Competition - post a photo you've taken of nature >>

WJEC A2 Computing CG3 Exam - 22nd June 2016

Scroll to see replies

I always lose 6 marks on the psuedocode question.
I also lose marks on the 3rd normal form question (I can never get exactly the same fields/final table correct as the mark scheme)
I also lose marks on the BNF diagrams + entity diagrams.

Any advice?

Reply 21

Carpe Diem Jay

Original post by Okami1997

I'n taking the exam too! I do hope that HCI is the 12 marker, but i think that programming languages would be okay too. I hope they're nice with the questions this year lol

Same, I'm revising by looking at all the past papers, I really really hope that they just repeat the questions. I want an easy paper :/

Reply 22

jay_em

With the pseudocode, do you find knowing what to write difficult, or do you struggle to get it exactly right? Normally the mark scheme says that there are different methods that can get marks if correct, so it's not a case of getting every line correct. Generally you get marks for initialising variables, outputing, incrementing a loop, and stuff like comparing values if that's part of the question.

With third normal form, it may be a help to do an entity relationship diagram (the ones where you draw 'one to many' lines between the different entities), even if the question doesn't ask for it, just to get the relationships clear in your head. I agree that those are tough to get exactly right- sometimes they want composite keys, sometimes not, and sometimes the question is just so ambiguous that you can't tell, like last year.

Backus Naur Forms are frustrating. I'm not sure what to suggest. Is it a case of not knowing what to write, or not being able to get it identical to the mark scheme? Mine are rarely identical to the MS in terms of what I put on each line, and how I split things up, because the MS is just one of many different possibilities. WJEC really should credit different answers if they're still technically correct, but I'm not sure if they do :/

Reply 23

Deathinmycoffee

Original post by Carpe Diem Jay

Yup, really hope they're kind, especially with pseudocode - in one past paper they wanted an array. :frown:

But yeah, good idea to just be doing the past papers at this point and go over any gaps in your knowledge. Good luck for tomorrow :smile:

Reply 24

Carpe Diem Jay

Original post by Deathinmycoffee

Yup, really hope they're kind, especially with pseudocode - in one past paper they wanted an array. :frown:

But yeah, good idea to just be doing the past papers at this point and go over any gaps in your knowledge. Good luck for tomorrow :smile:

Yeah that was a nice paper lol

Thanks and good luck to you too.

Reply 25

Dunphero

I'm in the exact same boat as you, even when I think I've got either of these right, the MS always turns out to be slightly or massively different, which when dealing with pseudo code, BNF or entity diagrams, it means that they would not correctly function in the real world, and there for marks are dropped. My hope is that marks are at least rewarded for correct elements of each..

Reply 26

Dunphero

The prediction that programming languages might come up has been a real scare, it's the only topic that I have not really covered. Provided to us by our teacher was only the PP's from 2011>. Which of course did not come up in any of them so it was not something I considered to be possible. I didn't know there were past papers that date back further than this. Looks like it will have to be added to my last few hours of cramming..

Can anyone provide me with a PP/MS that includes a programming language question?

(edited 7 years ago)

Reply 27

username2450799

Original post by Dunphero

June 2014 has one on question 28
MS: http://pastpapers.download.wjec.co.uk/s14-computing-ms.pdf
PP: http://pastpapers.download.wjec.co.uk/s14-1103-01.pdf

all you need is 4th gen, visual, scripting, object oriented, special purpose and package with programming facilities.

This site is good for some revision: http://moodle.nptcgroup.ac.uk/file.php/1/Content_Library_CIT/TheTeacher2009/theory/topics_wjec_cg3.htm

All the best!

Reply 28

Dunphero

Original post by ggdtg

Thank you!

Reply 29

jay_em

Good night and good luck everyone!

Anyone know the latest, WJEC approved definitions for 'internet' and 'data structure'?

Reply 30

Dunphero

Original post by jay_em

Good night and good luck everyone!

Anyone know the latest, WJEC approved definitions for 'internet' and 'data structure'?

I've always known the internet as a publicly accessible, international collection of interconnected networks. It's a mouthful but would this not be accepted if it doesn't match their definition?

After all it is correct so why would it not :s-smilie:

Reply 31

username1860653

Hey all, what were your thoughts on the paper? I thought it was quite nice! I messed up the pseudoode question though lol
And i was right about the 12 marker being HCI :biggrin:

I was surprised to not see any maths questions though!

Reply 32

The_Big_E

Original post by Okami1997

Hey all, what were your thoughts on the paper? I thought it was quite nice! I messed up the pseudoode question though lol
And i was right about the 12 marker being HCI :biggrin:

I was surprised to not see any maths questions though!

Was a great paper, lots of easy marks from the definition questions. I too messed up the pseudocode, I couldn't figure out how to make it output the longest time stayed, and also the type of visitor. I could do it in a high level language but pseudocode is just annoying.

The 12 marker on HCI was nice, managed to write a few pages for that question due to the generous 3 hours ☺

Also screwed up the binary tree but that was only three marks and i think you get one mark just for the root node anyway.

(edited 7 years ago)

Reply 33

Dasaw

Original post by The_Big_E

yeah good paper overall wasnt sure about the file generation question anyone got an idea?

Reply 34

caley

Original post by Dasaw

yeah good paper overall wasnt sure about the file generation question anyone got an idea?

The three backup files are usually called Grandfather, Father, Son.

If one version is corrupted: the previous version(s) is still avaliable

Reply 35

Mattyh1997

Original post by caley

The three backup files are usually called Grandfather, Father, Son.

If one version is corrupted: the previous version(s) is still avaliable

Yeah it's the idea that the generation of backup system includes the most recent master files with the transaction files so that if the master file becomes corrupt you can apply the transactions in the transaction file to be master file in order to update it to the state immediately before corruption so no data is lost

Reply 36

jay_em

Thought it was pretty alright. A few things I had nailed down in my memory like multiprogramming and compilation stages didn't come up, which was a shame. But the 12 marker was nice.

Did you guys do a composite key for the bookings table, with just CustomerID and HolidayID?

What did you put for the entity relationship diagram question? Think that was new.

Predictions on grade boundary?

Reply 37

username1860653

Original post by caley

The three backup files are usually called Grandfather, Father, Son.

If one version is corrupted: the previous version(s) is still avaliable

oooh right, i got the definition wrong (i was thinking over version number) but i think i got the second part right

Reply 38

username1860653

Original post by jay_em

For the ERD question I think I explained that you couldn't have a "many to many" relationship so you have ti place another entity between the two entities and have either a many to one relationship or a one to one relationship.

Reply 39

jay_em

Original post by Okami1997

I put something similar- I said that there would be replicated data on both sides, so it wouldn't be fully normalised, and you'd have to make a new table in between, perhaps with composite keys. I wasn't sure whether to talk about it in terms of entities, or to bring tables and keys into it.

Oh, and I just realised I talked about attributes when normalising, rather than groups/entities :/