Retroarch interface confusing?

Sorry for my MAJOR rant, but i’m just so sadly frustrated that I’ll be uninstalling retroarch AGAIN.

Some users obviously want to tinker, and its great for them to have every single little granular option exposed. But other people don’t need to put every little bit of ikea furniture together one piece at a time.

The main purpose of this app is for people to be able to play old games. Why so many hoops just to get started? The interface is not intuitive at all, rather still a half baked rip off of the ps3? Why are game cores still not called Emulators, like everywhere else. Why am i bug testing just to get games working. Why do I have to look up youtube videos on how to get a game started? Who in the world would think that: “install SNES emulator”

would equal

“online updater > core updater > (pick one of the 15 SNES emulators at random that all have truncated names)” and hope for the best!!! After all this time is there really NO recommended emulator? No recommended settings?

Like the title says. I install retroarch every 6 months or so when I get a new pc or new work setup, then uninstall it after a few days every time. I want to LOVE this project so much. But every single time I have to sift through menu after menu searching for things just to make it work. Then backtrack and try more things to make something else work. Then sift through the forums AGAIN. Its a jack of all trades master of none situation. The forums are filled with users trying to figure out how to get this working properly.

I’m sure a lot of fanboys have switched me off by now, but I have some thoughts on how to improve the new user experience.

Something as simple as a WIZARD section when you launch would help IMMENSELY. The wizard is optional, with a plain old “close” for all those who want to manually sift through it all.

Current UX is so bad you need to get launchbox to cover it up. Why has this never been updated?

UX should be:

  • User installs retroarch
  • User inputs Roms folder
  • Waits for install/update to complete
  • User presented immediately with wall of playable games.

Behind the scenes:

  1. Have a default roms path. This folder is scanned on launch without user input.
  2. Have user option to input rom path.
  3. Retroarch scans folder for file types, and self identifies file formats to Emulators.
  4. Retroarch self tests for the matching emulator, and if one is not installed, it installs the best one for that rom.
  5. Upon detecting new rom, perform scrape of igdb (or similar open source game database) for individual game artwork and data. (Because your servers are s.u.p.e.r.s.l.o.w)
  6. Scrape for platform artwork, icons and sounds.
  7. On first install, retroarch scans GPU, CPU, RAM, and matches that with minimum specs to run games. Flagging potential underpowered systems.
  8. User presented with a wall of immediately playable games with default plain posters. Posters, art and sounds update in front of users eyes as the scraping completes. (see plex/kodi)

I just want to play my games. Launchbox and individual emulators are fine, but it just seems that if this app was a bit more seamless, it would be the one ring to rule them all.

If anyone is actually in making this project look and feel nice, I will put my money where my mouth is and help for free. (i’m a designer and 3d artist)

See you in 6 months.

EDIT: formatting

Retroarch is a multi-platform engine and as such it is not easy to offer a one-button setup. It runs under Windows, Unix, OSX, Android, it has to support different graphics/sound APIs, different joysticks, …

It is an amazing piece of software but it requires some setup. I did it once, creating all emulated machines i wanted, bezels, joystick configurations and now i just update to latest versions without the need to change anything in my configurations… And i also use Launchbox, making it very easy to add new roms…

Give it another try…

1 Like

I’ve explained this a few places elsewhere, but I’ll do it here, too:

As @Crush mentioned, RetroArch runs on pretty much everything, which makes it hard to do a lot of UXy stuff.

RetroArch has a couple of missions, one of which is to be good for playing games, but it’s also “the reference frontend for the libretro API,” which means anybody that wants to hook into the API should be able to look at RetroArch’s source code and see how it works.

While most of our cores are emulators–emus are an especially good use-case for the libretro API–a number of them (and growing, we hope!) are game engines and/or other multimedia applications. So, it just doesn’t make sense to call them all “emulators”, and we don’t want to pigeonhole the API into “just emus”.

With that in mind, we think of RetroArch/libretro as a platform. While recommending/automatically selecting cores to end-users would be helpful to them, it means that we become king-makers. That is, if someone is considering adopting the libretro API for their, say, NES emu, they may ask themselves “why should I even bother when it’s not as fast as QuickNES or as accurate as Mesen, therefore it will not be recommended and thus no one will ever even see it?” It’s even worse for cores that are otherwise comparable.

Setting a default “content” directory, even if just for Windows like we do with the ‘system’ directory, seems like a pretty good idea to me. Many people will have/want to change it immediately, but that’s no worse than the current situation.

I see what you are trying to say, but if you think it cant be streamlined and optimised just because it has “options” i feel you are selling developers short. Its nothing more than a bunch of if/thens with some UX.

Its currently clunky, unintuitive, full of jargon, and poorly designed. None of this is to do with the amount of options.

checkPlatform if Android, then androidSetup if Windows, then windowsSetup …etc

windowsSetup; (and then I listed the if/then concepts previously)

I’ve already given this software many tries, and the developers have no care for ease of use. Ill stick with individual emulators until someone works on this element.

Maybe that person should be me. hmmmmmm

Download just the cores that are counterparts of the standalones you are familiar with. All cores have default settings

.

It’s the same for stand-alones, nobody tells them which standalones are good.

You just gave yourself the answer. Just go with the first one in your search result, there should the latest libretro counterpart, just the same amount of time.

There is something that @hunterk said that you are overlooking, not everybody would be pleased if they phased out one emulator in favor of a “recommended” one. Let’s say, for example, they decide to make higan the “go to” SNES emulator, that could demotivate the developers of Snes9x, and that is MY favorite SNES core, you can’t make everybody happy like that.

There is no “best”. It’s like the 16-bit console war.

EDIT: I don’t think I get your point about the user using their time to decide which emulation core suits them best. The way I see it, is that there is always a point you don’t know what emulator is right for you, there is no exception, standalones and libretro alike, you have to learn something first.

Snes9x, of course! and that would make some higan fans burn me alive.

Because I couldn’t care less about the amount of accuracy necessary to emulate a puny airplane shade of a game I couldn’t care even less for. That’s it.

I’m talking about Air Strike Patrol, the plane has a shadow that is hard to emulate. I don’t like that game, just to clear that out too. Then, there is the thing about performance vs. accuracy, accuracy is NOT free and not everybody has the latest PC, and if they have it, they may prefer employing that muscle power in more important things like elimiminating input lag.
EDIT: Oops, I didn’t mean Snes9x is inaccurate, it’s good enough, it’s just that higan goes overboard.

There is no demotivation, the users have the last word on it.
EDIT: And by that I mean there is equal oportunities for both emulators, they can coexist and try to best each other, because there is no clear winner yet (hopefully the competition will last endlessly).

At least I enjoyed talking to you. I wouldn’t consider chatting about common interests a waste of time.

I don’t disagree with you intent of improving. I’m just concerned about that specific point about choosing winners, that’s all. You can voice your ideas and hopefully, you’ll get reception (if you make sense of yourself). That’s the beauty of this community, empowering users, giving them choices.

So yes, starting to learn about emulators is a good step to help you deal with your own shortcomings.
Nobody cares if you want to use LaunchBox.

Same in 6 months.

typical useless answer from someone acting superior because they know something technical that someone else doesn’t. Good job, that will help.

So an app created to unify and simplify is needlessly complicated? I didn’t try this app so that I could make things MORE complicated.

So I try and use an app to emulate games, but now I have to google every different setting and how they all function.

Judging by the community in this forum, this scene is too toxic for me. Its all yours.

I think part of the disagreement here also comes down to what the goals of RetroArch really are. Unfortunately, ease of use is very far down that list, as you’ve seen. The top 3 (and they’re pretty equal) are: portability (run everywhere), advanced features (shaders, netplay, runahead) and libretro reference frontend. (and no, the part about looking at the code isn’t meant for you as an end-user, but rather for other developers who want to make their own libretro frontend unrelated to RetroArch; if those people don’t understand something about the API, they should be able to look at RetroArch and come away with a better understanding of what the API is trying to do).

MAME has a lot of the same user complaints because their top goal is preservation. Actually playing games is further down their list, so they sometimes make design decisions that makes those of us who just want to play games groan.

We aren’t trying to simplify emulation or make a babysitter for people’s toddlers (I don’t mean that as an insult, but rather literally; a common UX complaint is that they can’t leave their child unattended with it, though we now have a kiosk mode to help with that). We think of RetroArch not as a simplifier but as a power-tool. Think Photoshop Elements vs Photoshop CS. We obviously would like to make it as friendly as possible, but those other aspects come first.

1 Like

I’ve locked this thread and reverted all of the deleted posts. This discussion is worthwhile and it’s not cool to try and erase history.