Cool.

I do not get the 0 beat jumps, how do you keep speed going sideways.

I'm pressing forward and never let go of it when switching sides. If you know how half beating or A/D works, W/S is the sideways equivalent of that.

I recommend everyone to learn the no-beat strafing (first time I hear of the name though).

When you can do normal strafing, half beat and no-beat, it feels like you have complete freedom of moving where you want at full speed while aiming at targets in different directions.

This is actually why I don't like CPM / Anarki movement, because half beat and no-beat don't work anymore!

And yeah like Smile said, on the keyboard you just keep pressing +forward, that's it. The mouse does everything else.

To Smile: there's a really cool move on Aerowalk where you are up at RA, somewhere between LG and rocket, and then you no-beat jump your way on to the RA, while popping a rail or nade to the spawnpoint down at shotgun. Tricky, but feels good when you land it and it's a good example of why no-beat strafing can be very practical in a game.

No-beat style refers to using only one pair or single key. Not just +forward, but for example using only +left or any of the diagonal directions.

This is actually why I don't like CPM / Anarki movement, because half beat and no-beat don't work anymore!

I have the same issue with CPMA movement, you can turn around your opponent but it's a very different technique. If your angle is even just a bit wrong either of these will happen: 1. You start turning to your viewangle linearly. 2. You adapt some weird strafe jump angle, gain speed but lose the air control.

But when it works, it's kinda like sideways bhop without the speed gain. I demonstrate that somewhere here. Works also in CPMA, Warsow, Xonotic and presumably Anarki as well if +forward is linear aircontrol.

Eh, how do you do that with CPM movement? When I do it like you're doing in the video, I just straight jump to the point where I'm looking at.

Take the sideways angle before you start pressing +forward, look how I did it at 0:21. Also try keeping the speed between 450-550 when you turn. If you're too slow you start turning linearly, if you're too fast you lose aircontrol. The sweetspot speed probably varies between games though.

Aha I think I see it. I'll try this later. Doesn't look as stable and usable as in QW for example though.

If they would fix that in Reflex and add wall clipping we have the perfect movement game.

Broken sideways bhop is probably not intended side effect of what they wanna use +forward for. The ambition to have that linear behavior is basically what is in the way, so it will probably be never fixed.

Reflex already has wallclipping, did a video on that too lol. Only differences are that you can wallclip from every jump and ceiling clipping doesn't work.

Lol @ 1:47.

For me right now the best movement is Slash in QC. With the crouch sliding you have great mobility and you can still hop sideways. Also, circle jumps are done pretty quickly and give a nice speed of the bat.

A vid with showinputcmds would be nice, because I absolutely don't get it.

Here's the technique with keypresses: https://www.youtube.com/watch?v=PzzRZl-qtcE

Thx.

When it comes to strafe jumping in VQ3, all that matters is that your acceleration direction (where you're indicating you want to go with your keys and mouse) is the same relative to the current direction you're going in.

So you can indicate that acceleration direction 4 different ways using a single WASD key and another 4 different ways using two adjacent keys at the same time. Each of the 8 combinations requires you to point a different direction with your mouse to get the same acceleration direction.

For instance, holding left only (A) makes your acceleration direction 90 degrees to the left of where you are facing, while left and forward (WA) makes it 45 degrees to the left of where you are facing, etc.

As I understand it, "no beat" refers to strafe jumping using only the single key combinations.

Here's a video of w3sp showing a whole range of combinations on the jump pads:

https://www.youtube.com/watch?v=hbX2INvtkBY

Now, if only it were as easy to do as understand! :D

It can be confusing, because when you want to go forward: you have to steer towards sideways angles. And when you want to go sideways: you have to steer forwards or backwards angles.

Yep, the faster you go, the closer the ideal acceleration direction is to being 90 degrees from your direction of travel.

I think normal strafe jumping is pretty easy to get a feel for, not so much these tricks. :D

I think much more people would be into learning strafe jumping if the variety of techniques were made clear. Because if it feels unnatural to strafe with A/D+W, then you might find it more natural to use A/D only or W only.

W-only is the most common strafing technique in Urban Terror for example.

A/D is the most common airstrafing technique in CS/TF2/HL, and QL's equivalent of that is essentially with just less air control.

I guess it is all about the ease of hitting whatever angles are needed in the particular game that determines which method becomes the standard sort of way of doing it.

When all of the tutorials teach the A/D+W method, then obviously that is going to become a standard. But A/D is quite literally as efficient, the view angles are just inverted.

The thing is, that method came about long before any tutorials.

I think it is easier to move a large amount left and right with the mouse while pressing the same direction on A and D than to move the mouse a small amount and in the opposite direction of the key you are pressing. That's the different between holding W and not.

Also, the angle by which you change the direction you are looking gets wider in the first case as you accelerate but gets narrower in the second case. This probably makes it harder to be more precise since it is harder to get a feel for the angles involved.

It may be different in other games with different velocity and acceleration values changing the sweet spot.

It's mostly all just placebo. There's no differences to acceleration rates or how the angles spread, that's why half beating works and is practical.

There might be small velocity snapping related differences, but as far as human performance is concerned there might as well not be.

Cool video. :D

Nah, I don't think this has to do with velocity snapping since the angles you are inputting at any given time are exactly the same. If there is a subtle difference, it's probably in the way your acceleration vector is composed from your key inputs.

I think the difference is in what feels easier to do when you're starting.

Let's pull an angle out of my ass, say, 15 degrees.
I can input that holding D while facing 15 degrees to the left, and I will turn to the right (kinda confusing since my mouse and keyboard input are opposite), or I can input that holding W and D while facing 30 degrees to the right (now my strafe and mouse are in sync). Maybe it is easier to feel 30 degrees than 15 also. When I accelerate, 15 might become 5, while 30 becomes 40. Maybe swinging harder as you accelerate is more intuitive.

Also, when most people figured this out, they didn't have any clue how it worked; it was figured out by experimentation, so the idea of releasing forward to go forward would be unintuitive.

so the idea of releasing forward to go forward would be unintuitive.

Except you don't go forward, you circle strafe left and right, which adjusts your forward going direction. When you call the jump technique "strafe jumping" and you're actually steering diagonally, isn't that unintiutive as well? Although Q3 movement is unintuitive period lol

There isn't any technical advantage to A/D+W. When using A/D, you don't have to start jumps from larger degrees and you don't steer to sides with less turning, you just think you do because you're used to dealing with another view angle. All placebo my friend.

I think the difference is in what feels easier to do when you're starting.

Probably hard to tell what feels easier when you haven't been practicing the options equally, after using A/D+W for 15 years it's obviously going to feel the easiest. New players might end up with a variety of alternatives though, as long as that variety was just introduced to begin with.

In the beginning, you didn't have people reverse engineering q2/q3 source code for this.

There were just some vague ideas about what needed to be done to get the speed improvement found out through experimentation.

I guess if normally you press forward to go forward, it is more natural to have a slight modification of that by pressing strafe keys at the same time. I don't think anyone but the nerds were thinking about vectors and the stupid normalization technique used by id.

I know there is no technical difference. We're just talking about what is more natural to discover. There must be a reason that is the method used. It could just be the way that was discovered first, and then that spread by word of mouth, just like the console commands that were completely undocumented. Why was it discovered first? Because during experimentation releasing the forward key would be the last thing on people's minds, I guess.

Strafe jumping isn't really intuitive, but you do get a feel for the angles pretty quickly. Probably the big thing about releasing the forward key is that if your angle is too small, your acceleration will drop off abruptly, whereas for the same thing to happen while holding forward, you need to swing too far out, which is obviously much less likely to happen during that experimentation stage.

I know there is no technical difference. We're just talking about what is more natural to discover.

I'm still not convinced that A/D+W is the most natural technique to discover. Some nerds in the past discovered it that way and passed it on via tutorials and demonstration - would be more reasonable assumption in my opinion. When you discover something, it doesn't mean you understand the discovery as a whole - so you naturally pass it forward with the circumstances that led to your discovery.

Let's take Quakeworld for example, where it is in fact A/D technique that is normalized instead: https://www.youtube.com/watch?v=ItyfXLeYgGU

So why did it become a norm there to hold A/D instead of A/D+W? At the ground level the movement works the same as in Q3 = turning in air means you gain speed. The major differences are that you have less linear acceleration and more aircontrol. A/D is not any more intuitive than A/D+W, it's just what gets taught in majority of the tutorials.

which is obviously much less likely to happen during that experimentation stage.

But we're past the experimentation stage, were at the stage where we pass the knowledge forward - which includes variety of technique. It should be made more clear what that variety is so people have more options through learning the technique - and not just be limited to the knowledge that the technique was originally discovered with.

That is the method that was discovered. We can only speculate about why that is what they discovered and not some other method of inputting the same angles. I've already given reasons why the A-D method might not be the most obvious thing in the world to try.

I don't think people were trying completely random things. They probably noticed a bit more speed when holding strafe while going forward, something they'd be likely to do just in the course of the normal game, and then tried to expand on that to exploit it better.

Your video, although cool, is an example of the sort of thing people would not be trying at that stage.

I'm sure the vast majority of people learning strafe jumping in 2000 or earlier learnt it by word of mouth. There wasn't even a youtube back then. Tutorials are pretty much irrelevant to what got popularized. I never learnt strafe jumping from a tutorial. I learnt it from other players.

As far as I'm aware, QW doesn't have strafe jumping. It works differently. The angles involved are different, and so what people had to try to make it happen was different, and it took a long time for people to figure it out too. If you watch early demos, no one is doing that.

"But we're past the experimentation stage, were at the stage where we pass the knowledge forward - which includes variety of technique."

I agree. This thread started where I was doing exactly that. :D

As far as I'm aware, QW doesn't have strafe jumping.

Wrong, it has strafe jumping in every sense of the word. Except you have more aircontrol and more aircontrol means you need to turn more in order to gain the equivalent amount of speed to Q3. To put that in perspective, Quetoo has a little more aircontrol than VQ3 but still not as much as QW: https://youtu.be/z48cW6kAl28?t=2m45s

But you're sort of right, it's called "bunnyhopping" there instead, but it works in the same principle: You turn in air and you gain speed. The angles aren't that much different that would explain the normalization of A/D that happened over there.

If A/D+W was a more intuitive thing to discover, it would of emerged the popular style in bunnyhopping as well, but it didn't. That's why I think your speculation is faulty. I also don't think pressing forward is intuitive if your objective is to steer in the air. You never actually go forward while you strafe jump even in Q3 and there's no reason to think so either unless you were taught the technique wrong.

Your video, although cool, is an example of the sort of thing people would not be trying at that stage.

I think it's as much natural to attempt these as it is simply running sideways or backwards. Because the movement direction you press corresponds to the direction that you airstrafe. So as soon as you have the knowledge that "turning in air = speed" the knowledge about the possibility of turning to all of the 8 directions should come naturally as well. Unless again, you were taught the technique wrong.

If you watch early demos, no one is doing that.

Yup, but we're talking about a time where mouse aiming wasn't even a standard in FPS.

QW has strafe jumping in the sense that you gain speed by jumping and strafing. It works the same way technically, but the acceleration is very different, and no one calls it "strafe jumping". At least, I've never heard it called that.

In any case, whatever you call it, the angles are quite different. Hitting full strafe while turning fast is normal in QW, so the angle is 90 degrees or more from your direction of travel. Normal strafe jumping in q3 never gets that high. The sweet spot is in a different place. It'll only get close to 90 when you are going very fast.

Anyway, that is my only explanation for why they discovered different keys for doing it. It led to the weirdness of having to press the opposite strafe key to the direction you were turning your mouse. If that doesn't explain it, then it was just random luck that they discovered that way over any other way.

We'll never know. :D

Q3 is essentially just a combination of linear acceleration and linear turning. Strafe jumping is quite literally just linear turning, you gain speed when one linear direction combines with another linear direction, this is achieved by turning. You're never actually strafing, holding strafe keys just happens to be one way to achieve turning...

...You can in fact "strafe jump" without mouse as well: https://www.youtube.com/watch?v=74YeUdvCQjw

The sweet spot is in a different place. It'll only get close to 90 when you are going very fast.

This is all due to linear acceleration. Following with what I said above, the reason why you have to keep 90 degreeing it is because you're constantly turning and combining linear directions with one another. And you can't combine directions if you hold one direction for too long or if you repel towards opposite directions. (Turning mouse too slow or too fast). If it weren't for linear acceleration the movement would be quite identical to QW, even if the implementation is different. Look no further than CPMA/PQL.

While CPMA is a good example of QW simulation on Q3 engine, it's not really the best example of the two movement techniques corresponding to same values. Because the mod by design aims to balance between maneuverability and speed gain.

I know that QW movement is not implemented the same way and I'm not sure how. But they're definitely both from the same mold.

Strafe jumping, and the bunny hopping in QW, is purely the result of the flawed method of capping the speed when you add your acceleration vector to your velocity. Who knows how they dreamt it up, because it is incredibly stupid, but it had some nice side effects.

I guess the word "turning" is confusing because you have turning as in "changing the direction you are facing" and turning as in "following a curved path". The former is just a byproduct of the fact that to input the sweet spot angle (the angle that provides maximum acceleration) you have to turn your mouse. You wouldn't need to turn at all if you could input that acceleration vector some other way, using the keys only for example. (Either way, you will follow a curved path, the other kind of turning. :D)

If you use keys only, you can only input angles that are multiples of 45 degrees, so you can't hit the sweet spot unless that angle just happens to be a multiple of 45 degrees. Of course, you can get some acceleration that way, but it won't be the optimal amount.

Anyway, I don't think I'm necessarily contradicting you here, just describing it in my language. :D
I guess I don't know what "linear turning" is. Although I kinda see turning (as in following a curved path) as a side effect rather than the method itself. The method to me is just ensuring the correct angle is being input at all times.

The QW method of acceleration is virtually identical to q3's. The big difference is just in the size of the vectors involved, which create different sweet spots. :D

Anyway, all I was saying is that the angles being different may well account for one input method being discovered instead of another.

If you use keys only, you can only input angles that are multiples of 45 degrees, so you can't hit the sweet spot unless that angle just happens to be a multiple of 45 degrees.

You're absolutely in the right, but the fact that you can strafe with keys only was just an example of the linear turning (caused by linear acceleration). You can't do this technique in QW, because there is almost no linear acceleration. This makes it very impractical to attempt turning in the air without using the mouse.

Anyway, all I was saying is that the angles being different may well account for one input method being discovered instead of another.

But if angles determine what people discover naturally, then shouldn't Q3 ppl have discovered A/D primarily as well? As you know when strafing diagonally, your view angle starts to bend 90 degrees away from your direction of travel. This behavior however is inverted in A/D where the turning starts from the side, but centers towards your direction of travel. It's the same exact angles and sightlines, but with inverted logic (hence people calling it inverted strafing). I don't think either technique is more natural than the other, its essentially 50/50 chance to be discovered first. The implementations may differ between QW and Q3, but the logic is the same. Corresponding key presses give corresponding benefits.

You could well be right. Which was discovered first could just be a random occurrence. There isn't really any way to find out. Maybe we can dig up the guy who figured it out first and ask him. :D

Anyway, nowadays we know better. Go forth and spread the word. :D

Good music.

very nice, except those stupid icons in config. still 10

Good stuff I only have seen few of those jumps before. I discovered the cp5 jump years ago https://www.youtube.com/watch?v=Xh4nqnAn5DA

Nice, it's hard to keep track of whats been discovered. Though I'd argue one plasma and no rocket is a decent improvement :P

EDIT: Also, I aimed the plasma at the ramp rather than the ground, because you actually can get little bit higher the more lower you start sliding up the ramp.

EDIT2: And you get more of the plasma boost if you do it while youre sliding.

I agree you have the more elegant solution and execution but I did it under the aspect of functionality for ctf which means speed and efficiency. I did it ingame a handful of times and it was useful probably once or twice. anyway nice stuff.

Only reason I discovered that trick was because I was inspired by trickjumping in Wolfenstein: ET which has lots of terrain, angled walls and ramps like that: https://www.youtube.com/watch?v=hq2oatJNdhU

It's a shame that most QL maps are almost entirely flat, you could have so much more depth with even slightly less flat geometry - which is something I tried to illustrate in my video from those tiny angled spots I could find everywhere :P

nice music & tricks °_°

well done

Nice effort, but had to mute the audio after 10 s..

Why not a big fat 1 minute instead

Couldn't bare it for that long.

this style reminds me a little bit (and i mean LITTLE) of the insanity that is trick'em all 3 by soul_brother, so for that i have to give it a 10.

Most of the insanity in trickem all 3 is impossible in vQL :P

- Overbounces
- Plasma slick
- Whatever those wall glitches are called
- bridge to rail with a single jump LOL, maybe in PQL.