ROBOT.LK

Posted: **Sun Apr 01, 2012 5:04 pm**

Hello my friends,
I like to study about Microprocessor's architecture in simply.Now i want to know about RISC and CISC.i have some small idea about this.i have read Wikipedia articles.But,i cant understand it completely.

Please someone tell me about this.Or give links or e-Book which this describe simply.

Thans

Posted: **Sun Apr 01, 2012 9:22 pm**

We had about these in BIT Semester 1 Computer Systems 1 subject, Even though I don't have a clear idea

Posted: **Mon Apr 02, 2012 12:34 am**

Nipuna wrote:We had about these in BIT Semester 1 Computer Systems 1 subject, Even though I don't have a clear idea

thank for your post Nipuna.

Please Can another friend get me a solution?

Posted: **Mon Apr 02, 2012 8:25 am**

Try these 2 books I think in these 2 you will fins them.

Computer Organization and Architecture by William stalling
Essentials of Computer Organization and Architecture by Linda Null & Julia Lobur

Posted: **Mon Apr 02, 2012 8:30 pm**

Best book for this is Computer Architecture: A Quantitative Approach, 4th Edition by John L. Hennessy (Author), David A. Patterson. This is the standard university reference.

The major difference is that RISC chips use simpler instructions sets to achieve higher clock frequencies and process more instructions per clock cycle than CISC processors. Typically, CISC chips have a large amount of different and complex instructions. The philosophy behind it is that hardware is always faster than software, therefore one should make a powerful instruction set, which provides programmers with assembly instructions to do a lot with short programs.
In common, CISC chips are relatively slow (compared to RISC chips) per instruction, but use little (less than RISC) instructions. Intel and AMD, for example, develop CISC processors (x86), while Apple and SUN use RISC architecture. Major problem of RISC - they don't afford the widespread compatibility, that x86 chips do.

There is still considerable controversy among experts about which architecture is better. Some say that RISC is cheaper and faster and therefor the architecture of the future. Others note that by making the hardware simpler, RISC puts a greater burden on the software. Software needs to become more complex. Software developers need to write more lines for the same tasks. Therefore they argue that RISC is not the architecture of the future, since conventional CISC chips are becoming faster and cheaper anyway.

RISC and CISC architectures are becoming more and more alike. Many of today's RISC chips support just as many instructions as yesterday's CISC chips. The PowerPC 601, for example, supports more instructions than the Pentium. Yet the 601 is considered a RISC chip, while the Pentium is definitely CISC. Further more, today's CISC chips use many techniques formerly associated with RISC chips.

So simply said: RISC and CISC are growing to each other.

Posted: **Tue Apr 03, 2012 11:03 pm**

This is my short note on CISC vs RISC(hard hard time finding this).

Code: Select all


RISC                                                        CISC

stands for reduced instruction set architecture             stands for complex instruction set computing.
(examples-ARM,IA64,PowerPC)                                 (examples-x86,x64)

Simple and Fixed length instructions                         Variable length instructions.
Because of this it's easy to design compilers.               There are around 10,000 of opcode+operand combinations
                                                             so compiler backend have hard time in emitting optimized
                                                             instructions, Probablly heuristic algorithms have been used.
                                                             Because those algorithms are all NP complete when it comes
                                                             to CISC.

Load and Store Architecture.                                 No need a Load and store architecture, you can directly
(read the wiki page on this)                                 access memory.

Instructions can be directly placed into the                 Instructions are typically grinded into the microcode before 
pipeline.                                                    putting them into the pipeline.

Have higher number of registers.                             Have higher number of cache memory cells.
(to load and store)                                          (to increase the performance of memory accessing).

Have high CPI ratio.                                         Low CPI ratio.
(wiki for CPI)  



and also one Instruction in CISC matched with many instructions in RISC for a example, take multiply operation.

RISC way
========
   LOAD A,memory_loc_1
   LOAD B,memory_loc_2
   PROD A,B
   STORE A

(did you read about Load Store architecture ?if not this will be geek).

CISC way,
=========

    mov eax, ds:[offset memory_loc_1]
    mul eax,ds:[offset memory_loc_2]
    mov ds:[offset result],eax

Likewise CISC have compat instructions,so it will definitely increase the CPI ratio and discrease the memory
traffic too.And CISC undirectly exploit the same benifits of piplined processors just by griding the CISC 
instructions into microcode by the processor front end. ( google for 'microcode' and 'processor font end' ).
So personally what I'm thinking is CISC will dominate the markert, good example is Mac World have already 
landed on the CISC x86_64.

Posted: **Wed Apr 04, 2012 8:27 pm**

Thanks both ZevenZero,Sandun and Nipuna,
The ZevenZero's short description is well.
And Sandun's answer is more difficult to understand.However,I understood some using google.but some

I must search others.ZevenZero and Sandun motivate me for it.
Thanx all

I also used following resources,I hope,those will be important to ROBOT.LK readers...

I think,i will be able to post later for this topic.
PowerPC601/601v RISC Microprocessor Hardware Specifications
http://www.datasheetcatalog.org/datashe ... C601EC.pdf
PowerPC
http://en.wikipedia.org/wiki/PowerPC#Endian_modes

Posted: **Wed Apr 04, 2012 8:32 pm**

Hi All,

SevenZero wrote: .....
.....
In common, CISC chips are relatively slow (compared to RISC chips) per instruction, but use little (less than RISC) instructions. Intel and AMD, for example, develop CISC processors (x86), while Apple and SUN use RISC architecture. Major problem of RISC - they don't afford the widespread compatibility, that x86 chips do.
.....
.....

What is Widespread compatibility which above ZevenZero said?
And,
What is CPI ratio of microprocessor?This is has several location of Sandun's Short note.

Thans

Posted: **Wed Apr 04, 2012 11:30 pm**

What is Widespread compatibility which above ZevenZero said?

This happens to anybody who is trying to jump in to the subject without learning it systematically.
It means the RISC processors are incompatible with popular CISC architecture chips such as x86 made by both Intel and AMD. CISC architecture chips are widespread (popular and distributed in massive quantities) than RISC these days. When you step out, you see this reality. How many Apple machines (mostly with PowerPC chip) you have seen and how many IBM compatibles with Intel or AMD chips.
So nothing complicated as you think in that word.

What is CPI ratio of microprocessor?This is has several location of Sandun's Short note.

CPI is for cycles per instructions. This is one of the ratings used to measure performance of a CPU.
In most of the processors we have today are single-CPI (or 1 CPI) for instructions like addition. Say you a 1GHz chip. That means you can perform 1,000,000,000 number of additions per second.

A more popular rating is MIPS or Million Instructions Per Second since almost all chips can perform millions of instructions per second. Even the small Rs. 500 RISC microcontrollers work in 20MHz and those can perform most of the functions such as ADD, SUB in single-cycle. Say you have 20 Million ADDs to perform. This type of cheap microcontroller can complete that 1 second.

This is a very important advise to you. Systematic learning is very important when it comes to technology subjects. You can become a 'Jack of all trades, master of none'. This is not worth in technology field. You'll never be able to do anything useful. The subject you are referring is called "Computer Architecture" and the university standard reference has been given to you before. The Indian publication is around Rs. 1000 (according to my memory). This must be also available in public libraries. When you learn through such a standard book, you will get to know that you are learning the subject in full. So at the end of book, you will feel confident on yourself about the subject. It doesn't required to go to a university to do that. If you can follow these books, ask questions in places like ROBOT.LK and clear out problems, I'm sure you can become a master.

Good luck Nandika buddy!!!!!

Posted: **Fri Apr 06, 2012 10:37 am**

Thank you ZevenZero,
You are correct.my knowledge is not clear.i must improve my knowledge systematically.it's essential for me.thank again your for your advice.
And,i do it for C/C++ now.However,i like to increase my unclear knowledge.it will be clear future with ROBOT.LK.

I can find book which you said before.but,not now.because,i study C++ now.

Thank you

ROBOT.LK

What are RISC and CISE?

What are RISC and CISE?

Re: What are RISC and CISE?

Re: What are RISC and CISE?

Re: What are RISC and CISE?

Re: What are RISC and CISC?

Re: What are RISC and CISE?

Re: What are RISC and CISE?

Re: What are RISC and CISC?

Re: What are RISC and CISE?

Re: What are RISC and CISE?