Last night's EE testing results

edited October 2015 in EmptyEpsilon
Finally was able to get some multi-player testing on EmptyEpsilon for 2 hours last night with a couple of friends who help me run games at conventions.

I had some crashing issues on Windows 8 when trying to connect to a Windows 7 server. Older builds ran fine back in September. It might be my build of it that's causing problems. Will have to test the official builds to see if the results are the same. Also had issues detecting and connecting to the same server from my Linux build. Ran fine under Wine on that machine though.

My touchscreen Windows 7 x86 tablet didn't want to press buttons when using the touch screen. I didn't configure the options.ini for touchscreen though. We had enough problems with some of my boxes, as mentioned above, so we hooked up a keyboard and mouse to that tablet.


One thing we discussed after playing it is that Science scans are too hard for new players. This would be especially true for conventions, where time to get used to those controls are limited. It would be nice for an option to vary the difficulty/number of scans on the server screen. What I'm thinking is 3 settings: One that is the same as it is now; the second that reduces the amount of "slider screens" from 4 to 2 for full scans; the third that is a click & wait similar to Artemis. The click and wait should take longer than it would if you do the sliders by hand. The 2nd option (just 2 slider screens) should have some sort of delay for "processing of data" since you aren't doing it by hand. This should even out the reduced difficulty and make it more advantageous to run the full 4 slider scans.

They were inpressed on how it looks and how it plays as compared to Artemis. They also mentioned that the AI is much smarter.

In the next couple of weeks we should be able to get together and do some more testing since I'm wanting to run EE at our next Convention on Nov 14-15.

Edit: changes for clarification

Comments

  • I'll see what I can do for the science scans. Shouldn't be to hard to make that configurable. Note that the old "just wait" code is still in there, it just never runs normally. (Actually, it's configurable per "object" that you are trying to scan, and some objects could still be marked as just a delay instead of a slider set)

    Right now, ships are always set that you need to do 2 "cycles" of scanning. With 1 slider per cycle for the basic scan, and 2 sliders per cycle for the full scan.
    I never had much issues with the 2 sliders. But I can understand that it's hard to grasp the mechanic. I can however, solve the 4 slider variant quite well now. Which is why I set the "normal" level to 2 sliders.

    Maybe making it an option when starting the server:
    **scanning type**
    =Basic delay - Old method, just a delay
    =Simple - New method, 1 cycle per scan, 1 bar for both
    =Advanced - New method, 2 cycles per scan, 1 bar for first, 2 bars for 2nd scan (Current setting)


    Not sure how much "smarter" the AI is in EE. AI is tough to program. But there are some randomizations, and decisions made by the AI. It's not all "just attack the player, always". (Actually, the Artemis AI is stack based. But that does not allow for much dynamic behavior)
    EE AI does take in a lot of factors for choices on who to attack when there are multiple targets available, which is especially noticeable when you have helper fighters or are near stations. And it fires missiles into your current flight path, so missiles will hit you if you do not maneuver or change speed. Even if the missiles would not home a bit.


    (As for the crash, I also noticed a random crash when starting/connecting to a server. It's rare, so it's a bit tough to pin down)
  • Maybe making it an option when starting the server:
    **scanning type**
    =Basic delay - Old method, just a delay
    =Simple - New method, 1 cycle per scan, 1 bar for both
    =Advanced - New method, 2 cycles per scan, 1 bar for first, 2 bars for 2nd scan (Current setting)

    That's what I was thinking.


    I think the crash bug I'm seeing has to do with my builds (Win 32bit built on Debian Linux). They work fine on XP and seem to work OK on Windows 7. Windows 8 (32bit) doesn't like them.
Sign In or Register to comment.