Joe doesn't *need* that speed, but it does make the computing experience more enjoyable.
Not for most of them. If you look at the average user, they are disk-bound. A 2mhz and a 3mhz P4 are, for most people, identical in performance, all else being the same. Just look at how few cycles are actually being used in the average PC. Some years back, I'd have agreed with you, but I'm still running an AMD Athlon XP 2600+, not exactly state of the art, and I really don't find a compelling reason to upgrade, even though I typically have lots of windows open and more background tasks than the average user.
Actually, while IBM and the clone makers skipped on the 80186 as a core system CPU, the Unisys ICON Series I and II desktop computers both ran off the Intel 80186 (running QNX at that).
There were also other 80186 PCs. The Tandy TRS-80 Model 2000, the Research Machines (RM) Nimbus PCs, and the Acorn Business Computer used the 80186. It was also used in the add-on PC-compatibility unit for the BBC Micro.
So that chip was (at one time) suitable for desktop systems:).
So was a 4 mhz Zilog Z80. I know because I had one. It was my pride and joy at the time. But I think that we can all agree (okay, except perhaps for my nemesis in this discussion) that the average embedded system CPU is not going to work for a modern desktop.
How amusing; yet another thing you don't know anything about that yet you feel to need to speak with authority.
I do speak with authority and expertise. It is you who seem unable to address even the most rudimentary aspects of this discussion.
For the record, I'm a graduate engineer with specific training in embedded system design and over a decade of real-world experience.
Exactly what is your "real-world experience"? Working on databases? Writing HTML? What kind of engineer are you -- perhaps a software engineer?
But I guess you're trying to fall back on questionable claims of credentials rather than answering any of the points I made in the previous post. Specifically:
The fact that two CPUs share the same architecture does not mean that they are both suitable for the same uses.
We were discussing the G5 CPUs that Apple uses for desktop systems and you tried to swing the conversation around to PowerPC CPUs for embedded systems.
You incorrectly implied that the PowerPC CPUs used in your household appliances and comms devices are suitable for use in Apple's desktop systems.
You speculated that PPC architecture CPUs for embedded systems cost less than desktop CPUs from Intel and AMD -- which has nothing to do with what Apple pays for the G5 PowerPC chips used in their desktop PCs.
When the part cannot be used for your intended purpose, it has no "bang for the buck" in that application.
You falsely accused me of saying that the PPC is inadequate for a desktop -- and then weren't man enough to admit that your claim was untrue.
Your claim that embedded systems engineers normally choose CPUs based on the "$ per MIP[sic]" was simply wrong. Embedded systems engineers look for the lowest cost system implementation that meets the design goals -- not the most CPU performance per dollar.
Certainly seems that I hit a nerve there.
You didn't "hit a nerve." You just pissed me off by falsely attibuting statements to me.
Now, answer the points or go away and quit wasting bandwidth.
You are in way over your head. You don't know the difference between an architecture and a specific CPU. We talk about the G5 CPUs that Apple uses for desktop systems and try to swing the conversation around to PowerPC CPUs for embedded systems:
"PowerPC is used in a lot of embedded systems. The next generation of video games consoles are exclusively PowerPC. I've got PowerPC chips in my household appliances and various comms devices."
So what? Are you suggesting that the PowerPC used in your household appliances comms devices are suitable for use in Apple's desktop systems? I have a couple of Garmin GPSs that use 80186 family CPUs and the CPUs only cost a few bucks. That doesn't mean that those chips are suitable for use in desktop systems.
Then when several people point out that PowerPC is cheaper than AMD or Intel you say the price is irrelevant and what matters is "bang for buck".
No, you speculated that the low-end PPC architecture CPUs for embedded systems cost less than desktop CPUs from Intel and AMD, which is irrelevent to the discussion.
Then when I point out that price can never be irrelevant when you're talking "bang for buck"
You didn't point anything out. You made an incorrect assertion. When the part cannot be used for your intended purpose, is has no "bang for the buck." If you are designing a passenger car, it doesn't matter how cheap a Weed Whacker engine is. 1 hp isn't going to power the car and a PowerPC CPU out of a thermostat isn't going to power your desktop computer.
you go back to your initial argument that the PPC isn't adequate for a desktop.
No, I never claimed that the PowerPC architecture was inadequate for desktop use. You are either a liar or an idiot. I defy you to show any instance where I said that the PPC isn't adequate for a desktop. Either show me where I said that or admit that what you just wrote was untrue.
Hadn't you noticed that PowerPC desktops are rather popular? Apple is selling millions of them.
Were you actually under the impression that the CPUs in the PowerPC desktops are the same as the PPC CPUs in embedded systems?
I think I spot the underlying problem here; you don't actually know how much the PowerPC costs. The PowerPC is one of the most popular embedded CPUs - used in cars, phones, routers, consoles, multimedia devices - because systems designed around the PowerPC are inexpensive. Embedded systems are the litmus test for cost efficiency. They measure CPUs in $ per MIP; that's "bang for buck" by another name.
Your ignorance has reached new heights! First off, it's "MIPS", not "MIP" and it stands for Million Instructions Per Second. Only a small percentage of embedded systems designers care little about CPU performance (as if MIPS were a measurement of that anyway!). They care about total system cost. For example, the AVR series of microcontrollers are incredibly popular and rather slow. Their cost per MIP is horrible when compared with a desktop CPU. But they include such things as A/D converters, built-in flash memory, built-in RAM, and built-in EEPROM. If they are designing a setback thermostat with 8 buttons and a 12 character display, they care about cost, not MIPS per dollar. It doesn't matter to them if the CPU is.75 MIPS or 1000 MIPS. It's totally irrelevent.
Are all the embedded designers stupid?
No, but you clearly don't understand anything about what we do.
Would they save more money by listening to you?
That's why I've been paid to design, and review designs for, embedded systems.
Have fun with your stupid register-starved CPU which needs 5 times more power to achieve the same computing performance.
Consumers of PCs don't buy them based on their register complement or their CPU wattage requirements. They buy them based on price, performance, and features.
While you are at it, load a niche operating system from Redmond that can hardly interoperate with anything different and finish it with a shitvidia video card with super-proprietary drivers that will delete every.c file it finds while installing.
Niche? I suggest that you invest in a dictionary. Microsoft is the dominant OS on desktops today and is probably the dominant server OS, too. As to your silly comment that it cannot "interoperate with anything different", open your eyes. It can operate as a web server, mail server, FTP server, Telnet server, SSH server, etc. It can print to Unix LPD printers, access servers running on Linux systems -- all using standard TCP/IP, UDP, and other standard protocols. Your inability to configure Windows to interoperate speaks more to your lack of experience than flaws with the Windows OS.
As to your comment than drivers for certain popular video cards "delete every.c file it finds while installing", that's just plain hyperbole and stupidity.
There are more PowerPC architecture chips floating around in the world than there are and ever will be x86 desktop processors.
So what? Some slow, cheap, embedded PPC architecture CPU isn't going to have any effect on the R&D for desktop CPUs. That's like arguing that the 80186 CPUs that can be found in everything from GPSs to fishfinders are funding the R&D for the P4.
IBM afford the R&D by being IBM - a company people keep saying "is so small and insignificant compared to Intel" made $1.8bn profit last year which was a little more than Intel did, and have a total revenue of almost triple Intel's.
And they didn't become that big by investing large amounts of R&D into low-volume CPU lines.
IBM's fortunes rest in many many other places than "in boxed and OEM processors for the desktop PC market". Market share is moot
That's simply ridiculous. Market share is what drives R&D. IBM is not going to divert money from a profitable line if minicomputers to fund a low-volume, low-profit chip line like the one used by Apple.
- you could say that Intel have a much lower market share than AMD in some areas, does that make AMD the R&D behemoth that people should be entertaining for a switch?
In many areas, yes. That's why you're seeing AMD being adopted in areas like servers.
If Apple switch to x86 then I will eat my laptop.
I have no desire to see you die -- just to admit that you were wrong.
The news articles clearly state that they are switching to Intel Chips - nobody has yet dared say that this means Intel IA-32 with or without EM64T. I wonder what Intel could produce that would replace IBM's efforts in this field.
The x86 architecture, while not elegant, is extremely well-tuned and refined and is an viable choice for Apple. It's not like the IBM chips are outrunning the latest Intel and AMD offerings.
In fact, I would argue that it's the only rational choice for Apple. Apple needs to adopt technology that already has a big market share. That's why it was smart for them to adopt PCI, IDE, USB, and other technologies for which the R&D cost was shared with Apple's competitors. The big problem Apple has now is that IBM doesn't sell enough PPC Gx CPUs to support R&D that would be needed to keep them competitive with Intel and AMD's x86 offerings. Going to another "non-standard" CPU architecture would leave Apple in the same position that they are today.
The PowerPC is certainly adequate for a modern desktop system. The popularity of Apple desktops is proof enough of that.
Yes, but it has poor bang for the buck because it costs considerable more than an AMD or Intel CPU which has equivalent performance in a desktop system.
I don't see how it's possible for you to argue "bang for buck" while disregarding the price.
0 / any$ = 0. If the CPU has no value in a given application, it does not matter how cheap it is. A Zilog Z80 can be had for probably $5. That doesn't mean that it has great bang for the buck if your goal is to build a desktop system. So it's irrelevent to this dicussion. If Apple could build a system with a $20 gaming console PC and it satisfied most users, then that would be great. But they can't. So let's ignore those, ARMs, Atmel AVRs, Zilong Z80s, and every other CPU which really isn't viable for a modern desktop PC. They are irrelevent to this discussion.
Of course, let's not forget this question: outside of the gaming arena, just how fast does a personal computer have to be, anyways?
My 4mhz Z80 CP/M system does seem a bit slow now...
It all depends on what you expect of it. If the GUI interface is slow on the OS, it can be annoying -- and the OSs are getting more and more demanding all of the time.
I don't think performance is necessarily the driving force behind Apple's decisionmaking, which means there has to be something else behind it.
Yes: Price. If Apple has to spend $200 to get a CPU that performs as well as the one that Dell pays $100 for, then they will have trouble competing. Most people don't need the speed that they have in their PC, but it was a selling point. It's like the Suzuki Hayabusa. It will do 186mph. How many people need to go that fast? Okay, now how many people bought the bike because it can go that fast? I bet the second number is much higher than the first.
Basically this is the same argument for ARM. The ARM is not a fast processor, but ARM is used in lots of systems because it represents exceedingly good "bang for buck".
But it doesn't mean that it's got adequate performance for a modern desktop system. So, in that usage, it has incredibly poor bang for the buck, regardless of price.
I don't disagree with you about the sluggish performance of PowerPC. If you can afford it then you will get much better performance from a P4 than from a G5. The Apple fanboys annoy the hell out of me as well. I'm running a G4 here but I've no allegiance to the cult of Apple.
Nor do I have any allegiance to Intel (as I compose this on my AMD-powered system). I get annoyed with all of the fanboys on all sides. It's a CPU, not a girlfriend. You don't owe it loyalty, fidelity, and love. It's like getting misty-eyed about the brand of RAM in your system.
So after a long hard day at work, you like to come home and kick back with a game or two.
Yep. Usually on my PC. I just can't get into game consoles. I find the games for PCs better and the graphics from my ATI video to be superior. I also like the upgrade rather than replace philosophy with PCs. If I get a new PC, my old games will still work.
It's little wonder that IBM found themselves with more interesting things to do than ship apple their G5 chips.
Sorry, I didn't realize that was the point of what you were saying. But it's a very good point indeed. IBM will invest where they think they can make the most profit.
Shipping out volume in Linux servers, XBox 360s, PS3s and Nintendo Revolutions would be my first guess as to how they afford the R&D.
When they get a contract for the XBox 360 or some other console, they don't have to improve the CPUs constantly. The first one and the last one are probably going to be about the same. Nor do they have to compete with the desktop offerings from Intel and AMD. And really, how many PowerPC Linux servers do they really sell compared to Intel and AMDs sales in the server market?
IBM doesn't build these chip fab facilities just for the tax writeoff:)
That's exactly what the heart of the problem (for Apple) is: IBM is looking at this as a business and isn't willing to invest huge sums in R&D to compete with Intel and AMD. Apple wants a CPU that lets them compete with Windows PCs. They want to be able to compete on both performance and price. Without similar R&D expenditures, so IBM has not been able to offer competitive performance to Intel or AMD at the same price point.
This isn't a slam at Apple or IBM nor is it a form of high praise for Intel (I'm really no fan of Intel and prefer AMD). It's just economic realities. If Apple had 50% of the PC market, then IBM could afford to do a lot more R&D and might well come up with a CPU that was considerably faster than the Intel or AMD offerings.
Cutie Pi 3.1415926535897932384626433832795028841972....
Actually, the last digit shown should be a 1, not a 2, if you are going to end with "...". The 2 is the result of rounding up because the digit that follows is a 6.
Yes, I'm an embedded systems engineer with over 20 years of experience.
Are you suggesting that IBM is going to take money from profitable embedded processor lines and plow that into R&D on desktop PowerPC CPUs? I somehow doubt it. Most of it doesn't apply anyway. In Intel's case, many of their embedded CPUs are based on now-outdated desktop technology, so while the desktop R&D eventually benefits their embedded processors, it seldom goes in the other direction.
Also, you will find that Atmel, ARM, Zilog, Microchip Technology, Motorola, and others, who are not big players in the desktop market, have a huge share of the embedded systems market. You also have to consider the dollar figures. When Intel or IBM sells a desktop CPU, they probably get about $100 or more for it, on averate. The average embedded CPU is less than $15.
In answer to your question, I'm an engineer. I get my chip info from the vendors and I get my benchmarks from independent, respected benchmarks like SPEC (Standard Performance Evaluation Corporation). Where do you get your chip info?
but I don't see what Intel has in 2006 that IBM can't match, or AMD, or whoever.
Intel has the market share so they have the R&D money as well as economies of scale. The PowerPC line is like the Sun SPARC line -- they have a limited market share and they can't afford the R&D to keep up with Intel and AMD when it comes to performance and price. Apple has less than 2% of the PC market. Intel has about 80%. When Intel sells about 40 times more CPUs, how can IBM afford the R&D to stay competitive?
Oh, if only those of us with x86 Intel and AMD CPUs could have the sluggish performance and high prices that one can get with a PowerPC system. I feel the jealousy welling up inside of me...:)
Seriously, grow up. It's just a CPU. Intel's CPUs offer more bang for the buck than IBM's PowerPC CPUs. It's not surprising given how many more Intel sells and how much more they have for R&D as a result. If the story is true, then Apple recognizes this and realizes that they can make more money with Intel CPUs while giving their customer base better performance. It's not religion. It's business. Just like the past Apple decisions to support PCI bus, IDE drives, USB, etc. Apple is a for-profit company and they base their product design decisions on that.
Also, not too many pieces of PC hardware at that price point have Firewire 400 and DVI.
And you think that Linux is the best OS to take advantage of those interfaces? Do you think that Linux is going to offer better, more polished apps than OS/X?
Installing Linux on the Mac Mini is like buying a Mercedes and then replacing the engine with one pulled from a Saturn.
That's not confusing, that's just faulty logic in which only a licensed taxi service may contract to drive you to the airport (or wherever).
No, it's not faulty logic. I never claimed that only a licensed taxi could drive you to/from the airport. If you want to pay your next door neighbor to drive you there, you can. But your next door neighbor can't legally hang out at the airport offering to drive random strangers home for a fee.
Completely contrived example: by default it is illegal to hire or otherwise engage someone to do something illegal. That's why the law punishes the guy who hires the hit man as well as the hit man himself. Prostitution is illegal, which is why solicitation is illegal as well.
But you are ignoring the point: That you are legally permitted to do something for yourself does not imply that you can legally hire someone else to do it for you -- or that someone is legally permitted to provide said service for a fee.
By your comprehension of a derivative work, it is illegal for me to let my next door neighbor use the DVD remote when watching a disk at my place.
No one knows my comprehension of "derivative work" better than I do, and I can state, categorically, that I never thought illegal the hypothetical situation you just posed.
Ever heard the term "reasonable man" as it applies to the law? The law is not binary. It recognizes shades of gray. Judges use judgment when making decisions: Did the framers of copyright law intend to leave a loophole for on-the-fly creation of derivative works? Did they want a for-profit third party to be able to provide the kind of service that ClearPlay does? I don't think so. If they prohibited companies from selling you a derivative work on film, tape, LP, or other media, why do you think that they would have wanted a company like ClearPlay to provide on-the-fly editing for profit? That's the key question: What is the intent of the copyright law? Obviously, this administration felt that they had to specifically make a law permitting what ClearPlay is doing, so they must have thought that it was either previously illegal or of questionable legality.
If there were any credit to the claim that this was about copyright violation, this matter would never have gotten as far as it has - ClearPlay doesn't have that kind of legal muscle.
But you forget that these are the same right-wing zealots who held Congressional hearings about Janet Jackson's nipple being flashed for a split second. They are the same right-wing zealots who levied record fines for "indecency" against radio and television stations. This is how they kiss-up to the ultra-conservative Christian voters.
This is about your rights as a consumer to do what you will with your legally acquired DVD movie. And that includes the right to hire a third party to prepare a list of scenes to be cut or bleeped. If it's legal for me to do it myself, why isn't it legal for me to hire someone else to do it for me?
The first sale doctrine gives the purchaser the right to create derivative works for their own use. Clearplay is not the purchaser of the DVD and, therefore, does not have those rights. It is legal for you to hire a third party to edit your DVD, but it is not legal for the third party to do the editing. If that's confusing, think of it this way: There is no law against you paying someone for a ride to the airport. But there is a law against that person operating an unlicensed taxi servive.
That you can legally do something yourself does not imply that you can legally hire someone to do it for you. You can bring youself to orgasm legally, but you cannot legally hire someone else to bring you to orgasm.
The issue is the rights of the consumer versus the rights of the producer. All this talk about ClearPlay distributing a "derivative work" ignores the fact that every ClearPlay customer has bought or rented the DVDs they play on their ClearPlay equipped DVD player.
It's not being ignored at all. But, since the consumer is not watching the DVD that they bought and is, instead, viewing a derivative work put together on-the-fly, it's a side issue.
But they aren't doing that: they are proposing a service that allows people who have legally bought or rented a DVD to exercise their legal rights in choosing how they watch their DVD.
No, they are providing a service which produces an on-the-fly derivative work based not on what the consumer wants, but, instead, based on what ClearPlay thinks is appropriate (divided into categories). Is a slap against the cheek of a hysterical person violence? Is a mother nursing a baby sexual content? These are decisions being made by the movie editors at ClearPlay, not by the consumer.
At least that's the way the most business friendly congress in recent history seems to see it.
There's no doubt that the right-wing zealots in Congress, many of whom have been whining about the evil, liberals in Hollywood, think that it's a good idea. They think that levying millions of dollars in fines because of one exposed nipple is reasonable, too, so it's not like Congress has the most progressive values in the world. I think that the Right Wing members of Congress are like the O.J. jury -- they are using the law to send a message rather than trying to make the right decision.
Fine. You go to 290 million people in the U.S. and educate them -- every man, woman, and child -- on how to deal-with, operate, and upgrade computers.
It's not my job to personally educate everyone in the U.S. about computers. Just as it's not my job to teach them traffic laws and how to operate a motor vehicle -- yet we expect them to learn those thing before we license them to drive.
God knows that learning about computers should be the key goal in everyone's life. The guy investigating prostate cancer online after bad news from his doctor? He should stop what he's doing and take lessons from you about computers.
Again, I'm not offering lessens. The guy investigating prostate cancer online should have learned about computers long before being diagnosed with prostate cancer. That's like saying that he should be allowed to drive to the library on the wrong side of the road at 90mph because, damnit, he has prostate cancer and doesn't have time to learn about traffic laws and cars.
Everyone should not have to know about everything just to avoid being victimized.
People shouldn't have to take self-defense courses and carry pepper spray. They shouldn't have to worry about identity theft. No one should get prostate cancer. Life's not fair. Deal with it.
'We have to reduce the complexity. People do not want privacy software, firewall, antivirus, spyware, adware and blended threats. The average individual is saying I don't want this anymore. Just fix it. What we are doing is reducing the complexity.'
I am sick and tired of companies pandering to people too lazy to learn anything. Computers are complex and they aren't getting simpler -- because people demand more and more from them. If someone doesn't want to invest the time to master the use of a computer, then they shouldn't operate one.
These are the same stupid whiners who are in tears when their hard disc crashes because they never backed up anything. They are the ones who are standing in line in front of us at the computer store having hissy fits because they want the techs to install a DVD-R/W drive while they wait -- because there's no way that they would consider learning how to put in four screws and three cables (power, IDE, and audio) themselves. They are the ones who install adware on their systems and are then shocked at the results. They are the ones who stare blankly like deer looking at headlights when the salesman asks if they have USB ports into which their new digital camera can be plugged.
All that this kind of initiative by Symantec does is help the ignorant remain so while increasing the under-the-hood complexity even further. So when something in the automated mass of hundreds of megs of files breaks, the user doesn't have a prayer of being able to fix it.
If you aren't willing to learn how a radio works, then the FCC won't give you a HAM radio license. If you aren't willing to learn the rules of the road, the DMV won't give you a driver's license. We need a computer license and people who won't learn the basics about computers shouldn't be allowed to own and operate a computer.
No, the final product may be a derivative work, but the list of time codes to blank or bleep aren't. By your standards, *any* kind of editing software or hardware - including the fast forward and rewind buttons on the remote control - would constitute a "derivative work". That's plainly ridiculous.
It is ridiculous. And it's not my contention.
The final product is the issue. Customers aren't paying to have a list of timecodes and edit instructions. They're paying to see a version of Spiderman, The Matrix, or Titanic without the "dirty bits." So what they are buying are derivative works.
If they want to fast-forward or hit mute based on some cue-sheet sent to them from ClearPlay, that's fine. Then the cue-sheet really would be the issue because that's what the consumer would be paying for. Now they're skipping that and going right to paying for a derivative work to be presented on their television screen.
Slashdot Mirror
Joe doesn't *need* that speed, but it does make the computing experience more enjoyable.
Not for most of them. If you look at the average user, they are disk-bound. A 2mhz and a 3mhz P4 are, for most people, identical in performance, all else being the same. Just look at how few cycles are actually being used in the average PC. Some years back, I'd have agreed with you, but I'm still running an AMD Athlon XP 2600+, not exactly state of the art, and I really don't find a compelling reason to upgrade, even though I typically have lots of windows open and more background tasks than the average user.
Actually, while IBM and the clone makers skipped on the 80186 as a core system CPU, the Unisys ICON Series I and II desktop computers both ran off the Intel 80186 (running QNX at that).
:).
There were also other 80186 PCs. The Tandy TRS-80 Model 2000, the Research Machines (RM) Nimbus PCs, and the Acorn Business Computer used the 80186. It was also used in the add-on PC-compatibility unit for the BBC Micro.
So that chip was (at one time) suitable for desktop systems
So was a 4 mhz Zilog Z80. I know because I had one. It was my pride and joy at the time. But I think that we can all agree (okay, except perhaps for my nemesis in this discussion) that the average embedded system CPU is not going to work for a modern desktop.
I do speak with authority and expertise. It is you who seem unable to address even the most rudimentary aspects of this discussion.
For the record, I'm a graduate engineer with specific training in embedded system design and over a decade of real-world experience.
Exactly what is your "real-world experience"? Working on databases? Writing HTML? What kind of engineer are you -- perhaps a software engineer?
But I guess you're trying to fall back on questionable claims of credentials rather than answering any of the points I made in the previous post. Specifically:
- The fact that two CPUs share the same architecture does not mean that they are both suitable for the same uses.
- We were discussing the G5 CPUs that Apple uses for desktop systems and you tried to swing the conversation around to PowerPC CPUs for embedded systems.
- You incorrectly implied that the PowerPC CPUs used in your household appliances and comms devices are suitable for use in Apple's desktop systems.
- You speculated that PPC architecture CPUs for embedded systems cost less than desktop CPUs from Intel and AMD -- which has nothing to do with what Apple pays for the G5 PowerPC chips used in their desktop PCs.
- When the part cannot be used for your intended purpose, it has no "bang for the buck" in that application.
- You falsely accused me of saying that the PPC is inadequate for a desktop -- and then weren't man enough to admit that your claim was untrue.
- Your claim that embedded systems engineers normally choose CPUs based on the "$ per MIP[sic]" was simply wrong. Embedded systems engineers look for the lowest cost system implementation that meets the design goals -- not the most CPU performance per dollar.
Certainly seems that I hit a nerve there.You didn't "hit a nerve." You just pissed me off by falsely attibuting statements to me.
Now, answer the points or go away and quit wasting bandwidth.
I wrote:
Only a small percentage of embedded systems designers care little about CPU performance (as if MIPS were a measurement of that anyway!).
I meant to write:
Only a small percentage of embedded systems designers care about CPU performance (as if MIPS were a measurement of that anyway!).
You are in way over your head. You don't know the difference between an architecture and a specific CPU. We talk about the G5 CPUs that Apple uses for desktop systems and try to swing the conversation around to PowerPC CPUs for embedded systems:
.75 MIPS or 1000 MIPS. It's totally irrelevent.
"PowerPC is used in a lot of embedded systems. The next generation of video games consoles are exclusively PowerPC. I've got PowerPC chips in my household appliances and various comms devices."
So what? Are you suggesting that the PowerPC used in your household appliances comms devices are suitable for use in Apple's desktop systems? I have a couple of Garmin GPSs that use 80186 family CPUs and the CPUs only cost a few bucks. That doesn't mean that those chips are suitable for use in desktop systems.
Then when several people point out that PowerPC is cheaper than AMD or Intel you say the price is irrelevant and what matters is "bang for buck".
No, you speculated that the low-end PPC architecture CPUs for embedded systems cost less than desktop CPUs from Intel and AMD, which is irrelevent to the discussion.
Then when I point out that price can never be irrelevant when you're talking "bang for buck"
You didn't point anything out. You made an incorrect assertion. When the part cannot be used for your intended purpose, is has no "bang for the buck." If you are designing a passenger car, it doesn't matter how cheap a Weed Whacker engine is. 1 hp isn't going to power the car and a PowerPC CPU out of a thermostat isn't going to power your desktop computer.
you go back to your initial argument that the PPC isn't adequate for a desktop.
No, I never claimed that the PowerPC architecture was inadequate for desktop use. You are either a liar or an idiot. I defy you to show any instance where I said that the PPC isn't adequate for a desktop. Either show me where I said that or admit that what you just wrote was untrue.
Hadn't you noticed that PowerPC desktops are rather popular? Apple is selling millions of them.
Were you actually under the impression that the CPUs in the PowerPC desktops are the same as the PPC CPUs in embedded systems?
I think I spot the underlying problem here; you don't actually know how much the PowerPC costs. The PowerPC is one of the most popular embedded CPUs - used in cars, phones, routers, consoles, multimedia devices - because systems designed around the PowerPC are inexpensive. Embedded systems are the litmus test for cost efficiency. They measure CPUs in $ per MIP; that's "bang for buck" by another name.
Your ignorance has reached new heights! First off, it's "MIPS", not "MIP" and it stands for Million Instructions Per Second. Only a small percentage of embedded systems designers care little about CPU performance (as if MIPS were a measurement of that anyway!). They care about total system cost. For example, the AVR series of microcontrollers are incredibly popular and rather slow. Their cost per MIP is horrible when compared with a desktop CPU. But they include such things as A/D converters, built-in flash memory, built-in RAM, and built-in EEPROM. If they are designing a setback thermostat with 8 buttons and a 12 character display, they care about cost, not MIPS per dollar. It doesn't matter to them if the CPU is
Are all the embedded designers stupid?
No, but you clearly don't understand anything about what we do.
Would they save more money by listening to you?
That's why I've been paid to design, and review designs for, embedded systems.
Have fun with your stupid register-starved CPU which needs 5 times more power to achieve the same computing performance.
.c file it finds while installing.
.c file it finds while installing", that's just plain hyperbole and stupidity.
Consumers of PCs don't buy them based on their register complement or their CPU wattage requirements. They buy them based on price, performance, and features.
While you are at it, load a niche operating system from Redmond that can hardly interoperate with anything different and finish it with a shitvidia video card with super-proprietary drivers that will delete every
Niche? I suggest that you invest in a dictionary. Microsoft is the dominant OS on desktops today and is probably the dominant server OS, too. As to your silly comment that it cannot "interoperate with anything different", open your eyes. It can operate as a web server, mail server, FTP server, Telnet server, SSH server, etc. It can print to Unix LPD printers, access servers running on Linux systems -- all using standard TCP/IP, UDP, and other standard protocols. Your inability to configure Windows to interoperate speaks more to your lack of experience than flaws with the Windows OS.
As to your comment than drivers for certain popular video cards "delete every
Bzzt.
Bzzt yourself.
There are more PowerPC architecture chips floating around in the world than there
are and ever will be x86 desktop processors.
So what? Some slow, cheap, embedded PPC architecture CPU isn't going to have any effect on the R&D for desktop CPUs. That's like arguing that the 80186 CPUs that can be found in everything from GPSs to fishfinders are funding the R&D for the P4.
IBM afford the R&D by being IBM -
a company people keep saying "is so small and insignificant compared to Intel"
made $1.8bn profit last year which was a little more than Intel did, and have a
total revenue of almost triple Intel's.
And they didn't become that big by investing large amounts of R&D into low-volume CPU lines.
IBM's fortunes rest in many many other places than "in boxed and OEM processors
for the desktop PC market". Market share is moot
That's simply ridiculous. Market share is what drives R&D. IBM is not going to divert money from a profitable line if minicomputers to fund a low-volume, low-profit chip line like the one used by Apple.
- you could say that Intel have
a much lower market share than AMD in some areas, does that make AMD the R&D
behemoth that people should be entertaining for a switch?
In many areas, yes. That's why you're seeing AMD being adopted in areas like servers.
If Apple switch to x86 then I will eat my laptop.
I have no desire to see you die -- just to admit that you were wrong.
The news articles clearly state
that they are switching to Intel Chips - nobody has yet dared say that this means
Intel IA-32 with or without EM64T. I wonder what Intel could produce that would
replace IBM's efforts in this field.
The x86 architecture, while not elegant, is extremely well-tuned and refined and is an viable choice for Apple. It's not like the IBM chips are outrunning the latest Intel and AMD offerings.
In fact, I would argue that it's the only rational choice for Apple. Apple needs to adopt technology that already has a big market share. That's why it was smart for them to adopt PCI, IDE, USB, and other technologies for which the R&D cost was shared with Apple's competitors. The big problem Apple has now is that IBM doesn't sell enough PPC Gx CPUs to support R&D that would be needed to keep them competitive with Intel and AMD's x86 offerings. Going to another "non-standard" CPU architecture would leave Apple in the same position that they are today.
The PowerPC is certainly adequate for a modern desktop system. The popularity of Apple desktops is proof enough of that.
Yes, but it has poor bang for the buck because it costs considerable more than an AMD or Intel CPU which has equivalent performance in a desktop system.
I don't see how it's possible for you to argue "bang for buck" while disregarding the price.
0 / any$ = 0. If the CPU has no value in a given application, it does not matter how cheap it is. A Zilog Z80 can be had for probably $5. That doesn't mean that it has great bang for the buck if your goal is to build a desktop system. So it's irrelevent to this dicussion. If Apple could build a system with a $20 gaming console PC and it satisfied most users, then that would be great. But they can't. So let's ignore those, ARMs, Atmel AVRs, Zilong Z80s, and every other CPU which really isn't viable for a modern desktop PC. They are irrelevent to this discussion.
Of course, let's not forget this question: outside of the gaming arena, just how fast does a personal computer have to be, anyways?
My 4mhz Z80 CP/M system does seem a bit slow now...
It all depends on what you expect of it. If the GUI interface is slow on the OS, it can be annoying -- and the OSs are getting more and more demanding all of the time.
I don't think performance is necessarily the driving force behind Apple's decisionmaking, which means there has to be something else behind it.
Yes: Price. If Apple has to spend $200 to get a CPU that performs as well as the one that Dell pays $100 for, then they will have trouble competing. Most people don't need the speed that they have in their PC, but it was a selling point. It's like the Suzuki Hayabusa. It will do 186mph. How many people need to go that fast? Okay, now how many people bought the bike because it can go that fast? I bet the second number is much higher than the first.
Basically this is the same argument for ARM. The ARM is not a fast processor, but ARM is used in lots of systems because it represents exceedingly good "bang for buck".
But it doesn't mean that it's got adequate performance for a modern desktop system. So, in that usage, it has incredibly poor bang for the buck, regardless of price.
I don't disagree with you about the sluggish performance of PowerPC. If you can afford it then you will get much better performance from a P4 than from a G5. The Apple fanboys annoy the hell out of me as well. I'm running a G4 here but I've no allegiance to the cult of Apple.
Nor do I have any allegiance to Intel (as I compose this on my AMD-powered system). I get annoyed with all of the fanboys on all sides. It's a CPU, not a girlfriend. You don't owe it loyalty, fidelity, and love. It's like getting misty-eyed about the brand of RAM in your system.
So after a long hard day at work, you like to come home and kick back with a game or two.
Yep. Usually on my PC. I just can't get into game consoles. I find the games for PCs better and the graphics from my ATI video to be superior. I also like the upgrade rather than replace philosophy with PCs. If I get a new PC, my old games will still work.
It's little wonder that IBM found themselves with more interesting things to do than ship apple their G5 chips.
Sorry, I didn't realize that was the point of what you were saying. But it's a very good point indeed. IBM will invest where they think they can make the most profit.
Shipping out volume in Linux servers, XBox 360s, PS3s and Nintendo Revolutions would be my first guess as to how they afford the R&D.
:)
When they get a contract for the XBox 360 or some other console, they don't have to improve the CPUs constantly. The first one and the last one are probably going to be about the same. Nor do they have to compete with the desktop offerings from Intel and AMD. And really, how many PowerPC Linux servers do they really sell compared to Intel and AMDs sales in the server market?
IBM doesn't build these chip fab facilities just for the tax writeoff
That's exactly what the heart of the problem (for Apple) is: IBM is looking at this as a business and isn't willing to invest huge sums in R&D to compete with Intel and AMD. Apple wants a CPU that lets them compete with Windows PCs. They want to be able to compete on both performance and price. Without similar R&D expenditures, so IBM has not been able to offer competitive performance to Intel or AMD at the same price point.
This isn't a slam at Apple or IBM nor is it a form of high praise for Intel (I'm really no fan of Intel and prefer AMD). It's just economic realities. If Apple had 50% of the PC market, then IBM could afford to do a lot more R&D and might well come up with a CPU that was considerably faster than the Intel or AMD offerings.
Cutie Pi 3.1415926535897932384626433832795028841972....
:)
Actually, the last digit shown should be a 1, not a 2, if you are going to end with "...". The 2 is the result of rounding up because the digit that follows is a 6.
No, I didn't have to look that up.
Uhmmmm, ever hear of embedded processors???
Yes, I'm an embedded systems engineer with over 20 years of experience.
Are you suggesting that IBM is going to take money from profitable embedded processor lines and plow that into R&D on desktop PowerPC CPUs? I somehow doubt it. Most of it doesn't apply anyway. In Intel's case, many of their embedded CPUs are based on now-outdated desktop technology, so while the desktop R&D eventually benefits their embedded processors, it seldom goes in the other direction.
Also, you will find that Atmel, ARM, Zilog, Microchip Technology, Motorola, and others, who are not big players in the desktop market, have a huge share of the embedded systems market. You also have to consider the dollar figures. When Intel or IBM sells a desktop CPU, they probably get about $100 or more for it, on averate. The average embedded CPU is less than $15.
What an angry little man you are...
In answer to your question, I'm an engineer. I get my chip info from the vendors and I get my benchmarks from independent, respected benchmarks like SPEC (Standard Performance Evaluation Corporation). Where do you get your chip info?
but I don't see what Intel has in 2006 that IBM can't match, or AMD, or whoever.
Intel has the market share so they have the R&D money as well as economies of scale. The PowerPC line is like the Sun SPARC line -- they have a limited market share and they can't afford the R&D to keep up with Intel and AMD when it comes to performance and price. Apple has less than 2% of the PC market. Intel has about 80%. When Intel sells about 40 times more CPUs, how can IBM afford the R&D to stay competitive?
Dream on x86 fanboys...
:)
Oh, if only those of us with x86 Intel and AMD CPUs could have the sluggish performance and high prices that one can get with a PowerPC system. I feel the jealousy welling up inside of me...
Seriously, grow up. It's just a CPU. Intel's CPUs offer more bang for the buck than IBM's PowerPC CPUs. It's not surprising given how many more Intel sells and how much more they have for R&D as a result. If the story is true, then Apple recognizes this and realizes that they can make more money with Intel CPUs while giving their customer base better performance. It's not religion. It's business. Just like the past Apple decisions to support PCI bus, IDE drives, USB, etc. Apple is a for-profit company and they base their product design decisions on that.
Also, not too many pieces of PC hardware at that price point have Firewire 400 and DVI.
And you think that Linux is the best OS to take advantage of those interfaces? Do you think that Linux is going to offer better, more polished apps than OS/X?
Installing Linux on the Mac Mini is like buying a Mercedes and then replacing the engine with one pulled from a Saturn.
That's not confusing, that's just faulty logic in which only a licensed taxi service may contract to drive you to the airport (or wherever).
No, it's not faulty logic. I never claimed that only a licensed taxi could drive you to/from the airport. If you want to pay your next door neighbor to drive you there, you can. But your next door neighbor can't legally hang out at the airport offering to drive random strangers home for a fee.
Completely contrived example: by default it is illegal to hire or otherwise engage someone to do something illegal. That's why the law punishes the guy who hires the hit man as well as the hit man himself. Prostitution is illegal, which is why solicitation is illegal as well.
But you are ignoring the point: That you are legally permitted to do something for yourself does not imply that you can legally hire someone else to do it for you -- or that someone is legally permitted to provide said service for a fee.
By your comprehension of a derivative work, it is illegal for me to let my next door neighbor use the DVD remote when watching a disk at my place.
No one knows my comprehension of "derivative work" better than I do, and I can state, categorically, that I never thought illegal the hypothetical situation you just posed.
Ever heard the term "reasonable man" as it applies to the law? The law is not binary. It recognizes shades of gray. Judges use judgment when making decisions: Did the framers of copyright law intend to leave a loophole for on-the-fly creation of derivative works? Did they want a for-profit third party to be able to provide the kind of service that ClearPlay does? I don't think so. If they prohibited companies from selling you a derivative work on film, tape, LP, or other media, why do you think that they would have wanted a company like ClearPlay to provide on-the-fly editing for profit? That's the key question: What is the intent of the copyright law? Obviously, this administration felt that they had to specifically make a law permitting what ClearPlay is doing, so they must have thought that it was either previously illegal or of questionable legality.
If there were any credit to the claim that this was about copyright violation, this matter would never have gotten as far as it has - ClearPlay doesn't have that kind of legal muscle.
But you forget that these are the same right-wing zealots who held Congressional hearings about Janet Jackson's nipple being flashed for a split second. They are the same right-wing zealots who levied record fines for "indecency" against radio and television stations. This is how they kiss-up to the ultra-conservative Christian voters.
This is about your rights as a consumer to do what you will with your legally acquired DVD movie. And that includes the right to hire a third party to prepare a list of scenes to be cut or bleeped. If it's legal for me to do it myself, why isn't it legal for me to hire someone else to do it for me?
The first sale doctrine gives the purchaser the right to create derivative works for their own use. Clearplay is not the purchaser of the DVD and, therefore, does not have those rights. It is legal for you to hire a third party to edit your DVD, but it is not legal for the third party to do the editing. If that's confusing, think of it this way: There is no law against you paying someone for a ride to the airport. But there is a law against that person operating an unlicensed taxi servive.
That you can legally do something yourself does not imply that you can legally hire someone to do it for you. You can bring youself to orgasm legally, but you cannot legally hire someone else to bring you to orgasm.
The issue is the rights of the consumer versus the rights of the producer. All this talk about ClearPlay distributing a "derivative work" ignores the fact that every ClearPlay customer has bought or rented the DVDs they play on their ClearPlay equipped DVD player.
It's not being ignored at all. But, since the consumer is not watching the DVD that they bought and is, instead, viewing a derivative work put together on-the-fly, it's a side issue.
But they aren't doing that: they are proposing a service that allows people who have legally bought or rented a DVD to exercise their legal rights in choosing how they watch their DVD.
No, they are providing a service which produces an on-the-fly derivative work based not on what the consumer wants, but, instead, based on what ClearPlay thinks is appropriate (divided into categories). Is a slap against the cheek of a hysterical person violence? Is a mother nursing a baby sexual content? These are decisions being made by the movie editors at ClearPlay, not by the consumer.
At least that's the way the most business friendly congress in recent history seems to see it.
There's no doubt that the right-wing zealots in Congress, many of whom have been whining about the evil, liberals in Hollywood, think that it's a good idea. They think that levying millions of dollars in fines because of one exposed nipple is reasonable, too, so it's not like Congress has the most progressive values in the world. I think that the Right Wing members of Congress are like the O.J. jury -- they are using the law to send a message rather than trying to make the right decision.
Fine. You go to 290 million people in the U.S. and educate them -- every man, woman, and child -- on how to deal-with, operate, and upgrade computers.
It's not my job to personally educate everyone in the U.S. about computers. Just as it's not my job to teach them traffic laws and how to operate a motor vehicle -- yet we expect them to learn those thing before we license them to drive.
God knows that learning about computers should be the key goal in everyone's life. The guy investigating prostate cancer online after bad news from his doctor? He should stop what he's doing and take lessons from you about computers.
Again, I'm not offering lessens. The guy investigating prostate cancer online should have learned about computers long before being diagnosed with prostate cancer. That's like saying that he should be allowed to drive to the library on the wrong side of the road at 90mph because, damnit, he has prostate cancer and doesn't have time to learn about traffic laws and cars.
Everyone should not have to know about everything just to avoid being victimized.
People shouldn't have to take self-defense courses and carry pepper spray. They shouldn't have to worry about identity theft. No one should get prostate cancer. Life's not fair. Deal with it.
This sucks. My vegetarian friends will soon be telling me that the therizinosauroids would still be alive today if they hadn't started eating meat.
'We have to reduce the complexity. People do not want privacy software, firewall, antivirus, spyware, adware and blended threats. The average individual is saying I don't want this anymore. Just fix it. What we are doing is reducing the complexity.'
I am sick and tired of companies pandering to people too lazy to learn anything. Computers are complex and they aren't getting simpler -- because people demand more and more from them. If someone doesn't want to invest the time to master the use of a computer, then they shouldn't operate one.
These are the same stupid whiners who are in tears when their hard disc crashes because they never backed up anything. They are the ones who are standing in line in front of us at the computer store having hissy fits because they want the techs to install a DVD-R/W drive while they wait -- because there's no way that they would consider learning how to put in four screws and three cables (power, IDE, and audio) themselves. They are the ones who install adware on their systems and are then shocked at the results. They are the ones who stare blankly like deer looking at headlights when the salesman asks if they have USB ports into which their new digital camera can be plugged.
All that this kind of initiative by Symantec does is help the ignorant remain so while increasing the under-the-hood complexity even further. So when something in the automated mass of hundreds of megs of files breaks, the user doesn't have a prayer of being able to fix it.
If you aren't willing to learn how a radio works, then the FCC won't give you a HAM radio license. If you aren't willing to learn the rules of the road, the DMV won't give you a driver's license. We need a computer license and people who won't learn the basics about computers shouldn't be allowed to own and operate a computer.
No, the final product may be a derivative work, but the list of time codes to blank or bleep aren't. By your standards, *any* kind of editing software or hardware - including the fast forward and rewind buttons on the remote control - would constitute a "derivative work". That's plainly ridiculous.
It is ridiculous. And it's not my contention.
The final product is the issue. Customers aren't paying to have a list of timecodes and edit instructions. They're paying to see a version of Spiderman, The Matrix, or Titanic without the "dirty bits." So what they are buying are derivative works.
If they want to fast-forward or hit mute based on some cue-sheet sent to them from ClearPlay, that's fine. Then the cue-sheet really would be the issue because that's what the consumer would be paying for. Now they're skipping that and going right to paying for a derivative work to be presented on their television screen.