Username:
B
I
U
S
"
url
img
#
code
sup
sub
font
size
color
smiley
embarassed
thumbsup
happy
Huh?
Angry
Roll Eyes
Undecided
Lips Sealed
Kiss
Cry
Grin
Wink
Tongue
Shocked
Cheesy
Smiley
Sad
1 page
--
--
List results:
Search options:
Use \ before commas in usernames
Game Page: Doesn't exist yet

God Hand (ntscus) (ps2) [Any %] [Segmented] [Easy]

Decision: Reject

Reason: A few areas were unoptimal (especially related to the strategies involved) and this led to significant time losses.

https://queue.speeddemosarchive.com/verificationfiles/1976/

This run will be available for a month. After that these link(s) will no longer work.
Thread title:  
Run Information

God Hand (ntscus) (ps2) [Any %] [Segmented] [Easy]

Verification Files

http://v.speeddemosarchive.com/GodHand-20161115/

Please refer to the Verification Guidelines before posting.

Please post your opinions about the run and be certain to conclude your post with a verdict (Accept/Reject). If you wish to remain anonymous, you can also send a pm with your reply to 'sdaverification' (please state clearly in that case which run you have verified). This is not a contest where the majority wins - Each verification will be judged on its content.
Doing this.
The run is honestly pretty sloppy, and does a bunch of things that are clearly not fast (taunting is a great example). I kept saying the square combo was terrible, but it continued all run long. If you want a perfect example of just how bad that combo is, look at the Demon Shannon fight (1:18:43 -> 1:19:43 in the video). Notice how the second hit doesn't even connect with her without unleashing.

Many, many missed opportunities for High Side Kick cancels too, which is the bread and butter of New Game runs. Absolutely no death animation cancels. No duck cancels for the unleashing animation either, though I guess that's a slightly less painful mistake overall. No Fist of Justice cancels on Azel struck me as odd, but I don't think it really cost that much time. Swapping out Charged Punch 1 in stage 2 was a huge mistake that destroyed a lot of his go fast options from stage 2 on.

Post-stage shopping menus were very slow. Depending on how this is timed, they definitely should've been in separate segments. Regardless, there was obvious confusion over where the moves were in the list, and pressing left or right to page up/down the menu was clearly missing.

On the bright side, there were a few spots that I thought were pretty well done: Iron Bridge Fashion Show (4-6), Electric Deathmatch (4-7), and Death Shudder (8-1). Unfortunately, those spots couldn't make up for the rest of the run.

I wish I didn't have to do this, but I'm voting reject on this one. It's a God Hand run and SDA badly needs one, but this isn't it.
Because the audio commentary is already done (and pretty long) I guess we'll keep all the loading stuff for the actual published run as well, if this gets accepted.

Molotov: To save ktwo the trouble, and to inform my own decision, can you point at some videos somewhere showing comparable fights (i.e. not NG+) where some bosses or fights are done faster with the techniques you mentioned? Perhaps in a SS NG run? Also have all those techniques been known about since when the runner started the run/before they got too far in to make it worth restarting? Here's the thread if anyone else wants a look. Looks like some of them you most certainly brought up in 2011 already, and had a mock segment showing considerably faster completion of stage 3 (18 seconds?). The proof is in the pudding as they say, so if such demonstrations exist and there's no negative splash effects on upcoming segments, there's certainly room for improvement. Are you able to give any kind of estimate on how much time overall could be saved with the improvements you're suggesting?

I'm surprised you're saying the menuing would have needed segmentation seeing as it's not quite that heavy in this game seemingly (not like some mad JRPGs), but missing fast menuing options is rough of course. The menus are all timed up until the point when there's no more inputs, so as the game is saving, and gameplay is timed from the first frame of control/visible. Thus it might be more efficient to do menuing in the same segment as done here. That having been said it seems this game has an in-game timer... The runner is saying it's not quite accurate?

If the runner agrees that the strategies are sub-optimal, would it be possible to patch the run up by redoing certain segments or do you think it's "beyond repair"? You can redo segments so long as you manipulate the same (or better) drops/stats/whatever and aren't getting impossible RNG in the process (the next segment giving you something you couldn't have gotten without splicing segments together).

Finally if you know some other runners who might be interested in joining the discussion, feel free to bring it up. The runner himself is, of course, welcome to comment on anything.
I feel bad for all the work that Redlimit did, and I do have respect for it, but I echo what Molotov is saying. Also, if you needed any NG comparison, mine definitely doesn't have optimal combos but the heavy slow hits for square combos have never been considered efficient without considerable manipulation. http://www.speedrun.com/run/kz5roeey

I'd say the run is beyond repair as literally none of stage 1 is correct unless it's been completely redone since the last time I saw it. Also redoing individual segments, I wouldn't agree with that because your resource count carries from stage to stage, if this were allowed in God Hand's case then you could, in theory, have a segmented run where you start all segments with the best possible moves, max roulettes, and full God Hand. It's also worth noting that this is barely faster than my SS PB, and I think that my SS PB in this category is shit.
As everyone has said, redoing individual segments really isn't realistic as too much carries over into each following segment (gold, tension bar, health, powerups, etc). Timing wise it doesn't cause issues if the in-game isn't used but redoing one would mean redoing everything afterwards as well - hence why the early portions of the run were not re-done once I was able to pick the run up again.

The run isn't without its faults, but it showcases what is possible and opens the door to other submissions in the future to beat it or have separate categories. That is what SDA is all about.

The commentary can be re-done if the loading is to be edited out, I was mainly getting it all done and encoded when I had a few free days before my crazy busy holiday work season - I just wanted it to be done with and submitted before I had no free time again to devote to anything.

I feel that the run from the second part of Stage 4 to the end is really tight (this is when I was able to pick the run up again) but it is the early stages that are more rough. I certainly welcome the feedback of everyone but I am not likely to ever have enough free time to devote to redoing the entire run again in the foreseeable or far future - hence why I submitted it in the first place "as-is".
Zanaso: Redoing segments is allowed only if you don't gain such advantages from having done so. We'd probably have to compare the pixels on the meters to make sure they're at least the same, but we could also take the verifiers' word (if they know the game like here) that it doesn't constitute cheating in each case. The timing would be problematic if it wasn't shown between each segment, in the save menu. I think we could just subtract seconds from the final IGT based on time saved in the in-between segments. I guess it's pretty tough trying to get the same or better everything that Redlimit listed.

The WR linked looks to be 1:23:25, whereas this is 1:07:53. Difference about 15 minutes. That's not "barely better" IMO if we're going with in-game time. However, I see three different times listed in the same entry on speedrun.com so I'm a bit confused. One of them is IGT + compensation for faster loading (loading times are included in the IGT??) and there's another one that looks to approximate real-time. So is it IGT for this game or what?

As I was saying earlier, there's some leeway in having sub-optimal strategies in segmented runs if better ones were found in the middle of it. That having been said, it looks like the other runners think you've made a misjudgment in the strategies from early on, as evidenced by looking at the thread. This would mean it's the runner's mistake, because even finding the best strategies in the first place, unless we're talking some really profound glitchy stuff, is part of the runner's duty. On the other hand the run doesn't look sloppy, at all, and sometimes that's all we get to go on with public-verified runs. Seeing as there's now two other runners verifying this, it's subject to the inevitably more rigorous private verification standards. It simply put IS easier to get a random run accepted than one for a popular game, no way to deny that, and nothing we can do. We still expect every run to be high-quality in routing and strats, we just often can't prove they're not with such games.

There's another thing about treating each run fairly: in some games that have gotten lots of attention (three runners is plenty for most games), the standards grow higher than what would be required for getting a run accepted. If the WR runner comes along, they more often than not can point out mistakes and might well have a better run under their belt, but we can't just wait forever: the record holder might never ever submit to SDA. The counterpoint to this is there's typically nothing preventing any runner from at least taking the best strategies despite execution tying in with it.

I think I'll abstain my vote a little longer.
Note about timing: The in-game timer at the end has long been known to be a horribly coded piece of crap. You can mash X while it's counting to get wildly different times from it. I seem to remember the post-stage timers being considered accurate, but I can't say for sure anymore. Real-time is probably the easiest way to go, since that's what single-segments use. I just did a super rough time on Redlimit's run (using real-time as if it was single-segment) and got 1:22:42 from it, for what that's worth. (I may have screwed up the spreadsheet though. Timing is hard.)

As for explaining my points with videos, that's honestly difficult. I used to actively run this game, but I don't have any real strategy videos at hand, so it's more of a "I just know" sort of thing that's hard to provide video examples for. I realize that weakens the argument, but oh well. (As for SDA standards, I'm aware.)
Segmented runs of this game are generally not done, for reasons which should be obvious to everyone at this point due to the insanely high RNG involved...so there isn't a lot to directly compare the run to. I stand by the run and it improves in quality throughout, culminating in fantastic Stages 7, 8 and 9. Most strategies for God Hand are designed around real time single segment running and I attempted to go in a different direction with this speed run. There is only one outright time loss (Shannon's Dance School boxes) but it comes at the cost of the rest of the RNG for that level lining up perfectly. The rest of everything is differences in strategy used and new techniques being found out during the production of the run. (in which some will be faster than others).

Another issue entirely is that there really aren't any runners of the game anymore (myself included). The world has moved on from pre-2010. I was aware of how much of a pain this would make verification but there isn't really anything for it - hence the mini-essay written notes and commentary to try to explain things further for viewers of the run.

Maybe the best way to describe the issue is this: If someone wants to showcase a faster segmented run then they are welcome to step up to the plate and make that happen. That is the entire philosophy behind SDA. If anyone wants to do single segment runs to show the differences to this one or to complement what SDA has then that is also always welcome.
The entire philosophy behind SDA, if I may be as bold as to attempt to outline it here, is to host high-quality A/V and gameplay -wise runs. It's not really stated anywhere it's to elicit competition, although the general standardization of rules is to facilitate that as well. The most helpful and constructive thing anyone can do in verification is objective comparisons between strategies, pointing out existing runs etc. What I'm not getting is anyone being able to show why they think as they do, and what's difficult for me will be difficult for ktwo making the call. I can see there's a ton of different things to consider and lots of different attacks to choose from, and seeing as you're running different categories, it's not surprising you can't all have the exact same expertise, but you can't all be right about everything either.

Redlimit: You can't both say "the run isn't without its faults" and "there is only one outright time loss". I'm sensing because of the effort you've put into this you're being more defensive of it than you would if it was someone else's run. It won't be for nothing even if it gets rejected as people can take all the good strats from it and make sure the next run isn't subject to the same criticism, if it's warranted. The RNG manipulation and overall quality of gameplay will be taken into account in judging the run, and those don't seem lacking in any way, it's just the fight strategies being questioned.

Indeed things may change when you can manipulate enemy patterns so the most reliable is no longer the fastest, but "the heavy slow hits for square combos have never been considered efficient without considerable manipulation", the comment by Zanaso, makes it seem he at least doesn't think that manipulation is present to justify using those combos? Can no-one really show a fight done faster using another strategy then? Molotov pointed at the Demon Shannon fight but you can't expect me to be able to see what you're saying just by looking at it. Can you not show me someone doing that fight faster then, using means that don't compromise some other fight later/earlier on?

Also I'd still like to hear about how much time is lost (in your opinion Molotov) to missing a High Side Kick cancel, death animation cancel etc. and in each menu? Are we talking 1 second each time? 0.1 second? What? How many times could they have been done? Can you think one more time about whether any of those techniques were not known about when Redlimit started recording?

Sadly I don't think we can take the stage times, even if they're accurate, and add RTA for the last stage because this leaves the shop gameplay unaccounted for. Redlimit: do you feel 1:22:42 is a good estimate of RTA (so to recount from first frame of gameplay to when the game starts saving) if IGT is unusable?
Yeah, the IGT is an absolute farce, when following how it counts with CE on emu, it is inconsistent at all times and on console it's no better.

Look, two people who are wildly familiar with this game have told you this segmented run isn't very good. All due respect to Redlimit, there's just a lot of things he doesn't know... though I will be borrowing that SHAOLIN BLAST dmg carryover on Azel phase 1 that I previously wasn't aware of. Past that, do whatever, and Red if you want to learn single segment runs I'll be willing to discuss routing and help with that.

Past all of that, do whatever, I don't see how it matters at the end of the day anyway.
Edit history:
Redlimit: 2016-12-21 05:56:25 pm
As far as timing goes I agree with Molotov and Zanaso - the in game time leaves a lot to be desired and was really only used to make it easy to identify segment starts and stops (file naming conventions, basically). I'm not an expert on frame counting in the slightest so we may need to get someone to do that to get an "official" time for this run.

Individual segments kept were simply because they were the fastest based on a separate timer - I would start the timer when pressing a button from "Press Start" and time as normal until the save was made.
Zanaso: look, both your being "wildly familiar" with it (and that extending to segmented runs) and everything else you two have brought up isn't apparent to me in any concrete way based on what you've written, hardly. You're just saying "it's so". Redlimit is just saying "it's not so". Who should I believe? The best way you could have helped is to show that concrete evidence, which is often something you can do putting some effort into it whether or NOT you know the game. That's not happening here. Molotov at least did a rough timing on the run (which was effectively missing completely).

Redlimit: Now that we know it's inaccurate, the timer will do his separate timing for sure, if this gets accepted. Because you didn't really comment on it, I'll assume 1:22:42 is a good-enough estimate of this run's time, which means Zanaso indeed looks to have a SS that's only about 7:30 minutes off that apparently wasn't put as much work into as other categories. I even found a faster run that Zanaso perhaps forgot about (or is it somehow incomparable?) over here. Looking at the first few areas, the SS...

saves 10 seconds with smarter strats up to entering the first building
15 more before the cutscene in the building
and maybe 1 more by the end of the 1st area (and has one more orb left even!)

2nd area up till the cage cutscene saves 2, but this is lost straight after

Next stage loses 5 seconds at first (killing an extra guy? and got blocked at first, saves a bit of time just by smashing through the doors instead of waiting them to open)
after cutscene up till curing the guy saves 6 for smarter strats
Boss fight room (with the Indians) saves 4 seconds to not farming for extra orbs and visibly more efficient combo attack
the part after entering the next building loses 8 seconds seemingly to controls, strat might have been faster though
room of many fights loses 20 seconds but that's with getting TWO demons: I think would have saved time otherwise
after leaving the room to the demon door: maybe saves 1 second

So what I'm getting from this is the early parts of the run really have bad strategies, but it's still a bit difficult to judge it because it's a game of enough complexity that unless you're like me (I've been working on a game for maybe 1½ years now without even attempting a single segment just to make sure I'm not missing anything), you usually won't have all the strats quite optimized. That having been said, I also feel like the runner was simply too focused on quantity of attempts (which shows for sure) instead of quality. I'm going to look at a later part of the run as well because everyone seems to agree the early parts are worse than later on.

(part where segmented run is at 1:09:38)
in fact this part (first jumping puzzles) is difficult to compare because of different menuing and having a different count of orbs to begin, skipping to the next, but making a note of the SS having one extra orb slot whether that lost time to get or not...
midgets fight: SS lost 20 seconds due to slower strategy because of wanting to save orbs, segmented run can ofc rely on orb drops
next part loses 40 seconds (but with one demon fight and using 2 less orbs): in fact the more orb-dependent the segmented run gets the more time it's supposed to save almost the whole time, so at least it does that here
boss fight: loses just 1 second despite the segmented run having spent 4 orbs on it... which it went out of the way to pick up earlier, so essentially saving time over it

next part (cigar statue): saves something like 7 seconds factoring in spending some in the tech menu
Shannon boss: loses roughly 25 seconds (saves 4 orbs)

(1:22:30 in segmented run) this boss is an example where the SS saves time because of the High Kick cancel technique Molotov mentioned, so despite being so far into the run, it's still not implemented here. Judging by the back-and-forth in the planning thread, this was a choice the runner made early on in the run (?) in stage 3. The boss itself: SS wins something like 14 seconds although switching to the high kick ability earlier factored in, maybe 9? If this technique is usable in all kinds of fights, it looks clearly faster overall. I know I can't see the big picture with all the shop strats and everything though.


Okay I think I'm done: despite this run having very good RNG manipulation and looking clean - except I think the movement can be slightly improved on -, it's the strategies themselves that probably were not looked into enough. The runner hasn't been able to show anything concrete on why their choices were better (and has been kinda dodgy and vague in general which won't help him in the future), I guess I've seen enough. If ktwo agrees with this, I think this'll be a textbook example of why no amount of RNG manipulation can save it if you're forgetting about so much in your planning.

weak reject
Decision posted.