You're right. I was confusing loudness and volume like the parent poster, and then thought you were doing the same! Now that we know we've all been talking about the same damn thing I've got to ask: Who's on first again?
Get a life. Math skills aren't deteriorating over this colloquialism. Everyone knows that what's meant by "three times less" is "one-third". If you want to be that pendantic then attack the poor use of the language. Math has nothing to do with it.
...in general 10db is considered to sound like "a doubling in volume"...
The parent poster wasn't talking about power. Volume, or how loud something appears to the senses, is measured in sones. 1 sone approximately equals 10 dB though that varies with the frequency and magnitude of the sound.
Well, maybe leave it plugged in if your power supply has a switch on the back, but I'm not sure that's wise for ATX. I've got a VP6 based system. That board has a power led mounted on it that stays lit for a few seconds even after unplugging the power supply. Presumably the leftover juice comes from PS capacitors. Anytime I muck around in the case I unplug and wait. I slacked off on this habit once, managed to bump a molex connector just right, and toasted the nice 20GB drive that held bootsector +/boot + some other stuff. You could leave your box plugged in, but there are better ways of avoiding static cling.
Re:So is it any better to have faster AGP?
on
Tackling AGP 8X
·
· Score: 1
John Carmack has talked about the idea of generating texture maps dynamically.
Why bother? I would think that texture maps play a role similar to precompiled bsp trees.
1) Color accuracy could be adjusted with individual gain/bias controls just as in CRTs today. Adding "headroom" to the max subpixel brightness (build the display twice as bright as the control electronics will allow it to operate) would facilitate this kind of tuning and extend the lifetime of the display. Problem already solved by precedent and common sense.
2) Absolutely not. An LCD's matrix must be made on a transparent substrate. IIRC, that is why they are limited to amorphous polysilicon transitors and have a 30% yield. In contrast, any substrate that can support microelectronics can be home to an OLED. The design challanges (which I think revolve around heat) may remain, but making the support electronics shouldn't require anything that hasn't been possible years now.
3) Why would higher current be a cause of failure? That may make the active matrix harder to design but if the components fail because they were conducting electricity I don't see how lowering the current would help them last longer. If an athlon gives off 50+ watts of heat and runs on 3+ volts (argument's sake) how many amps are passing through through it? Understand I'm not saying LCD current levels wouldn't be easier for the EEs to deal with. I just don't think we'll be seeing dead pixels on an OLED monitor for the same reasons one could find a blown fuse on a power supply. 20 years from now people might have to say, "My old monitor is starting to look kinda red." whereas today they might say, "My CRT needs recalibration." or, "My LCD has a stuck pixel." Then again, there's always counterpoint #1.
Footnote: Their biggest advantage won't be just in power consumption if and when they arrive. It will be that they are emmissive, light, durable, capable of high pixel densities with rich color saturation, possibly a wider color gamut, and cheap.
... and if you go here you'll find that G3 450 achieving 62% of a P3 733's performance at cross-compiling gcc. That would make a P3 clocked at 454.5 MHz a G3 450's equal. The numbers look worse for native compilation.
Regardless of how easily DDR integrates into x86 every disscussion of the latest PowerMacs on this forum and xlr8 have claimed that Apple's new DDR boards benefit everything but the FSB and that the limitation is inherent in the G4 itself.
The AC post comparing "PhysicsScholar" to "PhysicsGenius" has hit the nail on the head. This is a pale shadow of a fine Slashdot spoof artist. Nothing to see here folks. Move along...
I just googled for some info on these formats and found this page which describes SACD's "Direct Stream Digital" encoding. It's like 1 bit encoding at 2.82 MHz where each sample means "louder" or "softer" than the previous bit. Strange, eh? It seems to have a certain analog quality in that playing a DSD file one wouldn't have to worry about word or byte alignment -- start at bit number 13 and it will still make sense...
I also found an interesting discussion of DVD-V and it's use for audio. It can master at 24-bit/96KHz using desktop pc hardware which is the apparent limit of the human ear and nearly the limit at which thermal noise caps the precision of the electronics (see the above link). Could this be an "open" alternative to these audiophile formats?
Earlier tonight I gave my ti4200 the finger test for heat and wasn't pleased. It occured to me that since the AGP slot is always topmost on the motherboard there is no reason why the GPU shouldn't be mounted on the backside of the PCB where there is more room for active cooling. As it stands now GPU heatsinks are designed to make the best of a cramped situation.
They fortified the towers against an accidental impact. Think flaps down or an engine out with a pilot unable to avoid impact. The jets that hit the towers were travelling nearly 150% of their rated speed. I agree that fireproofing could be made with more "cling", but should we blame the engineers for not safeguarding against behemoth kamakazis? Where do we draw the line that says, "This building is fortress-like enough"?
The real failures were essentially management failures that marginalized warnings, and inter-agency rivalries.
Frontline just aired a documentary called "The Man Who Knew." FBI agent John O'Neil had Bin Laden in his sights for a decade. When he admitted to himself that bureaucratic politics would forever hobble his investigation he quit and took a job as Chief of Security at the World Trade Center. His famous last words: "We're due for something big."
Slashdot Mirror
You're right. I was confusing loudness and volume like the parent poster, and then thought you were doing the same! Now that we know we've all been talking about the same damn thing I've got to ask: Who's on first again?
Good ol' Nellie Schmoe... always up fer watchin' DVDs.
Can we not say, "Y is one-twentieth the size of X"? How about, "a twentyfold reduction"?
Get a life. Math skills aren't deteriorating over this colloquialism. Everyone knows that what's meant by "three times less" is "one-third". If you want to be that pendantic then attack the poor use of the language. Math has nothing to do with it.
The parent poster wasn't talking about power. Volume, or how loud something appears to the senses, is measured in sones. 1 sone approximately equals 10 dB though that varies with the frequency and magnitude of the sound.
Well, maybe leave it plugged in if your power supply has a switch on the back, but I'm not sure that's wise for ATX. I've got a VP6 based system. That board has a power led mounted on it that stays lit for a few seconds even after unplugging the power supply. Presumably the leftover juice comes from PS capacitors. Anytime I muck around in the case I unplug and wait. I slacked off on this habit once, managed to bump a molex connector just right, and toasted the nice 20GB drive that held bootsector + /boot + some other stuff. You could leave your box plugged in, but there are better ways of avoiding static cling.
John Carmack has talked about the idea of generating texture maps dynamically.
Why bother? I would think that texture maps play a role similar to precompiled bsp trees.
LOL
Toshiba made a 17" prototype back in May that you can read about here.
1) Color accuracy could be adjusted with individual gain/bias controls just as in CRTs today. Adding "headroom" to the max subpixel brightness (build the display twice as bright as the control electronics will allow it to operate) would facilitate this kind of tuning and extend the lifetime of the display. Problem already solved by precedent and common sense.
2) Absolutely not. An LCD's matrix must be made on a transparent substrate. IIRC, that is why they are limited to amorphous polysilicon transitors and have a 30% yield. In contrast, any substrate that can support microelectronics can be home to an OLED. The design challanges (which I think revolve around heat) may remain, but making the support electronics shouldn't require anything that hasn't been possible years now.
3) Why would higher current be a cause of failure? That may make the active matrix harder to design but if the components fail because they were conducting electricity I don't see how lowering the current would help them last longer. If an athlon gives off 50+ watts of heat and runs on 3+ volts (argument's sake) how many amps are passing through through it? Understand I'm not saying LCD current levels wouldn't be easier for the EEs to deal with. I just don't think we'll be seeing dead pixels on an OLED monitor for the same reasons one could find a blown fuse on a power supply. 20 years from now people might have to say, "My old monitor is starting to look kinda red." whereas today they might say, "My CRT needs recalibration." or, "My LCD has a stuck pixel." Then again, there's always counterpoint #1.
Footnote: Their biggest advantage won't be just in power consumption if and when they arrive. It will be that they are emmissive, light, durable, capable of high pixel densities with rich color saturation, possibly a wider color gamut, and cheap.
... and if you go here you'll find that G3 450 achieving 62% of a P3 733's performance at cross-compiling gcc. That would make a P3 clocked at 454.5 MHz a G3 450's equal. The numbers look worse for native compilation.
Regardless of how easily DDR integrates into x86 every disscussion of the latest PowerMacs on this forum and xlr8 have claimed that Apple's new DDR boards benefit everything but the FSB and that the limitation is inherent in the G4 itself.
These things should be easier to make than CRTs or LCDs. Ink-jet printing is one of the considered manufacuring techniques.
The AC post comparing "PhysicsScholar" to "PhysicsGenius" has hit the nail on the head. This is a pale shadow of a fine Slashdot spoof artist. Nothing to see here folks. Move along...
Is this possible?
I also found an interesting discussion of DVD-V and it's use for audio. It can master at 24-bit/96KHz using desktop pc hardware which is the apparent limit of the human ear and nearly the limit at which thermal noise caps the precision of the electronics (see the above link). Could this be an "open" alternative to these audiophile formats?
...or a decent looking GUI.
mozilla uses cvs.
Why did Apple put BSD on top of Mach? Why not just fork BSD directly to their own liking and skip the Mach layer?
I'll bet Linus will be pleased to find this out as well.
There's also compilercache . The homepage details some of the benefits of of using this to supplement make when, for example, changing compiler flags.
To quote /usr/src/linux/Documentation/Configure.help, "Linux can use up to 64 Gigabytes of physical memory on x86 systems."
So you're saying.. like, GW 'n Sadam should play Return to Castle Wolfenstein?
Earlier tonight I gave my ti4200 the finger test for heat and wasn't pleased. It occured to me that since the AGP slot is always topmost on the motherboard there is no reason why the GPU shouldn't be mounted on the backside of the PCB where there is more room for active cooling. As it stands now GPU heatsinks are designed to make the best of a cramped situation.
The real failures were essentially management failures that marginalized warnings, and inter-agency rivalries.
Frontline just aired a documentary called "The Man Who Knew." FBI agent John O'Neil had Bin Laden in his sights for a decade. When he admitted to himself that bureaucratic politics would forever hobble his investigation he quit and took a job as Chief of Security at the World Trade Center. His famous last words: "We're due for something big."