My friend claims he yells "Asshole alert!" into the receiver, then slams down the phone every time he gets a call from Aerotek/Maxim... My personal experience has been that any time I spent talking to Aerotek was a waste of time, and I felt that at least one company thought that being submitted by Aerotek was a major count against me, since their previous Aerotek consultant had walked off the job after 2 weeks. Your mileage may vary...
From what I've read about Transmeta, here are my guesses as to what they are doing:
The patent is somewhat of a red herring as any method for determining whether a physical address is memory or MMIO could be better used to determine whether an address resides in local memory or in shared memory in a multiple-processor system. This would be a much more useful application of this technology!
Transmeta is almost certainly working on a CPU that supports parallel processing very well (I beleive that Linus was hired for his experience with SMP). Probably using both large on-chip cache memory and large shared memory address space. I would assume that scalability up to massively parallel would also be a design goal. To make this chip useful, they also need a compiler that can generate code to take advantage of parallel processors.
Transmeta may or may not be working on emulating existing CPUs and OSes. The way to emulate an existing CPU is not by changing your microcode, but rather by using the type of technology used in Java JIT compilers. Instead of just translating Java byte code to native machine language, this could be extended to also translate X86, 68000, or PowerPC instructions on an on-demand basis into the new CPU's native instructions. If the JIT compiler also knew how to parallelize the tasks very well, it would then be possible to emulate older processors in faster-than-real-time by using parallel processors! But to really make this useful, you also have to emulate the run time environment, e.g. Linux, Windows, or MacOS. Net result would be a machine that could simultaneously run applications written for different operating systems and CPUs. Way cool, if you can get it to work! As a minimum, I would expect at least Java to be supported in the first release, as this immediately gives you loads of applications for the new chip.
Disclaimer: I am not privy to any inside information about Transmeta, and probably know less about this than you do! However, I would appreciate any feedback you can give me on how likely/unlikely my speculation is to be true.
Question: Does anybody know of any research done on the parallelizability of Java byte code, and if so, where are the researchers now?
That's based on a Dell Dimension V 333Mhz Celeron. Yes, I added the 10/100 ethernet card, and the monitor was included. I reduced the RAM from the standard 64MB to 32MB to equal the iMac. Unfortunately, I couldn't get a disk any smaller than 8.4 GBytes. Oh, and of course the PC has a floppy drive! And yes, saying a 333 Mhz Celeron is "comparable" is probably cheating, but I haven't seen any benchmark results that I trust yet. Anyway, the point was that the iMac and the PC are approximately equivalent in price/performance, therefore it is misleading to keep implying the the iMac is somehow a better value than a PC.
I stand by my original statement, Mr. AC
on
Apple Announcements
·
· Score: 1
What planet have you been living on? The Celeron 300A, 333, 366, and 400 ALL have onboard cache. Which is not to say that it's a better processor than a power PC (it isn't). Next time, check your facts before you make an ass of yourself in public.
The quote I questioned was in reference to the iMac at $1199. There is a significant leap from "a G3 400 is twice as fast as a Pentium II 450" to "a G3 266 is faster that the Dual PII 450 system you could buy for $4800". I love the Power PC and I'm sure it is faster than a Pentium II at a given clock rate. However, I see no evidence the the disk, video, or bus are any faster than a comparable PC, and the video is slower than most PCs. Which means in real life, you should see little difference between an iMac and a comparably priced PC. I like Apple and like to see them succeed. I just have a problem with them trying to bullshit me. Certainly, you misunderstood my statements; what exactly in it do you consider FUD? What part of is is less accurate than the Apple Marketing hype?
A Celeron 333 system from Dell with comparable features (except for bigger 8.4 GByte hard drive) costs $1268. Congratulations to Apple for bringing the price of a Mac down to the same price as a PC, but I certainly wouldn't call these puppies less expensive than a PC!
its processor humbles the Pentiums in office PCs four times its price
Sorry, but I can't even find an office PC that costs $4800, unless its a server. And while that 266Mhz G3 might be faster than the 266Mhz Cyrix in a $400 PC, I suspect it would be sadly embarrased by a 450Mhz PII. Yeah, they got pretty colors, but if they want to maintain some semblance of credibility, they should install some bullshit filters on their marketing dept. output!
Based on price/performance, I'd say Linux has one of the most effective marketing departments in existance! Think what they could accomplish if they actually spent money on marketing! Word of mouth is not only the cheapest, it is also the most effective form of marketing!
Seriously though, the mindset that good marketing can make up for bad technology is dangerously obsolete in the age of the Internet. Free trial downloads means anyone can instantly evaluate all competing software based on merit alone. Even Micro$oft's billions can't squelch the voice of dissent on the internet (although astroturf campaigns may be somehat effective -- M$ can afford to pay A LOT of people to post.) It may be that sooner than you think, your marketing budget won't mean a thing, unless you can deliver software as good as or better than your competitors. The Internet changes everything -- you either learn to play by the new rules or go home!
Slashdot Mirror
;-)
My friend claims he yells "Asshole alert!" into the receiver, then slams down the phone every time he gets a call from Aerotek/Maxim... My personal experience has been that any time I spent talking to Aerotek was a waste of time, and I felt that at least one company thought that being submitted by Aerotek was a major count against me, since their previous Aerotek consultant had walked off the job after 2 weeks. Your mileage may vary...
The patent is somewhat of a red herring as any method for determining whether a physical address is memory or MMIO could be better used to determine whether an address resides in local memory or in shared memory in a multiple-processor system. This would be a much more useful application of this technology!
Transmeta is almost certainly working on a CPU that supports parallel processing very well (I beleive that Linus was hired for his experience with SMP). Probably using both large on-chip cache memory and large shared memory address space. I would assume that scalability up to massively parallel would also be a design goal. To make this chip useful, they also need a compiler that can generate code to take advantage of parallel processors.
Transmeta may or may not be working on emulating existing CPUs and OSes. The way to emulate an existing CPU is not by changing your microcode, but rather by using the type of technology used in Java JIT compilers. Instead of just translating Java byte code to native machine language, this could be extended to also translate X86, 68000, or PowerPC instructions on an on-demand basis into the new CPU's native instructions. If the JIT compiler also knew how to parallelize the tasks very well, it would then be possible to emulate older processors in faster-than-real-time by using parallel processors! But to really make this useful, you also have to emulate the run time environment, e.g. Linux, Windows, or MacOS. Net result would be a machine that could simultaneously run applications written for different operating systems and CPUs. Way cool, if you can get it to work! As a minimum, I would expect at least Java to be supported in the first release, as this immediately gives you loads of applications for the new chip.
Disclaimer: I am not privy to any inside information about Transmeta, and probably know less about this than you do! However, I would appreciate any feedback you can give me on how likely/unlikely my speculation is to be true.
Question: Does anybody know of any research done on the parallelizability of Java byte code, and if so, where are the researchers now?
That's based on a Dell Dimension V 333Mhz Celeron. Yes, I added the 10/100 ethernet card, and the monitor was included. I reduced the RAM from the standard 64MB to 32MB to equal the iMac. Unfortunately, I couldn't get a disk any smaller than 8.4 GBytes. Oh, and of course the PC has a floppy drive! And yes, saying a 333 Mhz Celeron is "comparable" is probably cheating, but I haven't seen any benchmark results that I trust yet. Anyway, the point was that the iMac and the PC are approximately equivalent in price/performance, therefore it is misleading to keep implying the the iMac is somehow a better value than a PC.
What planet have you been living on? The Celeron 300A, 333, 366, and 400 ALL have onboard cache. Which is not to say that it's a better processor than a power PC (it isn't). Next time, check your facts before you make an ass of yourself in public.
The quote I questioned was in reference to the iMac at $1199. There is a significant leap from "a G3 400 is twice as fast as a Pentium II 450" to "a G3 266 is faster that the Dual PII 450 system you could buy for $4800". I love the Power PC and I'm sure it is faster than a Pentium II at a given clock rate. However, I see no evidence the the disk, video, or bus are any faster than a comparable PC, and the video is slower than most PCs. Which means in real life, you should see little difference between an iMac and a comparably priced PC. I like Apple and like to see them succeed. I just have a problem with them trying to bullshit me. Certainly, you misunderstood my statements; what exactly in it do you consider FUD? What part of is is less accurate than the Apple Marketing hype?
A Celeron 333 system from Dell with comparable features (except for bigger 8.4 GByte hard drive) costs $1268. Congratulations to Apple for bringing the price of a Mac down to the same price as a PC, but I certainly wouldn't call these puppies less expensive than a PC!
Pentiums in office PCs four times its price
Sorry, but I can't even find an office PC that costs $4800, unless its a server. And while that 266Mhz G3 might be faster than the 266Mhz Cyrix in a $400 PC, I suspect it would be sadly embarrased by a 450Mhz PII. Yeah, they got pretty colors, but if they want to maintain some semblance of credibility, they should install some bullshit filters on their marketing dept. output!
You know, maybe you could take a few hundred out of that $500K advertising budget, and maybe hire a web developer or two... just a thought... ;-)
Seriously though, the mindset that good marketing can make up for bad technology is dangerously obsolete in the age of the Internet. Free trial downloads means anyone can instantly evaluate all competing software based on merit alone. Even Micro$oft's billions can't squelch the voice of dissent on the internet (although astroturf campaigns may be somehat effective -- M$ can afford to pay A LOT of people to post.) It may be that sooner than you think, your marketing budget won't mean a thing, unless you can deliver software as good as or better than your competitors. The Internet changes everything -- you either learn to play by the new rules or go home!
How 'bout a link to a $400 box that does NOT come with Win98 and M$[rarely]Works preinstalled?