42 replies to this topic

[ShadowSpirit]

IraqiWalker, on 21 June 2015 - 06:18 AM, said:

This is dumb, and I don't say this lightly.

Heh. Please do continue to take yourself too seriously.

I'll restate it more succinctly so that you can fill in the blanks to your liking. In this way maybe you can argue less without missing the point.

"They should shift resources around and do something about the terrible new player experience in this game. "

[DustySkunk]

I totally agree that the new player experience is in serious need of work.

Mechs help pay the bills and give us something to keep working towards as players. The skills involved in mech creation are not closely related to those required to create a good NPE. Less mechs are bad for PGI and therefore not good for us.

Here's an alternative:

How about PGI keeps doing what they're doing in addition to hiring some more staff to focus solely on designing the NPE? Ideally these folks would be some industry vets who are looking for work and have designed other games' tutorials. Just a thought

[xWiredx]

Something to consider, gents, is that lately patches are RELATIVELY content heavy, but the word RELATIVELY is key here - it's only relative to how little we used to get. PGIs patches are still light compared to other games, despite having some of the higher real-cash prices. We are giving them money every month or every quarter (basically, heros, champions, mech packs) to continue developing at a somewhat slow pace AND they have a somewhat high rate of introducing flaws+unpopular additions.

There are only 2 ways to see this currently: PGI is hard-coding things in such a way that makes tiny changes hard to accomplish, or PGI is lazy and not making the best use of its resources.

[FuzzyLog1c]

MWO's problems stem from poor management and direction from Paul and Russ. Nothing's going to get better as long as they lead. Pumping more money into PGI won't help--it'll continue to get misused instead of being invested in network infrastructure and content.

Think about it: In the few instances where we've been able to directly communicate to the PGI rank-and-file, the general impression has been that they're competent, decent individuals. Compare that to Russ and Paul refusing to interact with the community, with Russ in particular troll-baiting people on Twitter (e.g. "I don't even play the TBR" after announcing the TBR nerfs), and banning everyone that disagrees with him. It's abusive. A cautionary tale of what's possible if you manage to acquire an exclusive license to a much-beloved franchise.

If this was any game except Battletech, everyone at PGI would be out on the streets begging for food. It's a real credit to the fundamental mechanics of this game that it's still fun, despite the plethora of technical and programmatic problems MWO as a project has.

Edited by FuzzyLog1c, 21 June 2015 - 03:18 PM.

[FuzzyLog1c]

As a testament to how borked PGI's implementation of CryEngine 3 is, I'd like to point out that people have placed mechs into large, singleplayer maps from Crysis 3. The result: perfect Crossfire and SLI support, with the engine running 3-4 times faster, and looking vastly better than what we have today in 2015. Let me emphasize: this is not hard. It's not expensive. Several people with vastly different skill sets did this, at different times and in different places, independently, without helping each other, in less than an afternoon.









Mixing Star Citizen and MWO assets:

Edited by FuzzyLog1c, 21 June 2015 - 02:58 PM.

[IraqiWalker]

ShadowSpirit, on 21 June 2015 - 12:39 PM, said:

Heh. Please do continue to take yourself too seriously.

I'll restate it more succinctly so that you can fill in the blanks to your liking. In this way maybe you can argue less without missing the point.

"They should shift resources around and do something about the terrible new player experience in this game. "

They don't have them, and relocating people from different teams to work on things they don't know about won't work.

I like the example gallowglass used. you're telling a heart surgeon to go treat cancer, because there's a backlog there. The two skillsets don't even intersect at any point.

[xWiredx]

FuzzyLog1c, on 21 June 2015 - 02:57 PM, said:

As a testament to how borked PGI's implementation of CryEngine 3 is, I'd like to point out that people have placed mechs into large, singleplayer maps from Crysis 3. The result: perfect Crossfire and SLI support, with the engine running 3-4 times faster, and looking vastly better than what we have today in 2015. Let me emphasize: this is not hard. It's not expensive. Several people with vastly different skill sets did this, at different times and in different places, independently, without helping each other, in less than an afternoon.

Those images are gorgeous. Yet here we are...

[Chimperator]

Ironwithin, on 16 June 2015 - 03:45 PM, said:

So you're saying you want the 3D-artists working on the game-code and the concept-artists fiddling around with AI for bots/NPCs ?
What could possibly go wrong there...

This alrdy happend dude

[Kiiyor]

Ironwithin, on 16 June 2015 - 03:45 PM, said:

So you're saying you want the 3D-artists working on the game-code and the concept-artists fiddling around with AI for bots/NPCs ?
What could possibly go wrong there...

HA, you took the words right out of my mouth!

I wish there were bigger fonts for emphasis, but here we go:

The teams that work on mechs are separate to those that work on everything else.

[Kiiyor]

FuzzyLog1c, on 21 June 2015 - 02:57 PM, said:

As a testament to how borked PGI's implementation of CryEngine 3 is, I'd like to point out that people have placed mechs into large, singleplayer maps from Crysis 3. The result: perfect Crossfire and SLI support, with the engine running 3-4 times faster, and looking vastly better than what we have today in 2015. Let me emphasize: this is not hard. It's not expensive. Several people with vastly different skill sets did this, at different times and in different places, independently, without helping each other, in less than an afternoon.

Except it is hard, and it is expensive - on the game servers especially.

It's not about simply cramming assets into a different game, it's about getting the netcode working with everything else, and trying to manage the ludicrous number of server calls that MWO has. The more complex you make the maps and terrain, the more data has to get fed through the game servers. Mechs have a horrific number of server calls as it is - lots of weapons, lots of different moving parts, lots of different damage registration points, lots of physics, lots of different keyboard inputs...

The above example is freaking gorgeous, but it's only a couple of models imported into another game. It's not a stunning technical achievement. Those models aren't rigged with animation, or physics, or damage models, or weapons, or HUD's, or particle emitters, or anything else. I defy you to find anyone with a lick of actual game engine expertise to say they could import 24 mechs into Crysis and have them fight the way ours do "3-4 times faster" than what we have now. I've asked around my embittered circle of friends, two of which have worked on the Cryengine before (one extensively, who worked on CryEngine games in the UK), and they laughed at me.

Apparently, all those elevation changes would have an enormous impact on performance and hit registration: Mech movement would be insanely complicated, as each change in elevation and collision with map terrain would have to be fed to the server and back to the client after calculating velocity and momentum changes and whatnot. Weapons fire would also be an order of magnitude more complex with environmental hitboxes that detailed. Remember when inverse kinematics were killed off in MWO to increase performance? Imagine trying to run that on a map as detailed as the ones above.

I'm informed that it would be indeed possible to have maps as detailed as those above - but not with anything even remotely approaching stability for a multiplayer title, and not without a huge development studio driving improvement.

I'm informed that It would apparently be possible to improve game performance and visuals with a newer game engine, but let's not forget that work started on this game quite a few years ago now, and switching the game engine of an existing game (while easier than starting from scratch) apparently requires almost as many man hours as starting from scratch, depending on the level of experience your development team has with the new engine, and the differences between the two engines.

I doubt PGI has the resources for that.

[Twilight Fenrir]

Kiiyor, on 21 June 2015 - 08:01 PM, said:

Except it is hard, and it is expensive - on the game servers especially.

It's not about simply cramming assets into a different game, it's about getting the netcode working with everything else, and trying to manage the ludicrous number of server calls that MWO has. The more complex you make the maps and terrain, the more data has to get fed through the game servers. Mechs have a horrific number of server calls as it is - lots of weapons, lots of different moving parts, lots of different damage registration points, lots of physics, lots of different keyboard inputs...

The above example is freaking gorgeous, but it's only a couple of models imported into another game. It's not a stunning technical achievement. Those models aren't rigged with animation, or physics, or damage models, or weapons, or HUD's, or particle emitters, or anything else. I defy you to find anyone with a lick of actual game engine expertise to say they could import 24 mechs into Crysis and have them fight the way ours do "3-4 times faster" than what we have now. I've asked around my embittered circle of friends, two of which have worked on the Cryengine before (one extensively, who worked on CryEngine games in the UK), and they laughed at me.

Apparently, all those elevation changes would have an enormous impact on performance and hit registration: Mech movement would be insanely complicated, as each change in elevation and collision with map terrain would have to be fed to the server and back to the client after calculating velocity and momentum changes and whatnot. Weapons fire would also be an order of magnitude more complex with environmental hitboxes that detailed. Remember when inverse kinematics were killed off in MWO to increase performance? Imagine trying to run that on a map as detailed as the ones above.

I'm informed that it would be indeed possible to have maps as detailed as those above - but not with anything even remotely approaching stability for a multiplayer title, and not without a huge development studio driving improvement.

I'm informed that It would apparently be possible to improve game performance and visuals with a newer game engine, but let's not forget that work started on this game quite a few years ago now, and switching the game engine of an existing game (while easier than starting from scratch) apparently requires almost as many man hours as starting from scratch, depending on the level of experience your development team has with the new engine, and the differences between the two engines.

I doubt PGI has the resources for that.

That was a very interesting, and informative post. Thank you

The game USED to look better, but it was only 8v8 at the time, I suspect our gorgeous game was sacrificed on the altar of 12v12... but, the game still looks pretty good, and runs perfectly for me, now. Pretty rock-solid, too.... Unless I click 'exit game' at which point it crashes 90% of the time... which is hilarious

Edited by Twilight Fenrir, 22 June 2015 - 03:51 PM.

[Gas Guzzler]

Well new mechs bring in money to support the game. Not saying we couldn't use some of the other stuff, but without mechs being released, they run out of money and we have no game. That being said, I love mech variety, and new clan packs really help a lot for that.

[Scam Newton]

CW is a total cluster. Pugging is short term gratification for explosions and watching people do excessively stupid things.

[Kiiyor]

Twilight Fenrir, on 22 June 2015 - 03:42 PM, said:

That was a very interesting, and informative post. Thank you

The game USED to look better, but it was only 8v8 at the time, I suspect our gorgeous game was sacrificed on the altar of 12v12... but, the game still looks pretty good, and runs perfectly for me, now. Pretty rock-solid, too.... Unless I click 'exit game' at which point it crashes 90% of the time... which is hilarious

Yep, it was. One of the MWO net techs posted about it a while ago - each added mech puts an exorbitant amount of strain on game performance, and adds more server calls. Stuff like the gorgeous damage textures we used to have, and inverse kinematics (feet standing at different levels on uneven terrain, and actually angling on sloped surfaces etc) were given to the battletech gods as sacrifice for being able to field a full company.

Let's hope that an optimization pass is made at some time or another. Maybe they're already doing it with the new River City map. I dread the performance if they aren't!

[D34DMetal]

Ironwithin, on 16 June 2015 - 03:45 PM, said:

So you're saying you want the 3D-artists working on the game-code and the concept-artists fiddling around with AI for bots/NPCs ?
What could possibly go wrong there...

OR you could put them to work on maps...

[IraqiWalker]

DeadMetal89, on 22 June 2015 - 07:34 PM, said:

OR you could put them to work on maps...

They are. They also work on mech models, and map assets.

[Dirty Old Man]

First off, great job PGI with this version of the PVP universe on a very old IP that "no one" else was interested at all for almost 10 years past. For that alone a lot of us will continue to support this fast pace PVP buzz trip.

After almost 5 years, I still get the shivers and my heart still pumps so hard like its gonna pop when a thrilling match ends... this is an incredible product as it is and we acknowledge its a labor of love and the business model is lead with micro transactions that has always been and will continue to be NOT P2W.

However, having congratulate on achieving your beta goals, there are more and more diverse levels of Gamers joining our ranks and getting thrust into the great universe.. and with them comes more demands for diverse ways to give them the entertainment they want.

Therefore, perhaps it would be a good time to share with the new collective of Gamers and Fans of your brand of MW experience on what would be the direction for this product for the next 5 years.

For example would it be tuned towards being enlisted as the CyberSports arena franchises? or its still going to be as it is with more releases of simplified content? or will it be inserted into a SANDBOX Universe were a more holistic Ultimate experience would be delivered where everyone can also be the engineer, the privateer? the explorer or the mechwarrior as well....

Thank you again for your hard work.

[Troutmonkey]

Tarogato, on 16 June 2015 - 04:24 PM, said:

People who work on mechs aren't necessarily equipped to work on things like "new player experience."

Every. Single. Patch. There's. Always. Somebody... that tries to blame the lack of development on the new mech production.

Yes, there's slow development, but it's not the mech-packs' fault.

Reduce their hours, and use the savings to hire actual programmers and UI designers.

[FatYak]

ShadowSpirit, on 16 June 2015 - 01:02 PM, said:

Seriously. We don't need new mechs. We need improved new player experience, better small group mechanics (why is everything 12 vs 12 cluster muck?), and some sort of campaign mode to get players introduced to the game.

This game has one of the _WORST_ new player experiences of most of its contemporaries: WoT, WoPlanes, War Thunder, etc.

Steam release is going to do jack ____ for this game if you don 't deal with the incredible learning curve. Get people in better mechs more quickly. Let mastering a chassis be part of the new player learning experience. Something.

/end mini-rant

Its not as bad a Hawken, but hawken is dying so probably not a good benchmark

[Ironwithin]

FatYak, on 03 July 2015 - 10:58 PM, said:

Its not as bad a Hawken, but hawken is dying so probably not a good benchmark

Hawken deserves to die ... a horrible abomination, that one.