(WIP) **GE AC4400CW** In Development from Searchlight Simulations!

[mrennie]

I think he has. Theres blue in the front part.

So there is

[spartantrain32]

I think he has. Theres blue in the front part.

So there is

Im sure without a doubt, its CSX

[5292nate]

We actually have not started texturing as of yet. We both still have much to learn in that department. The blue you see is the photo texture applied to it for model referencing. Texturing isn't too far off, though! Also, thank you all for the encouraging words!

[mrennie]

We actually have not started texturing as of yet. We both still have much to learn in that department. The blue you see is the photo texture applied to it for model referencing. Texturing isn't too far off, though! Also, thank you all for the encouraging words!

Well, I did wonder if that blue was a genuine texture or just a reference. The reason I asked if you'd started is that I find that it's better to texture as you go along rather than waiting until the model is done, because quite often the texturing has a significant influence on the way to make the geometry (it's hard to explain why, but when you do it, you realise - it has a lot to do with normals smoothing groups, shadows and highlighting). Also, when you have parts that are replicated many times (nuts, bolts and rivets are the most obvious examples), you can save yourself a lot of time and work if you make one, texture it and then replicate it, rather than make one, replicate it and then texture each one separately.

Anyway, the detail in the geometry is looking superb.

[bnsfsubdivision]

We actually have not started texturing as of yet. We both still have much to learn in that department. The blue you see is the photo texture applied to it for model referencing. Texturing isn't too far off, though! Also, thank you all for the encouraging words!

Well, I did wonder if that blue was a genuine texture or just a reference. The reason I asked if you'd started is that I find that it's better to texture as you go along rather than waiting until the model is done, because quite often the texturing has a significant influence on the way to make the geometry (it's hard to explain why, but when you do it, you realise - it has a lot to do with normals smoothing groups, shadows and highlighting). Also, when you have parts that are replicated many times (nuts, bolts and rivets are the most obvious examples), you can save yourself a lot of time and work if you make one, texture it and then replicate it, rather than make one, replicate it and then texture each one separately.

Anyway, the detail in the geometry is looking superb.

Thank you Mike! Our initial plan was to texture it along but this will have to wait till spring.

[bnsfsubdivision]

[spartantrain32]

That is so amazing

[bnsfsubdivision]

okay I was liking this but there's a line where too much is too much. this has to be a texture in side there, no one is gonna look that closely and its gonna take forever to model with this amount of detail. Sorry for the negative opinion but its getting a bit ridiculous now . Although I must say this is amazing detail but I still don't support it

We have enough support out there and It's time to move on from the 15k poly locomotives you and most people are used to (no offense intended). Times have changed and we're not living anymore in 2007.
We are well on track with modeling and everything planed so there's no need to say that it's gonna take forever to model it without to even know exactly at which stage this thing is.

[PapaXpress]

We have enough support out there and It's time to move on from the 15k poly locomotives you and most people are used to (no offense intended). Times have changed and we're not living anymore in 2007.
We are well on track with modeling and everything planed so there's no need to say that it's gonna take forever to model it without to even know exactly at which stage this thing is.

+1

[artimrj]

okay I was liking this but there's a line where too much is too much. this has to be a texture in side there, no one is gonna look that closely and its gonna take forever to model with this amount of detail. Sorry for the negative opinion but its getting a bit ridiculous now . Although I must say this is amazing detail but I still don't support it

We have enough support out there and It's time to move on from the 15k poly locomotives you and most people are used to (no offense intended). Times have changed and we're not living anymore in 2007.
We are well on track with modeling and everything planed so there's no need to say that it's gonna take forever to model it without to even know exactly at which stage this thing is.

+1 more. Work it until Your satisfied. I'll be supporting you soon as it is done.

[Chacal]

Mike, you made similar comments in October and they answered it already:

The model will be trimmed for final export so as to keep the poly count at a reasonable level.

So I don't know why you bring it up again.

[5292nate]

Thank you for your continued support. If anyone wouldn't mind passing this around to other areas of the community, that would be awesome as our advertising resources are limited.

Thanks again, all!

Nate.

[buzz456]

If it's half as good as it looks it will sell itself.

[mrennie]

Let's see, about poly count - the FEF-3 is already twice as large as the Consolidation and I'm still getting similar fps. I've seen screenshots with five (yes, five) Consolidations side-by-side, all with steam coming out, no details suppressed, and no problems. TS2014 is much more capable than the older versions.

My recommendation for making sure that a model will work on lower-end computers is to do as I did with the Connie - put some of the finer details (nuts and bolts) into a group (or node) than can be deactivated at the press of a button. Either that or make two versions of the model - one fully detailed (HD) and the other that leaves out some of the finer details (replacing 3D nuts and bolts with bump-mapped images on the underlying textures). You could also make the non-HD version use compressed textures on everything and leave the uncompressed ones (*_NM) for the HD model.

From my experience, there are only two things that have a significant and immediate impact on fps - one is the shadows cast by lights, but the impact is not that high really, and the second is the density of particle emitters. These do have to be adjusted carefully to get the right balance.

[bnsfsubdivision]

Let's see, about poly count - the FEF-3 is already twice as large as the Consolidation and I'm still getting similar fps. I've seen screenshots with five (yes, five) Consolidations side-by-side, all with steam coming out, no details suppressed, and no problems. TS2014 is much more capable than the older versions.

My recommendation for making sure that a model will work on lower-end computers is to do as I did with the Connie - put some of the finer details (nuts and bolts) into a group (or node) than can be deactivated at the press of a button. Either that or make two versions of the model - one fully detailed (HD) and the other that leaves out some of the finer details (replacing 3D nuts and bolts with bump-mapped images on the underlying textures). You could also make the non-HD version use compressed textures on everything and leave the uncompressed ones (*_NM) for the HD model.

From my experience, there are only two things that have a significant and immediate impact on fps - one is the shadows cast by lights, but the impact is not that high really, and the second is the density of particle emitters. These do have to be adjusted carefully to get the right balance.

Thank you for the explanation Mike! It's much appreciated and will hopefully show people that polys are, (if grouped like you already explained above) no longer really a matter for the performance of a engine.
Also. Thank you Buzz!