typewriter

[oakchas]

Okay!

 

The guys at Hash Inc. saved my model... but I was in the process of rebuilding anyway. So, here's the progress so far.

 

[attachmentid=15509][attachmentid=15510]

 

1. The lighting's not right... But I've got to relearn a bit with the switch to v 12 from 11... apparently the lights are brighter now (the render doesn't indicate that until the render is complete and you click in the screen, then it gets washed out.

 

2. Decaling is accurate and maybe a bit much (higher res than needed).

 

3. I haven't aged it yet... this is what it looked like new.

 

I've got a roto (but not from this angle) included here as a reference to compare against... (Heck, many of you, having never seen a typewriter, may wonder where the "windows" key is )

 

Any suggestions at this point? Especially about the lighting?... I'll post updates in this thread as time goes by.

[Scottj3d]

Lookin good so far. At first I quickly clicked on the pic's before reading all of your post. Thought the roto was the model (WOW that look's like a photo!) Oh, it is! Can't offer any advice on the lighting and no crit's at this point. What I see look's good to me and the decaling look's fine. I want to see that chassis (main body) come together. Are you building it at the angle of the roto and planning to rotate it once it's finished? I never thought to model that way but I suppose with those key's (nice job on those by the way) it could be necessary.

[nimblepix]

Looking good so far.

A little early to say too much.

Are you going to thicken the metal all around,

and deboss the characters on the keys?

If I remember correctly,

these typewriters had a very unique texture.

Are you going to try to recreate that too?

[oakchas]

Yep, I'm way early with this one, so there isn't much to crit now... I'm starting on the case (under the cover where the strikers come out and around the dustplate for the keys). The compound curves are killin' me.

 

Are you building it at the angle of the roto and planning to rotate it once it's finished?

No, I had a guy take pics of all sides and angles of one for accurate rotos, but this other pic was high enough res for me to read all the keys... there weren't any or {}\| back then And I'd forgotten that most simply had commas above and below on one key, same with period.

 

and deboss the characters on the keys?

I didn't think the letters were indented any? is that what you mean? Like the W would actually be engraved into the surface of the key? I do have the circular indentation around the letters on the keys.

 

I'll double check that... and I think I'd have to use ver 13's displacement to do that.

 

Are you going to thicken the metal all around

I'm going to add the lip around the striker hole, as I did around all the key holes and where the bottom of the striker cover meets the top of the case.

 

Thanks for lookin', guys.. now, I'm back to splinin!

 

Here's an update.... and a wire

[attachmentid=15545][attachmentid=15546]

[Scottj3d]

I'll double check that... and I think I'd have to use ver 13's displacement to do that.

 

 

You might be able to get away with a bumpmap reversing the colors and a slight gaussian blur. It'd be worth a try. I've done some spaceships where I did the seam's for the plates making up a fuselage that it seem's to me would work well for that. Just a touch of blur should give you that edge around the letter's to make them appear indented. Here's an example of the indentation I was able to get using that technique. I can get you the percentages if you think it would be worth a shot.

[oakchas]

Help!

 

This is getting funky. I'm trying to model this case. I've tried porcelain to smooth it... with a .2% setting on the normal weight in the attributes surface (I've tried going up to 100% and places in between, and that's worse)... I'm still getting kinks. I've tried other methods of modeling it and the compound curves are killing me, no matter what I do. It looked better in older versions (like 11.1 t)... but v 12's seem worse.

 

I think it's me... but so much of it has come out good, I can't quit... any suggestions out there? I've tried Jin's (Waheed's) method... I've tried Rodger Reynold's.

 

I know I can do this, I've done such things before... Here are some pics so you can see what I'm getting. I don't really want to post the project 'cause I wont't learn much if somebody fixes it for me, but I really would like some suggestions... I've messed with the magnitude on the splines causing the kinks... I'v even messed with alpha and gamma... I'm stumped.

 

this is the chor shot

[attachmentid=16322]

 

this is the shaded from the modeling window

[attachmentid=16323]

 

this is the shaded wire from modeling window.

[attachmentid=16325]

 

Thanks for your suggestions

[luckbat]

You've got a number of dead-end splines that are going to cause problems. I've circled them below:

[attachmentid=16326]

 

The trick for rounded surfaces is three-point patches. Take a look at this model of a sink that I made a while back--it uses three-point patches on nearly every rounded corner.

sink.mdl.zip

[Rodney]

Charlie,

There are some spline intersections that are going to give you problems.

You have at least one 3 pointer that will work... but it might be reworked as well.

 

When in doubt and you can't get out of a problem its often best to let A:M work it out for you.

In your case I think if you locate the profile (see blue line in image below) and extrude out from there.

Then once your splines are in place in one profile select splines and turn to another profile and pull out the parts you need. Clear as mud... sorry.

 

Maybe the images will help.

 

Edit: Gotta get up early to beat Mike Stamm to the post!

Uploaded the wrong image too! Right one posted now.

[oakchas]

Thanks Mike, Rodney...

 

I'm trying. Very trying (I'm told)!

 

I'll attempt reworking it. The extrusion method is a bit tougher because of the cut out for the strikers in the top of the case... and the cutout where the keyplate rests... that's why I was using R.Reynold's method.

 

The three point method makes a beautiful compound curve. But, where it ends in a straight line(i.e. where the top of the case meets the bottom of the case) it creates problems for me.

 

Add into that mix, when I try to break a point and delete it I'm getting realtime render errors and A:M closes...

 

Frustration! But I shall overcome! Swing low, sweet chariot!

[R Reynolds]

I realize I'm really late to this thread but here's a quick suggestion.

[oakchas]

Rodger,

 

Thanks! I'll give that a go.

 

I haven't had much time of late... and I really would like to get this one down...