nesdev.com

The judging is something we will definitely be revisiting again for the next competition. The sheer volume of entries this year made it more difficult to judge mainly for the time it takes to judge everything. It's great that we keep learning with every competition we've had. Next year should be fantastic!

There is no question in my mind that more time should be invested playing the entries before judging. Especially if you are playing a bunch of games in a row. It may not be intentional, but I'm sure that an equal amount of gameplay isn't being dedicated towards each entry. It's true that some entries don't have a ton of gameplay, while others can have such a learning curve that you don't have time to progress before you need to move on to the next game. Maybe reducing the barriers to play the games might help. Having the games playable and judgeable in a browser would probably be useful. You wouldn't need to download a rom, unzip, read a text file, etc before playing. You could have it all onscreen at once.

I realize that we are still relatively small compared to a lot of software development competitions, but if this years showing has told me anything, it is that we are growing. It has already gotten to the point where doing things manually probably isn't the best option. It has probably also gotten to the point where we should start thinking about having an independent website devoted to the competition. There has been a lot of confusion about who hosts, who owns the competition, etc. The reality is that while the idea started on this forum, it has never been a competition just for NESDev. There are other sites besides NESDev and Nintendoage that have people who program for the NES, and I think a central gathering point for information would be useful. We would obviously have to do some cost analysis and see if the competition can self fund this, or if this is indeed the right direction to head. I have big plans, but sometimes I feel like I need to step back and remember that it isn't MY contest, that it was started by the community. I only make the decisions I am forced to by no one else giving other suggestions


Feedback is very important. This is the main reason we encourage people to make build threads or have a dedicated group of friends to do testing. I think with some tweaking to the control of your game it would help it out a lot. That being said, everyone who submitted their games, PLEASE keep working to improve them. The more polished you can make your games the better. We can't exactly put a game with a bunch of bugs on a production cartridge.

@Tepples Would I be right to assume you will be taking the reigns with your custom mapper again this year?

@Moderators Can we please get a 2017 subforum started for brainstorming?

I remember there being JSNES and em-fceux. Which web-based NES emulator is still maintained? And what kind of latency and Gamepad API support can one expect?


I guess so. I'll probably have to split it into two ROMs for prerelease testing because the PowerPak can't handle more than 4 Mbit (512 KiB) of PRG ROM at once. An 8 Mbit configuration would be the second largest PRG ROM ever (after Action 52) in a commercial, non-Chinese, non-pirate NES game. Hopefully Paul can handle it.


I'll look into it.

If I wasn't sick, it would have had all that and more. Really, entering at all was a long shot; I was happy just to make it in the door.


Me too, that's why I didn't participate in the judging - I could only finish one of them. Heck, I couldn't beat the first level of most of them. I'm not as coordinated as I used to be.



Some comments on the games, since I don't really have enough to say to go spamming every thread:

I made it to the final bosses of Twin Dragons and Filthy Kitchen, though only once each and I failed to beat either of them. There were some minor issues, like sprite flickering, but I thought both were excellent games and deserved the placement they got.

I'm a bit disappointed Flappy Jack wasn't up there with them, because, surprisingly, it felt like the most polished entry of the lot - I'd have given it a perfect score. And I don't normally like this kind of game!

Nebs 'n Debs was fun, but the level design was too punishing and getting killed by the glitches really killed the mood. Bumped my head and died too many times to count... but I did actually beat this one! Or made it to the credits, at least. It's a bit unfinished as-is but has a lot of potential.

The three Mojon Twins games were technically spot-on, but I have a bee in my bonnet about games made by and for horny teenage boys and would have given them all the minimum possible score, because I'm asinine like that. Another reason I decided not to vote. Moving on, there were just a handful of enemy and platform placements that cost me a life every single time, and one game-breakingly wide pit in Lala. If there was a way across that thing, I couldn't find it, even with two platforms.

Sinking Feeling made me laugh. I didn't get the reference but it was funny anyway.

Karate Kick was very smooth, very intuitive, and I actually manage to do quite well at it. But I imagine diagonals wouldn't be so much fun on an actual controller and there's really not much to it.

The Paths of Bridewell really caught my attention. It obviously lacked polish, and was too hard and tedious for me to finish, but I think it could really go places. Keep at it!

Jupiter Scope 2 and CYO both looked, sounded and felt great, but the gameplay was a complete letdown. Maybe I've played too much Touhou.

Spacey McRacey was way, way, way too fast for me to handle. It was, however, hilarious.

I never figured out Mini Brix Battle. I found it completely unplayable; I can't bounce the balls back and there aren't enough balls to finish the levels without doing so. I know it's my nervous system at fault but I can't judge what I can't play.

My experience with Waddles the Duck went like this: QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK QUACK *ALT-F4*. The jumping sound effect drove me bananas. So did the jump physics, for that matter. The warping mechanic was fun and intuitive, but the rest of the game was torture.

Super Tilt Bro.: I'm still trying to wrap my head around the fact that this is a thing. I'm also still unable to play Super Smash Brothers, and managed to lose to the non-existant AI. (Yes, lose to a character doing absolutely nothing. That's how bad my motor control is nowdays.) But still, wow.

Rock Paper Scissors Lizard Sbock makes no sense. Maybe I need a sense of humor.

And finally, I have no idea how to rate something like the 240p test suite. You keep doing what you do, tepples.

Finally, real feedback!

It's probably not your fault... there are two major issues and a minor one that I wanted to fix but didn't have time:

- levels don't have enough space between the player and the lowest tile: this happened because I chose to do 16x16 and that doesn't leave much space to do stuff. I tried to work around it by creating "strategies" in each level so the player can release the ball and break as much as possible before reflecting it again against the paddle was needed, but that only goes so far. Redoing that would require drastic code changes, I realized it too late to fix it.

- heightmap-angle relationship was wonky - The fixed angles I introduced weren't enough to give the player good control of the ball trajectory, and how this was chosen was counterintuitive in relation to the shape of the paddle (umbrella). I feel like there's something wrong with what angles are chosen depending on the ball position. This is not a hard fix but it could have introduced hard to find bugs...

- paddle is too small - this could have been fixed more easily but I had no time to make a bigger paddle and/or powerups for enlarging it. That would have made the game more manageable.


Anyway I had a blast creating and playing it, and so did my sister (my beta tester ), the difficulty issue was noted but we didn't think it was "unplayable". It is certainly unfair sometimes though. It's not a well designed game, the rushed development really shows. Thanks again for the feedback.

Fun fact: I've designed the ball object to be able to move in 1 to 8 vector units per frame, but the game has them locked at speed 2 since it's too difficult at that speed. Also the time counter doesn't actually count real seconds, 99 seconds is not nearly enough for most of the levels.

_________________
This is a block of text that can be added to posts you make. There is a 255 character limit.

Heh, it was a little bit slower until close to the deadline, when one of my main testers said "it just needs to be faster!". Maybe I paid too much attention to that advice

_________________
My games: http://www.bitethechili.com

I thought it was good. Had a lot of fun with the wife

Thanks Rahsennor for the full reviews! I love it.



From experience:

JSNES is not really usable as it have no sound and strange glitches.

Em-fceux is awesome: all the features of fceux + asmjs speed. It also handles gamepads nicely and is easy to patch. I use it for the itch.io page of SuperTiltBro and it is awesome. Note that the demo page of em-fceux is not up-to-date with their git repo, try directly the master

Cycloa-js. This ones does the job and is easier to integrate than em-fceux as it is native javascript. It may be a bit slow at times especially when playing music, does only handle NROM if I remember and is unable to run in PAL video mode. I used it before finding em-fceux.

_________________
Super Tilt Bro.: https://sgadrat.itch.io/super-tilt-bro

I thought so too at first. But, actually, you have infinite platform powers. I'll explain. You probably tried... run and jump, and make a platform and jump immediately... and fall to your death. No. You run and jump and make a platform and WAIT until a millisecond before your platform disappears, then jump and make another platform closer to your destination.

However, I do agree with (rainwarrior ? I think) that platform making should be ONE BUTTON. There were so many times I thought I pressed correctly for platform making, but really she just waved her wand... and I fell to my death. Grr.

_________________
nesdoug.com -- blog/tutorial on programming for the NES

I just sent the cash prizes the top 5 entrants!



Yeah do you still have your kazzo programmer? I could put together a 32Mbit or 64Mbit mapper 28 cartridge for you for testing. Prob will end up producing on 1Mbit, but may as well send bigger version so it's potentially useful in the future.

Perhaps it's time for a separate cartridge thread to come up with a plan with this carts release. I get frequent requests from the public for purchase of vol1. It saddens me to have to constantly turn these homebrew gaming fans away and tell them to come back in a few years. If there's no chance for this year's cartridge to fit within 4Mbit, then we're effectively forced to jump to 3v flash. And once all the level shifters are on the board to support 3v parallel flash, the cost differential between 8-16Mbit or even 32Mbit is minor on the order of 12cents per doubling. So that presents an opportunity to have a long term release of vol1 titles that I think we need to consider.

_________________
If you're gonna play the Game Boy, you gotta learn to play it right. -Kenny Rogers

How much room is left on the cart?

If there's room to spare, I'd be happy if Ralph 4 could be included. Also some other games to consider from people who might not post in this thread: Prez, MashyMashy, BSides.

Congrats on the best nesdev competition ever.

There is a decent amount of room left on the cart. Feel free to offer any non-compo entries up for inclusion and we'll do our best to include what we can.

As discussed in A53 vol3 cart thread entries have until end of March to make any changes to their entry. If you find a bug past that date reach out to us and we may still be able to patch it in.

I'm proposing this schedule for future compos as well. Jan 31st concludes compo. Judging complete during Feb. Game improvements, and bug fixes to be submitted by Mar 31st for guaranteed inclusion on the cartridge. Artwork and printed materials to be finalized in April. Printing and cartridge assembly during May/June, cartidge release in June/July.

Now that the compo is effectively complete, it's probably time to start discussing any changes were looking to implement for next year's compo. Maybe we try to officially announce the next years compo by end of March? That would align with the transition from last min tweaking of last compo entry to starting on next year's entry.

Note to self: Also about time for another financial report!

_________________
If you're gonna play the Game Boy, you gotta learn to play it right. -Kenny Rogers

At this point, I'd recommend reserving the main category for games only, even if they do fit the technical requirement (64K discrete), as I can't think of criteria to judge both games and "toys" (non-games).

Makes sense, they can always go in the free-for-all category if people would like them included.

I have older / unreleased material if you want to fill up free space, by the way.

_________________
http://www.mojontwins.com

I produced another pair of truncated builds for the PowerPak with the submultis, and a round of testing yesterday revealed several problems that I'll need to fix over the next few weeks.

• Attribute corruption in Rock Paper Scissors Lizard Sbock and Spacey McRacey. I'm not sure if I overestimated how many zeroes at the end are unused or if they actually have problems with uninitialized memory.
• Filthy Kitchen freezes at Game Over, with some stray letters at the upper left corner. I forgot to write down exactly what the letters were.
• Cheril the Goddess map is scrambled.

I am also aware of several problems that I can fix without knowledge of each game's internals:

• Builder: Reset patch works only for NROM games and automatic submultis, not for UNROM games. I'll try to fix this in the builder.
• Menu: Load all banks of CNROM if activity requests it
• Builder: Write whether activity record requests all CNROM banks
• Menu: Integrate JRoatch's coredump tool

About frequency of my status updates to this topic: Should I keep quiet until I have fixed all four of these?