Results 1 to 6 of 6

Thread: Developer's Journal

  1. #1
    Administrator
    Join Date
    Oct 2010
    Location
    Texas
    Posts
    2,119

    Developer's Journal

    15 July, 2015

    Man, what a CRAZY month it's been. I mean, I knew when eAccess released there would be a lot of additional work that would go along with it (website input, PR, forums, tech support, etc, etc); but WOW... Managing the forums alone was taking about three hours every morning for the first couple of weeks, and then perhaps another hour or so before bed. Admittedly, too, I had to take a few days off after the release to recharge a bit. I was really burning the candle at both ends a couple weeks before. Then of course I fell a bit ill. Blah, blah, blah.... Point is, I lost too much time for development and a couple items didn't make the update.

    All that said, what a great month it has been. First, I can't BEGIN to tell you how jazzed I am about the initial reactions. Apart from the few comments I've read about "ships in the future would be automatic this," and "wouldn't have as many buttons that", reactions seem to range from cautiously-optimistic to REALLY good. I can live with that, especially considering the lack of content at the moment. I think a big help there was that I was completely open about that fact, rather than trying to hide it.

    ISI and I expected sales to be slow at first--new game from an "unknown", one-man developer currently, early access, etc., etc. That said, we had more eAccess supporters now than I had Kickstarter backers in the same number of days. Not a HUGE gain, but a gain none-the-less. For example, thanks to you all I was FINALLY able to start compensating Michael Manning for the great sound and music work he did to get early access started; so, that made me feel really good, honestly. He certainly deserves it. Michael is getting married this week by the way, so congrats to him.

    Apart from sales though, RogSys has also had a good bit of exposure lately and a lot of people do KNOW about it now. At this point, I think many folks are kind of waiting to see where this goes before making a commitment. There's lots of reason for that, including some less-than-stellar results from other crowd-funded projects. It's been putting a bad taste in people's mouths, I think. That's fine--it just means I need to work a little harder to let people know that I'll make every effort to actually do what I say I'm going to do. That's why I felt it was REALLY important to get the update out today as I said it would. These things need to occur reliably; and each needs to show progress. As I said, my only regret was a couple items didn't get in.

    If you followed the change log thread, you pretty much know what I DID get accomplished this past month. With the first update out of the way then, it's immediately time to start talking about what the plans are for update 2 (tentatively scheduled for August 15th). I had listed a few key elements in that change log thread; but after the issue today, I think I need to alter plans a bit.

    Item 1--Mission condition management and initialization refinements:
    If you don't know, one of the missed items for Update 1 was a tutorial mission for the FLUX Drive. The only reason it's not there was because I couldn't put a station in orbit around another planet. The code worked a couple months ago, but obviously something I changed in support of something else broke it. What this says to me is that the mission start condition code is too cumbersome--too reliant on other elements. SO, the first order of business then will be to spend about a week on revamping this so that it does what its supposed to do, and does it with less dependency. What I don't want is to try and start adding real game-play missions later on and get thwarted every step of the way. Better to deal with the issue now.

    Item 2--Combat-related items:
    From there, I need to spend time on combat and damage related items. For example, currently ship components are not being deployed to their proper locations for mission-created ships (again, mission stuff). This means you can shoot at a ship all day and not hit ANYTHING other than armor. This is the sole reason why the Predator test AI has not shown up in any missions. It does fly, steer and shoot, but what's the point if it amounts to nothing? I'd also like to get ship-to-ship collisions re-enabled. I turned it off for the initial eAccess release due to some bugs. I NEED to get those hammered out since everything that comes next depends on it. I also want to make an effort to get at LEAST a dumbfire rocket put in. Does it make much sense in space combat? Not really. However, the systems to manage the missiles and deploy them are critical. Finally, I'd like to investigate a rudimentary system of having parts get blown/ripped off a ship. It won't be a flashy implementation where bits and pieces break off--that comes later. Right now I just want to explore the idea of a ship part having a certain joint strength, and what happens when it's hit with a greater impact force. That's a lot for two weeks of work, so I don't think it will ALL get done.

    Item 3--SAN:
    I want to get a basic version of Suspended Animation working. This is really going to have to be done in steps because it's so all-encompassing. When you SAN, although it will seem like you skipped time, you really haven't--you just "slept" though it. That missing time has to be applied to EVERYTHING--from the motions of planets and moons, to ship trajectories, to ship systems. The first step then will be to apply the time to the orbital trajectories of all celestial objects and ships (currently assuming that no piloting is being done and the ships are in drift). Along with this then I'll need to implement a simple method to enter SAN, a simple effect to represent it (a black screen will do for now), and some trigger to pull you out of it.

    Internal Docking?
    Because of the addition of the mission infrastructure work, I may have to push internal docking to update 3. The code works for the most part; but it will require art support which I may not now have time for. At this point I think it's off the plate; but let's see how the month unfolds...

    Finish up the burn calculator
    I've got a good start on this, but there's still a lot of issues to sort through. If I had to pick between this and internal docking, I'd say I'd have a better chance of completing this so it's probably what I'd do if time allows...

    And of course there's the little bug fixes and tweaks that I find time for here and there. I don't plan for these. If I happen to be near the code where something needs attention then I take a few extra moments and do it. It's good to be sort of free-form with these things rather than letting myself get bogged down by them.

    And there you have it! This month and next in a rather large nutshell I'm going to lock this thread so it doesn't get cluttered with responses, but by all means PLEASE feel free to create a thread to discuss anything here if you have the urge to do so. Otherwise, one last time, let me thank you all for the tremendous support, great suggestions, and outstanding discussions we've had here thus far. I'm VERY happy with the growing community and the communication that's been going on.

    Thanks,
    Michael...

    PS -- please excuse grammar and spelling errors if any. It's 2AM here and I AM starting to get a bit tired now
    __________________
    "There is no spoon..."

  2. #2
    Administrator
    Join Date
    Oct 2010
    Location
    Texas
    Posts
    2,119
    15 August, 2015

    Ok, before we go any further--the update It IS done and has been uploaded so Tim (the gentleman from ISI who handles the website, download mirrors, etc) can take care of things on his end. However, as I mentioned in the change log thread, he had plans for this evening, and I don't expect him home early enough to get the update out today. So, look for it early tomorrow (Sunday, US Eastern). As soon as I know it's up I will make the announcement post, and edit this entry to reflect that.

    EDIT: Update is available now here: http://imagespaceinc.com/rogsys/p/downloads/

    With that out of the way then, these past few weeks were MUCH better in terms of development time. The preliminary rush from the launch finally settled down, and the forums are great so far with people answering various questions, and folks helping other when they can (thank you all SO MUCH for this). I'm sure you'll notice the difference in length of the change logs between update one and two. So, what did we get done?

    First, I did a LOT of work on mission initialization, event, and goal management. It's also now possible for mission ships to begin anywhere within the system, and the apoapsis and periapsis of starting orbits can be defined. You'll notice this functionality with the new single mission, which is a rescue mission in which you have to save a training platform from crashing into the moon it is orbiting. Any flight's tasks can be marked as a mission secondary, or critical, goal. You can also define how many critical goals must be completed for the mission to be considered a success. You'll also notice that various tutorial missions begin in different locations within the local star system.

    Next up was fixing several elements that were breaking combat-related items, such as damage from impact forces. With these items fixed I was able to re-introduce the AI Test mission. This is where I'll be developing pilot AI functionality. Currently, there is one AI ship in this mission that will target and attack you as long as it is able. You'll immediately notice that the AI currently doesn't aim all that well. The main reason for this is that the AI has to fly exactly as the player does--by feeding steering input into the flight controls. This means I can't do the typical simple rotations and point-at's that most shooters use. Instead, the AI has to know where the target is, which way to steer towards it, and then determine the inputs to make. This is MUCH harder to do, and it will take some time to get it working well. The initial goal will be precise aiming. From there, I can make use of the values and conditions that will determine AI pilot aiming skill.

    The other combat-related item that went in was the preliminary secondary weapon system--missile systems basically. This introduces the idea of hardpoints, missile racks, and the missiles themselves. The goal for this update was to define the data management required, get the systems up and running, be able to launch a powered weapon, and have it leave the rack and ignite its engine. This all works now. Please note though that there is no hit detection yet, so no impacts nor damage will occur.

    Finally, the last big-ticket item was Sleep/SAN (Suspended ANimation) time acceleration. This is critical for long-duration flights of hours, days, even weeks. Since we're dealing with real orbital mechanics, scales, and velocities there HAS to be a way to accelerate time to make these trips practical from a gameplay standpoint. I mean, you certainly COULD fly a hours (or days) long mission in real time if you wanted, but I don't know many people that would actually have the time to do that The first phase of implementing this was to get the orbits of celestial bodies and ships to update properly. Also, a ship's velocity needed to be updated during acceleration by Flux Drive propulsion at least. I'm happy to say this all works (and works pretty well, if I'm honest). In the testing I've done over the past couple of days, I haven't noticed any orbital weirdness caused by the accelerated time.

    To go along with this, I added a very preliminary "destination alarm" system. Once you select a destination, you can set an alarm for midway, or end point (edit: it appears there may be a bug with setting the endpoint alarm--I've noticed a few instances of it not working as expected). The alarm will not go active until you begin a FLUX or MTS Boost burn. Once on your way you can go to sleep. When you are within 15000 km of the alarm's goal, you will wake up to make adjustments and begin your braking burn if needed. From there you can set an endpoint alarm, sleep, and wake up again at your destination (or close to). Otherwise, you currently will wake up every 20 minutes. Obviously, there's a lot of refinement and functionality to add, but now that the basic system works tweaks are a non-issue. Getting the physics to play nice was the important part.

    Otherwise, lots of small fixes and refinements across the board based on the feedback here. Thanks for all the great posts with ideas, suggestions, and bugs reports. You all ARE making a big difference! The one item I didn't get to for this update was further refinement to head-tracking. I know there were a couple suggestions regarding the changes made for update 1. I hope to get some more refinement for this done for update 3.

    That's all in the past though. Let's take a look at what I have planned for update 3....

    Item 1 -- Art for the first final player-flyable ship
    It's time to start getting final ship art going in. I'll start by working on the "Flying Fox" combat support ship. I'm going with this one first because the design is further along, and it's mission profile will get us a certain bang-for-our-buck in that we can add a bunch of diversified test missions that will hopefully give you all new things to do while the combat elements continue to take shape. Plus, it's a great design and I really want to fly it For update 3 I want to have the basic exterior modeled and textured, and a preliminary data file set up so it can be flown. I'll need to create a placeholder interior (probably using elements from the current FireArc's cockpit module). It WON'T be pretty on the inside--just fair warning.

    Item 2 -- Burn Calculator (FLUX Propulsion) and Destination Aim-At indicators
    Pretty much self-explanatory, really--I want to take some of the guess-work out of orbital transfers from one body to another. In my couple days of testing with Sleep mode, it became quickly obviously that while constant-g acceleration makes transfers more straight-forward (no pun intended), they can still be unpredictable. This is especially true when transferring to faster-moving bodies.

    item 3 -- AI Piloting
    Need to get the AI aiming more accurately. Apart from close-in fighting (which agreeably may not occur nearly as often as stand-off), being able to aim the ship well is critical for docking and other precision maneuvers. The sooner this is working properly, the better.

    Item 4 -- Preliminary dynamic mission generator elements
    This will be mostly foundation work--data structures, class and method frameworks, etc. By update 3 I want these fundamentals in place.

    Internal launching and docking
    Once again, we'll try to get this back up and running if time allows. Since I have the fundamentals already in place for this (and that we have a functioning external docking system) I'm not as concerned about this as some of the other items on the list. It needs to be done, but it can slip an update or two if required.

    Of course, I'll continue to hit bugs and make refinements when and where I can. Oh, I have not forgotten VR support. Honestly though, if I have to pick between VR and all the gameplay critical elements that need to be done the choice is obvious. Again, I'll be sure to make the proper post as soon as I know Tim has gotten the update ready to go. Sorry for the slight delay on it. Going forward I think I'm going to try to aim for a weekday for updates as weekends are always iffy...

    Thanks,
    Michael
    __________________
    "There is no spoon..."

  3. #3
    Administrator
    Join Date
    Oct 2010
    Location
    Texas
    Posts
    2,119
    21 September, 2015

    It's been an interesting month, both in terms of development, as well as some personal issues that kind of crept in and broadsided me. I'll talk about those in a minute. RogSys work itself was going fine the first two weeks. I was making good progress on the "Flying Fox". I don't want to post any images because it IS floating around somewhere in tutorial 7, in orbit around one of the bodies in the star system. Anyway, I got the exterior modeled, ran a nice first pass Ambient Occlusion map (helps with definition), as well as a self-illumination map (spot lights to light up the ship so it's visible, tail numbers can be read, etc).

    I spent the second week fleshing out the interior, which was a bit tricky because I designed it for use with the current, simple character controller, as well as the hopefully upcoming zero-g controller. I have to say, trying to think of each surface as a potential wall, floor and ceiling was an interesting challenge. I fully expect to redesign the interior once we know for sure which sort of character controller will be used. It's design requirements will be more focused, and we can get more out of the internal space then. In any case, what I came up with is pretty nice, with beds that rotate so the occupant is always pushed down regardless of the acceleration vector. The same goes for the "kitchen" area. There are two service decks below the main deck where system computers and such will be mounted. Although a bit hard to read, here's an image of the roughed out interior space:



    As you can see, I'm currently using the pilot's seat from the FireArc cockpit module, but that will be changing soon...

    So, that was a bit over two weeks QUICKLY chewed up and swallowed by art work. If you notice the shortened change log for this update, that's the main reason why. As I mentioned in, I believe it was the PC Gamer article, just ONE artist would cut my workload by half, leaving me more time for code. This update is the perfect example. Just think if I'd spent the entire month on art--not only would the ship still not be finished, NO coding would have taken place. It's a big balancing act to continue to make progress in all areas. I'm not complaining mind you--it is what it is. Just stating the obvious fact

    For the second half of the month then I moved back to code, with a few goals in mind (as I mentioned in the last dev journal). Of prime importance was the burn calculator (which has been renamed to the XFER Panel on the HMD Navigation page). Without this, the visual indicators for the pilot, and autopilot coordination, body to body travel would be problematic at best. I'm happy to say I have a good foundation working now in Update 3, but it came at a cost. Getting this to work was no small feat for me. I have a certain knack for self-teaching, so once I find the right material I can work it out. However, that doesn't always translate to what the sim needs in order for everything to work right.

    Worse, for several days I was fighting a losing battle. There were two DEEP-rooted bugs that I didn't know existed, and they were defeating my solutions without my ever knowing it. Fortunately, after the third day, Angel (my wife) suggested I look deeper down than to the level I thought things were breaking at. In hindsight it was an obvious course of action. Being stuck in the middle of it--sometimes it's hard to see clearly Anyway, sure enough I found the two issues.

    The first was more obvious--gravity well determination for ships was skipping the inner most planetary system, so it was impossible to achieve an orbit. The other was much more subtle--a VERY slight yaw drift caused by FLUX propulsion was throwing off the solutions. Absolutely not noticeable normally, but thanks to SLEEP, after time skipping for a couple hours, it became obvious. Tracking THAT down was not fun, but I found it. From there, the constant-g transfer solutions could be validated by flying the trip manually. It's VERY important to be able to fly manually just in case autopilot is lost. Happily, with just two new indicators you have the bulk of the information you need for a constant-g flight path.

    Getting the autopilot to then tie in to the NAS solutions became straight-forward. By this time though, it was nearing the end of last week. I knew the update wouldn't be ready in time, hence the delay. Knowing I have to get a move on with Update 4's work, I decided I had to go into overtime for the weekend. I got the remainder of the NAS work done, and worked up the data file so I could get the Flying Fox to appear in the sim. It's not flyable, which I was kind of shooting for, but you can't win 'em all sadly. Needless to say, there's still a lot of enhancements to be made for constant-g transfers (not to mention the work still needed for typical "burn and drift" transfers) but we've got a good start on it now.

    I did get to look briefly at AI aiming. While I didn't get time to actually dive into it I now have a few good ideas where the improvements can be made (in large part based on the work I had to do to aim a ship at a target several million kilometers away, and fly to it so accurately you hit it dead center). I'm pretty confident the improvements I'm wanting to make will come quickly (famous last words there!). I also DID manage to get the general data structure and framework set up for the dynamic mission generator. Obviously it does nothing yet so you won't notice anything; but it did allow me to get my head wrapped around what NEEDS to happen....

    Something else took a good chunk of time away from me lately. This is a bit personal, so feel free to skip it if you'd like. Both my daughters left for college just after mid-August. They earned it--both worked hard in school and got some nice grants and scholarships to show for it. I have to tell you though that after devoting the last 20 years of my life to them--making sure they were cared for, provided for, grounded, open-minded, etc., etc.; to have them suddenly not be here... honestly I was lost for a few days. Things weren't making a lot of sense to me all of a sudden; and it was incredibly hard to really concentrate. Once I got over them just not BEING here though things sharpened up again. That was actually minor, because no matter how bad I may have felt, Angel was devastated.

    Let's face it, they are my daughters, but they are her BABIES--now gone from home. That's a tremendous hole that I can't imagine. I can't even PRETEND to imagine. I'll tell you that I took a couple good hours a day off and spent them with her. Taking her to a movie, tooling around the mall, grocery shopping--just keeping going. I was just doing my best to let her know she was still loved and needed. I'm happy to say she's on the other side of the tunnel now. She still hurts, I know; but she's remembering that we're still us and we have our own life.

    Don't get me wrong. I mean, I KNOW we still HAVE daughters. In fact, we're spending Thanksgiving week in Estes Park, CO (they both go to school in Colorado, which is where I want to move to because I LOVE it there) and we're all looking forward to that. It's just when they're not there everyday like there were for what seems like FOREVER... You older guys know what I'm talking about In any case, we're working through it, and as you could see by the big push I made these last few days, I've got my focus back...

    Blah, blah blah... So, what's on the plate for Update 4?

    Item 1 -- Continue working on Flying Fox
    I want to get a nice set of textures on this thing. More important though is to get the interior working so you can actually fly the ship, move around inside it, etc. To do this, I really need to finish the proper implementation of the VMS display. The art has already been built to support it, so it's more code than anything else at this point. Otherwise, I probably will still use cockpit assets from the FireArc for now as I'm not quite ready to start designing final control panels yet. After that, if time allows, I want to explore the idea of being able to pick something up and carry it around. Part of the reason to have an FPS element is so you can DO THINGS inside (and outside) your ship, such as making repairs. OR, I might do some work on controlling the robot arm.

    At this point though, knowing the amount of art required, I'll be happy to just get the ship flyable. Besides, the designs for the smaller "fighter", larger "bomber" and your home station are coming and I'll need to get those started ASAP....

    Item 2 -- AI Piloting
    I REALLY want to get this knocked out (aiming and steering). I also want to see if I can get an AI pilot to fly his ship from one moon to another, and establish a simple orbit once there. If so, then I want to see how hard it is to fly formation manually. This will give me a good idea of the information leaders will need to pass off to those in their flight so they can stay together. Getting the AI aiming sorted out is the goal though.

    Item 3 -- Anything else I can squeeze in
    This is kind of vague, but considering how much time art takes I'm not 100% sure what else I'll be able to get to. I don't want to make too many hard goals then. Rather, I'll just try to knock out a few items I've been thinking about that SHOULDN'T take too much time, assuming I can get to them. Just keep an eye on the change log to see what I'm working on...

    Lastly, thanks again for all of your patience with the delayed update. I know I make a big deal about the punctuality of these, and I have many reasons for that. The biggest is that they REALLY do help keep me on schedule. It's VERY easy to get lost in all of everything that needs to be done. Defining clear goals when I can with solid deadlines really does help maintain focus--at least for me As always, if you have any comments feel free to start a thread to discuss...
    __________________
    "There is no spoon..."

  4. #4
    Administrator
    Join Date
    Oct 2010
    Location
    Texas
    Posts
    2,119
    26 October, 2015

    As always, it's been rather hectic the past week or so with preparations for Update 4. Art, again, took a fair chunk of time from the development cycle; so much so in fact that I almost thought I wouldn't get everything done. But, I'm glad to say everything I wanted to finish, I did... kind of.

    My goal for this update was to do both the interior AND exterior art for the new Flying Fox. As I was working on exterior art though, I realized that if I rushed it to be able to move on to the interior, BOTH elements would suffer visually. Since this art probably won't get another pass until later in the Alpha stage, I decided it was best to hold the interior, and work the exterior up to 70% of completion. Doing this, I'm much more confident the art will stand up for WIP screenshots and videos. Plus, we'll be looking at this for a few good months--I didn't want it to become hard on the eyes. Obviously then, I'll be beginning work on the interior art for the next couple of updates. More on that in a minute...

    Another goal for this update was to improve AI direct aiming. I DID have time to work on this, and have made an improvement. Once the AI is lined up, I think you'll find it now scores more hits, more often, from a greater distance. What's throwing the solution off a bit now is the slowness at which the AI aims the ship. It's easy to defeat right now because they simply don't rotate with their full potential to precision. I was hoping to be able to work on this aspect, too; but that time escaped me in favor of the art task. So, I'll be working to improve this next.

    Beyond that, I did find time to work on a fair amount of smaller items, one of which actually turned into a BIG item. If you recall from Update 3, the Flying Fox was disappearing from the orbit it was in around a distant moon. I THOUGHT this was a simple mission code bug concerning object placement. Turns out that this was only PART of a bigger problem. The placement was in fact bugged. For example, if you expect an object to be in orbit with a periapsis of 400 km and an apoapsis of 500 km, but the mission code places the object at a periapsis of 900 km, obviously the orbital velocity is all screwed up and the orbit degrades quickly. Once I had this fixed I still found I had a problem though.

    One issue was Leadwerks was not returning 64-bit object position data. The truncated 32-bit data was causing errors in orbital calculations, and that was degrading orbits. Switching to 64-bit position data (as was originally intended) fixed this problem instantly. However, there STILL was another problem--entering SAN/SLEEP was causing celestial bodies to "jump" slightly out of position, which also adversely affected orbital calculations. Finally, when an alarm triggered the player to wake from SLEEP, the SLEEP event was not cleaning up correctly, and this was also causing celestial bodies to jump out of position by many kilometers.

    Once all of these items were fixed, the orbital simulation was once again stable. So much in fact that my new test "sandbox" mission finally worked properly. The sandbox mission puts at least one object around each planet and moon in the system. You start docked, cold and dark. Your two critical goals are to undock, and eventually dock again. What you do inbetween is up to you....

    Beyond this, Michael and I were able to add a few new audio items, one of which was added to demo some new signal functionality. It is intended that the SAI will generate feedback audio for things the pilot normally wouldn't hear (weapon hits, for example). While there are no volume controls for this yet, we did get in SAI feedback for FLUX-generated thrust, weapon firing, reloading, and hits on an object. More importantly though is the new training platform music channels.

    It's been a feature goal that signals not only travel through space, but also can be blocked by celestial bodies, AND signal data can degrade over range. You may have noticed before that you could lock on to a localizer signal for a ship that was orbiting a completely different body. This is no longer the case. Signal data will degrade over range, and if a celestial body comes between you and the sender, the signal will be blocked. To demo this, Michael was kind enough to put together a few original songs in a FMOD playlist, and I assigned a music channel to each training platform. Music frequencies, for now, always end with .301 So, if you entered something like 45.2.301, and the signal actually reaches your ship, you'll hear music. As you get further away, you'll hear more and more background noise and static as the signal degrades. Eventually, the signal is completely "lost".

    This audio demo is not wasted, as it directly applies to an ATIS-like feature. Any ship with an STC will broadcast information to update you on local conditions. Likewise, eventually voice communications will have proper audio, too; and this will also need to be affected by the same type of blockage and degradation.

    So, let's look ahead to the next update. First, Thanksgiving is coming up and so that is going to truncate development time in November. What's more, we won't be spending Thanksgiving at home. Instead, I will be out of town for the holiday week with my wife; and we are going to pick our daughters up from college (in Colorado). We're all going to spend the week in Estes Park, CO, in the mountains. We normally try to do this every summer, but it just didn't work out this year. So, we decided to do this for Thanksgiving. We LOVE the mountains, and mountain hiking, so I'm REALLY looking forward to unplugging, getting some fresh air (and maybe even some snow), and clearing my head.

    What this all means is that I'm looking at only a 3-week update cycle. That is nowhere NEAR enough time to get the FF interior art done, AND do the control art so the ship is actually flyable with a proper cockpit interface. It just won't happen. So, here's what I'm thinking: I'm going to do what I can on interior art during the day, and get some coding done in the evening. I will then release what I have stable on November 17th as "Update 4.5". After the holiday, because of Christmas, we'll be looking at a similar situation for December; so I'll work until about the 20th of December and then release that as "Update 5" proper. By Update 5, it's my intent to have the Flying Fox flyable.

    Of course, I could just say there won't be any update in November, but I REALLY don't want to do that. I think I can get enough done to make an update worthwhile. So, like it or not, this is what I'm doing Considering this then, for update 4.5, we're looking at:

    Item 1 -- Flying Fox interior work, part 1
    Simply, anything and everything that needs to be done to get the interior art completed to 70%, get the cockpit controls working, get VMS implementation finalized--anything I can get done by Nov 17th towards this goal....

    Item 2 -- Control Binding enhancements, part 1
    There's been a lot of suggestions for additional binding functionality and features. I'll prioritize these items tomorrow and then start knocking them out as I can.

    Item 3 -- Ship Part detachment due to force and damage, part 1
    An example of this is damaging a radiator by running it into a station during docking, including the possibility of having it become detached from the ship. This does not include more refined forms of damage such as bending or shattering. Right now we're looking strictly at "coarse" representations.

    ISI and I do have a goal for getting Rogue System to Steam in six months, and I'm developing a roadmap of items that should be done to make this a successful attempt. Basically it's business as usual, but I'll need to shuffle some tasks around to facilitate some things I want to achieve for the 1st Steam eAccess version. I'll be talking more about this soon.

    Otherwise, as always, thanks for all the kind words, suggestions, and constructive feedback. Also, thanks for the extra effort in trying to help spread the word about RogSys. I will tell you that there are noticeably positive results whenever you guys promote a video, or get a popular discussion going about RogSys. I truly appreciate the extra effort you all are putting forth in trying to make RogSys' early access a success!
    __________________
    "There is no spoon..."

  5. #5
    Administrator
    Join Date
    Oct 2010
    Location
    Texas
    Posts
    2,119
    01 January, 2016

    HAPPY NEW YEAR! I seriously can't believe another year has come and gone--seems like it was just yesterday that I was beginning to ramp up for the 1st eAccess release around this time last year. Now it's 2016 and I'm ramping up for the upcoming Steam early access release. I'm not sure of an exact time frame yet, but it is coming. So, before I go any further I just want to take a quick moment and say THANK YOU to all of the current eAccess community. The support and feedback you all have given thus far has been thoughtful, well-written, and very useful (although the usefulness may not be readily apparent yet). With that, let's talk about a few things going into the new year.

    As noted in the FAQ and other places, RogSys is privately funded to a minimum level of completion, hence RogSys is NOT crowd-funded; but rather its expedited development is crowd-supported. ISI is the publisher, and specific eAccess purchase data is their private information. While it's not my place to publicly divulge this info, I do like to update you all from time to time on eAccess by giving you a general rundown of the current state of affairs. As I mentioned previously, in order to ethically and responsibly be able to hire other team members for a reasonable amount of time, eAccess sales had to reach a specific level. As I wrote here, 10K eAccess package purchases at an average of $60 would cover ISI's initial investment and allow me to hire some additional help. To date, we have not reached this. In fact, we're WELL under this mark; but sales still continue at a steady (albeit slow) pace. Considering RogSys has NO ADVERTISING, is still in very early access, AND is currently being sold directly from ISI's commerce system, I don't consider this a failure by any means! Obviously though, additional team members are not yet an option.

    What this means is that since I'm still a team of one I remain forced to continue aiming for the minimum feature set for the 1.0 release. This means that things like character interactions in between missions, storytelling cutscenes, and other elements of this nature will have to come in a post 1.0 update. There is still a chance though for crowd-supported extended development to work, and this is with the upcoming Steam early access release. There are elements that ISI and I have identified to give Steam early access the best chance of success. These include the introduction of more final gameplay elements (beginning with the first "final" flyable ship--the Flying Fox). Likewise, we're coming up with a few ways to help further spread the word about RogSys once it hits Steam. You wouldn't BELIEVE how often I still here, "I didn't even know this game existed until today." Obviously, my focus will be to work on the gameplay and art needed for this in the immediate future.

    I have been working hard and am proud of what's been done thus far. Most of all, I'm VERY proud of the overall stability of each RogSys update (especially when compared to other titles with far larger teams and budgets). I try VERY hard to make sure things work reasonably well without crashing the game, OR your system, every few minutes. Considering this, I will admit that watching people continue to throw hundreds of thousands of dollars at Star Citizen (which Chris Roberts has said has enough money tucked away to finish SC if all funding stopped immediately, if you believe that), while I struggle to hire a couple artists to help out, has been maddeningly frustrating at times. I KNOW RogSys can be incredibly unique, fun, and challenging. What's frustrating is that I can't make the progress I want each month because I'm stretched thin.

    Does this stop me, or make me want to quit? No! In fact, it is a somewhat effective motivation in some ways Regardless, it is what it is; and so all I can do is continue slowly and steadily. We'll get there eventually, I have no doubt....

    Moving on, you may or may not have seen that I've been dropping some updates here and there over the holiday season with continued work on the Flying Fox interior. The most recent update showed that I had gotten the interior structure and pilot's station art in game and was able to move around in it.



    Obviously the interior art isn't final. I left it at 70%--still in need of some detailing and texture tweaking, which will get hit in the revision pass later in Alpha.

    I'm currently setting up control art, which I hoped to have done already. Why am I running late? You all know I don't beat around the bush, and the honest truth is I started burning out shortly after the start of December. As I've said before, I've been doing this long enough to know when I'm reaching my limit; and I started feeling this right after the Thanksgiving break. I immediately decreased my work day and made sure I got more sleep, got outside and did some more R&R; and was trying to get more exercise. You HAVE to do this or you're going to hit a brick wall and everything will just come apart (Josh over at Limit Theory knows what I mean--hope he's feeling better). All you can do is slow down until you recover. Uncontrolled burn-outs can be devastating to the mind and body--severe depression can set in; and anyone who suffers from depression knows how that can make you feel physically.

    You might ask why I didn't just take two weeks off completely, or something similar. That actually does not help at all. What happens is you just get accustomed to not working, and then when you HAVE to go back to work it hits you all over again. This kind of happened to me over Thanksgiving--I took the 10 days off and started feeling very light and relaxed. As soon as I got home and got back to work the weight hit me all over again. That was the big clue that I was burning out. No, all you can do is throttle back, continue to work, and push through steadily until you come out on the other side. Meditation also helps a lot to refocus. You also need to remind yourself over and over again that it's all worth the time and effort. If you don't believe in what you're doing then there's no point in carrying on.

    So yes. All you can really do when you're facing a burn-out is to manage its effects as best as possible until you recover. It's just like being sick--you rest, take the "medicine" you need to fight it, and try to get well soon. I'm happy to say I'm on the other side and am feeling a LOT better. I'm refreshed and ready to push again

    Posting updates was a way to prove to you all (and myself) that I was still making progress. That's very important for self-motivation. Sadly though, I didn't make the progress I had hoped I would. The problem with working solo is that there's no one to help out when you're down--there's no support. Hence, there is a reason why this developer's entry comes without an update for the first time. The honest fact is it's just not ready yet. I do understand if this disappoints you; and I'm sorry for the slip. All I can do is promise you that I will get back on schedule as quickly as possible; and release the update as soon as the work is finished.

    So, all that said, let's talk about what's coming up. Top priority is to get the pilot's station for the Fox working ASAP (which includes final VMS integration). This needs to get done, and get done SOON. Period. I also want to finish up the first round of control binding enhancements, as well as the detachable ship parts. Next, I'm going to do a bit of visual FX work, rework all the current tutorial missions to work with the Flying Fox, and also add as many Fox-related single missions as possible. These include ship rescue, satellite recovery using the robot arm, and whatever else I can do. With this, I hope to have enough gameplay to create a worthy Steam early access release. We could PROBABLY go now, but I think it's very important to launch with more gameplay.

    During all this work, ship systems will start being finalized, and the physical pilot reactions that you saw in the kickstarter prototype will start coming online. I'll also start introducing FPS gameplay. For now, this mainly will be repair work both within the ship, and externally via EVA (hence the need for the Fox's airlock and EVA suit storage area). By time we get to Steam, you should have a much clearer idea of what you'll be doing in RogSys (at least with regards to the combat service squadron)....

    From there, we need to get the player's station up and running, as well as get the first final enemy ships--the "fighter" and carrier--put in. We have a great design for both--think you're going to like them a LOT. As soon as I can do this we'll have the very basic elements required to get the dynamic mission system up and running. From there, we start expanding combat aspects (weapons, sensor and defensive systems). Some of the freighter designs that you all have been submitting will start making an appearance after that.

    And that's where we're at. I'm off-schedule by about a month now; and working as quickly as possible to recover as much of that time as I can. eAccess sales, while low, are still steady; but we're not where we need to be to hire additional team members yet. Finally, we're gearing up and working towards our Steam early access release. By the way, if you're wondering, anyone currently in eAccess will be given a Steam key if they want one. ISI has been doing this with their rFactor1 & 2 customers and they've indicated no problems or negative ramifications from this. So, when the time comes, there will be a way for you to request a Steam key. Again, as for the current delay, all I can do is apologize. I managed it as best as I could to come back as quickly as possible; and again I appreciate your understanding and patience during this time.

    With that, I sincerely hope that everyone had a truly happy and safe holiday season. I greatly look forward to the coming year and the coming updates to Rogue System. As always, if ANYONE has ANY questions about anything I've written, please feel free to start a thread and ask away. I will be happy to answer questions as best I'm able.

    Best Regards,
    Michael
    __________________
    "There is no spoon..."

  6. #6
    Administrator
    Join Date
    Oct 2010
    Location
    Texas
    Posts
    2,119
    13 March, 2016

    Meh... Been FAR too long since I've done one of these I've really had my head buried in the work, which I HOPE is somewhat obvious based on the new 4.9 "interim test" update. If not, it wouldn't surprise me much if you felt that way--a LOT of the work was lower-level revision....

    As you hopefully know, I've been trying to get the first "final" player-flyable ship up and running--the Flying Fox (our "combat services" ship). Again, the reason I started with this one is that there is a lot of gameplay I can add with this in fairly short order that DOESN'T require the campaign and mission code to be fleshed out further. For example, using single missions we can do some deploy and recovery scenarios with the robot arm, pretty easily. It also provides a functional interior that can be used to build on some of the required First-Person gameplay, including the zero-g character controller (if time allows). So, there's some good short-term payoff here.

    Part of doing this though, was the final removal of all UI hard-coded elements and getting the new externalized frameworks up and running. This framework was something I'd always hoped to do, but until the Noesis UI was implemented I was never sure if I could. Long before that ever went in I had developed the hard-coded system (which is what the early Kickstarter prototype used, as well as the majority of the FireArc cockpit). It was okay, but didn't allow a lot of flexibility. The new system provides a way for both me and modders to link new Noesis UI schemes to ships. This actually allows ships to share the same schemes; as well as use various elements from different schemes; AND easily add new schemes or modify existing ones.

    What it all boils down to is that I AND modders will have a FAR easier time getting new ships up and running. Also, because all the UI text is externalized, this now allows a ship's control button labels and instrument text to be localized. Obviously this is good news for non-English players.

    Along with that, as always, the Fox's interior art work sucked up a large amount of time. I mean, I certainly COULD have just put in simple, non-textured geometry as placeholder, but this really doesn't get us anywhere. I doesn't provide feedback on style (which everyone so far seems to like for the most part); nor on performance. These are two things we need to learn about sooner rather than later (when there's no more time to do anything about it). Besides, it could be another few months until the next ship goes in, so it's much nicer to have something that is at least somewhat pleasing to the eye.

    The other big work that was done was to update the VMS system to use actual render-to-texture, rather than the old "looking out of a window" tech that we had. Setting up the render target wasn't a big deal, really. What WAS though was all the camera synchronization that had to be circumvented for the new system. I wanted to keep it in case we have a case of an older ship with actual windows; so that meant having to set up conditions to use one or the other. This synchronization is actually very low-level in RogSys, so I had to proceed very methodically. For the most part, I think it's all working now (although some people are reporting TrackIR issues that I'll need to look at--I'm not surprised honestly); but there's still some clean-up left to do.

    I am somewhat surprised at both a visual quality loss, as well as some performance issues, in the viewport image. Well, not so much performance--I'm using a wider FOV for the view now, and with those awfully unoptimized planetary shaders the lower FPS makes sense. I REALLY need to work on those shaders (someone had volunteered to help; but... oh well... ). The blurriness of the view does surprise though (as if the render texture is too low-res). This REALLY shouldn't be the case; so I need to investigate further. I'm not too concerned about it yet--it's PROBABLY nothing more than an improper setting when I define the render texture buffer. I just need some time to dig in and see what's what....

    There was a LOT of other work done, too, across the board, from ship system functionality revisions to a wide range of bug fixes, based on the great feedback on the FireArc. These are not ALL done yet, either. One really "cool" update was with MFD control. Before, if you remember, we had to click into a MFD in order to control it. Now, all you have to do is look at it and the MFD caret tracks to where you look--it's seamless. Some people have already commented that the MFD caret is probably pointless now and in some ways I agree. Once all the MFD functionality is back in and I verify that everything is working as expected I will PROBABLY remove the caret (or at least make an option to do so).

    Personally, things have gotten better for the most part. You may recall from the last entry that I was working through a potential burn out. With the corrective actions I took I was able to halt this and start to mend. While I'm still somewhat easy to distraction (a symptom), I'm MUCH better than I was; and feeling better week to week. I'm now putting in proper hours again and feeling strong. I will say my wife and I had a good scare not long after the new year started.

    Angel and I were taking our daughter back to college after the winter break. One the way home, at about 2 AM, just outside of Colorado, we hit a bull elk (you may have seen the pictures I posted on twitter). He was standing just past the crest of a hill right in the middle of the highway. I Didn't see him until I was right on top of him--absolutely nothing I could do. The Pathfinder was a total loss; but happily we walked away with only a few bruises. I still feel pretty bad about the elk though, honestly. I'm pretty soft-hearted when it comes to wildlife, truth be told. Long story short though, dealing with all of the ramifications of that (back and forth with the insurance company, car rentals, finding a new vehicle, etc., etc.) was a bit time consuming. All sorted now--we're back in a Pathfinder. I never had a day's trouble with the old one, and it saved our lives--seem to be good enough reasons to get another one, in my opinion

    Lastly, I should talk about what's upcoming. Obviously, I need to finish the instruments, MFD, and VMS HUD for the Fox. That will let me get it working with all the old tutorials (need to update the scripts accordingly); and that will let us move to a new current build (which will be update 5.0). From there we push forward with Steam integration. I already have a version of RogSys running on Steam, and everything LOOKS like it's working fine. Before we release there though we need some more gameplay in so we get off to a strong start. I AM excited by the fact that updating will be SO much easier; and, as pointed out, CAN happen more often if completed work warrants (or we need to hot-fix, etc). Once on Steam we start working on all the other assets needed for the campaign--mainly ships.

    I DO expect that once on Steam we will no longer sell RogSys using ISI's system, which means the current early access period will end. At this point we'll be in "Alpha" (out of prototype, but not all features implemented yet). Of course, Steam still considers that "early-access". I've already talked with Tim about making sure we give everyone a month's notice so they can update their current package if they want (this is mainly for the "Early-Access" customers who may want to move up to the "Core Module" package). I expect to have the first Steam release candidate ready in about three weeks; so you can probably expect that one-month countdown to begin around April 1st.

    That said, I have every intention of continuing to make periodic non-Steam builds for those who wish to continue to use that version. ISI seems to be agreeable to that, so I don't see any issues preventing me from doing so...

    Otherwise, that's about it for now. I think I should be able to get back to a regular update cycle now that all this low-level, tedious work has been finished. So, I fully expect to release a new update on time, and get back to regular developer entries again. As I've already said many times, I TRULY appreciate your patience and understanding over that past three months--letting me keep my head down and getting this work knocked out. This is a great community, and I'm very happy to have you all along for the ride!
    __________________
    "There is no spoon..."

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •