What smd75 is talking about (but doesn't convey well) is the depth of the image. Early digital was strictly 8-bits per pixel, and a very small gamut. RAW gives us 12 to 14 bits of depth per pixel. In cantrast, my 4x5 Provia transparencies, wet-scanned on a drum scanner, give me 16-bits of depth per color channel per pixel: 48 bits per pixel. This lets me get to the detail in zone 2 that's locked up in black on a digital image.
Regarding megapixels, my 4"x5" analog sensor (film) is able to give me the equivalent of about 130 megapixels, and the sensor costs me about $2.25 a sheet.
For comparison, Leaf is selling a 56megapixel camera back for medium format for only $33,000.
Look at Better Light scanners, but keep your eye on your wallet. These are $20,000 for the top-of-the-line. The downside to scanning in-camera is that the scans take quite a while (for a scan equal to a 4x5 transparency). Don't try to take a picture of anything that might move in that time period - like leaves moving in a mild breeze, ocean waves, people,... and expect it to look 'right'. They are best used on static oblects like making archival images of paintings and the like.
4x5 cameras don't have 'through the lens metering'. Just about the only way to be sure you got the right exposure is to expose a polaroid 4x5 sheet in a special polaroid 4x5 sheet holder. This lets you check focus, exposure, see if there's any vignetting, etc. If it looks right on the polaroid, then stick in a sheet of 'regular' film and get your negative/transparency image.
Distributed Interactive Simulation (DIS) protocol supports a number of dead reckoning algorithms, from simple first order position to rate of turn and acceleration. An IEEE standard since '97, I believe. See here.
but to Linux developers/development. Many Linux/*BSD/OpenOffice users are moochers, contributing nothing but load to furthering the goals and needs of free/open source development. If they're going to give something back, let it be usable, modifiable open source code and interface specs, not canned apps. Then if someone wants to make a client side interface to TOMTOM, they can see the API clearly and scratch their own itch.
Dyson sphere around a now-extinct star. The clouds surrounding it were exhaust gasses that result from their ion-powered generators that scavenged the energy from the star when the star was young.
Your ISP isn't the guy that'd be blocking the email from delivery. It's the recipient's mail server that'd drop it on the floor. So either your argument is spurious or you really don't know what you're talking about.
I worked for General Atomics in '79-'80, on the Doublet-III tokamak (an experimental fusion reactor). The 'B' coils on the machine put out 240 KGauss during a 5-second 'shot' That was enough to cause all the images on the CRTs in the control room to collapse, and enough to deflect magnetic north on compasses within a mile or so of the device.
No one in the lab felt anything physically, although you could sometimes feel keys on steel keyrings rearrange themselves in your pocket if you were working just the other side of the radiation-blocking concrete wall.
Remember, also, that magnetism decreases by the cube of the distance from the source.
When I worked on General Atomic's Doublet-III experimental fusion rector, in the early 80s the energy for the machine was supplied by a three-story motor-generator constructed below-ground at the site. The motor ran off 440V mains and when powered spun itself, the generator and a 400-ton flywheel at 480rpm. It took twenty minutes to get the thing up to speed.
During a 5-second 'shot', when the stored energy was released, the motor, generator and flywheel would go from 480 to ~100 rpm, and dump 960 mega joules of energy into the coils of the experiment. You could feel the vibration in your feet anywhere you stood at the site, all the CRT's images would collapse due to the intense magnetic field generated. Then it was another twenty minutes before they could do it again.
Non-Tactical Video, a system for training the sailors on the (at the time) 2/3rds of the Navy ships that didn't have NTDS (Navy Tactical Data System) computers (typically, the supply ships and older combatants).
The ships that had NTDS used Radar Video Simulators (typically from Norden, the company famous for their bombsight), which cost about $80K. I replaced it with a stock PC in an Anvil roadcase that we took aboard those ships, with a controller that did the hardware-level radar video simulation.
For it I designed an ISA card that had a 128 byte FIFO and an 8-bit parallel-to-serial TTL chip that drove a hi-speed current amplifier, and a 10-bit up-counter that drove a D/S (digital-to-synchro) conveter, with some support logic, and a program that drove it based on data it was fed over a serial port. This gave 1024 possible blips at any angle, and 1024 possible angles at which I could draw these blips.
The serial port's data came from a Hayes 1200 baud modem whose phone line input came from an RF receiver tuned to receive the information transmitted from the Navy combat systems training school in San Diego. I would hook the D/S converter to a spare channel on the ship's radar switchboard (they always had a spare channel).
Our whole setup cost about $2K each, so they had us make 30 the first run, and a subsequent run of 40 more, I think. I'd moved off the project when it got to the production phase. What i thought the cool thing was that we'd made a 20480x2048 display, but instead of cartesian coordinates it was in polar coordinates, and had a very slow refresh rate. Our slogan at the time was 'We Got Your NTV!' - Sting and Mark Knopfler's MTV tune was at the time getting serious airplay.
Oh - at the transmit end we used a military transceiver and wired our schoolhouse-side Hayes modem to the _microphone_ input of the transmitter by way of a Radioshack audio transformer (modems put out about -13dbM and microphone inputs expect a smaller signal).
The datapath was one-way, so we wrapped the data up with a 16-bit CRC, and dropped bad packets on the received end. The data was very redundant so it worked.
Two:
A 3-D volumetric display using a CNC-machined 13-inch diameter two-bladed helix, a 5-watt Krypton-Argon laser and three pairs of acousto-optical crystals. The crystals were stimulated by electrical energy applied at ~40MHz, frequency modulated by D/A converters driven by memory buffers on the controller card.
The electrical energy coupled into the crystals were translated into mechanical energy by piezoelectric effect, which set up a diffraction in the crystals, which allowed me to steer the beam in the X and Y axis. The laser's output passed through a prism.
The red, blue and green portions were sent on separate paths to these pairs of crystals, so I could pait light onto any spot in the volume occupied by the blades. The controller got a signal when blade 1 passed through the zero point (a fixed hall-effect device, and a small magnet on the helix).
My job was supposed to be simply to write diagnostic software to help the controller board designer debug the controller, but the code evolved into the principal tool used to demonstrate the display's capabilities.
The main limitation with the display was that it took about 1.3uS for one oscillation frequency to 'die down' in the crystal before the new frequency was dominant, limiting us to about 4000 'voxels' per color per revolution (the helix spun at 600 rpm, so we had a refresh rate of 20Hz, which is barely above tolerable flicker. Faster gave less flicker, but the image started to smear. Slower and the flicker became very noticable.
Wrong on three counts: make(1), source code management systems and tools like diff and patch, and python source.
Make requires a tab to begin a command line. diff will show you and your coworkers all the silly changes you just committed to the repository because you set your editor to convert tabs to spaces. And Python is easily confused when tabs and spaces are mixed.
Make the URL modereated. But don't let moderation's low point trip invisibility of the link. It must always show, to give the poster the creds.
URL moderation categories:
'Low Bandwidth' ie: easily slashdotted out-of-existence 'Google Page Rank Whore' 'Related to Story' 'Obnoxious' ie: g0teSx, GNA in addition to the usual 'insightful', 'informative', etc.
My feeling is that if this guy is delivering good articles, I'm OK with us slashdotters indirectly helping him financially (hitting his page, upping view counts for his linked ads) - he works for those views more than most of us who just sit here and participate in the comment-stream.
I'm not happy he's poppin' stuff up on us, and think he ought to go for a better class of customers than those that want to 'get in our face' in such a rude way, but he does seem to come up with better-than-average stories.
Tell me when it's 01111111111111111111111111111111BINARY cuz that's when all hell's gonna break loose. All the code that uses a signed it to hold the time will enter a time warp the likes of which will not be seen again unitl - wait! we'll be stuck in an infinite loop! AAAAAAUUGH!!
Slashdot Mirror
It's 150M degrees Electron Temperature.
Coincidence? I hope so...
What smd75 is talking about (but doesn't convey well) is the depth of the image. Early digital was strictly 8-bits per pixel, and a very small gamut. RAW gives us 12 to 14 bits of depth per pixel. In cantrast, my 4x5 Provia transparencies, wet-scanned on a drum scanner, give me 16-bits of depth per color channel per pixel: 48 bits per pixel. This lets me get to the detail in zone 2 that's locked up in black on a digital image. Regarding megapixels, my 4"x5" analog sensor (film) is able to give me the equivalent of about 130 megapixels, and the sensor costs me about $2.25 a sheet. For comparison, Leaf is selling a 56megapixel camera back for medium format for only $33,000.
Look at Better Light scanners, but keep your eye on your wallet. These are $20,000 for the top-of-the-line. The downside to scanning in-camera is that the scans take quite a while (for a scan equal to a 4x5 transparency). Don't try to take a picture of anything that might move in that time period - like leaves moving in a mild breeze, ocean waves, people, ... and expect it to look 'right'. They are best used on static oblects like making archival images of paintings and the like.
4x5 cameras don't have 'through the lens metering'. Just about the only way to be sure you got the right exposure is to expose a polaroid 4x5 sheet in a special polaroid 4x5 sheet holder. This lets you check focus, exposure, see if there's any vignetting, etc. If it looks right on the polaroid, then stick in a sheet of 'regular' film and get your negative/transparency image.
.. I thought it would be an article about sniping eBay bids at the last second from my iPod.
Distributed Interactive Simulation (DIS) protocol supports a number of dead reckoning algorithms, from simple first order position to rate of turn and acceleration. An IEEE standard since '97, I believe. See here.
Use D instead of dL and save a keystroke.
I was told by a staff scientist at Caltech that a common and useful definition is:
'Any computer that performs within an order of magnitude of the current fastest computer is a supercomputer'.
but to Linux developers/development. Many Linux/*BSD/OpenOffice users are moochers, contributing nothing but load to furthering the goals and needs of free/open source development. If they're going to give something back, let it be usable, modifiable open source code and interface specs, not canned apps. Then if someone wants to make a client side interface to TOMTOM, they can see the API clearly and scratch their own itch.
Dyson sphere around a now-extinct star. The clouds surrounding it were exhaust gasses that result from their ion-powered generators that scavenged the energy from the star when the star was young.
Your ISP isn't the guy that'd be blocking the email from delivery. It's the recipient's mail server that'd drop it on the floor. So either your argument is spurious or you really don't know what you're talking about.
I worked for General Atomics in '79-'80, on the Doublet-III tokamak (an experimental fusion reactor). The 'B' coils on the machine put out 240 KGauss during a 5-second 'shot' That was enough to cause all the images on the CRTs in the control room to collapse, and enough to deflect magnetic north on compasses within a mile or so of the device.
No one in the lab felt anything physically, although you could sometimes feel keys on steel keyrings rearrange themselves in your pocket if you were working just the other side of the radiation-blocking concrete wall.
Remember, also, that magnetism decreases by the cube of the distance from the source.
During a 5-second 'shot', when the stored energy was released, the motor, generator and flywheel would go from 480 to ~100 rpm, and dump 960 mega joules of energy into the coils of the experiment. You could feel the vibration in your feet anywhere you stood at the site, all the CRT's images would collapse due to the intense magnetic field generated. Then it was another twenty minutes before they could do it again.
One:
Non-Tactical Video, a system for training the sailors on the (at the time) 2/3rds of the Navy ships that didn't have NTDS (Navy Tactical Data System) computers (typically, the supply ships and older combatants).
The ships that had NTDS used Radar Video Simulators (typically from Norden, the company famous for their bombsight), which cost about $80K. I replaced it with a stock PC in an Anvil roadcase that we took aboard those ships, with a controller that did the hardware-level radar video simulation.
For it I designed an ISA card that had a 128 byte FIFO and an 8-bit parallel-to-serial TTL chip that drove a hi-speed current amplifier, and a 10-bit up-counter that drove a D/S (digital-to-synchro) conveter, with some support logic, and a program that drove it based on data it was fed over a serial port. This gave 1024 possible blips at any angle, and 1024 possible angles at which I could draw these blips.
The serial port's data came from a Hayes 1200 baud modem whose phone line input came from an RF receiver tuned to receive the information transmitted from the Navy combat systems training school in San Diego. I would hook the D/S converter to a spare channel on the ship's radar switchboard (they always had a spare channel).
Our whole setup cost about $2K each, so they had us make 30 the first run, and a subsequent run of 40 more, I think. I'd moved off the project when it got to the production phase. What i thought the cool thing was that we'd made a 20480x2048 display, but instead of cartesian coordinates it was in polar coordinates, and had a very slow refresh rate. Our slogan at the time was 'We Got Your NTV!' - Sting and Mark Knopfler's MTV tune was at the time getting serious airplay.
Oh - at the transmit end we used a military transceiver and wired our schoolhouse-side Hayes modem to the _microphone_ input of the transmitter by way of a Radioshack audio transformer (modems put out about -13dbM and microphone inputs expect a smaller signal).
The datapath was one-way, so we wrapped the data up with a 16-bit CRC, and dropped bad packets on the received end. The data was very redundant so it worked.
Two:
A 3-D volumetric display using a CNC-machined 13-inch diameter two-bladed helix, a 5-watt Krypton-Argon laser and three pairs of acousto-optical crystals. The crystals were stimulated by electrical energy applied at ~40MHz, frequency modulated by D/A converters driven by memory buffers on the controller card.
The electrical energy coupled into the crystals were translated into mechanical energy by piezoelectric effect, which set up a diffraction in the crystals, which allowed me to steer the beam in the X and Y axis. The laser's output passed through a prism.
The red, blue and green portions were sent on separate paths to these pairs of crystals, so I could pait light onto any spot in the volume occupied by the blades. The controller got a signal when blade 1 passed through the zero point (a fixed hall-effect device, and a small magnet on the helix).
My job was supposed to be simply to write diagnostic software to help the controller board designer debug the controller, but the code evolved into the principal tool used to demonstrate the display's capabilities.
The main limitation with the display was that it took about 1.3uS for one oscillation frequency to 'die down' in the crystal before the new frequency was dominant, limiting us to about 4000 'voxels' per color per revolution (the helix spun at 600 rpm, so we had a refresh rate of 20Hz, which is barely above tolerable flicker. Faster gave less flicker, but the image started to smear. Slower and the flicker became very noticable.
Red/Black isolation and COMSEC requirements are described in Orange book.
Wrong on three counts: make(1), source code management systems and tools like diff and patch, and python source. Make requires a tab to begin a command line. diff will show you and your coworkers all the silly changes you just committed to the repository because you set your editor to convert tabs to spaces. And Python is easily confused when tabs and spaces are mixed.
They're used in the larger optical telescopes. Very expensive, and often only greyscale, they offer huge dynamic range.
Make the URL modereated. But don't let moderation's low point trip invisibility of the link. It must always show, to give the poster the creds.
URL moderation categories:
'Low Bandwidth' ie: easily slashdotted out-of-existence
'Google Page Rank Whore'
'Related to Story'
'Obnoxious' ie: g0teSx, GNA
in addition to the usual 'insightful', 'informative', etc.
My feeling is that if this guy is delivering good articles, I'm OK with us slashdotters indirectly helping him financially (hitting his page, upping view counts for his linked ads) - he works for those views more than most of us who just sit here and participate in the comment-stream.
I'm not happy he's poppin' stuff up on us, and think he ought to go for a better class of customers than those that want to 'get in our face' in such a rude way, but he does seem to come up with better-than-average stories.
More significantly, it's the 50th anniversary of Elvis's recording of "Heartbreak Hotel".
No need to wait for a repost - this one dupes itself!
Seems kinda brutal to hit them with another DDOS.
Give up. You can't 'escape'.
Tell me when it's 01111111111111111111111111111111 BINARY cuz that's when all hell's gonna break loose. All the code that uses a signed it to hold the time will enter a time warp the likes of which will not be seen again unitl - wait! we'll be stuck in an infinite loop! AAAAAAUUGH!!