Home World Forum
Stars! AutoHost web forums

Jump to Stars! AutoHost


 
 
Home » Stars! 2.6/7 » The Academy » Hello again Stars-Autohost! (Old-time player returns.)
Re: Hello again Stars-Autohost! Wed, 18 January 2017 04:19 Go to previous messageGo to next message
XAPBob is currently offline XAPBob

 
Lt. Commander

Messages: 957
Registered: August 2012
I suspect that CC and M9M are actually agreeing in this case...

No benchmark is perfect, and rarely do they survive contact with the enemy, but this one is 'the least bad we have'.

It somewhat overvalues the 'all expense except W' races - because the other tech requirements are pretty low - it's not a bad ship design, but compared with ships in other eras it does make 'five expensive' research much more attractive than it should be.

Note that the 3.5 cheap tech gets a 55% boost in terms of the long game (the Omega Nubian example) but has to survive through the lower boost in the early game...
This is one of the perpetual balances of the game IMHO - but it's worth being aware of what it is you are testing with any test bed.

In this case the expensive tech needs to buy something reasonably significant in a long game - like 55% more resources, call it 30% to account for the earlier periods of lower differential.

Report message to a moderator

Re: Hello again Stars-Autohost! Wed, 18 January 2017 09:31 Go to previous messageGo to next message
Entropicurity is currently offline Entropicurity

 
Petty Officer 1st Class

Messages: 62
Registered: November 2007
Location: Northern Indiana, USA

Wow! I started off saying "Hello everyone!" and already getting a lot of feedback on things I've since lost touch of. Thanks to EVERYONE who has provided a response so far as I'm learning a lot with each new reply.

From what I can understand the Testbeds mentioned provide "snapshots" of how a race could work and has a secondary benefit of forcing a mindset that can be applied in real games. If one can't maximize their own Economy (by shooting for 25k by 2450) then they will have a significant difficulty getting the most out of their race in tight situations.

With the BB Test, you are applying yourself to War Tech and progress, though the Testbed uses Maximum minerals to get them out as soon as possible and assumes the best case scenario. Furthermore the Test takes into account tech that is available to everyone. The one thing I didn't know about was the following from CC:

ccmaster wrote on Tue, 17 January 2017 23:45
Frigates weapon 8 ( beams)
Frigates Weapon 9/10 ( beams)
CC Weapon 10 ( beams)
CC Weapon 12 ( Missile)
BB Weapon 12 ( Missile)
BB Weapon 16 ( Beams some times Missile)
BB Weapon 22 ( Beams)


This piece of history gives me a good snapshot of what to expect in each of the "Eras" and something I've had a lot of difficulty finding. I'm trying to piece together my own "guide" on each of the different In-Game "Eras" on a Google Site project I've been working on that will take in all this information and chew it out into my own words. Being able to apply this with each different economy helps get a good feeling of what to expect from other races as well with enough experience.

What it boils down to like with any evaluative scenario that YMMV (Your Mileage May Vary) with each testbed, though they allow you to get a good snapshot of what to expect in the most ideal situation from yourself and from enemy civilizations!

Please let me know if I am missing anything!



--
"A gem is not polished without rubbing, nor a man perfected without trials." - Chinese Proverb

Report message to a moderator

Re: Hello again Stars-Autohost! Wed, 18 January 2017 19:41 Go to previous messageGo to next message
talkingbologna is currently offline talkingbologna

 
Senior Chief Petty Officer

Messages: 86
Registered: November 2016
Location: 1947
In a small/packed, regular minerals universe, BB-gorillas, organic, IS-10, s.computers, and armageddons by '60 isn't that hard. This is also with AI enemies. I never could stand to sit here and play out, properly, a game with no enemy. Always got sloppy at about 2420. The '60 mark didn't include producing 100 of them, nor with jammers. I did, however, always have regular defensive fleets of hundreds of yakorma DDs and colloidal CCs, and orbitals with Driver-7 or -10, as is required to keep from losing all those last planets I had to take from the PP AIs.

Even if I played out a testbed, though, I don't think I could get anything like that by the '40s. I'm usually struggling to get EITHER bazooka CCs or colloidal DDs by '30-35, depending on setbacks and luck.

Report message to a moderator

Re: Hello again Stars-Autohost! Wed, 18 January 2017 23:17 Go to previous messageGo to next message
magic9mushroom is currently offline magic9mushroom

 
Commander

Messages: 1361
Registered: May 2008
talkingbologna wrote on Thu, 19 January 2017 11:41
In a small/packed, regular minerals universe, BB-gorillas, organic, IS-10, s.computers, and armageddons by '60 isn't that hard. This is also with AI enemies. I never could stand to sit here and play out, properly, a game with no enemy. Always got sloppy at about 2420. The '60 mark didn't include producing 100 of them, nor with jammers. I did, however, always have regular defensive fleets of hundreds of yakorma DDs and colloidal CCs, and orbitals with Driver-7 or -10, as is required to keep from losing all those last planets I had to take from the PP AIs.

Even if I played out a testbed, though, I don't think I could get anything like that by the '40s. I'm usually struggling to get EITHER bazooka CCs or colloidal DDs by '30-35, depending on setbacks and luck.


- Accelerated BBS start and Beginner: Maximum Minerals are assumed for this benchmark, IIRC.
- The benchmark's set for a true testbed, no enemies.
- The 2440s claim probably comes from very finicky and optimised races, monster CA or -f HE, that are meticulously micromanaged. It's a far, far higher bar than 25k by 2450; my guess is that the sort of races that could do it would get at least 50k by 2450. Certainly, something like my Mimigas (a playable 3i HE that make 45k-50k by 2450) couldn't do it; I'm guessing they'd clock in at something like late 2450s.

Report message to a moderator

Re: Hello again Stars-Autohost! Thu, 19 January 2017 09:00 Go to previous messageGo to next message
Entropicurity is currently offline Entropicurity

 
Petty Officer 1st Class

Messages: 62
Registered: November 2007
Location: Northern Indiana, USA

talkingbologna wrote on Wed, 18 January 2017 19:41
<snip>I never could stand to sit here and play out, properly, a game with no enemy. Always got sloppy at about 2420...<snip>


I would imagine then that Testbeds aren't necessarily your cup of tea, which there is nothing wrong with that. The "testbeds" are nice to run A) When you have the time and B) to get a good snapshot of when you will expect to meet each "tier" or "era" of ships. If you think you can do that just find in a contested universe, that works well!

Another benefit of this is also in situations where possibly you are able to use diplomacy really well to reach a key benchmark and then race for a win after you've reached it. Personally I got sloppy with these around 2440, though by that point people typically have a good idea of what to expect by the 2450.

The Acc BB and Max Mins is typically there to reduce micromanagement, and to provide "emulated" situations where you are assumed to have what resources available. In a real game, this is filled in with micromanagement, diplomacy, technology trading, etc.

Another factor to consider is that the AI are sloppy, and popdropping to gather Tech from the neighboring races who tend to colonize EVERYTHING can truly impact and alter your testbed in a way that can't easily be managed by the time you get in with other humans who can project their weight around more effectively. Something to consider I wager.

Personally although I may not enjoy it as much, it does help me apply my MM skills and provides a good snapshot of each "step" of the game. Right now I'm realizing about what time to expect the Frigate/Cruiser Rush with each of my race designs that I testbed and when looking at a larger strategy for how I plan to "win" I can apply that knowledge each step of the way!



--
"A gem is not polished without rubbing, nor a man perfected without trials." - Chinese Proverb

Report message to a moderator

Re: Hello again Stars-Autohost! Thu, 19 January 2017 10:06 Go to previous messageGo to next message
XAPBob is currently offline XAPBob

 
Lt. Commander

Messages: 957
Registered: August 2012
Testbeds are good for comparing races/playstyles yourself...

Beware of trusting the AI though - they cheat (not in the game, just in the race wizard) so if you enable PPS you will find yourself a mile behind at 2420 - that's quite normal...

Report message to a moderator

Re: Hello again Stars-Autohost! Thu, 19 January 2017 12:15 Go to previous messageGo to next message
Entropicurity is currently offline Entropicurity

 
Petty Officer 1st Class

Messages: 62
Registered: November 2007
Location: Northern Indiana, USA

On the topic of testbeds, I was reading up on the Challenges and found the following very interesting/entertaining:

http://starsautohost.org/sahforum2/index.php?t=msg&th=54 12&start=0&rid=1146

Forced gened to 2450, looks like the challenge was bested by 2519 (about 70 years!) May have to give this one a try myself!



--
"A gem is not polished without rubbing, nor a man perfected without trials." - Chinese Proverb

Report message to a moderator

Re: Hello again Stars-Autohost! Tue, 31 January 2017 14:08 Go to previous messageGo to next message
talkingbologna is currently offline talkingbologna

 
Senior Chief Petty Officer

Messages: 86
Registered: November 2016
Location: 1947
Quote:
popdropping to gather Tech from the neighboring races who tend to colonize EVERYTHING can truly impact and alter your testbed in a way that can't easily be managed by the time you get in with other humans who can project their weight around more effectively


Very true. I guess I just can't stand to spend all those hours and days slightly altering a race for each new test verse.

I also tried that challenge, Entropicurity, but only the once. I tried to produce yakorma dd's as my first fleet and was clobbered before I even had the tech. I read in that thread that the AI tend to ignore you if you don't expand or leave HW, so maybe it was my scouting that drew them in.

Quote:
if you enable PPS you will find yourself a mile behind at 2420


That's not been my experience. Usually, I'm a mile ahead of them by 2420, and the HE and PP catch and pass me by about year 60 or 70 (cause of planet and base points). I still ruin them in combat, though. For some reason, they never seem to produce more than a certain number of ships. In the beginning, back when I was playing on normal difficulty, I saw a fleet of 400 dd's belonging to a race (Cattle Prods, if I remember right).

Report message to a moderator

Re: Hello again Stars-Autohost! Thu, 02 February 2017 04:54 Go to previous messageGo to next message
iztok is currently offline iztok

 
Commander

Messages: 1202
Registered: April 2003
Location: Slovenia, Europe
talkingbologna wrote on Tue, 31 January 2017 20:08
[(AI)...
For some reason, they never seem to produce more than a certain number of ships.

You can check what AI is doing by looking into AI turns files. The password is "viewai".

Report message to a moderator

Re: Hello again Stars-Autohost! Thu, 02 February 2017 22:40 Go to previous messageGo to next message
magic9mushroom is currently offline magic9mushroom

 
Commander

Messages: 1361
Registered: May 2008
iztok wrote on Thu, 02 February 2017 20:54
talkingbologna wrote on Tue, 31 January 2017 20:08
[(AI)...
For some reason, they never seem to produce more than a certain number of ships.

You can check what AI is doing by looking into AI turns files. The password is "viewai".

If you press F8 while looking at the AI, you can see their races. They're about 1000 points overdrawn on Expert. Not particularly efficiently designed, though.

Report message to a moderator

Re: Hello again Stars-Autohost! Fri, 03 February 2017 14:38 Go to previous messageGo to next message
Entropicurity is currently offline Entropicurity

 
Petty Officer 1st Class

Messages: 62
Registered: November 2007
Location: Northern Indiana, USA

To be honest, I am not entirely too sure as to what the AI consisted of for this game. It looks like they had some basic strings to work with and used the overspent races to make up for the lack of MicroManagement-enabled AI. Then proceeded to "COLONIZE ALL THE THINGS!"

As for ship design, I think the HE had the most interesting ones, but as of yet never really ran into any significant problems. Even if one of the AI's got a little ahead of me, dropping on their poorly maintained planets was enough to grab a few tech and push ahead.

Would be very curious to see how a more robust AI would handle a game like this?



--
"A gem is not polished without rubbing, nor a man perfected without trials." - Chinese Proverb

Report message to a moderator

Re: Hello again Stars-Autohost! Fri, 03 February 2017 15:55 Go to previous messageGo to next message
talkingbologna is currently offline talkingbologna

 
Senior Chief Petty Officer

Messages: 86
Registered: November 2016
Location: 1947
Yeah, me too, Entropicurity. Never played any games with great AI strategists. I suppose it would take true artificial intelligence, to which literary history has not been kind. Could you imagine, one of us inspiring the awakening of man's greatest potential enemy, just to have a better class of computer opponent?

Thank you for the pass, Iztok, but I knew it. I suppose a more accurate assessment would be that I never encounter more than a few of their ships--never enough to pose a threat unless I am woefully behind and then they clobber me. And yeah, AI seems to go for those expensive LRT plus' a lot like TT, UR, and MA.

Report message to a moderator

Re: Hello again Stars-Autohost! Sat, 04 February 2017 02:19 Go to previous message
magic9mushroom is currently offline magic9mushroom

 
Commander

Messages: 1361
Registered: May 2008
Entropicurity wrote on Sat, 04 February 2017 06:38
To be honest, I am not entirely too sure as to what the AI consisted of for this game. It looks like they had some basic strings to work with and used the overspent races to make up for the lack of MicroManagement-enabled AI. Then proceeded to "COLONIZE ALL THE THINGS!"

As for ship design, I think the HE had the most interesting ones, but as of yet never really ran into any significant problems. Even if one of the AI's got a little ahead of me, dropping on their poorly maintained planets was enough to grab a few tech and push ahead.

Would be very curious to see how a more robust AI would handle a game like this?

Micromanagement is actually one of the things the Stars AI does fairly well. Where it appears to do poorly is macromanagement and race/ship design.

Report message to a moderator

Previous Topic: Getting really frustrated with some things
Next Topic: How do you like your Germanium?
Goto Forum:
  


Current Time: Fri Apr 19 04:30:38 EDT 2024