paying XM for anything!
Another point: you are oversimplifying when you say that we'd need to squash 3*25 instructions on a misprediction, it could be more, it could be less.
Yup, I'm oversimplifying wildly. There are also non-BTC effects I left out. However, what would cause more then 3*25 instructions to be thrown away? Things living in the register rename buffers? If so how many? If there a separate re-order buffer later in the pipe, or is that done with the register rename ones?
Or you could store your own syscalls table for each process. At exec(), you init the table so that all syscalls are redirected to your handler, which then performs the overhead of checking the policy. If the syscall is allowed under the policy, then the process syscall map would be updated with the proper syscall information. Future calls to the same syscall would be passed immediately onto the kernel with no more overhead than your normal syscall mechanism.
Cool idea, and I know FreeBSD at least supports per-process syscall tables, that is how the Linux and BSD/OS "emulation" works. However having more syscall tables does have some real overhead, or could. When you do an indirect jump (table base address plus 4 times the syscall number is an indirect jump) if the jump can't be predicted you end up flushing the pipeline. Tieing that in with other slashdot articles that is 25 pipe stages times three instructions (max) for the P-III or 75 instructions worth of work or 6 cycles time 3 (or is it two?) instructions so 18 instructions for a PPC G4.
Having another table per process may make the branch prediction much harder since you'll have many more addresses for the branch target cache to hold. Er, maybe. It won't make a difference if the BTC has to be flushed when there is a process context changed (many CPUs have something like a PID that can be used for the different caches to reduce that kind of churn). Also if the CPU does the BTC based off of the physical address this may not be such a big deal (at least if an effort is used to share the tables).
the question is, will Altivec code compiled on a G4 slow down on this new chip?
All indications are the new CPU has AltiVec which probbably means AltiVec code on it is not slower then non-AltiVec code, and one hopes faster by about the same mesure as it is on the existing G4.
I don't understand how these vector instructions are implemented well enough (are they in the OS as special calls or passes as flags to the compiler which generates different machine code out of it?
Well they are implmented in Silicon, but somehow I think that is not what you really want to know:-)
The compiler can be given a flag that asks it to gennerate AV code if it can, but I don't think this helps a whole lot because compilers are not that good at finding places to use SIMD instructions. Esp. not in C code (as Cray about that kind of thing some day). One can also hand code the assembly, but that isn't so common these days either. Most people use C compiler extensions that look like normal functions. I beleve normal practice is to write one AltiVec version and one stright line version because the stright line version is faster then pretending to have AltiVec and packing and unpacking results "by hand".
Many of Apples library calls, and maybe OS calls will use AltiVec if your CPU has it. Kind of depends on what you define as "the OS". If the GUI stuff is "the OS" then your set. I don't know if the IP checksum code, or bzero(2) use it or not.
Ummm no, actually PowerPC is a subset of POWER, not an extention. PowerPC code would run on a POWER, not nessessarily vice versa.
It is a non-strict subset plus a non-strict superset. In other words they removed some POWER instructions (like the fused multiply-add) and added other (PPC has a whole set of single percision FP instructions). There are other POWER extentions PowerAS for example (which I think is strictly a superset for 256 bit addressing, but I'm not positave!).
Of corse many POWER CPUs implment the PowerPC instruction set also, but there is no requirment what so ever that they do so! The POWER4 does at least POWER, PowerPC, and PowerAS. However...
...the PowerPC instruction set doesn't include the AltiVec SIMD instructions, not technically. So you can have a fast PowerPC CPU that is mostly useless to Apple because while Apple doesn't depend on AltiVec (they run on the G3 after all which has no AltiVec!), they really really run faster with it. A 1.8Ghz AltiVec-less CPU may well run signifigantly slower then a 667Mhz AltiVec CPU for some commonish tasks (MP3 encoding for example, and some screen effects).
Slashdot Mirror
Some of the APM stuff is done by the BIOS in a way that is very hard for an OS to interfere with, or at least to alter other then preventing it. Some OSes use the "BIOS" (actually video card extensions to it) to change vid card modes. An OS could use it to do disk I/O, but then you can't do async I/O which is why no modern OS does it that way.
Sun uses their equivoent of the BIOS (OpenPROM I think) more, it actually does the console screen I/O (and use to suck because they read it out of 8 bit wide ROM with no caching, but that is over with now). Sun didn't make it palatable to do disk I/O or network I/O via OpenPROM though. "New World" Macs uses the same OpenPROM stuff Sun does, but I don't think they use any of it after the boot anyway.
Not exactly true. There isn't a way to try in most cases. You can't try to alter another processes memory because you can't express that except via an OS construct, and that ones doesn't kill you if it doesn't work, just returns an error code. Sometimes the response is more severe then an error code -- a signal. But almost never is your process just killed...not in any normal Unix (I think OpenBSD will panic for some security breaches, like a second chroot in the same process hierachy...but it might just do the equivoent of kill -9).
The text messaging on some phones is quite easy to use, and frequently costs less money then a short call...and is a lot less intrusave to recieve (and sometimes make). So for me, text messaging via the keypad is a win.
I can't say I've really found web browsing on a phone all that useful. Except once. I had taken a walk in a show storm and managed to get lost on twisty little roads. Mapquest even on the tiny phone was quite useful. A GPS might have been better, but I didn't have one. That was 3 years ago or so, I havn't had great use for a web browser in my phone since.
I don't think this new addressing scheme will have anything to do with how we use cell phones though, just how we use computers to talk to cell phones...
Plus, isn't this just Carl Malamute's tpc.int all over again? Same thing, revers the digit order, put dots between them...
I remember seeing a Sun press relase about Java being Y2K complient, how long it would last, and that Sun promised to fix it at least 3000 years beofre it became a problem. Or something like that. It amused me greatly at the time.
Um, during a DDoS attack the servers have to handle the normal load which includes the bozo queries. If you eliminate the bozos you reduce the peak demand load (DDoS + other traffic).
That may well not matter much since the DDoS may be 100x the normal load, so getting rid of the bozo queries would only let you save 1% of the peak demand (and 98% of the normal demand).
The time around take off and touchdown are the most interesting to take photos (or I assume movies), and almost all modern cameras have eletronic autofocus and lightmeteres (unless you have a Nikon FM3a at least...and eve there you have to remember to turn off the lightmeter). Also unless there is a ton of light one probbably wants to use an image stabalised lens, so that is also eletronic...
Yeah, maybe not a huge deal to most people, but I really like the view, and would rather frame a few then keep buying plane tickets to get it :-)
Money and time. The only software engenering project that I know of that has been aproached like this is the space shuttle's software. It has had zero bugs detected during a mission. It has had a small (like less then a dozen) bugs that were found in deployed code though.
The rather good article on the Space Shuttle code that I read years ago (in Fast Company??) one of the project members said "there is no reason we couldn't do with for MS Word...but the package would cost $36,000 a copy, and it would take 8 years to build even after throwing out most of the features -- who would buy that?"
I expect you can find similar defect rates in the embeded medical software world, and other places where a bug would kill people (or in the millatary where a bug might fail to kill the Bad People)... other places where the wait and cost for "bug free" code can't be managed...well, you won't see it.
You might come close in an airline reservation system though, or other things that have downtime worth milluions a minute... unless deploying the newer code can increse the $/min value enough to make the chance of having minutes without $ worth it... hmmmm..
That isn't new for Golf-type bodies, I had a US built '79 VW Rabbit, and it had the rubber timing "chain". It broke somewhere around 100k miles. Supprisingly it didn't destroy the whole engine, but it was a bitch removing enough stuff to change the belt (the VW approved method involved removing the engine from the car, if I recall corectly we managed to get by with removing parts of it, and susspendign it on stout boards...).
However...what does a desal use a timing chain for? There aren't spark plugs with carefully timed sparks fireing! What is the deal with timing on a diesel?
Mac OSX by default uses HFS+ rather then FFS, so there is a lot of Apple-specific code getting executed in there. Maybe they don't do namei cache invalidation correctly in their HFS+ file system code (for example).
Not a huge unforgivable bug to have, but one hopes they will try to fix it quickly. It would definitly re-enforces my opnion of OSX as very stable for a desktop OS, but not very stable as a server OS. Which is why I own an Apple laptop, but not an Apple rackmount computer ;-)
However if they don't fix this kind of bug fast they are less likely to sell Xserve systems...
...not that Sun didn't have a bug where if you ftruncate'd /dev/audio you got a panic for something like five years! Sure that is a little less serious because you could deny users access to /dev/audio on a share machine and not suffer, but still... and I think it worked on any streams object that lived int he file system, so....
...but it would be nice if Apple proved themselves to be better then that.
Well MACH isn't exactly an OS, it is more of an OS for running OSes, and one of the OSes it can run is the "BSD Single Server" which is a BSD4.3+/4.4ish derieved OS that isn't in my opnion as good as some of the other BSD4.4ish derived OSes (like FreeBSD).
One of the other OSes that runs under MACH is a modifyed MacOS9. I havn't run OS9 (aka "Classic") on purpose for months, but other people find it rather indepsnsable, and wouldn't use OSX without it.
As you say they could plop Carbon and Quartz ontop of FreeBSD just as easally as onto MACH's BSD Single Server. However getting OS9 to "run under" FreeBSD would have been a much larger pain.
I doubt it is SMP issues. I'm not even sure the FreeBSD people would reject the stuff needed to get OS9-under-FreeBSD working, after all it might not be that different from what WINE needs from the kernel...but it would have taken a whole lot more time then getting OS9 running under MACH more or less along side the BSD Single Server (kind of under it and off to one side I susspect...)
the device driver model is also different, and in a lot of ways better (and unfortunitly in a lot of ways worse) then FreeBSD.
The older TiVo software had this as an undocumeted backdoor feature (you could assign thumps up/down to writers, actors, show types and all, not just shows). It made my suggestions a bit better, but as far as I can tell thumbing shows really does the same thing (gives a thumb to the actors and direcotrs and all). I have heard it is back (still undocumeted, and nobody has figured out how to get to it, but string'ing the binary shows bits of it's UI).
Still TiVo doesn't do too bad for me. Stuff pops up that I don't like, I hit 'em with some thumbs down. I think the big problem is something like I watched one or two cartoons so it thought maybe I would like cartoons (reasonable), but it seems like it only redoes the list o' possiable suggestions every once in a while, so it tried to give me cartoons for a few days before it got the message. Thumbs down seem to have less effect then thumbs up (well a thumbs down'ed show doesn't get recorded again, but a thumb up seems to get similar shows while a thumb down doesn't do so much to discurage -- not supprising, most people that like sitcoms don't like all sitcoms!).
It doesn't really bug me since the suggestions are "free" anyway, but it does seem to bother other people.
Most people watch movies a small number of times, but listen to music a large number of times. I think I watch movies an unusually large number of times, I may have seen Apollo 13 20 times in the last 2 or so years. Most movies a lot fewer times. I have listened to a lot of the CD's I own (or MP3s of them) way way way more times. I'm positive I've listened to oh, say Tweekend more times then I have watch Apollo 13...and I'm not nearly as fond of Tweekend!
Why? Well few people will watch a movie and do much else at the same time. Many people will listen to music and do something else (drive, walk, run).
I don't know if that makes music "of more value", but I do know that kind of dashes the DVD's being more valuable strictly because they are a few hours vs. a bit under one hour.
US Air has them on their A333's, British Midlands's does not. So like you say definitely not ubiquitous, but they ain't non-exsistant. I'm thinkin' of this battery because the BMI flights are cheaper and much faster...but I know a paper back and an iPod is a cheaper solution (since I already own the iPod).
Because that is where most of the desktop CPU money is going, some of the high end, and frighteningly enough a fair bit of embedded CPU money too.
In short if you can navigate the patent mine field, the brutal competition mine field, and deal with the instruction set making things a royal bitch doing an x86 CPU is a total no-brainer.
Other then needing a whole new decoding front end, and being forced to use a trace cache because decoding multiple instructions in x86 land is very hard... the instruction thing isn't a big deal. Handling the odd-ball 80 bit FP format is. So is emulating all of the trap stuff and the other little odd bits close to the instructions set (like the MMU).
A big pain. But with much of the effort not being where folks think it is!
Speaking as an American who visited the UK recently...
Apparently they have to put signs up for the cameras, which makes them way way more bearable. In fact I would rather deal with that then all the unmarked police cars, and the cop hiding nooks in interstates we have here!
On the other hand they have pretty reasonable speed limits on their major highways, like 70 to 75mph, as opposed to here where there are 55mph limits in most places (at least in VA and MD!), but traffic tends to move at 70MPH!
The UK also tends to enforce the "pass only on the right" laws while here "faster traffic stay left" is totally ignored. Insane fuel tax, (costs around 75p/litre)
Every time I tried to work it out in My head it came to $4/gal or so. Pretty amazing. It makes the smaller cars thing more understandable. The Ford Fiesta is popular there. It was removed from the USA market because nobody wanted a 3cyl engine here.
That is a more generic thing then targeting autos. Their government provides more services, so it needs more money. Now I happen to beleve the government is the least efficient way to provide services, so that leads to higher costs then doing it privately. It may well lead to more universal service though. It also tends to lead to lower quality service for various economic reasons.
What is/are "chicanes"?
Toll charges are actually a pretty good way to deal with congestion. Especially if the money from the tolls is applied to make underlying problem better (more roads, more clover leaves, fewer stop lights, or better public transportation). Of corse toll road money in most countries just goes into a general fund and ends up paying for totally unrelated things.
I will note that the UK has way better public transportation then at the very least my part of the USA!
Well Apple did. Except it was NeXT, and it was 1990, but it was close enough. Had real apps and stuff. A real desktop, and a real Unix even. Now 12 years later it's almost the same OS, but in color, with the ability to run old MacOS code, and people are all excited. (and, hey, I'm one o f 'em, but I like the NeXT too)
Well it is, even Microsoft says it is. However I expect they are only talking about the GUI, not the stuff under it. You never know though, if there is an Apple store close by go and check. They will let you load in the book and give it a shot (put it on CD). Somehow I think it'll blow goats though :-(
I don't think there is a WordPerfect. On the other hand there is Claris' MacWrite, and before you dis it because it is part of a $79 office suite, I'll tell you in 1986 I worked for a company that did two 400 page manuals in MacWrite, and it had lots of layout, and changes to custom fonts (it was an APL manual, and APL uses it's own charactor set...). It may have gotten worse in the 15 years since then...but you know what? The original version still runs. On OSX, on a different CPU.
Or you could give Quark Express a shot, I don't think it is normally used for tons of pages, but it is a serious layout tool, many magazines use it.
However, I think maybe Lyx would be best. I mean it's LaTeX with a GUI, right?
Well it's a Unix machine other people can use? Oh, wait, that just means other people will fight you over it. Not so good.
Hmmmm, well last time I tryed to run Unix (other then OSX) on a laptop it didn't do so good. I mean when I buy a desktop I make sure it only has good parts with drivers for the Unixlike OS I'm going to run. On a laptop it is way harder to control everything. Buy a Mac laptop, any of the current ones, and many of the old ones, and it all works. That's nice. To me it's worth paying the extra that a Mac laptop costs. Plus, well, the TiBook really does rock.
As far as the desktops go? Well they seem over priced to me. I mean, sure, being able to pop open the side and swing the motherboard down and not be in a maze of cables is nice. Being able to slide in a new drive is nice. Not cutting yourself on something every time you fiddle with the machine, that's just nice. But I don't do that real frequently (I mean it would get in the way of my uptime!), so I don't see payin' a whole bunch extra for that!
Now a whole bunch of the place I work for uses MS Office, and can't be dissuaded from it. It is nice being able to run Office and look at the crap they send. That's hard to do on most Unixes, but not OSX. Of corse that costs more, but...
Hmmmm, what else? Well it does worth with dirt cheap USB photo printers, and it has nice photo manipulation software. Er, what do you mean I'm the only geek with a digital SLR? Ah well, not compelling to the masses then :-)
To be honest while I love free software, and even write it frequently, sometimes being able to shell out some money and get a nicely polished chunk o' software is nice. I can do that in OSX, I can't do that very much in Linux or FreeBSD.
It's even nicer in OSX because not only can I pay out $700 of PhotoShop, I can use the GIMP. I can keep using it until I'm sure PhotoShop is $700 better. So far I'm still using it. If I had Linux on a Wintel box I wouldn't really get to change my mind! It wouldn't be $700 for PhotoShop unless I could stomach Windows to run it under. Stomach Windows and not have any services running on the Unix box other stuff depends on. Plus maybe have to buy Windows. Ick! Nope, I would rather overpay on the Mac then buy Windows!
Sure I write real servers in real (and painful!) C++. I hack together a ton of Perl scripts. But....sometimes...I just want to fire up Civ III, and should I have to boot windows for that? Sometimes I want a web browser that renders most of the web correctly, should I have to fire up windows for that? (er, and Mozilla didn't work good 'nuf when I bought the Mac...now it's the browser I use most...'cept when I use OmniWeb). Sometimes my cable ISP dies and I have to convince them to fix it. For that I need to run a supported OS. Guess what, "Unix" of all flavors is absent from that list, except...OSX. Those are all good reasons to have a Mac.
There are good reasons not too. My Mac is outnumbered 5 to 1 by non-Mac Unix boxes in my house. 3 to 1 if you ignore the Unix machines managing my TV. Maybe 15 to 1 if you count all the computers I own but am not using! (my C=64 is in a box somewhere...and I insist that it is a computer, even if it is so outclassed by not only my Palm, but even my cell phone that it's not funny!)
Any you know what? Those arguments don't have to convince you to buy a Mac, just get you to see that people that do buy one might not be nuts.
Oh, and your mom needs a Mac. Now. Go get her one. You do love her don't you? Come on now...
Er, sorry, according to Apple's EULA I have to let Steve Jobs control my body 5 minutes a year. Hey, it's still better then what Bill wants....
So like hard links? (ln without the -s)
My mother has a name brand PC, it cost about $1000 when it was new. My mother-in-law has an iMac. It cost about the same.
The monitor on the iMac is way way way sharper, and edges and corners can be used.
The built in speakers on the iMac while they suck suck less then the speakers on the PC. I expect the sound hardware on the iMac is better too, but I don't know 'cause the speakers on the PC hide it.
Other then that, I don't see a reason the hard drive on the Mac would work better, or the RAM.
But hey, who cares about all that crap. The absolute most important thing? I get next to zero help calls about the Mac. It Just Works. Really. Honest. When they buy hardware that has a Mac sticker on it and plug it in the it doesn't screw up all the existing settings. They don't seem to get a bizzilion little auto-start crap-lets every few months. They don't end up with some commercial software they buy overwriting half a dozen important system files with some other version of the files an having stuff no longer work.
In short the Mac does the most important thing possible: it doesn't screw up as much as a windows box.
To me it is worth the extra money to hear from my relatives less. Or in a less cynical mind, to hear them talk about interesting stuff when I hear from them, not about computer problems.
Now maybe you want the fastest CPU in Mhz, I just want the one that "does the job" the fastest. "Does the job" includes time for the user to figure out how to do the job, and the time lost if it crashes part way through. For me "does the job the fastest" is frequently a Unix box. I mean if I do it a lot, I probably already wrote a program to do it, and I've been using Unix forever, so that'll be a Unix program. I'm not most people though. Most people can (gasp!) get stuff done faster on a box that coddles them. So a Mac or a Wintel box. And of the two? It seems the Mac really does a better job way more offen then people think.
Don't beleve me? I tell you what, for the price difference between my in-law's iMac, and my mom's PC will you take her tech support calls?
That's more or less true. If you ignore the fact that pretty much every Unix system that has a GUI except Apple's uses X11 the differences from "Apple's Unix" to anyone else's isn't really any bigger then the differences between any two other Unix-like systems. Sure Apple uses NetInfo, but it really isn't any different from Sun's YP or NIS. Yes, Apple has a ton of GUI admin tools that whizz all over /etc, but what is IBM's SMIT? Or HP's...er...what does HP call their admin tools again?
If you are talking about command line tools, Mac OS X is "just another Unix", period. One of the less common ones, so you may not find as many things compiling out of the box, but that isn't because OSX is more different from whatever Unixish system the author used (most likely Linux these days) then, say NetBSD or SunOS is, but just that whatever 3 random things that always seem to trip people up when going to a new platform weren't already spotted and fixed.
I remember when SunOS was king, and it was a slight pain to port stuff to Ultrix (DEC's Unix). This is no harder. Straight down to programs sometimes forgetting a htonl or the like.
Once you get to GUI's then it's a whole different thing (unless you remember when Suns came with Sun Tools, DEC had X11, AT&T had the BLiT, and everyone else had their own thing too). OSX is way different from other Unix-like systems. You could install X11 on it, but X apps will never feel like native apps, and most apps that are written for OSX that you might want to modify won't be using X. Then again, it's nice to learn a new thing once in a while, isn't it?
In what ways are aliases more powerful?
Um....well, the auto makers may have big pockets, but they have a deal to put receivers in their cars, one assumes they got the year's programming for free, and the receivers at a cheaper rate...not a deal to give XM and all money.
Yes, but there is no money in the short term, or even the medium term. Let's see, starting sometime next year XM gives free service so lots of people with new GM cars. That doesn't cost XM very much money (they might have product support costs, but they should be low, right?). For a year after that they have no extra money from those people. Then, after over a year they can start getting an idea about how many people keep the subscription.
Think about that.
We are still over a year from them getting money from the GM deal, and more importantly over a year from finding out how much money they will get from the GM deal. Do they have money to make it that long without going into bankruptcy?
How does that makes sense (to XM, not the listening public!)? It wouldn't make money for XM, and it would make the XM subscription slightly less valuable in as much as people that like the publicly funded stations would then have something they could listen to "whenever" without paying XM for anything!
I was, I even thought "P4" when I typed "P-III"!
Yup, I'm oversimplifying wildly. There are also non-BTC effects I left out. However, what would cause more then 3*25 instructions to be thrown away? Things living in the register rename buffers? If so how many? If there a separate re-order buffer later in the pipe, or is that done with the register rename ones?
Cool idea, and I know FreeBSD at least supports per-process syscall tables, that is how the Linux and BSD/OS "emulation" works. However having more syscall tables does have some real overhead, or could. When you do an indirect jump (table base address plus 4 times the syscall number is an indirect jump) if the jump can't be predicted you end up flushing the pipeline. Tieing that in with other slashdot articles that is 25 pipe stages times three instructions (max) for the P-III or 75 instructions worth of work or 6 cycles time 3 (or is it two?) instructions so 18 instructions for a PPC G4.
Having another table per process may make the branch prediction much harder since you'll have many more addresses for the branch target cache to hold. Er, maybe. It won't make a difference if the BTC has to be flushed when there is a process context changed (many CPUs have something like a PID that can be used for the different caches to reduce that kind of churn). Also if the CPU does the BTC based off of the physical address this may not be such a big deal (at least if an effort is used to share the tables).
Modern CPUs are a pain to optimize for.
All indications are the new CPU has AltiVec which probbably means AltiVec code on it is not slower then non-AltiVec code, and one hopes faster by about the same mesure as it is on the existing G4.
Well they are implmented in Silicon, but somehow I think that is not what you really want to know :-)
The compiler can be given a flag that asks it to gennerate AV code if it can, but I don't think this helps a whole lot because compilers are not that good at finding places to use SIMD instructions. Esp. not in C code (as Cray about that kind of thing some day). One can also hand code the assembly, but that isn't so common these days either. Most people use C compiler extensions that look like normal functions. I beleve normal practice is to write one AltiVec version and one stright line version because the stright line version is faster then pretending to have AltiVec and packing and unpacking results "by hand".
Many of Apples library calls, and maybe OS calls will use AltiVec if your CPU has it. Kind of depends on what you define as "the OS". If the GUI stuff is "the OS" then your set. I don't know if the IP checksum code, or bzero(2) use it or not.
It is a non-strict subset plus a non-strict superset. In other words they removed some POWER instructions (like the fused multiply-add) and added other (PPC has a whole set of single percision FP instructions). There are other POWER extentions PowerAS for example (which I think is strictly a superset for 256 bit addressing, but I'm not positave!).
Of corse many POWER CPUs implment the PowerPC instruction set also, but there is no requirment what so ever that they do so! The POWER4 does at least POWER, PowerPC, and PowerAS. However...
...the PowerPC instruction set doesn't include the AltiVec SIMD instructions, not technically. So you can have a fast PowerPC CPU that is mostly useless to Apple because while Apple doesn't depend on AltiVec (they run on the G3 after all which has no AltiVec!), they really really run faster with it. A 1.8Ghz AltiVec-less CPU may well run signifigantly slower then a 667Mhz AltiVec CPU for some commonish tasks (MP3 encoding for example, and some screen effects).
Complex innit? Welcome to IBM's world.