Talk:Eight-to-fourteen modulation

Untitled

I just wanted to clear up something before editing. I could just edit, but there are two ways in which consistency could be achieved, and I don't really know which one is right:

EFM is an NRZ code - uninverted. ECMA-130 clearly states that ones are encoded as a change from pit to land, and NRZI is the inverse (no change)
EFM is an NRZI code, but NRZI is wrongly descripted in the corresponding wiki page - I really doubt this one, as I understand NRZI intuitively just as the wiki describes it, making the previous option the best one IMO.

--Klaussfreire 00:44, 7 September 2006 (UTC)[reply]

The terms are a bit confusing. NRZI is not NRZ, inverted, and inverted NRZI is not NRZ. Inverted NRZI is still NRZI. Mirror Vax 01:36, 7 September 2006 (UTC)[reply]

I'm not sure what your point is, but regardless, it cannot mean that the current state of the wikipedia is correct, as it is clearly indicated in the NRZI page (which is linked in the EFM article and so unambiguously attached to EFM) that state transition represent zeroes, while ECMA-130 clearly states the opposite. The question, thus, is: If NRZI indeed represents zeroes with state transitions, is NRZ the propper classification of EFM? By reading the NRZ page (which I hadn't), the most accurate term seems to be NRZ, inverted (perhaps that's what you were trying to say?). --Klaussfreire 03:22, 7 September 2006 (UTC)[reply]

The premise of your question is that inverted NRZI is NRZ (or inverted NRZ). Since that's wrong, your question is illogical. The NRZI article should note that the opposite assignment of '1' and '0' is commonly used. Other than that, there's no problem. Mirror Vax 12:35, 7 September 2006 (UTC)[reply]

In fact, prior to May 3 edits, the NRZI article had the assignment the other way ('1' = transition). One is not more correct than the other; the choice is arbitrary. I don't know which way is more commonly used. Mirror Vax 13:06, 7 September 2006 (UTC)[reply]

You have a problem with your logic, Vax. if my question was based on that premise, it would only turn out to be irrelevant, but never illogical. In fact, no question is illogical, only reasoning process may. Anyway, my question isn't based on that premise, it just considers it as one of several possible cases. Anyway2, it's clearly confusing the way it is, and since I seemingly can't get a confirmation of what is actually understood by NRZI or NRZ, inverted or whatever, I guess a small clarification phrase may be an acceptable addition. --Klaussfreire 20:09, 7 September 2006 (UTC)[reply]

I agree that it is confusing. Mirror Vax 21:03, 7 September 2006 (UTC)[reply]

About the pourpose of EFM

One of the effects of the EFM code is that it reduces the posibility of failures of tracking, but from my point of view, its far from being the primary purpouse of the EMF codification. The primary purpouse of the EFM is to homogenize the density of bumps in the recorded signal, reducing the frequency at which the reading circuits would work. For the people who are not very familiar with the technical expressions here, the EFM would be the equivalent to an economic optimization of the use of disc space and the use over time of the circuits involved in the reading of information. Instead of having this mess __--_-__-__-_-___-_--_ we have this simpler mess __----__-----____----___ in the same distance with the same meaning of the info.

Consecutive ones

"If there are 2 zeroes between 2 consecutive ones" -- How can this be? D021317c 07:09, 18 August 2007 (UTC)[reply]

Because not all data patterns are used. You're putting 8 data bits into 14 bits on disc. Number774 (talk) 19:40, 6 December 2010 (UTC)[reply]
I agree this makes no sense, and have modified the text to match the example. Kendall-K1 (talk) 14:00, 19 April 2018 (UTC)[reply]
I agree too. The "How it workds" section is full of self-contradicting and confusing statements. In the example cited, the encoded sequence does not have any zeros between 2 consecutive ones. It looks like the article was written by people who have either a poor understanding of the topic, or of the meaning of the word "consecutive":

A sequence 0011 would be changed into 1101 or its inverse 0010 depending on the previous pit written.

AVM2019 (talk) 19:18, 23 June 2020 (UTC)[reply]

2,10 RLL or 3,11 RLL

I just had cause to look this up in the Red Book. (CD specification). To prevent anyone else getting confused, this article is talking about runs of zeroes (obviously bounded by 1 bits) whereas the Red Book talks of run lengths including the trailing 1. Number774 (talk) 19:40, 6 December 2010 (UTC)[reply]