Maker Pro
Maker Pro

Using photonics for real-time processing.

B

Bob Myers

Jan 1, 1970
0
Michael A. Terrell said:
You do know that "Jack" is "Dark Matter", don't you?

I certainly have the impression that large parts of him are, yes.

Most of those would be above the neck.



Bob M.
 
B

Bob Myers

Jan 1, 1970
0
JackShephard said:
The fact remains that you treated him like shit, and still do, and you
have been treating me like shit.

Oh, not at all, Jack, old man. For one thing, I could
respect shit.


Bob M.
 
M

Michael A. Terrell

Jan 1, 1970
0
Bob said:
In that case, would you please ask one of your toenail
clippings to start posting here in your place? I'm sure they
would provide far more mature and worthwhile contributions
than you've been able to provide to date.

Bob M.


He can't even figure out that all my kill files have been updated, so
he keeps mouthing off to the bit bucket.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida
 
M

Michael A. Terrell

Jan 1, 1970
0
Bob said:
Oh, not at all, Jack, old man. For one thing, I could
respect shit.


You're right. At least 'shit' has a reason to exist.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida
 
J

JackShephard

Jan 1, 1970
0
In that case, would you please ask one of your toenail
clippings to start posting here in your place? I'm sure they
would provide far more mature and worthwhile contributions
than you've been able to provide to date.


So, you are saying, like Terrell, that EE Times is lame, and that JPL is
lame, and that this holographic memory array is "vaporware?

I'd say that every one of you, except for myself and the OP, have made
ZERO "worthwhile contributions to this thread.

So what does that make you stupid bastards? or...

So what does that make you?... stupid bastards?
 
J

JackShephard

Jan 1, 1970
0
The word is "you're," a contraction of "you are."

As in, "you're welcome."


Bob M.

I am fully aware of that, just like I am fully aware of what can be
found in EE Times, and that it is a very good publication.

I don't need a primer from an idiot that spent several posts deriding
the author of a thread, and one of the respondents thereto.

It was a "heated, fast response mistake". Better known as a typo.

So **** you, and your little grammar, spelling, and usage dog too.
 
J

JackShephard

Jan 1, 1970
0
Oh, not at all, Jack, old man. For one thing, I could
respect shit.

In your case, that would be referred to as "self respect".
 
M

MooseFET

Jan 1, 1970
0
On May 27, 8:35 pm, Sjouke Burry
[....]
Radium invented trolling.
(Oh and he also invented questions with a maximum
stupidity content).

Trolling was around long before Radium. His ideas may be a waste of
time but he doesn't seem to be causing trouble on purpose. He also
doesn't interject into others peoples conversions. I have yet to see
him attempt to insult anyone.
 
B

Bob Myers

Jan 1, 1970
0
JackShephard said:
So **** you, and your little grammar, spelling, and usage dog too.

Again, I am in awe of the maturity and vocabulary skills
consistently displayed in your posts. Normally, in order
to see writing of such caliber, one must peruse the walls
of the men's room at the local bus station.

Bob M.
 
B

Bob Myers

Jan 1, 1970
0
JackShephard said:
So, you are saying, like Terrell, that EE Times is lame, and that JPL is
lame, and that this holographic memory array is "vaporware?

What dazzling powers of inference! For your next
trick, will you show us the authenticity of the Shroud of
Turin, demonstrate the conspiracy behind the Kennedy
assassination and the 9/11 attacks, and prove creationism,
all in the space of a single sentence?

The entire world waits with bated breath...

Bob M.
 
J

JackShephard

Jan 1, 1970
0
On May 27, 8:35 pm, Sjouke Burry
[....]
Radium invented trolling.
(Oh and he also invented questions with a maximum
stupidity content).

Trolling was around long before Radium. His ideas may be a waste of
time but he doesn't seem to be causing trouble on purpose. He also
doesn't interject into others peoples conversions. I have yet to see
him attempt to insult anyone.
Notice that the only people I "insulted" were those that insulted him,
and that was damned near everyone that responded to him.

Oh, and this is USENET. There are no "other people's conversations".

Nice try though, and thanks for not being yet another idiot on the
"lets beat up Radium" bandwagon, even though you do appear to be running
along side the "lets call Jack a troll" bandwagon.
 
M

MooseFET

Jan 1, 1970
0
Do you think the heat generated and power requirements will decrease
when photonic chips are available?

When/if chips go photonic, I fully expect the makers to cram as much
in as they can and not have the part melt. Faster chips gain the
maker more dollars per acre than slower ones. As a result, they will
always push the speed up to the limit of their technology.

The big thing that using light gains will be speed. Even there, the
speeds people are getting over copper interconnects is increasing with
time.

[....]
I am thinking of a purely optical computer that is powered by a main
400 nm laser. The main laser if of course powered by electricity.

Why ever would you want 400nM?
I'd suggest 780nM. Take a look at:
socrates.berkeley.edu/~budker/papers/pdfs/hole_burning_pdftex.pdf

Notice how using light right near the D2 line allows you to do things
that cause Lasers to interact.



This optical PC contains 400 nm lasers but no LEDs. AFAIK, lasers tend
to be more efficient that LEDs.

So do you think a chip based on lasers - instead of electricity - can
be as real-time, hardware, and digital as possible while using the
least amount of buffering required [hopefully none] and experiencing
the least amount of latency possible [again, hopefully none] and at
the same time being high-speed not getting hot enough to need any
cooling equipment?

I think it would be easier to do this in photonics that electronics.
Since electronics seem to easily overheat.
 
R

Radium

Jan 1, 1970
0
Do you think the heat generated and power requirements will decrease
when photonic chips are available?

When/if chips go photonic, I fully expect the makers to cram as much
in as they can and not have the part melt. Faster chips gain the
maker more dollars per acre than slower ones. As a result, they will
always push the speed up to the limit of their technology.

The big thing that using light gains will be speed. Even there, the
speeds people are getting over copper interconnects is increasing with
time.

[....]

Do you think chips will be made of photonic circuits -- instead of
electric -- in the next 20 years?
Why ever would you want 400nM?
I'd suggest 780nM. Take a look at:
socrates.berkeley.edu/~budker/papers/pdfs/hole_burning_pdftex.pdf

Shorter wavelengths usually allow more bandwidth. AFAIK, 400 nm is the
shortest wavelength in which you are safe from ionizing radiation. At
wavelengths shorter than 400 nm, your risk of cancer increases.

Conventional CDs use red lasers. Blu-ray uses 405 nm.
Notice how using light right near the D2 line allows you to do things
that cause Lasers to interact.


This optical PC contains 400 nm lasers but no LEDs. AFAIK, lasers tend
to be more efficient that LEDs.
So do you think a chip based on lasers - instead of electricity - can
be as real-time, hardware, and digital as possible while using the
least amount of buffering required [hopefully none] and experiencing
the least amount of latency possible [again, hopefully none] and at
the same time being high-speed not getting hot enough to need any
cooling equipment?
I think it would be easier to do this in photonics that electronics.
Since electronics seem to easily overheat.
 
M

MooseFET

Jan 1, 1970
0
Shorter wavelengths usually allow more bandwidth. AFAIK, 400 nm is the
shortest wavelength in which you are safe from ionizing radiation. At
wavelengths shorter than 400 nm, your risk of cancer increases.

Remember that data rate is log2 the number of levels times the
frequency. IIRC, You get 7 levels of the D2 absorption. This is
almost a factor of 3 more than you get with 2 states at 400nM.

The speed advantage therefor goes to 780nM. You could go with Cs and
use 752nM instead. Unfortunately, I think this would limit the high
temperature end of the operating range.

Conventional CDs use red lasers. Blu-ray uses 405 nm.
Notice how using light right near the D2 line allows you to do things
that cause Lasers to interact.
This optical PC contains 400 nm lasers but no LEDs. AFAIK, lasers tend
to be more efficient that LEDs.
So do you think a chip based on lasers - instead of electricity - can
be as real-time, hardware, and digital as possible while using the
least amount of buffering required [hopefully none] and experiencing
the least amount of latency possible [again, hopefully none] and at
the same time being high-speed not getting hot enough to need any
cooling equipment?
I think it would be easier to do this in photonics that electronics.
Since electronics seem to easily overheat.
 
B

Bob Myers

Jan 1, 1970
0
Do you think chips will be made of photonic circuits -- instead of
electric -- in the next 20 years?

If you mean a purely optical system, as opposed to having
just pieces here and there rely on some optical phenomenon
(e.g., the "holographic" memory which has been mentioned
here) - no, there is no likelihood of that at all. The reasons
have already been given here, at great length.
Shorter wavelengths usually allow more bandwidth. AFAIK, 400 nm is the
shortest wavelength in which you are safe from ionizing radiation. At
wavelengths shorter than 400 nm, your risk of cancer increases.

Conventional CDs use red lasers. Blu-ray uses 405 nm.

Shorter wavelengths in general permit signals of greater
bandwidth, but the available bandwidth of even infrared
light (longer than 780 nm or so) is so huge already that this
isn't the issue. The reason that Blu-Ray or HD-DVD use
blue lasers is not because of the bandwidth capability
implied by the shorter wavelength, but because it permits
smaller physical structures in the disc to be used for
storage (and hence increases the capacity of the disc per
unit area over their "red-laser" cousins). In other words,
"red" light could easily carry information at the rate necessary
for a Blu-Ray disc; it just can't "see" the disc properly to
read it.

Bob M.
 
R

Radium

Jan 1, 1970
0
If you mean a purely optical system, as opposed to having
just pieces here and there rely on some optical phenomenon
(e.g., the "holographic" memory which has been mentioned
here) - no, there is no likelihood of that at all. The reasons
have already been given here, at great length.

Well, this PC I describe is as photonic [with lasers and no LEDs] as
practical. Maybe not purely-optical, but as laser-based as it can be
and still be efficient.
Shorter wavelengths in general permit signals of greater
bandwidth, but the available bandwidth of even infrared
light (longer than 780 nm or so) is so huge already that this
isn't the issue. The reason that Blu-Ray or HD-DVD use
blue lasers is not because of the bandwidth capability
implied by the shorter wavelength, but because it permits
smaller physical structures in the disc to be used for
storage (and hence increases the capacity of the disc per
unit area over their "red-laser" cousins). In other words,
"red" light could easily carry information at the rate necessary
for a Blu-Ray disc; it just can't "see" the disc properly to
read it.

Yes. By the same token, let's say there are two laser-circuited chips
of the same capacity [let's say 4 GB]. Chip A uses 780nm lasers, while
chip B uses 400nm lasers. Chip B can be made smaller than chip A,
because B uses shorter wavelength, and hence can be use smaller
structures for the same amount of data.

Chip B would take up a lot less physical space [for the same amount of
info] than chip A.
 
J

JackShephard

Jan 1, 1970
0
If you mean a purely optical system, as opposed to having
just pieces here and there rely on some optical phenomenon
(e.g., the "holographic" memory which has been mentioned
here) - no, there is no likelihood of that at all. The reasons
have already been given here, at great length.


There are already segments of optical network links that are all
optical.
 
K

krw

Jan 1, 1970
0
There are already segments of optical network links that are all
optical.

No shit? Who wudda ever thunk that optical networks would be
optical. Dimmy, you're the dumbest.
 
R

Rich Grise

Jan 1, 1970
0
On May 27, 8:35 pm, Sjouke Burry
[....]
Radium invented trolling.
(Oh and he also invented questions with a maximum
stupidity content).

Trolling was around long before Radium. His ideas may be a waste of
time but he doesn't seem to be causing trouble on purpose. He also
doesn't interject into others peoples conversions. I have yet to see
him attempt to insult anyone.
Notice that the only people I "insulted" were those that insulted him,
and that was damned near everyone that responded to him.

Oh, and this is USENET. There are no "other people's conversations".

Nice try though, and thanks for not being yet another idiot on the
"lets beat up Radium" bandwagon, even though you do appear to be running
along side the "lets call Jack a troll" bandwagon.

Now I'm a little worried - does telling him, "build one and show us how
it's done - so far, everyone thinks it's impossible" constitute picking
on him?

Thanks,
Rich
 
A

Alex Colvin

Jan 1, 1970
0
There are already segments of optical network links that are all
No shit? Who wudda ever thunk that optical networks would be
optical.

and wireless networks? are they all wireless?
 
Top