#savemwo Townhall #1: Discussion
#61
Posted 26 July 2013 - 06:43 PM
#62
Posted 26 July 2013 - 07:20 PM
#63
Posted 26 July 2013 - 07:38 PM
Geck0, on 26 July 2013 - 07:20 PM, said:
3pv was mentioned quite a bit at the meeting.
#64
Posted 26 July 2013 - 09:08 PM
Sybreed, on 26 July 2013 - 06:10 PM, said:
hat is one of the thing that is extremely unpleasant for me about MWO: mechs don't feel unique, they feel soulless. My question in the last ATD was answered by a big "I don't know", but the dev seemed to understand my concern and agreed with it. Won't be until a while we see something to make mechs feel more distinct, but I think it's an important point to discuss.
We can talk about the "uniqueness" factor of different mechs perhaps at the next meeting. In reality there was only so much we could talk about in the given time frame. Hell it still took 3 hours to get through it all.
I for one look forward to hearing the opinions of a bunch of the speakers again.
#65
Posted 26 July 2013 - 11:05 PM
#66
Posted 27 July 2013 - 01:16 AM
Gwaihir, on 26 July 2013 - 05:22 PM, said:
We actually play a good bit of Megamek, and someone even made this guy, forest colony in TT form: (It works surprisingly well there)
This was pre-boat in the lake, but close enough.
e: More fun with megamek, 20 fleas vs an Emperor, Nightstar, and Templar (aka Jumpjets + LBX20(s)). AoE ammo explosions and catastrophic engine failures enabled, naturally!
I have a shiny little card to show for it!
I have actually used that map form time to time, I know that a goon made that map, kudos to him.
Edited by Stormwolf, 27 July 2013 - 01:17 AM.
#67
Posted 27 July 2013 - 06:35 AM
Geck0, on 26 July 2013 - 07:20 PM, said:
I think part of the problem with 3rd person view is that it's such a "WTF?" that we don't really know how to react to it without seeing more of what PGI has in mind. I actually kind of think some form of external view (via the UAV drone, for instance) would be cool, but I just think it's inherently against the feel of an in-cockpit mechwarrior game to have it be the actual way you pilot. It would simply be a very, very different game.
#68
Posted 27 July 2013 - 07:04 AM
#69
Posted 27 July 2013 - 08:08 AM
Well, I'll just answer the last question. Megamek is just "Tabletop: the game, on your computer." It looks like those screenshots I posted above, so it's not the most beautiful thing in the world, but it has support for every type of unit and rule out there.
#70
Posted 27 July 2013 - 08:40 AM
Jackson Jax Teller, on 27 July 2013 - 08:14 AM, said:
Oh well, there was a very important breakthrough in the game balance department that made coolant viable, even though it helped out the massive alpha striking meta we have now....
..they realised they could make money out of it.
Edited by Rippthrough, 27 July 2013 - 08:40 AM.
#71
Posted 27 July 2013 - 08:45 AM
Gwaihir, on 27 July 2013 - 08:08 AM, said:
Well, I'll just answer the last question. Megamek is just "Tabletop: the game, on your computer." It looks like those screenshots I posted above, so it's not the most beautiful thing in the world, but it has support for every type of unit and rule out there.
Now I wanna play Megamek, I know what I am doing this afternoon
Welcome back to the game Jackson, I would be happy to answer some of those questions you have.
I will start at the top of your first long post, in regards to the beta test server. It is not as silly as it sounds. Beta is kinda loose term in gaming now, and in the free to play market, open beta really seems to translate into viable market product, something they can legally sell you to play. This accounts for the fact that anyone can play the game, nobody is selected based on computer set up, experience, etc. to actually test and provide feedback (like you see when you apply to alpha/closed beta projects), and they actually have a product to sell (MWOs case MC/Paint/Hero). Open beta to me is more of a test at how viable the game will be in the market.
Another recent game I would like to site that went from open beta to launch in a matter of months is Neverwinter. They run both normal and test servers during their entire time in open beta. This seemed to stream line major game mechanic changes, so that it did not effect the regular player base. Dedicated testers, who usually play on the test servers, got their hands on the content first, and would help in eliminating problems before it went to everyone else. Think of what this could have done for PGI when they first implemented Artemis, UAC/5 Jam mechanics, Etc Etc; all the broken they released. This helps keep the casual player from having to deal with events like LRMaggedon.
As for who set up this shindig, it was the Goons, more specifically Word of Lowtax, they sent out the invite, and the rest followed. As for why doesn't PGI do this? Well that is one of the spear points of our initiative:
To open a clear and direct line to the PGI development team so that we can assist in shaping this game to be the MOST successful Battletech/Mech Warrior product to date.
I want this game to be so damn good that someone who has never even IMAGINED a big stompy robot has interest in being a premium member. Let this game be the next pillar that keeps alive the IP for the next 30 years.
Edit: Just saw the signatures keep it up guys! Spread the word! It is so great to see so many familiar names that I haven't seen in a long time.
Edited by Miekael, 27 July 2013 - 09:20 AM.
#72
Posted 27 July 2013 - 09:54 AM
Gwaihir, on 26 July 2013 - 12:23 PM, said:
When a lot of us think about mech combat, we tend to think about limping around with half a machine left, only a couple of weapons, and bashing on each other until finally the last enemy falls. That doesn't really happen now, we generally go from "Everything's good!" to "Red CT/Side torso armor" to "Deep red internals/dead"
e: A larger ratio of internals to armor buff also indirectly helps LBX weapons, MGs, and other crit happy weapons with spread, like SRMs.
This so hard.
My favourite part of the WoL Birthday Tourny was when I was fighting a mirror match in trial atlases. It took forever and in the end I was lost my right torso and all my center armour. We both overheated at least once.
The fight was all torso twisting, heat management and precise targeting. My strategy of concentrating only on the CT was a close win over my opponents choice of shooting off my right torso(reducing my DPS by more than half).
Edited by Quincy80, 27 July 2013 - 09:55 AM.
#73
Posted 27 July 2013 - 09:58 AM
Sybreed, on 26 July 2013 - 06:10 PM, said:
hat is one of the thing that is extremely unpleasant for me about MWO: mechs don't feel unique, they feel soulless. My question in the last ATD was answered by a big "I don't know", but the dev seemed to understand my concern and agreed with it. Won't be until a while we see something to make mechs feel more distinct, but I think it's an important point to discuss.
Chassis variant utility was mentioned once or twice. I agree it's an issue worth discussing, but its also one the PGI seems aware of. I'm actually more concerned with the changes in their mech design process that have lead to some mechs (awesome, and to a degree the hunchback) being inferior even in their intended purpose compared to newer releases. Every mech shouldn't be good at everything, I think that's a given. (I mentioned missile tubes, for instance. Correcting/enforcing the launcher-to-tube system would help out missile-oriented chassis like the cat and treb almost immediately).
Looking forward to continuing the conversation, and finding a way we can meaningfully contribute to the future of the game, other than just playing it.
#74
Posted 27 July 2013 - 10:26 AM
They introduced the quirks system, which should have helped, but they haven't done anything to really make quirks worth the tradeoff in hardpoints.
PGI has a lot of variables to play with -- without requiring any radical coding or adjustments, just changing a couple numbers in some xml files.
They ought to be doing more tinkering, a little trial and error by playing with a lot of numbers at once. I realize that PGI is a big fan of "baby-steps" incremental approaches, but those haven't been working out so well. Maybe it's time to plug some bigger changes and see what sticks.
#75
Posted 27 July 2013 - 11:37 AM
Protection, on 27 July 2013 - 10:26 AM, said:
They introduced the quirks system, which should have helped, but they haven't done anything to really make quirks worth the tradeoff in hardpoints.
PGI has a lot of variables to play with -- without requiring any radical coding or adjustments, just changing a couple numbers in some xml files.
They ought to be doing more tinkering, a little trial and error by playing with a lot of numbers at once. I realize that PGI is a big fan of "baby-steps" incremental approaches, but those haven't been working out so well. Maybe it's time to plug some bigger changes and see what sticks.
We probably shouldn't discount the possibility that only having one useful variant per chassis might well be deliberate - I mean, look at the Raven. How many people GXPd the 2X and 4X so they could speed tweak their 3L? I know I did, because those other two are pretty horrible. If it's a deliberate choice to increase revenue, I could see why it would be like that.
#76
Posted 27 July 2013 - 11:45 AM
Much appreciated to all those who are keeping this thread clean and troll free. Please remember to REPORT any posts that are not productive.
Cheers.
#77
Posted 27 July 2013 - 12:09 PM
#78
Posted 27 July 2013 - 12:25 PM
Jackson Jax Teller, on 27 July 2013 - 12:20 PM, said:
Maybe you could bump this up to the higher ups in a "why arent WE doing this?" type train of thought?
Although I already have , there was no need. The individuals who are organizing this discussion have already done an excellent job communicating this to those who they are trying to reach.
I am sure they will continue to do so, but I will still assist whenever I can.
Cheers.
#79
Posted 27 July 2013 - 12:29 PM
10 user(s) are reading this topic
0 members, 10 guests, 0 anonymous users