RubyPress

So a while back I had thought my WordPress install was getting exploited through some bug and decided I wanted to write my own replacement.. Turns out that bug had nothing to do with wordpress but was rather an issue with my host.. It’s been resolved, but that niggling desire to create my own blogging software stuck..

This is something I wanted to create from the ground up that would be a 100% “my own” solution to include even the look&feel theme of the thing and I decided that it would also be 100% open source from the start.

I wanted to use this project as a possible way to expand my horizons but I also wanted something functional that I could use and expand on quickly;  the more I thought (and still think) about this the sooner I wanted off WordPress, not because WordPress is a bad thing (it’s not) but because I wanted something that was fully mine that I could modify to my hearts content.

So with that, I moved forward with my project and started looking at which language I wanted to write it in.   After some consideration I decided to limit myself to writing the code in Python, Ruby, Node, or PHP.

I knew from the get-go that I didn’t really want to write this thing in PHP.   PHP is a great language but it’s getting a little long in the tooth and frankly some of the frameworks out there leave a lot to be desired.   Plus I have been writing PHP since ’96 or so.. I’m ready to move on..

RubyPress Admin

So with my original intent in mind I decided to branch out and try NodeJS as I really like javascript and I think node is fairly clever.  I played with node for a couple weeks even making an initial start on the blogging software and while I do believe Node is pretty cool, I just don’t think its “quite there” yet.  I think node has a little bit of growing to do before it’s fully matured, which isn’t to take away from it at all but rather just my opinion on it’s current state.  For whatever reason, node didn’t excite me.

This left me in a bit of a pickle;  Suddenly my language options went to Python, Ruby or back to PHP.   At the time I knew nothing about Python or Ruby compared to my PHP knowledge.

I had to do some serious considering here and what I ultimately came back to was one simple fact;  PHP and the myriad of MVC frameworks I’ve worked with (Zend, Kohana, Cake, Symfony etc…) as well as Python and it’s MVC frameworks (Django’s really the only one I have any experience with) all do their damnedest to do what Rails is doing.

Rails does all the big stuff right out of the box and then you have the whole Gem library thing… dear god.. RoR seems awesome.. okay lets give that a try!  I’ve never so much as cracked a book on Ruby before so this will be a 100% new thing for me.. *shrug* why not?!  Off to the races I went.

Side note:  I found that Lynda.com has some nice beginner intro courses on Ruby and Rails and of course Ryan Bates at Railscasts.com has lots of good video information for the watching.. both are totally worth checking out..

Now set the “not so” way-forward machine to 10 days later:    I’ve been working on it for 10 days or so, on and off, and in that amount of time I have created a system that allows me to import content directly from wordpress and display it on the site as seen in the screenshot above using my new theme.    I can create users, posts, pages, messages, have a functional (if sparse yet) admin area that forces authentication etc..  and they all work!

So, in 10 days using a language and framework I don’t know and have never used before,  using plugins and modules I don’t know and have never used before, even a brand new IDE..  etc..  I have managed to implement the core functionality of WordPress..    Holy crap, Ruby and Rails are easy as hell..   Seriously this stuff writes itself..  (I’m using Twitter Bootstrap and Devise which I think are both brilliant and likely will be using CanCan and a few other gems as I need them. )

I’m jokingly calling this project “RubyPress”, more as a codename than a real product name since it’s now become a Ruby blogging software to replace WordPress..

I have no idea if it will grow beyond my getting the basic blogging functionality working.. I still have a lot of stuff that are requirements like categories, tags, theme support..etc..  Lots and lots to do.. but I’m really enjoying it, and learning Rails at the same time.. so win win..   There is still a lot of “magic” to Ruby for me, stuff that I’m not sure how or why it works the way it does..etc.. but slowly I’m getting it..  I’m sure this thing will require a dozen rewrites before it’s not embarrassing.. heh  What can I say.. I’m a Ruby and Rails noob.. bear with me..

Though I have to admin, it’s coming together so quickly that I do entertain the notion of having something that might one day actually compete with WordPress.. Wordpress itself has been around since 2003 and has a lot of very cool functionality that I doubt I’ll ever match in my little toy project, but wouldn’t it be fun?  =D

It is 100% open source; https://github.com/palamedes/Press

I will keep working on it and eventually have something worth using I hope..  For now I’m still on WordPress..

Click for Example

Final Update:  Okay.. I owe a huge apology to the wordpress folks..  After doing a bunch of digging this isn’t a wordpress bug at all, but rather a server exploit that someone used to edit a good portion of the .php files on my site in order to inject their shitty obfuscated iframe crap.  There were files that were edited completely outside the realm of wordpress, on my server that the wordpress user didn’t even have access too.. So wordpress is completely off the hook!  Sorry guys.

Now having said that I’m still planning on making my own blogging software for funsies.. but I am not feeling as much pressure now since I now know how to purge the iframes off and wordpress isn’t to blame.

Update: Click the image to the right to see the highlighted code that got injected into the site tonight.. I have no clue how it got in there and am actively taking steps to figure it out.. it’s magically there one second then a refresh shows its gone..   I have tested my codebase thoroughly and I can’t find it anywhere, nor can I find it in any of the javascript that is being called..  Either way,  Please don’t install or run anything that you download from my site.  Note: I’m deactivating all extra plugins for the time being until I figure out whats causing this.

As many of you know, I am very much my own special kind of nerd.  I enjoy doing many a’geeky thing that others wouldn’t even dream of.  To be honest, I think all software engineers have to be a bit brain damaged just to do their job.  Recently my own affliction has been leading me down the path of writing my own blogging and content management system.

I have used a lot of different packages over the years from full on CMS systems like Joomla and Mambo to more specific blogging solutions like WordPress and Blogger..  They all have their good points and their bad points..  I have stuck with wordpress the longest as it most fit my need, but the more I dig into programming for wordpress the more I hate it..  I really think that wordpress needs to be rewritten from the ground up, but that’s a post for a different thread..

Long story short, I have decided to begin work on my very own blogging and content management system that is fully featured and professionally developed.    My ultimate goal is to make it a fully featured suite of tools that folks can install and use for their own site that has a very low bar to entry.    It should be easy to install and work seamlessly out of the box..etc..

The main reason lately that this has become an issue is that I’m tired of having folks hack/inject code into my headers and footers.   I have done everything I know to do to harden my wordpress install, yet I am still constantly fighting random crap showing up in the various theme files on the site.

I hate it!

It’s frustrating and not fun.  More over, if they have the ability to do that then they have the ability to inject all manner of malicious code..

So with that I’m going to begin designing and building something that can do all the work I need and take all the stuff already done (all of my wordpress blog) and convert it.. etc..

Gonna be a lot of work.. design will be huge, but should be good in the long run if for no other reason than something fun to do..  yes.. I see this as fun..

Hostgator Sucks

I have moved my hosting off Hostgator because frankly, they suck.     My site was suffering repeated down times, crashes, database disconnects, SSH daemon crashes/hangups, ..etc..

The big pisser for me though was the SSH restricted access that they had given me, combined with the more recent downtime..

For whatever reason I’d try to SSH into my account and it would fail acting as if there was no SSH daemon running..  I’d have to contact them via their live chat system, wait in queue forever for some support monkey to help me, only to have them either tell me that;  A) nothings wrong.  B) I don’t have an account with them or best of all C) “Oh yeah we changed the SSH port to ### now.. use that.. ”

Seriously?  You just up and changed the SSH port?  Why?!  It’s been 22 for ever.. Now suddenly its 2222? … WHY?!

Apparently their support team is filled with more assholes than a donkey in an iron maiden..

Okay that probably wasn’t fair, but it was funny =)..  But I mean seriously, If I can’t SSH/SCP into my box then there is a serious issue here..  They also have it setup apparently so that you can’t have more than 2 SSH connections at a time, and each time you SCP a file to/from the box it creates a new connection, thus that’s the reason for the failure of the third connection attempt and the reason you have to wait for 3 minutes for those connections to time out.  (Or so their tech support drone said.. )  What a crock of shit..

And when I did finally manage to SSH in they had me stuck in a jail shell so I can’t do dick anyway.. I just wanted to use screen people.. that’s all!!

It wasn’t so bad 2 years ago when I first got the account, but it’s been getting worse and recently my site was down for nearly 48 hours.. Done.. Stick a fork in me..

I’m kicking them to the curb..  Their shit stinks and if they continue they are going to be hurting for business..

EDIT:  Okay so I did talk to them at length over the phone about the service I was getting and they didn’t make a lot of excuses or try to be shady about it, but rather owned up to the issues and apologized.    While I am unhappy with their service, I am impressed at how they handled the whole thing..  They were very professional and courteous..   So with that I’m still moving on, but I am willing to give them the benefit of the doubt and say that they are still a good company, if maybe just not the right ones for me.

SU 152 -- Current Render

I have been playing a lot of World of Tanks lately — more specifically, I have been playing the Soviet Tank Destroyers and I finally got into the tier 7 vehicle;  The SU-152.   I liked the look of the vehicle enough that I thought it would be fun to model one up.

I also thought I would twist this project up a bit and do two things I haven’t done before;

1. I will be creating this model in 1 hour increments, and will only spend a grand total of 10 hours on it.
2. At the end of 10 hours, I will release this model to the wild for free.

Now I may work on the model more than 10 hours once things are all said and done, but my goal is to get the model mostly done in that 10 hour period and release it.  Anything I do beyond the 10 hour period is mine =)

Also, I will say that as far as WoT goes, the SU-85 is much more fun to play and will likely also get a model from me once I get this 152 bug out of my system.

So with that here we go.

SU 152 -- Hour 1

Hour 1 — Setup and Initial shape

The first hour I spent setting up my backdrop images that I would be using to model this sucker as well as creating the basic lighting rig, and render settings.

Once I got all that crap out of the way (and I’m not sure it was fully fair of me to count that as part of the 10 hours.. but I did..) I started on the basic shape of the vehicle and had hoped to get a lot farther along than I did.

SU 152 -- Hour 2

Hour 2 — Refine basic shape

The second hour I spent getting the main body to match up with the backdrops and to get all the curvy bits to curve, and the straight bits to be straight.   The body you see is still synchronous along the x axis, but wont be for long.  Again I had hoped and planned to get more done in the second hour.

These hours are going by fast!  I wrote out a plan of what I thought I could get done in each hour and I’m failing to make the mark.

SU 152 -- Hour 3

Hour 3 — Rear end work and beginnings of road wheel

While I added a lot of rear end basic shape, the majority of the work in this hour was on the beginnings of that roadwheel.

The reason I spent so much time on it is because its a 100% subpatch, where as the main body of the tank isn’t at all.   This means the road wheel once done, is done.  Wont need to revisit it to get the detail out of it.  Even so again this hour flew by.

SU 152 -- Hour 4

Hour 4 — Making the body 100% subdividable

Okay so the subdivision thing had to be resolved as it’s unacceptable to me to model up something that can’t be subdivided.  So with that, this hour was spent converting the entire tank body to sub-d.   This pretty much ends the synchronous body work as the details I start to add are all going to be off center one way or another.

This was an ugly sub-d conversion as I really was pressed for time, but it works and renders out nicely as you can see.

SU 152 -- Hour 5

Hour 5 — More Roadwheel work

I spent this hour working on the road wheels.  The tracks are going to take the longest amount of time for sure, but I had hoped to have all of the road wheels done by this hour.  I had hour six set aside as the hour in which I would make the single track, and then in hour seven I’d get it correctly arrayed around the road wheels.

Well at this point I don’t have all the road wheels done.  I still have the six main road wheels, their spring assemblies, and the three return wheels to do.. =(

It’s becoming increasingly evident that 10 hours will not be enough time at my skill level to get this thing done.   I’ll likely release the 10hour model, then continue on with it and release more as I get more done.

SU 152 -- Hour 6

Hour 6 — Even More Road Wheels!

I spent this hour adding the six primary road wheels to the model.  Thankfully, its the same road wheel cloned a bunch of times so not a lot of extra work was needed.. I just had to create one wheel.    Even so it took a long time to get it looking the way my plans show.. its actually a very detailed wheel..

I still need to model up the linkages that go between the backs of the wheels and the hull, create the three return wheels and if I have time, add some of the details that are missing.

Once that’s done.. Track!

SU 152 -- Hour 7

Hour 7 — Road Wheel details and return wheels etc..

This hour (which in all honesty was closer to an hour and a half) was spent getting all the road wheel details in place.. The wheel pistons, spring assemblies, and return wheels are all in now.

Again thankfully I only had to do each one time, then clone them over and over.   I also skimped a little on some of the details just due to time.  Like for example, none of the wheels are actually connected to anything.. They just kinda float there.. I’ll get to that eventually, but so long as it doesn’t show up in a render…

In hour 8 I hope to get the track done, leaving hour 9 and 10 for the body details and gun.

No way in hell I’m making it in 10 hours..

Modo CTD

Hour 8 — Crash To Desktop

I love Modo.  I really think its easily one of the best modeling packages out there, but its instability pisses me off..

The application just randomly crashes to desktop sometimes for no apparent reason.   In 301 I used to avoid certain tools cause they just about always caused a CTD, but in 401 its mind bogglingly random.  You can use the same tool a 100 times on the exact same model, and on the 101′st time it will crash.

I end up saving my models quite often for fear that this will happen, cause I know it will.. I hate it..

SU 152 -- Hour 8

Hour 8 — The real thing..

Okay so after storming away from the computer in a fit of pique I came back and got back to working on what I was doing.  I was working on the roof doors and spring hatch assemblies.

Not much for an hours worth of work I know.. The crash really took it out of me..

Hour 9 will be my finishing off the roof and rear nurnies/greebles and then save hour 10 for the main gun.

I have accepted that there is no way I’m going to get the track on the tank within the 10 hours, so that will have to come at a later time. *sigh* sorry.

Either way, at hour 10 I’ll release a link to this model.

SU 152 -- Hour 9

Hour 9 — Roof Nurnies and Greebles

I went as fast as I could to get all the little roof details in and it still took me an age.   Most likely, for a 10 hour project I am employing way too much detail.. I just don’t know how else to do it..  The more I think about it the more I think that making extremely low poly models (for games and the like) is an art form that I’m not sure I’ll ever master.  (Modo says I’m up to 2.6 million polygons.. The WOT game version probably has only a couple hundred.. )

The rear end of the vehicles is still entirely unfinished and that sucks.  I’ll get to it, I wont let this one go unfinished as I am enjoying it but it wont happen in the 10 hours

I will be using hour 10 to work on the gun.

SU 152 -- Hour 10

Hour 10 — Gun!

Well I got as much done as I could in the short period of time that I had left for the main gun.. A lot is missing, and I’ll have to get to it as I can.

All in all this has been a very good experience.. 10 hours is no where near enough time to do anything remotely close to a fully finished project.

Plus, with the 2 CTD’s that happened, setup, odd breaks and overages.. this 10 hours was likely more like 13 and that just wasn’t enough..

My plan is to continue updating this guy and post the results here.

But as I promised above — Here’s the model – SU152 at 10 hours in Modo format:   SU152.zip

I would ask that you please not directly link to that file on your sites, but rather host it yourself.  If too many people start linking it I’ll be forced to remove it.

Usage;  You’re free to use this model in anyway you see fit so long as you please give credit where credit is due by mentioning my name {Jason ”Palamedes” Ellis} and give me a link back to this page so that others can use the model too.

Any and all comments are welcome..  Feel free to email me at;  

SU 152 -- Hour 11

And so it continues!

Hour 11 — Chin!

So I have gotten over my disappointment at not being able to finish the model in the alloted time period and continued to work on it.  During this hour I added the “chin” stuff that sits under the gun mantle.

I wont be updating the downloadable model on the site every time I do work.. But I will update it every couple of hours or as I see fit.

Also I might stretch the hours in to two or three hour sessions instead of a single hour in an attempt to get more done.  One hour is a very short amount of time and passes very quickly.

SU 152 -- Tracks

Hour 12 through 16 — Tracks!

Okay here you see the rear sprocket with 4 tracks wrapping around it..  I’m embarrassed at how many hours I have spent on not only modeling up a single track foot but also getting it correctly sized such that it can  actually articulate around the road wheels and sprockets.

I’m still no where near done.  There is no thickness to the tracks right now at all and I don’t have any pictures of the inside of the tracks, so I’m going to have to kind of make up some of that.

This is also a really bad render.. sorry about that..   Either way I’ll get it done and on the model so that it looks good eventually.. I like where I am headed with it finally, its just taking a long time..  Each track tread (foot? shoe?) has a lot of curves to it for traction that I want to get right.. More tomorrow.

SU 152 -- Hour 17/18

Hour 17 – 18 — Rear details

I stepped away from the tracks for a bit in order to do some rear details..  There is still a lot of work to be done back there and I need to get to it sooner or later.

I also changed the render light rig a tad bit by removing the directional light in favor of an area light.  The directional light was behaving oddly..

Also, the more I look at it the more I think my gas tanks aren’t large enough.  I’m looking at some research photos and in all photos they are clearly bigger than in my plans.. I’m going to scale them up a tad I think..

My BOINC Stats

I really like the idea of community computing, or grid computing..  I’ve always thought it was a damn good idea and I always try to leave one running. I mean think about all the wasted computer cycles that your machine could be using to do good things with when you’re not around.

Especially if you set it up to be completely invisible like I do..  I never know its running because I have it setup to run only when my machine is idle, and I disable the screen saver portion of it in favor of just showing a blank screen.  Doing it this way I have generated over a hundred thousand points and turned in a lot of good results for them to chew on and I never even noticed.  Yay doing my part, by doing nothing.  Seriously it costs me nothing; why not?

The down side of my method is that it’s so invisible that I didn’t notice it hadn’t been running for the last couple months.  I forgot to reinstall it when I formatted and reinstalled windows.  Woops.

So today I got BOINC reinstalled and set back up the way I like it and it’s now chewing away at the bigger problems of life, the universe and everything.

And since some will be curious, I typically do some World Community Grid projects (Help Conquer Cancer, Human Proteome Folding, FightAIDS@Home, Genome Comparison..etc), SETI@home Einstein@home, Docking@home and the MilkyWay@home stuff.  (See the graphic above)

I have used the Berkeley Open Infrastructure for Network Computing (BOINC) since 2004.  And I used a very scientific method for picking it; It was the first one I heard of and it was the one doing SETI@Home.  Okay not so scientific but it’s served me well (or rather I have served it well) for a while and I hope it’s doing some good.

Now-a-days, one of my main requirements is that they use the CPU and GPU to calculate results.   It seems right now the GPU calculation stuff is a tad underdeveloped or at least you have poor control over it.  I have two GPU’s in my machine, but it only uses one.  Or says it only uses one, who knows how many it actually uses.  But one is better than none.  The Nvidia CUDA core calculations are way faster than CPU only.. like 10x faster..  If you don’t use the GPU you’re wasting your time.

This got me to thinking, are there better grid systems out there?   Do some deserve our computer cycles more than another?   Why?   What do you use?    Do you have a grid machine setup?  Why not?  I honestly can’t think of a down side… Can you?

Okay that’s not true, I guess the major down side is your computer is always hyperactive.  It’s never “at rest”.  But is that a problem?  Computers aren’t like people in that they don’t rest.  A CPU using 10% of its processing power is effectively the same as a CPU using 100% of its processing power in all ways except one;  Heat.

So a computer that’s chewing away at @Home stuff is going to be hotter than an idle machine.  And there will be a slight spike in bandwidth when transferring the files back and forth, but its hardly worth mentioning as these files are tiny.   So that’s the two “major” downsides I can think of.

Thoughts?

Edit: I swear to god I had more points than this at one time.. Either I used a different login at some point in the past, or something got erased..  Oh well.. It’s not about the points..

Simple Grot Tank

So for giggles I have been doing some Modo work this weekend..   Warhammer 40k Grot Tanks amuse me so I thought it would be fun to make one in 3D..

I spent the first part of the weekend completely wiping my system and reinstalling windows, redownloading my copy of Modo from luxology and getting everything set back up..  It takes a while, but not too bad to be honest..

Once that was done I started working on the roadwheels and the track..

Tread Demo 1

First 4 hours — It took me forever to figure out how to get the one tread I created to clone around a bezier curve such that all the points matched up.. But once I did I was very excited!!  I think it looks great and it should be fairly easy to animate too..

The tread itself isn’t finished, but since it’s just 55ish clones of one tread anything I do to the parent magically appears on all the children.. Good times..

It’s WAAAY too clean for a grot tank right now but I figured I’d make the clean version of everything then dirty it up my mangling the mesh and adding dirty textures..

I had to ask via the Luxology forums how to group the instances correctly such that I could move them.  Otherwise they go all wonky when you click and drag..  But once done it made sense and I was ready to move on to the rest of the tank..

Grot Tank

+4 hours — The next thing I did was build in the basic body shape..  I made sure to build subdivision into this model from the start so it wouldn’t be something I would have to go back and do later..

I also spent some time adjusting my environment so that my renders weren’t straight up black and white..

I have had a couple folks ask me recently if I thought Modo was worth the money..  For me I have to say yes.. Even though I don’t do this professionally, It’s just a hobby play thing I still think it was well worth the cost.  For professionals I’d say it’s a must have..   Just the amount of work that one can do in a very short time is amazing..  It takes some training and experience, but still its fast going..

At this point I’m going to continue to work on the model, getting the “clean” tank finished before starting to add the bits of scrap metal, plating, rust and grime..etc..

Exhaust

+2 hours — So I spent a couple hours tonight beginning the exhaust system on the back of the tank.. Far from done of course, but looks good.. Again I went through and made sure that the mesh was 100% subdivision compliant and it looks good..

Its far from finished though as I’m not sure what to put inside the cutouts.. This is one place where the model can’t help me because its just solid..   I’ll come up with something.. Likely something silly.. this is after all a grot tank..

Exhaust 2

+4 hours — I spent the next four hours on this model getting the exhaust on the back of the tank looking the way I wanted it to look.   It’s pretty close to the model I have and should look darn good once textures are applied.

I probably went a bit overboard on the polycount but oh well that’s something I always do anyway..   There is a piece still missing, and that’s the little bar/strap that goes from the band around the middle to the tank body.   My practical model only has a stick of resin with no detail connecting the two, and that just wont do.. I’ll figure out something.. I have some ideas already..

Grot Tank

+2 hours — I went through and added a bunch more greebles, nurnies and what’s-it’s around the tank.. You can’t really tell in this image but there are a lot more.  Plus I started working on the tank body though most of the work is on the other side.. heh..

I have a lot more work to do but the main portion will be taking the perfectly looking model and distorting it to make it look beat up and made up scrap metal parts..  Should be fun..

I haven’t started the turret yet as I want to get a good handle on the rest of the tank first and there is a lot left to do.. The turret would just get in the way and make me want to work on it..  It’s the coolest part after all.. Saving the best for last!

Metal Plating

+4 hours — I have been adding a lot of little deformed plates to the model as per my practical..  Basically they look like extra bits that have been bolted on, but all of them are beat up as befitting a true Grot vehicle..  Eventually I’ll have to give the main tank body this same treatment..  But for now adding all the extra plating is taking time..  I’m not have way done with that..

Once I get the bulk of the plating done, I’ll start working on deforming the tank body itself, before adding the rivets and holes etc..  Texturing the noise and destruction on to the model will really make the tank come to life eventually.  Should be good.

More Detail

+ a bunch of hours — I have been adding more plating, and generally cleaning up the model..

The main body is just about ready to start getting its deformation stuff before I get on to working on the turret.  At this point though I think I need to research a better way to make some of these details.. Right now I’m actually modeling everything in which is a huge time sink as well as making the model be an unbelievable number of polygons..

Sheet Metal

Take for example the sheet metal image to the right.. that’s a lot of polygons for what amounts to a pretty small detail on the back of the model..   I’m sure I’ll get some points for actually modeling it, but it puts a lot of strain on the renderer..  Right now I’m over 5 million polys and I’m no where near done..

I think this is where the experts know super secret tricks and things that you can do to get that kind of detail without having to actually place pixels.. Bump, normal and parallax maps for example..  No idea how to use/create those though..

Last night I had yet another one bite the dust..  Thankfully it was my secondary game drive, not my primary OS drive.  I have learned to make my primary OS drive a SSD, they seem to last longer..

I’m likely going to go get another SSD today as my new game drive..  I’m just tired of the 1950′s technology that is HD’s failing on me..

C# Confusion

I’m really kind of frustrated at the general lack of helpful input from other people on the internet when it comes to trying to learn a new programming language..  It almost seems like folks just don’t want to be bothered or that they have the opinion that “by god it was hard for them, so it should be hard for you.. ”

Human beings are blessed with the ability to learn from one another, and while we are also often cursed with the disinclination to do so, quite often it seems that those “in the know” don’t want to share that knowledge when others are clambering for it..

I’m fairly new to C# and am interested in learning it..  I’m not new to programming though so I understand most of the basic concepts, I just need to learn the syntax and the “flavor” of the language..  The IDE is pretty familiar because I’m an old time Delphi programmer (C# was written by the same guy apparently)..

I decided I wanted to write an application that take a screenshot of a DirectX game while it was in the background (not minimized)..  I’m having a lot of trouble getting it to work correctly, and trying to find people to help is very much like trying to find Venuses arms..

I can’t help but feel very alone in my struggle and that plain sucks..

Latest Image

So I have always liked submarines, I think they are pretty neat.  I have wanted to create a 3D sub for a while and finally decided to get started on one..  Instead of going with a more modern design I went with one of the most popular submarine designs of old; the German type VII.   I also have to admit that another reason for this design is that I like the game “Silent Hunter” and the fifth installment of the game is coming out this year.

As I work on it, I’ll post images here to show where I am at and as with all my other posts the top most image will show the latest image or render.

I’m creating this entirely in Modo 401 using my Wacom tablet with the intent to make it as photo-realistic as possible.  That’s gonna be tough because it will force me to model everything in sub-d and it will really make me have to pay very special attention to my models materials, the shader tree and lighting..

The first 3 or 4 hours was scouring the net trying to find a set of plans for this boat.  Believe it or not it was harder than one might think..

The initial shape

+3 hours – Getting the initial shape - After getting my work area setup and getting the backdrops in place I was able to do this.  Not perfect, but not a bad start for getting the basic body shape down.  Next I’m going to try to form the outer hull actually into the bulb shape you see there.. for now that’s an additional mesh just sitting there.

Shape Refinement

+3 hours – Shape refinement – So I know it doesn’t look much different but it is.. I have refined the hull shape a lot so that now the top deck area is actually the shape it should be for holding the conning tower and what not.  But more importantly I have actually added those side bulges to the model such that they are part of the main layer and can be subdivided nicely.  No more having one model poke through another model.  Ding! “You have gained a level!”

It may sound like a simple thing but it’s actually complex if you’re coming from hard body modeling to subdivision modeling.

Front Wire Cutters

+3 hours – Front wire cutters – So I spent the better part of these three hours getting the front wire cutters (I guess that’s what they are) looking right under subd and making sure they smoothly transitioned into the rest of the mesh.  Good times.  It took longer than I wanted because I had some serious symmetry issues that I wasn’t sure how to sort out.  Thankfully, Modo has a Symmetry tool that does it all for you!  It’s just magic.. No clue how the hell that works.

I have also been researching what I want the top part of the boat to look like.. I wanted to keep the boat as accurate as possible and my research has lead me to believe that every fricken one of these boats was different..  For some reason I had it in my mind that manufacturing back then would be similar to how it is today.. But I guess I have to keep in mind that these things were built one at a time and each one was hand crafted.. So they tend to be a tad different from one another..

The more I work on this the more I think I would really like to visit a real one of these if for no other reason than to just gawk at it..

Conning Tower

+ 5 hours – The beginnings of a conning tower - I started and restarted this conning tower probably 10 times and even now I’m not 100% happy with it.   I kept going back and forth with the design as well as the number of tiers it would have.  This one has 2 which means this is a type VII-C or later as it will have two flak guns as opposed to only one.

It’s been hard to find plans with anything close to this so a lot of what I am doing comes from this kit built by Guillermo Martinez Baz..  Nice looking model Guillermo..

The shape of the tower has a lot of gentle curves and odd shapes so I have had to play with and learn the falloff tools a lot better in modo.. It was a great learning experience but I didn’t get a lot of work done for the 5 hours I spent on it..

Also the more I study Guillermo’s real world plastic model, the more I think my model is way off in some places..  My decking at the top of the model for example isn’t correct by a long shot.  Probably want I’m going to do is actually build an inner cigar shaped tube where the people would be to get some idea of why the exterior is shaped the way it is..

First Actual Render

+2 hours – The first actual render – So here you have the very first render of my entirely incomplete Type VII-C German U-Boat.  My goal is to have it render out with that underwater look but for now I’m going to make it really bright so I can see all the defects..

I still have a very long way to go.. I’m still working on the conning tower area as that will have a lot of detail.  I need to merge the two layers too so that the conning tower actually connects to the sub body, as right now they don’t.  Still having fun!  That’s what counts..

Details and Lighting

+4 hours – Adding detail and lighting – So there isn’t much of a change to the model here other than for details that are hard to see, and the obvious lighting differences.  I’m going to be making the necessary detail changes to the model to make it look better and better slowly over time, but I’m also going to be working on texturing and lighting while I can.

I have also learned that its not necessary to make everything part of the main mesh.. It’s okay to have sub-d elements in different layers so long as it looks okay.  The idea being if in the real world something is made of a single piece of metal then sure one layer is appropriate, otherwise more than one is fine..

Tail Features

+7 hours – Tail Features – I have spent a good portion of the day working on the tail features of my sub as well as generally mucking around with the over all look of the boat.   But primarily, the rear dive plains, propeller shafts, rudders, supports, struts..etc..  I have a long way to go but this put a nice dent into the rear features that up to this point were completely missing.

During this process it occurred to me that I don’t know which way the props should be twisted.. And I kind of assumed that they counter rotate.. So if that is way off some one please tell me..  The rudders are actually to scale according to the plans I have but damn they look big.. And my plans arent accurate at all with how far apart the propellers are, and so I did my best to “Guesstimate” and I think I blew it..  They seem a little too close together..

The more I work on this thing the more I hate how some of it has come out.. I spent such a long time getting those side bulges working and now I think they are too prominent..  The top decking is all wrong too in a few places..   Still I know I should give myself a break.. I mean this wont be a perfect thing no matter how hard I try.. I just don’t have the skills yet..  But I am going to work at it..  But my dreams of making this thing photo realistic likely wont happen at all..

And I haven’t even started to ruin the textures..

+10 hours – Still going! – Okay so I don’t have a picture this update..  I’m in the process of correcting a horrible mistake I made..   At some point I subdivided the entire main hull and it caused a lot of odd rendering artifacts.. I was pretty upset that I had done this at first because apparently I did it days ago and didn’t even notice!  Luckily I make daily backups of my model and I work in a lot of layers.. so thankfully its only a matter of using the old hull and bringing it back up to date, preferably without making the same mistake..

Yay for rails

+5 hours – Yay for rails – I’m guessing todays installment was roughly five hours worth of work..  Hard to say really in that it wasn’t a straight piece of work..  It’s really a continuation of work I did last night, and what little work I did today in between bug fixing for work which has been going on at a rabid pace..  Every hour or two I spend 20 minutes or so on the model… The bug fixing is killing me so this makes a nice change of pace..

In this update you’ll see I’m starting to add lots and lots of railing everywhere… Ah good times..  I have also completely fixed the ugly striation crap that managed to work its way into the model because I was a dolt and did something stupid..

Part of this fix was also to correct some big errors with the conning tower.. I still have a lot of work to do.. but its coming along nicely..

Steel Rope

+5 hours – Steel Rope – I just got done spending the better part of the entire 5 hours learning to make rope.  3D rope that is..  The sub has quite a lot of steel rope hanging from various points on the hull and I wanted to get that to look right.

At first I just created a really long cylinder and sliced it up with the intent of just adding a 3D texture or displacement to make it look like strands of steel rope.. but it just looked crappy.. So, I had to actually create stranded rope..  This took me a long time and I had to get pretty intimate with Modo’s Twist tool and the Soft Move tool..

I still think the way I ultimately did it, was the wrong way..  I should have used a path and cloned a single rope piece along that path, but it didn’t work..  No matter what I tried, I just couldn’t get the curve clone tool to output something that looked reasonable..  Plus it was REALLY laggy bringing modo to a stop trying to clone a <1000 polygon item 100 times..  100,000 polygons isn’t that heavy so I have no idea why it was struggling so hard with it.. I also tried to use a custom python script suggested to me on the modo forums called CurveDeformer, but it failed too.. (in odd ways) .. SO I was left to my own devices..

It took me too long to do this though.. I mean seriously ~5 hours to figure out this ONE thing.. sheesh..

Steel Rope Details

+3 hours – Steel Rope Detail – So I spent the majority of this time getting the steel rope finished on the back side of the boat.  It now hangs correctly between the two rail uprights on the back of the boat in such a way as to imply weight..  In the tail of the boat it rides over the risers in a ring then down into a hull point where its anchored.  And its simply wrapped around the lower conning tower platform rail and clamped together using a pair of wire clamps.  This isn’t how it was actually done, but I thought having the wire simply wrapped around the rail and secured with a couple wire clamps made it look more real..   Very excited as to how it turned out.. I think it looks fantastic..

This model is starting to get rather beefy.. It’s already a 5 meg file and renders out at over 3 million polygons, 15 million with subsampling turned on.  I call that starting to get heavy..  I know that a lot of the models the big boys work with are vast by comparison, but then so is the hardware they use..

Gonna keep working on this thing.. I haven’t been playing any games at all since I started this project, its kept me that entertained.. Tomorrow I’m going to continue poking the detail holes in the side of the boat.. Good times.

Poking holes

+8 hours – Poking holes in the model – For todays installment of “make a sub” I have been concentrating on poking holes in the side of the main hull mesh. It’s actually exceedingly difficult because of the non-linear off axis subd modeling method I’m using, combined with round holes.. I have to say so far so good! I have a lot more holes to poke, and I need to add inner piping and details that could be visible from the outside.. This means I need to build a very basic model of what the internal cigar tube will look like and the basic support structures.. I have no clue what those actually should look like so I might just make it up..

Waist Steel Cable

I also spent a chunk of my time adding the waist steel cables to the waist railings.. I went to a lot of trouble too wrapping the cable around the railings in a believable way.. I think it turned out pretty damn good if I do say so myself.. Though looking at it now in some areas I should have added a tad more twist to the cable, but since each cable is actually a strand of six cylinders its a pain in the rump to go back and change it now.. I’m really glad I went this route though as opposed to the one cylinder that gets textured into wire.. The actual twisted polygons while heavy in terms of polycount, really adds a lot in terms of details..

I have to say I’m still have a great deal of fun making this model.. that bodes very well.. Typically when I do these things, they last for a couple of weeks before I want to move on to something else, but this model is really inspiring me to keep going.. It’s been really cool seeing how each piece comes together to add up to more and more detail.. Every so often I scroll up to the top of this post and look at where it started and then look at where I am.. I realize I have a long way to go, but I have come a LONG way..

I want to find some more pictures of real Type VII’s up close so I can get a good idea of what the hull plating should actually look like. I dunno if I should model that or just texture it.. I’ll likely model the rivets, but texture the plates.. we shall see..

At this point this model is easily the most complex and well done model I have done to date.. Not only that but its by far the largest in terms of polygon count, file size.. render time..etc..

Conning Tower Details

+10 hours – Conning Tower Details – So after a day or two break I am back at it and have added a lot of detail to the conning tower.  I have a lot more I want to do and shall continue to work on it..  For this installment I have added several rails around the front of the tower for hand holds, as well as adding several doors and panels with hinges and handles..   I’m likely adding way too much in the way of details in that no one will ever see some of it.. I mean whats the likelihood someone will zoom all the way in to a single hinge on one of those doors..  heh but hey what the hell..

It’s a learning process as much as it is anything else..  I mean I want this thing to look good and having a high-poly model that renders out well is the beginnings of that..

The real issue at this point is that the model is getting big enough that rendering it is starting to run me out of memory..  Any form of adaptive subdivision is right out of the question..   Plus I can only do one or two renders before I have to reload the model fresh.. memory leak I guess..  It’s only mildly annoying..

The Arjuna theme is outstanding..  I especially like that it works in the exact same way as EOS and so theme switching is 100% seamless!  Thank you guys a bunch.. This is awesome..

I highly recommend this theme, or the EOS theme.