Welcome, Guest. Please login or register.
March 28, 2024, 05:34:32 AM

Login with username, password and session length

Search:     Advanced search
we're back, baby
*
Home Help Search Login Register
f13.net  |  f13.net General Forums  |  The Gaming Graveyard  |  MechWarrior Online  |  Topic: UI 2.0 0 Members and 1 Guest are viewing this topic.
Pages: 1 2 [3] Go Down Print
Author Topic: UI 2.0  (Read 35913 times)
Falconeer
Terracotta Army
Posts: 11124

a polyamorous pansexual genderqueer born and living in the wrong country


WWW
Reply #70 on: September 11, 2013, 03:42:17 PM

UI 2.0 Engineering History Lesson, by Sean Cove (UI Dev).


Quote
Summary:

UI 2.0 was not only a redesign, but a reengineering of the whole UI system that drives it. Named UI 2.0, but it is the third iteration of the UI, along with UI 1.0, and UI 1.5. To help understand why with UI 2.0 we decided to do a full rebuild of the UI, I need to tell you a little about UI 1.0 and UI 1.5 first.


History:

UI 1.0 was an iteration that much of the general public didn't get to see. It was largely placeholder work to prototype the large grand design of the UI. Many features were stubbed in and non-functional, but it served to show the flow of the final product. This early UI contained stubs for things like Merc Corp management, an Inner Sphere map and a Leaderboard, all of which were non-functional, but you could navigate and see what could be there in the future.

UI 1.0 was also used as a sandbox, for we hadn’t worked with Crytek, or Scaleform UI technologies. We were learning what features Scaleform supports of Flash, and how it differs. Learning how Crytek’s code interfaces with Scaleform, and how to setup the flash screen assets, and communicate with them from C++ layers. It was a prototype build not a generic re-usable system. Only the smallest of flash assets were reusable, such as buttons, display lists, and other components that make up a screen. Much of the C++ code and Actionscript driving the screen layers were built to function for a specific design.

Up To Now:

Then we started work on the current UI in the game, UI 1.5. UI 1.5 has been the version that has gone through the most iteration, beginning with stripping out, and disabling stubbed screens, and reworking many of the existing functional screens. Using the base of UI 1.0, and what we learned about Crytek and Scaleform, to create a few central systems and centralize more of the code. During the course of UI 1.5 much of the Flash layer was rebuilt to fit the new look, leaving the C++ layer much untouched, for it was mostly used to manage and store data from the User and backend database.

Throughout the beta tests, many of the flash screens would go through much iteration. One example being the Mechlab, where in the beginning, all you could change were weapons, armour points, and the engine. There was no upgrade system (artemis, armour type, heat sink types), no paint scheme customization, and no engine heat sinks. The Mechlab must have changed at least once a month, though not all of it visible to the end user. This left the Mechlab code in a bit of a Frankenstein state, having gone through many iterations, with new features and multiple systems talking to each other to manage all the data. So when design started talking about a new UI, a UI 2.0, we jumped at the idea to reengineer the UI systems we were using.


Where We're Going:

Much of the reengineering started before the design of UI 2.0 was complete. Over the course of a few months, a handful of engineers rewrote many of the core systems, with the goal to data drive as much as possible and be able to run as much of the logic for the UI in C++ rather than on the Flash side with Actionscript.

We structured the new system to make the Flash side completely reusable, it became a data display system - with a little screen logic - making it possible to reuse Flash assets in full. Screens became a Flash layer, with a C++ driver layer that can be changed independent of the Flash asset. To accomplish this we centralized many of the systems that were in UI 1.5 into one or more sub systems that were common to all screens. This also reduced the amount of work required to bring a screen online. This has allowed UI 2.0 to separate from the effects of taking new engine drops from Crytek, changes in our backend and gameplay systems without having to change any screen logic, for the UI central system is the only piece of code that talks to and is aware of the systems underneath.

In UI 1.5, many of the screens were directly talking to these systems, which meant that if any of them changed in a major way, the UI would have to be adjusted to bring the UI online again. With all this in place, it leaves us with less maintenance work and allows us to focus more on usability of the UI, rather than keeping it functional.

UI 2.0 was a return to square one, using our better understanding of the systems involved; we rebuilt the foundation of the UI to better support the product as we continue to develop features. The new foundation is more flexible and less specialized, allowing the Designers and Artists the ability to work free of as many limitations as possible.

It would be great if any of you had any knowledge of what he's talking about to comment about it. Unfortunately it is rocket science to me so I have no idea if what he says makes any kind of sense or shows his lack of talent.
« Last Edit: September 11, 2013, 03:44:41 PM by Falconeer »

Goreschach
Terracotta Army
Posts: 1546


Reply #71 on: September 11, 2013, 05:05:43 PM

To put it siimply, the original UI was basically a placeholder and the reason UI2 has taken so long is because they had to completely rewrite all the code for it.

Answering whether or not this is a good thing or if it shows a lack of capability is deceptively complex. If you've ever payed any attention to the game development subforum here, you'll probably have seen people mentioning the value of early prototyping. Ignoring most software development methodology arguments(which is largely a liliputian war), there is a number of good advantages to early prototyping.

If you have no knowledge of programming, all you really need to understand is that doing software design the fast and easy way isn't the same thing as doing it the right way. Usually, if you want to get something up and running, you can typically get the system up and running rather quickly. However, this hacky system will have a number of problems, and it will start complaining or just fail to work whenever you change anything it talks to. This means you will need to go back and tweak and change your system whenever anyone changes some other part of the game. This was UI 1.

A good system lives in it's own little bubble, and doesn't really care what happens elsewhere. This takes longer to develop, but there is far less maintenance to deal with. The fast way, as mentioned above, takes less time to implement but will eventually require more total time due to all the work and code rewrite needed to keep it running as the entire program evolves. A well designed system needs very little work after it has been finished. This will be UI 2.

When game developers talk about 'early prototyping' they are taking about this problem. Early prototyping is the idea that you lower total production cost or time by making a 'preview' version of a game. This preview version isn't like a game demo, rather, it's essentially the entire game design(minus most content) made out of quick, crappy systems that are only intended to get up and running immediately. The main advantage of this is that you can get something resembling your game up ASAP, and you don't have to wait for two years to spot any major top level design failures that keep the game from working as it should or simply keep it from being fun.

The downside is that you have have to go into early prototyping 100% committed to tearing out all of this crappy early work and rebuilding everything, correctly, from the ground up. This is often a hard sell, because once you have something up and running, people will balk when you tell them to delete it all and start all over. They will wan to 'fix it'. After all, it's just code, and you can simply delete lines and write new lines. Those old lines are gone, they won't just sneak back in. The problem with this idea, however, is that 'fixing it' is exactly the problem that prevents you from using the fast, hacky code. Once you have a program sitting on a crappy, unstable base, you will spend forever fixing it. Only terrible designers will make a crappy, unstable base, and then try to tweak and fix it into shipping code. This is what PGI did. This is UI 1.5.

PGI made quick, crappy code, and then instead of tearing it out and starting over, they dragged it along trying to polish a turd. Did they start out on the assumption that they were prototyping? Maybe. Maybe they got lazy and decided the prototype was 'good enough to fix'. Maybe management didn't want to spare the time to rewrite something already there. Maybe PGI just isn't very good and their crappy UI 1 really was supposed to be production code. It doesn't matter. They are now, apparently, doing the right thing in rewriting a system that needs to be rewritten.

The problem is that they have blown a ton of time trying to maintain a crappy system that they should have torn down and rewritten a year ago. Look at all the UI bugs they've had. Random, unpredictable things, caused by ugly lists of pointers floating about, referencing systems talking to systems, talking to systems. A system so large nobody unstands how it works. A design so complicated it took a guy that programmed it an entire page just to explain what the WWWWWWWW problem was. Judging by the way bugs seem to keep lingering in this game for months at a time, only to be briefly fixed and then pop up again a few patches later, I'd say the entire project is like the UI. Crappy. The entire game is most likely a prototype, and it needs to be torn down and rebuilt.

Too bad it's launching next week.
Hoax
Terracotta Army
Posts: 8110

l33t kiddie


Reply #72 on: September 11, 2013, 09:13:03 PM

Good post/10, thanks for taking the time with that.

A nation consists of its laws. A nation does not consist of its situation at a given time. If an individual's morals are situational, then that individual is without morals. If a nation's laws are situational, that nation has no laws, and soon isn't a nation.
-William Gibson
Falconeer
Terracotta Army
Posts: 11124

a polyamorous pansexual genderqueer born and living in the wrong country


WWW
Reply #73 on: September 12, 2013, 12:19:12 AM

Yes, thanks a lot Goreschach, your post is what I was hoping for and much much more.

Falconeer
Terracotta Army
Posts: 11124

a polyamorous pansexual genderqueer born and living in the wrong country


WWW
Reply #74 on: October 21, 2013, 04:26:52 PM

Recorded demonstration of the UI 2.0. It looks nice but... so many placeholders.  ACK!

EDIT: At 22:17 you can actually see a teaser of the moon map, coming in December.
« Last Edit: October 21, 2013, 04:32:18 PM by Falconeer »

Shannow
Terracotta Army
Posts: 3703


Reply #75 on: October 22, 2013, 04:18:38 AM

moon map looks kinda awesome...do we get low grav rules?..:D

Someone liked something? Who the fuzzy fuck was this heretic? You don't come to this website and enjoy something. Fuck that. ~ The Walrus
eldaec
Terracotta Army
Posts: 11838


Reply #76 on: October 22, 2013, 05:02:35 AM

Also, ambient temp -273?

"People will not assume that what they read on the internet is trustworthy or that it carries any particular ­assurance or accuracy" - Lord Leveson
"Hyperbole is a cancer" - Lakov Sanite
Sir T
Terracotta Army
Posts: 14223


Reply #77 on: October 22, 2013, 05:03:59 AM

We don't need no stinking heatsinks!!

Hic sunt dracones.
Falconeer
Terracotta Army
Posts: 11124

a polyamorous pansexual genderqueer born and living in the wrong country


WWW
Reply #78 on: October 22, 2013, 05:36:13 AM

No gravity changes. They said so multiple times.

I also think it looks fantastic. So Aliens, among other things.

Ironwood
Terracotta Army
Posts: 28240


Reply #79 on: October 22, 2013, 06:08:37 AM

Also, ambient temp -273?

This map was already in one of the Mechwarrior games, as I recall.  Had strange gravity and the heat was awesome.  (also, do you think gravity being weird would affect ballistics ?  Always wondered about that...)

As I got Terra Therma last night, all I could think about was 'MORE MAPS PLEASE'.  Launching the galaxy nonsense with this shit would be bad.

"Mr Soft Owl has Seen Some Shit." - Sun Tzu
Shannow
Terracotta Army
Posts: 3703


Reply #80 on: October 22, 2013, 06:09:20 AM

Howabout vacumn rules? What you lost all the armor on your CT?...oh sorry bye bye CT!

Heh.

edit: I got terra therma 3 maps in a row last night.  FUCK THAT.

Someone liked something? Who the fuzzy fuck was this heretic? You don't come to this website and enjoy something. Fuck that. ~ The Walrus
Falconeer
Terracotta Army
Posts: 11124

a polyamorous pansexual genderqueer born and living in the wrong country


WWW
Reply #81 on: October 22, 2013, 06:22:29 AM

Anyway, moon base (or actually "HyperPulse Generator Manifold", which is the actual name) will hit with the first patch of December. And then "jungle map" in February.

eldaec
Terracotta Army
Posts: 11838


Reply #82 on: October 23, 2013, 02:10:44 PM

Weird gravity should impact ballistics.

If you fire a bullet out of a barrel straight and level it will initially rise as for reasons no-one fully understands it generates lift. As the bullet slows, lift reduces and trajectory drops. Low gravity should impact that curve. But the curve doesn't exist in mechwarrior.

Lift also requires a atmosphere ofc. No atmosphere should eliminate lift but also drag.
« Last Edit: October 23, 2013, 03:05:03 PM by eldaec »

"People will not assume that what they read on the internet is trustworthy or that it carries any particular ­assurance or accuracy" - Lord Leveson
"Hyperbole is a cancer" - Lakov Sanite
Typhon
Terracotta Army
Posts: 2493


Reply #83 on: October 24, 2013, 05:55:47 AM

Do slugs experience lift?  Or just bullets with angular momentum?


Edit: it's early, this is a dumb question - I was thinking some strange coriolis interaction changing the angle of the bullet but "just as the bullet leaves the barrel" and not "kilometers after the bullet leaves the barrel" make this line of thought dumb.  Sorry. 

I don't think bullets experience any lift after leaving the barrel, probably a folklore/perception thing to do with the arc that the bullet flies along and correction for that arc with the aiming of the barrel.
« Last Edit: October 24, 2013, 06:12:43 AM by Typhon »
eldaec
Terracotta Army
Posts: 11838


Reply #84 on: October 24, 2013, 06:09:29 AM

I guess they must at least generate Coanda and flat plate lift as soon as gravity pulls them out of a zero angle of attack.

Most bulllets are fired at a low enough altitude to also benefit from ground effect.

So "I guess so" but whether it's the same lift I have no idea.

"People will not assume that what they read on the internet is trustworthy or that it carries any particular ­assurance or accuracy" - Lord Leveson
"Hyperbole is a cancer" - Lakov Sanite
Falconeer
Terracotta Army
Posts: 11124

a polyamorous pansexual genderqueer born and living in the wrong country


WWW
Reply #85 on: October 29, 2013, 12:52:08 AM

UI 2.0 is on the Public Test Server today.
To say that looks like it's a fucking pre-alpha version that is MONTHS away from being ready for live deployement would be an understatement. According to what Ekman said in September, UI 2.0 was going to be launched (not tested) by the beginning of November. These people are YEARS behind their schedules, it is between pathetic and ridiculous. It's proving harder and longer for them to make a new interface than to make a new game.

Quote
Hello MechWarriors!

You've heard about our progress on UI2.0 and we're so excited that we wanted to bring you an Alpha PREVIEW of our current build.

As you may know, we've been reviewing a slew of feedback from our Command Chair Updates which has been extremely well-received by our team. Thank you for taking the time to express your vision for UI 2.0, we're ready to move forward with construction.

This UI 2.0 build is a work in progress. We're releasing this early access build to ensure that our most dedicated players have the chance to try it out and provide feedback while elements are still in production. Internally, our testers have identified a variety of issues that we're already working on. We value your feedback on functionality and screen presentation. Please feel free to leave your feedback in the Public Test Forums.

There's still a lot to do! Please take a moment to join us in this exciting preliminary test.

Patch Notes

Notes:
In this build you can:
Access the new UI
Mech Lab
Customize a mech’s weapon + equipment + armor
Weapon and Mech Tool Tips
Save loadout
Work in Progress:
NO paint scheme customization
NO module add on
NO COMPARE functionality
NO Social functionality


Settings

Skill Tree
Only Mech Tree
Work In Progress:
NO Pilot Tree


Upon log-in, we provide users with following:

All 4 Founders `Mechs
Sarah's `Mech
All 4 Phoenix `Mechs (Overlord Package)
50,000 MC
500,000,000 C-Bills
1,000,000 GXP


You CANNOT launch into a match
But you can goto Testing Ground


Known Issues:

Sometimes the FE will get stuck at saving state while a user clicks on checkout during Mech customization
Client will crash when closed
Client may crash if Options, Loadout and Skill menus are visited in the same session
Client will crash if internet connection is lost whilst in tutorials
All menus do not adjust to all resolution changes.
Tabbing back and forth from the Mech Lab and Skills tabs causes the Mech skill list to not load properly
When launching Testing Grounds, users will not launch into their chosen map
The price of a Mech Bay is not currently displayed
Tab is currently non-functional at the login screen
Typing a comma into the login page generates an 'unknown error' message.
A placeholder message will be seen when attempting login whilst servers are offline
Logging out may lead to a black, unresponsive screen
Buttons that direct to website links do not work


Features Not Yet Completed:

Legend for hardpoints and item restrictions icons
Cannot see loadout and hardpoints available of `Mechs before a `Mech purchase is made.
Checks for engine and heatsinks are not made to ensure a viable `Mech is saved.
Options (now SETTINGS): joystick support is missing
Options: Vsync is missing
`Mech upgrades are not implemented


Quote
Oct 29th 2013
Focus: Early Preview of UI 2.0
Morning Test: 10am-12pm PST
Afternoon Test: 4:00pm - 6:00pm PST


You can download the client here : http://mwomercs.com/forums/topic/126210-welcome-to-the-public-test-program/

Falconeer
Terracotta Army
Posts: 11124

a polyamorous pansexual genderqueer born and living in the wrong country


WWW
Reply #86 on: October 29, 2013, 04:35:14 PM

Ok, I am fiddling around with this, and while, as anticipated, nothing works, I gotta say that it looks and feels prety darn good. Nothing like finally seeing your own 'mechs in all their glory. Plus lots of other little touches. While depressed for the endless delays, I gotta say this is really nice.

Sir T
Terracotta Army
Posts: 14223


Reply #87 on: October 30, 2013, 05:56:18 PM

Do you think this will be ready by the end of the year?

Hic sunt dracones.
Falconeer
Terracotta Army
Posts: 11124

a polyamorous pansexual genderqueer born and living in the wrong country


WWW
Reply #88 on: October 30, 2013, 06:16:24 PM

I would be surprised. I think it won't be ready, no. Too many missing features. If they weren't confident enough to show them, knowing how hungry the community is for it, I have to assume they are really behind schedule.

But again, the little I saw is much prettier than it looked in all the videos.

Sir T
Terracotta Army
Posts: 14223


Reply #89 on: October 30, 2013, 06:21:33 PM

Consider me  cry

Hic sunt dracones.
Falconeer
Terracotta Army
Posts: 11124

a polyamorous pansexual genderqueer born and living in the wrong country


WWW
Reply #90 on: October 30, 2013, 06:26:00 PM

Me too, man. Me too. I am so fucking bored of waiting.

Pages: 1 2 [3] Go Up Print 
f13.net  |  f13.net General Forums  |  The Gaming Graveyard  |  MechWarrior Online  |  Topic: UI 2.0  
Jump to:  

Powered by SMF 1.1.10 | SMF © 2006-2009, Simple Machines LLC