Random stuff

U48 passed and this time it was a success. I was pleased with the resulting product as long as my part in it was concerned at least. And here’s some photos still from 2010. Summer approached and the editing queue is growing as I handle clients first but I’m not stressing it too much.

I have a folder that I gather summerphotos to (in here.) and I added a bunch again so dropping couple of in this side too. This covers some spring pictures and pictures from Uneton48 competition where we failed miserably last year. Anyway it is again relevant as in May this year we’re going to try failing less in the same competition again. =)

Added photos from short trip to Milan & Venice last spring. Here is the usual bunch of favorite stuff and rest of the photos can be found from here.

Fast recap of year 2010 on my behalf:

I changed to work at Rightware and have been learning to use Autodesk Softimage (previously known as XSI) lately. XSI, in my opinion, seems to have quite a lot of burden (e.g. default keyboard shortcuts) from the old times. If my opinion changes I will tell. =)

Also I bought Canon Powershot S95 to carry around every day and didn’t really have time to edit my photos from last summer due the fact that I had too much going on in the process of getting a new job and such. Probably will get an inspiration at some stage for the editing but anyway – I’m slowly trying to move to taking photos according to the usage. So actually I’m using the S95 for snapshots that I’ll never edit and 5DM2 for the other side of the spectrum. Therefore the workload on processing the heavier stuff should get easier as I shoot a bit less with SLR (yes it might be time to drop that D from the start) cameras. But I’m happily taking photos with less technical/theoretical quality anyway. =D

Have been reading and doing physical exercise a bit more than before. Peopleware and Drive are books I’d recommend for anyone working in IT-industry at least.

Assembly 2010 came and went. I landed an entry to both Graphics and Fast graphics competitions where the entries resulted to places 11. (Graphics competition) and 1. (Fast graphics competition)

The fast graphics competition had a timelimit of 1,5h used for the piece and submission of the entry to the system. It also had a theme that it has to follow. This time the theme was “You’re holding it wrong.” The obvious iPhone themes crossed my mind but luckily I decided not to go that way and tried to come up with something at least a bit more original. I suppose not starting right away with the first idea and even sketching a few was a good choice this time. I used first 30min without starting the work at all and that left me approx. 1h to complete the painting which I pretty much used all the way. And like I said it ended up first in the competition. And to lend a bit of thoughts from my old pal, who ended up 3rd in the same competition, it worked well because it was indicating what would happen right after the moment instead of just showing it straight on.

I used two evenings on this piece. It wasn’t really in any way a sweat-and-tears effort. More likely a good mood doodling-kind of effort instead. Also I had a pile of errands to run for at the same time so I didn’t really polish or finish it. First the idea started from a boy fishing at night-time and the screen would be split to underwater and ground-view straight from the side where the underwater would be what the imagination of the boy would suggest (at least when I was a small boy the dark water was a bit scary at nights. =) The original composition might have worked to some extent but not really in this kind of competition as it wouldn’t be dynamic in any way. So the composition made a bit of a perspective twist and after a while I ended up in to the layout seen in the final version. Also the boy turned out to something else too. I think the original idea is not that present anymore. But anyway – the voyage was fun.

The Graphics competition had a great quality in the entries. There was 28 entries in the whole competition and 22 of the passed from the jury to the screen. One of them were disqualified after that due to copyright issues (or a rip-off to be more exact.) I ended up 11th but am not in any way disappointed by that. It’s quite there where it belongs anyway. At least when the two first ones were very different but still very high quality entries. Although otherwise the end results looked a bit messed up for me.

The event was great again. No surprises but good quality demos and intros, friends, meeting people, hanging out – all fun. Thanks to all!

Remember how just a ten years back we still were used to get loads of photos from print and then slip them to very uncomfortable slots or glued or slipped the corners in holders? Of course it was nice and memorable that way but I suppose nowadays people start to be content on having most of the pictures as low resolution versions in their hard-drives (with no backups.) From somewhere around the start of this millennium there has been a possibility to do those family albums relatively easily with a bearable price-tag using photo service providers with dedicated pagination applications with a focus on high-level controls. I did my first one in 2005 using services from Ifi, which is currently known as Ifolor, and back then both the paper and printing quality was low enough that I wasn’t keen of ordering more for any purpose. That was of course 5 years a go from now and I do not have knowledge on the current quality yet as the whole system in their print has changed as far as I recall.

Recently I got a bundle from a chocolate package to get photobook for a very small price and that got me involved again. The bundle was from TicTacPhoto and the price was spot on to do something fun. So we compiled a book of cat-photos for kids where we included all our own and some of our friends cats we had taken photos of. The software they provide for pagination is not too good quality. It does not handle other formats than .jpg and it is a bit slow as it can not handle more than maximum of 12.5% from my CPU power. Also the system and templates for layout were OK but not very good. Very slow to use in any case. Although we got it done in couple of hours and when we finally received the book the quality was OK. At least a lot better than the one 5 years back and we all liked it a lot. So later on I loaded my family-trip photos from last summer and did an another book but more about that when I receive the physical book itself.

Also that spawned an zeal to complete another one and I tried to find more about those services in Finland. There seems to be loads of it but my first on the line to test is kuvakirja.fi which is a site from Fotoyks. I created a very similar book from a summer-trip that I ordered from TicTacPhoto: ~60 pages and landscape layout. The software they provide for the layouting is easier to use, faster and has more features than the one TicTacPhoto provides. But still it could be a lot better and faster anyway. (I command you software developers: whatever you do, consider multithreading! =) Of course the material is from a different trip so they’re not straightly comparable.

I will post more about the results and services later with some pictures and as well will try out more of these services. The TicTacPhoto and Fotoyks are very similiar in cost. Next ones in the testing queue are Ifolor and Blurb. Feel free to suggest some.

Time passes. I started creating a realtime portfolio which runs realtime. More about that later when I really get it to a even first draft stage. Technical side is rolling already thanks to the versatility and ease of Unity3D.

Another 48 hour film project completed just a week a go. Boy does working around the clock multitasking directing, lighting, filming and acting deplete the energy quite fast! Was close to health-meltdown in the end. Although we did run out of time in the end – it was a project full of learning-moments, thoughts of processes and fun. We didn’t of course go with too serious of a goal as we are not trying to get break in the industry but still it was interesting to try multi-cam full-HD filming with two 5D Mark II’s and audio recorded separately with a good mic. Also we had a bunch of not-too-pro lights and took some time using them too. Although we didn’t have too much time to play with them to achieve something extraordinary but managed to pinch some experience out of them anyway. short trailer

Edited a bunch of photos from January-March. Some of it in: Winter 2009-2010 album.

And here’s a selection again:

For all you seemingly huge amount of iPad owners over there on the other side of the globe… I will shamelessly advertise anything I have been involved with…

ZB2 for iPad is now available:

iTunes

yay. Which reminds me to update my projects page with some stuff. =)

I updated the projects page.

I added info about the ZenBound project and moved all the projects on the same page. Now the link is http://random.gfxile.net/?page_id=69 if you need it.

In the meanwhile I updated a bit of my C.V. too. You can find it from here. I just added some and clarified couple of things. Dropped some old stuff too.

Actually.. a certain game is landing to be more exact…drumroll…:

Touch Arcade

Kotaku.com

Cnet

IGN

Pocket Gamer

Macnn

\o/

Even though I’ve worked as a freelancer for some time for just for a part of the stuff, it is always nice to see it getting picked up quite extensively.

I started to use JIRA and GreenHopper for scheduling, estimating and logging hours for personal projects and tasks somewhere around December 2009. It has now proven to be valuable tool for that. Before that I have had a bunch of different experiences with different softwares. I’ve used several different online and offline TODO-programs, Calendars, Wikis, etc. Of course that is kind of a wide variety of different applications that are meant for different uses but for task estimating and tracking it seems that JIRA is the thing. Although I think it is worth mentioning that I wouldn’t use it without GreenHopper Agile plugin anymore. I have plenty of professional history with JIRA anyway so it was easy to get going with it.

Now I have actually started to timetable my personal tasks at home etc. to quarters so that I have some kind of clue what I have planned to do even that I would not do that. =D Anyway the best part about that is that I can see if I have overbooked myself and can postpone things. Actually one of the very first things I noticed that when I am really being honest estimating the hours I would use for a task and put them in the line, I realize that I think I can do almost double the things that I really can and even when I estimate the hours I still underestimate the time they take. Being able to track that has an serious impact on stress levels and prevents the “everything is due at the same time and my walls fall on me when I lie awake feeling like tied with handcuffs inside a coconut”-phenomena. And those are probably the first and maybe the only reasons you should use some kind of hour estimation/task tracking at work. It helps you. When you know how and why you use it, it will not be an evil plan from your boss to track your breathing but a tool for you. _{(Of course it does not hurt to know how much time was used on which things and micromanagers tend to ditch the processes anyway as they do not want to get their micromanagement tracked.)}

Let’s take a small scale company with couple of guys working. They have excellent communication as they work together in a small office and they think that they can handle every issue with a small chat. That far it works well if the chemistry is in synchronization between the guys. But estimating hours might be a painful thing to do; in most cases the guys just decide that we do X and Y in T time. It goes by gut-feeling or the guys break it down to smaller pieces and estimate. Anyway in this way the estimating and communication may still work. Even calendar is used and they mark that milestone Z should be ready in T time. So far so good. If the guys did all this they are already reaching the basics (simplified it a bit though.) But this is not the stage where the stress pours in.

The stress pours in when a) the estimations do not hold b) goals are not met c) unknown factors arise. The guys will possibly meet a, b and c by the time of T in some form – it may be now harder to go back to estimations, see the whole picture and see that yeah: Here it went wrong or that thing X takes too much time and can be dropped. Disagree? Yes, it can be done but the difference in using tracking tools is that you get a new pair of goggles which give you a future vision to see the wall before you hit it. It is always a less painful experience to say I was wrong and we have to adjust the sails a bit instead of hitting the wall and then turning. Also fast turns are more stressful than slight and peaceful turns. If you’re in the IT-industry you probably have the picture there. =D

Also having a system that gives you a bit more versatile view on the big picture of an project that is ongoing may help with an adhoc-hog. If the adhoc-hog can be given a even a glimpse on how the “small things” that he tend to drop in affects the development he might reconsider in some cases. Of course this is not possible with all the breeds. Task tracking systems are not really tools for dropping tasks in randomly so there has to be some guidelines.

—

JIRA & GreenHopper

Pros:
- Easy to use
- Enough plugins (not plenty but enough)
- Solid and simple to customize to your own needs
- Cheap for my level of usage (as I have up to 10 users-license)

Cons:
- The Agile plugin (GreenHopper) has some problems working with Sprints (Moving issues forward from sprint to an another needs duplicating issues per sprint if you want to keep time tracking on track.)
- JIRA needs the Greenhopper to be good so why is it sold separately?
- takes 500M of RAM to run. / Java application

I am slowly getting through my photofolders from 2009. Now that my backlog is only one month instead of over six months it already seems that it is really possible to get it done some day. =D

I updated my Portfolio at random.kuvat.fi but as I am not done with the 2009 yet you can assume some more later on.

The publicly available trips from the recent developing sessions are listed here:

Vilnius Part I
Vilnius Part II
Fall 2009
Winter 2009-2010 (Not finished though)

And here is the quick selection of stuff I have been working with lately:

After getting portfolios done and some stock photo things on the way I am going to lean on the digital 3d side for a while. ZBrush provides a bunch of quite interesting 3d workflows these days…

We had a small fast painting session with friends and the randomized theme for it was Cyborg Banana Interceptor. We had a 1 hour deadline and I probably used at least 45-50 minutes to the painting which was kind of effective as it normally tends to be a bit lower.

Here is what I did:

Also I have been sorting out a lot of photo-backlog lately and here is some picks from the whole set:

Another issue with Strobist stuff is to find gels that are good for ya hot shoe flash. As I live in Europe I bought my set of gels from Flashgels and can recommend that easily.

You can read a bit more about using gels from here. But in short if you are up to white balancing your shots, some of you keep it on automatic and that is not white balancing – it is more like white randomizing with a guess, you might want to keep the flash same colored as the surrounding light to minimize the oddities in color already in the shooting phase. That is actually very easy by sticking that gel with a Velcro on your flash. By having Orange (CTO) gel for Tungsten and (Window) Green gel for fluorescents you’re already covering 95% of the situations. It is worth it.

As I might have noticed I have been looking in to the Strobist stuff lately and I bought a bit of cheap hardware to get started and to know what I really want from my equipment.

While a go I bought a new flash. What was particularly different was that this time the flash was a brand called Yongnuo and it costs under 40€. Which is cheap as butter for a flash. The accurate full name is Yongnuo Manual Flash YN460. So what are the drawbacks?

The flash ain’t particularly powerful – Guide number goes somewhere around 23. (If you don’t know what the guide number means, check this.) Even though, when shooting inside, I do not use it in full power anyway.

Another thing is that there is no way to set the focal length or “zoom” for it. I should do some straight-to-wall-tests to get some kind of feeling of the beam spread.

I do not know if I can use it on my camera hot shoe and am afraid to do so. It might be that it is not safe for that.

The flash is manual only. For me this is not a drawback – rather something I am really up to. It is actually easier to use in manual than my Canon 430EX because it is dedicated for that. It is always manual and the buttons in the back are always accessible. With the Canon 430EX I need to hold buttons to get to set the manual ratios.

It might sound that my Canon 430EX, costing 6 times more, would be expensive in comparison, but you have to take in account that it is more powerful, can adjust beam and have all the automatic syncing stuff etc. the price is probably in the right spot for both of them. Currently I’m using both flashes but both are always on the manual mode. They both do have their own usages.

planetmule.com

Finally! An excellent remake of a legendary game called M.U.L.E. One of the first multiplayer games developed originally for Atari 400/800 and from where the 4-player gameplay originates from, but then also ported to C64 (and some others) on which probably most remember it from. It is dated to my birthyear and the cover graphics for the original is great in an phenomenal level. It doesn’t stop there: The game itself has one of the best tunes to my remembering from the 8-bit era.

Following text is about the 1.0 version of the game:

The java-version is developed by Blue Systems and Turborilla and is played over the internet while being an very accurate copy of the original gameplay wise. There is some slight cosmetic changes though and the graphics and sounds are totally new. What this really provides is to play the great original game again on a modern computer with friends sitting in their own homes. Which is one of the biggest downsides: This game was designed to be played together in real life. Even though this version does have the in-game chat there, it is still just a vague illusion of talking about the game in real time. Also what tells about the lack of intensity on that side is that every time there is waiting period in the game, it feels like forever even though it might be less than in the original. One downside is also the A.I: It currently cripples the gameplay by selling all the good straight away. But in general the game is very good and well done. Some slight bugs are still to be tackled but the 1.1 version is already on its way.

Skip this if you’re familiar with the game:

Although the gameplay probably attracts more the nerds of us it is still a great simplification of economy. In the game you buy M.U.L.E.’s to gather resources from your land plots and then use, sell and buy them between players and the common storage. Most of the resources have connection to the gameplay in clear ways and the value of certain good is dependent on the needs of both the common store and other players.

Wikipedia – M.U.L.E.

I have not revealed or summed up for a while what I am currently doing and I feel that it helps, and probably does not do anything else, even myself to organize my thoughts to write something down. Normally in my year cycle I am a bit depressed in the fall-time when there is a lot less sun and I actually see it like once in a two weeks… With a luck. So this year I have been concentrating to not start that many projects and slept a bit better. Although I have been buying new hardware and learning new things. Learning and fiddling around with new things always gives a small mood-boost. All together I have again noticed that I should always do one project at the time and remember to have some time off. Which I actually did by visiting Lithuania for couple of days.

But currently I’m actually on to these things:

Started an oil painting course. 1/2 of it behind and 3 practice paintings done. Mainly those concentrated only on mixing colors and the subjects were still life cups and bottles. It was good to concentrate on on simple things subject-wise first as the color mixing was the real concern. I do think this was the best thing this fall. Every session gave me really good feeling. Continues on January and possibly then we move further on from the color mixing practices. Really worth it. Probably helps in a lot of other areas of my jack-of-all-trades nature in arts.

Built a new computer. I already wrote about that. 10Gb RAM, Windows 7 etc. Really made my living easier after got it working.

Modeling 3d for a game. Modeling just real-time 3d objects might sound like eating porridge at this stage of my career but the modeling part is actually just the process of carving out an idea. Plus there is some special stuff that matters in this project. Same as in any job – there is always a lot of specifications that can not be seen by people not familiar with the particular subject. Which makes it interesting and dull at the same time. It is a paid job though so dull things goes fine in my radar. In scale from dull to misusing icicles I have been a lot to the icicle end lately and this is not so much of it really.

Creating a layout animation at work for a real-time short/demo. Really nice project otherwise but I do get stuck a lot – especially having creative locks. Layout animation is the first time in the project that I gather up the placeholder stuff on the table to a more whole set of things. That makes it possible to get stuck on anything in the process even that it sounds easy. Although I do like to go on with a process as that might make it a lot more progressive to work with in the future. The layout animation I try to go for is much like what Pixar is doing in their process. Everything is collected to be in placeholder stage together to get some kind of idea of the whole package. Then it is refined for over 3 years in different levels and pieces by hundredfold of people and the end result is the movie. The article I wrote about that creative work should be teamwork really, really, really underlines all the problems in this project. Yes I am doing it alone and no change to that in the radar. So if you are still with me – remember the talk about icicles in the last paragraph? I do not wonder at all why even this short took 8 years: The Passenger.

Learning lighting. The thing I would like to spend a lot more money and time on anyway. Loses for oil painting as for level of interest but a subject where I can be learning forehead-deep for infinity but it does not actually make it go further without the practice. I am now in the stage that I do have a light-scale set of Strobist stuff and played around a while with it but I would need to test things more in practice. Quite sure I have red enough of the theory-side for now. I am planning to take on a project later on to do portraits of some close relatives. Moving slowly with all the photography stuff anyway as I do not have the time to run for full-time photographer-job anyway. I hope this way could teach me a lot faster about 3d-lighting too – the same way as starting photography in the first place has gave me a nice boost in understanding a lot of different creative and technical things connected to composing, painting, 3d, physics, etc.

Rethinking my photo workflow (which is made possible by my new comp.) This kind of a backfired after I found out that my super-indestructible-workflow would not really manage to fly that well. I first planned to have a pipeline where all my .CR2-files would be converted to .DNG’s and after development opened as smart-layers in Photoshop and all the edits would be done non-destructively on top of the smart layer as smart filters and additional layers, saved as .PSD’s and then when needed flattened and resized according to needs in to different sizes and formats. Actually there was only one particularly weak part and it was the filesize and saving/opening-time of the .PSD. *flush* Not entirely though. I will probably move to use DNG’s as it is better and faster than my ancient way of using TIFF as an intermediate-format.

Bought some advanced texturing tutorials from Eat3D.com. Just to go it through and learn in practice. I really need some prep-up on that one. Might drop a word or two later on. Requires some undisturbed time.

Bought JoeyL Sessions Photography tutorial DVD. Already watched a bit from the start. This is totally more “serious” kind of a thing than the first one (which I did buy too and liked a lot.) Probably will drop line or two about it when I have more to say. Link

Bought 5 books on 3d, photography, drawing and animation. The Photographers survival guide almost done. I can recommend for anyone trying to push to advertisement and such business as a photographer. Written by two art-buyers so it is kind of a different perspective but a very valuable one for understanding the business. It does concentrate mainly stating couple of points: Get to know your style, be coherent with it when marketing yourself (using a outsourced designer mentioned at least 500 times), market a lot. In addition to that there is also very practical stuff and scenarios that might happen and everything. Bundled with the book is a bunch of forms to use in different stages. Very well laid out book so it really does not feel hard to carry on. Other ones bought: Ideas for the animated short, Dynamic figure drawing, Force – dynamic life drawing for animators, Digital Lighting and rendering.

Created a maxscript exporter. For a fileformat that started from Ambrose but got tweaked so far that it is not anywhere close to it anymore. Might give it out if I get it cleaned up and simplified a bit and someone is interested. Currently it pushes out a lot of raw data and is very vulnerable for nonspecific data. Got a big help from Phaser when doing it.

Had a one week trip to Saaremaa in the summer. Here is some family album pictures for ya:

Saaremaa holiday trip photos

The place is great: silent small island with a lot of places to see. Rented a cottage with a sauna in the northern part but in a weeks time we had to skip a lot of potential nice places to go. Here’s a handful of picks:

As an user of workflow that starts from Adobe Bridge and goes through Adobe Camera raw to Adobe Photoshop extended I have not really found anything useful in Google Picasa for me. Now that they introduced the Face recognition and face tagging tools in it I had to test how it performs and mainly test it in purpose to find if it succeeds to do it quick enough, with moderate user interaction and somewhat accurately. That would really help tagging photos. Although I must say that I would be a lot more interested in a tool that recognizes more subjects from a photo than just faces =D.

First problem rose when I tried to get the Picasa 3.5 as it wasn’t available from my country thorugh picasa.google.com. The site gave me Finnish speaking localization and pointed to the installer of Picasa 3.1.0. Of course I first installed it and didn’t realize until I found some unfamiliar looking menus in some tutorials when I wanted to find out how to find the face tagging in the UI. The help-link given from the menu in Picasa didn’t help me either as it was again pointing to Finnish localization which didn’t exist yet (I did change the language of the UI in Picasa to English before that already.) There was also a suggestion to change the language in the help but when I did – that didn’t change the fact that the help didn’t find anything.

So finally when I found out that I was using 3.1.0 instead of 3.5. I started to look where I could get the real deal. There was some “download-sites” that listed it but I didn’t want to get any “uncomfortable bundles” so I decided to find the package from Google. Even going to google.picasa.co.uk didn’t help as even that it did (oh, finally) have English language in there, it actually also had the 3.1.0. installer.

Luckily I had a friend outside of the country who linked me straight to the 3.5 package. Phew… Got it installed, rolling and some test pictures in finally. As I do have tens of thousands, probably hundreds of thousands of photos in my hard-drives, I didn’t (for some odd reason) want it to scan all of them to the database for this evaluation.

I took about 60GB of data which consisted of ~4000 photos. Mainly holiday and family snapshots to have quite random types of pictures and angles but including a lot of same people. Practically stuff where there is enough challenge for the heuristics when at the same time the help for tagging would be very welcome in that kind of data. I also had doubles of many of the pictures in one folder.

So how it goes…

First of all: Do not try to tag people at the same time as the Picasa is listing and analyzing the photos. The unnamed list grows and shrinks wildly back and forth while you are trying to edit it and sometimes photos go inactive so that you are not able to select them and realize after a while that those that couldn’t be selected a while a go moved away from unnamed list and so on. It is a mess at that stage and it did get frustrating. At least in the Windows-version I was testing. (Friend of mine reported that Mac-version didn’t seem to have this kind of problems.)

While the Picasa does learn to find suggestions that match quite well it does mix kids faces quite easily. Also note that side profiles seem to be very hard to match plus in great percentage they do not get recognized as faces at all. Clear faces from the front and random noisy blankets in the background gets in most easily. One thing that probably everyone had pointed out in their reviews already is that having photos from public places is kind a fun as there is a lot of unknown faces you wouldn’t notice otherwise. Also they are a pain because you need to ignore the people from the list too. I even had some very old black & white group photos on a wall of a restaurant in some photo and all of them were included in the list =D.

First impression is that it is fun and feeling very quick. Although I did find that it is still quite far stretched to use it in generic situations. I ended up using a lot of time tagging people one by one and I suppose it was something around 40% of the pictures came in as suggestions and rest should be manually checked. Also it did mix up some people quite often but that might have something to do with being relatives . A lot of faces were left out even that they seemed clear to me and 100% duplicates of an identified picture was never identified by the automation. Even funnier I found out that the system could NOT use more than one core of my CPU. So the whole system was running on maximum of 12,5% CPU and was of course a bit slow because of that. Considering I have high-end machine exactly because I want to keep photo-editing swift this didn’t make me happy either.

So to re-cap…

The best things in it:

The UI is quite smooth to use and the system actually does recognize a lot of faces and even learns when constantly reminded. Like kids normally do. But this kid needs a bit of time to develop further still. I’d recommend it though for different scope that I have. Someone that has a lot of pictures of friends in very monotonous situations might find it OK to tag them this way.

The greatest drawbacks were:

1) Obviously the lousy job on recognizing profiles of faces even as faces affected a lot
2) The system actually left out a lot of faces. Even some that were a lot more clearly faces than most of the automatically found ones. And to point that out here’s raw-data examples of pictures including my daughter Fiona that do not include faces according to the system (After I had identified 93 pictures of her, many of them from the same situations):

3) The heuristics didn’t take in account that there could be duplicates of the pictures AND after I had identified the other it never accepted the duplicate of that picture as a same person. So all of the duplicates has to be went through manually even that they’re 100% the same picture.
4) The system doesn’t seem to care about multithreading. Which is a shame.

I finally figured out how to get the computer running. There was this wonderful led-screen that should show me information of the motherboard status. It wasn’t showing anything so I removed it. The computer started working after that. =D

After all I was originally just fixing the cooling problem of my chipset. It did get down to 65-70 degrees celsius from the 82 when I added some silver and an additional small fan. But I still think that is kind of high. Even though the computer has not crashed after that so I suppose the change was enough. If there will be more I’ll add new cooling unit or get the board swapped. Perfect example of a very small mistake in manufacturing that can affect everything in such a critical way. Switching the motherboard in the first place would be the easiest choice I suppose but I’m too stubborn to do it after all the assembling and fixing. =D