Prince Of Persia Classic - December 13, 2013

Edit history:

SDAVerification: 2013-12-13 02:01:51 am

2013-11-29 02:15:14 pm

Game Page: Doesn't exist yet

Prince Of Persia Classic (Any %) (Single Segment)

Decision: Accept

Congratulations to 'Jaguar King'!

Thread title:

SDAVerification

2013-11-29 02:15:18 pm

Run Information

Prince Of Persia Classic (Any %) (Single Segment)

Verification Files

http://v.speeddemosarchive.com/popclassic/

Please refer to the Verification Guidelines before posting. Verifications are due by Dec. 13, 2013.

Please post your opinions about the run and be certain to conclude your post with a verdict (Accept/Reject). This is not a contest where the majority wins - I will judge each verification on its content. Please keep your verification brief unless you have a good reason otherwise.

After 2 weeks I will read all of the verifications and move this thread to the main verification board and post my verdict.

bangerra

2013-11-29 07:01:51 pm

A/V is fantastic (except for LQ, it looks a lot blurrier than it should, especially around moving objects)
No cheating as far as I can tell.
The run is exceptionally smooth with very few minor mistakes.
The route seems very well thought out and the execution is a joy to behold.
The few unskippable combat scenes could be potentially a bit faster, but they seem RNG ridden combined with clunky mechanics.
A shame there are no run notes though.

Overall a very solid run! ACCEPT

UraniumAnchor

2013-11-30 03:42:36 pm

Not a walrus

A/V is good, LQ is just... well, it's LQ, only so much you can do with trying to jam source footage like this into such a low bitrate. There's one audio drop during the ending but it's short and not during gameplay so no big deal.

According to the timer at the end the run is 12:46.02 unless my math is bad, so I'm not sure why real time (which isn't even right, 15:31 would put the ending somewhere in the Jaffar fight) was used for the StatID.

I'm much more familiar with the "true" classic version of this game which has very different controls and moves, but I have played this version and nothing jumped out at me as a planning error and there are very few execution mishaps outside of the RNG-laden fighting sequences.

I say accept, though I hope the issue with the StatID is fixable.

MrWalrus

2013-11-30 05:37:41 pm

I am a walrus

No cheating, A/V looks good.

Very nice routing and overall execution.

RNG is RNG and even that doesn't seem to cost much time.

ACCEPT

TheVampire

2013-12-01 09:39:29 am

A/V Good, no signs of cheating.

A very solid run of the game with a few minor mistakes that don't cost too much time in the end.

Another strong ACCEPT by me.

djcj

Edit history:

djcj: 2013-12-01 10:53:00 am

2013-12-01 10:50:01 am

The mode he had chosen in the menu is single-player - time attack.

A/V quality is nice (watched the HQ version). At 17:56 the sound is mute for a second, but it's the credits, so it doesn't matter.

There are only a very few minor mistakes:
3:30 He jumped a bit too far.
6:51 Got hit there by a flagstone.
9:03 I think he waited a bit too long there.
9:25 He jumped/ran a bit too far.

10:11 Why did he run back to this place?

UraniumAnchor already pointed it out, the time in the StatID seems to be incorrect. I would say manual timing starts at either ~1:11 or ~1:14 and it stops at ~16:58. Therefore the time would be somewhere around 15:44-15:47.
If calculating the time from the remaining time shown at the end, it would be 0:12:46.02 (according to my office suite's spreadsheet calculation).
If the runner doesn't have the original video source anymore he or someone else could try to remux the MP4 files and split off the first and last 5 seconds. That might work (depending on how the encoder set the key frames).

The run is very straight forward and fast, from what I could tell. I like how he often avoids fights. For me this is an accept.

PS: I love how the character always takes his time to draw back his sword, even if there's an enemy standing in front of him.

The mode he had chosen in the menu is single-player - time attack.

A/V quality is nice (watched the HQ version). At 17:56 the sound is mute for a second, but it's the credits, so it doesn't matter.

There are only a very few minor mistakes:
3:30 He jumped a bit too far.
6:51 Got hit there by a flagstone.
9:03 I think he waited a bit too long there.
9:25 He jumped/ran a bit too far.

10:11 Why did he run back to this place?

UraniumAnchor already pointed it out, the time in the StatID seems to be incorrect. I would say manual timing starts at either ~1:11 or ~1:14 and it stops at ~16:58. Therefore the time would be somewhere around 15:44-15:47.
If calculating the time from the remaining time shown at the end, it would be 0:12:46.02 (according to my office suite's spreadsheet calculation).
If the runner doesn't have the original video source anymore he or someone else could try to remux the MP4 files and split off the first and last 5 seconds. That [i]might[/i] work (depending on how the encoder set the key frames).

The run is very straight forward and fast, from what I could tell. I like how he often avoids fights. For me this is an accept.

PS: I love how the character always takes his time to draw back his sword, even if there's an enemy standing in front of him.

UraniumAnchor

2013-12-02 03:56:28 pm

Not a walrus

Quote from djcj:

10:11 Why did he run back to this place?

Dropping the tile on a switch that'll open a required door elsewhere.

AlecK47

2013-12-02 05:01:29 pm

A/V good, no cheating detected.

Pretty much everything seems to have been said already, and the run looks good to me. Accept

Uilnslcoap

2013-12-02 06:39:38 pm

F*ckin' sanity effects...

I've only played the PC and SNES versions of PoP, but this looks like a largely faithful remake (only they made the platforming a little more forgiving/fun).

The mistakes are mostly minor and scattered and I saw nothing to discount it on A/V or cheating; it's certainly worthy of SDA. Accept.

Efreeti

2013-12-03 03:58:36 am

Clear as a crisp spring morning!

A/V good other than LQ, no cheating evident

3:30 it looks like it's faster to jump over and catch the edge rather than slow down before the jump.
There are a few genuine mistakes but they cost 2 seconds each at most.

Circling around and avoiding fights is a genuine tactic even in the original version of the game, in non-speedrunning contexts.
The worst part of this version is how much less visceral the sound of the chomping blade obstacles are Tongue

I see no reason not to accept the run.

I time real time from first control to last control at 15:44.
However, I think as the game times itself that time is more apt to use. with 47:13.98 left of an hour that time is 12:46.02

SDAVerification

2013-12-13 02:02:36 am

Decision posted.

Koning Pinguin

2016-01-07 05:11:05 pm

Xbox Speedrunner

Quote from SDAVerification:

Decision posted.

He cheated.
Here’s why: at the end of lvl12 you can see a total time remaining of 47m14s10 (this is legit so far, the game gives you 60 minutes to complete it). Then he does lvl13 which takes him 1m20s40, however the total time remaining now states 47m13s98.. This is a time difference between lvl12 and lvl13 of 0m00s12. Then he does lvl14 in 1m15s32 and now the time remaining states again 47m13s98 just like it did at the end of lvl13. So now there is no time difference between lvl13 and lvl14. This is not possible if it was one connected run.

One could make the same vid by recording a ‘’time attack’’ trail up until lvl12 and then cut and paste lvl13 and lvl14 done in ‘’normal”. ‘’Time attack’’ and ‘’Normal” are 2 different modes in single player.

ktwo

Edit history:

ktwo: 2016-01-10 01:55:35 am

2016-01-08 08:12:16 am

First of all, thanks for pointing this out Koning Pinguin. But let's not jump to conclusions just yet. I've contacted Jaguar King and invited him to reply here has well. I also looked up another playthrough on the net (unfortunately, I could only find one of the time attack mode for this version of the game):

[Edit: updated the link]
If you go the time stamps 31:49, 33:20 and 33:23, you'll see that the total time passed has augmented correctly, but the remaining time has been messed up just like in JK's video. However, in JK's video, the total score has also been messed up, while it's been correctly added in the youtube playthrough.

Let's hope Jaguar King sees my message and can reply here. If anyone reading this has direct experience with the game, it would of course be interesting to hear what you have to say as well.

KoningPinguin

Edit history:

KoningPinguin: 2016-01-09 04:11:41 pm

2016-01-09 04:10:17 pm

@ktwo thanks for your fast reply. I see no remaining time issue in the video you posted here. Every time he dies in lvl9 and restarts it the timer says 39m13 remaining. This is normal, it doesn't work like the old pc version where the timer just keeps on counting no matter what. Also this gameplay was recorded in "normal" imo, not "time attack". Because in "time attack" you can't choose which lvl you want to play and one can see this person selecting different lvl's throughout the vid. In "normal" you can replay any lvl a hundred times and it will only count your fastest completion time of that lvl and carries that time to the end result.

I'm using a different profile btw because I can't login anymore with the old one, and the password recovery link doesn't work. Cheers

ktwo

2016-01-10 02:00:29 am

I had several youtube playthroughs open in different tabs and did the classic mistake of linking to the wrong one... I've updated the link now though. Can you check again and see if you agree that the playthrough in the updated link shows the same type of inconsistency as in the sda-run?

KoningPinguin

2016-01-13 02:44:07 pm

@ktwo Thanks again for your reply. You are right in this new link I can clearly see that it has the same countdown timer issue that Jaguar King had in his video. I stand corrected and take back my accusation of him cheating (sorry Jaguar King). He did not cheat, however imo it is fair to say that he did not do the run in 12m46 but in 15m18s62.
The thing is that I'm doing a any% speedrun of the game now and will be posting it for approval on this site. And it looks now that I'm just gonna beat that glitched time by only 1 or 2 sec. Do you agree that he did the run in 15m18?

IsraeliRD

Edit history:

IsraeliRD: 2016-01-13 03:11:56 pm

2016-01-13 03:01:52 pm

Dragon Power Supreme

... is this something that happens every time? is that a consistent issue or does the timing change every time?
If the latter, then we'll use real-time for this game and when an improvement is submitted, re-time the old run and compare to new one.
If the former (always same amount of time at the same points) then we can probably go with in-game timer quietly.

KoningPinguin

Edit history:

KoningPinguin: 2016-01-13 06:23:19 pm

2016-01-13 06:22:13 pm

@IsraeliRD Just did one playthrough of "Time Attack" and I didn't had that timer glitch. Nor do I see any inconsistencies on the time attack leaderboards where one can see the lvl times of other players. Based on this one run I can't provide any explanation for this countdown-glitch. And like ktwo stated the (point)score of one of these vids is correct unlike the other vid. I only play "Normal" mode so I don't have much experience with "Time Attack" btw.

IsraeliRD

2016-01-14 12:17:56 am

Dragon Power Supreme

Alright, we will us real-time on next submission. Thanks for bringing this up.

KoningPinguin

2016-01-14 07:22:44 am

Or just correct the description Of Jaguar King's vid to it's real completion time of 15m18s62. And use the real ingame timer of the game, in the old and new submissions. If you look at the vids you can see the real timer works perfectly, much more accurate then someone watching a vid with a stopwatch in his hand. It is the additional countdown-timer that apparently can be glitchy in "Time Attack".

If one would follow IsraeliRD's suggestion then the new submissions would be timed in a non perfect accurate way, and the old submission gets away with murder..
That doesn't make any sense to me.

IsraeliRD

Edit history:

IsraeliRD: 2016-01-14 04:11:28 pm

2016-01-14 04:10:28 pm

Dragon Power Supreme

I ended up counting it all up myself and the actual time is 15:21.62, so you missed 3 seconds there. I also went the extra step and checked how accurate the in-game timer is. It isn't. It seems to drop frames at random places, or adding at others.
For example the third level was reported as a 1:00.20 but really took 58 seconds real-time.
Also turns out the game's timer starts AFTER player control is gained (by 0.2 seconds or so), and ends when the character is halfway through the exit (post-player control, by half a second). Minor thing but just a thing I noted trying to figure out why timing is so weird.

Therefore saying he did the run as 15:21.62 is also incorrect.

KoningPinguin

2016-01-14 05:31:24 pm

@IsraeliRD 15m21s61 is the correct total time, unlike you I made a mistake while doing simple math.

I disagree on your lvl3 example, put his run and 3 of mine on that lvl in a video editor and the timing looks correct. How did you make the assumption that it took him 58s real time? He did it in 1m just like I did it in 52s in the 3 vids I compared it to. Willing to send you my vids, if you're interested. Lvl's like 1 and 7 start with an animation, maybe that is the reason why you suspect the in-game timer is incorrect. Can you give me one more example of the drop of frame rates at random places you say you've found?
Is there one lvl on the internet where the real in-game timer times 2 separate runs differently?

Imo 15m21s62 is the correct total time for his run.

IsraeliRD

Edit history:

IsraeliRD: 2016-01-15 01:40:39 am

IsraeliRD: 2016-01-15 01:40:19 am

IsraeliRD: 2016-01-15 01:37:57 am

IsraeliRD: 2016-01-15 01:37:03 am

IsraeliRD: 2016-01-15 01:36:55 am

2016-01-15 01:28:36 am

Dragon Power Supreme

For player control, on the current run on the site (level 2 for example) if you look at the countdown clock from the get go:
Level 1 was 57.00.00 flat. That means the clock starts at 59:03.00 and should count down instantly, right?
Level 2 starts at 2:08.445 but only at 2:08.845 the clock finally goes down to 59:02, and JK moves left a few tiles during this time too.
At level 3, start is frame 11529 [3:12.342] (mind you, ingame clock says it has 58:00.57 remaining) but only at 11585 [3:13.276] it switches to 57:59, and JK is so far to the right side that 3 frames later the screen transitions.

For the drop frames:
Level 2, frame 7723 is where the timer hit 59:02 (this is 2:08.845 on the video), and at 7781 it goes to 59:01. This is 59 frames.
Level 2, frame 7781 is where the timer hit 59:01, and 7838 is where it hits 59:00. This is 58 frames.
The next second is 59 frames, and so on.
Mind you, this keeps happening on levels 1 and 3 and I'd image pretty much every other level, but I did not test further than level 3 to see if occurrences of two seconds in a row where there are 58 or less frames consecutively.

For timing level 2:
Start frame is 7699 [2:08.445] (since a frame before the was 'Press A to continue'), and clock has 59:03.
End frame is 11289 [3:08.338] (camera starts zooming into the exit door) and clock has 58:01.
At 59.94fps, this is 0:59.89 seconds real-time. In-game clock however claims the level time to be 1:02.43.

For end of player control, take level 3:
At 14980 [4:09.916], clock turns to 57:00. At 14990 [4:10.083] the game begins zooming into the exit door. However, the game claims the remaining time is 57:00.37, so that means that part of the zoom in process (which has no player control) is actually taken into account before the timer actually stops. If we assume game time to be 100% correct no frame-drops etc., then end of level/control is frame 15018 [4:10.550]. This however goes into territory I'm not going to pour too much energy into, hence this paragraph is in smaller size text.

I could keep going but the above is sufficient proof to show why this ingame timer is incorrect.

For player control, on the current run on the site (level 2 for example) if you look at the countdown clock from the get go:
Level 1 was 57.00.00 flat. That means the clock starts at 59:03.00 and should count down instantly, right?
Level 2 starts at 2:08.445 but only at 2:08.845 the clock finally goes down to 59:02, and JK moves left a few tiles during this time too.
At level 3, start is frame 11529 [3:12.342] (mind you, ingame clock says it has 58:00.57 remaining) but only at 11585 [3:13.276] it switches to 57:59, and JK is so far to the right side that 3 frames later the screen transitions.

For the drop frames: 
Level 2, frame 7723 is where the timer hit 59:02 (this is 2:08.845 on the video), and at 7781 it goes to 59:01. This is 59 frames.
Level 2, frame 7781 is where the timer hit 59:01, and 7838 is where it hits 59:00. This is 58 frames.
The next second is 59 frames, and so on.
Mind you, this keeps happening on levels 1 and 3 and I'd image pretty much every other level, but I [i]did not test further than level 3[/i] to see if occurrences of two seconds in a row where there are 58 or less frames consecutively.

For timing level 2:
Start frame is 7699 [2:08.445] (since a frame before the was 'Press A to continue'), and clock has 59:03.
End frame is 11289 [3:08.338] (camera starts zooming into the exit door) and clock has 58:01.
At 59.94fps, this is 0:59.89 seconds real-time. In-game clock however claims the level time to be 1:02.43.

[size=9]For end of player control, take level 3:
At 14980 [4:09.916], clock turns to 57:00. At 14990 [4:10.083] the game begins zooming into the exit door. However, the game claims the remaining time is 57:00.37, so that means that part of the zoom in process (which has no player control) is actually taken into account before the timer actually stops. [i]If[/i] we assume game time to be 100% correct no frame-drops etc., then end of level/control is frame 15018 [4:10.550]. This however goes into territory I'm not going to pour too much energy into, hence this paragraph is in smaller size text.[/size]

I could keep going but the above is sufficient proof to show why this ingame timer is incorrect.

Jaguar King

Edit history:

Jaguar King: 2016-01-15 11:16:35 am

Jaguar King: 2016-01-15 11:16:02 am

2016-01-15 11:15:24 am

wow sorry for the late response, I just noticed it

yes the timer can be glitched and inaccurate, I didn't know this till last year when someone told me about it. I think the reason is due to the inconsistency for loading the levels in the game, some levels can load way way longer than its suppose to be making the ingame timer glitch out for no reason. stage 15 is a good example for it to load way longer than it should be.

This only happens if you play xbox360, I don't know if the same thing happens with the PS3 version of the game

also one thing we all can agree on is the ingame timer stops at stage 14 and 15. making the use of inagme timer is out of the question, so yes we should start using real time for future submissions. the real time for my run is 15:47 you can check it in this link here http://www.twitch.tv/jaguar_king/c/5104941