[TMA] New Probing Mechanics

[Alkeena]

Incarna changed they way probing works, though I'm not quite sure of the specifics. What I do know:

-All 8 probes now return a result
-It is no longer possible to be unprobeable, though it's still supposed to be hard to get a result

So, what's the best probe formation? I would assume a cube, but I've heard suggestion that that may not be optimal?

I have a few unprobeable gang-link ships hanging around--what's the ETA of hostiles pushing my shit in once I uncloak and fire up the links, assuming I continue to use an old unprobeable fit?

Thanks.

[Mr Marram]

An undocumented change was that all probes now have half the scan stength of before, this was either by a bug in the coding or permenant, one of the mini-patche notes touches on ships not giving bonuses.

[Tsubutai]

You need a virtue set plugged in to probe down "unprobeable" setups. There does appear to have been a reduction in individual probe strength, although I'm not sure that it's been halved. For just probing sigs and regular probeable ships, the cross formation still works fine. For "unprobeable" setups, the cube works but is a little tiresome to set up. The octahedron also works and is rather quicker to set up, but you're going to need all the skills, implants, rigs, etc. turned up to 11 to hit unprobeable setups. Nabbing unprobeable setups requires a hit with at least six probes at 0.5 AU radius in a rather tight formation, so the answer to "how long does it take to probe one down" is "the amount of time required for your virtued prober to get a fix of that quality." I've done it in about a minute on an oblivious target and rather longer on a second who was wary and kept warping around when the probes got near.

[Samp]

I'm going to run some expermements tommorow - on the face of it formations with a center probe will be more effective since the boost in % it gives now won't be offset by a loss of positional accuracy, where as previously it would discount one of the box verticies. This will ofc always mean having one probe scannable by your victim even in a wide scan.

[Mr Marram]

Someone also mentioned that probes ontop of each other will only give the the strongest signal and not duplicates (it disregards one probe).

This means the old 7 probe method of having 4 in a cross and 3 of different sizes in the middle is useless.

[Dray]

I did some testing this morning and the results weren't to impressive, I used my main Dray with Astrometrics 5 and the support skills at 4 in a cheetah, I have cov ops 5 and used sisters kit as well as fitting 2 t1 grav cap upgrade rigs.

The tengu alt is max skilled with 3 eccm mods and fury missiles loaded.

With out implants I didn't get close but this was expected, I plugged in a low grade virtue set and the hardwire for scan str, best result was 43% with 8 probes overlapping, in fact until I switched off 2 of the eccm mods I couldn't get the tengu, with 2 running and furies loaded best hit was 87%.

The sensor str and sig on the tengu was 194.8/205 with 3 eccm, 125.8/205 with 2 eccm, and 68.6/205 with 1 eccm.

I'm guessing you will need a max skilled prober with a virtue set and hardwires to get a max skilled eccm tengu but I was surprised that I didn't get closer then 43%, of course I don't know how broke Sisi is so maybe it's different on Tranq.

Anyone else been testing?

[Dray]

Someone also mentioned that probes ontop of each other will only give the the strongest signal and not duplicates (it disregards one probe).

This means the old 7 probe method of having 4 in a cross and 3 of different sizes in the middle is useless.

I tried this method and variations with a centre probe and the results were strange got a 70% hit with 1 centre and 7 around in octagonal pattern -1 for the centre probe, when I moved the centre probe out to complete the octagon I got a 74% hit then moved it back to the centre and tried to reduce the overlap while still maintaining coverage and the hit percentage dropped off to 60% +/- a few, I wasn't expecting to get a 100% hit without max skills but I was expecting to get pretty close with astro 5 and support skills at 4 and all the virtue imps/hardwire, so I guess I was on course for what I expected still a bit surprised by the inconsistency of hit str.

[sassy b]

Dray - it sounds like all your probes are arranged as a 2d shape on a single plane, which is going to give pretty strange/bad results.

The optimal arrangement for scan strength will still be to have the probes evenly distributed around the signature in 3d space.
Therefore: 4 probes = tetrahedron, 6 probes = octahedron, 8 probes = cube

Reposting my cube technique from the incarna patch notes thread:

Launch 4 probes, Shift-drag probes vertically an arbitrary distance, launch remaining 4 probes, then make squares on the two horizontal plane

imo not significantly more mouse-intensive than an octahedral method - two extra probes launched, one extra click-drag.

[Cydo]

I do something like this:



One probe at each point and one in the centre (7total). If I really need it (i haven't yet) i can put a second centre probe down with a smaller radius.

Only got astrometrics + supports @ 4, cov ops @ 4, no virtues. Having no problem probing out anything, albeit I haven't found an "unprobable" to probe out since the patch.

[Dray]

Dray - it sounds like all your probes are arranged as a 2d shape on a single plane, which is going to give pretty strange/bad results.

The optimal arrangement for scan strength will still be to have the probes evenly distributed around the signature in 3d space.
Therefore: 4 probes = tetrahedron, 6 probes = octahedron, 8 probes = cube

Reposting my cube technique from the incarna patch notes thread:

Launch 4 probes, Shift-drag probes vertically an arbitrary distance, launch remaining 4 probes, then make squares on the two horizontal plane

imo not significantly more mouse-intensive than an octahedral method - two extra probes launched, one extra click-drag.

Yeah I was probing on a single plane, more consistency with the 2 plane method, but I don't have a max skilled prober to see how easy it is to catch a 3x eccm tengu.

[joe space]

i tested cube verse octahedron. unprobable loki v. cov ops (no virtues, no sisters, rangefinding 4, covops 5, 2x).

warped to loki and sorted probes with loki in center.
octahedron: low 70s
cube: mid 60s

i only did it a couple times. afterward i thought it may have been because i was getting the eight probes as well centered as the 6 probes.

anyway, what i would like to know - can you probe an unprobable with every bonus except rangefinding V?
I will train it, but i want to know if i should plug in virtues and hardwirings now or wait until rangefinding is complete,

[Dray]

i tested cube verse octahedron. unprobable loki v. cov ops (no virtues, no sisters, rangefinding 4, covops 5, 2x).

warped to loki and sorted probes with loki in center.
octahedron: low 70s
cube: mid 60s

i only did it a couple times. afterward i thought it may have been because i was getting the eight probes as well centered as the 6 probes.

anyway, what i would like to know - can you probe an unprobable with every bonus except rangefinding V?
I will train it, but i want to know if i should plug in virtues and hardwirings now or wait until rangefinding is complete,

Common sense says you should but this is Eve, I would wait.

[Vortex]

I've personally found that a 7 Probe configuration works best. One probe in the center point at resolution X, 6 probes around it distributed evenly at resolution Y, covering the up/down/left/right/nadir/zenith positions. Resolution Y is set to one increment higher than X, and you use the center X probe as the targeting point and just stick it on the result directly. Using the alt-key, you can easily adjust the entire set of probes positions relative to the center (which doesn't move) with one movement, and physically move all of the probes as a bundle to reposition. Setting the Y probes to be touching the X probe sphere is very easy to do and seems to give the best results.

This method has been giving me roughly ~70% stronger results compared to just 4 probes in the left/right/up/down positions, and its almost as fast to set up and move around. I haven't been able to figure out an 8 probe configuration that is stronger without doing really weird configurations, and those are too impractically slow to employ. An 8th probe could conceivably just be set to maximum range as a spotter, but a spotter probe is too far out to improve the hit result, and would probably just slow you down with additional time spent dicking around in the interface.

I have not seen a decrease in probe scanning strength compared to pre-patch. My probing alt is in an orca ATM, but her probe launcher on it exactly matches what EFT (non-updated) is giving me. If you saw a decrease in strength, maybe it was the same damage-mod bug that people were getting earlier?

[Tsubutai]

i tested cube verse octahedron. unprobable loki v. cov ops (no virtues, no sisters, rangefinding 4, covops 5, 2x).

warped to loki and sorted probes with loki in center.
octahedron: low 70s
cube: mid 60s

i only did it a couple times. afterward i thought it may have been because i was getting the eight probes as well centered as the 6 probes.

anyway, what i would like to know - can you probe an unprobable with every bonus except rangefinding V?
I will train it, but i want to know if i should plug in virtues and hardwirings now or wait until rangefinding is complete,

You should be OK without rangefinding V if everything else is maxed. Using a Tengu with the electronics subsystem skill at IV, a single T2 gravity capacitor rig, and the 6% hardwiring rather than the 10% one, I was able to probe out an unprobeable dude today (although he escaped my clutches ). Since that put me around 19% below the maximum possible scan strength, compared to the 10% you lose by not having rangefinding V, you should be good to go with a set of virtues plugged in.

It's worth noting that IME, you need a considerably tighter probe formation to hit unprobeables than you do for 'regular' signatures or ships.

[joe space]

i tested cube verse octahedron. unprobable loki v. cov ops (no virtues, no sisters, rangefinding 4, covops 5, 2x).

warped to loki and sorted probes with loki in center.
octahedron: low 70s
cube: mid 60s

i only did it a couple times. afterward i thought it may have been because i was getting the eight probes as well centered as the 6 probes.

anyway, what i would like to know - can you probe an unprobable with every bonus except rangefinding V?
I will train it, but i want to know if i should plug in virtues and hardwirings now or wait until rangefinding is complete,

You should be OK without rangefinding V if everything else is maxed. Using a Tengu with the electronics subsystem skill at IV, a single T2 gravity capacitor rig, and the 6% hardwiring rather than the 10% one, I was able to probe out an unprobeable dude today (although he escaped my clutches ). Since that put me around 19% below the maximum possible scan strength, compared to the 10% you lose by not having rangefinding V, you should be good to go with a set of virtues plugged in.

It's worth noting that IME, you need a considerably tighter probe formation to hit unprobeables than you do for 'regular' signatures or ships.

thanks! unfortunately it looks like i'm going to have to sell the virtues i just bought and may not afford another set until after i've got rangefinding V anyway. derp. anyone want to buy a virtue set for 1.9 bil??

Vortex - that's helpful, but i already assumed 8 and 7 probe formations in the manner you described were best for narrowing a signature. what doesn't seem to be established is what is the best configuration granted all probes are at their smallest resolution and in the tightest configuration.

[Tsubutai]

thanks! unfortunately it looks like i'm going to have to sell the virtues i just bought and may not afford another set until after i've got rangefinding V anyway.

Sadness. FWIW, I've been told by People Who Know Stuff that if you have everything else at V, you can drop the Omega from the Virtue set and still nail unprobeables; that's consistent with my experience that you don't need absolutely everything maxed, just to be 90-95% or so of the way there.

[Vortex]

[quote=joe space]

i tested cube verse octahedron. unprobable loki v. cov ops (no virtues, no sisters, rangefinding 4, covops 5, 2x).

warped to loki and sorted probes with loki in center.
octahedron: low 70s
cube: mid 60s

i only did it a couple times. afterward i thought it may have been because i was getting the eight probes as well centered as the 6 probes.

anyway, what i would like to know - can you probe an unprobable with every bonus except rangefinding V?
I will train it, but i want to know if i should plug in virtues and hardwirings now or wait until rangefinding is complete,

You should be OK without rangefinding V if everything else is maxed. Using a Tengu with the electronics subsystem skill at IV, a single T2 gravity capacitor rig, and the 6% hardwiring rather than the 10% one, I was able to probe out an unprobeable dude today (although he escaped my clutches ). Since that put me around 19% below the maximum possible scan strength, compared to the 10% you lose by not having rangefinding V, you should be good to go with a set of virtues plugged in.

It's worth noting that IME, you need a considerably tighter probe formation to hit unprobeables than you do for 'regular' signatures or ships.

thanks! unfortunately it looks like i'm going to have to sell the virtues i just bought and may not afford another set until after i've got rangefinding V anyway. derp. anyone want to buy a virtue set for 1.9 bil??

Vortex - that's helpful, but i already assumed 8 and 7 probe formations in the manner you described were best for narrowing a signature. what doesn't seem to be established is what is the best configuration granted all probes are at their smallest resolution and in the tightest configuration.[/quotenn3ue0h]

I'm not sure I follow - the 7 probe configuration just scales its position with the alt key, so 32AU or 0.25AU scans are all the same pattern, you just scale them into position after adjusting the center probe's aimpoint. Can you clarify what you mean? As for 8 Probes, I haven't found a way to add an 8th probe that is helpful without adding way too much complexity to the pattern, maybe somebody is more clever than me in that regard vOv.

[joe space]

[quote=Vortex]

i tested cube verse octahedron. unprobable loki v. cov ops (no virtues, no sisters, rangefinding 4, covops 5, 2x).

warped to loki and sorted probes with loki in center.
octahedron: low 70s
cube: mid 60s

i only did it a couple times. afterward i thought it may have been because i was getting the eight probes as well centered as the 6 probes.

anyway, what i would like to know - can you probe an unprobable with every bonus except rangefinding V?
I will train it, but i want to know if i should plug in virtues and hardwirings now or wait until rangefinding is complete,

You should be OK without rangefinding V if everything else is maxed. Using a Tengu with the electronics subsystem skill at IV, a single T2 gravity capacitor rig, and the 6% hardwiring rather than the 10% one, I was able to probe out an unprobeable dude today (although he escaped my clutches ). Since that put me around 19% below the maximum possible scan strength, compared to the 10% you lose by not having rangefinding V, you should be good to go with a set of virtues plugged in.

It's worth noting that IME, you need a considerably tighter probe formation to hit unprobeables than you do for 'regular' signatures or ships.

thanks! unfortunately it looks like i'm going to have to sell the virtues i just bought and may not afford another set until after i've got rangefinding V anyway. derp. anyone want to buy a virtue set for 1.9 bil??

Vortex - that's helpful, but i already assumed 8 and 7 probe formations in the manner you described were best for narrowing a signature. what doesn't seem to be established is what is the best configuration granted all probes are at their smallest resolution and in the tightest configuration.

I'm not sure I follow - the 7 probe configuration just scales its position with the alt key, so 32AU or 0.25AU scans are all the same pattern, you just scale them into position after adjusting the center probe's aimpoint. Can you clarify what you mean? As for 8 Probes, I haven't found a way to add an 8th probe that is helpful without adding way too much complexity to the pattern, maybe somebody is more clever than me in that regard vOv.[/quote:1twa1pgw]

see sassy b's post. the cube uses 8.

i think i misunderstood what you wrote first. anyway, still would like to know whether 8 (cube), 6 (octahedron), or 7 (octahedron + 1 in the middile) are best

[Vortex]

I tested all of those various configurations a few days ago, and in my tests the 7 probe configuration (with center probe at 1 lower scan res increment) produced stronger results than any other. Its by definition a stronger set of probes than 6 due to the addition of the center point probe, and the cubical 8 probe configuration lacks a closer-range scan probe to increase the hit strength. Even if the cubical 8 probe could beat it out though, moving those probes is a serious PITA, and so any gains are lost due to the need to manually reconfigure each probe's position.

[sassy b]

Even if the cubical 8 probe could beat it out though, moving those probes is a serious PITA, and so any gains are lost due to the need to manually reconfigure each probe's position.

I don't get what you're talking about here. It's exactly the same number of click-drags as any other probe formation. Scale probes with alt key then recenter on the sig.