Post Reply 
 
Thread Rating:
  • 1 Votes - 5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
DEVELOPMENT ROADMAP
2016-07-22, 02:28 PM (This post was last modified: 2016-07-22 02:28 PM by Gladyon.)
Post: #31
RE: DEVELOPMENT ROADMAP
(2016-07-22 01:55 PM)hellatze Wrote:  3.didnt try empryon. but the graphic and enginer are weak. that why the game smooth.
Yes, but they don't stutter, and stuttering is usually a different problem than lag.

(2016-07-22 01:55 PM)hellatze Wrote:  4. Starmade are actually block only. in FTD we have detailed block. but starmade are boxy. all of it. so the polygon are small. yes its complex, but its takes on space. didn't need to render water doesn't it?
But they are speaking of beginning to have problems wiht 10 000 000 blocks, so that's about 500x FtD.


I'm not asking for millions blocks in FtD, well, it would be great, but 200 000 with reasonable speed and no stuttering would be enough to give another dimension to the gameplay.
We could have battles between several large ships and a score of small ones.
Fleets would have entierely another meaning in FtD, and that would add greatly to gameplay.

I know that the calculations are intensive, I have optimized quite a few softwares in my job and in my own devs.
But there are nearly always ways to improve greatly.
I have a 35 000 blocks ship that consume about 9% in feeding intakes, that doesn't feel right from a dev point of view, it should be faster.
Find all posts by this user
Quote this message in a reply
2016-07-22, 02:58 PM
Post: #32
RE: DEVELOPMENT ROADMAP
(2016-07-22 02:28 PM)Gladyon Wrote:  
(2016-07-22 01:55 PM)hellatze Wrote:  3.didnt try empryon. but the graphic and enginer are weak. that why the game smooth.
Yes, but they don't stutter, and stuttering is usually a different problem than lag.

(2016-07-22 01:55 PM)hellatze Wrote:  4. Starmade are actually block only. in FTD we have detailed block. but starmade are boxy. all of it. so the polygon are small. yes its complex, but its takes on space. didn't need to render water doesn't it?
But they are speaking of beginning to have problems wiht 10 000 000 blocks, so that's about 500x FtD.


I'm not asking for millions blocks in FtD, well, it would be great, but 200 000 with reasonable speed and no stuttering would be enough to give another dimension to the gameplay.
We could have battles between several large ships and a score of small ones.
Fleets would have entierely another meaning in FtD, and that would add greatly to gameplay.

I know that the calculations are intensive, I have optimized quite a few softwares in my job and in my own devs.
But there are nearly always ways to improve greatly.
I have a 35 000 blocks ship that consume about 9% in feeding intakes, that doesn't feel right from a dev point of view, it should be faster.

3. Of course they didnt stuter. They use a weak engine.

4. Of course it could support a lot of blocks. They requre small polygond and didnt calculate drag. Waters. Etc. While FTD need a load calculation with heavy complexity. This game actually beta stage

And this game still in ALPHA stage. Also you dont know how many code that work just 1 cram cannon? No mention buoyancy. Laser. Ai etc.

Give him time.
Find all posts by this user
Quote this message in a reply
2016-07-22, 03:02 PM
Post: #33
RE: DEVELOPMENT ROADMAP
starmade has vastly different weapon systems and stuff in terms of complexity. In FtD a cannon consists of 12+ different parts.
In starmade a cannon consists of 2-7 different parts (colored bullets included) which are randomly placed and just add up for more damage. Also, in starmade a physical connection isn't required to hold two parts of a ship together, reducing physic calculations to almost nothing.

But the only thing that is cost intensive in starmade are ship collisions, even though collision damage is normally switched off.

It's also important to note that Starmade uses his own engine, that was purely created for and alongside starmade, and thus is highly optimized.

Once in a while I decide to do something stupid and try to perfect it. That is how some of my strongest designs were created.
Find all posts by this user
Quote this message in a reply
2016-07-22, 05:56 PM
Post: #34
RE: DEVELOPMENT ROADMAP
you can't really compare FtD to other games.
There's a lot more calculation going on than in Empyrion or in space engineers.

Furthermore you can't compare it block to block. Place hundreds of shields and your Fps will go down. No matter how many blocks you have. Have 5 projectile weapon or 30 Missles fires and your framerate will drop, doesn't matter how many blocks you have.

There's a lot going on in this game.

I don't know why I still use my nickname, which no one can read but germans

From the Depths is like life: You'll never stop learning.
Find all posts by this user
Quote this message in a reply
2016-07-22, 06:11 PM
Post: #35
RE: DEVELOPMENT ROADMAP
Its all about RAM.
Find all posts by this user
Quote this message in a reply
2016-07-22, 06:22 PM
Post: #36
RE: DEVELOPMENT ROADMAP
Nope, I have got
16 Gb Ram,
i5-6600k,
gtx970
and its still not running fluently even at when there are only a few ships.

It would make a huge different when the game uses all the CPU power, not only a little bit.

Le Sir's gaming YouTube channel
Find all posts by this user
Quote this message in a reply
2016-07-22, 06:43 PM (This post was last modified: 2016-07-22 07:11 PM by Njord.)
Post: #37
RE: DEVELOPMENT ROADMAP
(2016-07-22 12:38 PM)Nick Smart Wrote:  out of interest how does it compare optimisation wise to other games on the market?

Do they offer 10s of thousands of blocks at smooth frame rates?

Hell of a lot better I only have experience with one game like it (starmade) as I moved over to this game when the Redshift server went down. I played that game for 4 years.

Whereas in starmade the development was slow and when he did bring out something new it was almost always broken, and even when he didn't break the new feature he almost assuredly broke something (or multiple things which was usually the case) completely unrelated to the update altogether. I could go on for the next couple hours telling you of schemas farces. But long story short he refused getting development help for the longest time and when he finally did get help. It was from two guys whose only coding experience was in minecraft modding.

Whereas... with from the depths the scale of ships that can be handled is much smaller but that is within reason as each block does a lot more than just serve as Armour. (buoyancy weight etc) The game is a hell of a lot better optimized then starmade, and the development is a lot more transparent than schemas approach. Not to mention, I haven't seen you break a single thing in any of the branches.

Overall I would have to say your doing an outstanding job compared to starmades development considering you both are pretty much in the same boat, In terms of how you started and team size, funding etc.

Edit:
Just noticed some one made a comment on starmade

(2016-07-22 01:33 PM)Gladyon Wrote:  StarMade:
Don't have it, can't say for sure.
From what I hear, it's very similar to FtD in terms of complexity (except perhpas than FtD is younger, so less complex).
On the performance side I read bout 100k-500k blocks as standard numbers
someone complained that he got 25fps only and a minute lag at loading when loading his 10 000 000 blocks ship, and some other said that a million block is where you can be comfortable with.
https://steamcommunity.com/app/244770/di...324314415/


I believe Starmade is the closest to FtD in terms of complexity & gameplay, I'm still considering buying it but I have no time to play it due to FtD...
Does someone here owns Starmade and can compare performances?

Yep played it for four a couple years nonstop.
Standard block size on Redshift (most popular server for about two or three years) was about 3 to 4 million blocks. The second Best server elewyn eternity had a standard ship size of over 10 million blocks. But this was not for the better as entering a sector with another player in it could easily crash his game.

In terms of complexity the game is not that complex.. the only "complex" part of starmade would have to be crafting system, and if you compare that crafting system with From the depths.. well its a lot more complex. And a hell of a lot of people love it.

In terms of gameplay there really is not much to do in single player. making it where most of the gameplay is structured around multiplayer. but most of the player interactions would be the same in the two games If FTD had a multiplayer that was server based (more like starmades or minecrafts)


But overall I would have to agree with you that Starmade is a lot like FTD. Although Schema has made several Fatal Errors in the Coding that make the game a hell of a lot less efficient. Whereas FTD runs very smoothly and efficiently.

(2016-07-22 03:02 PM)CubeMaster_1 Wrote:  Also, in starmade a physical connection isn't required to hold two parts of a ship together, reducing physic calculations to almost nothing.

It's also important to note that Starmade uses his own engine, that was purely created for and alongside starmade, and thus is highly optimized.

Funny that you would mention physical connection..

Also. Highly optimized my ass Tongue There was a big push a couple years ago by the player base to have hull beams Like we do in FTD to help increase FPS, ship size etc only to find out the game engine makes it impossible to put that in without having to pretty much rebuild the game.

Do you not know, my son, with how little wisdom the world is governed?
~Axel Oxenstierna 1648
Find all posts by this user
Quote this message in a reply
2016-07-22, 11:45 PM
Post: #38
RE: DEVELOPMENT ROADMAP
(2016-07-22 11:15 AM)Nick Smart Wrote:  Optimisation is done as we go along.

It's running faster now than it ever has done thanks to the LOD system.

I assume you are aware that (apparently) in (at least) Unity 5 LINQ and foreach loops can produce garbage, if they are in hot code paths the GC tends to feel it's needed right now, even when not explicitly triggered by the game. Of course that will then happen at the worst moments ... and cause stutter.
Find all posts by this user
Quote this message in a reply
2016-07-23, 12:02 AM
Post: #39
RE: DEVELOPMENT ROADMAP
(2016-07-22 11:45 PM)Antonin Dvorak Wrote:  
(2016-07-22 11:15 AM)Nick Smart Wrote:  Optimisation is done as we go along.

It's running faster now than it ever has done thanks to the LOD system.

I assume you are aware that (apparently) in (at least) Unity 5 LINQ and foreach loops can produce garbage, if they are in hot code paths the GC tends to feel it's needed right now, even when not explicitly triggered by the game. Of course that will then happen at the worst moments ... and cause stutter.

Some foreach loops works without garbage, it depends on how the iterator is developped. Arrays cause no trouble but lists are bad (except if you re-program the iterator of course).
About LINQ, you're right, it's evil (for performance, for dev speed it's fabulous).

There are a lot of things in Unity that produces garbage, and it's not always because of Unity.
For example, when you transform a float or an integer to a string, you create garbage, it's .NET, not Unity that does that.
It can be avoided but you need to re-program the conversion.
Even giving a string to Unity creates garbage, because a copy is done. You can avoid that but it's even more un-natural and few people are even aware of how to do it.

But there's worse (and on the Unity side this time), when you modify a mesh at runtime you create garbage. To my knowledge it's unavoidable, the only solution is to reduce the dynamic mesh modification/creation at the strict minimum.
And I think that FtD may do that quite often (probably each time a block is removed/destroyed/added).

There are so many things that we cannot list all of them, but Unity 5 profiler shows the garbage creation for each each method, so it helps, but I don't know if Unity 4 profiler works the same.
Find all posts by this user
Quote this message in a reply
2016-07-23, 04:33 AM
Post: #40
RE: DEVELOPMENT ROADMAP
(2016-07-22 01:22 AM)Nick Smart Wrote:  Starting 22nd July 2016.
1) Polish and finish all detection related code and the single resource system. Stable release. Update aim point selection logic to only pick detectable and visible aim points. Add passive sonar and acoustic sensors.
2) Several weeks of general polish, balancing leading to a stable release that should be very stable.
3) [5th Aug?] [3 weeks?] Begin work on strategic AI and campaign mechanic changes to bring campaigns alive and add challenge. Stable release.
4)[September][1 week] Add new content to adventure mode to bring it alive. Stable release.
5)[September][1 week] Finish plasma weaponry. Stable release.
6)[September][1 week] Finish steam engines. Stable release.
7)[September][1 week] Add optional tech system for making campaign more interesting. Stable release.
8)[October][1 month]Revisit multiplayer. Stable release.
9)[November] add fire.

In parallel to these activities we will aim to achieve:
1) All story missions brought up to scratch to tell the story of Neter
2) New geometries for structural blocks
3) New tutorial content for all new features
4) New multiplayer maps and modes and old ones refreshed
5) Responding to small requests, balance changes
6) Adding any appropriate and finished mods into the vanilla game
>Browsing forums casually
>Reading this
>General POlish? Ooo sounds good, i've been wanting this
>Campaign AI Hype!
>Plasma? Awesome, Plasma Bomber is a go.
>Steam? cool.
>"November: add fire"
Oh Nick i love you, adding Fire on my Birthday Month~!
Tho when is the Dediblade/AirPumps/Antigrav thing happening pl0x

Centralised Communities Front Directorate
Kiso best Light Cruiser. Nagato Best BB.
3100 Hours Ingame and Counting
Find all posts by this user
Quote this message in a reply
Post Reply 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)