Astro City Mini is a closed source emulator and it plays Virtua Fighter accurately. I'm confident that's your only option. Comes with 2 save state slots but can't see or adjust the DIP switch settings. Still sold new today. Don't buy the V by mistake.

Shmups Junkie on YouTube showed that the Mini and Mini V have very fast screen displays. You probably add 1 frame of lag using the HDMI out to display + capture card. I think recording the Mini with cell phone cam on tripod would be very funny but that is my plan.

SMB legend @Kosmic

I watched one @Kosmic video and he refused to name the cheater like he's above the fray but relentlessly bullied them anyway. Didn't seem like how a legend should act to someone who didn't lie under oath or steal YouTube revenue.

Beyond the games, joey, who runs therun.gg, is working on the website becoming a front page for all of speedrunning.

I know video game journalism is a low barrier but since Speedrun.com is listed as a patron on the website, you should disclose that. If someone else makes a similar website, do they get a free writeup in the official newsletter? How do you know joey hasn't done a speedrun in nine years? Is he your friend or an admin here? If people know, they can decide for themselves.

You know what, the current situation with Hierophant and Priestess isn't so bad. Recruit Tristan and Rauny, go male. Recruit Tristan, go female. If someone goes male and only gets Tristan, the run could count as Priestess. Less confusing that my other suggestions.

Requiring "full ending" shouldn't be a strict rule. I say that after finding two any% World runs from Hyoko Ball on their channel. As in, they got Tristan, Rauny and Aisha, Brunhild, Armband and Grail and the 12 Zodiac Stones. The old Japanese TAS is also any% World.

Makes me think: Does anyone have a problem with changing World to not needing the full ending? 100% remains the same, I mean the World that doesn't need Deneb/Ashe/Lyon. Not needing to recruit most characters is more distinct from 100% and can skip more stages.

You're just claiming lag is not an issue and fail to mention you're using an unofficial emulator. Runs can be rejected solely for inaccurate emulation without an explicit rule. You're not disproving the extra lag frames are an unfair advantage over real console either. That the moderator spells out the final boss being easier with the extra lag sounds like an easy reject to me:

I play Virtua Fighter 1 Arcade and in that game MAME is banned because it's too inaccurate. The game doesn't play the same. There's thus no ability to compete with an unofficial emulator for the arcade release. Fortunately, the Astro City Mini arcade emulation is accurate and mods told me they would allow.

I play some fighting games but I can't view the most prolific runners' profiles without risking freezing or crashing my browser. Like, say, DBcade - Speedrun.com and FelipeNascimento83 - Speedrun.com. On Windows 10 PC, sometimes a page loads in a few seconds and sometime never. My phone stops loading results at 2 games.

The answer is at hand. I did this at my job. You can limit the number of results straight from the SQL query. You can also use a COUNT command to determine the number of results which runs very fast because it doesn't return any rows of data. You can even return a partial list of the results to display on one page, versus getting the whole 1000+ dataset and paginating the whole thing. Maybe have a different protocol for mobile users.

Bathroom reading material has arrived. The truth is within reach.

One day people noticed emulators giving the same RNG on each bootup of Super Off-Road, unlike real console. This is because, perhaps uniquely on SNES, Super Off-Road uses uninitialized RAM for starting RNG. A gentleman and scholar named ghost says it's bits 4D, 4E, 4F, 50 and 51 in the RNG subroutine at 99C2. That's right, there are 2^5 = 32 possible starting combinations for RNG. Perhaps 32 total RNG states for just 5 bits of entropy.

Real console, uninitialized RAM is sort of randomly 0 or 1 at bootup and games initialize to all 0s or whatever they want. I say "sort of" because the patterns have bias. The combinations are not equally likely, nor would their odds match on every console and some may never occur. It's debatable how to handle this on emulator. The common answer is...not to.

I looked at the first item spawn while staying idle on different emulators. Let's number the spawn points and use "c" for cash and "n" for nitro:

Reload the ROM or hard reset, same thing. SNES9X v1.60 "Reset" is a soft reset, which I'll get to:

BizHawk 2.9.1, BSNESv115+

• Entropy High: 4n
• Entropy Low: 3c
• Entropy None: Very glitched

BizHawk 2.9.1, BSNES

• "Randomized" Initial State: 4n
• Not "Randomized" Initial State: 4c

SNES9X v1.60: 4c aresv115: Actually random! At least 2c, 2n, 3c, 4n are possible.

Entropy None was useful for spawning with 0 delay. Proved a max 8 items are possible with a max of 4 cash and 4 nitros.

Soft resetting actually gives different starting RNG because it doesn't reset the RAM for it to be reinitialized by emulator. Should get same behavior on every emulator and real console. Whatever values the 5 bits were at on console reset is what they start at. You could mod the savestate bits to have a set of all 32 possible starts.

In this process I learned that nitroing at the start does not preserve the starting RNG. It instead advances it a consistent amount. Could memory watch and determine. Also memory watch to determine the RNG spread of not nitroing at the start. Watch driver movement and know what pattern you're on. Top Speed first purchase potential.

So what's the dream? Check out z_con's luck with grey driver getting blocked the whole race. Not manipulatable in that case but we already got delaying to third lap with minimal effort.

Have you seen grey driving backwards? I've seen it a few times on the last lap so didn't seem manipulatable. Then I saw d0re's run. Nitros at start but not immediately and gets a slow pickup at 6 seconds. First lap...grey drives backwards and finishes last! Grey also decides to u-turn on the sixth race at 5:45 and finishes third.

Trick is, the run is from 2015 and no SNES emulator version from this era is accurate enough to be allowed today. As in, ZSNES or pre-SNES9X v1.54. Still, I think first lap backwards is possible today. We can have the RNG arbitrarily start where we want, monitor it when grey decides to lose and reverse engineer the subroutine.

I like @Sircunch's idea too. It's realistic to how people play and fair to real console. I was interested in IL because I had no chance of beating @master_of_hilarity's insane luck + skill. I don't want to race for 2nd. I don't feel strongly that IL has to exist though.

I spent an hour in looking for good fullgame starts because emulator can amusingly begin off savestate. Check this ZIP out with 4 savestates for BizHawk. The "fff" was just my convention to keep track of saves. My goal was to win with only buying 1 nitro at the start for $90k leftover:

• fff: At the shop with nothing bought.
• fff start lap 1 pressure: I collect the nitro and grey car pressures to pass at the end of lap 1.
• fff start lap 3 pressure: I collect the nitro perhaps 1 to 6 frames earlier and the grey car only pressures to pass at the start of lap 3. This extends the fixed RNG since blue driver got 2 nitros. There's a guaranteed 3 nitro pickup since it spawns before grey driver calls the RNG to interact.
• fff start lap 3 win: Victory with 2 nitros to spare! I end up with an extra $10k due to a bug that's new to me. Someone else got a cash bag just before I finished that was awarded me instead. I should have $240k at the next shop but have $250k. Can see "$10000" under Fandango sign.
• The strat: Use 1 nitro at the start, try to get the nitro pickup on the same frame. Easy to tell if you succeeded from blue getting 2 nitros and grey staying behind.

I found I could drive under the first nitro pickup on the first lap to continue the fixed RNG. IL can easily use 3-4 nitros to pass before it spawns to maintain RNG. I also tried collecting it since that's preferable for fullgame and then the frame you collect matters...greatly as shown.

By the way, when I load a savestate, I can't have the gas pressed down. BizHawk treats it as not pressed, requiring me to let go and repress. Mo technology, mo problems.

I think it's premature to say RNG for maxed stats doesn't matter when no one had studied it a week ago or competed for IL. Maxed stats aren't limited to 12 nitros when I want at least 4 per lap on the first race.

If we give 99 nitros to avoid a need to collect, the RNG still determines the initial 3 car pileup. Can find a pattern with no items you have to dodge to maintain racer movements, which I think matters when you go to lap yellow or blue.

If we say RNG matters, could have 1 ROM patch for each track to be the first race. Also lets real console compete on flash cart. Keyboard shortcut a hard reset each attempt and enter initials, which is faster and less dumb than completing the race before.

Check out 38.2 on first track with 10 starting nitros and no upgrades. I maniped the 2 nitro pickup and got lucky on 3 and 4 nitro pickups with fast enough spawn times.

I looked into RNG manip more. Starting from save state is a ridiculous advantage for your first race and we're talking 13 completed in a run. You get the same starting racer AI with same item locations in the same order and approximately the same times.

Whether it's cash or nitro and the exact time the next item appears is based on:

• If the RNG has advanced from other racers being near you and interacting. You can defeat this at the start by nitroing ahead and collect a predetermined cash or nitro that appears at the same time each attempt. Until a driver catches up and your interaction isn't identical. Next item will still be in the same location but could be cash or nitro and the exact time is changed.
• The time you or another racer collects the previous item. While that seems obvious given only one item exists at a time, it's not simple addition. The next appearance can vary by several in-game if the last one is collected a few frames apart. Again, next item is still in the same location.

Good news is the following race would take extreme luck or skill to be on the same RNG position. All items would have to be collected on or near the same frame and your consistency of how you drive around the others would need to match. Other racers collect items on same frame if avoided but you can't avoid some item locations and you're losing time or money by dodging.

Of course, the RNG system limits IL single race randomness when you save state on the shop where the race RNG has already been fixed. Unless you save state from the shop of the race before or earlier. Or we do want a save state pack with same race RNG + maxed stats for equal playing field, or both options?

I think even with no manip, the optimal start is upgrading Top Speed before the first race. Finish second and collect enough items with no nitro use to make up for the lost $50k. Grey racer finishes 4 to 5 in-game seconds faster thanks your Top Speed purchase.

You mean just the final race of the run?

Yes, the race(s) you're submitting for IL. Finishing second or third in other races wouldn't disqualify. But requiring first in every race is interesting.

Are we expecting runners to submit their entire run of the game up to that IL attempt to show it wasn't savestated?

Yes. Alternative: Separate console and emulator leaderboards in IL. Or have no IL.

wouldn't it just devolve into who's savestate had the perfect RNG setup for enemy cars to be encountered at the perfect places/times, and less so about the driving skill?

Yes. One optimal pattern per race that could be in a save state pack. I think we all agree that evolution/devolution takes the spirit out of the game but maybe someone likes the idea of being able to match a TAS.

I'm not very creative. I don't know the elegant solution.

So the fastest fandango time would just be someone who saved all their nitros up and waited to use them all on this one particular track at that moment for an IL?

About what I would do but I don't think that's a bad thing. An alternative and shorter way to play a 15 minute game gives variety. Maximizing nitro use is skillful.

Save states would be allowed since you are just competing for a specific time on a specific track.

• The problem is being allowed to start from save state, which is banned in all competitive games. I like to play on real console and cart while that's still possible. I can't repeatedly reload the 2nd or whatever stage to increase attempts per hour.
• Save states also rig the RNG of racer movement and turbo drops. This is cheating. I save stated at the trophy stand after the first race on 2 attempts. The second race's RNG has just 2 possibilities: Whether I buy Top Speed or not. First attempt, staying idle, grey car finishes in 46.0 if I buy it and 48.6 if I don't. Second attempt, it's 45.8 and 48.1, respectively. Each of these 4 times has a distinct nitro/cash drop pattern and I want cash if I'm speedrunning the next level.
• You can RNG manip on real console as well but there's no shortcut.

Besides requiring to start from power on or reset, I request having to finish in first for an individual level. Would look bad to see grey car turbo ahead but still be a record. This won't be much upkeep unless a big Tuber makes a speedrun history video.

I looked at what other RPGs are doing on SNES. Some combine English and Japanese, some separate and Final Fantasy III/VI combines for some categories but not others. Let's wait 1 week for additional responses and split if there's no pushback.

Nintendo Power included with original SFC won't get played. People might run it included with English for text advantage but that would be criticized by active runners. For rules simplicity, let's put it with SFC. Can revisit in the future.

There is no difficulty difference for Fireseal and Krayzar's and bizarreluck's runs on SFC would still be #1 and #2 if done in English. Seems fine then keeping Fireseal together. If I ever submit Japanese Fireseal and someone doesn't like it ranked with English then that can be discussed. Both versions have instant text display so maybe the text difference is less significant than the PS2 model loading times.

I feel like there are too many awards. The prestige of each becomes lower like a Grammy. Maybe next time remove the last 4 awards or at least replace Top Dog with Top Cat. This is the internet we're talking about. But good choices and ideas for the first 6. I don't like the concept of non-video games allowed here but I much prefer seeing baking over speed reading.

I like how you beat your English WR on the first day you played in Japanese. The 1-2 level advantage gives more routing flexibility and less risk. Each extra STR or INT point on Opinion Leader is 1-2 points in Tarot card damage.

90+ seconds is larger than it seems considering most of any run is spent idling. Can save 30-60 seconds in Any% by never drawing a card at liberation and never pulling buried treasure but you pay for it in Joker cost and missed opportunity. We also didn't have as much competition a year ago. I was originally neutral on separation.

I like separation if we don't separate Fireseal where there is no difference in leveling or routing and the WR odds are steep. I recorded a few hundred Playstation attempts with no success.

If Fireseal is reclassified as a Level versus Miscellaneous Category then SNES/SFC could stay combined there. Sarjem Island as a Level wouldn't show up anymore for SNES/SFC/Playstation.

The (Japanese) Nintendo Power SFC version has the same nerfed EXP gain as English SNES. I would group that with English. It was released on flash cart and flash carts have always been allowed. Edit or do we ban it for being the least accessible and most obscure?

Cool, thanks for adding the run and explaining! I definitely don't think the run should be rejected. Just maybe adding time to it. That's a significant advantage getting 1 free round out of 16.

You convinced me to leave the glitch situation alone for now. It's not some unbeatable time. I'm interested in arcade but if I succeed then I may as well play on Saturn. If the free ring out can also happen on Remix then it's more fair. Maybe with enough recorded runs we can spot a difference for when the game will glitch or not.

I've been making gradual progress in RNG research. One way of maniping a Dream Crown or Trade Ticket is on Galf's stage that 100% and Devil paths play. There's a 4 Phantom + Sorcerer unit that the Sun card leaves the Sorcerer weak enough that my level 5 Canopus finishes off with Thunder and moves next. Then there's the 4 Wraith + Lich unit that, uniquely, is completely wiped out by Sun.

Obviously, this strategy dodges the half of the problem of maniping a neutral encounter. I recorded a Sharom Border test run on Saturn where I did pull a neutral Hawk Man. No item but the EXP gets the Opinion Leader a level up at Usar. Can see how slow this approach is:

Meanwhile, on Galf's stage, a level 16 Opinion Leader would level up from either Sun kill or get to 18 from both. This makes the approach low investment and you can deploy another unit to do it. Quite a bit more enemy units to work with using Sun + Death combo or defeating a 2nd or 3rd stage unit with Death, but either approach loses more time.

Of the 32 item pool, I wonder if Crystal Ball last in the list isn't obtainable in SNES, similar to how Nue's Shield isn't. 1 in 31 for the first 8 items, which includes Dream Crown and Undead Staff, I'll take it. I hacked in 200 Summoning Chimes to generate neutral encounters to get "the feel" of what I'm dealing with. The shortest amount of encounters between items was 2 and the longest 28. The theoretical 1 in 10 odds seem correct.

If there are 2 desirable items for 1 in ~16 odds, 1 in 160 odds is reasonable to manip. Summoning Charm sells for 25k goth or can sub as a Fool card with chance of failure. Blood Kiss sells for 31k, a Promotion is decent since it gives the full stat range on level up and AGI Potion on Debonair/Tristan has a 5-10 point spread. Kind of like raising accuracy 5%. Maybe 1 in 80 of a good outcome.

I tested Tarot cards for one Skeleton encounter I got an item on. Damage ones besides Judgment didn't change anything, proving the game checks if damage is possible and doesn't waste calling the RNG. I saw before Tower doesn't either if every target is Low/High Sky. Sun didn't change the item, implying item is rolled for once the unit is defeated. Judgment prevented obtaining any item, as did using a Joker. Not surprising but does mean we can advance the RNG, probably making any consistent starting frame from reset viable.

PlayStation would be perfect since the 5 class icons seem to uniquely determine what frame you loaded on but its RNG advances even on the item and battle menus and the time spent on text boxes.

Speaking of neutral encounters, I found the mechanics in my Sega Saturn official strategy guide. Plus other mechanics but all have been seen on Japanese websites. Main point is there was much less reverse engineering done in Ogre Battle than I thought. We're still in the Wild West.

I have Remix on Saturn and the arcade on Astro City Mini to finally compete with Janet.

• @K3ftc calls out the glitch at 4:32 on his Dural run submission that's part of the same Beat the Game run. He definitely didn't do anything wrong. The mod who approved it I guess was Janet who disappeared and never played on Saturn. Is this glitch known and repeatable? How can I do it? If it's some very rare "act of god" that just happens on real console + disc, I think the other round time vs Akira should count double.
• Janet's Arcade Beat the Game run also beat Dural. Can @Molotok or @Tenka submit it again for Beat Dural? Mods can submit runs on behalf of other players. I want this competition as the only person who's interested and able to play in the category. Not asking to submit to Beat Kage. Seems like a very lolzy category but that's a separate discussion.

Edit: I fixed my math error. Janet's Timing for Beat Dural is 1'56"26 Beat the Game +

1. 30 - 10.03 = 19.97
2. 30 - 4.16 = 25.84
3. 30 - 10.0 = 20.00
4. 30 - 18.70 = 11.30
5. 30 - 4.30 = 25.70
6. 19.97 + 25.84 + 20.00 + 11.30 + 25.70 = 1m 42s 81ms. Total time is 3m 39s 07ms. The Dural calculator under Resources agrees:

The moderators can request Super Dodge Ball Advance be added and become the moderators due to the clause of being another game in the same series. Commonly done in fighting and puzzle games.

I don't particularly like the same people modding multiple game in a series that they may or may not be intimately familiar with. Been a problem for me in obscure games that need rules changes. Nice to see they haven't requested it.

If you request a new game to mod, such as this one, you are required to add at least 2 public social media links to your profile. Can see for instance I have Twitch, YouTube and Discord linked. Also helps to have a run uploaded ready to submit.

Super Dodge Ball Advance would definitely be approved.

Yeah, that must be nice to make the rules, self-verify your runs and combine every platform and emulator so you have the greatest possible advantage by playing on NTSC Saturn, PAL Saturn, PC and Arcade emulator. SEGARALLY deleting their profile and not responding doesn't look good either.

Settling on North American Saturn emulator for Sega Car Touring Championship, that must have taken serious research. All platforms combined in that game too.

I have a Japanese Saturn, guess I I'm not getting any WRs.

@Lonne, can we get a separation of some sort? Why wouldn't this have been done by now?

I think it's amazing that PAL is combined with NTSC. Racing games are easier to play in 50 Hz. Also amazing that emulators are combined with real console. Even more amazing that all platforms are combined. There needs to be a split of some sort.