EVE General Discussion

It is theoretically impossible to warp scramble a pod that hits warp straight away, whether that be from an exploding ship or from landing and warping off again. Don't let anyone convince you otherwise. Note: 'theoretically'

This is simply due to the fact that the server runs on a 1Hz tick, thus meaning when two things that take less than 1 second and start on the same tick they will both finish at the exact same time too. The server effectively rounds up to the nearest second how long an action takes. Meaning when 2 processes happen that require multiple <1 second actions the process which requires the least amount of actions should in theory finish first.

This is important because a pod will align to warp in <1 second. Instalocking ceptors will also lock in <1 second, as I've already pointed out this means they will both happen in 1 second no matter how stupidly high you crank that sensor strength up.
So when scramming a pod trying to warp off the success depends on which requires the most actions. A pod will align in that first tick and then be invulnerable as it enters warp in the second tick, the instalocker will require the first tick to lock and the second tick to scram but as I've already said the pod is invulnerable by the second tick.

Therefore a pod is impossible to scram before it warps out... in theory...

As the difference between getting scrammed and warping out safely comes down to 1 server tick network latency plays a big roll in deciding if the pod gets out or not, if the warp command doesn't go through in that first tick, that pod is toast.

Just wanted to dispel a myth.

Warping to something directly in front of you doesn't shorten your align time if your are not moving. If your at 0 m/s it takes the same amount of time to align to any celestial in any direction to warp out.

That may seem counter-intuitive but seriously, try it.

This!

In fact, I recall some Dev explicitely stating that "spamming the warp button is actually counterproductive" as it messes with the ticks, causes lag or whatever.
Don´t know if that was fixed, though (probably not, so "a moderate 1x "warp to" every second or so" is the best advice given so far).

This is the most important post here.
The do it now.....warp nownownownow omg plz is the most important part of the process.

Your align time as well as the lock time on the stiletto probably fell within the same server tick, causing him to get you pointed before you hit warp. Unfortunately you are bound to the timing of the server ticks when trying to min/max mechanics that happen that quickly.

Yes, I've been understanding more about UI lag now that I know about the 1s server tick.

It is indeed possible for an interceptor to catch a pod.

Everything in Eve happens in 1 second intervals. If the inty has less than 1 second locking time and is spamming lock, he will insta-lock you. In these situations, just having a bad ping time can get you killed.

For example, I was once using a Nemesis to salvage T2 wrecks after a battle. Was orbiting a wreck outside of decloak range and inside of salvage range (~3000m). Had my finger on the cloak button when an interceptor decloaked from jumping into system. I quite literally instantly pressed F1 and instead of cloaking I got the "You cannot cloak because you are being targeted by someone." message. Never mind that I had pressed the button less than 1 second after seeing the ship in space. Nor did my overview show a yellow box. At the next server tick the yellow box came up. Next tick red box. gg lag.

If your ship is about to explode then this will work for you but if you are coming out of warp that is another story. Coming out of warp and his scram is already active and just waiting to lock you then its no good spamming warp.

You killed my Bhaalgorn a couple weeks ago. I hope you're happy

This is what i thought also, the (1/TiDi)Hz server tick is designed to help level the playing field among people with crappy machines, connections or long distances to cover. Being Aussie, I am used to rotten copper phone lines and thousands of kilometers of trunk.
I beleive each action you log in the UI is timestamped not just to each server tick but to the milli'second' within that tick

For clarification (or probably not), I thought i'd explain how I see it as a round of pokemon moves.

After the tick, the server recognises the Pikachu Pod was ordered to warp 0.2 into the tick, and would complete say 0.7 ticks later, at t=.9

Unfortunately for him, the Ratatta Stiletto attacks at t=0.1, and takes 0.7 ticks to lock. While locking the pilot presses the scram key ONCE, registering that he wants that activated instantly upon successful lock.
the pod pilot was too late, better luck next time.

Interesting point is that you only need spam things a few times to ensure earliest input of the scram/warp/target signal.

I'd like to point out that if this is really how things work (i'm no expert) I think any latency up to half server tick time (oyu need to see AND react in time) would be mitigated by this process.
Of course there are still ways you can improve your system, by getting a fast computer, reliable ISP, and minimising client lag from having ALL BRACKETS ON ALL DA TIME among other fun things like inappropriately high graphics.