[gpz]

 Eldorado.Railroad, on 19 February 2014 - 05:22 PM, said:

Yes, that is exactly what I have been accessing on the SVN server. However, I have not had success in backdating/reverting to a specific eXperimental version of OR, lock, stock, and barrel. This would help me do comparisons in both results and the code, which would help solve problems.

If TortoiseSVN is installed on your computer, just right-click on project folder, select menü TortoiseSVN -> Update to revision. A form appears with a text field, into which you can enter the revision number you want to checkout. It is very easy to go back to even the very beginning. :clapping:

If you need to see two versions parallelly, then copy the whole project directory, so you can checkout two different versions to the two directories.

[Eldorado.Railroad]

 From 19 February 2014 - 11:22 PM:

If TortoiseSVN is installed on your computer, just right-click on project folder, select menü TortoiseSVN -> Update to revision. A form appears with a text field, into which you can enter the revision number you want to checkout. It is very easy to go back to even the very beginning. :sign_rockon:

If you need to see two versions parallelly, then copy the whole project directory, so you can checkout two different versions to the two directories.

Peter,

That solved the problem. I was using a standalone SVN program, which is handy because it does not need to be installed, but it lacked certain features. I don't really like the shell integration of Tortoise, but at the moment there is no other free choice that I can think of. When a machine is being used for multiple purposes (not just coding per say) shell integration is too distracting. That right click menu can get to be way to long and complex! I wish there was a portable version of Tortoise someplace as even now this is not the machine I would use to code with.

Thank you sir!

[Eldorado.Railroad]

 From 17 February 2014 - 04:44 AM:

ShadowMapBlur = (user set) False
ShadowMapResolution = (user set) 6144

My graphics card is an Nvidia GTX640 with 1GB of memory, which is fairly modest by today's standards.

Although I have not been able to do so since July of last year, I ran OR on a SLI Nvidia 780 rig. Quite simply, outstanding and expen$ive! On lower hardware I tried your settings (except I left ShadowMapBlur as True). When shadow maps are that size, this ATI rig drops framerates by at least 33% into the 40 FPS territory. I wonder what framrates you are achieving when the HUD indicates large numbers (>2000) for primitives and/or dynamic shadows?

[markus_GE]

Can tell you that, having tried out (and adpted, yet not adopted) these settings: 25 on the MONON, with that drop in the low 40s on everything else.

Reduced map resolution to (I think) 2048 and enabled two maps... gives reasonable shadows and reasonable performance.

Cheers, Markus

PS: This is a SONY VAIO Laptop, i5 M430 @ 2.27 GHz, 4 GB RAM, ATI Radeon Mobility HD5650 GPU with 1 GB VRAM

[dennisat]

 Eldorado.Railroad, on 23 February 2014 - 08:38 PM, said:

Although I have not been able to do so since July of last year, I ran OR on a SLI Nvidia 780 rig. Quite simply, outstanding and expen$ive! On lower hardware I tried your settings (except I left ShadowMapBlur as True). When shadow maps are that size, this ATI rig drops framerates by at least 33% into the 40 FPS territory. I wonder what framrates you are achieving when the HUD indicates large numbers (>2000) for primitives and/or dynamic shadows?

I have VSync enabled so never see frame rates above 60. They start to drop into the 50s when I hit about 3,000/1,500 Render/Shadow primitives and into the 40s at 4,500/1,200 primitives. It really starts to stagger, less than 20fps, when I hit about 7,000/4,000 primitives but I don't have many route locations that are this dense.

Dennis

Edit:
Having re-read this topic from the start I think I'll experiment again with the shadow options.

[Eldorado.Railroad]

 From 24 February 2014 - 01:27 AM:

I have VSync enabled so never see frame rates above 60. They start to drop into the 50s when I hit about 3,000/1,500 Render/Shadow primitives and into the 40s at 4,500/1,200 primitives. It really starts to stagger, less than 20fps, when I hit about 7,000/4,000 primitives but I don't have many route locations that are this dense.

Yes, the moment you turn the blur on, frame rates will drop quite a lot when the maps are that big. In your case you are not running the sim with blur enabled. If you enable ShadowAllShapes, you will see quite an increase in Shadow Primitives for a dense route. Try it, you might like it, but you will have to leave the blur effect off.

Thanks for the test data.

[BB25187]

Hi,

On my side, I noticed another change since the rev #1597. To illustrate it, I prepared a video. I used the Swiss Crocodile, first with the revision #1596, and then with the latest revision #2067 (the result would be nearly the same with any revision since #1597). I removed the textures from the engine to make the difference more visible.

The video to illustrate (you can enlarge it up to 1080pixels)

We note two important changes with the revision higher than #1597, compared to the revision #1596:
- The shadow of the middle body onto the rear "boggie" is no longer visible
- The pantographs don't produce any shadow on the roof anymore (let's forget about the "shivering effect" which is only visible because of the rapid change of the lighting).
If we look closer, we can see that many other shadows which were produced by some parts of the machine are no longer visible either. For instance: the upper part of the machine created very nice shadows on the wheels and lower parts. This is no longer the case today.
Setting the ShadowAllShapes option to true/false does not modifies things.
Is there a way (option or whatever) to restore some of the shadows we got with rev #1596? They really enhanced the aspect of the rolling stock.

Regards

[James Ross]

 BB25187, on 01 March 2014 - 12:31 PM, said:

Is there a way (option or whatever) to restore some of the shadows we got with rev #1596? They really enhanced the aspect of the rolling stock.

No, and I'd really appreciate people stop asking for previous version of things. If you want to help, identify where the bug is in the changes that have been made, but at no point are we going to wholesale go back to a previous version.

[BB25187]

Hi,

 James Ross, on 01 March 2014 - 03:04 PM, said:

No

OK, this is clear.

 James Ross, on 01 March 2014 - 03:04 PM, said:

and I'd really appreciate people stop asking for previous version of things.

Apologize but this was not my point. It may have happened that I missed something with the changes at rev #1597. I also thought that this section was the right place to ask questions on possible malfunctions or enhancements before entering records in the tracker.

 James Ross, on 01 March 2014 - 03:04 PM, said:

If you want to help, identify where the bug is in the changes that have been made, but at no point are we going to wholesale go back to a previous version.

OK. Do you confirm that the first step is then to enter a defect in the bug tracker for this specific point?

Regards

[James Ross]

 BB25187, on 02 March 2014 - 05:27 AM, said:

OK. Do you confirm that the first step is then to enter a defect in the bug tracker for this specific point?

You're always free to enter bugs for things you think should be changed/made better. I doubt the issue here will go anywhere unless someone actually finds a bug in the code, though.