Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Die photos and analysis of the revolutionary 8008 microprocessor (2016)

righto.com

Post reply on HN
Print view View original

11–17 of 17 posts

Re: Die photos and analysis of the revolutionary 8008 microprocessor (2016)

#11

Post by Mitzz »

bogomipz wrote: post #4

The picture under the heading "The structure of the chip" has two arrows pointing to polysilicon regions. Does anyone know why the polysilicon region that the upper arrows points to is os much larger than the polysilicon regions that the lower arrow points to? They seem like a couple of order of magnitude larger.

I know slightly less than nothing about this topic, but I remember reading this from a while back:

"Notes for reversers Electron holes have more effective mass, and therefore less mobility. P-type regions are thus made bigger than N-type regions to compensate. The size disparity often differentiates P-type and N-type silicon."

Source: http://siliconzoo.org/tutorial.html

Maybe that is helpful...?

Re: Die photos and analysis of the revolutionary 8008 microprocessor (2016)

#14

Post by kens »

bogomipz wrote: post #4

The picture under the heading "The structure of the chip" has two arrows pointing to polysilicon regions. Does anyone know why the polysilicon region that the upper arrows points to is os much larger than the polysilicon regions that the lower arrow points to? They seem like a couple of order of magnitude larger.

The "extra" polysilicon region is a capacitor made of silicon and polysilicon. These capacitors were a "bootstrap load", essentially a way to drive a transistor harder to get a stronger output. The capacitors you're looking at are for the instruction register to drive the signals through the instruction decoder (basically a PLA). Since the decoder is fairly large, the signals into it need to be boosted. I haven't looked at this for several years, so I may be wrong about some of this.

http://www.intel4004.com/btstrp.htm

Re: Die photos and analysis of the revolutionary 8008 microprocessor (2016)

#15

Post by close04 »

sand500 wrote: post #6
Arbalest wrote: post #5

I'm curious, what could be made out of a processor in the order of 4000 transistors as here, but with entirely modern chip design and fabrication tech? What could you put out on TSMC's 7nm, presumably an extremely small wafer.

maybe the cpu in a sd card?

Apparently those are derivative of a much newer controller, the Intel 8051 (1980), heavily modified, running at up to 450MHz, and with added hardware accelerators.

https://en.wikipedia.org/wiki/Intel_MCS-51#Use_as_intellectu...

Re: Die photos and analysis of the revolutionary 8008 microprocessor (2016)

#16

Post by bogomipz »

Mitzz wrote: post #11
bogomipz wrote: post #4

The picture under the heading "The structure of the chip" has two arrows pointing to polysilicon regions. Does anyone know why the polysilicon region that the upper arrows points to is os much larger than the polysilicon regions that the lower arrow points to? They seem like a couple of order of magnitude larger.

I know slightly less than nothing about this topic, but I remember reading this from a while back: "Notes for reversers Electron holes have more effective mass, and therefore less mobility. P-type regions are thus made bigger than N-type regions to compensate. The size disparity often differentiates P-type and N-type silicon." Source: http://siliconzoo.org/tutorial.html Maybe that is helpful...?

What a great link, thanks! I had not seen this before.

Re: Die photos and analysis of the revolutionary 8008 microprocessor (2016)

#17

Post by bogomipz »

kens wrote: post #14
bogomipz wrote: post #4

The picture under the heading "The structure of the chip" has two arrows pointing to polysilicon regions. Does anyone know why the polysilicon region that the upper arrows points to is os much larger than the polysilicon regions that the lower arrow points to? They seem like a couple of order of magnitude larger.

The "extra" polysilicon region is a capacitor made of silicon and polysilicon. These capacitors were a "bootstrap load", essentially a way to drive a transistor harder to get a stronger output. The capacitors you're looking at are for the instruction register to drive the signals through the instruction decoder (basically a PLA). Since the decoder is fairly large, the signals into it need to be boosted. I haven't loo…

Your link has a great pic of the original schematic print out. Might you or anyone else know if there is a high-res version of the printout online anywhere? Cheers.
Post reply on HN