[MOC] Honda NSX (NA1) – 42143 Alternate Build

[brunojj1]

6 hours ago, Mestari said:

I always wander how smart you have to be to be able to build such masterpieces. Hats off!

It's not so much about being smart, but more about dedication, bravery and patience. Did you ever ask yourself: why does nobody on earth do this? And give yourself the one answer: I'll have to do it myself! Necessity is the mother of invention. Lack of somthing leads to more creativity with the limited material you have.

[Romanista]

On 9/18/2023 at 5:27 PM, Migeta22 said:

Wow, you've captured the detail and proportions of the NSX very well.  For a first MOC, this is one of the cleanest B Models I've seen.  Pease consider making instructions because I will definitely grab another 42143 set just to build this.  Amazing job!  I rarely post on here but had to share my thoughts. Regardless of your choice,  Can't wait to see what you'll build next.

On 9/18/2023 at 10:54 PM, R0Sch said:

Wow! Now this is what I call a proper B-model for a 1:8 supercar. Even the 5-spoke rims fit the NSX well, plus the proportions and functions are phenomenal. 

On 9/19/2023 at 8:44 AM, Mestari said:

I always wander how smart you have to be to be able to build such masterpieces. Hats off!

19 hours ago, MinusAndy said:

Outstanding! And a way cooler car than a Ferrari!

Thanks, it's really amazing! i am thrilled to see such a great feedback!

On 9/18/2023 at 10:54 PM, R0Sch said:

How many parts were left over from this @Romanista?

I don't know the exact count, I suppose 200-300 parts. Majority of them are useless system parts used for Ferrari headlights. Technic parts were almost fully utilized.

19 hours ago, brunojj1 said:

It's not so much about being smart, but more about dedication, bravery and patience. Did you ever ask yourself: why does nobody on earth do this? And give yourself the one answer: I'll have to do it myself! Necessity is the mother of invention. Lack of somthing leads to more creativity with the limited material you have.

Maybe it will sound too pathetic in relation to Lego   But I agree with Bruno - it is about dedication, patience... and readiness to spend unlimited time for every aspect of the build.

Anyway, I can't say that my build is masterpiece  There are a lot of compromise and not optimal solutions because of parts limitation. 

[gyenesvi]

Wow, I have just seen this, it looks fantastic, really clean lines, nice part usages, and glad you filled it with interesting functions as well. Great B model!

7 hours ago, Romanista said:

and readiness to spend unlimited time for every aspect of the build.

I can really relate to that one :)

[930 Turbo]

Love it man! Congrats on the awesome build! now im thinking i should make public my 1/8th porsche 906...

[Romanista]

On 9/20/2023 at 5:53 PM, gyenesvi said:

Wow, I have just seen this, it looks fantastic, really clean lines, nice part usages, and glad you filled it with interesting functions as well. Great B model!

 

On 9/25/2023 at 7:40 PM, 930 Turbo said:

Love it man! Congrats on the awesome build! now im thinking i should make public my 1/8th porsche 906...

Great thanks! 

I have a question regarding technic parts in Studio. Sorry if it should be raised in a specific thread.

Why does the main palette (or custom one - imported from the official set) not contain such parts like 65766 (wheel hub), 79717 (shock absorber) and x71 (rubber)? 

I can admit some problems with printed parts, but what is the issue with commonly used parts...

Edited September 27, 2023 by Romanista

[gyenesvi]

39 minutes ago, Romanista said:

Why does the main palette (or custom one - imported from the official set) not contain such parts like 65766 (wheel hub), 79717 (shock absorber) and x71 (rubber)? 

Well, because the developers don't care? I don't know, but following up with new parts should be pretty top priority I agree. Many people also raised this. There are parts that are years old and have not yet made it into the palette, even ones that are available from elsewhere (Philo's page). Though the wheel hub is in there (under Technic - Steering). Studio has a ton of small irritating bugs and weird behaviors about connecting / moving / snapping parts. I have filed a couple of bug reports - nothing happened so far..

At the same time they have time to start adding experimental features like auto-generation of build sequences, something that is super difficult to achieve properly. I believe they should focus on the basics first. Also the graphics rendering could be improved a lot. Making animations is pretty impossible if you don't have a powerhouse at home. But this graphics workload is nothing compared to today's games, which can run in real time.. So I think it's just weak implementation..

[grum64]

When I first saw this I thought I’d missed hearing about a new official Lego release. I’d never have known it was an alternate build, in my opinion it’s so much better than the Daytona. Great work.

[Romanista]

Good news! I've just finished the instruction (generated from Studio).

I need 1-2 weeks for beta-test build by another person. I hope it wouldn't require a lot of fixes. 

On 9/28/2023 at 9:08 PM, grum64 said:

When I first saw this I thought I’d missed hearing about a new official Lego release. I’d never have known it was an alternate build, in my opinion it’s so much better than the Daytona. Great work.

Thank you for your kind words!

[GirchyGirchy]

On 10/15/2023 at 1:50 PM, Romanista said:

Good news! I've just finished the instruction (generated from Studio).

I need 1-2 weeks for beta-test build by another person. I hope it wouldn't require a lot of fixes.

Hooray, thanks a bunch!  We appreciate your work on this.  It really does look great.

I wish I had the time to make one of these, I have something I'd love to make.  Maybe next year.

[Romanista]

Hi everyone, 

My first MOC on Rebrickable.com is complete and approved. In addition to the instructions, I've also added 15 recommended modifications of the build and the sticker sheet. 

I would really appreciate any support from Eurobricks community. 

P.S. IMHO, Studio software is a total nightmare. 

Edited November 1, 2023 by Romanista

[Romanista]

There is a naive question to experienced MOC designers   Is any legal and not irritating way how to promote the MOC on Rebrickable?

Despite the fact that I see some interest to the model (which is very pleasant for me), I think the only way how users observe new MOCs is the starting page with 'new' and 'hot' MOCs. Without having an army of followers, every MOC becomes overlooked in just a few days, isn't it? 

[Alex Ilea]

6 hours ago, Romanista said:

There is a naive question to experienced MOC designers   Is any legal and not irritating way how to promote the MOC on Rebrickable?

Despite the fact that I see some interest to the model (which is very pleasant for me), I think the only way how users observe new MOCs is the starting page with 'new' and 'hot' MOCs. Without having an army of followers, every MOC becomes overlooked in just a few days, isn't it? 

You need social media presence or just rely on Rebrickable recommending it to people...

[Jay Psi]

On 11/2/2023 at 8:15 AM, Romanista said:

There is a naive question to experienced MOC designers   Is any legal and not irritating way how to promote the MOC on Rebrickable?

Despite the fact that I see some interest to the model (which is very pleasant for me), I think the only way how users observe new MOCs is the starting page with 'new' and 'hot' MOCs. Without having an army of followers, every MOC becomes overlooked in just a few days, isn't it? 

I've seen some calls for Rebrickable staff to review your MOC, that would certainly help, but directly asking is frowned upon.

On topic, the modifications suggestions are a great idea, not only for those who have the parts to do so, but for less experienced builders to see the difference part availability and choice makes.

[Timorzelorzworz]

On 9/27/2023 at 3:35 PM, gyenesvi said:

Also the graphics rendering could be improved a lot. Making animations is pretty impossible if you don't have a powerhouse at home.

To my knowledge stud.io is using some version of Unity Game Engine for rendering, which is improved for Games instead CAD design. The performance is indeed not that good but a "Powerhouse" is not needed. I do all my models and renderings only with business PC (Graphics Chip onboard) without dedicated Graphics Card. For Animation you can do it frame by frame

[gyenesvi]

2 hours ago, Timorzelorzworz said:

I do all my models and renderings only with business PC (Graphics Chip onboard) without dedicated Graphics Card. For Animation you can do it frame by frame

And how long does it take for you to do your animations? I am not willing to wait hours or even days for them to get done. Last time I tried on my laptop it would have taken so much. If I compare that to the real-time performance of games, it seems completely out of proportion. I don't need photo-realistic ray-traced rendering for simple animations..

Also, are you using Studio for your models? I always thought you are using LDD, because that has the ability to animate hinge sequences and add pneumatic hoses, unlike Studio, as far as I know..

[Ryokeen]

First off, rly nice model i like it and the added modifications for ppl who have some more parts are just amazing. Good job there

@gyenesvi

There is just a slight difference between games and a render from a cad software or studIO. Games render with tricks, a lot of tricks to maintain high fps, where as studIO/LDD povray just bruteforce computes an image. Rendered images in studIO or animations(which are just a series of images) are made with raytracing/pathtracing which gives more realistic calculations and is a technique that in itself is not easy to optimize.
The thing is just that without proper data on surfaces, light, environment, even pathtracing looks like some image from a early 2000s game. That's where studIO is laking.
So yeah it does take a long time and could be optimized better but then again, studIO doesn't use lowpoly meshes for a renderere so that's just a lot of computations to be done.
If you're interested in that topic pm me, any more explanation would be way to offtopic.

[Timorzelorzworz]

10 hours ago, gyenesvi said:

Also, are you using Studio for your models?

I prefere stud.io as all-in-one solution for doing models, instructions and raytraced images but only on CPU-level without dedicated graphics card for that PC. A photorealistic image takes up to 45 minutes and requires some time and patience, but when you have nothing else to do on that PC it is ok to give them 100% CPU for doing that stuff. 360° animations with atleast 96 individual frames takes up to 3 days of none-stop rendering.

[gyenesvi]

4 minutes ago, Timorzelorzworz said:

A photorealistic image takes up to 45 minutes and requires some time and patience, but when you have nothing else to do on that PC it is ok to give them 100% CPU for doing that stuff. 360° animations with atleast 96 individual frames takes up to 3 days of none-stop rendering.

Yeah, that's what I calculated too, and just gave up there..

@Ryokeen, yes I know all that, I used to work in the field of computer graphics, that is why I believe it could be better. Basically, I'd be fine with non-raytraced, non-photorealistic renders for animations. Something similar to what is currently available in the main build view of Studio. Why can't I make an animation with that level of realism? Maybe even better lighting/shadow casting can be done, it would be okay to have 1 fps rendering instead of real-time, that could still give us reasonable times for short animations. But 1 min / frame or even 45 mins / frame is way too much and unnecessary I think.

7 hours ago, Ryokeen said:

studIO doesn't use lowpoly meshes for a renderere so that's just a lot of computations to be done.

To me it seems it can use low poly meshes, you can set the detail level for the build view and if you set it low, it gets blocky.

But anyways, sorry for the off-topic, agree that we should not go on here..

[Timorzelorzworz]

6 hours ago, Ryokeen said:

So yeah it does take a long time and could be optimized better but then again, studIO doesn't use lowpoly meshes for a renderere so that's just a lot of computations to be done.

stud.io uses high-res primitives from the LDraw path when there are high-res primitives available for a specific part. Realtime rendering (not raytracing) of Lego bricks is today not that hardware-hungry like many people suppose. Problem with the renderer is not the GPU performance, but how the program and CPU handles the data for the GPU to render. Lets assume you have 1000 parts in the scene and lets assume that 1 part consists of 100 polygons. For sure some parts uses many more polygons, but other parts also uses less. For example a 1x1 Tile (3070b) has way less polygons. Following this example the GPU has to render only 100.000 polygons in the scene, keeping shading or outlines not in mind. For todays hardware, this is not a workload that people are required to have a hardcore gaming-PC and this amount of meshes is done by a onboard GPU on the fly incidentally.

Sorry for going off-topic. Btw saw this forum post for the first time. Its a friendly and informative discussion here.