[TMA] New Probing Mechanics

[Vortex]

[quote=sassy b]

I note that the two strongest individual results from those tests were from the cube and antiprism respectively.

I note that the weakest individual results from those tests were form the cube and anti-prism respectively.

What's your point? Mine is that those results agree with my hypothesis that formations with max number of probes evenly distributed around the sig are optimal for maximising scan strength.
At no point have I said that its the quickest/easiest/best method (all of those being subjective and dependent on a lot of variables). Tbh in the majority of combat probing I'd still expect 4 probe tetra to be fastest for me. But cube/antiprism should be best of the above methods for obtaining the strongest possible result, which the above tests corroborate.[/quote:kvsec7u4]

Well if your just intellectual curious then yea, there isn't much more to glean (though 7 with center lowered range I bet beats it until the entire pattern is at maximum resolution). That the results had both the highest strengths and by far the lowest strengths, suggests that the pattern is highly sensitive to minor deviations or imperfections in probe placement, which makes it rather impractical to use for actual probing application.

For this, what's important is the greatest reliability and speed. I'm sticking with tetrahedrons. Combat probing unprobeables may be a different game altogether.

Curious - when probing regular (i.e. not "unprobeable") ships, do you generally aim to get a 100% on the first scan? If so, what radius do you go for on that first scan? For my part, I normally used a 4-probe square at 4 AU prior to the changes, which was enough to get most things (om nom nom tengus), but am not sure that'd be sufficient for a 100% hit with the new changes, even with my new virtues plugged in. Will have to test it, but I think it may be advantageous to move to the octahedron or centered octahedron for routine probing of ships in safes/missions/sites at 4 AU; I'm not sufficiently confident in my initial d-scan based assessment of the target's position to be comfortable running the initial scan at 2 AU.

There's a lot of variables that go into this. How much time you have, how aware is the target of your probing, what ship type they are in, and how accurate you think you can preposition your probes. Small ship in a weird safe your going to have to start with much longer range scans to guarantee you lock onto it than a 100k km scanpoint to a stargate safespot. If your in wormspace with time to really set up the scan angle and initial probe position, you could conceivably get very narrow results. A lowsec system with just you and your target, well, your going to have to fire from the hip quickly unless your probing alt looks like the most cuddly character ever (mine doesn't).

Also, e-jealous of virtues. Wish I could afford a set

[sassy b]

That the results had both the highest strengths and by far the lowest strengths, suggests that the pattern is highly sensitive to minor deviations or imperfections in probe placement, which makes it rather impractical to use for actual probing application.

Lolwut. I'd be more inclined to look at the high level of variance across all the results, due to deviation on the initial weak scan exacerbated by going down two sizes at once. The two lowest hits you mention only had three probes getting a valid result, meaning that the patterns weren't even vaguely centered on the sigs.

Maxing scan strength isn't a purely intellectual concern, its what you want to aim for if scanning sigs at the limit of your skills and equipment (eg. getting tengus without using virtues).

[Hydro]

[quote=sassy b]

I note that the two strongest individual results from those tests were from the cube and antiprism respectively.

I note that the weakest individual results from those tests were form the cube and anti-prism respectively.

What's your point? Mine is that those results agree with my hypothesis that formations with max number of probes evenly distributed around the sig are optimal for maximising scan strength.
At no point have I said that its the quickest/easiest/best method (all of those being subjective and dependent on a lot of variables). Tbh in the majority of combat probing I'd still expect 4 probe tetra to be fastest for me. But cube/antiprism should be best of the above methods for obtaining the strongest possible result, which the above tests corroborate.[/quote:235kgcmp]

Cube (8 probes)
34.78
Average: 19.11


Square Antiprism (8 probes)
30.51
Average: 18.25


Centered Octahedron (7 probes)
29.70
Average: 24.73

The tests seem to indicate that Cube is the "best" of the tested formations. The difference between square antiprism and centered octahedron imo is too small to say for certain it one or the other is actually better.

[Yankunytjatjara]

A useful trick, in case anyone didn't know, is that you can aim your dscanner with the system map open, so you can line up your ship and your probe cube with the camera in the system map, and use a 15-degree scan to see if you've got the probe in the right spot.

Really? I've always had the impression that the D-scan direction is weird with the map open... Will try that!

[Wrack]

Cube (8 probes)
Average: 19.11

Square Antiprism (8 probes)
Average: 18.25

Centered Octahedron (7 probes)
Average: 24.73

The tests seem to indicate that Cube is the "best" of the tested formations. The difference between square antiprism and centered octahedron imo is too small to say for certain it one or the other is actually better.

They really don't. The fact that the cube and antiprism each got a very strong hit 1 time in 5 does nothing to support a claim that they are "best." That's why I did it 5 times, because probe deviation is important, and probing reliability is important, and the cube and antiprism failed there.

The choice of how small to shrink each formation is subjective and inexact (which is why I gave typical screenshots, so you can see how I'm doing it) I'm sure if I shrank the octahedral patterns down farther, I could increase the strength of their hits while making them unreliable too. Would that make octahedral best all of a sudden?

[Lady Spank]

Drop 6 probes, pull 1 up. 1 down, pull the others out NSEW, start scanning. Haven't had any concerns with the speed or practicality of this yet. Max skills, no imps. :shrug:

[joe space]

Really? I've always had the impression that the D-scan direction is weird with the map open... Will try that!

dscan isn't weird with the map open so long as you remember to center system map on your current location. it works well with things outside your own cluster, but not with things in your own cluster.

also, you can use the tactical overlay in map mode and it will give up a tactical overlay in AU. so after you use dscan to get distance and direction you can use tactical overlay to get a pretty accurate placement of probes on the first go.

using the dscan to make sure your probes are in the right spot, however, is only useful if you completely missed on the first go (since your probes only move to wherever you've placed them after you start probing). if you didn't completely miss your probe results are probably good enough to show you where to readjust to.

[Tsubutai]

Not quite sure what you're saying there, joe.

I will repost sekrit tuskers tricks from our forums regarding the use of the d-scanner/system-map/navigation panel to pin down the location of a probe target at an arbitrary point in space without dropping probes and then secure a 100% hit on the first scan:

1) Launch probes from an offscan spot, get them hidden by dragging them 20+ AU above or below the plane of the system and running a scan, then warp to a safespot in scan range of the target
2) Locate the target with your directional scanner. You'll need to get the range to the nearest AU* and narrow the angle down to at least 15 degrees; 5 is preferable
3) Open the solar system map, activate the tactical overlay, and hit F11 to bring up the navigation panel on the right of your screen.
4) Move the map camera such that you are looking 'down' on the plane of the system, then rotate the view such that the orientation of the system in the main map window matches that in the bottom window of the navigation panel.
5) Note the green cone** in the lowest window of the navigation panel. This indicates the area of the system covered by your narrowed-down d-scan. By comparing it to the main map and using the distance indicators provided by the tactical overlay, you can tell where the target is to within one cubic AU or so.
6) Run your first scan with probes at 4 AU, at the appropriate location. I use a 5-probe formation like that shown by EKR in the OP, but with the 'outer' probes at 4 AU and the 'inner' at 1. With max skills, this should net you a 100% hit on anything the size of a shield-tanked cruiser or larger. If your first scan is slightly off, move your probes to 'centre' of the first hit and re-scan without adjusting your probes' formation or radii; the second should certainly nail them.

* Note that 1 AU = 150,000,000 km (and so 7 AU = 1050,000,000 km and 14 AU = 2100,000,000 km, etc. etc.).

**Sometimes this cone doesn't show up when you first hit F11. If so, adjust the d-scan angle up a notch, then back down to its original value; this should cause the cone to appear.

The stuff about the 5-probe formation is outdated; nowadays I use an octahedron for regular targets and a cube for things I suspect are "unprobeable."

[joe space]

wasn't aware of f11 and a green cone thingee. other than that, pretty much what i do when i think the thing i am probing cares about being probed (except i try to get to a off d-scan celestial before launching probes and sometimes drag +15 au between scans if not getting 100%.)

[Smuggo]

Drop 6 probes, pull 1 up. 1 down, pull the others out NSEW, start scanning. Haven't had any concerns with the speed or practicality of this yet. Max skills, no imps. :shrug:

Aside from the initial setup it shouldn't affect speed as you can simply use the alt+drag to collapse the arrangement down so once the shape is set it's just as quick to narrow down.

[Jalif]

Intresting thread and sorry for diggin up a old thread (still seems relevant tho)

I read that the border of being "un-probable" was around 1 to 1.1 and that it would take maxed skills to get you.

But what if you have a signature radius of 168 & a sensor strength of 253? Probably overkill, but would that take you 1h before you have a hit? Very skilled prober maybe 15min?

Or would it be better to "downgrade" to just 164 sensor strength?

[Tsubutai]

Intresting thread and sorry for diggin up a old thread (still seems relevant tho)

I read that the border of being "un-probable" was around 1 to 1.1 and that it would take maxed skills to get you.

But what if you have a signature radius of 168 & a sensor strength of 253? Probably overkill, but would that take you 1h before you have a hit? Very skilled prober maybe 15min?

Or would it be better to "downgrade" to just 164 sensor strength?

There is a hard cap on your effective sig radius: sensor strength ratio for the purposes of probing; once you're below the cap value, further increases in sensor strength or decreases in sig radius have no effect. In your specific example, going to 253 SS would have no impact whatsoever. A skilled prober can hit a threshold ship in less than a minute under ideal conditions; I'm kind of rusty these days, so it sometimes takes me a bit longer.

(FWIW, the old unprobeability threshold was 1.08)

[Daneel Trevize]

Ah but what's the graph of the impact if they're slightly lower skilled probers/don't have all possible related implants? Is there a ratio range that's still unprobable except to the few?