[center]LOCKED - Refer to bottom post for information as to why[/center]
Original post:
(I'll format this better when I can be bothered)
Right, so after having done a run PSP to test the fucking thing out, the results are in and they're fairly interesting.
The thing saves 2:28 over PS2. That means, if you own a PSP then go for the real time WR right now by all means. Mentioned test run can be viewed here. It's a horrid run though, IGT adding together to 38:27.92, where my PS2 PB's IGT is 35:57.30. Yet it's somehow still only 2 seconds behind, go figure. Oh, and it's 50:49.045 in real-time, I fuck up the timer right after Roo so it ends up looking a bit fake but who CARES.
What it also means is that this game not only becomes far less accessible, but also far less competitive.
Removing loads allows for most versions to be atleast viable. The only thing separating them would be lag, but this is ALREADY the case, so not much changes. All this does is eliminate most console differences, not console model differences. (This is the option I'm most fond of.)
So, there are a few ways we can go on about this. The primary objective is to kill differences between consoles. Meaning, I can accurately time loads out from runs that are 51 or under and then use an average load time for the rest of the boards.
Another option is to start timing with IGT (+ hub world movement). This is really dangerous ground 'cause it means restarts will have to be checked so people don't cheese IGT. That's also an incredibly arbitrary thing to enforce so, ha. (And timing hub world movement on top of IGT is just REALLY SILLY.)
A third option is to "equalize" the versions. We have good enough data to be able to convert different PS2 models and consoles to have faster or slower times. The standard should be comparing to PS1 and then editing time accordingly. That being said no one does PS1 runs so PS2 will probably end up being the standard and PS1 just gets time removed off of it. (PSP obviously having time added.)
Now because of loads being removed, emulator inaccuracies shine even brighter as a lot of them tend to lag less than consoles do. As in, half the boards are nuked. Gone. BizHawk probably has something good going for it now, so people could run on that. But previous emulator runs would gain a significant advantage over actual consoles in not having any real lag to speak of.
In conclusion, I am very determined to remove loads from this game, especially so because SRcom allows you to sort runs by Real Time if you so desire, so it's not entirely stupid. Give your two cents on how the loads should be removed and I'll get going with it. Also complain about how your emulator PB is now gone, I guess. Someone more knowledgeable about current day emulators also pitch in on their accuracy yes yes.
And to add, misc. categories aren't getting any of this treatment from my end 'cause all the categories there are bad jokes.
Would you be able to get the sum of the load times for each console, and use that as a standard to subtract time from the RTA for each run depending on which console is submitted? Or are loading times not particularly consistent for something like that to work?
"Now because of loads being removed, emulator inaccuracies shine even brighter as a lot of them tend to lag less than consoles do"
What sort of lags do you mean exactly ?
"And to add, misc. categories aren't getting any of this treatment from my end 'cause all the categories there are bad jokes."
Thanks, I appreciate it Kappa
Yeah, that's pretty much what I'm going for. The thing is that different models even for the same console have load differences though, so people would need to specify whether they're on 30k or 75k and stuff like that. I don't want to add ugly variables to the boards though, so I'll probably toss a mention of it to the rules. If there's no mention I'll just use the shortest loads, probably. (Meaning, you might lose time in loads, but it still balances PSP out for example. However, if they're on the fastest version and I'd cut more than need be, then that's just fake timesave.) 51s and other top time material (Deceptive SDS/PS1) get special treatment though, but that's just for <the rest of the boards>, 'cause I'm really not timing 100+ runs. I'll work it out but that's pretty much how it works over at CNK, 'cept better.
On most slims, PS1, PSP and whatever, tracks actually have notable lag in them. TA and OS are very nice examples of it. Also stuff like HAS start, Blizzard entry, hub loads and such. Emulators usually don't have any of that lag at all. Pretty much this means they're just saving time everywhere there should be proper lag but isn't.
Update:
PSP literally runs the game faster, so it's essentially just a shit VC. Moderators don't seem to agree on having a "VC" category, much less banning PSN releases. Discuss this in either the Discord server or make another thread entirely. Loadless doesn't achieve its goal and ultimately ends up troublesome to put up on the boards for no real reason. I'll still keep the data around incase someone wants to do something with it.
gg
