Logic Blog - updated 2011/02/06

[Rogar]

Sure, the wobble bolt was an eye-opener, and I have never had the need for an XOR-gate, so even the piston version was useful to see.

Can't wait to see the negative thermo OR-gate...

[comphermc]

Now, is an XOR something that triggers on an odd input, or just a single input?

Can we not just use a 3-way AND gate from the Logic Pack, and position the key-switch/key to activate when only one of the pistons extends? Then, this can be extended for any number of inputs (pistons).

[rtm223]

Odd number of inputs, sadly.

It just happens that with two inputs, the only odd number is 1

.

[comphermc]

Now, pardon my dense-ness but what is the function of a XOR beyond 2 inputs. I understand the idea of binary adder, yes, but it would seem to me that the "exclusive" part of XOR loses its pizazz when you get an activation upon three inputs.

My understanding (which is apparently low), would suggest that XOR should only activate when there are exactly 1 true input.

Am I missing something basic?

[rtm223]

Well, OK. I don't think they actually make XORs in the form of greater than 2 bit, so in a technical sense it's probably undefined.

But:

a b c

Should be the same as

( a b ) c

Right? A two bit XOR, whose output is XOR'd with the third signal, like in the adder diagram. Well that comes out as an "odd" function. So conceptually, a 3 bit XOR works out as an odd. Or at least in my head it is!

[comphermc]

Well, OK. I don't think they actually make XORs in the form of greater than 2 bit, so in a technical sense it's probably undefined.

But:

a b c

Should be the same as

( a b ) c

Right? A two bit XOR, whose output is XOR'd with the third signal, like in the adder diagram. Well that comes out as an "odd" function. So conceptually, a 3 bit XOR works out as an odd. Or at least in my head it is!

a b c

Is the same as (a+b+c) • !(a•b) • !(a•c) • !(b•c) (If this is wrong, then ignore me)

While, ( a b ) c

Is the same as [((a+b) • !(a•b)) + c)] • ![((a+b) • !(a•b)) • c]

...

I'm not sure that helped, but the point is that (a+b+c) • !(a•b) • !(a•c) • !(b•c) is not true as soon as you have 3 inputs. You can see that from all of the not, AND conditions. As soon as one of those is false, the whole thing kerplodes.

The second one does allow three inputs, but the first one does not... They are not quite the same.

Dangit, I hate it when you are right. XOR is indeed associative (upon some research). You win again...

[rtm223]

That's weird, I'm the complete opposite. I love it when I'm right

If it makes you feel better, I can never remember the words like associative


Edit: Just to confirm, there will be no logic blog this week (23rd), blame water, christmas and grant for getting me all excited about doing other stuff. Also it's gonna be another big'un (c. 3k words again, so it takes a little while, like). Normal service will resume next week

[rtm223]

Yay!!!!

(new blog out)

[Rogar]

About time! Leaving us starving for so long...tsk.

[rtm223]

Yeah I know Between losing almost the entire thing when I'd nearly completed it the first time, various real-world issues and generally not feeling like rewriting it all, it's taken quite a while to get this together. I make a lot of effort to be less tardy in future!!

[Rogar]

No worries, at least you produce something worthwhile once in a while, which is more than some of us can say...

[rtm223]

Well, I just wrote this as a comment on the last blog, but I figure it makes sense to post it here as well.

Kinda ****** off here. I went to revisit my latency tests last night, and as it turns out, it's not that difficult to get stuff switching at 0.05s in both directions. Now I swear I tried a good few times and failed before, but last night thrice (yes thrice) in a row it was working. Which is a shame because I came up with some clever stuff to get around a problem that apparently doesn't exist.

I'm blaming leerdammer.

You do have to be really careful with your mag switch tweaking still though, and latency is quite an interesting topic, so there is probably some mileage in a latency blog.

[Tamland]

I'm crawing for a new blog-post about logic!

Give it to me! please...

[Rogar]

Gee, I recently discovered how to subscribe to blogs, but how do I test I did it right...?

[rtm223]

I wouldn't advise sitting next to your email inbox waiting.... Not unless you have a LOT of snacks. Logic blog will be returning as soon as possible, but I'm giving no guarantees on when that will be.

Worst case I'll just knock together some explanations of some unique logic pieces I have knocking around that would allow me to talk about some interesting concepts.