2020-DEC-01 Info: You can shape the upcoming "Community Scramble" - quite literally. Check out the sample collection thread for SWC041 / January 2021

Songwriting Competition - General Gossip Thread

Ask us a question, give feedback, join surveys, make suggestions
Olli H
Backer
Posts: 130
Joined: Fri Dec 29, 2017 16:08 CET

Re: Songwriting Competition - General Gossip Thread

#141

Post by Olli H »

Double Tap wrote:
Wed Oct 14, 2020 10:17 CEST
If the proposal is that you can enter a song and take a 10 point hit for not bothering to provide feedback, but still have your song in the competition, then why bother voting? It takes a substantial amount of effort to listen to everything and write feedback. I doubt many people would be cynical about it, but it would be tempting on eg October 29 to think, oh, I'll do it later, and never quite get round to doing feedback.
Agreed. The real feedback is probably the most valuable part of these competition. It shouldn't be risked.

User avatar
Mister Fox
Site Admin
Posts: 1910
Joined: Fri Mar 31, 2017 16:15 CEST
Location: Berlin, Germany

Re: Songwriting Competition - General Gossip Thread

#142

Post by Mister Fox »

I fully understand the discontent with the "every time it is extremely painful for me to categorize others as losers and total losers". But rest assured, you are already winning by joining the game (if you stick to the very basic rule of "create something from scratch", that is). You merely set "your" personal preference for this month. Next month it might be completely different.

However I also get what you two infer. On one hand, the regulars (those that are with us for longer than 3-5 games) mainly want feedback and improve. We had it a couple of times already that the top spots passed on their "freebies". On the other, the newcomers want to get on the score board since their incentive might be different (maybe focus on the bonus licenses).

I get that people might see a way more relaxed rule set as "meh, why should I even participate - it's just points, I won't win anyway?!".



:?: My main question would be: how could we ideally handle this in the future?

Do we still keep the rule that it is mandatory to fully participate in the voting process, else you're disqualified?
Should this particular one be reworded, or can it remain as is?



The thing is, with the bonus incentive to win free goods, we (sadly) need a certain "voting" system. I really try to find a balance between "learning factor" and "frustration". Being fair to all participants without handing out participation medals.

Olli H
Backer
Posts: 130
Joined: Fri Dec 29, 2017 16:08 CET

Re: Songwriting Competition - General Gossip Thread

#143

Post by Olli H »

Mister Fox wrote:
Wed Oct 14, 2020 19:59 CEST
:?: My main question would be: how could we ideally handle this in the future?

Do we still keep the rule that it is mandatory to fully participate in the voting process, else you're disqualified?
Should this particular one be reworded, or can it remain as is?
I don’t see any problems with testing the new rules. If there’s a new idea, why not implement it for a couple of months and let’s see what happens. I’m glad to speculate what might happen, and I’m glad to share my feelings, and I’m glad to change my opinions on daily basis.

TrojakEW
Posts: 138
Joined: Sun Sep 23, 2018 18:15 CEST

Re: Songwriting Competition - General Gossip Thread

#144

Post by TrojakEW »

I think the new idea behind bonus points is good approach. It may be need to balance those points. I do agree that feedback is main part of this competition so it should be most ideal that biggest penalty will be for not providing feedback, like half of the points. There will will be always somebody who will not like the rules and the way how they are set up, so bending them over and over can cause more problems or rather said complications.

User avatar
Mister Fox
Site Admin
Posts: 1910
Joined: Fri Mar 31, 2017 16:15 CEST
Location: Berlin, Germany

Re: Songwriting Competition - General Gossip Thread

#145

Post by Mister Fox »

Thank you for the feedback so far. It took a couple of days to think about the topic. And I come to two conclusions.

1) we stay a bit more strict, but the main point for disqualification will be, if you do not participate in the final feedback round
2) we will drop the "disqualification" mechanic, and completely rely on the bonus points

Both can have it's strengths and weaknesses. The first one makes clear, that this is not just "Here, have my track... thank you for the feedback, but you won't get any from me". The second one is like "meh, i don't care if I'm at last position - I'm not disqualified, and I got both free exposure and feedback". So I'm currently heavily leaning towards sticking to the known game mechanic, but slightly overhaul it.





:arrow: Which brings me to the following as a test run for SWC038 already:

 ! Bonus Info
Participants of the Songwriting Competition can score up to thirty (number: 30) global bonus points on the score sheet for the following situations
  • 1 bonus point for properly naming your files and making them downloadable by the time the deadline has ended, or up until the "submission summary" has been collected and posted (usually happening between 6-12 hours after the submission period has ended)
  • 2 bonus points each (up to 4 points max) for submitting a mix equal or lower than -14 LUFS ILk (+-0 LU Integrated) and -1 dBTP max -- if you only got 2 points, your track either exceeded -14 LUFS ILk or it clipped (-1.00 dBTP max). Please have an eye on that in the future.
  • 5 bonus points for properly documenting what equipment was used during the production (you don't have to go into super detail regarding plugins per channel), and what your idea behind the production was
  • up to 20 bonus points for creating your "top 20 entries" during the voting process (part of the "Winners Podium" mechanic to get a chance to win bonus software licenses). One point per participant. If there are less than 20 participants, then only as many points as there were entries, excluding your own (example: 12 participants, then 1 point per vote, result: 11 bonus points)


Listing on the score sheet in the future:
  • File Consistency (1 pt)
  • Loudness / dBTP (4 pts in form of 2+2 pts)
  • Documentation (5 pts)
  • Voting Process (up to 20 pts)
Total: 30 pts



Change of wording in the rules:

From:

Each participant must(!) vote for his/her personal Top 20 favorite compositions

Important: it is mandatory for everyone to participate in the voting process. If you do not give final feedback and/or cast your vote, your entry will be disqualified from the final results



To (draft):

Important: In order to declare a Winners Podium, each participant must(!) vote for his/her personal Top 20 favorite compositions. If you do not cast your personal top 20 vote, or maybe just your top 5 (since you are not feeling comfortable declaring your top entries this month), you will not be disqualified. But you will loose out on the majority of your bonus points (see Bonus Point Mechanic)

Important: it is mandatory for everyone to participate in the so called "voting process". If you do not give some final words of encouragement how to maybe improve one's production, your entry will be disqualified from the final results. Giving constructive feedback and interacting is an essential part of the Songwriting Competition.
While the "up to 20 bonus points for 20 votes" is a bit more work, I think it adds an additional dynamic to the overall concept. Also, as some of you pointed out, if you're not feeling okay with casting "Top 20" a certain month, but only like a "Top 10", you would still contribute. You won't be disqualified, but you are(!) drastically set back. I just hope this doesn't open doors for misusing/boycotting this game mechanic.

The rest of the rule book on "voting" remains the same.




:arrow: Technically, this can still be trimmed in "fat". Especially since the last paragraph is basically doubled in the "Rule Book" with this:
It is encouraged to commit to your vote sheet. There is absolutely no shame in not liking a specific production at a certain given time, or because it's not your particular cup of tea this time around. Everyone joining the Songwriting Competition is a winner - you are not worth less by getting an overall lower score. But in order be able to address bonus content, we can only have a winners podium of three.


But the whole "Voting Process" should feel a bit "less strict" now, yet will still sting in the long if you do not select your top 20 entries.

I might need to think of a fall-back mechanic though, to prevent the upper mentioned "why should I even vote?!"
Example: if we have 10 participants and less than 6 top 20 votes, there will be no winners podium




We still have 3,5 days to refine this and set clear ground rules that are engaging, yet fair for everyone involved.

:!: Your additional feedback is important in this case, and more than welcome.
Thank you

A Future in Noise
Posts: 80
Joined: Wed Nov 07, 2018 09:26 CET

Re: Songwriting Competition - General Gossip Thread

#146

Post by A Future in Noise »

It looks like the Youlean loudness meter is all you need to measure LUFS. But I want to know what LUFS is! OK – Loudness Unit Full Scale. But it must be some kind of mean value. So how often is the loudness measured? Once per second? If a program calculates the mean value for the whole track, that would mean that I can be extremely loud at the end, if the beginning of the track is very quiet, but I can't imagine that this is how it works. I have been trying to find information about this on the Internet, but I haven't found any. So if you know about the time aspect, please let me know.

User avatar
Mister Fox
Site Admin
Posts: 1910
Joined: Fri Mar 31, 2017 16:15 CEST
Location: Berlin, Germany

Re: Songwriting Competition - General Gossip Thread

#147

Post by Mister Fox »

A Future in Noise wrote:
Sun Nov 15, 2020 14:09 CET
So if you know about the time aspect, please let me know.
As this can turn into a really long topic, I highly advise you to read up on the whole concept and technical aspects with the following two PDFs:

Recommendation ITU-R BS.1770-4 (10/2015 - Algorithms to measure audio programme loudness and true-peak audio level)
EBU tech 3341 (LOUDNESS METERING: 'EBU MODE' metering to supplement EBU R 128 Loudness Normalization)



The TL;DR answer is:
  • Momentary Loudness is measured with a time frame of 400ms
  • Short Term Loudness is measured with a time frame of 3 seconds
  • Integrated Loudness is an average based upon the "smoothed" (and in case of ITU-R BS.1770-4 and EBU R-128, also gated) MLk values, picked up/re-calculated every 1 Hz (every 1 second) bar minimum, and is then "averaged" over the course whole length of the program stream (it is similar to Leq(m), which was most known to be used by the MPAA to declare loudness for Cinema content)
Slightly different rules apply for "live broadcasting" (as there really is no "definite end" to a stream). But for finished productions (songs, podcasts, re-cut radio shows, Youtube videos, etc), measuring the "program stream" once from start to finish once gives you your desired ILk readout.

I might turn this into a separate thread some time. But this is a good "simplified" starting point

Olli H
Backer
Posts: 130
Joined: Fri Dec 29, 2017 16:08 CET

Re: Songwriting Competition - General Gossip Thread

#148

Post by Olli H »

CALCULATING THE LOUDNESS FOR COMPETITIONS

About measuring loudness to match competition requirements:
Integrated loudness is calculated from the whole song.

Just use this free online tool
https://youlean.co/file-loudness-meter/

It calculates the integrated loudness value from the whole file. Simple, quick and easy.


HACKING THE LOUDNESS
A Future in Noise wrote:
Sun Nov 15, 2020 14:09 CET
If a program calculates the mean value for the whole track, that would mean that I can be extremely loud at the end, if the beginning of the track is very quiet, but I can't imagine that this is how it works. I have been trying to find information about this on the Internet, but I haven't found any. So if you know about the time aspect, please let me know.
Here’s Youlean article about hacking the LUFS calculation with quiet verses and loud choruses.

https://youlean.co/how-to-hack-lufs-normalization/

I haven't tried it but the end result seems to be: with that kind of trick you can manage to make your chorus couple dbs louder than others, but you may have to turn down the quiet parts too much. End result would be musically poor.

A Future in Noise
Posts: 80
Joined: Wed Nov 07, 2018 09:26 CET

Re: Songwriting Competition - General Gossip Thread

#149

Post by A Future in Noise »

Olli H wrote:
Wed Nov 18, 2020 13:25 CET
Here’s Youlean article about hacking the LUFS calculation with quiet verses and loud choruses.
:tu:

A Future in Noise
Posts: 80
Joined: Wed Nov 07, 2018 09:26 CET

Re: Songwriting Competition - General Gossip Thread

#150

Post by A Future in Noise »

I noticed something very strange (IMO). When I imported my rendered file (LUFS -14.3, true peak -1.3) into a new project, and tried to raise/increase that track's volume, it didn't affect the LUFS value. I had the Youlean loudness meter as FX in the master track / master fader, but I didn't touch that fader, but the fader of the track where the rendered file is. First I increased the volume just a little, and then I made it 1 dB higher. Youlean showed exactly the same values as before.

How can one understand this?

Post Reply