Self Assessment B Anwsers?

Do any of you guys have them?

(nervous smile)

Sorry no such luck. I suppose we'll get such things in the next few weeks, though?

There were some answer at the very back of some of the tutorial letters too. Maybe double check there?

Thanks Ed

How is the preperation going?

I havent gone through all the material yet, and I've been working a lot.

I also get nervous when I think about the chap 4 stuff, while I'm working on the chap 7 stuff.

I have now decided that I'm gonna use some of my time on the old stuff and some on the new stuff, preparing for this is very tricky.

How do you go about it?

I think some of the Self Assessment Questions are gonna be exactly what we are gonna be asked in the exam.

Havint TOUCHED chap 8 yet, very concerned about that. I just gonna spot some possible questions.

The previous exams look HECTIC

The Ch 4 stuff is the most interesting to me. That doesn't mean it's at all easy to me, mind. Just like the fact that this course is really interesting doesn't make it any less daunting, does it?

Give Ch8 a skim. It's not that hectic. It's almost "the first step of the next level". (Don't be surprised to find it fascinating either).

To make Ch4 more manageable, try cutting your matrix down to just one row (notionally).

So now you're going to "multiply" a row by a column from the right.

That's just a sum of products.

You get the entry of the Result Vector for that row. One sum; one result.
===

OK now reassemble it all in your mind. All you're doing is doing that four times over. Four results.

Grand final result? Just another vertex. Same old same old. It's been altered a bit by multiplications.

===
Now in all cases getting away from the How, the What of what you're trying to do is: For any single "point" (refine that to "vertex" if you're cool with the difference, if not, "point" will do here)..

er... Oh yes...

For every point [ x y z w] ^T you want to move it exactly the right distances.

So simplify again. For every x (in some point, of course) you want a new x. That's a row by your column.

You're saying "To get new x I might have to alter the y and z of the old one a bit too." (Remember the row is just an equation?) ax + by + cz = new x.

Do that four times. Most of the time b and c are just 0 (meaning "do nothing".

What you're trying to do is shuffle each (and every) point/ vertex around to some new next position.
===

Right. Last little bit for the time being:
When you call eg. glRotatef(....) you're actually "calling a MATRIX".

That was a point I had to get straight in my head at some point. The function sets up those rows of equations (the matrix), and then it right multiplies that with the one below it.

You could roughly think of
glTranslate.. // T
glRotate.. // R
glScale.. // S

as T(R(S))) (put those braces they drop for "elegance" back in, and the way it operates becomes clearer).

Right. I'd better stop there. Something tells me I might've got some of that last part slightly backward. Might've. But why risk crossing over into confusion?

There's no Self Assessment B under the additional resources on myUnisa, but having skimmed over the huge collection of exams with answers that our lecturers have supplied there, I think I've seen at least one of the questions answered in the exams. Could be wrong though. In the unlikely event that you haven't looked there lately, there's a ton of stuff under "Additional Resources". At very least it'll help us boost our understanding a lot. No harm in understanding what the course is all about eh?

Oh and I hope nobody's insulted by my "back to very basics" posting further up the thread. No such thing intended.

Hi Edd

No worries man, I'm still working through the text book, almost done with Chap 7,
I gonna skim Chap 8, and start with the Self Assessment B stuff.

I'm gonna tri and post my own answers here.
I think we are gonna get questions that look a lot like those. Maybe even the same ones. (gut feel)

I have seen those additional stuff, very nice to have the memo's, maybe I'll only tri the Self Assessment after reading and writing out the memos

onelove

Smoky Rots

On the one hand there's plenty of new lights that have been defined for the illumination of our brain cavities (Even some specular components, nogal). On the other hand the course looks almost more daunting than before. There's a fair bit to pack into a tiny space.

Don't miss out on Ch8, ROtti, even if you just visit it for a fleeting acquaintance. What you start doing there is fiddling with the buffers that are written out once your GPU has adjusted your current geometry. "The pipeline goes on"..

And it gets cooler. I downloaded a few of the UC Davis videos earlier in the semester, when it seemed like the clocks had not yet been fitted with accelerators. Turns out it was a postgrad course - just the seminars for it, though, so nothing to crazy intense. What they're doing is using GPUs to do massively parallel computation. Guys have been hacking them to make them crunch numbers as "textiles" etc., and now the graphics card companies are facilitating direct programming of the GPU ...

But don't worry, that's not in Ch8. Just scary that just the tip of the iceberg we encounter looks big, eh?

Whatever happens it was worth the journey.