According to the June Top500 List, LLNL's third-ranked MCR Xeon cluster has an Rmax of 7634.00, and an Rpeak of 11060.00.
Rmax is the best it's attained; Rpeak is the theoretical peak in a perfect world with 100% multiprocessing efficiency.
One of the really amazing things about the NEC Earth Simulator (aside from its sheer power) is that its Rmax is over 85% of its Rpeak. A lot of other systems only get Rmax of 60-70% of Rpeak.
I've read that PowerPC chips are pretty efficient in SMP scenarios, and if "Big Mac" is turning in an Rmax that's 80% of Rpeak (even on a fairly small number of processors), that would seem to validate that assertion.
Isn't it safe to say if the cost of a 'server' or 'workstation' 64bit chip is cheap enough to replace a 'standard' chip in a PC then why can't they claim the right to the first Desktop PC (64bit)...just because it can be used as a server or workstation doesn't mean it can't do all of the needs of a PC.
If that's how you choose to define things, the first 64-bit desktop PC's were made over a decade ago.
Oh, and if one 64-bit PC architecture weren't enough, how about this one? Of course, back then, Windows NT also ran on the PowerPC architecture.
So... either the fact that the Opteron, like the MIPS and Alpha chips, was not designed or intended for use in "personal computers" means that Opteron-based workstations are not "personal computers" any more than Alpha- or MIPS-based ones were... or neither AMD nor Apple has the first "64-bit personal computer" by an entire decade.
(Incidentally, Windows NT 4.0 CD's -- I have one -- still contain code for those other architectures!)
Aswell I don't know if I agree with the statement that is has to be a basically "Dell" or some big name company build to be a PC... I have a 17inch monitor, speakers, headphones a mouse and keyboard and athlon 1600+... I call it a PC..and I built it from parts.
It doesn't -- it just has to be substantively identical to one. "Personal computers" are -- or are substantively identical to things that are -- mass-produced and marketed to consumers. You didn't see commercials on VH-1 for your system, and you didn't buy it at the mall or a big-box computer store -- but you (or others) have probably seen commercials for substantively identical kit on VH-1, and you could have bought something substantively identical to it in a mall or big-box store.
You cannot (to my knowledge) walk into the mall, or a big-box computer store, right now, and walk out with a BOXX workstation or something substantively identical, nor will you see commercials for it on VH-1. You will see commercials on VH-1 for Apple's G5, and you can walk into an Apple Store or possibly even a CompUSA store (I don't know who's gotten shipments of what, ya know?) and walk out one. (They'll be upset, though, if you don't pay.)
I realize, of course, that this all reeks of people with advertising budgets getting to declare that their products are whatever the heck they darn well please. Apple could have claimed that the G5 was a desktop, a personal computer, a workstation, a server, even a cheese grater, and since they're spending the bucks on commercials, that's what The Masses(tm) will probably believe.:)
The first desktop PC? Of course it wasn't. That was the IBM model 5150, a couple decades ago. Don't be silly.:)
Oh, wait, did you mean the first 64-bit desktop PC? Well now, that's a little harder, since it depends on what "PC" means.
PC, as we all know, stands for piece of cra...^W^W^W, um, er, stands for Personal Computer. Another nice vague term, but here's a crack at it:
If it's mass-produced and you can buy it (or something basically equivalent to it in terms of componentry) at the mall or your local big-box discount computer superstore, like BestCompCircuit BuyUSACity, take it home and run stuff like Quicken and AOL Instant Messenger on it, there are good odds it's a personal computer. If you can't, there are good odds it's not.
So, if something's not a PC, what is it? Either a workstation (smaller production run, intended for specialized uses like CAD, visualization, rendering) or a server (possibly even smaller production run, intended for heavy-duty stuff, services, and running headless).
The people who say that the Power Mac G5 wasn't the first 64-bit desktop personal computer invariably point at products BOXX Technologies introduced in June.
They do this in spite of a few facts:
The products BOXX introduced in June use AMD Opteron processors, which AMD targeted at the server market - not the personal computer, or even workstation, market. AMD's Athlon64 processors, aimed at the desktop personal computer market, have only recently become available.
BOXX itself makes it clear that its products are workstations, produced in small quantities and intended for specialized uses.
64-bit workstations predate BOXX by a minimum of 7 years; most UNIX vendors had them by the mid-1990s.
Now, it's possible to use a workstation to do many of the things you can do with a personal computer, especially in this age of open-source software. OpenOffice, GAIM, GnuCash and the rest are getting to the point where they provide most of the basic functionality of traditional "personal computer" software, even if total replication of features is still far off. You can load 'em up on your 7-year-old SGI Indy, and maybe you've even got Photoshop 3.0 (IRIX version) on there too!
If you do all that, does that SGI Indy become a "PC"? I'd say no - it's not designed to be one, nor intended for use as one, no matter how much you might want to make it into one.
So... BOXX workstations with server chips inside do not, in my reading of things, constitute "personal computers." In which case Apple did, in fact, have the first 64-bit desktop personal computer (and the only one, until the Athlon64 came out). BOXX had the first workstations to use 64-bit AMD chips - and that's not a bad position to be in. AMD and BOXX didn't pioneer 64-bit workstations, though.
First: Yes, Apple has said that the Power Mac G5 is the fastest, most powerful desktop computer. I don't recall hearing them call it a "workstation." Why? 'Cos it ain't a workstation. It's a general-purpose personal computer that they expect people to run all sorts of stuff on. It doesn't even offer features that people in the workstation space care about, like ECC memory and other stuff.
Second: Yes, systems using the Opteron were for sale 2-3 months before people started taking delivery of Power Mac G5's. But... you do, of course, realize that AMD positioned the Opteron as a server chip, yes? Let's see... a 64-bit server chip. Hmmm. Uh, no, I can't really say I consider that a "first." You'd have to ignore a lot of IBM POWER chips, Sun SPARCs, HP PA-RISCs, and MIPS things over the last 10+ years.
But oh, you say! BOXX took those server chips and put them into specialty rendering workstations, some of which can sit on a desk! Isn't that a first? Hmmm... seems to me that at least 7 years ago I was working on an SGI Indy, and even with that goshawful-heavy huge monitor on it, the desk didn't collapse.
So, yes. AMD's Opteron -- intended for an entirely different sort of system than the PowerPC 970 used in the PowerMac G5 -- started becoming available a few months before people started getting their PowerMac G5's. This is every bit as relevant, and apt, a comparison as pointing out that sometimes one automaker will introduce their large new SUV at a different date than another automaker's compact coupe.
The Athlon64, on the other hand, is targeted at desktops, and looks like it'll outperform the G5. However, from the time the G5 launched to the time the Athlon64 launched, it sounds to me like Apple actually did have "the fastest, most powerful personal computer," and that was enough time for them to convince over 100,000 people to order them.
(As the The Register points out, Opteron sales seem to be off to a fairly modeststart . 5,000 or so units in a quarter? That's what, about 1/20th the rate of PowerMac G5 sales?
In 1998 (five years ago), a system with an Rpeak of 16 (theoretical maximum performance of 16 gigaflops) would have been near the bottom of the June 1998 TOP500 list. (I believe there were some 333MHz 24-processor Suns in that range?)
One Power Mac G5 2.0GHz Dual (introduced in June 2003) has an Rpeak of 16 (if you look solely at the multiply-add operation pair, thus giving each FPU 2 operations/cycle). And Virginia Tech went and got 1100 of them. I guess they didn't want their system to "fall off" the bottom of the list for a while.
Isn't progress wonderful? We live in a world where you can go to the mall (if your mall has an Apple Store), call up a mail-order place, or even order on-line, and get yourself an aluminum box with enough goodies inside that a mere five years ago it would have been on the Top500 list.
Umm, no, I didn't miss that there were 2 CPU's per computer. There are 2 CPU's and each CPU has 2 FPU's and each FPU can perform "at least" one double-precision 64-bit floating-point operation per clock cycle. Thus my quote of:
There are 1,100 computers. Times 4, that gives us 4,400 flops per clock.
They've got 1,100 machines. Each has 2 CPU's for a total of 2,200 CPU's. Each CPU has 2 FPU's for a total of 4,400 FPU's. Each FPU is capable of doing a 64-bit (double-precision, I presume) floating-point calculation each clock cycle. The machines are running at 2GHz, or 2 billion clock cycles per second. So... 4,400 * 2 billion = 8.8 trillion 64-bit floating-point calculations per clock cycle.
It appears (based on thesecomments) that the only "factor of two" I missed would be the G5 FPU's ability to perform a certain pair of double-precision 64-bit floating-point operations (multiply and add) in a single clock cycle -- a pair of operations which apparently come up frequently in scientific programming. (Something that, not being a scientific programming sort, I didn't know.)
(Gosh... it's almost like they designed it to be really fast, or something!)
So... hmm. I guess like everyone else, I'll have to wait and see what the numbers are on the TOP500 list. That's an interesting formula you've got there, by the way.
$1000 just to try a Mac? Geez! I wouldn't spend $1000 just to try a Mac, either. Certainly not when Apple sells brand new, complete eMac systems starting at $800, and frequently has refurbished ones available (look for the red "SAVE" tag on store.apple.com) with full warranty for as little as $600-$700. Heck, refurb-wise you can even get a laptop for under $750. And again, that's with full warranty.
Oh, and there are vendors that'll give you a no-questions-asked return policy for 2-4 weeks...
Er, that's $5 million, not $5 billion. Big difference.:) $5 billion would get you the 1,100,000-node G5 cluster that would crank out, er... 7-15 petaflops or so.
Cooling would be a bit*h, but maybe you could rig up some sort of cogeneration plant, using the immense amount of heat generated to produce electricity, heat your dorms in winter, etc.
Hmmm... according to my electric bill, over the last year, I've used anywhere from 377 to 469 kwh of electricity per month, at a cost of US $84.89 to $105.15. My daily usage (averaged by month) has ranged from 12.6 to 14.7 kwh.
The dual G5 that should be arriving soon (are you listening, Apple?) has, if I recall, a 600 watt power supply. 600 watts * 24 hours = 14.4 kwh per day.
Yowza.
Um... yeah, I think I had better leave all the power-saving features turned on, and put it to sleep at night and all that other good stuff. Don't want to have to explain to the wife why the electric bill is suddenly double...
If the G5 can do a fused multiply add at double precision in one clock cycle, the 17.6Tflop figure would then presumably depend on 100% of the code executed consisting of fused multiply adds, yes?:)
(Of course, there may be other fused 2-operation things it can do in one cycle... but it seems unlikely that code would consist of them so completely as to attain that theoretical peak.)
The only skepticism I have regarding this figure (and I have this skepticism as a current G3 owner who's waiting for a dual G5 that had #$%#% better ship by Tuesday like they said it would!;) is that in the past, Apple has published figures based on single-precision floating point performance, and TOP500 rankings use a benchmark that requires double-precision.
I had this pointed out to me in May of 2002 when Apple introduced the Xserve... in this Slashdot thread.
Apple's page about the G5's execution core states that its two FPU's are double-precision, and that it can thus "complete at least two 64-bit mathematical calculations per clock cycle." That's obviously an improvement over the G4, but let's see...
They've got 1,100 machines. Each has 2 CPU's for a total of 2,200 CPU's. Each CPU has 2 FPU's for a total of 4,400 FPU's. Each FPU is capable of doing a 64-bit (double-precision, I presume) floating-point calculation each clock cycle. The machines are running at 2GHz, or 2 billion clock cycles per second. So... 4,400 * 2 billion = 8.8 trillion 64-bit floating-point calculations per clock cycle.
To me, that's 8.8 TeraFlops, which is conveniently precisely half of the 17.6 figure. Did I forget to multiply by two, or did someone else multiply by two an extra time? Or... is the situation (most likely) that there are situations under which the G5's additional mathy bits can actually turn out a couple more FLOPs per cycle, but that's not going to be the case most of the time, and probably won't be the case running LINPACK?
Of course, it is important to point out that this is a theoretical maximum, and that things like interconnects, RAM limitations, chip failures, entropy and unauthorized use of the cluster to render hot babes in "Poser 5" will detract from it.
So... I don't think this will be the second-most-powerful machine in the world. Even relative to the previous list. I would not be at all surprised to see it in the top 5, though, and would be surprised if it didn't make the top 10.
When we moved to our current house, Verizon issued us a number that was still listed in the yellow pages (printed about... 5-6 months earlier) as the number for a local ball-bearing company. (In fact, the number is still listed on various online directories - it's 808-935-6496; feel free to Google for proof.)
After a week of getting phone calls every frickin' weekday morning from well-meaning folks who wanted to buy ball or roller bearings, I scrounged up the correct number for the company and started giving it out.
After the second week, I called Verizon, explained that they'd issued us a number that was still listed in the yellow pages as being someone else, and said that I wanted a number that had been out of service for more than six months. (They do know how long numbers have been out of service.) We got another one that same day.
If I'd been a little more irritated, I would have held out for one with no 1's or 0's that spelled some cool word, like "welcome" or "welfare" or "welders." (Oh, but then I'd get calls about welding, I just know it...)
Darn tootin'. Even a monoculture of Macs would be more secure (since that's the buzzword in play), although price and performance would be more of a wash.
but there is no language in their terms of sale which even implies that they won't build software into OSX that maps your ip address, or time zone, or configured ISP dialup number, to a physical location. In fact, they are pretty much putting you on notice that they may well do such a thing; there's no legal requirement that they state that they "may use technology to verify compliance", yet they include explicit clear mention of the possibility. I would consider the chances of it to be better than unlikely.
I'd consider the chances of it to be non-zero, but very, very small. IP addresses don't map well to geographic locations. Time zones don't map well to "outside the U.S." - think about military bases that are U.S. turf but located nowhere near the U.S. in terms of TZ's. Ditto for configured dialup numbers - if you're even using dialup.
More importantly, though, I think Apple's got better things to do. Making sure that your card is issued by a U.S. bank and your billing address passes an AVS check as a U.S. address catches upward of 99% of things, and beyond that, they get significantly diminishing returns for any further efforts at checking stuff.
The person I was replying to didn't mention Archos, sorry. Yeah, I'm aware of those. Aren't they kinda bigger/heavier than most of the "competing" ones, though?
As far as RCA... RCA's own site lists the Lyra's being on closeout. And a search for Lyra on their site doesn't find any new models. If your stores are still selling them for full price, oh well, folks in your area must not know to look at RCA's website for a better deal.:)
I'm not sure how a music store competes with a player.
Maybe you mean that those other players are going to compete with the iPod? Yeah, that's probably what you mean. Hmmm.
iPods are... $299 for 10GB, $399 for 20GB, $499 for 40GB.
(Who the #$%^&#$& has 40 frickin' gigs of MUSIC? I mean, I don't think there are radio stations with playlists that big!)
Rio's Karma 20 is... $399 for 20GB. Comes with a dock and stuff (though it looks like the dock will use up an electrical outlet, since it's not Firewire). Doesn't look like it's anything but a music player, didn't see mention of the extra apps an iPod has. But, well, same capacity, same price, okay, I guess that's competing.
iRiver's HP-100 is... $399 for 10GB. Hey, that's a great idea! "Apple's charging an arm and a leg for the iPod and it's selling like hotcakes covered with blow and naked babes... let's price ours even higher so we can sell even more! Yee-ha!" Uh... no, that's not competing. Even if they'd gotten away from the stale, er, ahem, time-honored "Sony Walkman" look.
Philips' HDD-100 is... well, I couldn't find anywhere selling them, so I'm not sure how much competing it's doing.
The RCA's someone mentioned are cheaper - $199 for 10GB and $249 for 20GB - but they appear to be discontinued and those are closeout prices.
I was talking to an analyst the other day who seemed convinced that the new tiny el-cheapo hard drive from Cornice were going to rule the roost. Well, maybe among the people who only want 1.5GB of storage and are happy that they can get it for significantly less than the cost of 1.5GB of flash RAM... but that's only a few hundred songs. Hardly the sort of thing you'd want to use as the emergency backup for a radio station.
Authorization has nothing whatsoever to do with your geographical location, and everything to do with the geographical location of the bank that issued your credit card, and the billing address for that card.
Period. Pure and simple. If you have a credit card issued by a U.S. bank, and you sign up for the iTunes Music Store, you can then proceed to drag your happy Mac-using behind all over the planet, and all your iTunes stuff will still work.
That one guy who had problems and raised a huge fuss? He had either gotten a card from a bank outside the U.S., or changed his billing address to be somewhere else. And then he'd tried to sign back up after nuking all his songs somehow.
Okay, so maybe it's a little bit hard to move without changing your billing address... but then, there are plenty of card companies who're perfectly happy to handle your billing and all that stuff on-line, since it's cheaper for them than using paper.:)
Oh, and for what it's worth, only accepting credit cards issued by U.S. banks is very common in the dot-com space. When I worked for Cheap Tickets years ago, I became aware of this. You have to hook into entirely different systems and stuff if you want to be able to run verification on cards that are drawn in other countries - even neighboring ones like Canada and Mexico. It's a real pain.
Slashdot Mirror
Rmax is the best it's attained; Rpeak is the theoretical peak in a perfect world with 100% multiprocessing efficiency.
One of the really amazing things about the NEC Earth Simulator (aside from its sheer power) is that its Rmax is over 85% of its Rpeak. A lot of other systems only get Rmax of 60-70% of Rpeak.
I've read that PowerPC chips are pretty efficient in SMP scenarios, and if "Big Mac" is turning in an Rmax that's 80% of Rpeak (even on a fairly small number of processors), that would seem to validate that assertion.
If that's how you choose to define things, the first 64-bit desktop PC's were made over a decade ago.
Oh, and if one 64-bit PC architecture weren't enough, how about this one? Of course, back then, Windows NT also ran on the PowerPC architecture.
So... either the fact that the Opteron, like the MIPS and Alpha chips, was not designed or intended for use in "personal computers" means that Opteron-based workstations are not "personal computers" any more than Alpha- or MIPS-based ones were... or neither AMD nor Apple has the first "64-bit personal computer" by an entire decade.
(Incidentally, Windows NT 4.0 CD's -- I have one -- still contain code for those other architectures!)
It doesn't -- it just has to be substantively identical to one. "Personal computers" are -- or are substantively identical to things that are -- mass-produced and marketed to consumers. You didn't see commercials on VH-1 for your system, and you didn't buy it at the mall or a big-box computer store -- but you (or others) have probably seen commercials for substantively identical kit on VH-1, and you could have bought something substantively identical to it in a mall or big-box store.
You cannot (to my knowledge) walk into the mall, or a big-box computer store, right now, and walk out with a BOXX workstation or something substantively identical, nor will you see commercials for it on VH-1. You will see commercials on VH-1 for Apple's G5, and you can walk into an Apple Store or possibly even a CompUSA store (I don't know who's gotten shipments of what, ya know?) and walk out one. (They'll be upset, though, if you don't pay.)
I realize, of course, that this all reeks of people with advertising budgets getting to declare that their products are whatever the heck they darn well please. Apple could have claimed that the G5 was a desktop, a personal computer, a workstation, a server, even a cheese grater, and since they're spending the bucks on commercials, that's what The Masses(tm) will probably believe. :)
Oh, wait, did you mean the first 64-bit desktop PC? Well now, that's a little harder, since it depends on what "PC" means.
PC, as we all know, stands for piece of cra...^W^W^W, um, er, stands for Personal Computer. Another nice vague term, but here's a crack at it:
So, if something's not a PC, what is it? Either a workstation (smaller production run, intended for specialized uses like CAD, visualization, rendering) or a server (possibly even smaller production run, intended for heavy-duty stuff, services, and running headless).
The people who say that the Power Mac G5 wasn't the first 64-bit desktop personal computer invariably point at products BOXX Technologies introduced in June. They do this in spite of a few facts:
Now, it's possible to use a workstation to do many of the things you can do with a personal computer, especially in this age of open-source software. OpenOffice, GAIM, GnuCash and the rest are getting to the point where they provide most of the basic functionality of traditional "personal computer" software, even if total replication of features is still far off. You can load 'em up on your 7-year-old SGI Indy, and maybe you've even got Photoshop 3.0 (IRIX version) on there too!
If you do all that, does that SGI Indy become a "PC"? I'd say no - it's not designed to be one, nor intended for use as one, no matter how much you might want to make it into one.
So... BOXX workstations with server chips inside do not, in my reading of things, constitute "personal computers." In which case Apple did, in fact, have the first 64-bit desktop personal computer (and the only one, until the Athlon64 came out). BOXX had the first workstations to use 64-bit AMD chips - and that's not a bad position to be in. AMD and BOXX didn't pioneer 64-bit workstations, though.
I dunno, man... buying a system with a bleeding-edge CPU that's only been available for a few months, without OEM support?! You're brave.
Second: Yes, systems using the Opteron were for sale 2-3 months before people started taking delivery of Power Mac G5's. But... you do, of course, realize that AMD positioned the Opteron as a server chip, yes? Let's see... a 64-bit server chip. Hmmm. Uh, no, I can't really say I consider that a "first." You'd have to ignore a lot of IBM POWER chips, Sun SPARCs, HP PA-RISCs, and MIPS things over the last 10+ years.
But oh, you say! BOXX took those server chips and put them into specialty rendering workstations, some of which can sit on a desk! Isn't that a first? Hmmm... seems to me that at least 7 years ago I was working on an SGI Indy, and even with that goshawful-heavy huge monitor on it, the desk didn't collapse.
So, yes. AMD's Opteron -- intended for an entirely different sort of system than the PowerPC 970 used in the PowerMac G5 -- started becoming available a few months before people started getting their PowerMac G5's. This is every bit as relevant, and apt, a comparison as pointing out that sometimes one automaker will introduce their large new SUV at a different date than another automaker's compact coupe.
The Athlon64, on the other hand, is targeted at desktops, and looks like it'll outperform the G5. However, from the time the G5 launched to the time the Athlon64 launched, it sounds to me like Apple actually did have "the fastest, most powerful personal computer," and that was enough time for them to convince over 100,000 people to order them.
(As the The Register points out, Opteron sales seem to be off to a fairly modest start . 5,000 or so units in a quarter? That's what, about 1/20th the rate of PowerMac G5 sales?
This article from China (in English) provides a good bit more information about Lieutenant Colonel Yang Liwei.
One Power Mac G5 2.0GHz Dual (introduced in June 2003) has an Rpeak of 16 (if you look solely at the multiply-add operation pair, thus giving each FPU 2 operations/cycle). And Virginia Tech went and got 1100 of them. I guess they didn't want their system to "fall off" the bottom of the list for a while.
Isn't progress wonderful? We live in a world where you can go to the mall (if your mall has an Apple Store), call up a mail-order place, or even order on-line, and get yourself an aluminum box with enough goodies inside that a mere five years ago it would have been on the Top500 list.
Just what the world needs - supercooled squirrels that run even faster.
Now no birdfeeder will be safe.
(Somewhere in here, Foamy from Ill Will Press's "Neurotically Yours" should get mentioned...)
I've got a Versa 2000C, 486DX4-75, but that's decommissioned now.
(Gosh... it's almost like they designed it to be really fast, or something!)
So... hmm. I guess like everyone else, I'll have to wait and see what the numbers are on the TOP500 list. That's an interesting formula you've got there, by the way.
(I wonder how fast it'll run The Sims?)
It's only a matter of time before they're flying jets...
(C'mon, guys, you have to say it more often to really get the "mantra" feel...)
Oh, and there are vendors that'll give you a no-questions-asked return policy for 2-4 weeks...
Each computer can do 4 fpu instructions/clock.
(and those are double-precision 64-bit flops).
There are 1,100 computers. Times 4, that gives us 4,400 flops per clock.
(Note: given 1 DP flop per clock per FPU, this ties in nicely with my use of 4,400 as the number of FPU's, in my comment.)
4,400 flops per clock * 2 billion clock cycles per second = 8.8 trillion flops per second.
I'm still getting the same number. Where's the factor of two I missed?
Cooling would be a bit*h, but maybe you could rig up some sort of cogeneration plant, using the immense amount of heat generated to produce electricity, heat your dorms in winter, etc.
The dual G5 that should be arriving soon (are you listening, Apple?) has, if I recall, a 600 watt power supply. 600 watts * 24 hours = 14.4 kwh per day.
Yowza.
Um... yeah, I think I had better leave all the power-saving features turned on, and put it to sleep at night and all that other good stuff. Don't want to have to explain to the wife why the electric bill is suddenly double...
(Of course, there may be other fused 2-operation things it can do in one cycle... but it seems unlikely that code would consist of them so completely as to attain that theoretical peak.)
I had this pointed out to me in May of 2002 when Apple introduced the Xserve... in this Slashdot thread.
Apple's page about the G5's execution core states that its two FPU's are double-precision, and that it can thus "complete at least two 64-bit mathematical calculations per clock cycle." That's obviously an improvement over the G4, but let's see...
They've got 1,100 machines. Each has 2 CPU's for a total of 2,200 CPU's. Each CPU has 2 FPU's for a total of 4,400 FPU's. Each FPU is capable of doing a 64-bit (double-precision, I presume) floating-point calculation each clock cycle. The machines are running at 2GHz, or 2 billion clock cycles per second. So... 4,400 * 2 billion = 8.8 trillion 64-bit floating-point calculations per clock cycle.
To me, that's 8.8 TeraFlops, which is conveniently precisely half of the 17.6 figure. Did I forget to multiply by two, or did someone else multiply by two an extra time? Or... is the situation (most likely) that there are situations under which the G5's additional mathy bits can actually turn out a couple more FLOPs per cycle, but that's not going to be the case most of the time, and probably won't be the case running LINPACK?
Of course, it is important to point out that this is a theoretical maximum, and that things like interconnects, RAM limitations, chip failures, entropy and unauthorized use of the cluster to render hot babes in "Poser 5" will detract from it.
So... I don't think this will be the second-most-powerful machine in the world. Even relative to the previous list. I would not be at all surprised to see it in the top 5, though, and would be surprised if it didn't make the top 10.
After a week of getting phone calls every frickin' weekday morning from well-meaning folks who wanted to buy ball or roller bearings, I scrounged up the correct number for the company and started giving it out.
After the second week, I called Verizon, explained that they'd issued us a number that was still listed in the yellow pages as being someone else, and said that I wanted a number that had been out of service for more than six months. (They do know how long numbers have been out of service.) We got another one that same day.
If I'd been a little more irritated, I would have held out for one with no 1's or 0's that spelled some cool word, like "welcome" or "welfare" or "welders." (Oh, but then I'd get calls about welding, I just know it...)
Darn tootin'. Even a monoculture of Macs would be more secure (since that's the buzzword in play), although price and performance would be more of a wash.
I'd consider the chances of it to be non-zero, but very, very small. IP addresses don't map well to geographic locations. Time zones don't map well to "outside the U.S." - think about military bases that are U.S. turf but located nowhere near the U.S. in terms of TZ's. Ditto for configured dialup numbers - if you're even using dialup.
More importantly, though, I think Apple's got better things to do. Making sure that your card is issued by a U.S. bank and your billing address passes an AVS check as a U.S. address catches upward of 99% of things, and beyond that, they get significantly diminishing returns for any further efforts at checking stuff.
As far as RCA... RCA's own site lists the Lyra's being on closeout. And a search for Lyra on their site doesn't find any new models. If your stores are still selling them for full price, oh well, folks in your area must not know to look at RCA's website for a better deal. :)
Of course, there's also per-album pricing, and if you find an album with, say, 15 songs, it might still be 10 bucks.
Also, where have you been buying CD's? I haven't seen them for 10 bucks in most music stores since... er... ever. $14-$18 seems far more common.
Maybe you mean that those other players are going to compete with the iPod? Yeah, that's probably what you mean. Hmmm.
iPods are... $299 for 10GB, $399 for 20GB, $499 for 40GB.
(Who the #$%^&#$& has 40 frickin' gigs of MUSIC? I mean, I don't think there are radio stations with playlists that big!)
Rio's Karma 20 is... $399 for 20GB. Comes with a dock and stuff (though it looks like the dock will use up an electrical outlet, since it's not Firewire). Doesn't look like it's anything but a music player, didn't see mention of the extra apps an iPod has. But, well, same capacity, same price, okay, I guess that's competing.
iRiver's HP-100 is... $399 for 10GB. Hey, that's a great idea! "Apple's charging an arm and a leg for the iPod and it's selling like hotcakes covered with blow and naked babes... let's price ours even higher so we can sell even more! Yee-ha!" Uh... no, that's not competing. Even if they'd gotten away from the stale, er, ahem, time-honored "Sony Walkman" look.
Philips' HDD-100 is... well, I couldn't find anywhere selling them, so I'm not sure how much competing it's doing.
The RCA's someone mentioned are cheaper - $199 for 10GB and $249 for 20GB - but they appear to be discontinued and those are closeout prices.
I was talking to an analyst the other day who seemed convinced that the new tiny el-cheapo hard drive from Cornice were going to rule the roost. Well, maybe among the people who only want 1.5GB of storage and are happy that they can get it for significantly less than the cost of 1.5GB of flash RAM... but that's only a few hundred songs. Hardly the sort of thing you'd want to use as the emergency backup for a radio station.
Period. Pure and simple. If you have a credit card issued by a U.S. bank, and you sign up for the iTunes Music Store, you can then proceed to drag your happy Mac-using behind all over the planet, and all your iTunes stuff will still work.
That one guy who had problems and raised a huge fuss? He had either gotten a card from a bank outside the U.S., or changed his billing address to be somewhere else. And then he'd tried to sign back up after nuking all his songs somehow.
Okay, so maybe it's a little bit hard to move without changing your billing address... but then, there are plenty of card companies who're perfectly happy to handle your billing and all that stuff on-line, since it's cheaper for them than using paper. :)
Oh, and for what it's worth, only accepting credit cards issued by U.S. banks is very common in the dot-com space. When I worked for Cheap Tickets years ago, I became aware of this. You have to hook into entirely different systems and stuff if you want to be able to run verification on cards that are drawn in other countries - even neighboring ones like Canada and Mexico. It's a real pain.