Bizhawk is the gold standard. Recent versions of SNES 9x and BSNES are also picks you can't go wrong with - anything with a high amount of accuracy.

I'd assume any officially released version of the game is fine. Bear in mind that PS1 and Saturn versions are likely much slower due to loads, though both might allow for RNG manipulation within stages.

I'm honestly not sure about the Nintendo Power Kiosk version of this game though. The one with in level saves on the SFC/SNES. It is an official release, but it's very obscure. I've been toying with it myself in order to manipulate RNG within stages.

Hey no prob LoZ! I understand. It's been pretty crazy in general these last few months.

Makes perfect sense!

The C64 had a few hard drive solutions, but they're exceedingly rare to find in working condition, and few games supported them. So unlike DOS, playing from floppy and floppy emulating flash carts are the only feasible options for most using real hardware/accurate emulation. There are some games that actually have problems with the drive timing due to the specific bootloader they use, so accurate read and write timings are actually necessary to running them at all. Castlevania C64 doesn't appear to be one of them luckily, so I think load-less emulation and flash carts aren't much of a concern with the edited files from that perspective. I use True Drive emulation and Drive Sounds in my runs for no reason other than I miss the sound of C64 reads. =P

One thing to watch out for is that most of the hacked versions, like the really good one disk "Nostalgia" or "NOS" release have trainers built in. In the opening screens for that one you can verify that everything has been turned off though, and since folks should be hard reseting it shouldn't be much of an issue to use that version. I'm not sure if any of the hacks force you to play with trainer cheats though, as there's at least a dozen different hacked versions.

Please, no rush on the C64 runs. Take your time!

No rush. I love the C64, but there doesn't seem to be much of a speedrunning scene for it, so not a lot to go on in terms of precedence.

Yes, just double checked. Only the PS3/X360 and PS2 should be able to access them. It's likely why no one plays them.

If I'm understanding the question correctly, the Turbo settings should be attached to the ability to toggle free select in this one if memory serves. Some of the arcade perfect releases like the PS3 and PS2 collections have the ability to set Turbo higher than a real arcade cab or MAME/FBA based emulator has options for. Since they're technically still arcade perfect, the option was added to the board.

So there's a lot of "hacked" versions of Castlevania on the C64. The vast majority of them are not functionally different than the main game, and in many respects are much better versions.

For anyone unfamiliar, the C64's floppy drive was essentially a computer in itself, and the default program to boot programs and disks was... Well... Not good. It was possible to override the standard bootloading program and by the time that Castlevania was put out on C64, this was luckily common practice.

This ability to modify the loading process was ALSO unfortunately used to write very complicated and unnecessary copy protection.

While Castlevania for C64's copyprotection wasn't as horrible and damaging to it's load times as other contemporaries, it did make it so the game could not fit on a single side of the floppy, and had to be "flipped" to load the opening screens and credits, restart levels, and save.

That said the loader they used leaves much to be desired, and is rather painfully slow in places. You can start the game without putting in disk side 1, and while the perfect deathless run would never need to switch disks due to this, the problem remains that newer hacked versions with different loaders will always be faster.

It's also of note that it's possible to edit the game files themselves to store them on one larger disk or a flash cartridge like solution. I was a little too young for much of this, but editing games and bootloading hacks were a huge part of the Commodore scene.

If things became competitive loads would probably be removed, but what are folks feelings on using hacked/edited disks?

Certainly makes the game less painful to play and might draw more folks in if the single disk / no copy protection hacks and edits are allowed, but I personally don't care much either way.

Hey there!

Yes any official release is valid. 1.4x started life as a collection of fan fixes, but it was endorsed and bundled by the publishers on GoG and Steam, so it's official!

Just fair warning though, there's some evidence that parts of the patches slowed things down for stability, so it may not be the fastest version in a lot of ways.

I don't know for sure if it's that or the DOSBOX settings you're experiencing though. For some games, max has the effect of slowing things down. You may want to try toggling the speed with DOSBOX hot keys to fine tune it.

I personally use an Windows XP virtual machine and my original copies v1.2 and v1.31 depending on what I'm going for. I think one of my runs is in the GoG version.

I recently set up a C64 emulator for other projects and decided to take the C64 version of Battle Chess for a spin. I grew up with this version (on an NTSC Commodore) long before I played the others and remembered not liking it as much, but from a speedrunning perspective it's actually pretty interesting. Got me wondering if any of the other versions are significantly different.

I figured I'd start a thread and detail some basic observations and speedruns of each version I play around with. Not really sure every version of this game merits tabs on the board, but please feel free to chime in and share observations and runs on obscure platforms! I might eventually try to run them all!

Commodore 64

1. They fixed the "Novice and Level 1 difficulties are functionally swapped" bug. Makes sense as this came out a year later than DOS/Amiga.

2. There appears to be a much reduced opening move set for the CPU compared to DOS and Amiga, which occasionally leads to very bizarre behavior in the early game. Only 6 difficulty levels as well. Makes sense considering what they had to work with.

3. The delaying tactics to improve the CPU variability work in this one like DOS and Amiga. It seems like cursor location matters much more in this one though. Overshooting your cursor before selecting often leads to way more favorable CPU moves. They likely didn't have as much to work with for RNG on C64.

4. Pressing the fire button like in the NES version to try to interrupt the CPU seemingly does nothing. There is an explicitly designed "force move" button called out in the manual that changes the CPU behavior, but that feels like straight up cheating.

5. The NES graphics are slightly more clean, detailed, and colorful versions of the C64's it looks like. Pawns are just plain weird and different though. Definitely see why they did cut scenes on NES - the battle part of the game is underwhelming.

6. I remembered the game being slightly unstable on real hardware, and that seems to hold true for modern emulation as well. Guess my C64 wasn't a dud! VICE (C64SC) is your best bet to run it with default settings and "True Drive Emulation". Using a C128 in 64 mode might reduce load times a little but it's not as stable and seems to lag more between moves. Maybe I have some bad settings though, so your mileage may vary.

Take a read through this thread if you're interested in USB 3 cards:

https://www.speedrun.com/Streaming_Recording_Equipment/thread/havyr

TL;DR: I recently got an Elgato HD60S refurbished for around 70$ and it's been treating me well so far.

After poking around with it on MAME - it does seem pretty accurate, but I guess that's a pretty low bar these days. It's a shame about the settings and button mapping not being accessible though.

Probably demons.

...well that and some hand and physical issues. I've always enjoyed watching and speedrunning non-competitively, but I took up actively competing and posting videos in part because I felt I didn't have much to lose if I was terrible any more. Without speedrunning, I probably wouldn't be gaming at all.

That said, certainly doesn't help to try to force things. Wait for inspiration. Nothing and no one is telling you you have to speedrun.

When you think about trying to speedrun again, what would you say prevents you from doing it? What isn't as appealing anymore? I think that's what you need to determine.

Ooh neat! I'd heard that was a thing but I've never tested it myself! I'll have to try it out if I ever do a non-tutorial run!

The US versions of RPGMaker and Xenogears can be used to perform the save glitch as well. I used to use RPGMaker (US) to do save hacking, a process similar to what's done here, so I was already familiar with the general premise. After seeing Pingval's video, I decided to check out how similar the JP and US disks really were and was surprised to find them so structurally similar. Figured it might work, and sure enough it did.

Here's a tutorial video on how this works:

Sorry for my weaker than usual voice, not having a great voice day. >_<

I recently went through figuring out HDMI capture for the switch and ended up buying a refurbished Elgato HD60s for around 70$.

Take a read though this thread for other recommendations and my own observations with it: https://www.speedrun.com/streaming_recording_equipment/thread/havyr

I totally forgot the Origin version existed. I have it (don't ask why >_< ) and can confirm as of last night that it's essentially the same files with the same DOSBOX settings. That said, at one point, Origin and GoG could have had way different settings or files due to how updates work on these platforms. Don't really think it's worth controlling what versions people use at this point anyways (or ever really) though.

Just to add, sometimes games separate OPL and emulation from actual disk runs just in case. Worst case scenario, maybe the game could do this if there's still a minor time difference.

Hi @Lizstar and other SimCity 2000 runners!

@beerfullofbelly and I were taking a look at some mechanics tonight and discovered/tested a few things.

I've been loading up a Scenario map to move the river location to the coast in my runs. This doesn't really save any time, it just offloads the RNG to whatever mountains and lakes would normally generate in place of the river. On average it allows for a few more good runs per map load and ensures that the River doesn't screw ya 10 times in a row and you throw your mouse at the wall.

However, editing the map ratios on the edit screen ALSO similarly affects the generation RNG. Using this, you can edit the map to be nearly flat and have almost no water. As @beerfullofbelly put it, this kinda feels dirty to do. I personally agree and kinda like the minimizing clearing time to building time pressure this puts on a run. It's pretty easy to see when this is done as the ratios are displayed on map generation, luckily, so assuming we don't want people doing this, it would be relatively easy to spot an invalid run. This might be appropriate for Glitched runs though, as it seems the intent is more "anything goes"!

Another thing of note is that most if not all of the DOS runners are using GoG.com version of SimCity 2000 DOS. This version sets the DOSBOX emulation to 20k CPU cycles fixed. Setting this game to MAX cycles makes things run much slower, but there is actually a sweet spot at around 300k cycles that doubles the game speed in certain areas. I kinda feel like this goes against the spirit of the current meta. No one likes to win simply due to load times. Luckily, again, this is very easy to spot. A normal series of months takes about 12.25 to 12.5 seconds to pass (minus any popup menus) and it's very noticeable if you time the months. Again, maybe something for the glitched category.

Another concern to this is the fact that a real machine running DOS could potentially be as fast or faster than 300k cycles DOSBOX. No idea how to handle that one. Maybe it just isn't feasible to limit things by DOSBOX cycle settings.

So I guess we're just wondering what everyone thinks on the legality of the following:

• Moving the river by loading a scenario.
• Use of Map Editor to edit water/land RNG.
• Use of CPU cycles other than the GoG.com defaults.
• Runs on actual DOS that are faster than the emulated runs.

For the record, I wouldn't mind if moving the river is banned. I'd be happy to remove my runs and redo them! Anything that gets me playing more SimCity 2000 is fine by me!

If the OBS logs aren't very specific, it can't hurt to check your disks out too. The one time I had something similar happen, it ended up being a bad disk that I was recording to. Something like Crystal Disk Info would be able to help you figure that out.

Usually it's done by screen or level transitions, or major game event milestones. The rules for the game category will usually highlight where to start and end.

There's no single correct way to time them. Do whatever works for you.

Ah okay, no worries! You got the point across. I apologize, didn't mean to draw attention to that.

Generally speaking, many fighting games default to the hardest settings for runs because it makes things more interesting. But hey, 10 minutes is actually really impressive in this one! Nothing wrong with that.

Like I said though, it might be interesting to see an any% setting. Let's see what @peck324 says.