Return to “[Archived] Daily Dev Logs, 2012 - 2015”

Post

Week of April 21, 2013

#1
Sunday, April 21, 2013

Summary

Today I've been working almost all day on playtesting!! There are so many details that crop up during a real playtest that I never really get to see when coding, but I'm finally getting around to taking care of them and trying to make the ride as smooth as possible for players. A couple of hours ago I began what I hope will be my "final" playtest...actually starting from scratch and playing like a normal player would. I've had to stop a bunch of times to make small improvements, but have yet to run into anything game-breaking, and my savegame integrity is holding strong! I've worked my way up to a few good weapons, a respectable shield, and another friendly fighter. He's actually pretty decent, and together, we're capable of handling bombers without much problem...although I don't know how we'll fare against corvettes. One step at a time :geek: Should be able to add a third member to the party soon! I'd like to try out leading a gang of fighters before I move up to bigger things :D

Ultimately, what I'd like to do is work my way all the way up to a fleet of at least a few good sub-capitals, at which point I'll be able to test drive fleet combat "for real," using my own, hard-earned assets. If I can get to this point without any huge problems and am able to have fun doing so...then the green light will be shining for LTP!

Unfortunately I'm a little caught up right now on a non-game-breaking bug, but hope to get back to some bigger & more intense missions after I finish this.

By the way...I am having fun :) Yes, it's true, LTP is capable of inducing fun! At least for me. I'm pleased with that...it was my goal :D

Hour Tally

Coding: 9.52
Internet: 2.72
Testing: 3.40
Thinking: 0.50
Total Logged Time: 16.14
“Whether you think you can, or you think you can't--you're right.” ~ Henry Ford
Post

Re: Week of April 21, 2013

#2
Monday, April 22, 2013

Summary

YES!!! It was such a good day :D

Playtesting has been even more fun that I ever imagined. I'm already having loads of fun with this thing, and I honestly wasn't expecting that to happen yet at this stage. It's a very weird feeling for me...because when I'm coding, I usually just code a feature, verify that it works, and move on. I never actually play long enough to start having fun, so I guess I just assume that the game isn't fun yet. But that's not true :) I just needed to spend some time playing it!

But I didn't just play today...it's absurd how many features I implemented. Strange, because I usually think I'm pretty efficient and productive when it comes to implementing things, but today just felt like another level. It was because my gameplay was driving the implementation - I would pause and think, huh, you know, I'd really like to be able to do this right now - and then I'd go implement that. I guess it's a more effective way to work, because you're excited to implement something so that you can have even more fun. Truly a great feedback loop :)

Most of what I implemented today was relevant to fleet mechanics, since I finally started getting wealthy enough to maintain a fleet. So...what did I do? Let's see...where to begin. I started off by rewriting the combat AI (again), to make the attack patterns more "fun." The previous circle-the-attacker-at-maximal-range was good...too good. It's extremely effective, but it's not fun. You end up chasing down fighters forever because you can't hit them and they're smart enough to stay at long range and keep pulling a heavy circular motion, which makes targeting near impossible. I had to add some variety (I guess, in a way, you could call them "mistakes") so as to make combat fun an interesting. I then made following AI a lot smarter. Like, a LOT smarter. Collision avoidance is really good now, and the AI is also smart enough to use cruise mode when they need to catch up or need to go to somewhere far off, and also smart enough to dock / undock alongside you. So if you dock at a station and the AI is following you, they will dock too, and likewise for undocking. Nice touch :D

I also implemented a "patrol" command when I realized that I didn't want to have to issue an attack order for every mission that I took up. Patrol just tells a ship to roam around the region and attack anything that's hostile to you, as you would expect. The result is that you can keep a bunch of ships patrolling outside the station, and as soon as you accept a mission and the hostiles show up on radar, the ships will break and go attack. You can then go out and join them in battle if you like, or just hang back and watch them dominate from afar (hey, I've used that phrase before..it's exactly what I said in the KS video! See! Promises are coming true already!!) Once they've wiped the hostiles clean, they'll go right on back to patrolling. I love not having to issue orders, and just watching my ships go do their job :) It's also incredibly exciting to fly into battle with them, especially if you've got some big ships!

Speaking of big ships, I worked my way up to my first frigate today :D It is really a rewarding experience getting to buy a big ship, since you've got to work hard for a while to be able to afford it. But seeing it in action is just...awesome. I decked mine out with the best guns I could find, four of the best shields I could fine (!!!! yes, FOUR shields on my frigate :D ), and some really thick armor. The thing is unstoppable! I very much enjoy watching it lay waste to other, less-lavishly-equipped enemy frigates.

Oh, and then I decided that I wanted to be able to name my ships, so I went ahead and implemented asset naming. Just a nice little touch I think. I named my frigate "Vivec" after a certain deity in a certain favorite game of mine ;)

Even though fleet combat is still very simple, it's surprisingly engaging. I purchased a bunch of smaller ships to escort good old Vivec, mainly for aesthetics (they really can't help him much..but it's fun to see them buzzing around). Building a simple formation is as simple as telling them to follow the lead ship, and then setting the lead ship to patrol or attack or whatever. The result is simple but effective...you get the classic "escort" look, but as soon as the escorts get in range of an enemy, they will help out. They don't break formation yet, but that should be easy enough to figure out later. Watching Vivec and his little helpers make the system safe for travelers has brought me several hours of fun so far :)

Well, I'm happy to say, this thing is about done! I think the last thing I want to do is implement mission cancellation, just in case you get in over your head and don't want to reload. Should be a simple thing. Then I'll play a little bit more, probably work up to a few more frigates (Almalexia, Sotha Sil? :) ) just for fun..and then...well, you know the rest!!!

PS ~ Can you tell I'm excited today :)

PPS ~ LTP is also turning out to be nice and stable, for which I am really grateful. Every crash that I encountered today could be explained easily, and I was able to fix them all. So there are no outstanding instabilities that I know of, although I'll certainly keep hunting.

Hour Tally

Business: 0.08
Coding: 9.77
Composing: 0.07
Internet: 0.96
Testing: 4.85
Thinking: 0.19
Total Logged Time: 15.92
“Whether you think you can, or you think you can't--you're right.” ~ Henry Ford
Post

Re: Week of April 21, 2013

#3
Tuesday, April 23, 2013

Summary

Huge day of work...I can't believe how much I still had left to do...but..dare I say it? It's done :) LTP is sitting in the Desura queue right now, waiting for an admin to "approve" it, whatever that means. I have no idea how long this takes, nor did they give any indication...but I'm hoping they're speedy about it. After they let me through, all that's left to do is get a verification from a small group of pre-pre-alpha testers that the thing actually works. And if it does...! :D

So I did manage to rack up my second frigate this morning ("Almalexia," go figure :) ), but stopped there. Very good fun :) At that point, though, I was worth 6 million credits, and I'm afraid there wasn't a whole lot left to challenge me - Vivec and Almalexia, along with their underlings, could pretty handily plow through enemy fleets of 15-20, making me the ultimate force to be reckoned with in the system. LTP is definitely too "easy" compared to how the real LT will be (it only takes a day or so to get quite powerful), but I'd rather that than have it be impossible. After all, this is supposed to be a fun little taste of LT. Overall, I'm surprisingly pleased with how the balance turned out. People will no doubt find lots of ways to rack up fast cash, but, at least to me, most of the numbers feel "reasonable," which is more than I hoped for when I took a stab at all the constants that define the balance of the game.

In-between all the last-minute fixes, I actually went a little crazy on the graphics side today, for what reason I am not sure. I rewrote pulse, beam, and even thruster graphics! They all look significantly better, IMO. The pulses look like they are really carrying energy, and same for the beams. As for the thrusters, they've been cleaned up but also given an anamorphic flare effect. I think it's a lovely touch, particularly when watching fleets in the command interface :)

My next task is to make a tutorial video, since the game has no help to speak of. I'll do that while I'm waiting on Desura. I'm slightly concerned about the monthly update, as I feel like I don't have much to show in terms of new gameplay...I mean sure, I added millions of little things for the sake of the prototype, but nothing really ground-breakingly new. I might have to try to keep pulling these hours until the end of the month to see if I can squeeze something cool out of this last week ;)

Really can't wait to get this thing out. I'm nervous as can be...but hoping for the best. Most of all, I can't wait to start reading about peoples' experiences and watching videos of them playing, because I'm sure you guys will think of more fun and creative things to do in the game than I could have!

PS ~ The hours lie. I think it's not tagging all the applications that I use...because I have not been to bed since the last dev log, and only took an hour or two out of the day to eat...you can do that math :shock: Goal is to not go to bed until the prototype is out. Hurry up Desura, I cannot wait for my reunion with good, long, quality sleep!!!

Hour Tally

Business: 1.79
Coding: 7.17
Internet: 2.52
Testing: 5.04
Total Logged Time: 16.53
“Whether you think you can, or you think you can't--you're right.” ~ Henry Ford
Post

Re: Week of April 21, 2013

#4
Wednesday, April 24, 2013

Summary

I'm sorry for this dev log :(

It was a very rough day, as the first round of pre-pre testers got their hands on LTP today. Things weren't going so poorly for a while. We were met with a very serious performance problem caused by the audio engine, which was making large battles unplayable. Like all good problems, this one was only rearing its head very infrequently for me, so it was hard to tell if something was wrong or not. But it was, and I get to blame someone else's code this time. For a while I was liking irrKlang, but now I'm just frustrated and mad with it. There's no doubt the performance of it is what caused the problems. The API calls into irrKlang are far too slow. I had to spend several hours working around this by writing my own, threaded priority-queue system to make sure not to give irrKlang too much work (by only feeding it the loudest sounds), because it obviously does not take any measures to do so on it's own...it just slows to a crawl and kills the game. My system works well, and successfully reduces the load on irrKlang to the point that large engagements are not suffering at all anymore! Problem #1 down :)

There was a slew of minor things to fix and improve, but nothing horrible, until we got word from an ATI (AMD) card user, and an intel integrated chip user. The proto fails very badly on both. Failing on an integrated chip isn't so bad...but failing on a high-end AMD card is not acceptable at all. After an hour or two of debugging and working with this very patient tester, we were able to get everything working - except the mesh generation technology. So...everything working except the most critical piece of the game :cry:

This is a very bad situation, because I don't have a machine that can reproduce this problem, and the best I can do is guess at it from looking at log files. Luckily, my laptop has a built in Intel HD4000 chip in addition to the (real) nvidia card, so I am going to switch to integrated graphics and see if I can get the game to run. I'm thinking that if I can fix everything to the point that it runs on an integrated chip, then I should be in the clear. This may or may not be a valid strategy, but it's already led me to a few fixed problems :| In the morning I will go try to find an AMD machine to buy in town and see if I can get it set up and running tomorrow.

In short, although the proto is now happily running on nvidia machines (and we've verified that performance is high), there is going to be an immense amount of pain between now and its release in order to get it working on other chipsets. The release of the proto is looking much more uncertain than usual, and I honestly can't say when it will come - I was hoping for today, then again for tomorrow, and now I just don't know...I'm going to be debugging 23/7 until I solve it, but with an issue like this, it's just impossible to say when that will be :(

I knew there would be pain when I took a different route than most games. It was a choice...the procedural generator in LT is doing things that most modern games would never dream of doing with respect to pushing the GPU. But that's what's going to make this game special, and I can't let it stop me :) After all, that's Limit Theory - everything just requires the right insight. Pray that this insight comes soon :|

The good news, my friends, is that for every massive low, there is a massive high just waiting to happen. I'm going to be the happiest man on earth when this is solved :)

Hour Tally

Business: 1.28
Coding: 8.07
Internet: 1.14
Testing: 3.83
Total Logged Time: 14.32
“Whether you think you can, or you think you can't--you're right.” ~ Henry Ford
Post

Re: Week of April 21, 2013

#5
Thursday, April 25, 2013

Summary

Well, I like intensity. Really, I do. But..days like today...they're a bit much even for me. I'm putting myself in 200% gear (a gear that I didn't know existed until today..) until these hardware issues are fixed. I've come so far, and I've got a lovely universe to show you all, I can't let some stupid driver differences stop me now.

Today I dug deep into the nastiest, darkest corners of the engine - the layer that sits right on top of OpenGL. It's the layer that has the most intense code, and also the one that would cause hardware incompatibilities. I switched my laptop into integrated chip mode, turning off the dedicated GPU, and went to work trying to stop LT from crashing on this terribly-underpowered thing. Quite a challenge, to say the least. Nonetheless, this is really a fantastic strategy for getting things right. The driver for the integrated chip is extremely unforgiving, quite unlike NVIDIA's drivers. But that's exactly what I was looking for. I'm really lucky I had this, else I would have had to frantically scour around for an ATI machine (and trust me, after calling a few places today, I figured out that rural Tennessee isn't the best place to be shopping for a gaming PC :P )

Although the progress is extremely slow with this kind of work (most of my time is spent trying to guess at what the incredibly-vague error messages mean, reading obscure forum posts on the OpenGL forums where people were having similar issues, etc.), I readily compensated with shear volume of time, as the numbers below will attest to. Ultimately, I made a good bit of progress. Yesterday, I couldn't even get past the launcher on this thing. Today, I can get to and see the main menu. It might not sound like much, but it was an awful lot of engine re-writing to get to that point :( But, more importantly, I can see that the mesh generator is working. That's a big triumph!! I still can't see the generated objects, because there's some obscure shader problem that's preventing them from displaying..but I know they're there.

I'm about ready to ship another attempt to the ATI folks, and pray for the best. It's a sad way to have to work, but it no doubt will make me better :) You know, in the same way that stabbing your hand makes you more resilient to hand injuries :roll:

Now, honestly, I'd really like to launch into a very long rant about OpenGL and NVIDIA. But that's not what these dev logs are for, so I won't I will only say these two teensy things: OpenGL is a horrendous API, and NVIDIA does a very large disservice to the graphics programming community by making their drivers "friendly." See? That wasn't so bad :geek:

Truly, I apologize to everyone who is hyped to play the prototype. I want to get it to you!!! And really, I'm doing everything physically possible to do so. I appreciate your patience. At least this is one more issue that will be done and dealt with for release time :)

Hour Tally

Coding: 13.60
Internet: 3.90
Testing: 2.39
Total Logged Time: 19.89
“Whether you think you can, or you think you can't--you're right.” ~ Henry Ford
Post

Re: Week of April 21, 2013

#6
Friday, April 26, 2013

Summary

Remember what I said about bad lows bringing about great highs? Yesterday was an awfully bad low...and today?

8-) 8-) 8-)

Ok, you got me, I'll admit it. Today I am proud of myself. Yesterday, I took a shot in the dark (well, from the hours, it was actually a lot more than just one shot). I had no way of knowing how to fix the hardware compatibility problem. I had no way of knowing if any of those 20 hours would end up being useful in any way. But...

THEY WERE!!

I was not expecting it when I heard back from the first ATI tester last night, who informed me that all was well. I was just...what....!?!??! First shot in the dark...landed square on that issue's forehead :D By the time the second and third successful ATI tests had come in, I was just elated. It worked. It really worked. And in only one day :)

I'm sorry, I'll stop now, I promise I'm not usually big into bragging. But I have to a little bit this time, because I was so miserable yesterday and so uncertain if anything I was doing would pull through, and today I find out it did in a very big way :D

With that little slice of misery out of my life, I am pushing forward with wrapping this thing up and getting it out ASAP!!! I'm trying to smooth over any remaining rough edges, and incorporate some of the feedback from the early testers. I think I'm going to drop Desura and push it out using a digital distribution service that I signed up for as plan B. It has been working flawlessly for distributing the files to the testers, and it doesn't require any silly admin authorization. So this means that as soon as I get this thing finished and the video made, you'll have the proto sitting on your doorstep :D It'll be a long night, and, yet again, the bet is back on - no sleep until release!!! This time hopefully for real ;)

To those of you who still read this rollercoaster-like pile of drivel, I sincerely thank you for your patience, and apologize for the ups and downs :shock:

:)

PS ~ Yes, I treated myself to some serious sleep last night when I got word of the success. It was glorious.

Hour Tally

Business: 0.07
Coding: 8.54
Internet: 1.55
Testing: 4.09
Total Logged Time: 14.24
“Whether you think you can, or you think you can't--you're right.” ~ Henry Ford
Post

Re: Week of April 21, 2013

#7
Saturday, April 27, 2013

Summary

Great day of polishing and finalizing!

Tried to get as many of the easier/medium-easy fixes and improvements done as I could. There's not enough excitement in any one feature to merit a lengthy talk.

Alright, let's face it, I have a problem with keeping things to myself - but I can't hold it in...I've packaged up what I think may be a "final" build...! If I get a final green light from the testers...then..... :) As you can see from the list in the Prototype forum that DWMagus was kind enough to compile, I've knocked out most of the big issues, and done a lot of the smaller fixes as well. But I don't want to hold up the release just to do a lot of extra polishing - this isn't intended to be a final, polished gem (although it did turn out to have a much higher degree of polish than I was initially planning :) ). I will keep working on those improvements, and will release patches as I get them done!

On a side note, I ordered a Thrustmaster joystick, and it came in today. I was super excited to get to play with it! Unfortunately, LT does not play well with a joystick at the moment. But that's not really surprising, since I haven't tried it before. The point is that now I have one, which means it's on my mind, which means it'll get done properly at some point :) Gamepad and joystick support is a go! I think I'm going to have loads of fun with them both.

Alright, I keep talking about this tutorial, but tonight, I'm actually making it :) For real.

No weekly summary until I finish 8-)

Hour Tally

Coding: 6.70
Internet: 3.89
Testing: 4.89
Total Logged Time: 15.49
“Whether you think you can, or you think you can't--you're right.” ~ Henry Ford
Post

Re: Week of April 21, 2013

#8
Week Summary

General Sentiment

Another killer week :shock: Since I'm writing this summary so late, I can hardly remember what happened - my mind already blocked it out because I think it was pretty traumatizing. But I do vaguely recall there being something about a hardware compatibility problem or two...? Maybe?

Major Accomplishments
  • Fixed ATI compatibility problem
  • Fixed *some* Intel HD Graphics compatibility problem (although some are still unable to run LTP)
  • Improved combat AI
  • Added a lot of new AI behavior, including following that can dock/undock, as well as patrolling
  • Improved mission balance
  • Implemented asset namer
  • Overhauled graphics for thrusters, beam weapons, and pulse weapons (much better!!)
Combined Hour Tally

Business: 3.22
Coding: 63.37
Composing: 0.07 (??)
Internet: 16.68
Testing: 28.49
Thinking: 0.69
Total Logged Time: 112.52

( :shock: )
“Whether you think you can, or you think you can't--you're right.” ~ Henry Ford

Online Now

Users browsing this forum: No registered users and 3 guests

cron