Talk:Final Scratch

This article has not yet been rated on Wikipedia's content assessment scale.

Can someone please post release dates in the History section with references? Thanks! —Preceding unsigned comment added by 71.76.24.36 (talk) 19:14, 12 May 2009 (UTC)[reply]

+1 It's not a history, if whe don't know WHEN things happened. --Panoramedia (talk) 23:19, 4 December 2017 (UTC)[reply]

Could the author of "=== Issues regarding time code errors ===" please clarify the definition of 'a timecode'? You state:

"A bit that has become unreadable due to a scratch can make an entire 40 bit long time code permanently unreadable."

The paragraph makes it sound like a speck of dust can make the whole record useless, which obviously isn't the case. How long is the 40 bit 'bit' in playback time (at 33.3 rpm)?--Djdannyq 12:12, 10 May 2007 (UTC)[reply]

What I meant is that if there is a scratch on the record that makes one of the 40 bits un readable or incorrect, a position code can not be found using those 40 bits. For example, supose that you have a binary string such as 10110 which equals 22 in decimal. If one of those five bits is read wrong, you won't get the correct decimal value. This isn't really an issue in most usage cases because it's rather hard to destroy the record groove enough for a bit to read incorrectly. However, there is still a chance that this could happen. Another analogy might be a non-error-checking transmition code: one bit wrong can corrupt the entire data stream.

My basic point in even bringing this point up in the article is that the coding scheme isn't as robust as other vinyl control schemes.

The code is a 1k hertz along the entire record groove. So at 33 rpm you can do some math to figure out how many degrees the record has to turn to get 40 bits (20 cycles of the time code). I always mess something up in the math when I try to do the conversion so I'll spare the humilation.--71.214.58.157 22:33, 12 May 2007 (UTC)[reply]

Actually, degrees of arc on the record don't really matter so much as the time itself. 20 time code cycles at 1 kHz (plus or minus 8%) is 0.02 seconds, or 20 ms. Then the question is, how does the software playback a corrupted bit? I've gotten some annoying pops in the audio, but was able to avoid them by turning off some Windows services, so I think that's a data-processing issue. As for hardware, I've found that if I have a bad needle or stylus contact on the turntable, I get relative playback—speed and pitch—just fine, but no absolute positioning (which is a real pain if I'm trying to cue up). In lieu of actual error-checking, it seems like the system is still designed to playback and plow right over bit errors in the time code.

My guess is, if 20 or 40 ms are corrupted by dust or scratches, it will continue playing at the pitch based on the last good signal it had. Is FS able to tell if a bit is corrupted, as in 0100110 to 010?110, would FS then know to ignore it or go ahead and misinterpret it? Does it fill in the data with whatever is closest, or does it check against some threshold, as in does [0.01,0.95,0.01,0.4,0.99,0.98,0.02] (analogue on vinyl) translate to 0100110 or something like [error at "0.4", unreliable bit] (digital)? --Djdannyq 09:18, 14 May 2007 (UTC)[reply]

I don't know how Traktor handles it, but in my own software (djDecks) I just check if three timecodes in a row actually point to time positions close together. If one or more bits are incorrect, it is very unlikely that the three time codes will still all point to similar time positions. The time for one timecode is about 8 ms I thought btw. A 1 khz sine wave has 2000 peaks per second (one positive and one negative) AdionC (talk) 14:30, 23 November 2007 (UTC)[reply]

I have moved the following from the article to here, because it seems too subjective and speculative for the article itself:

"With Stantons move to make Final Scratch Open, they decided to use the ScratchAmp for it's real purpose. If Stanton wished to take their product further, they could, instead of using the ScratchAmp for input and output they could use a PCI or PCMCIA sound device that has the same number of outputs."--Djdannyq 12:12, 10 May 2007 (UTC)[reply]

I agree that it is to speculative. However, the switch over to Final Scratch Open makes public what the interface box really is. A 4x input by 4x output audio device with phono amps and some other gadgetry on it. The only thing that the Final Scratch software needs (which ever version one chooses to use) is two stereo timecode audio streams. Someone useing cd decks could easily route the output into their sound cards and not have to worry about having the hardware interface at all. With turntables, phono amplification comes into play and would make the use of sound cards a bit more difficult.

Again, this information probably doesn't have to be in this article.--71.214.58.157 22:33, 12 May 2007 (UTC)[reply]

Final Scratch uses special vinyl records. As you have to use it again and again, don't they wear out within the first twenty tracks you have played? --Abdull 10:15, 26 September 2005 (UTC)[reply]

Yes they do, but you can get replacements for the standard price of a vinyl

It is mainly the high freqs that mostly fade out (of course it will all level out in the end, but not after 20 or 200 plays). I don't know what's on the special record but in my head it doesn't need to have high frequency content.

The timecoded vinyls contain a 1200Hz signal

I used the software pretty heavily for a couple of years before switching to Serato; the vinyl never wore out, though the software did have trouble picking up the timecode on dirty bits of the plate. And there is an option to move the lead-in on the records through software to mitigate needle burn from continuous cueing at the beginning of tracks. Deejaysomething 08:39, 20 July 2006 (UTC)[reply]

Speaking of the timecode on the FinalScratch records, is it standard SMPTE LTC (Longitudinal/Linear Time Code) that's used, much like on the audio (or dedicated linear timecode) tracks of videotapes, or is it proprietary to FinalScratch? misternuvistor 06:07, 12 September 2006 (UTC)[reply]

This page shows a breakdown of the code: http://www.9elements.com/scratchlib/index.php?f_categoryId=14 and I think confirms that is isn't SMPTE LTC. Serato timecode is different again 80.0.9.255 11:23, 15 October 2006 (UTC)[reply]

Here's the RZA thing. I think it should hang out here until there's someone to back up RZA's claim.