Message Areas - "Frame rates & certain engines"

Train-Sim.Com Message Areas

"Frame rates & certain engines"

Email this topic to a friend
Printer-friendly version of this topic
Bookmark this topic (Registered users only)
Previous Topic | Next Topic Conferences Microsoft Train Simulator (Protected) Original message
yardlimit (8 posts)
May-02-02, 09:23 PM (EDT) "Frame rates & certain engines" I'm curious. I've heard that textures/coloration on certain engines can have an adverse effect on frame rates. Now I can
run a csx intermodal activity using 3 d9s and 50 cars on NEC and get 20fps +. Most of the stuff runs pretty well.But recently I have tried using some GP7s/ds9s on a similar length coal train on Marias and bog down to 9-11 fps in a similar environment. Seems too specific to be a system problem. My system is pretty fair: 900mhz athlon, 384MB mem, GEforce3-64Mb. Can anyone shed some light on this phenomenon?What should I look for; what should I avoid? Anyone had a similar experience?Yardlimit. Alert Edit | Reply | Reply With Quote | Top
Table of Contents

Subject Author Message Date ID RE: Frame rates & certain engines burgoofj May-02-02 1 RE: Frame rates & certain engines pwillard May-02-02 2 RE: Frame rates & certain engines Rhennar May-02-02 3 RE: Frame rates & certain engines derekmorton May-03-02 4 RE: Frame rates & certain engines cimmer May-03-02 5 RE: Frame rates & certain engines yardlimit May-05-02 6 RE: Frame rates & certain engines plainsman May-06-02 7 RE: Frame rates & certain engines yardlimit May-06-02 8 RE: Frame rates & certain engines plainsman May-06-02 9 RE: Frame rates & certain engines yardlimit May-07-02 10 RE: Frame rates & certain engines plainsman May-07-02 11 RE: Frame rates & certain engines yardlimit May-07-02 12

Lobby | Topics | Previous Topic | Next Topic
Messages in this topic
burgoofj (43 posts)
May-02-02, 09:43 PM (EDT) 1. "RE: Frame rates & certain engines" All the trees on Marias kill frame rates Remove | Alert Edit | Reply | Reply With Quote | Top
pwillard (79 posts)
May-02-02, 10:03 PM (EDT) 2. "RE: Frame rates & certain engines" In other words, too much of anything kills framerates...
Pete Willard
Remove | Alert Edit | Reply | Reply With Quote | Top
Rhennar (191 posts)
May-02-02, 11:50 PM (EDT) 3. "RE: Frame rates & certain engines" It stands to reason that a bigger engine/consist, byte wise, needs more computer power and will slow things down.
T Remove | Alert Edit | Reply | Reply With Quote | Top
derekmorton (411 posts)
May-03-02, 01:03 AM (EDT) 4. "RE: Frame rates & certain engines" You can prove what the problem is by swapping engines in the consist you are testing.Derek Remove | Alert Edit | Reply | Reply With Quote | Top
cimmer (44 posts)
May-03-02, 01:25 PM (EDT) 5. "RE: Frame rates & certain engines" For your testing you shoulod use the same route and area on that route so that the only varible that you are changing is the engines and not the terrain. In your able example you changed too many things for it to be a valid test. IMHO..
Rick Remove | Alert Edit | Reply | Reply With Quote | Top
yardlimit (8 posts)
May-05-02, 07:26 PM (EDT) 6. "RE: Frame rates & certain engines" Tried new engine set using same environment.
Frame rates jump over 20%.
This seems to be a polygon density issue.
Only 3 models I have do it. The NP GP7, and the GN sd9s.
These have all been repainted at some point. They are
quite good looking, but they are killing my frame
rates. (sigh)Yardlimit.
Remove | Alert Edit | Reply | Reply With Quote | Top
plainsman (892 posts)
May-06-02, 03:37 AM (EDT) 7. "RE: Frame rates & certain engines" I am curious, are you running exactly the same trailing cars, both quantity and type and version? I run the NP GP18 a lot and don't experience your results, and I am usually running 3 units. I have less computer power than you. Some cars will really eat frame rates. If I run a mixture of liveries of Svantes' doublestacks, frame rates plummet (But they are beautiful cars).
As a test, try the NP Geep (say 2 units and 20 default gondolas), then the same consist with the T&P GP7, then the default GP38.
Thanks,
Bob Remove | Alert Edit | Reply | Reply With Quote | Top
yardlimit (8 posts)
May-06-02, 09:30 PM (EDT) 8. "RE: Frame rates & certain engines" YUP. I ran the same cars, starting point, weather.
I simply exchanged the 3 Gn sd9s for 4 CP S2s. Frame rate
jumped from 7-9 to 14-15. Search me. Also tried standard
GP38s as well as Alco Rs11s. All others improved frame rate
considerably. The only common denominator I could see was
that the GP18/7 and sd9s all originated with T-Trains whatever
that might imply. My frame rates are usually at least
satisfactory no matter what I use. This is the first such
experience I have encountered. I did notice one oddity
about the GP18. Its appears to be a GP7 body with GP18
stats. When I coupled 3 of them together the game engine
held them appart about a meter. Checked the unit length
for GP7 and reduced it in eng file to about 17.1 meters.
They went together nicely. That engine has been tinkered with
to some degree, but what do I know. Sure wish I could make
it work. Its perfect for an activity I'm working on.Yardlimit.
Remove | Alert Edit | Reply | Reply With Quote | Top
plainsman (892 posts)
May-06-02, 11:38 PM (EDT) 9. "RE: Frame rates & certain engines" A few things, the NP GP18 is NOT a T-Trains model but is the work of Patrick Camaro (3D-Perfection). The SD9 version you mention was a T-Trains apparent adaptation of 3D Trains GP9 textures apparently borrowed without 3D Trains' permission.
The length of a GP7 and a GP18 are both 56'2". Outwardly both locomotives appear very similar. While the GP18 designation may not satisfy purists rivet counters, it was the best model available to represent NP GP18 #378. It was skinned by Don Hewitt at my request. I suspect any performance issues intrinsic to this locomotive result from the underlying 3D-Perfection model.
Here is the actual GP18 upon which the skin is based

The length problem is because of the original .eng file used by 3D-Perfection was based on a GP38-2 file. The length of a GP38-2 is 59'2".
Bob
Remove | Alert Edit | Reply | Reply With Quote | Top
yardlimit (8 posts)
May-07-02, 12:21 PM (EDT) 10. "RE: Frame rates & certain engines" >A few things, the NP GP18
>is NOT a T-Trains model
>but is the work of
>Patrick Camaro (3D-Perfection). The SD9
>version you mention was a
>T-Trains apparent adaptation of 3D
>Trains GP9 textures apparently borrowed
>without 3D Trains' permission.
>The length of a GP7 and
>a GP18 are both 56'2".
>Outwardly both locomotives appear very
>similar. While the GP18 designation
>may not satisfy purists rivet
>counters, it was the best
>model available to represent NP
>GP18 #378. It was skinned
>by Don Hewitt at my
>request. I suspect any performance >issues intrinsic to this locomotive
>result from the underlying 3D-Perfection
>model.
>Here is the actual GP18 upon
>which the skin is based
>
>
>The length problem is because of
>the original .eng file used
>by 3D-Perfection was based on
>a GP38-2 file. The length
>of a GP38-2 is 59'2".
>
>Bob
You're quite right. I got the GP18n confused with the sd9s.
Don't get me wrong, I'm not criticizing. Anyone who donates
time and work here is a saint as far as I'm concerned.
I hope to join them. Interesting about the length issue
-that explains it! Sure wish I could find an NP or GN engine
from the 50s that doesnt misbehave. Theres so few.
I'm looking for a workaround though.Yardlimit. Remove | Alert Edit | Reply | Reply With Quote | Top
plainsman (892 posts)
May-07-02, 12:26 PM (EDT) 11. "RE: Frame rates & certain engines" Try the GN GP9M over at Ron Paludan's site. I run it a lot as well. <http://www.railwaystation.com/trainsim/free.html>
Bob Remove | Alert Edit | Reply | Reply With Quote | Top
yardlimit (8 posts)
May-07-02, 07:31 PM (EDT) 12. "RE: Frame rates & certain engines" YES, YES! There was an RS2 there. Exactly what I needed.
Haleluja!Yardlimit. Remove | Alert Edit | Reply | Reply With Quote | Top

Lock | Archive | Remove

Lobby | Topics | Previous Topic | Next Topic