Michael Mirn (michaelmirn)

94 replies · posted

Mirn's Book of Renders

Greeting, 3d-world stranger. In this humble place I'll share some of my works, thoughts, hopes and achievements.
All kinds of feedback -- specially critical -- are always welcome.

  • 1. Dali.


    2. Van Gogh.


    3. Durer.


    4. That's a hard one. Rene Magritt, "The Lovers".

  • The first motorcycle, Reitwagen. Step one:


    • Oh, that's a cool thing to model!

      (Just be aware that that is not an orthographic picture, but I'm sure you got that covered...)

      Looking forward to see more.

    • Yes, spikeyxxx, I've noticed that but thank you for the warning)
      Sometimes I wish I had a proper blueprint for that thing -- but that would be too boring I suppose. No reverse engineering and all that guessing.

  • It is a great model indeed. Frame, looks like a dual toy horse so far:

  • Step 3: Frame and side wheels:

    beefkeef thanks, Keith. I am trying to be a dedicated person and not a slacker so I'll do my best to keep it daily updated)

    That model is really fun to make and it is going well so far. But I wonder -- how would someone approach that or shape this? Do I really have to add that curve? Should I bypass that edge?
    I like all these little decisions that come along and form the whole picture eventually.  

  • What a beast that retro bike really is:

  • Part 4:


    Gonna fix that engine soon.

  • Part 5: Tried to retopo that  thing today (it was  a good practice and the very bad result). It is always good when overconfidence meets the lack of experience.
    Frame and wheels are both almost done. The engine part comes next (or the seat, haven't decided it yet).

    • It's really starting to take shape now michaelmirn! Keep it up :-)

    • Wait when you say retopo, Are you saying that you sculpted that entire thing? Or are you saying that you just wanted to clean up the topology to make a lo-poly version compared to the high-poly or subdivision surface version?

    • Nah, no sculpting, that would be crazy, blanchsb 

      I wasn't satisfied with the initial grid so I tried to change it for the better. Got rid of these holes - they were nice to have but completely useless. And tried to separate the original form in parts so I could put an edge loop over each piece and then connect them. Didn't work too well)

      Still it is much cleaner now so it wasn't a total waste of time I hope.

    • Gotcha. That is good practice to try retopo on something like this. Looking great so far.

      I would post before and after wires (without SubD active) when showcasing retopo to see how things flow if I wanted some specific feedback on that portion.

      You've definitelygot an eye for detail, I'll give you that!

    • Thanks, @blanchsb
      Unfortunatly there is nothing solid to post yet) The first take on that frame was a flawed one. You can compare the current state of the frame  (pic #5, #6) and, let's say, part 4 and see it for yourself)

    • Yeah I am honestly having a hard time telling which which is which haha.

      The upper part shows a little better detail. Do you have Cavity turned on for your viewport? Maybe bump it up a little, perhaps do Screen Space and World Space.
      I personally like a screen space set to 1.5 for ridges and valleys on my own stuff but that is just me.
      The lower model of the two with a faint blue tint is hard to read any fine details on. I wonder what the improvement in poly count is between the two.

    • Yes, @blanchsb the cavity is on.
      Here is how it looks like "raw":

      Basically, I am trying to find a better way to represent my work. Showing white on white is definitely makes it hard to read)

  • Part 6: some subtle tweaks and gearing.
    These belts aren't stitched properly but I love them still. I saw some photos without them too. Someone must have been lazy.

  • Part 7: a little bit of details here and there and the frame is finally done. Engine is going to be much more interesting.



  • Part 8:

    I was badly burned in the sun and fell out of the process for a couple of days) It is ok now so back to the drill.

    The central part of the engine.
    Unfortunately, even for the central part there are several models for performance and parts, not to mention museum exhibits and exhibits with varying degrees of equipment, sometimes made just for beauty, but still taking time to think through the mechanism.

    That upper-middle chamber (cube/pyramid/cylinder/half-sphere) part was really great to make) Managed to model it in one piece.

    Almost feel myself as a mechanical detective.

    • michaelmirn You'd fit right in on the Heavy Equipment Live stream. We are all going through similar "growing pains" lol/


      This is really looking great. What's your poly count up to now?

      Not sure you struggle looking at all white objects while working. I like to assign my individual objects a different color in the Object Properties Viewport display when I am working just so I can focus better. I don't go for exactly the real colors but something quickly similar helps a lot for me when I am working multiple multiple pieces. It's easy to switch the main viewport shading to go from Object Color to Single Color that way. You can even

    • Yes, I know about that way) I prefer to create a separate file for each (reasonably complex) part and then combine them all in a master file.
      Like Reitwagen_wheels, Reitwagen_engine etc.

      By "detective" I mean that there are so many variations of that mechanism that have been built since 1885. And I do not own a copy of that bike so I have to rely on these pictures.
      Sometimes it is just too confusing to separate mistakes and "interpretations".

      I have to say that this project is big -- for me at least. Gonna do something way smaller after this. Maybe it is gonna be that ball Wilson from the Outcast movie. Just to flex some muscles.
      And then -- another big one.

      Oh, and the polycount (no engine involved):

      Looks too fat I think. Trying to reduce the number by cutting off all the parts of the bolts I can spare.

    • 2 Million Faces it looks like. I think it is really looking good man. Yeah that is a bid project.

      Yeah I can totally relate to work scope being perhaps too big. I think my loaders arms are a big project and I am only doing one section of our heavy equipment. I'm still fairly new and it is enough to be challenging yet fun.

    • Well, my polycount is a bit high but the most difficult part for me is the design of the engine cause it is hidden below that saddle))

      Here is an example of incredible and hard work -- it took 5 years to finish this project. A lot of people contributed to that (archaeologists, linguists, photographers etc). I wish I could work on that level soon.

      https://skfb.ly/6SXZR

      You can descend the stairs and dive into the mystery of the past -- step by step -- century by the century.
      The music and the atmosphere of that tomb is so beautiful and mesmerizing . And yet -- 5.5k triangles)

  • Part 9:

    It grows and grows and grows and grows. But the end is palpable.
    A few parts left: 2 units of "really, is that a thing?" and 1 unit of "hmm, what it might be?" Plus some minors "what the hell".

    • Its certainly a very strange but intriguing engine. Great modelling and attention to detail! 

    • Thanks, frikkr

      It looks way more accurate and compact in museum (but that is not an original design):


      The original looks like this:

      The main purpose of that engine was to demonstrate it's ability to work. The total "travelling distance" of the very first model never made it past 2 km (maybe because it got burned down with the garage where it was kept).

      But I really like how savvy that engine works (crude but savvy). For the instance: if you push that front lever forward it lowers the brakes and removes supportive wheel from the engine belt. If you pull it back then you release the brakes and supportive wheel squeezes the belt so you can move.
      That bike is like a Cro-Magnon for the modern Ducati. And it is so great to examine it)

    • I've been checking out the thread every now and then and I have to say, this is a nice undertaking and you're doing a great job at it. I'm enjoying the details and looking forward to how the project pans out. Good luck :)

    • Thanks a million, drgnclw !


  • Part 10: the engine is done. The last part that is left -- brakes. Gonna finish it tomorrow, then do some beauty shots and start the unwrapping.

    Can't say I am totally satisfied with that result. There are two major flaws within its' design and I can not fix it (or reinvent in some way) because I do not own a blueprint. So this is not a historical and 100% genuine model, more like a decorative one. Bugs me a bit)
    It's been a good practice tho.

    • Hi michaelmirn I think this is looking great! Can tell you've put some time in here getting all that detail - I think it's really paying off. 


      Out of interest, what parts are you not satisfied with?

    • Thanks, beefkeef

      These:

      Not heartbroken about that little door but it just doesn't seem sweet. I know it is placed below the saddle and completely hidden but still.
      And that pole (for the lack of proper term, sorry). It goes up and down the road as a starter I believe. But the original design is completely messed up here, it is just a counterfeit mechanism.

      There is a piece I like tho) And I know that no one will look for this part:
       

  • Ok, the model is done and it looks like this:

    How am I suppose to texture that whole thing? Well, gonna find out that soon.

    • Just hit the "Fill" button, lol

    • Yeah, the temptation is high!

    • As most of it is metal, I wouldn't unwrap it (you mentioned unwrapping this, yesterday, but most of it can be done better procedurally..),maybe only the saddle...

    • I think I'll use Substance for this one, spikeyxxx

      I really think that texturing is a weak spot in Blender and I really wish that Ton Roosendaal would develop the base of the Blender a little bit more and put a little less effort into some side features. But I judge as a customer here. He sure has his reasons to do things his way. Maybe he tries to build up some sort of a freeware Pixar media conglomerate, idk.

      Anyway, until Blender struggles to become the Jack of all trades I can always use more suitable side programs to fill the gap.

    • Substance is actually on sale on steam. I'm thinking of biting. Not that I will give up on blender but it would be nice to have a license for it.

    • michaelmirn I just meant that you don't need to unwrap most of this, whatever software you use for texturing;)

    • I know that) unwrapping all these bolts would be crazy.

      Anyway, thank you, spikeyxxx
      Your advice is always welcome. I have the feeling that you know that program really well and I am barely a beginner.

    • The latest build of Substance comes with an auto unwrap feature that is selectable when you start a new project. I've no idea if it works well or not as I have yet to use it , but it may be a huge time saver for you if it does.... 

    • Yeah if Spikey says something I may know something about, I be like: cool, that is good to know! Because more often than not he knows how to do something way better than I can think of. That's what working on blender for 10 years get you haha

    • Somehow it always feels a bit less overwhelming once the larger bits are done (or at least to me it does!) so I would probably start those first. 

      This is going to look great once it's textured, can't wait to see it!

    • Thanks a lot, beefkeef

      I've just finished the procedural version of that model for the practice sake. So if you want to see it -- here it is:


      I am going to complete that with the Substance version for sure. I just need a couple more days for it)
       

  • In 1882, the Italian inventor Enrico Bernardi (Enrico Bernardi) presented to the public a gasoline engine used to power a sewing machine. Two years later, the engine was installed on a children's tricycle. Thus the Motrice Pia was born.
    In 1881, Lucius Copeland assembled the eponymous "Copeland" mechanism — a two-wheeled Bicycle with a steam engine.
    Almost twenty years earlier than Daimler, in 1867 and 1869, Louis-Guillem Perrault rolled out the Michaux-Perreaux and Roper, a pair of two
    -wheeled motorcycles with steam engines on the frame.
    But the first motorcycle is considered a product of Daimler-Maybach. Why? According to the set of features:
    — single-track unit;
    equipped with two wheels (formally with four, but two side wheels — only for support);
    — most importantly: with an internal combustion engine.

    Yes, steam engines appeared earlier, but they turned out to be a dead end. It is the internal combustion engines that have brought modern technology to life. Therefore, what was before the internal combustion engine –
     simply does not count.

    Gottlieb Daimler and Wilhelm Maybach met in 1872 at the Otto und Langen Gazmotoren Fabrik Deutz. Gottlieb worked as a technical Director, and Maybach headed the design Bureau. For about ten years, the engineers actively collaborated with each other, until in 1882 they left the factory and opened their own workshop for the production of motors (Daimler was a co-owner of the patent for the production of Otto engines).
    Joining forces with Maybach, the inventors improved the design of the engine in two years, creating a valve chamber and an ignition system from the incandescent tube.

    In 1884, the Reitwagen appeared. On November 10, 1885, a demonstration of the new device took place.
    On the first day of the test (a trip from Kannstadt to Untertorkheim), the seat of the "self-driving riding cart" was repeatedly ignited due to excessive heating of the combustion chamber located directly under the driver, and the total length of the route did not exceed three kilometers.
    However, the original design has proven its stability and ability to move in space. The design was patented and a year later Daimler assembled his first four-wheel car (almost simultaneously with Benz, who assembled and patented the tricycle in 1887).

    For several years, the Reitwagen remained in the garage as a test stand, where it burned down in 1903 during the Cannstadt fire, which destroyed the Daimler and Maybach workshops.
    In 1985, in honor of the centenary of the invention, replicas of the original Reitwagen were created. Replicas differ in layout and individual technical solutions (for example, the number of gears), are not decorative and are on the move.

    The General characteristics of products include:
    Engine power: 0.5 horsepower / 600 rpm.
    Engine capacity: 264 cm3;
    Fuel capacity: up to 2 liters;
    Maximum installed speed: 12 km / h;
    Number of gears: 2 (5 km / h in first gear, 11 km/h in second).
    Weight: 90 kg.


    ___
    It is done. I really appreciate all the feedback from you guys. My Reitwagen has some flaws for sure but I learn and I love every little bit of it.
    Gonna start a new one really soon)

    • I want to ride it now that it is finished. Great job michaelmirn ! Way to stick with it through the entire process. That actually means a lot to complete such an undertaking. And it will go great on your personal portfolio.

    • Thank you, blanchsb ) Yes, there were a few moments like "What am I getting myself into?" xD
      10-15 more projects like this and my portfolio would be a solid one.

      I am going to proceed with that series eventually so if you have any preferences, please, let me know. I am thinking about:

      The first photo camera (I won't go Daguerre's tho, it is going to be a Polaroid);
      The first machine gun;
      Printing press (or typewriter).
      Or name your own thing.

    • Really awesome work here Michael! I agree with Shawn, it's really an accomplishment to be able to finish such a complex project - so well done to you! 


      I'm looking forward to seeing what you create next and if I could pick one of your ideas it would definitely be the printing press! I'm slightly biased here because I operate printing presses in my day job! 

    • Great work , I'm really impressed with how this turned out!

      Looking through your list of potential projects , I would like to offer a very radical idea : How about you use the very first version of Blender to make the very first default cube :).... Just a thought..... 

    • It is settled then, beefkeef I'll do a press. I do love books so that is a pleasant choice.

      frikkr, appreciate that!
      Yes, I've seen that or very similar challenge some time ago. It is a fun challenge but I think I need to know the medium a bit deeper and better to perform such tricks)  There are so much to learn: Substance, DaVinci, Rigging and Proper texturing in Blender. Right now I am focused on honing my skills more than performing some tricks. But that's a fun challenge for sure)

    • I’ll add to the list of ideas: 

      For guns: how about the Browning Automatic Rifle (a truly iconic gun)


      The first turbine engine.


      Or perhaps one of Nikola Tesla’s inventions? The death ray/peace ray. The first hydro-electric power generator, etc....

    • Thanks, blanchsb)

      These are great ideas and creating something from Nikola Tesla is definitely in the plans. But I don't want to waste ideas before I can implement them on a high-quality level. For example, Tesla must be made with special effects of electricity, all that lightning and humming. I will think over the concept and will definitely do it.

      In general, I have three directions: the first of its kind inventions. Creatures from myths and legends; the most vivid images from cinema. As I grow, I choose one of the items in any of the three categories and do it. 

      Although I certainly jumped over a couple of steps with Reitwagen, but I'm glad that everything ended well in the end xD


    • Great job, man! :D

    • drgnclw

      Thanks! I like it too. Even with the fact that this is not a 100% replica and you can't build it up from that visualization (that was the plan).

  • Ok, getting ready for the #2.

    I've decided to pick Gutenberg's press (obviously) and this is how it looks like nowadays (this is just to give you a grasp of the thing, I'll try to find a really old one, preferably the original):

    • That is a cool video. Quite nostalgic!

      I sure hope you include the ink pads and the other tools.


      I guess a stretch would be to model the whole room but maybe that is too big of a scope.


      Ultimately you get to define the scope. I’m excited to see what comes of this

    • Not just ink pads. I am thinking about animating that whole cycle)

      As for the scope -- I think that press needs some kind of a base for sure, but that's it unless I want it to get lost in its surroundings.

    • Yeah that is probably a good idea. These projects remind me of a comment that J.R.R. Tolkien made when he was working his big project LOTR.  He said it was consuming him.

      Nothing like diving head first and drowning in it! I often feel this way about any project that I start. I eventually learn how to breathe underwater with each one though, so I stop drowning and gasping for air or something else that is distracting.

    • >>I eventually learn how to breathe underwater with each one though
      Something I need to learn.

      I have on my hands (in Blender) right now: an animation / 7 deadly sins statuettes / some learning flows and this Guten-press going on.
      But it's ok. Just my modus operandi. I even read 4-5 books simultaneously.

    • Nice! I'm looking forward to seeing this come to life 👍

  • Step 1:

    Fortunately, I wasn't able to find a solid 1440s press so I'll try to model it as close to the original as possible but with numerous interpretations and "artistic freedoms".
    For the instance: it feels right to me to give that press a descent standing place. It looks almost like a temple entrance or a sacred monument and requires some fundamental base to be placed upon. These first books were like relics after all.

    Plus here is a good one to dive deep into the history of that printing process:

    • OMG! It's Stephen Fry from "A bit of Fry and Laurie"!

      At first when I saw the length of this video I thought: that's a tad long, but now...

      I'll watch it tomorrow morning with some coffee and cookies;)

    • I had chance to watch that Stephen Fry film earlier today - that was interesting to see the craftsman build it using the old fashioned tools! Big respect to how he chiseled the thread on the spindle, he must be a patient guy I reckon.  :-)

  • Step 2:

    I have a feeling that this model is more historical than artistic)
    Well, sometimes you have to do plain work.

    • You will have to learn procedural shading in Blender for this one, with all those wooden parts (wood is 3 dimensional and not just a surface, you won't be able to use image textures for that...think of the wooden 'screw' for instance...)..

      But great project! Looking forward to seeing your progress!

    • Well, I don't use image textures. I use SP ).
      I tried to do some procedural surfaces in Blender too and I feel that I am getting there (pretty slowly I have to add). But it is a question of quality of life. It is so convenient to create all the textures I need in SP. I almost force myself to do that in Blender.

      Thanks, spikeyxxx I like that series too. Gonna do the first atomic bomb next, the Little Boy, the one that was dropped on Hiroshima .
      It is going to be a good contrast -- that print press and an atomic bomb. An explosion of wisdom and knowledge + an explosion of cruelty and literally hell. Both done by human kind.

    • Did I misunderstand how SP works? I thought it created maps (image textures). It creates them procedurally, but they are still 2D textures in the end. Or can you, without unwrapping, directly texture your model with SP?

      Edit: Sorry I was thinking about Substance Designer...But still can you paint without unwrapping?

    • You have to unwrap, yes.
      But it is ok with me cause I can export my model to wherever I want with textured maps of all kinds I need. I don't want to create a model just for the fancy screenshot via f12, I think it is kinda useless and locked within the program. Plus it is a good practice. For the instance: there are 64 textures in Reitwagen and it was really good to play with them.

      I wish there were more tightly packed courses that cover the whole process of creating an asset, not just the initial stage.
      Something like the Chest one + various complexity. I know that UVs might seem boring and dull but that's a part of the drill.

    • What I meant is that when you unwrap, you get 2 dimensional textures and this would not be doable (this specifically applies to wood):

      But maybe this is possible with SP?

    • As for portability, you could texture it using a procedural shader and then bake that into textures...

    • Thanks, spikeyxxx 

      For some reason I just avoid Blender texturing. I don't know why. I force myself into it but nothing happens. Maybe I just block it like some bad memories xD
      But I really love sculpting / modeling and animating tools.

      As for the model -- I hope that everything would turn out all right eventually) But I'll keep an eye on these naughty woods.

    • Yes, the main thing is you get a satisfying result, whatever software or method you use!

  • Step 3: it's almost done. Have to add these ink-pads and some letters for the set.
    Can't wait for the texturing part. The model has wood, metal, ink, paper, rust + so many scratches and spots.

    • It might be cool to research the first letters that were printed. Give a homage to the first page ever. but......that may be a lot more work lol.

    • Yes, blanchsb, I was thinking about that) I know that Gothic set and I know that Bible page. But that would be a ton of work for sure. Plus it's kinda repetitive and doesn't teach a thing.

      So I'd go for "Gutenberg press modelled by Michael Mirn" instead) this whole thing is a homage after all!