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: http://speeddemosarchive.com/DevilMayCry3.html

Devil May Cry 3 () (pc) [Any %] [Single Segment]

Verifier Responses

Quote from ahuynh:
The video quality seems fine. There are some spots that seem strange, but I'll attribute that to this version being a poor port to PC. Audio is also fine, except for the Windows notification beeps, which only last at most a minute.

The ingame timer by the final save screen is 1:38:08. However, even though this is what the current DMC3 runs on SDA use as ingame time, I feel that it is not a good measure of time taken in gameplay, as this time includes literally everything, even loading times. A better measure of gameplay time is to sum up the mission times, which ends up being 1:20:56. The mission timer only counts gameplay, and does not include things like loading times and menu time, such as the screen that shows up when you grab an item. This is actually a good measure that allows runs from any platform to be compared to each other.

So, while it's nice that someone finally did a speedrun of DMC3, and while it's focused on speed, I would have to say that this run is not up to standard. I would say the movement is fine, but a good amount of the game is fighting enemies, so a speedrun would show off good movement as well as efficient fighting through the mandatory mob fights and boss fights. It is this second part where this speedrun fails at. To me, it felt that the runner played through the game a few times, and then just started doing playthroughs rushing through to the end and hoping for the best. Now I doubt that this is the case, and I honestly do believe that a good amount of work was done on this, but that is the feeling I get from watching it.

The runner shows off a good amount of the knowledge of the game that I do not know of. However, there is plenty that the runner does not know, and I feel that the runner did not do enough of an effort to find out about the deeper mechanics of the game, which really would have allowed him to save more time.

Comments:
-There is very little reason to be taking damage, though I can understanding getting unintentionally hit a few times. However, I do think that the player takes damage too often. He could have gotten more S ranks such as in Mission 3, 4, 5, but because of the amount of damage he has taken, he gets A ranks, and losing 2000 red orbs per mission, which are particularly important early game. Most of your Red Orbs will come from ranks.
-The runner also shows a lack of understanding enemies. For example, he gets hit by a Sloth in Mission 3 when there was absolutely no reason to. The Sloth was clearly on screen and very easily visible to see it initiate its attack. There are other instances as well, such as getting hit by a Greed several times, even though it was clearly in the middle of attacking.
-Moreover, in order to make up for the fact that he is unable to avoid damage apparently, he uses Vital Stars, as well as spends orbs on Vital Stars, which could have been better spent on a 7th purple orb, or moves like Straight Level 2, as Stinger Straight combos are faster than Stinger Dash combos. Dash to Sky Star is also really slow movement and should only really be used in areas where you go past enemies. Depending on the type of damage, as well as the amount of orbs, it might even be a better investment and a time saver to go with Blue Orbs if you're having trouble with taking damage.
-The use of Crazy Combos is a poor one, for crazy combos do very little damage. It takes 10 million stabs hits to match the damage of the first hit of Rebellion. Though the intention of using Crazy Comboes was to get more style points, I also see a lack of taunting, which would have worked just as well to get more style points.
-The runner mentioned that Rudra the boss is weapon to Agni and Rudra the weapon, but that's not true. In fact, both Agni and Rudra the bosses take 80% damage from A&R the weapon. Rudra the boss just isn't weak to anything. Also mentioned is that Arkham is weak to Beowulf the weapon, but that's not true either.
-For jump cancelling Aerial Cross, calm down and slow down. Do not mash the button, because the runner ended up jump cancelling before the attack even comes out, as seen in mission 7 and 13, as well as against Beowulf, and the Cerberus refight. Good jump cancelling is controlled, well paced, and even slower than you might think. Jump cancelling Killer Bee is incredibly easy, but he fails to use the easier technique which makes it extremely trivial on most enemies that can be lifted by High Time, including the Abysses in Mission 17 and 19. This technique is one that is so basic to Jump Cancelling and so easy to implement, I actually think that the player does not know to use High Time then follow up with JC Killer Bees.
-Vergil 1 looks fine, but it is actually suboptimal. Since he uses the 3 hit combo with the ordinary finisher, the finisher actually causes Vergil to begin parrying earlier, and he only gets 10 hits in at a time, when had he finished with High Time, he would have gotten 13 hits in. If you use Royal Guard for Vergil 2 and 3, you should be using guard cancelled 2 slashes.
-For Vergil 3 though, since you have Beowulf, you should try to JC Killer Bees, but since he fails to JC, he goes with slower regular combos. Once again, he shows a lack of understanding of Vergil's parry mechanics. You should be switch cancelling and preventing yourself from doing knockback moves.
-The Nevan fight is very sloppy. He does not know how to knock her down, which makes her vulnerable longer, even though there were a few opportunities to do so. Also, you should not be using DT Cerberus, as it gives absolutely no damage boost, and I'm not sure if the speed boost is worth losing the DTE. Even DT Rebellion should not be used as it gives a 10% damage boost, and the DT would be better used for DTE. I'm not sure how good Cerberus is at breaking Nevan's bat shield, but it may have been more effective to use Agni and Rudra instead. Jump Cancelled Aerial Crosses are very effective at breaking Nevan's bat shield, as well as guaranteeing knocking down Nevan.
-During the first part of Geryon, I have no idea why the runner ran out and got hit. There is no reason for him to do that as he would have been perfectly safe in the corner. If you can't consistently Just Release Geryon as he is charging you, I don't think it's worth the risk.
-For the second part of Geryon, I seem to understand the intended strategy, but it seems like all he is doing is just running into Geryon and getting hit. The runner should have figured out the exact conditions that causes Geryon to fall over, because clearly, after the first knockdown, he is able to connect with Cerberus's Swing, but it does not knock Geryon down. If there is a way to guarantee making Geryon fall down, more practice should be done to make it more consistent, which the runner has trouble since he messes it up several times.
-The choice of leveling Trickster over Royal Guard ended up being a questionable decision, which we can see in Vergil 2 and 3. The player has trouble on properly using Level 1 RG, which is even harder than L2 RG due to the inability to jump guard. I also did not see an instance of roll guards being used, though to be honest, I'm not sure how much it can help.
-As a result of the player's lack of skill in using RG, the Vergil 2 and 3 fights are very sloppy, resulting in lots of failed guards, as well as using Devil Trigger to recooperate for the failed guards instead of using it for DTE, which is a much more effective use of the DT gauge. Both Vergil 2 and Vergil 3 can be so, so much faster if he knows how to properly guard Vergil and using the DT for DTE.
-Furthermore, Trickster is just not a good speedrunning style. The best style is Royal Guard, which would save huge chunks of time, especially once it reaches Level 2 and you gain the ability to Air Guard. With that, guarding becomes much easier, and tricks like using a Just Release on Geryon is so much easier. The offensive power that you get from Royal Guard is just insane. Bosses like Agni & Rudra or Beowulf would be defeated extremely quickly with Level 2 Royal Guard. The lost movement can be made up by guard cancelled Stingers, as well as learning to guard mobs to get Devil Trigger so you can DT in rooms with enemies in them.
-There's also a free Holy Water at the end of Mission 14 that he didn't get, which should have been gotten to use to save 20000 Red Orbs, which then could be used elsewhere.

The movement is fine, but the battles are just so much improveable. I would recommend the player to spend more time refining his skill and execution. However, this run is a reject. I honestly do believe that this run can save several minutes with much better execution and newer strategies that reflect a greater understanding of the game. The strategies used in this run aren't exactly optimal, and I'm pretty much the strategies that were used was due to the runner's lack of knowledge of the game.

Here is advice I will give to the runner, which will be similar advice that I give to aspiring DMC1 runners. Are you able to play through the game without focusing on speed? With the bosses, are you able to consistency perform some strategy well? You don't have to be able to do it 100% of the time, but I would say at least 75% is fine. One guideline I give for DMC1 is the ability to go through DMD! and get S ranks consistently. If you can do that, you'll laugh through how easy Normal mode becomes. Most people I tell this to do not follow this advice, and I found that their skill is extremely lacking. However, there is one other person besides myself who can go through DMD! consistently, and his speedruns reflect that. Do not aim to get good at Normal mode. Aim to get good overall, and your skill will scale accordingly. I don't know if going as far as DMD! is needed for DMC3, but it is definitely a good standard to have in order to learn a good amount of the game. My recommendation for DMC3 is not to do speedruns to get better, but to do a SS-rank run. There are many times you gotten hit, because I think you don't understand how the enemy mechanics work. There is a lot of common knowledge out there, or at least, what I think would be common knowledge, except that you demonstrated to me that you really don't know about it. You should learn to Jump Cancel Swings. Now, there's no reason for you to ever JC Swings in practicality, but if you can JC Swings, that provides a good foundation for other JC moves, especially Killer Bee. You should also properly learn how to use Royal Guard. There is no reason for you to use it if you can't properly guard. The goal here is to obtain a level of skill first BEFORE focusing on speed. Once you have that knowledge and skill, then you can go start focusing on speed.


Quote from Eternalspirit:
The play quality is on the weak side of an accept, and it was brought to my attention that it likely isn't much faster than the current comparable speed runs. That being said, I think I will still accept this as the first run available in this category.

The runner clearly displays a very high level of skill and knowledge with this game, but I think the showmanship is a big reason for many of the small errors. The runner admits to this in the comments and I will let it slide this time, but I would scrutinize these issues more closely in future submissions.

Most of the errors are pathing errors where the runner fails to enter a door or jump onto a platform properly. There were a couple of big blunders but I can respect that sometimes you have to let it slide with this game, as it is not very forgiving.

Overall it's a good enough run. I will be the first to admit though that improving on this will be quite difficult.


Decision: Reject

Reason: While this run is a good achievement, both of the verifiers indicate that many things could be improved. This seems to have been a correct assumption as the times have continued to be pushed down in this category. Once things start to settle down, we're hoping that the runner will return and submit their new best time.

https://queue.speeddemosarchive.com/verificationfiles/1567/

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