World's first digital camera

Trust me, recording Digital images onto Mag Tape was a lot of work.

Our first generation Digital Infrared Sensor used 7-track Mag Tape ~1981, our second generation Digital IR sensor used a 28-Track Sangamo Weston High-Density Digital Recorder, ~1984. It was expensive and loud. When we ran data through a VAX 11/750 two DRE11C double-buffered DMA boards were required, each on a dedicated Unibus. The console lights dimmed on the VAX when my FORTRAN code ran to acquire the data. I tuned the CPU and software with an oscilloscope. That was fun. Four racks of equipment were required for the sensor and the data acquisition system.

Working in a research lab can be interesting.
 
Last edited:
same with GPS

same with GPS

driving around N. Calif. windy roads with a proto GPS changing maps on cassette tapes ...

Brian Sweeney said:
Trust me, recording Digital images onto Mag Tape was a lot of work.

Our first generation Digital Infrared Sensor used 7-track Mag Tape ~1981, our second generation Digital IR sensor used a 28-Track Sangamo Weston High-Density Digital Recorder, ~1984. It was expensive and loud. When we ran data through a VAX 11/750 two DRE11C double-buffered DMA boards were required, each on a dedicated Unibus. The console lights dimmed on the VAX when my FORTRAN code ran to acquire the data. I tuned the CPU and software with an oscilloscope. That was fun. Four racks of equipment were required for the sensor and the data acquisition system.

Working in a research lab can be interesting.
Click to expand...
 
Not GPS- I built custom hardware to interface a Z80 based CPM computer to the Navigational computer of a P3 Orion. It recorded position as the sensor took data.
 
Brian Sweeney said:
Not GPS- I built custom hardware to interface a Z80 based CPM computer to the Navigational computer of a P3 Orion. It recorded position as the sensor took data.
Click to expand...

Z80 chips; would you believe in the US Army, we still had a communications system running from a Z80 as late as 1989? I forget now if it was JNAP or wrapped in JNAP to get on the signal corps system. It was started in 1979. Hot stuff for us then.
 
ZILOG is still around, and the Z80 architecture is still available as an embedded processor. I still program in 8-bit assembly language.

Beats Management!
 
Brian Sweeney said:
ZILOG is still around, and the Z80 architecture is still available as an embedded processor. I still program in 8-bit assembly language.

Beats Management!
Click to expand...

Just a friendly reminder. Without management you wouldn't have a job where you can play with all that stuff you obviously love. ;)
 
wow

wow

Those P3s were flying over our house every 15 minutes in the '70s.

I programmed Z80s for a bit in the early '80s, but more 6502s, then 8088/6's, etc.

Brian Sweeney said:
Not GPS- I built custom hardware to interface a Z80 based CPM computer to the Navigational computer of a P3 Orion. It recorded position as the sensor took data.
Click to expand...
 
Matus said:
Sure we know him. He is the guy who wastes 30% of the sensor data to recover some color information :p
Click to expand...

I wasn't referring to this post. The post would attract the knowledgeable. I met the RFF in general..You could go to a Foveon. You still have 1/3 and RGGB smoothing. I don't think this will change for a while. And don't blame Bayer for the work of marketing people. Have you ever had a studio size flash generator apart? Do the math on the Caps. for Joules (watt/sec) and compare to the advertised output.. One manufactures pixel isn't the same as one from another manufacturer.
 
Last edited:
tom.w.bn said:
Just a friendly reminder. Without management you wouldn't have a job where you can play with all that stuff you obviously love. ;)
Click to expand...

I was a Manager for 10 years, now "Senior Staff" and free to work on what I like.

WAY beats management.

We had a non-disclosure agreement with Kodak in the 80s for developing detectors. I had Kodak make the IR version of the DCS200.

There were Mosaic filters for color information before Bayer created and patented his famous pattern. I still prefer the term "Bayer Pattern Mosaic Filter". I have also seen sensors with beamsplitters and monochrome detectors, fast-spinning color wheels, and detectors formulated for specific color bands. The latter was used in the IR sensors that we built.
 
Last edited:
Brian Sweeney said:
I was a Manager for 10 years, now "Senior Staff" and free to work on what I like.

WAY beats management.

We had a non-disclosure agreement with Kodak in the 80s for developing detectors. I had Kodak make the IR version of the DCS200.

There were Mosaic filters for color information before Bayer created and patented his famous pattern. I still prefer the term "Bayer Pattern Mosaic Filter". I have also seen sensors with beamsplitters and monochrome detectors, fast-spinning color wheels, and detectors formulated for specific color bands. The latter was used in the IR sensors that we built.
Click to expand...

I'm on the "user" end of this technology. Where do you think it's going. I can't see the noise level going lower without some kind of cooling. I guess Peltier cooling would work.. but then you have moisture to deal with. I've seen some big sensors used on telescopes with cooing. But then they don't rely on batteries and aren't made to be portable.
 
I worked with sensors in the 80s, moved onto developing optical networks in 1990. Another fast moving target.

Some of the groups in the Division are building sensors, and do use coolers. Electronic noise tends to be an issue for noise more than the detectors themselves.

I suspect image processing and low-power chips will be the bigger push in consumer cameras for improving images. Detectors are really starting to hit limits of physics. But everytime someone makes the statement that they can't get better, someone else invents something new to prove them wrong.
 
Brian Sweeney said:
I worked with sensors in the 80s, moved onto developing optical networks in 1990. Another fast moving target.

Some of the groups in the Division are building sensors, and do use coolers. Electronic noise tends to be an issue for noise more than the detectors themselves.

I suspect image processing and low-power chips will be the bigger push in consumer cameras for improving images. Detectors are really starting to hit limits of physics. But everytime someone makes the statement that they can't get better, someone else invents something new to prove them wrong.
Click to expand...

I understand.. I know that most noise issues are a product of resistor noise (electron travel-white and pink- Johnson-Nyquist). Cooling is a solution to these issues too? If cameras are to be portable there are going to be some serious power issues. I saw the research and product of Lam's new Laser cut Si. That will change a lot of stuff as it doesn't need near the scrubbing that the current Si requires and can deal with a higher population density..

I think I read that Nikon's current high end sensor cost about $400/ea with product failure. Maybe they will go the route of the micro processor builders and build one and scale the speed/price in QA. I guess you have to hide the bad pixels with firmware..
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom