Hi,
I found this forum earlier this year and having gone through a few personal hurdles finally got a submission in for the first time. A big thanks to @Mister Fox for your work and dedication to detail. I’ve never been involved in this sort of thing before. It’s very educational and challenging and it’s all appreciated greatly. Listening to different people’s mix interpretations has really been eye opening for me. I have been learning a lot and just getting through and being cleared for evaluation is a nice step in my books. I wish everyone here the very best and I thank @Gaz for this wonderful opportunity to play with a really interesting track.
2024-NOV-01 Info: Thank you everyone, for making MC100 a resounding success. Please show Songwriting Competition 087 the same love.
MIX CHALLENGE - MC093 August 2023 - Winners announced
Re: MIX CHALLENGE - MC093 August 2023 - Mix Round 1 in evaluation
Hi All,
I noticed in the MC093_Statistics.pdf that I've been given a warning for my TP value being 0.99.
That's interesting because I used the new Export Mixdown function in Studio One which allows you to set final file parameters. These were set to -16LUFS/-1.0TP. I also checked this after the export using Youlean LM and it reported a TP value of -1.0. I just checked it again and it's definitely saying -1.0TP. There must be some slight variation in how the TP values are being calculated (possibly due to samplerate or the type of interpolation).
I realise this isn't a big deal but it's interesting nonetheless.
Cheers!
I noticed in the MC093_Statistics.pdf that I've been given a warning for my TP value being 0.99.
That's interesting because I used the new Export Mixdown function in Studio One which allows you to set final file parameters. These were set to -16LUFS/-1.0TP. I also checked this after the export using Youlean LM and it reported a TP value of -1.0. I just checked it again and it's definitely saying -1.0TP. There must be some slight variation in how the TP values are being calculated (possibly due to samplerate or the type of interpolation).
I realise this isn't a big deal but it's interesting nonetheless.
Cheers!
Re: MIX CHALLENGE - MC093 August 2023 - Mix Round 1 in evaluation
Also, I think that @Mr Fox is having to do too much, chasing up people with messages, checking reuploads, etc.
I'm happy with the rules being applied strictly, and if that means I'm disqualified for something in the future then "so be it"!
I'm happy with the rules being applied strictly, and if that means I'm disqualified for something in the future then "so be it"!
Re: MIX CHALLENGE - MC093 August 2023 - Mix Round 1 in evaluation
Good luck everybody!
-
- Posts: 36
- Joined: Tue Feb 07, 2023 10:40 CET
- Location: Bulgaria
- Contact:
Re: MIX CHALLENGE - MC093 August 2023 - Mix Round 1 in evaluation
Hi all i am sort of new here but i start for real to like the strict rules. After all u wil learn something from your mistakes. For example this time my true peak is under -6 db and the result is that my mix is 23 bit. Next time i will watch for this too. Just wanna ask something related then it turns out that if u are using some analog outboard gear and u gain stage at -18 u are leaving he daw at 23 bit too or i am wrong ? And one more thing come to my mind there are some plugins that can fill blanket bits -ozone for example is it safe to use them just to fix this issue ?
Re: MIX CHALLENGE - MC093 August 2023 - Mix Round 1 in evaluation
I always admired people like @Mister Fox, who voluntary took on the burden like this! And yes, we are lucky to have him running this site.scottfitz wrote: ↑Wed Aug 30, 2023 01:44 CEST
Hi,
it's tricky to run a competition which has new people arrive all the time. The easiest situation is that entrants would simply read through the rules and abide by them. We are lucky to have Mister Fox run this at all and if he suddenly couldn't do it anymore then I don't hold out much hope for someone anywhere near as good taking over (unless they were paid and then the whole nature of the place changes and probably for the worse).
That said, it's great to see new people and please stick around, it's an incredible learning experience. The rules are a little mystifying at first, but in time you will understand or just be enjoying the challenge too much to worry about it.
Cheers all
Again, if there are Rule set for, say, file name template (even for the font size!) then there is other than 'just because' reason for that! It may be not so obvious for the participants, but once you decided to play the game, you must learn the rules of the game first. Sadly, I did not So for me there's no point to complain.
Really hope that one day I'll pass Guidelines and Rules exam and submit my entry successfully.
-
- Posts: 26
- Joined: Wed Aug 16, 2023 18:57 CEST
Re: MIX CHALLENGE - MC093 August 2023 - Mix Round 1 in evaluation
23 bits is a thousand times better than 25 bits.kombainera wrote: ↑Wed Aug 30, 2023 07:19 CESTHi all i am sort of new here but i start for real to like the strict rules. After all u wil learn something from your mistakes. For example this time my true peak is under -6 db and the result is that my mix is 23 bit. Next time i will watch for this too. Just wanna ask something related then it turns out that if u are using some analog outboard gear and u gain stage at -18 u are leaving he daw at 23 bit too or i am wrong ? And one more thing come to my mind there are some plugins that can fill blanket bits -ozone for example is it safe to use them just to fix this issue ?
I would absolutely avoid any plugins like that or leave it up to a mastering engineer.
100%.Necaster wrote:other than 'just because' reason
To be able to use scripts or batch processing without having to manually go through every file name, for example.
There are better things to waste your time on, I think we can all agree on that.
Discipline is an essential trait to have/develop in this field, too.
Let's just be better than graphics designers okay?
Last edited by trackerjack on Wed Aug 30, 2023 11:17 CEST, edited 4 times in total.
"I love deadlines. I like the whooshing sound they make as they fly by." Douglas Adams
- Mister Fox
- Site Admin
- Posts: 3357
- Joined: Fri Mar 31, 2017 16:15 CEST
- Location: Berlin, Germany
MIX CHALLENGE - MC093 August 2023 - Mix Round 1 in evaluation
I need to step in for a (longer) moment to clear some confusion.
However, first and foremost, I kindly ask everyone to please brush up their knowledge on the topic of the "Statistic Sheet" and the "Wild Card Mechanic" again. I posted about this multiple times before. If you overlooked that, you can find the thread here:
Mix Challenge - Addendum: Statistic Sheet and Wild Card Mechanic
The TL;DR is basically:
All this has been explained in minute detail in the official Rules and Guidelines, the Addendum: Statistic Sheet and Wild Card Mechanic, and has been attached as sub 300 word long "TL;DR Rules.txt" in the Mix Pack. By submitting an entry to the Mix Challenge audio community, you agreed to adhere to the given Rules and Guidelines for the Mix(ing) Challenge.
Topic: "Mr Fox is having to do too much, chasing up people with messages, checking reuploads, etc.":
This could be prevented, if people would at least read the TL;DR Rules.txt -- this is why I bundled them as of April 2023. There were too many excuses of "too many rules -- this is not fun at all", or "I didn't know that this is a thing", or similar. With adding this TXT as part of a Mix Pack, there are no more grounds to "not have known that" - despite there also being clear and visual indicators where to find the Rule Books. (see first two posts of each thread, and every reminder)
Yes, it is more extra work for me to sometimes triple check entries. However, I am indeed doing this for the learning aspect of the community (in fact, I've been doing this in the early days of the Mix Challenge), and since things went a bit wild in terms of "interpretation of rules" throughout the years, and their corresponding outbursts through PM and email after the fact.
I want to provide fair playground for everyone. If there are no consequences for your mishaps, how can you learn from this experience and then port it for "the real world"? This is something that training facilities do not teach, and it can ultimately cost you your job. This is also why the "Wild Card Mechanic" was introduced - to not just be mercilessly punished for a (often quite literal) careless mistake, but learn from your mishap, then do better. In case of the Songwriting Competition - and since this particular game is way more encouraging to "keep working" - you will not be disqualified (unless you don't participate in the final feedback/voting process), but you will loose out on valuable bonus points.
Which brings me to the following...
Topic: File Name:
It is part of the game, it also shows if you have paid attention to detail, e.g. reading the TL;DR Rules.txt at bar minimum -- It is an uncomplicated copy/paste/edit job. It really can't get any more simple.
Having to bring this up multiple times every game, for years on end now, is exhausting. This is why it is a criteria for being "tagged disqualified". Unless you straight up refuse to use the filename template and not even add your forum username - then it will result in "tagged OUT". This setup will stay like that until at least MC095 / December 2023 (if we have access to new material) - after that, I will re-evaluate the situation and either keep the current mechanic, or turn this into a "tagged OUT" scenario.
The reason why I insist on a correct filename, is not only username association. It is also to speed up the Statistic Sheet creation process. I've explained this multiple times since 2021 -- here is an example from the stricter Rule Book enforcement (October 2022 Update) news blog post. I will not reach out to participants during MC094 and MC095 - this month (August) was an exception.
Topic: Song Length:
This is currently not a "disqualification criteria".
You are merely informed that you might have cut "too short" and therefore discarded maybe content which is important to the narrative of the song. "Cutting too short" can also result in unwanted transient smearing (tracks that don't have crescendo and decrescendos), and leaves no room for edits for a follow-up mastering process.
Topic: Bitrate "dropping off":
This is a mere information. The file itself is still within specs (24 bit).
The drop-off can happen due to you not filling up the whole dynamic range of the medium. In most cases, that is completely fine and will be a recurring topic again due to more and more platforms adapting lower loudness limits. For example: every platform distributing "Dolby Atmos for Music" have a maximum limit of -18,0 LUFS ILk and at absolute most about -1,00 dBTP. Or radio stations distributing at -24 LUFS ILk with a maximum ceiling of -2,0 dBTP.
If you don't have "super strong transients", you won't fill up all 24 bits for the whole part of the song. Again, this is fine, nothing to worry about. However, if your maximum signal strength usually hovers between -5,0 dBTP to -3,0 dBTP and it suddenly spikes to -0,5 dBTP, then you should definitely take a closer look for possible audio glitches.
If you exceed the bitrate on analysis (e.g. 25 bit on a 24 bit file), your signal is "overshooting" / clipping.
Topic: Loudness Specs and warnings:
Those that got a "warning" on the statistic sheet on the topic of loudness, and wonder why you can't reproduce the same readout -- not every analysis tool works the same.
In fact, everything ITU-R BS.1770-x / EBU R128 based has a certain allowed tolerance. This is mentioned in the white papers of each specification. In case of the loudness analysis, it is +-0,1 LU at this point. In case of maximum signal strength in dBTP, it depends on the implementation - although ideally, the tolerance should be anything +-0,02 dBTP or even lower. Some tools oversample 4x, others 8x, even others only by 2x. Additionally, some tools "sum" the maximum signal strength (dBTP), others show individual channels and also give you a "max readout" for the highest peaking channel on the numeric "max readout" section. This can result in various offsets.
The measurements for the statistic sheet are based on EBU R128 (or ITU-R BS.1770-4 as variant) - I am using Wavelab for Batch Analysis at the moment, which is quite picky compared to (let's say) Youlean Loudness Meter. This engine also doesn't sum the maximum signal strength, but lists them "per channel". As mentioned above, all these tools have different tolerances - even with the same used spec.
The Mix(ing) Challenge and Songwriting Competition offers higher tolerances than the known white papers, to compensate for each user having access to the most different tools. To be on the safe side, always aim to "target lower" than the allowed maximum. Additionally, if you're using safety limiters, not every limiter is "True Peak safe" (see this article by Saintpid Mastering from March 2016 - although that article hasn't been updated in years). In this case, it is recommended to always set it up -0,1dBTP and/or -0,1 LU lower than your final target. (I also mention this for the Songwriting Competition).
I hope this covers everything.
Any further commentary on the topic "Rules"... please get reacquainted with them first.
Should more questions arise, then please post them in either General Gossip for the Mix(ing) Challenge, or in the #gossip-questions-answers section on Discord. I will move / delete posts - especially harsh criticism on the rule set and angry posts.
I now (re-)open the field for everyone to give each other feedback (highly encouraged). This will be independent to the client feedback however.
Please watch this spot for the client feedback and Mix Round 2 participant announcement, or any other future announcement.
We (as in: the Song Provider and I) will keep you updated, and of course send out appropriate reminder newsletters!
However, first and foremost, I kindly ask everyone to please brush up their knowledge on the topic of the "Statistic Sheet" and the "Wild Card Mechanic" again. I posted about this multiple times before. If you overlooked that, you can find the thread here:
Mix Challenge - Addendum: Statistic Sheet and Wild Card Mechanic
The TL;DR is basically:
- only the Song Providers can decide who goes into Mix Round 2 (his/her/their Top 15 entries) - they offer you this possibility in a follow-up post
- you can not enforce going into Mix Round 2
- everyone that is within (technical) specs have the possibility to advance into Mix Round 2 (users that have no "tag" on the "disqualified" column on the chart)
- those that have been "tagged disqualified" (a bold "X" on the chart) can still get a chance to advance into Mix Round 2, at the cost of a Wild Card (from which each community member has two available) - should the Song Provider select your mix
- those that have been "tagged OUT" (a bold "OUT" on a red background) are fully out of the game
All this has been explained in minute detail in the official Rules and Guidelines, the Addendum: Statistic Sheet and Wild Card Mechanic, and has been attached as sub 300 word long "TL;DR Rules.txt" in the Mix Pack. By submitting an entry to the Mix Challenge audio community, you agreed to adhere to the given Rules and Guidelines for the Mix(ing) Challenge.
Topic: "Mr Fox is having to do too much, chasing up people with messages, checking reuploads, etc.":
This could be prevented, if people would at least read the TL;DR Rules.txt -- this is why I bundled them as of April 2023. There were too many excuses of "too many rules -- this is not fun at all", or "I didn't know that this is a thing", or similar. With adding this TXT as part of a Mix Pack, there are no more grounds to "not have known that" - despite there also being clear and visual indicators where to find the Rule Books. (see first two posts of each thread, and every reminder)
Yes, it is more extra work for me to sometimes triple check entries. However, I am indeed doing this for the learning aspect of the community (in fact, I've been doing this in the early days of the Mix Challenge), and since things went a bit wild in terms of "interpretation of rules" throughout the years, and their corresponding outbursts through PM and email after the fact.
I want to provide fair playground for everyone. If there are no consequences for your mishaps, how can you learn from this experience and then port it for "the real world"? This is something that training facilities do not teach, and it can ultimately cost you your job. This is also why the "Wild Card Mechanic" was introduced - to not just be mercilessly punished for a (often quite literal) careless mistake, but learn from your mishap, then do better. In case of the Songwriting Competition - and since this particular game is way more encouraging to "keep working" - you will not be disqualified (unless you don't participate in the final feedback/voting process), but you will loose out on valuable bonus points.
Which brings me to the following...
Topic: File Name:
It is part of the game, it also shows if you have paid attention to detail, e.g. reading the TL;DR Rules.txt at bar minimum -- It is an uncomplicated copy/paste/edit job. It really can't get any more simple.
Having to bring this up multiple times every game, for years on end now, is exhausting. This is why it is a criteria for being "tagged disqualified". Unless you straight up refuse to use the filename template and not even add your forum username - then it will result in "tagged OUT". This setup will stay like that until at least MC095 / December 2023 (if we have access to new material) - after that, I will re-evaluate the situation and either keep the current mechanic, or turn this into a "tagged OUT" scenario.
The reason why I insist on a correct filename, is not only username association. It is also to speed up the Statistic Sheet creation process. I've explained this multiple times since 2021 -- here is an example from the stricter Rule Book enforcement (October 2022 Update) news blog post. I will not reach out to participants during MC094 and MC095 - this month (August) was an exception.
Topic: Song Length:
This is currently not a "disqualification criteria".
You are merely informed that you might have cut "too short" and therefore discarded maybe content which is important to the narrative of the song. "Cutting too short" can also result in unwanted transient smearing (tracks that don't have crescendo and decrescendos), and leaves no room for edits for a follow-up mastering process.
Topic: Bitrate "dropping off":
This is a mere information. The file itself is still within specs (24 bit).
The drop-off can happen due to you not filling up the whole dynamic range of the medium. In most cases, that is completely fine and will be a recurring topic again due to more and more platforms adapting lower loudness limits. For example: every platform distributing "Dolby Atmos for Music" have a maximum limit of -18,0 LUFS ILk and at absolute most about -1,00 dBTP. Or radio stations distributing at -24 LUFS ILk with a maximum ceiling of -2,0 dBTP.
If you don't have "super strong transients", you won't fill up all 24 bits for the whole part of the song. Again, this is fine, nothing to worry about. However, if your maximum signal strength usually hovers between -5,0 dBTP to -3,0 dBTP and it suddenly spikes to -0,5 dBTP, then you should definitely take a closer look for possible audio glitches.
If you exceed the bitrate on analysis (e.g. 25 bit on a 24 bit file), your signal is "overshooting" / clipping.
Topic: Loudness Specs and warnings:
Those that got a "warning" on the statistic sheet on the topic of loudness, and wonder why you can't reproduce the same readout -- not every analysis tool works the same.
In fact, everything ITU-R BS.1770-x / EBU R128 based has a certain allowed tolerance. This is mentioned in the white papers of each specification. In case of the loudness analysis, it is +-0,1 LU at this point. In case of maximum signal strength in dBTP, it depends on the implementation - although ideally, the tolerance should be anything +-0,02 dBTP or even lower. Some tools oversample 4x, others 8x, even others only by 2x. Additionally, some tools "sum" the maximum signal strength (dBTP), others show individual channels and also give you a "max readout" for the highest peaking channel on the numeric "max readout" section. This can result in various offsets.
The measurements for the statistic sheet are based on EBU R128 (or ITU-R BS.1770-4 as variant) - I am using Wavelab for Batch Analysis at the moment, which is quite picky compared to (let's say) Youlean Loudness Meter. This engine also doesn't sum the maximum signal strength, but lists them "per channel". As mentioned above, all these tools have different tolerances - even with the same used spec.
The Mix(ing) Challenge and Songwriting Competition offers higher tolerances than the known white papers, to compensate for each user having access to the most different tools. To be on the safe side, always aim to "target lower" than the allowed maximum. Additionally, if you're using safety limiters, not every limiter is "True Peak safe" (see this article by Saintpid Mastering from March 2016 - although that article hasn't been updated in years). In this case, it is recommended to always set it up -0,1dBTP and/or -0,1 LU lower than your final target. (I also mention this for the Songwriting Competition).
I hope this covers everything.
Any further commentary on the topic "Rules"... please get reacquainted with them first.
Should more questions arise, then please post them in either General Gossip for the Mix(ing) Challenge, or in the #gossip-questions-answers section on Discord. I will move / delete posts - especially harsh criticism on the rule set and angry posts.
I now (re-)open the field for everyone to give each other feedback (highly encouraged). This will be independent to the client feedback however.
Please watch this spot for the client feedback and Mix Round 2 participant announcement, or any other future announcement.
We (as in: the Song Provider and I) will keep you updated, and of course send out appropriate reminder newsletters!
- Mister Fox
- Site Admin
- Posts: 3357
- Joined: Fri Mar 31, 2017 16:15 CEST
- Location: Berlin, Germany
MIX CHALLENGE - MC093 August 2023 - Mix Round 1 in evaluation
I am more than overjoyed to announce the following license sponsor for the Mix Challenge audio community
Audified is kind enough to give away one license of MixChecker Pro to the winner
License will turn into NFR
More info on Audified: https://audified.com
Their offer is retroactive (starting with MC093 / August 2023).
A huge thank you goes to Audified for providing the licenses.
Audified is kind enough to give away one license of MixChecker Pro to the winner
License will turn into NFR
More info on Audified: https://audified.com
Their offer is retroactive (starting with MC093 / August 2023).
A huge thank you goes to Audified for providing the licenses.
Re: MIX CHALLENGE - MC093 August 2023 - Mix Round 1 in evaluation
What's meaning an X in the overview sheet? Am I disqualified?