Automatic for the capsuleer

16th November 2012 – 5.34 pm

On the EVE Online forums, Ellente Fervens makes a request that crops up occasionally: the directional scanner should have an option to update automatically, shifting a tedious and repetitive task away from the player to the computer. Conceptually, I agree with this idea. This is sci-fi, and we are piloting complex ships no doubt stuffed with computing power. It should be possible to give the ship computer a set of commands such as, 'keep d-scan updated, and if an unknown ship or scanning probes are detected then initiate an emergency warp back to the tower'.

Practically, however, I do not agree with such a change. EVE Online is not just a sci-fi environment, it is also and primarily a game. The interest and entertainment comes from player skill, and the progression of that skill. If you take away the skill from the game, by placing control in the hands of a computer, you are taking away the reason to play the game. This, I would argue, is one reason why AFK-[anything] is considered wrong, or at least unsporting. You play a game to be involved, and removing that involvement is seen to be detrimental to enjoyment.

As an example, the act of scanning is obviously algorithmic. Arrange probes in a set pattern, scan, pick a signature, centre the probe pattern on that signature, reduce the range of probes, scan. Repeat until signature is 100% or probe range is minimised, bookmark and/or ignore the signature, and reset; repeat until all signatures are bookmarked or ignored. In reality, there should be no pilot interaction for scanning. Launch your probes, let the computer do everything, and analyse the results at your leisure. But the lack of interaction gives no reason to be at the controls, to be playing the game.

Similar arguments could be made for other aspects of New Eden life, such as planetary interaction. I don't know myself, but I suspect there is an algorithmic method for finding sources of planet goo, extracting it, and shipping it to space. It would save a lot of fiddling around if you could just tell your computer to do it for you. And it would probably do a better job, too. Manufacturing also has algorithmic properties, when considering what blueprints to buy and research, and when to install lots of separate jobs to get a resultant ship off the production line. This isn't unusual.

Part of the reason there are so many processes that can be reduced to automation is that EVE Online is essentially a simulation, already running on computers, and already reduced to algorithms. This is clearly demonstrated by acknowledging that mission-running can be automated. The rats are always the same, and will always react the same way, because that's how they've been designed. Modules, planetary equipment, and probes will never deteriorate, so each subsequent process will be exactly the same as the previous one. Anything that has been programmed can be automated.

That's not to say scanning shouldn't be done automatically. But if it is, there must be some reason for players to want to scan manually. Generally, a common theme in gaming is to reward player interaction over player non-interaction. In this case, it would be that automated scanning would take at least twice as long as manual scanning. Note how this still makes no sense from a realism viewpoint, as a computer would be able to arrange probes much more quickly than a human. But if the processes were realistic then player interaction would be detrimental to progress, which is antithetical to the idea of gameplay. And there is a firm precedent, as autopilot takes you more slowly to your destination than does manual navigation.

If you consider the parts of the game that cannot be automated, that are not algorithmic, you will see a common pattern: player involvement. PvP is the defining aspect of EVE Online, and with good reason. Once you get other players, other people, involved, predictability is thrown out of the window. There may be aspects of PvP combat that can feel repetitive, or could be done more quickly by a computer, but there is always variation, always uncertainty, and that cannot be easily accounted for by an algorithm. The heart of EVE Online shows that player interaction and skill is rewarded over mechanically reproducing certain actions.

Now to bring the discussion back to d-scan updating automatically. Yes, conceptually it should be possible. Computers can do this easily. And there seems little reason to object to it. After all, the player still needs to analyse the results, and ephemeral ships or probes can still be missed if a capsuleer takes a drink or is otherwise distracted. But this isn't futuristic spaceship life, this is a game. Player interaction needs to be rewarded over player non-interaction. Automated d-scanning, if it happens, would need to update slowly, and significantly slowly than is possible manually.

A thirty-second automatic update of d-scan may be acceptable to consider. Any less and d-scan would come close to being the w-space version of a populated local channel, negating the base concept behind w-space. You could argue that this period is too long, and a pilot would be better off updating d-scan manually. Well, that is pretty much my point. An automated d-scan updating every thirty seconds is better than nothing, but updating it manually is better still. There needs to be risk, but specifically risk that can be mitigated by players being alert to those risks. There is limited automation in games for a reason, even if those reasons deny realism. The game needs to be player driven, lest we end up with computer-vs-computer gameplay whilst us meatbags with our slow-firing neurons sit back and watch what is essentially alternative TV.

  1. 7 Responses to “Automatic for the capsuleer”

  2. I agree 100%. As much as I detest having to manually d-scan, making it automatic removes a key feature of WH space.

    But. How about making range and resolution a skill? How about mods that improve either?

    By Splatus on Nov 16, 2012

  3. My problem with this line of argument is that your equating interaction with choice, the act of data gathering with that of data analysis.

    The other concern I have with this is the hunter target issue. It is a system where the hunter needs to get lucky once while the target has to always be lucky.

    The result is that you end up testing two very different things. For the hunter you are testing skill/knowledge and some small amount of care (remaining cloaked etc). While on the target side you are testing attention span (measured over hours in some cases) and tedium tolerance.

    By tnankie on Nov 16, 2012

  4. I think the test of tolerance is more a problem with PvE vs PvP, tnankie. PvE is generally a mindless grind, and mashing d-scan only adds to that. No, I don't much like keeping d-scan updated when pew-pewing Sleepers, as it seems to demand constant attention for what is a relatively mindless task. I had the same problem when fishing in Azeroth, so you'd think I'd be more sympathetic.

    As I have replied on the forum:

    ...on further reflection, automated d-scanning would also allow d-scan to be actively monitored whilst looting, off-loading cargo to a can, communicating in chat channels, updating bookmarks, and any other action that would take focus away from the d-scan window. That, I would argue, is where automated d-scanning becomes unfair.

    and that maybe d-scan could be updated at five-second intervals when its UI window retains focus. Maybe ten seconds. I think it would need testing.

    I don't agree that hunters only need to get lucky once. It's more of a chain of luck that's required, from opening a wormhole, to entering the system, launching probes, getting a good scan, and reaching the target in time. This is complicated should a fleet comprising ships that can't cloak be involved. Getting caught at any of those steps leads to failure.

    Opportunities aren't so common, and failures generally result in lock-down, such that the balance needs to be tipped in favour of the hunted. I think it's pretty fair currently. And it feels to me that most suggestions to improve d-scan fall in two camps: change the range gate to be in AU, dammit; and any number of ways to make d-scan protect the hunted. Perhaps I'm biased in my perception.

    So, Splatus, I don't think there should be any modification to d-scan's basic functionality based on skills or modules. D-scan can't update any more quickly than it can now, and increasing the range would only serve to protect the hunted. I don't see how greater range benefits the hunter in any way. So d-scan would have to be throttled back for skills or modules to have an effect, with training bringing us back to where we already are. That is undesirable for all concerned.

    But the range gate really does need to be changed to read AU. There is no reason for it to be in kilometres, even if a rough conversion to AU is possible. Besides, the conversion requires typing a whole bunch of zeroes, whereas AU can be input with single digits. On top of that, the maximum d-scan range is arbitrary and therefore meaningless. There are occasions when I am 14·something AU from a planet and wondering if it is in d-scan range or not. This is silly. Change the range gate to read in AU and add a range cutoff to be at 14 AU exactly. This would greatly aid clarity to the interface.

    By pjharvey on Nov 16, 2012

  5. Hm. Not sure I agree. Probe scanning has 3 (4?) skills associated with it, d-scan none. Probe scanning has rigs, mods and even implants that improve the skills and efficiency. d-scan has nothing. It's just there. As with all things in eve, abilities could come by choice. "I choose to be a miner / hunter, so I need better d-scan skills.". Your argument is valid if you assume that the skills improve for the current level. I say it is possible that the base ability for d-scan could be nerfed. All in theory of course.

    Am I advocating this change? Not necessarily. I am advocating to rethink game mechanics occasionally and not take unfinished and ugly code (Pos!) for granted.

    And if they do make it skill based and nerf the base level, you read it here first and you may pod me. If you can find me. ;-)

    By Splatus on Nov 17, 2012

  6. I suggested d-scan would need to be throttled back for any skills to have effect. But rather than reducing range or angle, how about reducing the information given?

    Basic d-scan works out to 14 AU and 360 degrees still, but only shows objects in space. Skill training then lets you discern between ships, drones, and probes; then hull type—such as frigate, cruiser, battleship; and then ship type, drone type, and probe type—core or combat.

    That would let it still show information, but only need to be trained if d-scan is actually going to be used. The hunted only need to see ships and probes, but hunters will want to know ship types and so train further.

    By pjharvey on Nov 17, 2012

  7. Yes, that could work. Agreed

    By Splatus on Nov 17, 2012

  1. 1 Trackback(s)

  2. May 7, 2013: Tiger Ears » Blog Archive » Automatic for the capsuleer redux

Sorry, comments for this entry are closed.