Thundercraft
-
Posts
266 -
Joined
-
Last visited
-
Days Won
2
Content Type
Profiles
Forums
Events
Posts posted by Thundercraft
-
-
...and do you trust that new versions are not infected?
no thanks to notepadd++
If that's your only reason for not using Notepad++, then you should check out the news feed on their site. Their latest version checks the certificate validation of a certain Windows file to make sure that it has not been compromised with a fake/hack.
The way you wrote that, though, sounds like an accusation that the author of Notepad++ colluded with the agency to include a backdoor they could exploit. I doubt that is the case. At least, that's not what the Wiki/links article in question said.
Anyway, if security matters really bother you, then switching to Linux (or macOS) is probably the best thing you could do. Windows is much more vulnerable, to both agencies and malware/viruses. (I say this as someone who still uses Windows, not as a Linux fanboy.)
-
I really enjoy this mod and the changes you've made.
Q: Would it be possible to widen the variance of how much crew a module supports, based on rarity?
From what I can tell, Petty and Common modules support something like 20 or 25 crew. I've seen a few Exceptional or Exotic modules support in excess of 200 (each, engineers and mechanics). Usually, though, it's less than 200.
Sometimes, I find rarer Robotic Crew modules that do not support any more crew than less valuable versions, with the only difference being that the rare version requires a bit less energy. (Though, rarely, it might actually require more energy.)
It would be more useful to see a wider variance. By that, I mean it would be nice if we could see Petty Robotic Crew modules start at around 10 crew and using proportionally less energy. And it would be awesome if we could find Exotic and Legendary versions that support in excess of 500 crew, maybe even more than 1000.
Yes, I have seen ships that require a crew that large. Indeed, such large ships seem to be becoming more common.
On the server where I play, they're buffing all enemies to have much larger ships. And players have to build larger ships just to survive. But larger ships means requiring a much larger crew. And most players are unwilling to sacrifice 3 or more module slots just for Robotic Crew modules.
-
Question about your /inventory upgrade script:
Is there any chance - at all - that you could add support for the Extra Crew Workforce System Moduls mod?
The reason I ask is because I always design my ships in a creative-mode galaxy. But, even in creative, ships still take constant damage unless they have enough mechanics for crew. And, as I design a ship, I don't always have enough crew quarters to hold enough... a catch-22.
I can't emphasize enough how handy it would be to be able to spawn several legendary Robotic Crew modules for when I design really BIG ships. Those are rare enough, let alone trying to find several legendary modules...
-
Don't you find it frustrating to have a folder FULL of various ship designs, but absolutely no way to determine how large a ship design is, what the stats are, or how many resources it would require to build... until AFTER actually applying the design?
Just imagine the amount of time and frustration that could be saved by being able to glance at the stats of each ship design to easily determine which we can or can not currently afford (in terms of both resources and crew) to build and which have the stats we are looking for.
I honestly did not feel that this suggestion needed to be brought up because I assumed that it would be forthcoming very soon. It's just so obvious.
Indeed, I am familiar with several other games in which the stats for player-created vehicles and weapons are displayed on the load menu. I was surprised that Avorion did not do this. But I forgave this glaring omission because it is still in active development.
-
-
...added super fast sector checking routine, now searches finish in few seconds and only have little performance cost...
Could you elaborate on "only have little performance cost"? Will this mod definitely have at least a minor impact on performance? Is it easily noticeable? Or only barely noticeable?
Also: Would the performance hit be even more noticeable on a server that regularly has 10 to 24 players at any given time?
This was at least considered on the multiplayer server where I play. However, even after upgrading the server, there are lag issues. And the concern is that - because this is server-side - this mod might make existing lag/performance issues worse. (The thinking is that there is not enough "headroom", performance wise, for something like this.)
-
Update: Either NPC or player use of fighters, particularly lots and lots of fighters, seems to be a leading culprit of lag issues. (See the [Req] -Disable- all NPC fighters mod request topic for details. The admin said he would gladly disable all NPC fighters if he found a way to do this.)
It's such that the admin set a rule where players are only allowed 2 to 3 squads of fighters, max - at least when there are lots of players online.
Further, it's been noticed that when we find several sectors with ongoing faction wars, the lag becomes much worse. (Faction wars always seems to involve one side using lots of fighters.)
Anyway, it seems clear that either reducing the number of NPC fighters / carriers or optimizing the code for fighter AI behavior would go a significant way towards lessening multiplayer lag.
Perhaps, in the future, server admins could have a setting to control the maximum number of faction war battles that can occur simultaneously? Reducing that to 1 or 2 should help considerably.
-
On the server where I play, it's become pretty obvious that fighters are a leading culprit of lag issues. It's such that the admin set a rule where players are only allowed 2 to 3 squads of fighters, max, in salvage yards or sectors with lots of wrecks - at least when there are lots of players online. (Server uses the carrier commands mod to have salvage fighters.) Also, we were told to try to avoid using fighters at all if the lag becomes bad.
Further, it's been noticed that when we find several sectors with ongoing faction wars, the lag becomes much worse. (Faction wars always seems to involve one side using lots of fighters.) The admin stated, "If you come across sectors with swarms of fighters, you have my permission to destroy them." And he said he would try to buff the stats of fighters so players won't have to use as many.
Anyway, the admin would gladly disable NPC fighters if a way was found to do this. Hence, this request.
-
I had some free time, so I've added the Prototype Shapes that I've been planning to share.
Enjoy. ;D
-
Might I suggest use of the "spoiler" function to organize the designs a little bit more?
The stats and costs are now behind spoiler tags. Those were taking up far too much space and detracting from everything else.
For some reason, when I apply the either cutter blueprint the stats show zero breaking thrust yet your stats show there is breaking thrust...
Actually... .Now that I'm looking through the rest of them, none of them are showing any breaking thrust when I apply the plans... o.O
Sorry they don't work for you.
I'm playing the latest beta branch, which is currently v0.10.5 r7633. I suspect that you are not as that would explain this difference. As I explained in my OP, all of these use Directional Thrusters which, AFAIK, are still restricted to the beta branch.
Perhaps I should have made this more clear? But I did prefix my topic with [beta] and ended it with (DTU) to indicate that these are Directional Thrusters Upgraded.
-
fighters/scouts, will be very high, with extra emphasis on pitch & roll to aid evasion and minimising profile.
You emphasize Pitch and Roll in your fighters? :o Pitch, I can understand. But how do you even take advantage of Roll in combat? ???
Using Roll requires pressing the "Z" and "X" buttons. Myself, I barely find myself using strafe like "Q" and "W", but never roll.
And what about Yaw? Do you consider that important? That's left and right. Granted, one can get by with either a high Yaw or a high Pitch, but I think having high Yaw and Pitch is better. It allows much more freedom of movement.
As for minimizing profile, I think many players put too much emphasis in this. For most ship designs, I doubt it has as much impact on combat as they think it should. (See my post here in The Cube Meta about that.)
To be fair, having a small profile is an advantage. And, as some have pointed out, those really narrow "mace"-style or pole-shaped NPC ships are quite hard to get clean hits on. However, those NPC ships use a very extreme example and I just don't see players using that kind of shape. It would be too fragile.
-
(reserved)
-
Locus class multi-role (coming soon-ish™)
This one needs a thruster overhaul and rebalance to optimize for the beta branch:
Preview:
-
Azugalenos class medium freighter
I was surprised to learn that the Trilenos had better performance with three arm instead of four. As an experiment, I wanted to see if Trilen might end up with even better performance with only two arms - one horizontal and one vertical. So, I created this:
Stats and Cost for Thu_Azugalenos Ti-Tr-6 gW 54.7.xml (with shields, sans cargo):
Credits: 560341
Iron: 1259
Titanium: 50997
Trinium: 12801
Stats and Cost for Thu_Azugalenos Ti-Tr-7 gW 54.7 car-Ir 8.1 k.xml (with shields & 8100 unit Iron Cargo Bay):
Credits: 632341
Iron: 33658
Titanium: 50997
Trinium: 12801
Stats and Cost for Thu_Azugalenos Ti-Tr-7 gW 54.7 car-Ti 8.1 k.xml (with shields & 8100 unit Titanium Cargo Bay):
Credits: 657542
Iron: 1259
Titanium: 83397
Trinium: 12801
The odd-armed Azugalenos does have markedly better performance without a cargo bay. However, with the same 8100 unit Iron cargo bay, the Trilen has somewhat better stats. Interestingly, though, the Azugalenos requires nearly 5 GW less energy (and produces slightly more). Combined with how it reqires 19 less crew, this makes it ideal to use with the Extra Crew Workforce System Moduls mod. It's also has better Roll and is slightly cheaper.
Despite the slightly worse stats when hauling cargo, I've decided on the Azugalenos as my go-to freighter ship. The server where I play has the Extra Crew mod and I have the modules on hand.
(Trivia: I think Azugalenos translates to non-paired arms. Lenos is Greek for arms. And ázuga is Greek for 'non-pairs' or 'odd'.)
Thu_Azugalenos_Ti-Tr-6_gW_54.7.xml.zip
-
Trilen & Trilenos class medium freighters
As I've been playing multiplayer lately, I wondered if my use of void tech would be frowned upon as cheating. (Though, in fairness, NPC ships regularly overlap their blocks in a similar fashion.)
So, I endeavored to modify the Tetralenos to get rid of the void tech. Replacing the void tech engines was a snap. But, I had to completely remake the nacelles because of the size difference. With those void tech thrusters, the ship got five times the directional thrust as normal. Without that, I had to make the nacelles about five times larger!
Interestingly, in the process of remaking the nacelles, I discovered that the ship would actually have better performance with only three nacelles instead of four. (It was too much mass far from the center of mass.) So, I left the top one off for a better view of where the ship is going and what's in front.
Thu_Trilenos Ti-7 gW 38.2.xml:
[spoiler=Stats and Cost]
Credits: 286729
Iron: 7969
Titanium: 41657
Further experiments showed that my long thruster arms were just too long. Even down to three, they contributed too much mass. It negated the extra thruster efficiency.
So, I created a separate version with short thruster arms. This really improved performance and even saved some in cost and materials.
Then, as I felt that I might be using this a lot, I added about 30000 HP in shields. That jacked the price way up.
Thu_Trilen Ti-Tr-7 gW 53.5.xml (with shields, sans cargo):
[spoiler=Stats and Cost]
Credits: 630448
Iron: 1779
Titanium: 61935
Trinium: 12801
Thu_Trilen Ti-Tr-7 gW 53.5 cargo 8.1 k.xml (with shields and 8100 unit Iron Cargo Bay):
[spoiler=Stats and Cost]
Credits: 640700
Iron: 1779
Titanium: 87087
Trinium: 12801
The original, long-armed version I named "Trilenos" or "three-armed". And the final version with three short arms I named "Trilen" - a shorthand version of "Trilenos."
Trilen download: Thu_Trilen Ti-Tr-7 gW 53.5.zip
(Attachment below is for Trilenos.)
-
Tetralenos class medium freighter
Next, I wanted to design a rather big freighter. And I wanted to take maximum advantage of the weird way in which Cargo Bays scale. Single, big Cargo Bays are insanely more effective than many smaller ones. (E.g.: Despite being the same volume, building a single 10x10x10 Cargo Bay is 685% more efficient than building 1000 1x1x1 Cargo Bays. I'm not exaggerating. :()
So, I planned for a really big cube as my Cargo Bay and literally built a ship around it.
(Trivia: Tetralenos should translate to "four-armed". Tetra is Greek for four and Lenos is Greek for Arms.)
Thu_Tetralenos Ir-6 gW 18.7.xml (Iron version, sans cargo):
[spoiler=Stats and Cost]
Credits: 215814
Iron: 28064
Titanium: 12044
Thu_Tetralenos Ir-6 gW 18.7 cargo 8.1 k.xml (Iron version with 8100 unit Cargo Bay):
[spoiler=Stats and Cost]
Credits: 277814
Iron: 60464
Titanium: 12044
Thu_Tetralenos Ti-7 gW 19.xml (Titanium version, sans cargo):
[spoiler=Stats and Cost]
Credits: 227781
Iron: 277
Titanium: 39830
Thu_Tetralenos Ti-7 gW 19 cargo 8.1 k.xml (Titanium version with 8100 unit Iron Cargo Bay):
[spoiler=Stats and Cost]
Credits: 299782
Iron: 32677
Titanium: 39830
Do note that this is another cross design, taking advantage of Directional Thrusters on long leverage arms. The bulk of the ship's mass is going to be near the center of the huge Cargo Bay, which helps thruster arm effectiveness. Though, I loose some effectiveness due to how much mass I added around my thrusters. It would work better with less protection around the thrusters and thinner leverage arms. It would also work better if my Cargo Bay was made of anything other than Iron, or if the ship was built from Naonite or better.
Download: Thu_Tetralenos.zip
-
Monitor class mid-game multi-role
I've been using the Monitor class quite a lot. It handles well, even with a cargo container. I used what I learned from earlier designs. The thruster arms are tough. With IFG protection, it would require about 750 HP in focused damage to sheer a nacelle off. And versions with shields can take quite a bit of punishment. Like the Caravel class, it has thrusters in the nose and tail to save weight, yet give good Pitch.
Monitor Ti-3:
Stats and Cost for Thu_Monitor Ti-3 gW 1.6.xml:
Credits: 11583
Iron: 1020
Titanium: 1693
Stats and Cost for Thu_Monitor Ti-Na-3 gW 5.7.xml:
Credits: 52984
Iron: 364
Titanium: 4966
Naonite: 986
Stats and Cost for Thu_Monitor Ti-Na-3 gW 9.6.xml:
Credits: 81191
Iron: 364
Titanium: 7320
Naonite: 986
Stats and Cost for Thu_Monitor Ti-Tr-3 gW 9.6.xml:
Credits: 91181
Iron: 384
Titanium: 7313
Trinium: 1707
Monitor Ti-4 cargo 176:
Stats and Cost for Thu_Monitor Ti-Na-4 gW 5.9 cargo 176.xml:
Credits: 56223
Iron: 1658
Titanium: 4981
Naonite: 986
Monitor Ti-4 cargo 336:
Stats and Cost for Thu_Monitor Ti-4 gW 9.6 cargo 336.xml:
Credits: 76213
Iron: 207
Titanium: 9308
Stats and Cost for Thu_Monitor Ti-Na-4 gW 5.9 cargo 336.xml (w/shields):
Credits: 59988
Iron: 364
Titanium: 7095
Naonite: 986
Stats and Cost for Thu_Monitor Ti-Na-4 gW 9.6 cargo 336.xml (w/shields):
Credits: 88194
Iron: 364
Titanium: 9449
Naonite: 986
Monitor Ti-4 cargo 876:
Stats and Cost for Thu_Monitor Ti-4 gW 1.9 cargo 876.xml:
Credits: 26747
Iron: 208
Titanium: 7118
Stats and Cost for Thu_Monitor Ti-4 gW 9.7 cargo 876.xml:
Credits: 83787
Iron: 208
Titanium: 11792
Stats and Cost for Thu_Monitor Ti-Na-4 gW 9.9 cargo 876.xml (w/shields):
Credits: 95765
Iron: 370
Titanium: 11928
Naonite: 986
Stats and Cost for Thu_Monitor Ti-Tr-4 gW 9.9 cargo 876.xml (w/shields):
Credits: 105755
Iron: 390
Titanium: 11920
Naonite: 1707
Monitor Ti-4 cargo 1752:
Stats and Cost for Thu_Monitor Ti-5 gW 2.1 cargo 1752.xml:
Credits: 38550
Iron: 3797
Titanium: 8143
Stats and Cost for Thu_Monitor Ti-Tr-5 gW 10.1 cargo 1752.xml (w/shields):
Credits: 117558
Iron: 3978
Titanium: 12945
Trinium: 1707
Download: Thu_Monitor.zip
-
Caravel class starter/small freighter
A recent attempt to modify the Needletail to further improve performance and cost ended up being a near complete redesign. The only thing that was borrowed were the nacelles. Taking what was learned from the Predator class, the Caravel class (sci-fi vernacular for a small, speedy freighter) has the thruster arms and nacelles in the middle, for higher thrust and cost efficency. And directional thrusters are pointed up/down in the nose and tail for Pitch, which eliminates the need for an upper and/or lower thruster arm, saving significant weight.
Being mostly iron, this is another starter ship. But it seems to work pretty well for an early, small freighter. Though more fragile than the Predator series, it's also much cheaper. As usual, the whole ship is protected by IFGs. And like the Predator, the thruster arms can take considerable punishment. This isn't designed for combat, though. The whole thing is covered in iron solar panels for cheap energy.
Sans cargo:
[spoiler=Stats and Cost]
Credits: 3987
Iron: 796
Titanium: 209
With the 74 unit cargo container:
[spoiler=Stats and Cost]
Credits: 5735
Iron: 1500
Titanium: 218
With the 336 unit cargo container:
[spoiler=Stats and Cost]
Credits: 9186
Iron: 2830
Titanium: 247
-
Predator I & II class early-game multi-role
After designing the Needletail and Locus classes, I learned a few things about thruster placement and the value of Directional Thrusters on long leverage arms. The Predator class was the culmination of my experiments. (As one may guess, it's named for the Predator drone, which it sort of resembles.)
Some of you are probably thinking that thrusters on leverage arms is a weak point that can be exploited. It depends on how you design it. For ships that may see combat, like the Predator, my leverage arms consist of no more than four blocks - two if I can manage. And I place these in such a way that at least two neighboring blocks would need to be destroyed before it would break off.
In the case of this Predator class, each Nacelle is connected to the main body by 4 blocks, each with about 15 HP. However, because of the IFGs, each of those 4 blocks can take 10 times the damage or 150 HP. And all four would have to be destroyed, which amounts to about 600 HP. With the whole ship having 863 HP, each arm can take about 70% of the damage of the ship.
Predator Mk I class:
[spoiler=Stats and Cost]
Credits: 15493
Iron: 2004
Titanium: 1369
Predator Mk II class:
[spoiler=Stats and Cost]
Credits: 15497
Iron: 1934
Titanium: 1441
You can see from the stats that the Predator II has better Yaw and Pitch than the Predator I. Though, the only major change was the side thruster arms were relocated from the back to the middle. This demonstrates how - at least with Directional Thrusters - placing thruster arms near the middle is best. I would move the bottom thruster arm to the middle, too, except that would interfere with adding a cargo container. Also, it looks better this way.
(For more about thruster effectiveness and placement, see my Thruster Nacelles & Leverage Arms post. Click the spoiler tags to read about it. Also, there is my rant in "The Cube Meta" topic.)
-
Needletail Mk V class starter/small freighter
The Needletail was my early attempt at thruster arms to increase Yaw and Pitch. But instead of doing a classic cross shape - which tends to obscure the top of the ship and thus our view from a 1'st person perspective - I wanted to do three nacelles that are more-or-less evenly spaced. This is difficult to design because the blocks in Avorion are cubic. Attempting a truly triagular (three-sided) shape is a challenge.
(Trivia: The white-throated needletail is the fastest level flying bird. Aside for being designed for speed, it got the name for the needle-like antenna in the back.)
My first versions used Regular Thrusters:
When the thruster changes happened, I did a few major redesigns. Current "V" version:
Sans cargo:
[spoiler=Stats and Cost]
Credits: 3271
Iron: 1398
Titanium: 105
With the 336 unit cargo container:
[spoiler=Stats and Cost]
Credits: 8470
Iron: 3433
Titanium: 142
-
Cutter class iron starter/scout
As I said, my "Cutter" class is designed to be a cheap starter exploration vessel, made mostly of iron. The first version ended up being much larger than I had intended. I renamed this "Cutter XL". Then I used tip #6 in Tips for ship building to resize this ship, making a version of the Cutter exactly 1/8'th the size (1/2 in X, Y, and Z) of the XL.
Stats for the tiny Cutter:
Credits: 572
Iron: 621
Titanium: 98
Stats for the larger Cutter:
Credits: 1980
Iron: 1228
Titanium: 37
-
I've been hard at work designing ships. And, since I've been playing the beta branch, this is what my new ships have been designed for. Due to the changes, I had to redesign a few. But all of these take advantage of Directional Thrusters on leverage arms.
Thus far, all of my ships have been designed for performance - high Pitch and Yaw in particular. Also, most of my ships are designed to have the option to carry cargo. They have respectable maneuverability even when carrying cargo. And they're designed for room to add on one or two Cargo Containers. (See my ship parts catalog topic.) For ease of use, I've included separate versions with the cargo containers (or cargo bay) already added.
So far, the only exception would be my "Cutter" class, which is strictly a cheap scout or exploration vessel.
The classification system that I use is mostly explained by this post in the "Ship class" topic. Following the name is a two-letter abbreviation for the main material(s), followed by a number which indicates the number of module slots (to indicate size). In addition, my later designs also have "gW" with a number to indicate how many Gigawatts it generates, followed by "cargo" and a number (if it has any cargo space).
I choose to design for performance. And I think the stats speak for themselves.
-
Cutter class iron starter/scout
-
Needletail Mk V class starter/small freighter
-
Predator I & II class early-game multi-role
-
Caravel class starter/small freighter
-
Monitor class mid-game multi-role
-
Tetralenos class medium freighter
-
Trilen & Trilenos class medium freighters
-
Azugalenos class medium freighter
-
Locus class multi-role (coming soon-ish™)
- (reserved)
Note 1: Most of my ships use 'Void Tech' engines and thrusters. Many of them also use void tech generators, capacitors and/or shields. (See the Scytales Laboratories topic for details.) The exceptions would be my later ships: Trilen, Trilenos, and Azugalenos, which don't use Void Tech.
Note 2: All my ships have all blocks protected by IFGs.
Note 3: So far, my only ships that come with shields are the Azugalenos, the Trilen, and certain versions of the Monitor. However, it should be possible to add a shield generator to many of them as many have some room to spare (usually with dull, colorless framework placeholders).
Note 4: After learning the hard way, I usually to keep my core block buried deep inside.
-
Cutter class iron starter/scout
-
I strive to optimize Pitch and Yaw on all my ships, even freighters. It really makes a difference in combat. Appearance is secondary. Roll, I don't usually care much about. Though, I don't want it to be in the red or yellow.
For my experience and theories on thruster placement and effectiveness, see my Thruster Nacelles & Leverage Arms post in my Ship Parts topic. Click the spoiler tags to read about it.
For even more details, see my rant here in "The Cube Meta" topic.
-
I thought you could change the material of a template by using the drop down in the top right corner of the block picker?.
Huh... ??? I didn't see that. Has that always been there, or is it new?
How do you build in Avorion?
in Gameplay Discussion
Posted
I've tried different methods, myself. And I would recommend others try different methods to see what works best for them. Also, I recommend designing ships in a creative galaxy in single-player in a safe sector so as to be able to go wild and avoid needless distractions.
I've tried the inside-out approach. It works, though the outside may end up looking different from what you originally planned. I've also tried the outside-in approach - sort of. It also works. Though, you may end up with unused, hollow space or the stats may be hard to tweak like you want.
In my experience, I've noticed that to build the outside first, it helps to build around a framework. Then, as it nears completion, I can use the Transform tool to transform framework blocks into the internals I need.
Sometimes, I look at other ships. It gets my creative juices flowing and helps inspire, since I can see firsthand what is possible. For this reason I keep a folder with images of various ships from the Creations area and elsewhere. I've also saved some images of non-Avorion ships.
Several of my ships started with a goal in mind. When I started my Needletail, I wanted to see if I could do a swept-back three-nacelle design, spacing them equidistant. It was a challenge as I had to use lots of triangle blocks. I had a rough idea how I wanted the front to look and how the nacelles would look.
Since the thruster changes, nearly all of my designs have been cross-shaped in order to maximize the effectiveness of directional thrusters.
While performance is high on my list of priorities, I don't want to sacrifice looks too much. Also, I try to design all my ships to be both decent at combat and have room to attach an optional cargo module in case I want to use it for trade. Perhaps going for a jack-of-all-trades approach is not the best, but it's what I like to do.
My preferred method is to start with a concept and some goals, write down some stats to aim for (such as the approximate amount of shields and/or cargo), and figure out how big it will need to be to fulfill those stats. Then, I may lay down a lot of cube framework blocks (usually 1x1x1 or 2x2x2) to form a basic outline. An example:
You should be able to make out the crew quarters in the front. I had to add this to hold Mechanics to prevent it from breaking down while under construction. And I can Transform additional framework blocks into more crew quarters as it grows.
I regularly refer to this reddit with a chart that associates volume with the number of module slots. In this case, the framework prototype did not have the volume (module slots) I was aiming for. So, I had to start over and create a new, larger framework using larger blocks. Thus, laying down a framework prototype saved me a lot of time and effort.
After I'm satisfied with my prototype, I add a few more framework blocks vertically to add some distance. Then I start building the actual ship, using the framework below as a guideline for the shape and size, imagining it as the shadow it would cast.
I used a different approach when I designed Tetralenos. My goal was to build a freighter for trade, so I literally built a ship around a single, cube-shaped cargo bay block. Even so, my initial design and plan changed drastically several times.
For my latest ship, I'm trying a still different approach. After I figure out the stats I will aim for - shields, power, cargo space, etc. - I'm building a very rough and ugly prototype. For example, knowing that I want to aim for about 120000 HP in shields, I figured out that a 7x7x7 cube of Xanion would give me that and add about 0.34 million cubic meters to ship volume. And so I build a rough outline of the shape with the actual volume of shields, power generators, engines, etc. Here's an early (discarded) version:
I'll use this much like my framework prototype of earlier, except it should give me a much better idea of how big the cargo bay will be, how much room I should devote to thrusters, etc. And I'm relying on it more for a guideline on volume and size rather than for the actual shape.
BTW: Originally, I wasn't sure how much power I should aim for. A friend on the server where I play said that he aims "to have only a third of the [energy] bar filled." In other words, he tries to have excess energy so that no more than a third of the ship's energy is used, at rest (without modules). Myself, I would only feel comfortable with a wider safety margin, esp. since I like to use Robotic Crew modules. So, I'm aiming to have mine to use less than 1/4'th, at rest. And I may aim for even more power on a combat-oriented ship with power-hungry guns.
Such a guideline seems less useful on "small" (less than 5 modules) ships, though, as modules can consume many GW of power and even a few GW can be significant at that size. On large (7+ modules) ships, it seems more reliable.