Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
FractalFusion wrote:
Oh, I didn't realize DOS was the "main" version of the game. I thought NES was the main version at first. Also it somehow reminded me of Karateka, but now I know it was by the same author.
I like how when encountering guards, the Prince takes one stab each time and decides he's had enough and leaves. Anyway I'm not familiar with what is and isn't a glitch in this game so I can't really say much regarding that.
I wouldn't say "main" version, but the most popular version atleast in the speedrunning scene. It was certainly one of the cleaner ports of the original Apple II version.
Yeah after switching sides, prince can just put away the swords and run after either being hit (which pushes him far enough) or by stepping back far enough himself. Which of these 2 happens is purely dependent on RNG, I took whatever the game decided to give and glad it wasn't a huge problem in terms of health management. Thanks for checking the run!
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Yesterday in the RTA speedrunning community, we discussed about the sort of arbitrary nature of the ruleset and we decided its not very apt to call it "glitchless" and renamed the category to "No Major Glitches". So I decided to add that as the branch to this run. However if the judges feel that it is not necessary, feel free to remove the branch.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Just edited the submission to remove the branch. Wasn't sure if I had to remove it myself or the judges would have done it during the judging process. Either way, its done now.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Radiant wrote:
Note that your run still has a 100% vote despite the goal discussion :)
Also, I'd love to see a POP2 run.
Yeah I am very grateful for that. Honestly didn't expect that for my first submission, which certainly motivates to do more TASes.
I need to first learn the RTA run of PoP2. I know that it's not necessary to do a TAS but it's just my favourite way of learning the ins and outs of a game 😅. No promises but I expect to do it within the next few months.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
PLANET wrote:
Obvious Yes vote, happy to see it done here :)
First game I've played on PC (just after seeing a spinning coca-cola can animation demo : D) and naturally, a lot of fondness to it.
Love some parts of the run, especially Jaffar "fight" ; ) and the guard on level 10 falling just 2 stories.. to his death :P :D
Having said that, I'd really like to see the limits pushed on this game, that is the usual way TASes are done here :)
So all glitches/bugs/exploits possible. That would definitely be lovely :)
And also agree with c-square, the safest option would be to proceed with the run in simply any%, because categorizing what is glitch and what is not is ALWAYS problematic.
Great job man, please keep it up - and consider a total any% TAS of well... BOTH :D PoP 1 _AND_ PoP 2 : )
Thanks man, yeah I have learnt my lesson and will never do a arbitrary goal TAS again lol. PoP1 w/ glitches as I said is cooking and PoP2 any% is definitely one of my future goals.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
c-square wrote:
I seeing there's no argument over the quality of this submission. General consensus is that it is well done and easily falls within the standards we expect to see on this site. I think where people are struggling is deciding what the publication branch would be.
I concur with what some have already said. While it may make sense to separate frame-perfect input tricks to a separate category in an RTA forum, on this site frame perfection is the standard, not the exception. The only time we allow a branch that avoids "glitches" in a TAS is when it produces a markedly different run, requiring different strategies and resulting in a different end-product. Good examples that come to mind are the zipless mega man 2 and BLJ-less Mario 64 runs.
I don't feel the glitches avoided in this run fall into the run-altering category I cited above (with the exception of Practice mode, which I'll discuss in a minute). In TAS-philosophy, glitches are there to be exploited, and it doesn't feel like there's a good reason not to exploit them here.
In conclusion, DOS Prince of Persia is an iconic game from its time, and it would be a benefit to this site to have a run published here. In my opinion, this submission certainly shows off superhuman play, and definitely meets the bar of quality we expect. I propose that we accept this run as our first submission on the any% branch, with the understanding that any future submission that utilizes "glitches" to materially improve the run would then obsolete this one.
[Note: In regards to the Practice mode skip, I doubt that will be accepted here, as it would have the exact same content as a full run, just minus the first few levels. Those currently working on the "glitched" run would be wise to take this into account before going too far.]
I would not mind at all if it's accepted under the any% branch as it avoids both the problems of "avoiding glitches despite not being too game breaking" and "doing tricks that some people may consider glitches".
And yes the people doing the current TAS are well aware of that level skip is not allowed. We want to see the full game anyway as that's the primary any% category.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Radiant wrote:
I find it hard to believe that walking on air is meant to happen at all. So why isn't walking on air considered a glitch?
The crouch mechanic can be used to move underneath half-closed doors. Crouching to avoid damage from falling ceiling tiles (not to avoid being hit, but to avoid damage from being hit) doesn't sound like intended behavior to me. The manual, at least, suggests to avoid falling tiles by stepping out of the way, not by crouching.
I mean, I'm hoping for a better explanation of what counts as a glitch than "that's how speedrun.com has been doing it for years", because that doesn't really explain anything.
The "running is air" is quite literally just a parkour mechanic, which the franchise is well known for. I recommended taking a look at the PS3 port of the game where the same mechanic exists with a much cleaner and non-glitchy looking animation (same about swapping sides with the guards that was mentioned earlier).
The regarding the crouch, idk, I just tried to justify why it might have been considered not a glitch, but I honestly have no better explanation other than the fact that it's not considered a glitch as per speedrun.com ruleset, and the ones that existed before it. My goal here as I said was to make a run that would be comparable with those runs.
Edit: here is the RTA World Record of the PS3 Port I mentioned - https://www.twitch.tv/videos/852203450/
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Odongdong wrote:
Congratulations on completing your run! I really loved playing this game.
I have the same question about the definition of glitchless. As far as I know, one of the most famous bugs in the DOS version of Prince of Persia (I think it's in most ports) is avoiding damage from falling ceiling tile with a crouching animation, which you used on this run, like 17:40 on YouTube video. Isn't that consided a glitch?
Thanks for the kind words Odongdong.
That is actually a very good question. It is one of the questions I asked when I started running the game as well. It is not considered a glitch apparently and it's been the case historically since this category has been established. The logic here is, that animation is actually meant to avoid this damage apparently, as there is literally no use of it for anything else in the game, hence it actually might be an intended mechanic.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Radiant wrote:
I love this game and played the heck out of it when I was a kid, so this looks like a great run to me...
...but the goal strikes me as rather arbitrary. Why exactly is it that jumping through guards and turning off chompers are considered "a glitch", whereas walking on air, turning away from guards in mid-combat, and one-shotting Jafar are considered "not a glitch"? All of that appears more-or-less equally plausible in RTA conditions, and none of it is as weird as the various warp glitches.
And I'm a bit surprised this is only marginally faster than the speedrun.com record (18:50 vs 18:56).
Hey Radiant, Thanks for checking out the run.
The reason they are considered a glitch is because even though they tend to look "normal" even compared to other intended mechanics, their execution is so precise and specific that it's obvious they aren't meant to happen at all.
It is only marginally faster than the RTA record because the emulator that is used to do the RTA record runs the game quite faster in real time. It also varies from machine to machine, hence why we actually use the in-game time to compare the runs. But obviously here I have to use the TAS timing. Comparing in in-game time, the RTA record is 18:08 and this run is 17:40.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Kenttsu wrote:
I'm curious why guard-clipping while holding a sword isn't considered a glitch?
You mean swapping sides with him? Because it's just an intended mechanic in the game. It's even mentioned in the manual if I remember correctly. Its nothing special, you just press the direction to do it. Guard jump (without a sword) on the other hand is a very specific trick that is frame perfect in most cases.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Warp wrote:
No glitches are to be used in the run, these include: screen warps, guard jumps, clipping corners while falling, etc.
I am wondering if it would be good to actually specifically list all known glitches that are known, and avoided in this run. Also, if possible, techniques that are allowed but which might perhaps be interpreted by some to be "glitches".
I ask this because this kind of would have to be a requirement for any run that intends to obsolete this. Anybody making a run intending to obsolete this would need to follow the exact same restrictions, or else it wouldn't be comparable, and we would enter this whole problematic issue of what counts as "glitchless" and what doesn't. Maybe someone makes a new run that's faster than yours, which abuses a glitch that you merely collated into that "etc" part. If you didn't specify it, who's to say it can't be used?
(Of course even then there's the question of what exactly is the "official" list of banned glitches for the "glitchless" category of this game.)
Sure, I followed the same ruleset that the RTA glitchless runs follow, I will paste it here:
The following things are considered glitches:
Screen-wrapping
Guard-clipping (guard-jumping)
Falling or clipping through floors and closed gates (specifically corners)
Fall damage crouch cancel
Inactive chomper glitch
Guard through-gate lure glitch
Here is a guide that explains these with GIFs showing examples - https://www.speedrun.com/pop1/guide/m4j3u
You can skip cutscenes between levels. Sound & music must be enabled at all times during the run. Ctrl+A to restart a level is allowed, except for situations where it would cause a glitched advance to next level instead of restarting.
As I said the jingle skip is the only thing that some may consider a borderline glitch, but we had an extensive discussion about in the RTA community and decided to allow it.
Edit: edited the submission description as well
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Challenger wrote:
Just found a run of this version w/glitches today: https://www.youtube.com/watch?v=hWJBsOalD-k
That was even more unexpected than I thought. I should have put bruteforce when I was TASing that other version.
Yeah! That is the prototype to the glitched TAS I have been talking about!
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Challenger wrote:
If you have plans, there's a TAS of SegaCD version I've mentioned before.
Most of the gameplay was based on a old DOS run from 2007, and while there are some mistakes, some strategies can be useful.
Will certainly check it out. I must make it clear that my role in the glitched TAS is going to be very very minimal, but they actually use a bot to pretty much bruteforce through the rooms (while giving some general route/directions), so its going to be a very very optimized TAS when it finally comes out I hope.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
c-square wrote:
Welcome to TASVideos, GMP! I love seeing the old DOS games get TASsed, and this is certainly one of the most famous ones. Thanks for all your hard work putting it together.
I finally got to watch it, and it looks very well done. Yes vote from me for both entertainment and publication. I'm looking forward to the any% w/glitches run!
Thanks a lot for the kind words c-square! I am looking forward to the any% w/ glitches TAS just as much as you :D
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Challenger wrote:
Second, normally that second guard from level 10 is supposed to turn around before the prince reaches the gate (it prevents you for climbing directly to the point above), but you can avoid that with some crouching jumps. This isn't actually a glitch?
Third, some stuff are new for me, like skipping end level jingle, level 3 reset and OHKO Jaffar. Nice finds.
Crouch-jumping to avoid alerting guards is indeed considered to be an intended mechanic, atleast in the RTA speedrunning community. What I did do differently however is absolutely minimize the amount of crouch jumps required by carefully positioning my jump, if I start crouching one frame late he indeed turns.
And yeah all of those tricks you mentioned are indeed fairly recent finds, thanks for looking into the run :D
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Easy yes vote. A lot of downtime but well compensated by the crazy shenanigans pulled off in almost every level. It is even better once I watched the explanation video in Kosmic's channel. And also this concept is amazing and allows to show off glitches that are obsolete for other categories which is great. Would love to see if its improved in the future but super cool too see that this category is established.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Hikaruon wrote:
Yes vote, I've played this in my childhood, and how! Only 14 levels! I didn't remember that game was too short.
But part of me think is possible do some improvements, even in DOS version (skipping the guard without get the sword or lure him).
PS. I loved the "Final Fight" XD
Thank you. You might have played the SNES version which contains 20 levels and also gives 120 minutes to complete the game (as a sidenote, someone mentioned they would do a TAS of that version soon :D - https://www.speedrun.com/pop1_snes/thread/mnqfw). But yeah this version is the classical one, it has quite a lot of ports some of which have some quite unique elements.
Regarding improvements, I am sure the guard RNG can be manipulated better to gain time, but I am fairly confident that the overall route of each level cannot be improved more especially in glitchless category. I am a fan of "never say never" but I cant help but say the scope is less. In any% you can jump straight through that level 1 guard (which is considered a glitch) but here this is the most optimal way as far as we know.
And yeah the final fight lol. It is actually 4 frames slower compared to the strat used in RTA runs which measured are with in-game time which stops as soon as Jaffar is killed, but I used it because a) its cool and b) its faster in real time haha.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Thank you, yeah personally I love this game so much, more specifically this entire franchise, glad to see someone who shares the same opinion :D
Regarding difficulty in TASing, the main difficulty was that this is kind of a unconventional type of system to TAS in. And since this game runs at 12 fps, one frame in-game passed only once 5-6 snapshots of the system, that was a little weird to get used to. But once I set up stuff and settled in it was nothing too different from TASing in a traditional console system.
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
nymx wrote:
Welcome.
I'm not quite sure how to evaluate your efforts, seeing that another un-submitted video exists. The video below is also a DOS run, but seems to take advantages of all the tricks that are not present your TAS. From your submission, you've stated that you avoided glitches. Unless glitches are game breaking...I'm not sure that your run is any different that the video posted below....which is minutes faster.
Can you elaborate more on the differences?
https://www.youtube.com/watch?v=hfOocUkxJ08
Hey nymx, thanks for the welcome and for taking your time looking into my submission. Yeah I am aware about that video. The main timesave of that run comes from the fact that it skips the first 3 levels with a cheat code that is present in the game. It is what we call as the Any% Level Skip category in RTA speedruns, which is not the primary Any% category. Glitches and tricks have actually improved quite a lot since that TAS that actually RTA runs of that category are faster now. We (specifically eien86) are. actually working on the Any% with glitches TAS currently and should be able to finish soonish.
Now coming back to my run, yes it specifically avoids glitches that skip a lot of the game, which contain satisfying platforming sections. These sections also contain a lot of precise speed strategies performed without the use of glitches. The glitchless category is equally as popular as Any% among the RTA runners despite being arguably more arbitrary and my main intention was to show what a really good run of this category would look like. Hope this information helps a bit. More information can be found about the glitches, gltichless rules, other categories, etc. here: https://www.speedrun.com/pop1
And ofcourse further specific questions here are welcome!
Editor, Experienced Forum User, Published Author, Reviewer, Active player
(360)
Joined: 5/22/2020
Posts: 197
Location: Chennai, India
Hi guys, I am new to TASing and this is my first fully finished TAS, even though I have messed around with TAS tools for a while now. I have always been fascinated by TASes and am super happy to have completed one myself.
And ofcourse feel free to ask any questions :D, I will try to check this page as frequently as possible but if for some reason I fail to respond quick enough, feel free to ping me @G M P#0177 on discord :)
