港產特技大作 - 機械俠 Poster

[h2o]

Metal 既特性系有 BRDF, anisotropic or blurry reflection, color reflection 呢幾樣野都可以 base on Blinn, Phong shader 再加上去; 或者分開唔同 render pass, compositing 到 cheat 都得

[soSmall]

the metals seem not realistic enough,...still using blinn and phong based shader?

You know! For this project, around 250 FX-shot will be finish for FX&Render&Comp part within 2 month. And just use totally 10 man power to do FX&Render&Comp part . Actuality, we know that a lof of FX shot need to be impoved it by us. But we should be balance the quality and the schedule. Remember, a billion film production just for slogan. I think we never get money more than 5% in a billion dollar budget. Anyway,I wish we can improve more with next project. Add oil ! : )

agree..we are trying to do the ultimate quality by limited manpowers and under a tough production pipeline. as a employee, we pay attention to finish a huge project with a fair qaulity that will be accepted by film director.

and...johngalliano
what sorts of shader are realistic enough in your horizon?

Agree as well.

I don't get political points for being an idealist.
I'm sure they are doing the best they can with what they have.

[johngalliano]

and...johngalliano
what sorts of shader are realistic enough in your horizon?

Well, nothing wrong to use traditional shaders, but if there's time, i will try those microfacet-based shaders for the specular, with use of HDRI.
And, having a brief google search, i saw ppl said the mia_material of MR is good for realistic metal, with very high IOR like 10-50.
After all, I think the trend of cg moves to the physically based methods. But I know there's little/no time for research in "C 記".

[EricLo]

and...johngalliano
what sorts of shader are realistic enough in your horizon?

Well, nothing wrong to use traditional shaders, but if there's time, i will try those microfacet-based shaders for the specular, with use of HDRI.
And, having a brief google search, i saw ppl said the mia_material of MR is good for realistic metal, with very high IOR like 10-50.
After all, I think the trend of cg moves to the physically based methods. But I know there's little/no time for research in "C 記".

Frist of all, We render realistic metal which don't need using HDRI in every case. Many rendering & compositing method can cheat to like as realistic image. And if u have not suitable to use HDRI Image.The result sure is not your imaginable. I think profressional can decide what kind of method is best within limited schedule. And i think MIA_material in MR is not for best. Sometime, We use simple V-ray shader or FinalRender shader which also easy to setup better than MIA_material in MR. Important point, We never only trust internet information from search engine. We must prove right direction by different infomation way(Book/Internet/People). One day, If u have become a professional cg artist in HK. You will be understand what is that meaning. Research is not simple word. I saw many people confused to use this word of the meaning. I think that the experience also for part of the "Research"! Let you see!

[stantang]

and...johngalliano
what sorts of shader are realistic enough in your horizon?

Well, nothing wrong to use traditional shaders, but if there's time, i will try those microfacet-based shaders for the specular, with use of HDRI.
And, having a brief google search, i saw ppl said the mia_material of MR is good for realistic metal, with very high IOR like 10-50.
After all, I think the trend of cg moves to the physically based methods. But I know there's little/no time for research in "C 記".

Don't you think the team would already know that if you can search it out with a brief google search?....

Knowing what's the right/best way to do certain things doesn't mean you can always do that in reality, there's many other factors which I think anyone has production (esp. HK) experience should understand.

To be fair, from what I've seen, I don't think the shots in the trailer are final yet, correct?

[johngalliano]

guys, i am no outsider cause i've been a hk cg worker since 2001. Not always employed from start to end due to my study break (enjoying my 2nd study break, now), but have certainly spent time with some of u guys in "C".

Anyways, schedules are always tight, not an excuse for a studio, the audience won't care about it. U can only blame john chu on that.

If you've gone through the research process of developing more realistic looking metals and still find that the most feasible way to do it is the traditional way, that's fine. But i think there is just no time for the research process. That's what i meant.

[Dxxk Mx]

專做 research 個d 都意興闌珊走哂啦, 而且每走一個, 又帶走一d 技術.

呢套戲交到貨已經算系咁, 幾驚 quality 會跌破新低點.....

[sobig]

Metal 既特性系有 BRDF, anisotropic or blurry reflection, color reflection 呢幾樣野都可以 base on Blinn, Phong shader 再加上去; 或者分開唔同 render pass, compositing 到 cheat 都得

even we don't have time to do the render pass. also, max users also render only one render pass....just output a color image, then do the compositing.

[sobig]

and...johngalliano
what sorts of shader are realistic enough in your horizon?

Well, nothing wrong to use traditional shaders, but if there's time, i will try those microfacet-based shaders for the specular, with use of HDRI.
And, having a brief google search, i saw ppl said the mia_material of MR is good for realistic metal, with very high IOR like 10-50.
After all, I think the trend of cg moves to the physically based methods. But I know there's little/no time for research in "C 記".

Don't you think the team would already know that if you can search it out with a brief google search?....

Knowing what's the right/best way to do certain things doesn't mean you can always do that in reality, there's many other factors which I think anyone has production (esp. HK) experience should understand.

To be fair, from what I've seen, I don't think the shots in the trailer are final yet, correct?

ya..stantang,
everyone know the best way to do the VFX. i am not talking about the best way/secret way in hollywood production. however, the production pipeline is the limition. hkcgers would figure out that.

[EricLo]

guys, i am no outsider cause i've been a hk cg worker since 2001. Not always employed from start to end due to my study break (enjoying my 2nd study break, now), but have certainly spent time with some of u guys in "C".

Anyways, schedules are always tight, not an excuse for a studio, the audience won't care about it. U can only blame john chu on that.

If you've gone through the research process of developing more realistic looking metals and still find that the most feasible way to do it is the traditional way, that's fine. But i think there is just no time for the research process. That's what i meant.

Hi Guy! Sorry about my last post. I don't know u also have experience for CG field in HK. Anyway , Exclude IMAGI , Where can figure out technical problem & able to solve them within loving tight schedule? (C / M / MM / DD / FF )? If u just said u know what kind of the shader able render the most of the realistic image. This is nonsense. I welcome u suggest more opinion for improvement point. But if u think we don't know what is HDRI? and what shader is called "mia_metarial for MR"? then i think u have undervalue our knowledge. This feeling just for me . May be,u haven't this meaning. You know, what is "家家有本難諗的經" & "旁觀者不得依牙松鋼"? So, we never said that where studio cannot make good quality FX. We would like to appreciate somewhere able to do there best within limited schedule. that all.

[johngalliano]

I didn't underestimate you, Eric, I know you've been working and studying hard. I just said if there is time, I'll look for other methods for more realistic metals before going the traditional way. Such as, try out those microfacet-based shaders, to see whether the off-specular peak and color shift "effect" could bring more realism as they should be more physically correct and also the Lafortune shader, to see if it's too unintuitive to tweak, these are very old stuff too, but may be worth trying. For the mia_material, I don't know a high IOR like 10-50 is good for metals before, the notion of this high IOR for conductors is new to me. So you can see I haven't studied metals a lot, and i still need to refer to books for algorithms for shaders like cook-torrance to get things clear before implementing anything about it, so that's what i meant by "research" for a hk-style production. I don't mean merely an internet search nor test this and that in softwares, but also i don't mean inventing things completely new.
ok..., now i know there is just no time and it's not the employees fault at all, so u can forget what i said but I am sure you know what i'm talking about cause you have studied the physically based rendering book and very experienced.

[soSmall]

《機器俠》New Teaser + Old Trailer

2009年劉鎮偉科幻大片《機器俠》將以“中國首創、東方元素”於“8月13日”勇闖暑期檔，力求成為暑期重磅。
http://dailynews.sina.com/bg/ent/sinacn ... 09829.html

42秒的預告片:
http://video.sina.com.cn/ent/m/c/2009-0 ... 0681.shtml

42秒的預告片+ 3分30秒片花:
[flash width=400 height=300 loop=false]http://www.youtube.com/v/-FEuv11bw5o[/flash]

[h2o]

唔知 數碼港個場戲系講咩呢? 唔同戲裡戲外 機械俠都系個到出世?

[EricLo]

唔知 數碼港個場戲系講咩呢? 唔同戲裡戲外 機械俠都系個到出世?

No! That shot just for promotion . And don't have major relations with this story! : )

[sobig]

可能真系咁呢....