It is currently Tue Dec 11, 2018 9:24 am

All times are UTC - 7 hours





Post new topic Reply to topic  [ 24 posts ]  Go to page 1, 2  Next
Author Message
PostPosted: Sun Aug 12, 2018 9:42 am 
Offline
User avatar

Joined: Tue Mar 22, 2016 8:27 pm
Posts: 164
Location: Minnesota, USA
Similar to adding an AY-3-8910 synthesizer to an FME-7 cart to make a Gimmick! repro, it seems equally simple to add a dirt cheap YMF262 for expansion audio. Hardware-wise, it seems totally painless and straightforward. This could be added super easy to a mapper-0 cart for proof of concept and be a platform for test code.

Less than $4 for both the YMF262 and DAC chip, for example:
https://www.ebay.com/itm/162781399389

What are your thoughts on this? Would it be useful? Would it sound 'wrong' to have Soundblaster style FM sound coming out of an NES? Would there be too much overhead to make this chip sound right? It just seems to me that you could basically tack on these 2 chips super easily and open the door to way more versatile audio for future hacks and homebrews. Other than no emulator support, someone please bring me to reality because I see no problems with this idea and it seems too good to be true.


Top
 Profile  
 
PostPosted: Sun Aug 12, 2018 9:51 am 
Online

Joined: Sun Sep 19, 2004 11:12 pm
Posts: 20853
Location: NE Indiana, USA (NTSC)
OPL3 in a Famicom cassette would sound no more wrong than Lagrange Point, which includes a cut down version of OPLL, which in turn is a cut down version of OPL2. But practically, unless your game is in Japanese or Chinese or Russian or the language of some other country where famiclones are popular, you'd also need to include a jumper pack that fits in the bottom expansion port of an NES-001 Control Deck to route the audio. Paul at Infinite NES Lives had prototyped such a jumper pack, but it doesn't appear on the products page yet.


Top
 Profile  
 
PostPosted: Sun Aug 12, 2018 9:54 am 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 7807
Location: Seattle
... sniped by Tepples...

We've actually talked about that idea before, several times. I was pretty enthusiastic then, too.

I don't think an OPL3 sounds so radically different from the just-barely-not-OPLL in Lagrange Point... although to be fair, Lagrange Point's audio is extremely unique.

Adding any FM synth is not so different from the difference between the US SMS (with its sn76496 audio) vs the Mark 3 (additionally had a YM2413).


Top
 Profile  
 
PostPosted: Sun Aug 12, 2018 12:26 pm 
Offline
User avatar

Joined: Mon Feb 07, 2011 12:46 pm
Posts: 1024
You can just mention, "require audio jumper pack to play music in NES" (expecting the customer to provide their own audio jumper pack or Famicom)

_________________
.


Top
 Profile  
 
PostPosted: Sun Aug 12, 2018 1:46 pm 
Offline
User avatar

Joined: Fri Nov 12, 2004 2:49 pm
Posts: 7600
Location: Chexbres, VD, Switzerland
tepples wrote:
OPL3 in a Famicom cassette would sound no more wrong than Lagrange Point, which includes a cut down version of OPLL, which in turn is a cut down version of OPL2. But practically, unless your game is in Japanese or Chinese or Russian or the language of some other country where famiclones are popular, you'd also need to include a jumper pack that fits in the bottom expansion port of an NES-001 Control Deck to route the audio.

I doubt Chinese and Russian famiclones works with expansion audio without any modification. Many of them maps Audio In and Audio out intenrally, as well as they map CHR-RAM /CE and PPU /A13 internally, disallowing 4-screen nametables.


Top
 Profile  
 
PostPosted: Sun Aug 12, 2018 2:25 pm 
Offline

Joined: Tue Feb 14, 2017 9:50 am
Posts: 97
Some do while others don't. My Micro Genius IQ-1000 works with Gimmick!. I don't have a mid-late 90s clone to test but a 2000s Power Joy doesn't, for example.


Top
 Profile  
 
PostPosted: Fri Aug 31, 2018 12:03 pm 
Offline

Joined: Tue Aug 07, 2012 11:01 am
Posts: 16
Ben Boldt wrote:
Similar to adding an AY-3-8910 synthesizer to an FME-7 cart to make a Gimmick! repro, it seems equally simple to add a dirt cheap YMF262 for expansion audio. Hardware-wise, it seems totally painless and straightforward. This could be added super easy to a mapper-0 cart for proof of concept and be a platform for test code.


Hi

Certainly doable. I've done OPLL, OPN2 Famicom carts already, eventually I'll make some for every sound chip
(already started on OPL3).

See Teaser for my Famicom based synth in my tech promo for PortaFM
https://www.youtube.com/watch?v=bzIeyZrIMkM

Basically my own GAL mapper(kinda like INES Mapper 034) with 512KiB of PRG ROM and 8KiB of CHR RAM.
If there is demand I could scale this outside of my target chiptune synth head crowd


Top
 Profile  
 
PostPosted: Fri Sep 14, 2018 9:18 pm 
Offline
User avatar

Joined: Tue Mar 22, 2016 8:27 pm
Posts: 164
Location: Minnesota, USA
Very cool plgDavid.

Per the old "MMC7" discussion here:
http://forums.nesdev.com/viewtopic.php?t=8579

I think MMC5 is the best we have, and expanding on it with a better synthesizer and the cool features they talk about could be a path forward for really advanced Famicom games in the future. We might not have the driving force behind homebrews or cheap giant TTL compatible CPLDs to make it feasible at the moment, but as millennials start reaching retirement, I predict these things will change in a big way. Now is the right time to plan, spec, and standardize on MMC7 in my opinion. We should continue to brainstorm for ideas and improve our understanding of MMC5.

From your experience, what do you think is the most fitting synth for a future MMC7 NES game? For an actual synth chip, YMF262 is very easy and cheap to attain, and there is something special about using the real chip -- not to discount the accuracy of the emulation demonstrated in the video.


Top
 Profile  
 
PostPosted: Fri Sep 14, 2018 10:46 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 7807
Location: Seattle
No, standardizing on a pipe dream is definitely a bad choice.

Mappers should be designed responsive to a software developer's desires, not according to some abstract ideal "most advanced mapper everrrr!".

The MMC5 is markedly more complex than any game wants (no game needs the ability to switch between a dozen different PRG and CHR banking styles). The 8x8 attributes and 16384-tile mode drastically increase the complexity of nametable updates and consume twice as much bandwidth for nametables updates (And that bandwidth is already astoundingly limited).

Left-and-right split mode is nifty, but since they have to use the same fine X scroll, it's really hard to use interestingly. Three nametables are ... ok ... but really, more more-flexible nametables (either mostly-fixed, like GTROM; or arbitrarily configurable, like Namco's 163) are better. The 512-tile 8x16-sprite mode is a nice idea, but the NES's extremely limited overdraw drastically reduces the situations in which it's worth using 8x16 sprites.


In contrast, we've repeatedly heard from people developing games that they'd really like a dual-ported RAM interface to pattern tables and/or nametables. (Scheduling updates via the PPU's limited bandwidth interface sucks.) The MMC5 technically provides a dual-ported nametable, but only one, and if it's on-screen one has to be careful to avoid visible tearing. (And if it's not on-screen one had to use the standard bandwidth limited PPU interface for the other nametables).


Top
 Profile  
 
PostPosted: Sat Sep 15, 2018 12:11 pm 
Offline
User avatar

Joined: Tue Mar 22, 2016 8:27 pm
Posts: 164
Location: Minnesota, USA
As a firmware developer myself, I somewhat disagree. I think that developing an approachable platform ahead of time can make things possible that won't otherwise happen. This is where the divide between hardware and software people makes a difference. People tend to specialize in hardware or software, not often being equally proficient in both. If a software-minded person wants parallax scrolling, they are going to pick an existing mapper that supports it and go from there. It won't make much sense to that person to create a new mapper first. They will make tradeoffs so that they can dig right in with their cool ideas and not have to wait around and hope for a hardware person to swoop in and rescue them with a new mapper. Likewise, we should not build this mapper and expect that a software developer will jump right in. We just want to "pay it forward" and create this platform for them to use, as best as we can, if and when that person should come along. Let's be prepared for that amazing software person that may or may not show up one day.

What I am suggesting is to take time now to design a really easy to use and feature-rich mapper. Nobody will possibly use all of the features in the same game. Similar concept to an embedded OS. An embedded OS provides all sorts of features like timers, multitasking, hardware abstraction, etc. Blah blah blah. But it's there for us and we can delve in and use more OS provided features if/when we run into something during development where it can make life easier, or allow us to do something we could not have otherwise done. That is the kind of approachability that a standard, agreed-upon mapper can bring. I have seen software platforms designed repeatedly in my career and they work.

What is great about this, in contrast with an embedded OS, is that the only overhead from making a complicated mapper is the cost of the hardware, which will definitely go down by the time these will-be famicom developers rise from their retirement.


Top
 Profile  
 
PostPosted: Sat Sep 15, 2018 2:05 pm 
Offline
User avatar

Joined: Sun Jan 22, 2012 12:03 pm
Posts: 7007
Location: Canada
Well, the MMC5 itself has been "approachable" for one or two decades now. What has been the result?

Mostly just a handful of romhacks that barely use its features (e.g. infidelity's "Mario All Stars" compilation). The only homebrew I've seen is Demptronic Football, which is still in progress.

You can make specs for whatever mapper you'd like to see, no problem with that by itself.

Once you start asking emulator authors to implement it for you in all of their emulators, so that it will just be there for when you think somebody might "approach" it in the future...? Why would anyone want to do this for you? Do you realize that 99% of promised game projects never release anything? In this case you're not even promising a game, you're just hoping someone else will.

If we implemented every dream mapper anyone's ever had there would be a few hundred mappers in the spec with no ROMs to run on them. That's wasted work for every emulator that implements a dead mapper. (By the way, if you want to make it more possible for someone to implement your new mapper, write test ROMs for it. It's much, much easier to verify a correct implementation with that -- which is equally important part of the task as just knowing what it should do logically.)

If you need an emulator to work on your MMC7 game, modify it for yourself (or even ask for help in doing it). This is not a very difficult task compared to actually making an NES game (for any mapper). Maybe someone would even help you with that. You can even redistribute that mapper and/or the added code so that others won't have to do this first step. No big problem here.

...but calling for people to "standardize" and hold your place for you on this? No, that's just not going to happen.

Mapper ideas are generally very easy to have. Members of this board seem very capable of coming up with a new very good mapper idea every day of the week. Making software that's worth running on it is the hard problem, and without doing that it's not worth standardizing any mapper. To be frank: good mapper ideas are cheap. Sorry. This particular idea is not exceptional.


Top
 Profile  
 
PostPosted: Sat Sep 15, 2018 2:23 pm 
Offline
User avatar

Joined: Tue Mar 22, 2016 8:27 pm
Posts: 164
Location: Minnesota, USA
I am sorry, I did not realize how widely this has been discussed already here. I am still pretty new at this board. I should do more searching and reading before making posts like this. I don't mean to sound like I know a lot about the history or potential in NES homebrew. I do wish to help make advances to support people like you Mr. Lizard developer -- that is my true intention.

I came across this thread that especially explains a lot about custom MMC3/5, I will study it:
viewtopic.php?f=9&t=17291

Also worth considering, brainstorming is useful, even if not all ideas are winners. #1 rule of brainstorming is not to shoot down others. Expand on it or redirect them and something useful can come. Direct people to a mapper ideas thread. Though I love Lizard and think your work is amazing, I do actually feel pretty bad and disheartened by your comments and lost interest in this train of thought at the moment.


Top
 Profile  
 
PostPosted: Sat Sep 15, 2018 3:56 pm 
Offline
User avatar

Joined: Sun Jan 22, 2012 12:03 pm
Posts: 7007
Location: Canada
Hey, questions about potential new mappers are no problem. That happens all the time here, and people do like talking about it (usually). If you're having trouble prototyping one for yourself, I'm sure you could get advice for that too.

Promoting a mapper idea into a bona-fide iNES mapper is an effort of completely different magnitude though. It does happen, but you have to do something to convince people the work is worth it, and in general a promise of potential won't accomplish that.


Top
 Profile  
 
PostPosted: Sat Sep 15, 2018 4:47 pm 
Offline
User avatar

Joined: Tue Mar 22, 2016 8:27 pm
Posts: 164
Location: Minnesota, USA
I didn't mean to ask for anyone to do anything for me, or buy my concept and start working on it. It isn't about me, I am not looking for someone to make me a mapper or print a PCB or modify an emulator for me or anything like that. I have no plans of making money from this. I am just sharing my ideas and looking for feedback how we could work together as a group to create large-scale, well-planned platforms to support each other, and also to provide my skills wherever possible.

For infidelity's Legend of Link and Super Mario All-Stars, I made a reflashable MMC5 cart out of "Just Breed", and sent it to him along with a real Japanese AV modded Famicom, all for him to keep. He needed this platform for his development and debugging, and I made that happen, all on my own dime. These are my intentions; not everyone here is for the greater good, but I am, and I believe in the potential of a new standard platform even if we haven't come together and spec'd out any realistic details about it yet.


Top
 Profile  
 
PostPosted: Sat Sep 15, 2018 5:49 pm 
Offline
User avatar

Joined: Sun Jan 22, 2012 12:03 pm
Posts: 7007
Location: Canada
Who said anything about making money? O_o Asking people to standardize on something necessarily means that more than one person who isn't you has to implement it. (...and in the case of emulators, all of that is normally volunteer work.) Otherwise it's not a standard, it's just a prototype. Nobody minds prototypes. I think everyone here is interested in seeing those.

If you want an example of someone designing a platform that makes NES development easier, NES Maker is a good one. This project did not involve designing a new mapper, and in fact ended up reusing a simple one that already existed. The parts of this platform that are actually making NES development easier are 1. software tools for development and 2. a flash cart and USB flasher device.

Of course there's also PowerPak and Everdrive N8, and on a smaller scale maybe Loopy's FDSStick combined with an FDS makes a really nice Famicom development kit, but interest in the FDS format seems to be a minority. Each of these has been very helpful toward facilitating development.

I don't believe a bigger mapper is really the best approach to making NES development easier, pretty much in the same way that SNES development by most accounts is a bit harder than NES development. Even just adding a synthesizer chip to the cartridge adds a bunch of complications. More power isn't by itself effective, it has to apply to the right problems, and most of the important problems here aren't easily addressed by a mapper.

There's been hundreds of mappers proposed on these forums. I sincerely doubt you'd find any configuration that could become the "standard" one for homebrew, not just by designing it. You're welcome to openly speculate about it, but I expect you're going to be disappointed if you want it to become this.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 24 posts ]  Go to page 1, 2  Next

All times are UTC - 7 hours


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group