bboard

Posted by Wowfunhappy
Why are you building Dolphin yourself anyway? It doesn't sound like you're making any source code changes?

Posted by Wowfunhappy
...on a scale of 1–10, how much would you hate me if I suggested using Flatpak?

Posted by creaothceann
https://www.reddit.com/r/emulation/comments/ciy5tj/bsnes_v108_released/evbi18r/?context=3

Also the pages lack the first-level heading element — h1, incorrectly using the header element instead. The header element has a totally different meaning than heading elements like h1, h2, etc.

Posted by Kawa
You do realize you're talking about byuu here, right?

Posted by Wowfunhappy
I have one really big annoyance with how RetroArch handles controllers:

If my controller has an B button, and the system I'm emulating has an B button, I want real B to trigger emulated B, regardless of button placement. So if I'm playing an SNES game with an Xbox controller, Xbox B means SNES B.

However, if the real and emulated button labels don't match, I want RetroArch to map according to placement. So if I'm playing a Playstation game with an Xbox controller, Xbox B means Playstation O.

There's no way to set RetroArch to do both these things, short of configuring each system individually, which is quite annoying in RetroArch's UI.

I'm probably being overly picky, but this behavior does strike me as the logical way to do things.

Posted by CaptainJistuce

Posted by sureanem

Posted by CaptainJistuce
I'm assuming they're using some sort of embedded microcontroller version that runs at something like 200 MHz. I mean, if it is possible for the Z80, it is possible for the 80286.

The article says it runs at 20 MHz, make of it what you will.

I read it as saying the 80286 part wasn't available in speeds above 20 MHz, which is factually untrue(Intel's 80286 parts went up to 25). And then starts talking about the possibility of overclocking.
More importantly, there are no new 80286s available, and Boeing isn't building a plane reliant on eBay to source parts.

I think they heard it used a 286 core, did some googling to find out what a 286 is, and ran. I don't think they have the actual specs of the system.

That’s a Harris 80286.

Can it do floating point math? No.

It can execute, roughly, 0.21 instructions per clock.

Does that say “distributed system” to you?

Not me.

Thing is, according to reasonably reliable sources – that’s what the 737 MAX software runs on.

Before I go any further, I have to state that all airplanes run generations behind the bleeding edge of hardware design in the name of cost, safety, and reliability.

No one wants a zero-day CPU instruction set bug grounding an entire fleet because some dumbass at Intel designed on-chip WiFi for the consumer market, and because of a common fab, the enterprise chips still have the feature, you just have to run some bizarre emulator… You see where I’m going.

Bad news. So you can’t begrudge Boeing for using the 286 architecture to run the 737 MAX software, at all.

Airbus—just to be show-offs or something—when it comes to Kung Fu, they decided to write the entirety of the A32X fly-by-wire software to run on a combo of Motorola 68000s and 186s.

The problem with the Paleozoic-era hardware is that two AOA (angle of attack) inputs, when combined with an inelegantly coded MCAS, can overload the CPU in present iterations of the code.

The FAA calls issues related to this, that came up in the simulator, “catastrophic.” It’s not a term they mean to use to fear monger, of course. But it’s not a happy word all the same.

Software is a challenge, well no. Elegantly-coded software is a challenge.

There are also so many questions as what the FAA will consider “acceptable.” Parallelism would be a kludge that some may consider a vast divergence from the 737NG. New hardware would definitely be considered an extensive modification that, given the realities of a post-MCAS world, will be gone over by CS professors and experts the world over before it so much as is allowed to carry a mouse.

It doesn’t matter how Boeing, ultimately, solves this problem—it’s going to add considerably more time to the aircraft’s return to service than they and their shareholders would enjoy.

Until I know more, I am in no position or want to speculate. You can solve a software problem in ten seconds with fresh eyes, or it could take huge team years.

08-04 18:27:43.776 23532 23594 E AndroidRuntime: FATAL EXCEPTION: pool-1-thread-1
08-04 18:27:43.776 23532 23594 E AndroidRuntime: Process: com.bancoplaza2.p2p, PID: 23532
08-04 18:27:43.776 23532 23594 E AndroidRuntime: java.lang.UnsatisfiedLinkError: dalvik.system.PathClassLoader[DexPathList[[zip file "/data/app/com.bancoplaza2.p2p-2/base.apk"],
nativeLibraryDirectories=[/data/app/com.bancoplaza2.p2p-2/lib/arm, /system/lib, /vendor/lib]]] couldn't find "libsqlc-native-driver.so"
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at java.lang.Runtime.loadLibrary0(Runtime.java:984)
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at java.lang.System.loadLibrary(System.java:1530)
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at io.liteglue.SQLiteConnector.<init>(SQLiteConnector.java:8)
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at io.sqlc.SQLiteConnectorDatabase.<clinit>(SQLiteConnectorDatabase.java:39)
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at io.sqlc.SQLitePlugin.openDatabase(SQLitePlugin.java:212)
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at io.sqlc.SQLitePlugin.access$000(SQLitePlugin.java:28)
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at io.sqlc.SQLitePlugin$DBRunner.run(SQLitePlugin.java:328)
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1133)
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:607)
08-04 18:27:43.776 23532 23594 E AndroidRuntime:        at java.lang.Thread.run(Thread.java:761)
08-04 18:27:43.776  2185 14760 W ActivityManager:   Force finishing activity com.bancoplaza2.p2p/.MainActivity

Posted by CaptainJistuce
Actually, the thing holding mobile pay back now is card readers. Lots and lots of card readers out there without any way of talking to a pocket computer. And the card reader manufacturers are going to charge a premium to upgrade to a new model with more features(card reader manufacturers apparently treat their customers as poorly as can possibly be done, so of course they charge extra for everything).

Some places have NFC readers, and they support Google Pay and Apple Pay.
For everywhere else, there's Samsung Pay.

Samsung Pay works everywhere, because the PDAs that support it simulate ye olde magnetic stripe via actuation of an internal electromagnet. No one has to upgrade their hardware to include NFC readers. It is a clever hack, and I respect that, even if I hope to never own another Samsung phone ever again.

Posted by CaptainJistuce

That said, "banks are awful" is a universal truth.

Posted by tomman
Just last week I lost big time, when the debit card networks decided it was a great time to "chip rejection" (whatever that means), while eating my money in the act. Called $BANK, they say "wait 72 hours, then come back with a formal complaint letter". If I'm lucky, I'll get my money back in FIFTEEN BUSINESS DAYS (i.e. 3 weeks, maybe a month). Normally this is not a big deal, but then, this is not a normal country. When I get my money back, it will be worth half it was worth today (in the best of cases). And yes, this happens to any random guy (or gal) over here at least once a year (the only new bit this time is the "formal complaint letter" - they used to deal with this over the phone, but I guess their call center drones defected to Argentina last year).

Posted by CaptainJistuce
Why does Microsoft get credit for inventing button labels that read in the right order when Sega's been doing it since the 80s?

2019-08-10 Because the master copy has already been sent to the duplicators, Wily Beast and Weakest Creature 1.00b update released ahead of Comiket 96 for players to use to fix important bugs when the full game comes out.