As promised, here is / here are the data sheets for Mix Round 1 and it's follow-up evaluation process.
The Song Provider has been in the loop since pretty much the first week of MC067, and he already got the full sheet as well. Those that have been with the Mix Challenge since June 2014, might remember that we already had this type of data sheet once. But considering the workload, we stopped this 7 games in. However, as mentioned in my last post, I am of the opinion that we are
in desperate need of a "straight to the point" summary again. Not only to address the constant debates about the established
Rules and Guidelines, and discontent regarding "rule breakers". But to also show exactly what went wrong, and improve on that in follow up competitions.
For the future: since this is a very time consuming manual process, it might be that the data sheet for the public will arrive up to 72 hours after Mix Round 1 has ended. Just letting you know in advance - I am not a machine after all. The evaluation process by the Song Provider does shift accordingly.
The following paragraphs will also be a one-time occurrence, to set some ground rules for the rest of the year.
Now that this is out of the way, let's get down to the nitty-gritty, shall we?
OVERALL STATISTICS (to be adjusted above):
- we have 109 submissions this month
- We have 7 submission that could not be downloaded by the time of the deadline
- We have 43 disqualifications overall due to the inability to download these mixes, and/or technicalities
That is a 39,45% ratio of disqualifications.
Compared to MC064 / April (since I still have the statistics for that):
There were 45 entries and 17 disqualifications due to technicalities, which is a ratio of 37,78%.
What were the reasons for a disqualification:
According to the
Rules and Guidelines (last update being 2020-MAY-30)
- entries that are not in the correct sampling rate or bar minimum in 24bit (source material was 44/24)
- entries that overshot/undershot a certain Loudness value (lower -23LUFS / higher -16LUFS, +-0,3LU tolerance), or a certain maximum signal strength in dBTP (higher -1,00dBTP, +-0,05dBTP tolerance)
- entries where you could not(!) associate the forum user with (as in - different participation name to the forum username, or no username given at all)
- double entries and/or edited entries
What was not a reason for a disqualification (but pointed out in the additional info section):
- entries that were submitted in 32bit (they will be bit-truncated if you import them at 44/24 - the Rules and Guidelines clearly state to "try to prevent bitrate truncation", as this can introduce noise and glitches)
- entries that were not properly named (as in - differing from the given template on the first post, including typos)
- entries that do not have any documentation (Important this will change once and for all with MC068/August, see below)
| ! | Bonus Info | As of MC068 / August 2020
If there is no documentation about your edit/mix by the time the deadline has been reached (21th of the Month, 23:59 UTC+2 / CEST), your entry will be disqualified.
This particular rule has been in existence since 2014, and has been a constant topic for debates. Please time your submission in such a way, that it will be complete and doesn't need any further edits. The Mix Challenge (game) runs for 21 days. You have plenty of time to set everything up (unless you opt for entering at the last minute, which is not recommended!). Please don't just enter the competition, write "will add the documentation later" and then ultimately never do this.
|
Regarding "tolerances" for LUFS and dBTP
I've settled on +-0,3LU (from the minimum and maximum) and +-0,05dBTP (from the allowed maximum) because of known offsets in metering tools. Usually the tolerances are +-0,1LU and +-0,02dBTP. I've been a bit more lenient behind the scenes for a couple of months now. I also take into consideration that not every measurement tool out there has the same readouts. In fact, if you measure 5 times, you get the most different result every time.
In case of this data sheet, I've used Wavelab 10's Batch Analysis, since it offers me the readouts that you see in the PDF (although I had to clean up the data quite a bit to simplify it, and of course layout everything). Orban Loudness Meter would have given me different readouts. Same goes for Nugen Audio VisLM, Youlean Loudness Meter 2 (Pro), tc.electronic LM2, ToneBoosters EBU Loudness or whatever else is out there.
This does not mean that you can always scrape at the allowed tolerance limits, and then hope that the Song Provider (or me - until end of 2020) will have better results after the analysis. This also
does not mean that you can just mix away and then just... run a "Loudness Normalizer" in the end to hit exactly -16LUFS or similar.
This is not about loudness normalization according to broadcast standards. Keep in mind, this is still a mixing challenge, not mixing
and mastering (which is why the "for broadcast streams allowed +-1LU offset of the target" doesn't apply).
The idea of setting minimum/maximum loudness limits, is for you to learn to work and export within a certain range. Ideally with reference levels - like the known -18dBFS = 0VU (EBU convention R68) or -20dBFS = 0VU (SMPTE RP 0155 recommendation). Which is especially beneficial if you work ITB (In the Box) and with external hardware.
On the topic of "mix runtimes"
I also need to point out, that a lot of entries didn't take care of adding suitable silence at the beginning and at the end of a mix.
While this is not relevant for the competition, as in - this is not part of the rule set for contestants (mix engineers). This is something you should still look out for in the future. Especially if you're doing this professionally, or plan to. Adding suitable silence at both ends of the mix offers the mastering engineer some more leeway (wiggle room) in terms of setting suitable fades (fade-in/out and crossfades). It also means that your audio tools might not cause any hiccups. I actually speak form experience with his - I had tools that changed the first 500ms of a mix in very undesirable ways (think of a fade in, or altered transient - even though the tool in question wasn't a transient shaper). I'm fairly certain that future Mix(ing) Challenges will not have a suitable gaps of silence. Just something to keep in mind.
File Accessibility
And finally: Before you post the link to your mix/edit on the forum, triple check that people can access it without any special permission through the help of a separate browser (Portable App, Mobile Device, etc). This has been a big issue in recent months and I consider this to be one of many careless mistakes. For the future, I also don't want people to point out "file can't be accessed". Participants need to learn this themselves - the hard way (it seems).
| ! | Bonus Info | Revisited for MC068 / August (Rule existing since at least 2017)
If one can not access and/or download your edit/mix by the time the deadline has been reached (21th of the Month, 23:59 UTC+2 / CEST), your entry will be disqualified.
This particular rule has been in existence since 2014, and has been a constant topic for debates. Please triple check the accessibility of your files without any need of a specific login or permission through the help of another browser (portable app) or a mobile device. If you're unsure how to set that up, please also consult the FAQ/Documentation section of the service you're using.
|
With that said...
here is the "Data Sheet" in PDF Form for Mix Challenge 067 / July 2020
Overview of Submissions - full data sheet
Overview of Submissions - disqualifications only
I will update the thread accordingly. Same goes for the "
Global Statistics" on the homepage, and I will also update the "
Rule Books" section the coming days.
Thank you for reading.
