...If you have an old digital camera laying around...
No, I don't. I am still using the one I bought two years ago. Should I feel guilty now, because of not buying a new gadget in time? How often are you supposed to replace your digicam?
The question is also what they have measured. Situation where a reboot was really required or just simple reboots by number. Many clueless users believe that rebooting fixes many problems, and just do so without trying anything else. I am pretty sure the same userbase would also reboot Linux frequently.
The gist of that trial is that rambus patentend parts of a public JEDEC spec before they were published. And it seems very likely that the Rambus people knew of the Jedec specs. In fact Rambus has the entire memory industry against them, very unlikely they win.
No, its at 110nm. One vendor already moved to 90nm and more are likely to follow during the next month.
You can not compare modern dram to cpus. The problem with dram is often not the transistors, but the capacitors. The capacitors require extremely high aspect ratios which are very difficult to manufacture. Another problem is leakage in transistors. While relatively leaky transistors can be used in CPUs, additional effort needs to be spent in drams to avoid leakage.
Computers Nope, Konrad Zuse of germany was first. However the Nazis failed to take advantage of it.
RADAR Both sides had radar before the war, the achievement of the british was the magnetron, enable extremely high power and high frequency radar.
Supersonic 20,000lb bombs They were just scaled up conventional bombs. I fail too see the achievement. May also be interesting to note that the allies did not have any targets for 20,000lb bombs on their side.
Jet aircraft (before the Germans) No, not true. Germans were definitly first on this.
Seemingly unbreakable encryption (even today)
AFAIK RSA was developed in the 70ies..... but it was only their V2 program that was of any interest to anyone after the war
How about astonishing progress in aviation (Jet planes, Sangers concepts, guidance systems, control systems etc.), U-Boats (all modern submarines follow their concepts now),high speed roads (dont ask me but they developed special processes required to build these),chemical/process engineering (nazi germany virtually lived of coal and air!), chemistry.. etc etc.
This new Imac does only look good as long as you do not plug any cable into these misplaces connectors. How does it look with 10 cacbles coming out of the back?
Judging from the article, it sounds as if the drive would require a 50nm laser source. There are no cheap semiconductor lasers available for this length and never will be. And I doubt you'd want to have an excimer laser in your data drive. In addition to that it would imply that the disk has to be read in vacuum... Veeery fishy.
In addition I refard physorg.com as a highly unprofessional site. They spam the usenet and various web forums a lot. They also have the nerve to steal entire threads(!!) from the usenet and insert them to their forum, so it looks populated.
That has nothing to do with harvard architecture, and your 68040 wasn't a harvard arch.
That is not 100% accurate. Actually it is common to designate CPUs as a harvard architecture when they use separate data and code caches. For example it is impossible on the 68040 to modify code that resides in the code cache.
The used wavelength (afair 15.4nm) is stil far from hard x-ray. The technologies for generation, mask, and "optics" of x-ray and EUV radiation is very different.
Wait until they see how they can run most of their Windows software under GNU/Linux using Wine.
Wait until they find out they can run all Unix applications using remote X login and significantly decrease the administration cost at the same time. That is what I found in almost all companies that had to deal with some unix applications.
I think you caught the essentials. Its just a small step from strapping them all together with duck-tape. Its the epitome of a totally useless console mod, imo.
Possibly, although the Fraunhofer Institute seems to be doing it in a massively less efficient way... and your background which qualifies you do judge in this issue is?? I think the guys at the Fraunhofer institute know very well what they are doing.
Why cant RMS just try to improve the situation, make a counteroffering, give suggestions instead of ranting about everything that is not exactly along his line of view?
Of course, but at that time there was just no way of improving the x86 float performance any further. The alphas and MIPS of that time were of course faster, but also ten times more expensive. May be interesting to note that the PPC601 to 603 did not have faster FPUs either..
Btw, some background information the oh-so-inaccurate ars technica does not tell:
The Pentium I design team was headed by Vinod Dham, who left intel to form a CPU design company named Nexgen. The Nexgen design was later bought by AMD and formed the core of the K6, which was AMDs first competitive offering.
The Pentium Pro and later successors were mostly designed by a team which was acquired by intel when they bought an MIT spinoff in the beginning of the 90ies. Seems that they ran out of steam with the Pentium 4 design..
Slashdot Mirror
...If you have an old digital camera laying around ...
No, I don't. I am still using the one I bought two years ago. Should I feel guilty now, because of not buying a new gadget in time? How often are you supposed to replace your digicam?
Scientists never get old.
Two cesium atoms are walking down the street. Suddenly one stops and looks around.
The other cesium atom says "What's the matter?"
The first cesium atom replies "I think I just lost an electron!"
"Are you sure?"
"Yes, I'm positive."
The question is also what they have measured. Situation where a reboot was really required or just simple reboots by number. Many clueless users believe that rebooting fixes many problems, and just do so without trying anything else. I am pretty sure the same userbase would also reboot Linux frequently.
The gist of that trial is that rambus patentend parts of a public JEDEC spec before they were published. And it seems very likely that the Rambus people knew of the Jedec specs. In fact Rambus has the entire memory industry against them, very unlikely they win.
No, its at 110nm. One vendor already moved to 90nm and more are likely to follow during the next month.
You can not compare modern dram to cpus. The problem with dram is often not the transistors, but the capacitors. The capacitors require extremely high aspect ratios which are very difficult to manufacture. Another problem is leakage in transistors. While relatively leaky transistors can be used in CPUs, additional effort needs to be spent in drams to avoid leakage.
Funny.. because as far as I know Infineon is producing the memory for Mushkin. But don't worry, it is of exceptional quality.
Erm.. they just spend a lot of money, it must not necessarily have the expected results.
Besides, I have already seen dedicated speech recognition chips at a company presentation years ago.
It may help to use the right spelling, it is: P O M P O U S
And the allies came up with:
... but it was only their V2 program that was of any interest to anyone after the war
Nuclear weapons
True
Computers
Nope, Konrad Zuse of germany was first. However the Nazis failed to take advantage of it.
RADAR
Both sides had radar before the war, the achievement of the british was the magnetron, enable extremely high power and high frequency radar.
Supersonic 20,000lb bombs
They were just scaled up conventional bombs. I fail too see the achievement. May also be interesting to note that the allies did not have any targets for 20,000lb bombs on their side.
Jet aircraft (before the Germans)
No, not true. Germans were definitly first on this.
Seemingly unbreakable encryption (even today)
AFAIK RSA was developed in the 70ies..
How about astonishing progress in aviation (Jet planes, Sangers concepts, guidance systems, control systems etc.), U-Boats (all modern submarines follow their concepts now),high speed roads (dont ask me but they developed special processes required to build these),chemical/process engineering (nazi germany virtually lived of coal and air!), chemistry.. etc etc.
~£300 ... and in real money that is?
but really, if all of the others haven't really made a dent in the iTMS market share, how does MS intend to?.
I guess the same thing was said a while ago by the executives at netscape.
This new Imac does only look good as long as you do not plug any cable into these misplaces connectors. How does it look with 10 cacbles coming out of the back?
Ok, I change my estimate to sqrt(64000)~2^8
No, its just 0.8km since sqrt(64000) = 800
Judging from the article, it sounds as if the drive would require a 50nm laser source. There are no cheap semiconductor lasers available for this length and never will be. And I doubt you'd want to have an excimer laser in your data drive. In addition to that it would imply that the disk has to be read in vacuum... Veeery fishy.
In addition I refard physorg.com as a highly unprofessional site. They spam the usenet and various web forums a lot. They also have the nerve to steal entire threads(!!) from the usenet and insert them to their forum, so it looks populated.
That has nothing to do with harvard architecture, and your 68040 wasn't a harvard arch.
That is not 100% accurate. Actually it is common to designate CPUs as a harvard architecture when they use separate data and code caches. For example it is impossible on the 68040 to modify code that resides in the code cache.
What a load of c... inaccuracies...
The used wavelength (afair 15.4nm) is stil far from hard x-ray. The technologies for generation, mask, and "optics" of x-ray and EUV radiation is very different.
Wait until they see how they can run most of their Windows software under GNU/Linux using Wine.
Wait until they find out they can run all Unix applications using remote X login and significantly decrease the administration cost at the same time. That is what I found in almost all companies that had to deal with some unix applications.
I think you caught the essentials. Its just a small step from strapping them all together with duck-tape. Its the epitome of a totally useless console mod, imo.
Possibly, although the Fraunhofer Institute seems to be doing it in a massively less efficient way. .. and your background which qualifies you do judge in this issue is?? I think the guys at the Fraunhofer institute know very well what they are doing.
Why cant RMS just try to improve the situation, make a counteroffering, give suggestions instead of ranting about everything that is not exactly along his line of view?
$2793 - 3 nVentiv Prometeia Mach II @ $931 (1 for CPU 2 for GPUs)
What the heck are these? My computer does not need any. Besides, a 24" TFT is likely too slow for gaming.
I know I've got mine on order!
Congrats for spending +300% for +50% speed increase..
Of course, but at that time there was just no way of improving the x86 float performance any further. The alphas and MIPS of that time were of course faster, but also ten times more expensive. May be interesting to note that the PPC601 to 603 did not have faster FPUs either..
Btw, some background information the oh-so-inaccurate ars technica does not tell:
The Pentium I design team was headed by Vinod Dham, who left intel to form a CPU design company named Nexgen. The Nexgen design was later bought by AMD and formed the core of the K6, which was AMDs first competitive offering.
The Pentium Pro and later successors were mostly designed by a team which was acquired by intel when they bought an MIT spinoff in the beginning of the 90ies. Seems that they ran out of steam with the Pentium 4 design..