infrared for processor pins?

chrisalviola

chrisalviola

New Member
Is this possible? a processor without pins but infrared to connect it on motherboard, they say it reduces heat. it from intel they say.
 
That would increase both the power used by the chip, and the temperature of the chip IF the heat generated by both 1 emitter and 1 reciever is greater than that given off by 1 small pin.
 
As far as I know infrared cant transfer data that quickly. And how would the processor get power?
 
[-0MEGA-];425791 said:
As far as I know infrared cant transfer data that quickly. And how would the processor get power?
Click to expand...
The latest revisons of IrDA can peak at 100Mbit/s, not entirely sure what the data throughput on a CPU would be, or how to measure, it.... Bu i highyl doubt you could implement such a thing, especially in socket such as 939, thats 939 emitters, 939 recievers. Lot more space, lot more heat, just a ridiculous idea really

dragon
 
yes it sounds quite unessisary...and useless...i think it would rase the heat of the chip rather then reduce it, plus the data tranfer will be much slower
 
I'd like to know more about that...but I still don't see the point, espcially if it's just the interface of the chip. A direct connection would be the obious choice of any other forms of transfer... Plus, this certinally wouldn't help with heat because the pins don't generate the heat, it's the processor/transitors...
 
dragon2309 said:
The latest revisons of IrDA can peak at 100Mbit/s, not entirely sure what the data throughput on a CPU would be, or how to measure, it.... Bu i highyl doubt you could implement such a thing, especially in socket such as 939, thats 939 emitters, 939 recievers. Lot more space, lot more heat, just a ridiculous idea really

dragon
Click to expand...
Data throughput on a CPU is MUCH higher than 100Mbit/s, which would really render transmission using infrared useless.
 
computermaineack said:
I did hear a while back that Intel had developed an optical chip, though. They said it could go at speeds of up to 10ghz.
Click to expand...

yea, wireless routers run at like 2.4ghz (regularly at least) and get transfer rates of up to 54mbps. funny because my router is wired, the wire i use runs at 300mhz and i get transfer rates of 500mhz.... using metal substances transfers electricity, heat, and information faster. gold pins would be the most efficient... since i believe it has the highest electrical conductivity rating of any metal. air is an insulator... and a damn good one at that... not a very good conductor. lol
 
The latest revisons of IrDA can peak at 100Mbit/s, not entirely sure what the data throughput on a CPU would be, or how to measure, it.... Bu i highyl doubt you could implement such a thing, especially in socket such as 939, thats 939 emitters, 939 recievers. Lot more space, lot more heat, just a ridiculous idea really
Click to expand...
Well as a quick reference, L1 cache is in the 30-60GB/s range (obviously dependent on the proc involved)

yea, wireless routers run at like 2.4ghz (regularly at least) and get transfer rates of up to 54mbps. funny because my router is wired, the wire i use runs at 300mhz and i get transfer rates of 500mhz.... using metal substances transfers electricity, heat, and information faster. gold pins would be the most efficient... since i believe it has the highest electrical conductivity rating of any metal. air is an insulator... and a damn good one at that... not a very good conductor. lol
Click to expand...
Silver has the highest conductivity of any metal.

doesnt light travel faster that electicity?
Click to expand...
They both travel at the same speed in vaccums however their transmisson speeds through various mediums may vary
 
I suppose but copper traces are, for the time being, more effective (think of the cost it would take to implement say a million pices of fibre optic cabling per motherboard)
 
Praetor said:
They both travel at the same speed in vaccums however their transmisson speeds through various mediums may vary
Click to expand...

May? :P

and yes,.. using fiber optic's could be done.. but think, at the start of a fiber able, you need a transmitter and a reciever, and at the end. ( the good thing about fiber optic's is, it can be used 2 way simultainliously :) ) anyway.. at the end of the fiber optic cables.. youre going to electrical systems again, because there arnt Optical AND or OR's or whatever yet,.. so the processor would be still a chuck of silicium. and the costs would be immense, and without having optical processors... it wouldnt be worth it, because there wouldnt be a speed difference
 
Last edited:
May?
Click to expand...
Yes they may vary.

going to electrical systems again, because there arnt Optical AND or OR's or whatever yet,.. so the processor would be still a chuck of silicium. and the costs would be immense, and without having optical processors... it wouldnt be worth it, because there wouldnt be a speed difference
Click to expand...
Well just the sheel fine-ness of handling a zillion physical "things" that need to be mounted to the motherboard is obscene ... we cant very well etch fibre cables :P
 
Praetor said:
Yes they may vary.
Click to expand...

well... my 'may?' was actually ment to be sarcastic ^^ Im pretty sure they vary,.. and quite a lot even :P for our calculations at school, we just take half the speed of light for electrons going through copper. I know.. not very accurate.. but were a bit lazy on that one. but.. half the speed of light gets quite close :) i think it was about 2Inch per nanosec.. but im not 100% sure, i would have to look it up :o
 
well... my 'may?' was actually ment to be sarcastic ^^ Im pretty sure they vary,.. and quite a lot even for our calculations at school, we just take half the speed of light for electrons going through copper. I know.. not very accurate.. but were a bit lazy on that one. but.. half the speed of light gets quite close i think it was about 2Inch per nanosec.. but im not 100% sure, i would have to look it up
Click to expand...
We're getting off topic ... but i can give you a quick example of when they dont vary (ok im no physicist, maybe it varies but when i said doesnt vary i meant damn-near-no-freaking-difference :P) ... vaccuum :)

In any case, we digress :)
 
You must log in or register to reply here.
Back
Top