Thread: How many bits are '1' in an integer variable?

Originally Posted by itsme86

No. The << and >> operators are more equivalent to SHL and SHR. The difference being ROR and ROL slap the shifted-off bit back on the other end, kind of like a conveyer belt.

I wish you'd stop spreading bad information. If you don't know the answer, don't make it seem like you do.

Well I am not to familiar with the intel instruction set, I was just making the point that bit shifting operations are available in C, MacNilly mentioned these instructions so I assumed they are appropiate, and you may well be able to find a solution incorperating them.
Any way the fact they slap bit back on the end like a conveyer belt would not be an issue in anycase because you only test each bit once, there is a loop counter which counts the number of bits being tested so its not a problem at all.

Originally Posted by Salem

> The C >> and << commands are the equivilant of ROR and ROL
No they are not.
ROL would (in 8 bits) turn 0x80 into 0x01, where the C operator would turn it into 0
<< and >> propagate zeros (mostly), not the bits which fall off the end..

I was merely making the point that C has bit shifting operations, and that an equivilant effect could be achieved in C, whether he chose the most appropiate bit shift operation was not the point.

Originally Posted by esbo

Well I am not to familiar with the intel instruction set, I was just making the point that bit shifting operations are available in C, MacNilly mentioned these instructions so I assumed they are appropiate, and you may well be able to find a solution incorperating them.

MacNilly mentioned no such thing. Stop trying to wiggle out of admitting that you're posting nonsense. If you're not familiar with the IA-32 instruction set, don't assume you mean what ROR and ROL mean.

Any way the fact they slap bit back on the end like a conveyer belt would not be an issue in anycase because you only test each bit once, there is a loop counter which counts the number of bits being tested so its not a problem at all.

Not for the loop method. But if you care to read the link Salem posted, you'll see it mentioned that, using rotate instructions, there's an even better method than those coded out. (Admittedly, MacNilly's method would work with shifting and gain no advantage from rotating. Not my point.)

I was merely making the point that C has bit shifting operations, and that an equivilant effect could be achieved in C, whether he chose the most appropiate bit shift operation was not the point.

If that is your point, you would do well to actually say so. What you said, though, was:

The C >> and << commands are the equivilant of ROR and ROL

And this is wrong, plain and simple.

Originally Posted by esbo

I used the word equivilent which is defined as:- "Having similar or identical effects." So no it is not identical, but it does have a similar effect for the purpose the OP intended, ie counting bits.

You are wrong! Now shut up! This is why everyone here gets so mad at every single post you make! You're wrong. ALL THE TIME!

Damnit just look at what you wrote above. That's wrong too!

"Having similar or identical effects."

They do not have identical effects! There is no "reading out of context". All you have to do is follow the thread, and here you are being wrong again. There is no context you could fabricate that would make you right here.

Originally Posted by esbo

But I think it's pointless going into an arguement over semantics the language is C not English language, I don't claim to be an expert on the English language!! Nor C for that matter,

No. No it's not. The entire point of the C standard is to set up these semantics, and behaviors so you know exactly what is going to happen when you do what you're supposed to.

Originally Posted by esbo

I just like to write programs that work!!

Well it's a good thing you didn't say "programs that are correct", because you always are doing stupid ........, and when people call you on it, you say "I don't care! It works for me!"

Stop polluting the forums. Stop pretending you know what you're talking about. Stay out of other people's threads, because all you do is cloud the issue.


Quzah.

Originally Posted by quzah

No. No it's not. The entire point of the C standard is to set up these semantics, and behaviors so you know exactly what is going to happen when you do what you're
supposed to.

Exactly!
Discussing semantics within the context of a programming language is more than just being pedantic. It's a necessity.

Originally Posted by esbo

If you read it out of context then yes technically it is incorrect, I admit that. However I think most people reading the thread would understand the point I was making.

Well, I did esbo. I understood that "the C >> and << commands are the equivalent to ROR and ROL". As such, you misinformed me.

Which would be ok, really. Everybody makes mistakes. But the fact you do it all the time, but worst, you try to cover your tracks behind a bad excuse of "I didn't mean it like that", is what is driving everybody nuts.

Most people do a genuine effort to help others in here. They feel their efforts jeopardized by someone that they feel have to be constantly watching because is constantly wrong and directing posters the wrong way.

I think, esbo, this is perhaps the time for you to stop posting. Please.

This looks done. Closed.