[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/vr/ - Retro Games

View post   
File: 36 KB, 512x448, Final Fantasy (Square, NES 1989).png [View same] [iqdb] [saucenao] [google]
8719227 No.8719227 [Reply] [Original]

>game is bugged to hell and back, some of the spells don't work and others do the opposite of what they're supposed to
>it gets even better--the resident braintrust thought he was being a l33t hax0r for using the power-on state of the NES's WRAM for RNG generation (since it's full of garbage on power-on)
>this of course doesn't really work on emulation or Flash carts (Everdrives fill the RAM with 0s or 1s when you start a game) so the RNG generation in the game is bricked
The absolute state of Nasir Gebelli.

>> No.8719239

CHAOS made my game an unplayable mess.

>> No.8719246

And it's still one of the greatest, most enjoyable, and FUN games ever made.

It's astonishing comparing /v/ and /vr/ and seeing what a total useless, seething, rancid shithole cesspool this board is.

>> No.8719252

>>8719227
>>this of course doesn't really work on emulation or Flash carts (Everdrives fill the RAM with 0s or 1s when you start a game) so the RNG generation in the game is bricked
Based. Pirates get cucked by their own greed.

>> No.8719254

>>8719227
>emulation defeated
Based nippon glory

>> No.8719257

>>8719227
Stranger of paradise is the best version

>> No.8719272

>>8719227
>using the power-on state of the NES's WRAM for RNG generation (since it's full of garbage on power-on)
>this of course doesn't really work on emulation or Flash carts (Everdrives fill the RAM with 0s or 1s when you start a game) so the RNG generation in the game is bricked
Well, that is smart to use garbage as a seed for RNG. Is your complaint that emulation is non-seeded?

>> No.8719276

>>8719246
/v/ is garbage, you're retarded.

>> No.8719278

>>8719257
best at being cringe, maybe

>> No.8719279

>>8719239
Then I hate CHAOS

>> No.8719283

>>8719246
Completely braindead post, leave and never come back.

>> No.8719287

>>8719278
Soulless you are

>> No.8719291

>>8719227
>using the power-on state of the NES's WRAM for RNG generation (since it's full of garbage on power-on)
>this of course doesn't really work on emulation or Flash carts (Everdrives fill the RAM with 0s or 1s when you start a game) so the RNG generation in the game is bricked
What does that mean?

>> No.8719295

Everdrives don't clear the video RAM properly (that is to say junk from the menu is left behind) so you can get graphics glitches from that. I prefer my real Dragon Warrior 1 cart for that reason after I found it would have glitches on the Everdrive due to leftover bits of the menu that were still in there.

>> No.8719301

>>8719227
>(Everdrives fill the RAM with 0s or 1s when you start a game)
Why?

>> No.8719335

River City Ransom is another one that does this RNG trick.

>> No.8719341

>>8719291
what do you think it means, dummy? the RAM in the console is full of garbage values when power is first applied to it which the game uses to seed its RNG generator.

>> No.8719353

>>8719341
So the practical outcome of this is? The RNG is pre determined on emulation?

>> No.8719357

>>8719353
>The RNG is pre determined on emulation?
I'd assume so.

>> No.8719373

>>8719246
Jesus the seethe, can you handle even the lightest amount of criticism for a game you enjoy? OP didn't insult you directly.

>> No.8719382

>>8719227
in all fairness Nasir Gebelli probably didn't anticipate emulation or Flash carts, nor was he under any obligation to accommodate them

>> No.8719406
File: 40 KB, 711x548, Untitled.png [View same] [iqdb] [saucenao] [google]
8719406

>>8719227
>this of course doesn't really work on emulation
ENTER

>> No.8719410

>>8719406
The NES resets the PPU when the Reset button is pressed while the Famicom does not. This was probably done to eliminate spurious graphics glitches when the console was reset.

>> No.8719434

>>8719278
Bullshit

>> No.8719457

man, Nasir didn't even know exactly what he was programming he'd never seen or touched an RPG before

>> No.8719473

I hate when people shit talk Final Fantasy 1’s programming. You don’t understand the context at all! You need to read Super Mario Land’s credits.

>> No.8719491

>>8719353
worse than that, some pseudorandom sequences don't work if you seed them with all zeroes; you'll just keep getting all zeroes over and over again instead of advancing through the sequence. so if that was how it worked, the RNG wouldn't work at all rather than just being exploitable.

>> No.8719494

why can’t programming just generate a random number
every game i hear of uses a seed

>> No.8719503

>>8719494
Most consoles and computers don't have any means of true RNG so you can only use a pseudo-random hash. The Atari 8-bit and C64 were unique in that regard because POKEY and SID's white noise channels could be used for true RNG.

>> No.8719505

>>8719227
>>it gets even better--the resident braintrust thought he was being a l33t hax0r for using the power-on state of the NES's WRAM for RNG generation (since it's full of garbage on power-on)
Well gee, he should've somehow predicted that people would be pirating the game and playing it on different systems in a decade's time.

>> No.8719520

>>8719494
>>8719503
Yeah, this is why some games do RNG by going off of a seed of some sort (like Diablo), or by having a predefined table of relevant values to cycle through with every action in the game (like Doom).

>> No.8719542

>>8719227
The bugged spells are definitely a sore point where he fucked up bad with no testing done. Granted, that's also on Square. Developers make mistakes all the time, it's not technically their job to see if bugs occur.

What a stupid secondary point however, why would he be trying to think about a hardware case that didn't apply to the product being sold?

>> No.8719543

one thing that always annoyed me was people thinking that armor lowering your evade% was a bug. if you look at the tables its pretty obvious that its based on the supposed weight of the armor in question - which is a neat thing to emulate, especially for the time. also the everdrive RAM stuff smells like bullshit considering my experience with other RPGs on the system. maybe ill test it myself soon

>> No.8719571

>>8719543
>also the everdrive RAM stuff smells like bullshit considering my experience with other RPGs on the system
It's not necessarily a given that all RPGs do it, they may use other means of RNG.

>> No.8719685

>>8719543
The real shitty part is that you just have no idea about it. Like in FF2, weapons and armor reduce your spell accuracy. You can't tell (at least in the Dawn of Souls version I'm playing, no I don't care if this isn't "the right way" to play it), because that particular stat isn't shown on your character's status screen, and no mention of this is made in the game manual.
It would be easy to eventually notice that after an armor upgrade your guy seems to be getting hit a lot more, see no mention of it in the manual or any in-game screen, and conclude that it must be a bug.

>> No.8719796

>>8719494
Short explanation: Because it can't pull a number out of thin air.

It needs something to start with, and electrics flooding RAM with garbage data is one way of doing it. Another common way is the system clock, or simply have it roll through a list of preset numbers every frame (Suikoden 1 does this, this is often exploited in speedrunning) or when called (Fire Emblem, Etrian Odyssey, and some turn-based strategy games utilize this, sometimes as an anti-scumming measure).

To have true randomness, you need to take a constantly-changing number you cannot reasonably predict, and throw some math at it to twist it into a usable number. If you want an outside influence, you can use a camera pointed at something constantly-changing like a road crossing or even lava lamps, and have aspects like the checksum, the RGB value of any given pixel, or other changing variable be used as a seed.
https://en.wikipedia.org/wiki/Lavarand

>> No.8719802

>>8719796
>To have true randomness, you need to take a constantly-changing number you cannot reasonably predict
yes and only a few machines like C64 can actually do it (though BASIC's RND function is pseudo-random as it had to be machine-independent)

>> No.8719809

>>8719227
I'm sure glad i don't emulate.

>> No.8719850

>>8719246
Nobody on /v/ has ever played Final Fantasy.

>> No.8719916

https://www.romhacking.net/hacks/6239/

Is this the fix?

>> No.8720001

I think this and FF legend are the only jrpgs crpg addict gave a positive rating

>> No.8720013

>this of course doesn't really work on emulation

I don't believe you, unless you've been playing in Nesticle.

>> No.8720020

>>8719543
This, people are retarded.

>get better armor
>better protection but lower evasion because the armor is heavier

But oh no no no this is a bug because I can't fathom the game having a different logic than the dumbed down sequels, and why would I have to make a choice between 2 armors in a RPG? I mean come on you're clearly supposed to buy a new upgrade every town

>> No.8720021

>>8719276
Looking pretty good when shitty seething threads like this are constantly posted on this board.

>> No.8720028

>>8719227
>Bad programmer because it doesn’t work on emulator
OP you are the faggot here.

>> No.8720039

>>8719542
>The bugged spells are definitely a sore point where he fucked up bad with no testing done. Granted, that's also on Square. Developers make mistakes all the time, it's not technically their job to see if bugs occur.

This is true and especially for a RPG. From experience most developpers didn't even "beat" their game when it comes out, as in play through them from start to finish. The only devs that would do so are high position of influence like a director, and even then... Nobody expects of a programmer to thoroughly test a 30hours RPG.

Of course making mistakes is on the guy, but the load is shared with the publisher not owning to how much testing this kind of games requires to be thorough. Of course if the stories of Squaresoft being on its last breath at that point are true, this would explain it.

>> No.8720824
File: 109 KB, 400x300, knob.png [View same] [iqdb] [saucenao] [google]
8720824

>>8720028
Nasir is naturally a good programmer. Hell, a well programmed emulator will actually be able to work around this limitation, see: >>8719406

OP is of course, a faggot.

>> No.8720842

>>8719272
This, the real problem is inaccurate emulation.

>> No.8721628
File: 2 KB, 200x200, warmech.png [View same] [iqdb] [saucenao] [google]
8721628

>>8719353
Not saying FF1 exactly works like this, but when you get a random encounter in it, it can draw from one of eight predetermined setups based on the area you're in. For example, the first area might be 3 Imps for encounter 1, 5 Imps for encounter 2, 9 wolves for encounter 3, etc. To randomly determine what you'd get, the game would read this garbage data of 0s and 1s, convert them into a number between 0 and 7, and that's how it would decide your encounter.

Practically speaking, this would cause some weird issues like ALWAYS encountering the standard 3 imp group (because that's encounter type 1 and all of your random numbers are 0, so it always goes to that one) or encountering a pack of zombies if it were all 1s. This could mean that at a later area where a very difficult encounter spawns rarely, you encounter it every time you get into a fight, and Warmech wipes the floor with your party over and over again.

>> No.8721635

>>8719406
Dunno what you can do about Everdrives though.