New answers tagged

1

Figure 1. Windows Caculator in Programmer mode. (Menu: View | Programmer.) You may find your Windows (or whatever OS you use) calculator helpful in this regard. Calc makes the distinction clear by using Word for 16-bit, Dword (double) for 32-bit and Qword (quad) for 64-bit. In the screengrab shown in Figure 1 I have selected Qword (4 x 16-bit = 64-bit) ...


2

Your interpretation is correct. In terms of the line: "the total is 4 words x 4 bytes = 16 bytes which is 128 bits" That is the total size of the memory. Each word in that memory is 32 bits, and you have 4 of them giving a total size of 128 bits.


0

Very few designs of CPU are able to adjust their speed of operation based on continuous variations in the speed of the internal logic. Where this has been tried, its likely to be purely for research projects. Such a design might be termed asynchronous (even if it uses variants of classical sequential cells and a recognisable pipeline structure) due to the ...


3

If a CPU produces any errors at all in computations, we consider it "broken". If the CPUs temp specification gives it's normal operating temp range as 0 ~ 95 ˚C (for example) it will produce no errors over this entire range. However, in general, semiconductor devices will draw less power at lower temperatures, so, ironically, if you spend some power to cool ...


8

methinks you are referring to this: (from the Intel website) and the description reads (again from the Intel website) Micro-FCBGA (Flip Chip Ball Grid Array) package for surface mount boards consists of a die placed face-down on an organic substrate. An epoxy material surrounds the die, forming a smooth, relatively clear fillet. Instead of using ...


17

In most if not all modern processors, the silicon is flip-chip bonded onto an interposer which then has all the connection pads on it. As a result the back of the silicon die is then at the top - pointing to where the heatsink is attached. In desktop processors, this is then typically bonded with thermal compound to the top metal shell, thus allowing good ...


0

simulate this circuit – Schematic created using CircuitLab Figure 1a. CMOS output configuration. 1b. Hard-wired switch representation. ... we do not have direct current transfer from Vdd to Vss ... We do this by ensuring that only the pull-up or pull-down transistor (M1 or M2) is turned on but never both together. Figure 1b might make this a ...


1

This is note quite clearly forumlated, but I guess what the book meant is that typical gate chips are not drawing current at steady state. This wasn't true on older days (when TTL ruled the world), but nowadays, this is almost true with the CMOS chips. Take a look at a the implementation of a NAND gate, for example: Whathever the state of the inputs, you ...


0

For sure there is energy loss - each time the gate output or input changes state there is capacitance charged or discharged - this energy can never be recovered so that gets dissipated as heat. Energy loss is therefore proportional to switching frequency.


2

In the absence of any better answer, I'll try to answer my own question based on findings since I first submitted the question some weeks ago: die area requirements for various MIPS architectures is available here. The table is incomplete, and the closest information to a MIPS32 74K is the MIPS32 24K at 0.83mm^2 (unclear for which process geometry). ...


3

The CPU chips that you can buy (for example, Intel i7) are optimized for the maximum speed and lowest power available at the current level of high-volume manufacturing process technology. To be sure they are always working on the feature set, circuit design, device technology, and the manufacturing processes that will help make the next generations of CPU ...


6

I worked for a decade at a startup doing computerised logic improvement for chip designers; we were eventually bought by Cadence. We had, among a whole bunch of other tools, a heuristic logic minimiser. It was a fairly simple thing that just pushed "bubbles" (negations) around, and attempted to absorb gates into the slightly more efficient AOI or OAI ...



Top 50 recent answers are included