If corn -> ethanol -> hydrogen -> fuel cell takes off those diesel powered farming machines will be relegated to a bootstrapping process. Keep in mind that the size, mechanical simplicity, and relative quiet of electric motors make them naturally preferable to internal combustion engines. There aren't electric tractors today because gasoline has a much better energy density than any battery yet made. The "Hydrogen economy" need only equal the economic efficiency of the oil economy to become an inevitability.
I think you guys are over-thinking and/or not addressing the question. Overclocking aside, the guy is wondering if digital circuits that use light will generate heat like electric ones do -- through simple Resistance -- or if they would behave like a PIII made from superconductors.
This is a much simpler question that I too am curious about. When Intel does their HDTV demo will the 5 mile coil of fibre get warm -- however imperceptibly -- because photons are "flowing" through it? If so, would that show up dramatically at small scales?
The processor has 64 GPRs, with the following specialized semantics:
* %r63 (%zero) always reads 0 when used as a source operand
* %r62 (%sink) is a discarded destination (e.g., for compares); it is never read
That makes sense, but there is still the flexibility and spring of Titanium that comes into play when comparing the two. It's easy to trace LCD "trip out" mood ring patterns with a finger tip on the lid of a TiBook for the same reasons Titanium makes excellent eyeglass frames. It's got a lot of give.
Oh, wait. I see what you're saying. The older form would still have been less solid feeling with aluminum. Check.
Yeah. Ti sounds great for marketing but it's a poor choice of material -- especially for the LCD enclosure. I'm glad they went aluminum, but I wish they had retained the older form and style.
I like the fact that my TiBook makes a boot sound, but I can't stand the sound Apple chose. It's odd that this topic has come up. Just last night I was googling for firmware hacks that would let me change that sound to something less annoying (to me personally).
When this book was first published I read every discussion of it I encountered in the non-scientfic print media. Wolfram's biography and academic career are certainly interesting. One detail in all the gloss stuck with me though. The man has never read his mother's publications. She is (was?) a philosophy don at Oxford. I don't understand how any intellectually curious individual could outright dismiss the work of a highly accomplished parent as being beneath his time. Even if his mindset is such that he dismisses the entire philosophic cannon outright one would think he'd have a bit of curiosity left over for something that was probably left around the house.
You don't even need to cp/boot/config-foo to./.config. The new kbuild will grab whatever it thinks is appropriate (presumably from/lib/modules/`uname -r`/build/.config) and plug it in to a fresh source tree. This actully freaked me a bit the first time I compiled a 2.6 kernel as I hadn't noticed what was going on and wondered how the damn thing "knew" that I had certain hardware.
Any unconfigured 2.6 source tree will avoid sticking you with a default config automagically if it can glean something about your current config even if you're running 2.4.
A previous poster noted that his 2 GHz 970 dissipated 24 Watts. While googling around for more info on 970s and heat I found a short blurb about the 970FX which noted that this more recent 970 was the beneficiary of a die shrink done on a 90nm process. As an aside, the author noted that Intel's die-shrinks were working against them thermally and attributed this to IBM's use of SOI, which Intel apparently doesn't use.
I don't know how valid that claim is -- though it does make sense if SOI makes that serious a dent in current leakage -- but if it's true Intel does have an option to tap in the future.
This old fashioned romanticism is touching and quaint, but really now, you lover boys are screwing things up for the rest of us. It's hard enough pleasing a modern woman with all the mushy "emailing" and "hanging out while I build a computer". I know you're just trying to warm her heart, but we're talking Men and Women here. It's war son and they'll take a mile if you give 'em an inch. We all managed to adjust when they started wanting us to whisper SSE opcodes in their ears but I think this shit might break us.
Back off dude. If we stick together they'll never know the difference.
Here's another wierd one for you guys to listen to; the video card in your computer. I'm not talking 'bout yer CRT. I really mean your video card. The first time I noticed this it was on a Matrox G400. I was screwing around with framebuffer consoles and noticed I could control the (extremely faint) noise by scrolling the buffer. The ear-to-card-with-the-case-door-off test cinched it. I figured it had something to do with the post-DAC analog output circuits or somesuch, but I can hear the same thing on my TiBook which uses LVDS or TMDS to talk to the LCD. Any theories here? I'm talking about a noise that shows up with heavy blitting. Is it possible for a motherboard's traces to hum just like a transformer? Or even a chip?
Not necessarily. It looks like the parent might have unpacked the makeself archive (get the pkg0 files -- they're much smaller as they come without precompiled modules for stock distro kernels) to compile the module by hand. The makefile should detect KVERS and KSRC in the environment so you can export those values by hand. That will ensure you're really getting what you want and it must be done if the automagic detection fails. The admonition against a/usr/src/linux symlink has sound merit as it's easy to break kernel specific compiles on a system with more than one kernel source tree. `make modules_install' puts a link to the source as it is at/lib/modules/$(uname -r)/build so linux -> linux-2.4.24 isn't really necessary.
I don't think MIDI -> standard notation would work very reliably, completely, or efficiently. It would be like using.wav files to represent a collection of sine waves (all you need is an FFT browser plugin), except.wav could actually represent all those sines for a given spectrum without losing information. There is much in human notation that MIDI ignores and resynthesizing that information would be needlessly expensive (computationally) or fundamentally indeterminate. E# and F are different notes to a good violinist playing tonal music. Does MIDI preserve that distinction? If it does do you want your webserver to spend all it's time crunching MIDI->Lillypond->TeX->png to end up with an unscalable image in a site visitors browser? If these are worthy points in this debate will MusicML stop you from using the tools and formats you use today?
When I did this with my TiBook, it ran for about 1:45 after the menu bar said it was at 1% power.
Shouldn't that mean that your TiBook is now mis-calibrated? One would think that if the power indicator reads 1%, then you've got 1/100th of the total running time left. Of course, you could just have a lucky battery that can last over 7 days...
Slashdot Mirror
No comment.
This is a much simpler question that I too am curious about. When Intel does their HDTV demo will the 5 mile coil of fibre get warm -- however imperceptibly -- because photons are "flowing" through it? If so, would that show up dramatically at small scales?
Wow. /dev/zero and /dev/null in silicon.
Oh, wait. I see what you're saying. The older form would still have been less solid feeling with aluminum. Check.
He was talking about the TiBooks, which did use Titanium in the casing, not the newer Powerbooks made with aluminum.
Yeah. Ti sounds great for marketing but it's a poor choice of material -- especially for the LCD enclosure. I'm glad they went aluminum, but I wish they had retained the older form and style.
I like the fact that my TiBook makes a boot sound, but I can't stand the sound Apple chose. It's odd that this topic has come up. Just last night I was googling for firmware hacks that would let me change that sound to something less annoying (to me personally).
When this book was first published I read every discussion of it I encountered in the non-scientfic print media. Wolfram's biography and academic career are certainly interesting. One detail in all the gloss stuck with me though. The man has never read his mother's publications. She is (was?) a philosophy don at Oxford. I don't understand how any intellectually curious individual could outright dismiss the work of a highly accomplished parent as being beneath his time. Even if his mindset is such that he dismisses the entire philosophic cannon outright one would think he'd have a bit of curiosity left over for something that was probably left around the house.
You don't even need to cp /boot/config-foo to ./.config. The new kbuild will grab whatever it thinks is appropriate (presumably from /lib/modules/`uname -r`/build/.config) and plug it in to a fresh source tree. This actully freaked me a bit the first time I compiled a 2.6 kernel as I hadn't noticed what was going on and wondered how the damn thing "knew" that I had certain hardware.
Any unconfigured 2.6 source tree will avoid sticking you with a default config automagically if it can glean something about your current config even if you're running 2.4.
So they make a 15.5% profit. Is that an uncommonly low number for a company?
I don't know how valid that claim is -- though it does make sense if SOI makes that serious a dent in current leakage -- but if it's true Intel does have an option to tap in the future.
Whoops! You guys are right. Plex86 has changed course, and yes, it is disappointing.
That's why we shouldn't regard contemporary x86 chips as being hobbled by a legacy. We might as well think of them as hardware emulators.
That's too literary. mod_lo.
Plex86 is the VMware alternative.
Uhh...
Iff yew byte intoo ah turd, itss gowing tue tayst lyck schitt. Know ahmownt uf sugar aynd spise iz gowyng tew mayck itt uhthorwize.
I'm sorry.
Really.
I know it shouldn't have been done...
But I did it anyway.
Back off dude. If we stick together they'll never know the difference.
Here's another wierd one for you guys to listen to; the video card in your computer. I'm not talking 'bout yer CRT. I really mean your video card. The first time I noticed this it was on a Matrox G400. I was screwing around with framebuffer consoles and noticed I could control the (extremely faint) noise by scrolling the buffer. The ear-to-card-with-the-case-door-off test cinched it. I figured it had something to do with the post-DAC analog output circuits or somesuch, but I can hear the same thing on my TiBook which uses LVDS or TMDS to talk to the LCD. Any theories here? I'm talking about a noise that shows up with heavy blitting. Is it possible for a motherboard's traces to hum just like a transformer? Or even a chip?
Not necessarily. It looks like the parent might have unpacked the makeself archive (get the pkg0 files -- they're much smaller as they come without precompiled modules for stock distro kernels) to compile the module by hand. The makefile should detect KVERS and KSRC in the environment so you can export those values by hand. That will ensure you're really getting what you want and it must be done if the automagic detection fails. The admonition against a /usr/src/linux symlink has sound merit as it's easy to break kernel specific compiles on a system with more than one kernel source tree. `make modules_install' puts a link to the source as it is at /lib/modules/$(uname -r)/build so linux -> linux-2.4.24 isn't really necessary.
Which means the mods aren't RTFA-ing if "whoring" "works".
Oh, OK. I guess I'll have to look elsewhere for humor fodder... (dang)
Classy and humorous.
Of course he has. He just didn't like the sluggish rendering and glitchiness.
I don't think MIDI -> standard notation would work very reliably, completely, or efficiently. It would be like using .wav files to represent a collection of sine waves (all you need is an FFT browser plugin), except .wav could actually represent all those sines for a given spectrum without losing information. There is much in human notation that MIDI ignores and resynthesizing that information would be needlessly expensive (computationally) or fundamentally indeterminate. E# and F are different notes to a good violinist playing tonal music. Does MIDI preserve that distinction? If it does do you want your webserver to spend all it's time crunching MIDI->Lillypond->TeX->png to end up with an unscalable image in a site visitors browser? If these are worthy points in this debate will MusicML stop you from using the tools and formats you use today?
Shouldn't that mean that your TiBook is now mis-calibrated? One would think that if the power indicator reads 1%, then you've got 1/100th of the total running time left. Of course, you could just have a lucky battery that can last over 7 days...