• Hey Guest! Ever feel like entering a Game Jam, but the time limit is always too much pressure? We get it... You lead a hectic life and dedicating 3 whole days to make a game just doesn't work for you! So, why not enter the GMC SLOW JAM? Take your time! Kick back and make your game over 4 months! Interested? Then just click here!

GMC Jam Discussion GMC Jam 40 - 10 YEAR ANNIVERSARY SPECIAL 🎉

Status
Not open for further replies.

Mr Magnus

Viking King
I'm not very sold on migrating the Jam to Itch.io when this is a GMC jam. Especially since we're at best expecting a surge in entries for this specific jam that's less than a month away. Even in the long ago when we had 50+ entries we managed fine enough managing that. At any rate I'm not even sure we have a "must be solved now" problem beyond large .zip downloads.
 

The M

Member
@Alice That's a very fair point. It would definitely be easier to tamper with the files (even accidentally) if you're in charge of them yourself.
 

Yal

🐧 *penguin noises*
GMC Elder
I'm not very sold on migrating the Jam to Itch.io when this is a GMC jam. Especially since we're at best expecting a surge in entries for this specific jam that's less than a month away. Even in the long ago when we had 50+ entries we managed fine enough managing that. At any rate I'm not even sure we have a "must be solved now" problem beyond large .zip downloads.
Judging from historical data, a lot of the ZIP download size seems to come from GM defaulting to the "silently convert this OGG/MP3 file to a WAV for maximum quality, file size be damned" option for audio... and it bloats the file size to ~10x of what it could've been if someone misses that. Sure, it's not on a nuclear meltdown level of urgency, but the chance of the ZIP being really big is pretty monumental according to the existing data.
 

FrostyCat

Redemption Seeker
Why don't we consider using a torrent for the Jam zip? That way we can seed the download to each other, and the checksum feature helps prevent tampering and accidental loss during download. It's also a standard way to publicly share files of the size range we've been seeing over the past few events.
 

HayManMarc

Member
Perhaps there should be a rule in the GMC JAM participation rules that declares audio files MUST be in this less-bloaty format? (Or risk forfeiture of your entry.) Just a thought.
 

FrostyCat

Redemption Seeker
Perhaps there should be a rule in the GMC JAM participation rules that declares audio files MUST be in this less-bloaty format? (Or risk forfeiture of your entry.) Just a thought.
There shouldn't, it's completely unenforceable. You can't tell from the size alone whether it is that big from having the audio files uncompressed, or just from having many audio files that are already compressed. A better and much more enforceable alternative is a simple hard size limit (e.g. 20MiB).

On a separate note, I support the suggestion from Cloaked Games about piloting itch.io's game jam hosting feature. This will save time otherwise spent building and maintaining our own infrastructure, which at its current cobbled-together state leaves a lot to be desired. If any Jam administrator is open to it, we can do a trial run this weekend (February 6-7) where we simulate a jam by uploading past entries, then emulate a voting procedure as before and see how it turns out.
 

dadio

Potato Overlord
GMC Elder
@Cloaked Games
I think the main concern with this would be (1) now GMC users would be required to have an itchio account to participate. (2) Uh... Let me know if I missed anything?

Yeah, thing is, this is the GMC Jam. It's not an itchio Jam.
It has a long (10 year) history of taking place/being hosted/all feedback taking place on a standard forum.
With a post/post/post format.
Itchio is a different beast - with a different audience.

These forums make for a more personal, smaller, friendly event with piles of back and forth casual chatting and feedback. It's meant to be fun.
Itchio makes for a more detached, larger, less friendly event with little back and forth feedback. More like a serious competition.

This is a pretty tightly knit community - most of us know each other at this stage....
and I think most of us value the back and forth chit chat, the hype build up etc. that takes place here.
Anything that shifts focus away from these forums kinda works against the reason these Jams were created in the first place (a fun event to encourage creativity and community interaction and feedback.)

Just my 2 cents.
Hope that made some kinda sense.

🍋

(Having said all that - I *do* think it's a good idea for folks to upload their games to sites like itchio, just so there's a greater likelihood of all the games still being available in 10 years time. Alice having a big zip for download works well for this Jam, but tragically many games have been lost along the way - I'm currently still trying to cobble together games from Jams 1-4 in particular, that are missing a huge chunk... so yeah, I'm all for the usual standard big zip download (+ optional uploading of games to itchio for a more reliable, longterm backup.)

Also @FrostyCat : Yeah, having a Torrent available as well as the usual direct download would be good. I brought that up a few Jams back.
 

FrostyCat

Redemption Seeker
Yeah, thing is, this is the GMC Jam. It's not an itchio Jam.
It has a long (10 year) history of taking place/being hosted/all feedback taking place on a standard forum.
With a post/post/post format.
Itchio is a different beast - with a different audience.

These forums make for a more personal, smaller, friendly event with piles of back and forth casual chatting and feedback. It's meant to be fun.
Itchio makes for a more detached, larger, less friendly event with little back and forth feedback. More like a serious competition.

This is a pretty tightly knit community - most of us know each other at this stage....
and I think most of us value the back and forth chit chat, the hype build up etc. that takes place here.
Anything that shifts focus away from these forums kinda works against the reason these Jams were created in the first place (a fun event to encourage creativity and community interaction and feedback.)
The plan is to outsource only the hosting and voting infrastructure to itch.io, the rest of the conversation can still be here.

If there's anything that isn't fun, that would be the manual setup and maintenance of the Jam zip, plus the manual tabulation of the votes that come after.
 

dadio

Potato Overlord
GMC Elder
Lol!
Hmmm... I'd be interested to get more opinions on this proposal.
A poll (of entrants only) would be good.

(I think there may be some that quite like manually formatting their review/vote posts the way they like, having a single handy massive zip for spitfiring through entries etc.)
 
Tons of other communities host their game jams on itchio. It's just a tool to make the jam process less laborious to set up. I'm not suggesting we host the feedback or conversation over there! I really don't think the location the files are uploaded being different makes this any less of the GM Community jam than before. Considering, as I mentioned, the itchio jam can be made to be a private link and all the voting/feedback can still happen on the forum.
 

pixeltroid

Member
These forums make for a more personal, smaller, friendly event with piles of back and forth casual chatting and feedback. It's meant to be fun.
Itchio makes for a more detached, larger, less friendly event with little back and forth feedback. More like a serious competition.
Yeah. The GMC jam should take place here on the GMC forum.

Itch is a good site, but it lacks the "warmth" of this community.
 

HayManMarc

Member
Yep, I'm with the " keep the jam on the GMC" crowd. Itchio is too far removed. If this were a community challenge or something else other than the main jam, then I wouldn't have a huge problem with it. But I agree with dadio's and pixeltroid's take on it.

This is OUR jam, and it should remain local to the GMC. (Even tho I hosted it one time during a forum snafu.)
 
Why don't we consider using a torrent for the Jam zip? That way we can seed the download to each other, and the checksum feature helps prevent tampering and accidental loss during download. It's also a standard way to publicly share files of the size range we've been seeing over the past few events.
Yeah, I suggested this on the last page. Like, I said there too, if the size of all the games is large again, you can easily selectively download the games you want to try in your torrent client.

Anyway, the first jam I joined had 80 bloody entries, and no one seemed concerned at the size then. It did have a torrent too which was great.
 

Micah_DS

Member
Personally, I'm hoping we can wrap this conversation up soon and just focus on jam hype and theme guesses again. As for me, I'll just give my last 2 cents before getting back on the hype train:
  • Yes to Torrent + the usual Mega link
    I honestly don't think the turnout will be THAT huge. A single zip should still suffice, especially if we do something to help awareness on audio optimization, like occasional reminders/tips throughout the jam or something.
  • No to itch.io
    Though I'm not totally against hosting on itch.io, it's the 10th anniversary, so it just feels... wrong? Plus, it seems to me like the history will be better preserved the more we keep things on here. I tend to agree with the reasons others have given as well.
With those two points in mind, I may be overlooking something, but I believe the Jam Player needs no changes.

Judging from historical data, a lot of the ZIP download size seems to come from GM defaulting to the "silently convert this OGG/MP3 file to a WAV for maximum quality, file size be damned" option for audio... and it bloats the file size to ~10x of what it could've been if someone misses that. Sure, it's not on a nuclear meltdown level of urgency, but the chance of the ZIP being really big is pretty monumental according to the existing data.
THIS. Which leads me to...

~ MICAH'S AUDIO OPTIMIZATION GUIDE ~

This is where I generally land with my audio settings, as of GMS2.3:


(Except you can totally ignore the bit rate for sounds - I'll explain why below)

Alright! Time to go DEEEEEP into why I choose the above settings, but don't worry, I'll make the core takeaways in bold and in slightly larger text, to act as a sort of TL;DR: :p
  • SOUND EFFECTS
    \
    • Uncompressed - Not Streamed
      Sound effects generally need to start playing as soon as your code tells them to, because they often play a very important feedback role for the player, and the sounds need to play right along with various actions or it feels bad, like it's disconnected or laggy in some way.
      Since uncompressed size is obviously the biggest, it's important to make sounds short when you can.
      Remember, if you have any silence in the beginning of a sound, that's bad for both saving space and for responsive playback. I recommend editing your sounds with something like Audacity, to ensure you cut off any needless silence on the sound's ends and also to shorten the sounds if you can.
    • Output: Mono
      If your sound doesn't absolutely need the stereo information, you can cut a stereo sound's size IN HALF just by choosing mono. :oops: That's some major space saving there.
      Often times, converting a stereo sound to mono will be an acceptable sacrifice, especially if you make use of the audio_play_sound_at function to play sound at a position relative to your listener position. In that way, your sounds won't all sound like they're coming from the center, but actually from around you (assuming your audio system isn't mono..... if it is... whyyy?).
      So I HIGHLY recommend you get familiar with positional sounds if you aren't already, because it's totally worth it.
    • Sample Rate: 44100 (Hz)
      If you grab free audio and don't know its sample rate, I think you're safe to assume it's 44100 Hz (a.k.a. 44.1 kHz). What a sample rate of 44.1 kHz essentially means is that it's fast/dense enough to store frequency information just above 20 kHz, which you probably already know to be around the human hearing peak. Lowering the sample rate (a.k.a. downsampling) is almost always a bad idea, as it will remove some or all of the high "clarity" and "definition" frequencies, making it sound 'muffled' and possibly even 'muddy'. And the muddy sound is partly due to another issue related to downsampling called 'aliasing', which I will avoid diving into, but the short description of aliasing is that it's a digital kind of distortion, which can sound pretty nasty and harsh since it's not a harmonic distortion.
      Look into "Nyquist (foldback) frequency" if you want to know more about what aliasing is and why it occurs. To be perfectly transparent, I'm still learning about it and I feel shaky on the details myself.
    • Quality: 16 bit
      Even if you think you have an 8 bit sound, because it was labeled as such, it doesn't mean it is. Lots of people call sounds 8 bit when they aren't true 8 bit. I honestly wouldn't be surprised if some 8 bit sound generators saved in 16 bit.
      When in doubt, please keep it 16 bit, because when you convert a 16 bit sound to 8 bit, you risk introducing some very harsh and bright artifacts that can fatigue or even really hurt people's ears, among other bad things.
      Okay, so "quality" is technically "bit depth" (not to be confused with 'bit rate'). Bit depth is the amplitude's (volume) resolution.
      8 bit is far more 'steppy' than 16 bit, because 8 bit covers the same vertical distance as 16 bit, yet with less 'blocks', so each block is taller with 8 bit, thus the wave becomes more steppy.

      So what does 8 bit being more steppy mean? Let's look at a sine wave vs a square wave to find out:
      The sine wave sounds soft and dull, while a square wave at the same amplitude sounds bright, perhaps even harsh, not to mention it sounds louder than the sine wave.
      What's happening is that the sine wave reaches the same peak as the square wave, but the sine wave comes off less bright/harsh because it reaches the peak more gradually, and the sine wave spends less time at its peak volume, so it also sounds quieter than the square wave. Well, the square wave being brighter also makes it sound louder, since the ear is more sensitive in the high range.

      So there's quite a lot to think of here, but the main takeaway is that, generally, the quicker a waveform's amplitude goes from zero db to peak volume, the brighter and harsher it will sound, because that translates to a physical speaker's cone moving faster, creating a more abrupt push to the air particles.

      Unfortunately, this is about as far as my own understanding goes. It's a bit more complicated, because apparently the noise floor for 16 bit is -96 dB, while the noise floor for 8 bit is -48 dB, so that affects things a lot as well, and then you throw in the aliasing (yes, more aliasing, yay!) that happens when converting from 16 to 8 bits, etc. And that aliasing is one of the nastiest things, perhaps even creating a wretched sparse distorted static shredding/clicking of sorts. It does not make the ears happy, I'll tell you that.

      If that was too hard to follow, I apologize. The short of it is that your sounds can REALLY get screwed up when you convert from 16 to 8 bit.
    • Bit Rate: N/A
      From my tests that I did just now (details in spoiler below), I'm convinced setting the bit rate for uncompressed (WAV) format does nothing at all. I always wondered this about WAV format, but now I'm pretty sure WAV format decides bit rate based on sample rate and number of audio streams. I'm not 100% sure on this though. All I know is that you can ignore bit rate for uncompressed audio - it won't change a thing.
      TEST 1:
      I listened for audible differences by taking a full spectrum white noise sample, duplicating it, then giving each sample different settings. When running the test program, I made it so I could switch between the sounds to listen for differences. It didn't sound like bit rate made a difference at all, but sample rate and 'quality' (bit depth) did.
      TEST 2:
      I needed to be sure the bit rate setting wasn't changing anything, so the obvious test was to see if it'd make any file size difference. So I exported the first project, then I simply duplicated the first project and changed each sound to 8 kbps and re-compiled. Comparing both, the sizes were totally identical.
      -
      If anyone wants, here is my first test project with the audio debugger on and compiled to an executable, if you want to hear the differences yourself:
      Direct Download from my DropBox
      VOLUME WARNING - The white noise is quite loud, sorry. Probably best to turn the volume down in advance.
  • MUSIC
    \
    • Compressed - Streamed
      I didn't always recommend this, but at some point the pause between audio looping was fixed (thank you YYG!), so now even the 'Compressed - Streamed' option loops with zero flaws (admittedly, I only tested this on a SSD). Before I always recommended 'Uncompress on Load - Not Streamed'. It really shouldn't make much of a difference for a game with just a few music tracks though. If it were a game with tons of music, I'd think streamed would be best, of course, since you don't want to keep a ton of music in memory for no reason, and loading a lot of music on startup isn't the best either.
    • Output: Stereo
      Even if your game's music has a super old school sound, it's still likely to be stereo. For layman: Mono will sound like everything is coming from pretty much one place in the center, but things can still feel closer or further from you. With stereo, it will feel wider. Or, if you have some kind of L/R audio meter in your soundcard/whatever software, you can tell a sound is mono by playing it while watching if both L and R channels are always the same level. If they differ from each other at all, the music is stereo.
      You can obviously convert stereo music to mono, but - especially with free music - there's a good chance it was mixed to sound best in stereo, and it's very possible a mixdown to mono will ruin the mix, which is due to something called "phase cancellation" or "phasing".
    • Sample Rate: 44100 (Hz)
      My reasons for setting music to this is the same as sounds, so you may refer to that info above.
    • Quality: 16 bit
      My reasons for setting music to this is the same as sounds, so you may refer to that info above.
    • Bit Rate: No more than 192 kbps
      Unlike uncompressed (WAV) audio, bit rate matters with compressed (OGG) audio, so you'll need to choose a good bit rate for your music, since your music WILL be compressed *intense stare* (Please remember to compress your music/narration/ambience/etc. - uncompressed music appears to be the #1 culprit of unnecessarily large file sizes in previous jams. - Thank you - :)).
      I've found 192 kbps to be great quality. When I tested on my studio monitors, I could not tell the difference in quality above that. Honestly - and this is coming from someone who loves music - from my past audio tests, I think 128 kbps sounded very good and nearly identical to 192 kbps. I don't know all the conversion/container mumbo-jumbo that's done for these OGG files, but they seem very efficient. I just want to make a point that 128 kbps can sound different between formats. For e.g., mp3 at 128 kbps does NOT sound good, in my experience.
      Admittedly, I usually go with 192 kbps when 128 kbps would be just fine, but it's only because I'm stubborn.
 
Last edited:

GameDevDan

Former Jam Host
Moderator
GMC Elder
OOOooooooKeeeeeyyyy Doooooookkkeeeeeey - It just ain't a GMC Jam without a meandering discussion about the logistics / practicalities is it? 😂

Dan's the man in charge of the Jam (currently) so I'll leave it to him to respond.
Personally, I'm hoping we can wrap this conversation up soon
YOUR WISH IS MY COMMAND 🧞‍♂️

This GMC Jam, as run by me (your benevolent dictator), will feature a ZIP and a randomiser which spits out links. End of story. If anyone here wants to create their own alternative they are very welcome to put in the added time & effort for their own amusement but I won't be sanctioning it.

My two cents on hosting the whole jam on itch.io: I love itchio and host all my entries there, but if the bulk of the jam takes place there it ceases to be the "GMC" Jam and just becomes another itch.io jam like any other. The GMC Jam may be messy and cobbled together - but it always has been and that's the way I like it. If the community chooses to go that way in future then that's fine - but GMC Jam 40 is being held here.
 

Micah_DS

Member
I just realized there are two letters revealed already. For some reason I thought the "I" was a "|", like some kind of divider.
Well, I was about to go with FAKE BIDETS, but now it's back to the drawing board. Sad.
 

Evanski

Raccoon Lord
Forum Staff
Moderator
3 weeks to go

? ? E ? I ? ? ? ? T ?
13 Results

Most likely
simi likely
50/50
unlikely
least likely

Arenicolite (This would be epic) An ancient wormhole in sand, preserved in the rocks.
Chelicerata (I hope not)
Chelicerate (I hope not)

Credibility
Flexibility
Pre-Isolate

Precipitate
Prefigurate
Preliterate
Specialists
Specificate

Specificity
Theriodonta
 
Status
Not open for further replies.
Top