Welcome! Look forward to seeing your guy's projects! :)
-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
-
We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.
Indie Game Development |OT 2018| Come, help, learn, show and tell!
- Thread starter Popstar
- Start date
- Status
Welcome! Look forward to seeing your guy's projects! :)
Thanks! Can't wait to show both upcoming personal and professional projects. <3
Again, I can't speak about Mercurial or Bazaar, as I haven't used them, but Git is strictly better than SVN; as in, Git's features are a superset of SVN's, and what it does above SVN is frankly essential to me now (offline work, local repos and branches, distributed repos, the works). I don't claim that Git is better than every single other option out there, but it's better than SVN, which is much better than CVS, which is better than SourceSafe. I couldn't care less about popularity; I'm strictly taking about feature set and workflow.
To be fair, I didn't get all the fuss about Git when I worked with Subversion (professionally, for nearly a decade), but Git made a convert out of me in a few months.
OP
OP
Git is not strictly better than Subversion. The thing I need most from a version control system is that if I commit something into it, it should be there forever barring some sort of hard drive crash or an explicit obliterate-style command. This is "table stakes" for version control.
Git does not guarantee this. You can commit stuff into Git and have it disappear. It's a consequence of its design. Stuff can get orphaned from the internal graph and then garbage collected. Gone forever. All the things you mention are just lipstick on a pig when it has a central flaw like that.
EDIT: for people wondering what I'm talking about -> How to use git to lose data
Git does not guarantee this. You can commit stuff into Git and have it disappear. It's a consequence of its design. Stuff can get orphaned from the internal graph and then garbage collected. Gone forever. All the things you mention are just lipstick on a pig when it has a central flaw like that.
EDIT: for people wondering what I'm talking about -> How to use git to lose data
Last edited:
Well, it's a tool, and if you don't use it correctly, it might not work as intended. :P If you don't push a local branch to the repo, of course the data will be gone. If you have files in your .gitignore, they will be gone too. Your stash will be gone, etc. This is all intended behaviour. Don't keep committing on a local branch for days without pushing. Don't add files to your .gitignore if you want them to be secure in the repo (it should only be auto-generated files or per-user configs). Use the stash only for temporarily backing up small changes that you don't want to commit yet, so you can quickly switch to another branch, or try a different solution on your current one.
And use git push -f only if you know what you're doing. I don't have much experience with SVN (used it for about a year ten years ago, and only Git since then), but I'm sure there are also ways to lose data in an SVN repo if you mess up.
I can recommend using a tool like SmartGit, so you can have a much better overview of what's in the remote repository vs. what's only local.
OP
OP
Spare me the straw man arguments and condescension. I'm talking about the fact you can push data into a repo, only for it to vanish from the repo you pushed it to.
You're sure, yet I doubt you can give any examples.
I knew was shaking a hornets nest to dare say something negative about Git, but like you said - it's a tool. And like any tool it has it's tradeoffs. Best to be aware of them.
The article is talking about a different thing that you are. I will address both.
1) Orphaned nodes are a non-issue. There is no reason at all to orphan work that is relevant. You pretty much have to actively and maliciously have to do this yourself, by deleting a local branch without merging it (which promts warnings), or something similar. And even then, Git gives you a month to realize your mistake and fetch the data from the reflog. And further, this would only affect your own repo, unless you actively push with --force or similar. At this point you might as well complain that if you delete all copies of your repository, your work is gone (in SVN, all it would take is to lose the central remote repository to lose the history).
2) The article talks about stuff that isn't pushed to other repositories, like local branches, and will be lost in the case of local loss of data (hard drive crash, etc.). This is something to be mindful of, that is, that branches aren't automatically remotely tracked and you need to make them so if you want them to behave like SVN branches. The uptake, again, is that Git provides more functionality than SVN, which only has remote branches. This allows for a much cleaner history and remote repos, where looking for relevant changes or performing bisect (another wonderful functionality that Git has over SVN) is far more straightforward.
Again, Git is a functionality superset of SVN. The entire point of being a superset is that it allows you to do more things than SVN does; this includes deliberate misuse, in the same way that SVN allows misuse. For example, under SVN, if you don' rename or move files from within SVN, it's not smart enough to know it's the same file, so it will unlink the history of the file (and once it's unlinked, you're screwed, there's no way to relink it that I know of). Git automatically knows what files are renamed and moved, so this isn't a possiblity.
In SVN, this is indeed a problem, because the remote repo is the only copy of the history.
In Git it's a non-issue, because 1) it's still there in the reflog, and 2) you still have a copy of the work, alongside the entire history, in your repo, and you will immediately see something is wrong when you try to push again and it doesn't let you (at this point you would hopefully talk with the repo admins and ask why did they roll back your work). The only way to lose it is for all of these to happen at once: 1) have someone roll your work back remotely (itself something that shouldn't happen at all), 2) don't push work on it for a month, 3) lose your computer or hard drive. The actual chance for this theorethical scenario to happen is virtually zero.
Your misgivings are very understandable if you're used to working with SVN, because Git is a pretty big paradigm change that I tried for years to wrap my head around (and indeed the possibility of "losing story" was anathema to me), and only succeeded when I actually started using it for a year. I'm not trying to say this in a condescending manner, you're where I was a year and change ago; I used to sing SVN's praises and be very skeptical of Git (more than anything, I didn't see "that much of a difference", but the difference indeed is pretty huge).
But we're straying far off topic and I didn't realize this would be such a controversial topic. For the average user Git is better than SVN simply because BitBucket exists; that's reason enough.
OP
OP
Really not much more point in arguing with you if you've already decided to dismiss anything I might say because of your belief that you have greater experience. I'm not trying to say this in a condescending manner, but perhaps a discussion of Git's strengths and weaknesses will have to wait until you moved on from singing it's praises and can be a bit more objective, the way you say you moved on from singing SVN's.
I don't think any of the above is warranted, and I don't know why you're making it personal. I haven't dismissed anything you said; I replied with a step by step explanation why it's not an issue in the 99.999% of cases. What exactly did you expect me to do if I disagree with you, smile politely and nod? If anything, it's you who has dismissed my explanation as me saying "well you think you know more than me but you don't". If you're really interested in discussing the merits of each, feel free to point out where you think I'm wrong in my argument, otherwise we'll just have to agree to disagree.
Last edited:
To wrap up my weekend engine work, I added event markers to my renderer. These markers group DirectX API calls into sections which allow easier reading in tools like RenderDoc and PIX.
In the RenderDoc profile below, you can see all of the major rendering passes in my engine (RENDER_G_BUFFER, RENDER_POST_PROCESS, etc)
It's not glamorous work, but it makes debugging so much easier.
Also, Unity has built-in support for RenderDoc profiles.
Edit:
Specified that the screenshot is from RenderDoc.
In the RenderDoc profile below, you can see all of the major rendering passes in my engine (RENDER_G_BUFFER, RENDER_POST_PROCESS, etc)
It's not glamorous work, but it makes debugging so much easier.
Also, Unity has built-in support for RenderDoc profiles.
Edit:
Specified that the screenshot is from RenderDoc.
Last edited:
I have a DK2! Unfortunately, it's not setup so I'll have to get to that at some point. But I think that your art style is well suited for VR. Everything is clear and readable which I imagine helps with visual comfort.
Anyways, good job so far! This is a lot of fun :)
My main project, a 3D collectathon platformer, has a rough half year behind it, because of various life challenges (phd, new job, relocating, birth of my son with a subsequent stay in hospital for three months) that ate away almost all of my free time, so progress slowed down to a crawl. This extended weekend I could finally find some time outside the regular fortnitely meetings with my friend who does the level design with me to work on the game. I managed to design two of the 100 main tasks, which means the level I am currently working on (which is a late-game level) is at 7/10 tasks done now. I also managed to find two instances of unoptimised code that I could still optimise. I am always happy if I can find some code improvements without being currently hard pressed for it, because reaching the performance goal of 60 fps on Wii U continues to be challenging with every new level (near) completion.
So, while there is still a lot to be done for the game (four full levels, three tasks in the current level, three tasks in the overworld, which my friend is working on currently), finally getting the chance to work on the game again outside of the regular, slow schedule of meetings, has given me some enthusiasm and regained hope that we may still manage to finish the game by the end of the year. It will be interesting though, whether we can go much beyond our current bland simply textured looks without harming framerate - which we would not accept. Visual optimisation (as much as a computer scientist and a mathematician without any education on arts can do, anyway) is something we postponed as the final step before beta testing so that we can ensure to keep the target performance after each visual enhancement. The exception are visual enhancements that are crucial to gameplay (e.g. there are some places in the game where gratings are required as textures, so I made them already).
So, while there is still a lot to be done for the game (four full levels, three tasks in the current level, three tasks in the overworld, which my friend is working on currently), finally getting the chance to work on the game again outside of the regular, slow schedule of meetings, has given me some enthusiasm and regained hope that we may still manage to finish the game by the end of the year. It will be interesting though, whether we can go much beyond our current bland simply textured looks without harming framerate - which we would not accept. Visual optimisation (as much as a computer scientist and a mathematician without any education on arts can do, anyway) is something we postponed as the final step before beta testing so that we can ensure to keep the target performance after each visual enhancement. The exception are visual enhancements that are crucial to gameplay (e.g. there are some places in the game where gratings are required as textures, so I made them already).
Sounds like a cool project any screenshots to share?
Wii U? Is there sales potential left on that platform?
You shal share a screenshot of your project as well.
The gif you posted back then was really cool.
Anyway question:
Can you remind me the name of the game which the Dev showed here and it was a punch-out like game with students?
Sorry, really wasn't my intention to sound condescending. As I said I don't have much experience with SVN but I do have much with Git, so I can only elaborate on what you (and that article's author) wrote about Git and why it really is a non-issue when you actually use Git and not try to lose relevant data on purpose.
The following screenshot shows one task from the factory level I am currently working on, though this specific task is the last one I finished before this weekend:
You see that the game uses a very simple style, with large single coloured surfaces. The blocky geometry will also remain in the final game, this is a conscious design decision (it reduces computational cost on collision detection and communicates clearly with the player).
Regarding Wii U: I suppose there is not much sales potential left, but it is not exactly easy to get on other console's Indie programmes with a game that looks like this. Also, I have invested a lot of work to make a 3D Unity game with reasonably large levels work at consistent 60fps on Wii U, so even if I manage to get the game on Switch or Xbox One, the Wii U version will still release. Since we are making basically our dream game here, we would still release it even if we were guaranteed no one will ever buy it.
One day.. actually I have not implemented backgrounds yet so it is just a enemies and blue screen for background so far.
As soon as there is more I will let you know. I really like your thread and all the amazing games you have found!
The following screenshot shows one task from the factory level I am currently working on, though this specific task is the last one I finished before this weekend:
You see that the game uses a very simple style, with large single coloured surfaces. The blocky geometry will also remain in the final game, this is a conscious design decision (it reduces computational cost on collision detection and communicates clearly with the player).
Regarding Wii U: I suppose there is not much sales potential left, but it is not exactly easy to get on other console's Indie programmes with a game that looks like this. Also, I have invested a lot of work to make a 3D Unity game with reasonably large levels work at consistent 60fps on Wii U, so even if I manage to get the game on Switch or Xbox One, the Wii U version will still release. Since we are making basically our dream game here, we would still release it even if we were guaranteed no one will ever buy it.
Looks very cool. 3D platformers are not a saturated market so my guess would be that some of the big guys would be interested to release on their current console.
Last edited:
Now that I learned what really is Sreenshot Sunday is, do you know some good places which I can check them daily?
Thank you!!! That is the one I ment!!!
Oh you. You make me blushing.
Thank you!!! That is the one I ment!!!
Oh you. You make me blushing.
A big weekend for Carovita. My RPG finally has sound effects! The first bits of sound are being played on attacks in combat, and I'm finishing up the engine to handle sounds dynamically. It's another huge step for the game coming to life.
Thanks, very kind words. I'd be glad if Microsoft or Nintendo were to decide it can go to either platform, but luckily, as it is a complete hobby project (and cost of dev kits is chalked up for me as paying for my hobby), I am not dependent on this.
By the way, there exists a gif of your project? What is it?
This looks great, do you have a professional art background?
No.. but... I have a professional artist working on the project :-P.
https://wolfenoctis.deviantart.com/
(I do programming and enemy animations)
"... Metal Gear?!".
If I did that I'd lose the physics-based bobbing of the boat, and the sinking effect when it's destroyed. I could, of course, code these myself, but I would sooner simply fix the boat into place horizontally, since it doesn't move (and even if I did, I could force a constant speed).
I don't use AddForce anywhere indeed, I either change the speed (over time if needed be) or in extreme (and rare) cases, the position.
On it!
Edit: My best:
My criticism:
+ Pretty relaxing, paradoxically enough.
+ Love the different random events, making it more varied on successive runs (train going by, crashing into a mart).
+ For such a short, simple loop, the music didn't get maddening.
- Almost every time I crashed, it didn't feel like I had done anything differently than in the other dozens of jumps. In particular, jumping slightly too early feels more unforgiving than it should be. Of about a dozen "deaths", in at least five I didn't know what I had done differently, and one was due to the "color change" key firing twice in a single press (maybe add a slight, fraction of second cooldown to color change). Of the ones that felt like legitimate mess ups, about half were due to not changing the color to match the block, and the other seemed like legitimate misjumps.
- If the game ramps up in difficulty, it takes far too long, especially considering you start from the beginning every time. You basically randomly die from the above issues, not because the game is actually giving you a challenge. I think you should throw sequences of blocks at the players not too far in, to make it more interesting.
- The people in the background are translucent (you can see the background behind them). They basically seem like ghosts, which is strange considering the otherwise mundane setting.
- The blocky style is nice, but it feels too "high res" if that makes any sense. It might be personal taste but I feel a blocky style like this works best when done in lower resolution. Some of the blocks here are far too small to "sell" a retro aesthetic. For a game with blocky look that I love, check out Super Time Force Ultra:
https://www.youtube.com/watch?v=Kyxrll5Bc-Q
It uses blocks and boxy shapes a lot, but it sells it very well because it has a relatively big pixel size. Smaller pixel sizes coupled with simple shapes make a game feel "cheaper", like a Flash game. Or at least that's my opinion.
Thanks so much for checking out the game, I really appreciate the feedback. Sorry for the late response, I was at BitSummit in Japan over the weekend.
I really appreciate you giving clear and detailed feedback, it really helps a lot and I am also glad that you generally enjoyed it! The jumping is something that has had lots of praise and also lots of negative feedback. It was hard to strike a balance between it feeling good and getting into your muscle memory, but also to make it feel like when you were doing well it made it clear you could feel a misjump. I could have done stuff visually to tell you about timing like a "NICE" or "BAD", kinda like a rhythm game, but I didn't feel like it was going to fit. It's definitely something I would work on next time.
In regards to the translucent NPCs, I changed the opacity of the background NPCs as I wanted to convey a feeling of everyone just blending together and into the background. In Japan there are just so many people and everyone is minding their own business, that everyone just sorta becomes a blurry background. I felt like making them seem a little transparent would give that sense of being surrounded by many, but not really paying attention :)
Thanks for playing, I am glad someone got a kick out of it.
Also your score was pretty damn good! A push towards 5000m and you would see the next season :)
Sure thing! And don't worry, many people here have busy schedules and can't reply right away.
How was BitSummit? If you want to describe your experience, I feel it would be relevant for this thread. :)
Actually I think that could be great! I like the idea of this addition. When you say it wouldn't fit, you mean thematically?
Ironically it was the opposite for me. :D I was constantly distracted by the fact they were transparent and I could see the background behind them.
If you want to convey a sense of "blurring into the background", I recommend using single-color silhouettes (gray, typically), and then using multiples of them so that they literally merge together. I did that with the farthest buildings in my own game and it looks pretty good.
That was my best run, but I admit the one thing that prevented me from playing more is that it feels somewhat random whether I'll get to 4000 or 400. With no feedback on how good I'm doing, I don't know that my timing is off until I die. I think adding a "perfect!" "good" "bad" "miss" indicator would help a lot. As would, perhaps, giving the player three "lives" or chances to miss; this would also reduce the feeling of randomness simply by the law of averages.
had an off day because of a cold so after i felt a bit better decided to try and implement an 'enemy leaping at player' state.
i failed.
but i got enemy projectiles working and i think i can probably adapt the logic, so maybe i'll give it another go in a few days
i failed.
but i got enemy projectiles working and i think i can probably adapt the logic, so maybe i'll give it another go in a few days
I mentioned a page ago that I finished up some work on my generic exception handling. It's really nothing special, but I thought I would share it. If anyone is using unity you can add this script to your game manager or some other singleton and it should capture most unhandled exceptions. It takes a screen shot, attaches it to the email, and puts the stack in the body.
You'll have to use your own smtp of course, so make sure to update that part.
You'll have to use your own smtp of course, so make sure to update that part.
Code:
using UnityEngine;
using System.Net;
using System.Net.Mail;
using System.Net.Security;
using System.Security.Cryptography.X509Certificates;
public class ExceptionManager : MonoBehaviour
{
private int emailsAttempted;
void Awake()
{
Application.logMessageReceived += HandleException;
DontDestroyOnLoad(gameObject);
emailsAttempted = 0;
}
void HandleException(string condition, string stackTrace, LogType type)
{
if (type == LogType.Exception)
{
Debug.Log(stackTrace);
if(emailsAttempted <= 3)
SendEmail (condition, stackTrace);
}
}
void SendEmail(string condition, string stack)
{
try
{
MailMessage mail = new MailMessage ();
mail.From = new MailAddress("FROM@??.com");
mail.To.Add("TO@??.com");
mail.Subject = "Carovita Crash Email";
mail.Body = condition + "\n\n" + stack;
string crashImagePath = Application.persistentDataPath + System.IO.Path.DirectorySeparatorChar + "CrashCapture.png";
ScreenCapture.CaptureScreenshot (crashImagePath);
mail.Attachments.Add(new Attachment(crashImagePath));
SmtpClient smtpServer = new SmtpClient("YOUR_SMTP");
smtpServer.Port = 587;
smtpServer.Credentials = new System.Net.NetworkCredential("LOGIN", "PASSWORD") as ICredentialsByHost;
smtpServer.EnableSsl = true;
ServicePointManager.ServerCertificateValidationCallback =
delegate(object s, X509Certificate certificate, X509Chain chain, SslPolicyErrors sslPolicyErrors)
{ return true; };
smtpServer.Send(mail);
emailsAttempted++;
}
catch(System.Exception ex)
{
Debug.Log("Could not send email: " + ex.Message);
Debug.Log("Original Error Stack: " + stack);
}
}
}
I've been checking this thread out off and on but I haven't posted much at all on here. I'm working with a friend to make The Iron Oath. We ran a successful kickstarter last year. Here is a trailer if anyone is interested.
My main reason for posting though, I was curious if anyone was going to DreamHack in Austin in a few weeks. We entered into their contest thing they ran to get a free booth at the event so we'll be there. Was wondering if I'd see anyone from here at the event.
My main reason for posting though, I was curious if anyone was going to DreamHack in Austin in a few weeks. We entered into their contest thing they ran to get a free booth at the event so we'll be there. Was wondering if I'd see anyone from here at the event.
Way late here as I'm incredibly lazy when I'm at home. I'm just following these instructions now and even after following yours and Unity's I'm still confused and can't figure out why the project still isn't opening in Unity.
I hadn't heard about your game, congrats on the successful Kickstarter, game looks pretty good!
Was checking up to see if you did any postmortems on the Kickstarter and seems like you did, I always appreciate when devs write articles detailing what went right and wrong. I'll link it here in case someone is interested, it's a good read: A Kickstarter Postmortem for The Iron Oath
I have one real task left to do on Jack B. Nimble, but I'm out of energy again :(
Shipping a game that has been "in my life" for so long is so fucking tiring... especially when I know that not a whole lot will come from it (other than achievement).
In other news I have just agreed to work on something with a friend... it's exciting to finally collaborate on something, so I feel that has something to do with my dwindling interest in my own game.
Shipping a game that has been "in my life" for so long is so fucking tiring... especially when I know that not a whole lot will come from it (other than achievement).
In other news I have just agreed to work on something with a friend... it's exciting to finally collaborate on something, so I feel that has something to do with my dwindling interest in my own game.
Wow, totally missed your post. Thank your very much for the feedback! I'm really happy to hear that :) I'll upscale them and correct the error. Good job catching that reference "Weltall Zero" :D
btw. I made two new areas. A flooded forest and the interior of a lab/factory/airship/future stuff :D
OK, can you post exactly what's in the project folder? The easiest way is to:
1) Hold shift and right click the folder in Windows Explorer, then select "open command window here".
2) In the command window, type "dir /a > Contents.txt" without the quotes.
3) Open the newly created file Contents.txt with e.g. notepad, and copy-paste its contents here (ideally within quote tags so as to not spam the thread).
Well, it's the first and most visible text in the back of the game's case, and that game was my favorite ever for many years. :)
Holy shit, man. I don't know how you can work that fast. It makes me a bit self-conscious, actually. :D
This is amazing stuff. I think your game has the potential to make big waves! It's awesome that people like you and Zeboyd games are picking up Square's slack and making great games filling that huge void left by them pretending Chrono Trigger never existed. :P
Actually, I have a suggestion, but it may be a hard one to stomach: I would change your game's name to something far more unique. You need a title that when googled, takes you directly to your game. I googled Grimoire Game and this is all the first hits:
https://store.steampowered.com/app/650670/Grimoire__Heralds_of_the_Winged_Exemplar/
Contrast and compare my game, where I simply put a video on Youtube, with zero positioning or even a site, and now googling "Divinoids" takes you straight there.
Haha thank you. That's really motivating! And yeah I feel the same :D
Oh... parting with your beloved game title is very hard. That's a decision I would have to sleep over a night... or more. xD Well at least searching for "Grand Grimoire Game" lists my website and my twitter on the first page. I'm only online for about 1 1/2 months mabye I can even make it to the top in a few more month without changing the name. Hmmm...
I see that typing it with quotes does turn out your game, but doing so without quotes, even adding "game", gives a bunch of other stuff. That's... not a good sign. Ideally you should pick something that turns out very few results on google.
I feel you on how hard it is to let go of one's title, but the thing is that, even beyond the fact that it turns up so many unrelated hits, the other and perhaps even more important problem is that it's also somewhat generic and not very memorable. I was given this exact same advice about my own game's name, which was originally Mechanical Legend, and it hurt, but also felt true. It took many months, but after much soul-searching, I came up with Divinoids. It conveys the same meaning of mixing mythology with technology / scifi, but is more unique and memorable. I bounced it against several of my friends and they all liked it, so I kept it. So, yeah; nobody is expecting you to come up with an unique and memorable new name on the spot; let it stew for a few weeks or months, talk it with friends, and so on.
Speaking of which, I find that, as a lone developer myself, I have ton of blind spots and make mistakes or odd choices that are quite obvious in retrospect, so it's absolutely crucial for me to get feedback as soon and often as possible. If you haven't yet, assemble a circle of close friends and send them versions of the game for them to test.
Just wanted to drop a line and say that I love everything about your game. It's a top notch example on how to stand out. I followed the kickstarter closely, but couldn't back it. I do look forward to playing it on release though.
You can fudge the google results if you have time, especially if you game has a somewhat unique name (like yours does). Just once a day, search google for it, find your website in the results and click it. Eventually it will bump up the rankings (Though dont do it too much as google doesn't really like this)
I did it with mine, took a few weeks. Annoyingly though it has slipped to 3rd/4th result as I haven't been active with the whole internet side of things.
He's right though. While it's a cumbersome and frustrating decision to make, it would benefit your game in the long run to have a more unique name. Still plenty of time to make that call, better now than near release when you already have your hands full with a million different things and very little time to spare.
Completely forgot to reply to this. Thanks a lot for this; I don't have time right now but I'll be sure to check it out in a few hours. :)
For the lazy like myself, gmail's SMTP settings are (in theory):
Server: smtp.gmail.com
SMTP port (TLS): 587
SMTP port (SSL): 465
Thanks everyone for the feedback. :) I'll think this over and consider all the pros and cons (well actually that's what I've been doing the past hours). :D
- Status