It looks like you're using an Ad Blocker.
Please white-list or disable AboveTopSecret.com in your ad-blocking tool.
Thank you.
Some features of ATS will be disabled while you continue to use an ad-blocker.
originally posted by: benwyatt
sounds good.
have you ever ran a live operating system? from a cd or dvd? PAINFULLY SLOW. as these will be. cool none the less.
originally posted by: StargateSG7
originally posted by: benwyatt
sounds good.
have you ever ran a live operating system? from a cd or dvd? PAINFULLY SLOW. as these will be. cool none the less.
These discs don't spin --- They are JUST BURNED THAT WAY !!!!
When completed and hooked up via edge connectors to
an external display, they will be running at 10 GHZ !!!
That's three times faster than many Intel i7 CPU's !!!
Stack the discs a hundred high and you have a massively parallel
supercomputer that waaaaaay more powerful than today's
100 million dollar supers!
The spinning BluRay technology is merely used to BURN-IN
the final connections and soft-logic of a virtual CPU design.
Once completed you can run at FULL SPEED -- 10 GHZ!!!!!
When completed and hooked up via edge connectors to
an external display, they will be running at 10 GHZ !!!
originally posted by: ChaoticOrder
a reply to: StargateSG7
When completed and hooked up via edge connectors to
an external display, they will be running at 10 GHZ !!!
So you would need a special type of CPU socket? Has this been done already?
originally posted by: DaRAGE
This thread is a total farce and i will explain why.
Current intel cpu's are 14 nm 3d Finfet designs.
You are talking about blu-ray which is 360 - 480nm
AMD K6 8,800,000 transistors 1997 AMD 0.35 µm This is 350nm. Built in 1997. Also that same size is....
Pentium II Klamath 7,500,000 transistors 1997 Intel 0.35 µm 195 mm²
Intel Haswell-E 8 Core 22nm 2.6 Billion transistors 356 mm2
Transistors turning on and off cause heat as electricity runs through them and some energy is lost.
You wont be getting a blu-ray disk worth of Core i7's. You will be getting pentium 2's and let me tell you... They run damn hot.
I don't know where you are getting your numbers from.
This is all jargonal bullshiot.
I have no doubt that maybe the technology you are talking about could definitely work, and it would be cheap, but you will not be creating a powerhouse of computing.
You should probably read this article to look at performance differences between intel cpu's from 10 years ago compared to intels cpu's two generations ago. 1997 cpu's would be much worse haha.
originally posted by: ChaoticOrder
a reply to: StargateSG7
When completed and hooked up via edge connectors to
an external display, they will be running at 10 GHZ !!!
So you would need a special type of CPU socket? Has this been done already?
originally posted by: theantediluvian
I've had a couple drinks but...
35 core i7s? The latest manufacturing process for Intel chips is 14nm. The interconnect pitch is 54nm which is about a third the minimum pit size on a blu-ray (150nm). Blu-rays and other optical discs are burned/read as a spiral track of pits and the track pitch of a blu-ray disc is 320nm.
Explain the manufacturing process of the discs and how you arrived at $2. What is the substrate? How will it diffract the beam? What are the wires made of? We're talking a unique set of properties there: a laser would have to be able to break the connection without melting everything in its proximity and it would simultaneously need to of course be suitable for building the circuits.
As a concept, I think it's a neat idea and I've been wondering about how we might evolve 3D printers in the near future to print microelectronics. In some regards, I think this sounds like an interesting concept but not with blu-ray. I can see the appeal in using established technology but I suspect it's not doable and if it were, not at anything approaching the the efficiency you're imagining.
originally posted by: StargateSG7
originally posted by: theantediluvian
I've had a couple drinks but...
35 core i7s? The latest manufacturing process for Intel chips is 14nm. The interconnect pitch is 54nm which is about a third the minimum pit size on a blu-ray (150nm). Blu-rays and other optical discs are burned/read as a spiral track of pits and the track pitch of a blu-ray disc is 320nm.
Explain the manufacturing process of the discs and how you arrived at $2. What is the substrate? How will it diffract the beam? What are the wires made of? We're talking a unique set of properties there: a laser would have to be able to break the connection without melting everything in its proximity and it would simultaneously need to of course be suitable for building the circuits.
As a concept, I think it's a neat idea and I've been wondering about how we might evolve 3D printers in the near future to print microelectronics. In some regards, I think this sounds like an interesting concept but not with blu-ray. I can see the appeal in using established technology but I suspect it's not doable and if it were, not at anything approaching the the efficiency you're imagining.
---
Each polycarbonate layer is stamped using a inverted glass master
that contains a 2D-XY array of pits or wells from 150 nm to 480 nm
in size depending upon the wavelength of the lasers ( now is blue
but in the future could be ultraviolet lasers!).
Each stamped well or pit have sharp edged walls that are
U-NOTCHED on each face on the flat plane containing a
thin-film vapor deposit of aluminum film that at the
root of the U-NOTCH forms a pathway between
another well or pit.
At the boot of each well or pit is a doped P and N silicon crystal
containing a blue light sensitive charge carrier or separator that
when actived by a pulse from a blue laser pulse at a specific
power level will POP the charge carrier thus performing
one of three actions:
1) Create a Forward bias P-N junction.
2) Create a Backward bias P-N junction.
3) Create an electrical charge storage well
forming the basis of an ON/OFF bit-based
memory storage unit.
Separating each layer is another layer of microtubules containing
a single vertical microwire (created by charge-based deposition)
that passes immediately to the above or below P-N Junction
well or pit. This allows electrical circuit pathways to be
constructed on a 3D stacked basis by using ANOTHER pulsed
beam to break or keep intact the vertical electrical connection
between vertically-stacked layers of P-N junctions.
The first laser pulse keeps or breaks the charge carrier
between deposited mass-produced P-N silicon crystals
via an amplitude-specific blue light-sensitive reactant
which affects the doping in the charge carrier.
Another power level of the pulsed laser can
create a charge-holding well to form a bit-based
storage unit by literally melting the entire P-N junction.
Then on the SAME LAYER, the blue laser can focus
and explode the U-NOTCH on any sidewall to break
a 2D-XY connection to another neighboring pit or
well. Only the connection to the desired neighbouring
P-N junction or charge storage well is kept.
For the connections BETWEEN each 2D-XY layer of
P-N junctions, the laser can focus on the vertical
microtubule that are installed between each 2D-XY
layer of P-N junctions and keep or break the connection
between the junction immediate below (or above)
the current layer by using amplitude-specific pulses
to burst the vertically oriented microwire (i.e. break a fuse)
Each layer of 2D-XY P-N junctions and layer of vertical
microwires are combined to form a stack array of junctions
that can be connected and disconnected to form a 3D-XYZ
structure similar in function to a field programmable gate array
(FPGA) but on a write-once basis.
A VHDL script at the END-USER stage tells a bluRay disc writer
to pop or keep intact connections between P-N junctions which
can form sequences that are diodes, transistors, larger logic
circuits and then soft-CPU cores using instructions that can
be loaded and run from a text-file based VHDL script which
can be either user-designed or bought as a file from an
online app-store as a Soft-CPU/GPU/DSP design.
Since Polycarbonate is stable up to 150 Celcius,
you can pre-dope the polycarbonate itself in a
flat planar or cubic array to form areas suspectible
to popping at specific amplitudes from a blue laser
pulse. This can be used to accurately break or keep
electrical connection pathways that are enabled
via thin film aluminum vapor deposition at the
disc manufacturing factory without ruining
the rest of the disc.
For ACTIVE COOLING, the factory can insert a pre-etched
layers of fluid filled microchannels that can dissipate over
100 watts to external fins. Calculate the thermal conductivity
from having spread-out CPU cores interspersed with
in-between layers of fluid filled micro-etched cooling
channels and the large SURFACE AREA of the actual
electrical pathways, and it is VERY DOABLE to run
a single disc at between 7-to-10 GHZ!
Plastic is CHEAP! Mass Produced P-N junctions are also
cheap and depositable into a factory stamped pit or well
using blu-ray disc manufacturing and disc layer alignment
and gluing techniques. The software to BURN the designs
is now mature so COSTS based upon what I have from
mass-market CD-ROM/DVD and Blue Ray manufacturers
themselves have given me a final wholesale price of
$2.00 per final CPU/GPU/DSP disc in blocks of
100,000+ discs! Order a million discs at once and
I get a price quote that is downto $1.50 per disc!
Finally the square area of the program area of
a Blue ray is about 86 square centimeters or
8600 square millimetres ON EACH LAYER
(total of 86,000 square mm on ten layers!)
which is MORE THAN ENOUGH ROOM to
burn the square area of MULTIPLE cores
of a software-burned CPU/GPU/DSP design.
Timewise and depending upon the complexity of the
VHDL script file, disc burning times are in the range
of one to four hours for a ten layer disc containing
an array of 35 equivalent 2012 era 1.6 BILLION transistor count
silicon-based Intel i7 CPU's that would normally be running
at 3.4 GHZ but on-disc would run between 7 GHZ to 10 GHZ!