[Csantucci]

"Show shape warnings" is enabled by default. This causes:
- slowdowns that can even lead to watchdog triggering in some critical cases
- more cumbersome analysis of Openrals logfiles.

I propose that the default for this option is "disabled".

[James Ross]

Csantucci, on 04 June 2016 - 06:19 AM, said:

"Show shape warnings" is enabled by default. This causes:
- slowdowns that can even lead to watchdog triggering in some critical cases
- more cumbersome analysis of Openrals logfiles.

I propose that the default for this option is "disabled".

I would not lose any sleep over Dave's suggestion to just remove the code entirely, but if there are any actually-useful warnings [1] I would not object to keeping just those ones. But disabled by default, 100% yes please. :)

[1] The untextured faces is interesting, as OR doesn't crash, but it also doesn't support setting the colour of such faces. I believe the shape format does support that, though whether any exporters can generate the necessary bits I don't know. Maybe one warning per shape with this problem would be okay, but I'm happy with zero.

[Goku]

James Ross, on 04 June 2016 - 12:13 PM, said:

I believe the shape format does support that, though whether any exporters can generate the necessary bits I don't know. Maybe one warning per shape with this problem would be okay, but I'm happy with zero.

Sketchup exporter can generate proper color material. I'm surprised if OR doesn't use color shaders.

[James Ross]

Goku, on 04 June 2016 - 12:24 PM, said:

Sketchup exporter can generate proper color material. I'm surprised if OR doesn't use color shaders.

Interesting. It's probably trivial to do in OR, but nobody to my knowledge has found a shape where it was meant to have coloured rather than textured faces, so it's never been hooked up.

[Genma Saotome]

Untextured faces in OR are shown as flat gray. The OR software already reports its found an unexpected condition:

Quote

nformation: Skipped unknown shader name Diffuse first seen in shape e:\bin\open rails - us west 1930-1952\global\shapes\wp_util_4a_45ft_xd.s

.

AFAIK, the code mentioned by Carlo is completely redundant to the above message and serves no useful purpose.

[Eldorado.Railroad]

We know why this code was added to the code base and for what reasons. In the end, the reason for it being there, was to "prevent" OpenRails from even starting, which I am very glad was shunned by the overseers. The purpose of said code is interesting, in that it shows the "deficiencies" of the shape compilers that are used to generate MSTS files. However, said code is rather useless in helping the casual user fix those "deficiencies". To my knowledge, even MSTS will load those shapes with those shortcomings. I always viewed the purpose of that code as "punishment" for certain payware vendors, where the originator of this shape verification code was most "unhappy" with.

So yes, please keep it in the code base, as a curiosity, should somebody want to use it. But until there is an automagical way of fixing the "critical" errors identified, where said shapes load up in OpenRails or MSTS anyways, I say good riddance. The default should be OFF. Period.

[Csantucci]

Blueprint filed https://blueprints.l...ssed-as-default .

[Jovet]

I like the ability to see the warnings, so disabling by default is fine but don't remove it entirely!

[Csantucci]

As I got the blueprint approved and the change is very small and didn't got objections, I uploaded the change with x.3553. Code is still there and can be enabled by checking the box.