Wednesday, January 22, 2014

Long knife project

I was practicing for one of my examns, and I ended up having some left over materials.
This included a 45 cm long piece of steel.

Let's do something crazy and make a makeshift long knife out of it.
While we're at it, let's do this the cheap-ass manly way and not use any power tools whatsoever.
Sure, there are smarter ways to do this, but I just want to see if I can pull it off this way.

Materials used so far:
- Steel profile, used half - 1m length, 2cm width, 4mm thickness - 4,5€ / 2 = 2,25€
- Sanding paper - 1,5€
- Small handsaw - 3€, not sure, will verify
Most of the stuff bought at action (

Materials planned for use:
- Cork sheets to be used for the handle.

Actions taken so far:
- Sawed profile to length
- Sawed one end to create tip of the blade
- Started sanding to create still blunt edge, tip is already sharp.

- Beveling relevant angles and creating edge with sanding paper: Really slow process, cleaner finish. Sanding paper wears out really fast.
- Use saw as makeshift file: much faster, much messier

Observations thus far:
- This takes a really, really long time. (Duh!)
- Might need better sanding paper

Work time:
One week so far, during breaks and combined with a ton of other projects.

Pictures coming soon

Thursday, November 7, 2013

Tablet drawing app experiment

I've just started experimenting around with sketcher free on my cheap tablet, and it's quite a bit of fun.
Let's see if I can make this better in gimp.
Let's start off with the original image.

Let the editing begin!
First off I tried doing some layer duplication, with one layer using edge detect.
The difference is barely notable.
After that, I tried adding a background image with a clothify filter added to it, and set the original image as a "multiply layer"
Also noticed I forgot to delete the alpha layer on the original image. I might have wasted a lot of time on this later on, had I neglected this..
Won't add this intermediary step, but I experimented with the "old photo filter as well.

I dropped this version and added some lighting effects instead.
That's enough fun with this one for now.

Monday, September 23, 2013

New project: bullet hell game

I've been looking around on the android market after buying a cheap tablet pc, which I'll be reviewing at some point as well.
I've been test-playing some of the bullet hell games around, and the ones I've played have a lot of room for improvement to say the least.

I have few criteria for bullet hell games that I actually consider rules rather than guidelines:
- The more difficult parts in the game are defined by the small size of the screen NOT occupied by bullets, lasers or other things that will kill you upon contact.
- Hitboxes are well defined and the hitbox for the player object in particular is small (solid guideline: either one pixel or approximately the size of the cockpit)
- These are adrenaline pumped games, and when done properly, art games. The pattern formed by the enemy projectiles is just supposed to be a wall of death, it's supposed to be a work of art, distracting the player even further, adding to the level of difficulty. Also, the music isn't supposed to be just there, it's supposed to accommodate the action. This last part isn't a must, but it really adds to the quality of the work.

What I've seen in some of these games:
- Bullets and other collision objects mesh into the background, and are therefore hard to see, resulting in a "what the hell just hit me?" feeling of frustration.
- Hitboxes were too large for a bullet hell game, or worse were poorly defined, resulting in a "that bullet was a mile away, how the hell did it just hit me?" feeling of frustation.
- Really poor controls. If I put my finger on a part of the screen, that's where I want my bloody ship to go, not two inches above it, not some awkward relative control system, but right on that bloody spot.
- Boring, tedious bosses only defined by having a butload of hitpoints, rather than any real exciting features, repeating the same monotonous boring pattern over and over again.
- Not enough weapons/features on the player ship/character/object.
- Levels too long without anything really excited happening.

The plan for my little project:
- Built using unity with blender made graphics
- One single level to start
- Multistage boss
- An interesting bullet pattern that meshes well with the music
- A small player hitbox and implementation of a system that rewards bullet grazing in some fashion.

After that:
- Take everything I've learned and implement to make a full fledged game for the android platform.

Tuesday, July 16, 2013

Back to doing some more 3D modeling: cargo ship design process

Hi, It's been a while since I've done some stuff.
I'm modeling a cargo ship and going through it stage by stage.
(Actually, I'm more focussed on making modular components that can be combined into more designs. May or may not be game design related, still thinking about that :p )
To start off: here's the basic layout for the design:
Now for the refining process, let's start off with the core block first.

Next up, the cargo blocks have been reworked for the V2.
It's been a while since I last updated yet again.
Cockpit area and engine block added. Time to add some point defense guns in case it needs to handle some floating debris.

Added a pair of railguns, but for now, the thing still looks a bit boring, right? Time to add some textures
After some fiddling around with the materials, here's what I got
... Probably going to rework that soon, but for now, I'm going to experiment with the modules I've done so far.

Gunship variant:

Well, you can't really make up anything out of these screenshots...
How to highlight the edges in blender: Under render options - Post Processing - Edge

Saturday, June 15, 2013

Dealing with ransomware

I've just had a fun run-in with the so-called police virus.

The version I encountered basically locks up your computer, attempts to access your webcam and shows a full screen notification with several police logos and a note that if I don't deposit 100€ as a fine, legal action will be taken.
Starting the computer in safe mode did not work, because windows immediately rebooted upon startup.

Don't panic if it happens to you. The authorities are not involved.
Also, do not pay, it will only encourage others to pull off similar stunts.

Steps taken to solve the problem:

- Intall Hitman pro kickstart (Link) on a usb drive using another computer

- Boot from the usb drive ( For asus, boot selection activates by holding esc from the moment you press the power button)

- Choose option 3

- System will start up and hitman pro kickstart will override all other processes and appear fullscreen.

- Start the scan and let it run

- Reboot

- Full system scan with malwarebytes and avast for good measure

Monday, May 20, 2013

Latest blender experiment

I'm in the process of rendering a tanker (ship).
Still early progress'ish, but coming along quite nicely.

The materials were acquired from blender open material database
Only around 366 faces for this one, but only one problem.
It takes forever to render, so I doubt it's suitable for ingame use yet.
I'm going to try to bake it next.

First attempt, copy the mesh, unwrap and bake to selected.
Result: too dark
 Used two standard lights in the baking process.

Second attempt: used a normal baking method this time, results are better this time, and I might actually be able to use it as an ingame passive prop now.

Thursday, May 9, 2013

3D scanning experiment

I've had this fun little experiment in my head for a short while now, and I decided to give it a go.

I've basically just improvised a crude 3D scanner.

Components used:
- Trust spotlight webcam (paid around 9,99€)
- Laser level square (was lying around in garage, price unknown, but didn't feel like getting the parts required to build diy line laser)
- David laserscanner software (free)
- Sheet of cardboard/foam plates at a 90° angle to keep calibration sheet in place.

Here's a single angle attempt at scanning so far:
 It's a scan of a little sculpt I did a few years ago.

Issues I have yet to properly resolve:
- The residues you see around the scan results. Can probably resolved through more controlled sweeping of the laser line and tweaking of the scanning corner.

Issues I have run across and found a solution for:
- Can't start camera error: under hardware setup, click the camera format button, and set up the format. The camera should start up now.
- Calibration issues: Place the camera close enough to the scanning corner, make sure enough of the markers are in frame, use a flashlight/lamp to illuminate the scanning corner and tweak the cam properties (contrast and brightness specifically)
- Laser not detected: Change the cam properties, again tweak contrast and brightness.

Software can be downloaded at: