|
Home > Archive > Flight simulator > October 2005 > More frame rate test results
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
More frame rate test results
|
|
| Dallas 2005-10-30, 7:35 pm |
| My "frame rate benchmark" testing system is working well. A couple of
interesting things:
1) Sometimes restarting FS9 will improve frame rates.
I was testing 4 aircraft when FS9 crashed to the desktop. When it came back
up, I tested all 4 aircraft again and all 4 of them had gained 4fps.
2) FSNav has zero effect on frame rates.
This was a surprise considering all the numbers being crunched in the
background.
3) A short list of aircraft frame rates ranked in order:
Default Cessna 172 - 24.2fps
RealAir Decathlon - 24.0fps
Flight1 Meridian - 22.6fps
Flight1 Cessna 172R - 22.1fps
Aeroworx King Air B200 - 19.6fps
Dallas
| |
| Marcel Kuijper 2005-10-30, 7:35 pm |
| On Sun, 30 Oct 2005 16:44:46 GMT, Dallas wrote:
quote:
> 1) Sometimes restarting FS9 will improve frame rates.
>
> I was testing 4 aircraft when FS9 crashed to the desktop. When it came back
> up, I tested all 4 aircraft again and all 4 of them had gained 4fps.
Interesting development. Sounds like the cache was dumped.
quote:
> 3) A short list of aircraft frame rates ranked in order:
>
> Default Cessna 172 - 24.2fps
> RealAir Decathlon - 24.0fps
> Flight1 Meridian - 22.6fps
> Flight1 Cessna 172R - 22.1fps
> Aeroworx King Air B200 - 19.6fps
I take it these are average framerates?
The RealAir and Flight1 birds aren't doing so bad....but the King Air is
really putting your pc to work. I take it it's more realistic than the
other two?
--
Marcel (SAG-21)
(Mmmmmm.....beer and doughnuts. - Homer Simpson)
| |
| Not4wood 2005-10-30, 7:35 pm |
| Dallas,
With those numbers may I ask what you had your frame rates locked at??
Also, what airport whether it was default or add on were you doing your
testing at??? Did you have real weather or just the standard, and do you
have the sky, clouds and water download textures??
I have all three downloads, sky environment, sky colors and water zone.
I was testing the sky conditions and colors of my new sky download with the
default Cessna 172 and had everything (all texture sliders) up to full
detail. I did a pattern at real weather which was heavy overcast and about
4300' cloud cover.
I averaged anywhere from 9-18 fps and had the fps locked at 25 and was at
default KSEA which is a high detailed airport usually with low frame rates
for me.
Mark G
Not4wood
"Dallas" <Cybnorm@spam_me_not.Hotmail.Com> wrote in message
news:2M69f.1652$8c5.739@newsread3.news.pas.earthlink.net...
quote:
> My "frame rate benchmark" testing system is working well. A couple of
> interesting things:
>
> 1) Sometimes restarting FS9 will improve frame rates.
>
> I was testing 4 aircraft when FS9 crashed to the desktop. When it came
> back
> up, I tested all 4 aircraft again and all 4 of them had gained 4fps.
>
> 2) FSNav has zero effect on frame rates.
>
> This was a surprise considering all the numbers being crunched in the
> background.
>
>
> 3) A short list of aircraft frame rates ranked in order:
>
> Default Cessna 172 - 24.2fps
> RealAir Decathlon - 24.0fps
> Flight1 Meridian - 22.6fps
> Flight1 Cessna 172R - 22.1fps
> Aeroworx King Air B200 - 19.6fps
>
> Dallas
>
>
>
>
>
>
| |
| Dallas 2005-10-30, 7:35 pm |
|
"Not4wood"
quote:
> Also, what airport whether it was default or add on were you doing your
> testing at???
Ok.. here is the test setup:
-----------------------------------------------
First modify your FS9.cfg to show average frame rates by adding this line in
the [MAIN] section of FS9.cfg:
Ave_Frame_Rate_Display=1
This makes your frame rate counter look like this:
[ 003.6 019.6 024.8]
The three numbers are lowest, average, highest fps.
----------------------------------------------------
Then set this up:
Weather off
Traffic off
Start 30 nm north east outside of New Orleans tracking the inbound VOR
radial
Autopilot set to Alt hold 3000ft and NAV mode is engaged.
Throttle full
Once the flight is stable, save it for use later.
Record the frame rate aver as you hit DME 20nm
Engage slew and pull the aircraft back to DME 30nm and make any test changes
and unpause the flight and do it again.
I know it works because it gives me exactly the same numbers every time I
run it for every aircraft tested which is what a benchmark should do.
Oh, BTW my frame rate lock is set to 25 and I have Ultimate Terrain USA and
BirdsEyeView summer. The city and airports are default.
Dallas
| |
| Dallas 2005-10-30, 7:35 pm |
|
"Marcel Kuijper"
quote:
> but the King Air is really putting your pc to work.
> I take it it's more realistic than the other two?
The Aeroworx EFIS King Air suffers from .XML gauge programming. Yes there
is a lot of detail, but it's the EFIS gauges that put a drag on fps.
Dallas
| |
| Gregory 2005-10-30, 7:35 pm |
| On Sun, 30 Oct 2005 16:44:46 GMT, "Dallas"
<Cybnorm@spam_me_not.Hotmail.Com> brought the following to our
attention:
quote:
>My "frame rate benchmark" testing system is working well. A couple of
>interesting things:
>
>1) Sometimes restarting FS9 will improve frame rates.
>
>I was testing 4 aircraft when FS9 crashed to the desktop. When it came back
>up, I tested all 4 aircraft again and all 4 of them had gained 4fps.
>
>2) FSNav has zero effect on frame rates.
>
>This was a surprise considering all the numbers being crunched in the
>background.
>
>
>3) A short list of aircraft frame rates ranked in order:
>
>Default Cessna 172 - 24.2fps
>RealAir Decathlon - 24.0fps
>Flight1 Meridian - 22.6fps
>Flight1 Cessna 172R - 22.1fps
>Aeroworx King Air B200 - 19.6fps
>
>Dallas
>
Interesting!! what would be the total Exterior Texture count (in MB)?
Another column could be VC.. complexity or total texture count.
Since they are all relatively small craft.. let's ignore MDL size.
Good test.. what would you say is the main reason for the B200 being
the biggest `resource challenge' ?
-Gregory
| |
| Not4wood 2005-10-31, 2:36 am |
| thanks, thats pretty interesting. Have you done the same testing with
adding more textures to the sliders, and or traffic? Or how bout maybe
certain planes that we all use like the Posky, MelJet, Pad and so on.
Turning off the sliders adds an interesting touch. To see what can be had
as opposed to what we do in reality to try to get the most out of texture
details while flying in lower frame rates.
Mark
Not4wood
"Dallas" <Cybnorm@spam_me_not.Hotmail.Com> wrote in message
news:jgc9f.3680$Rl1.2216@newsread1.news.pas.earthlink.net...
quote:
>
> "Not4wood"
>
> Ok.. here is the test setup:
> -----------------------------------------------
> First modify your FS9.cfg to show average frame rates by adding this line
> in
> the [MAIN] section of FS9.cfg:
> Ave_Frame_Rate_Display=1
>
> This makes your frame rate counter look like this:
> [ 003.6 019.6 024.8]
> The three numbers are lowest, average, highest fps.
>
> ----------------------------------------------------
> Then set this up:
>
> Weather off
>
> Traffic off
>
> Start 30 nm north east outside of New Orleans tracking the inbound VOR
> radial
>
> Autopilot set to Alt hold 3000ft and NAV mode is engaged.
>
> Throttle full
>
> Once the flight is stable, save it for use later.
>
> Record the frame rate aver as you hit DME 20nm
>
> Engage slew and pull the aircraft back to DME 30nm and make any test
> changes
> and unpause the flight and do it again.
>
> I know it works because it gives me exactly the same numbers every time I
> run it for every aircraft tested which is what a benchmark should do.
>
> Oh, BTW my frame rate lock is set to 25 and I have Ultimate Terrain USA
> and
> BirdsEyeView summer. The city and airports are default.
>
> Dallas
>
>
| |
| Dallas 2005-10-31, 2:36 am |
|
"Gregory"
quote:
> Another column could be VC.. complexity or total texture count.
Yep, that's a good idea, in fact I noticed some of the newer add ons do
better in the VC which is backwards from a fews years ago.
I like the test because it takes the subjectivity out of just looking at the
frame rate display and guesstimating. (And, yeah... I use a spreadsheet.
:-)
It's a useful tool to test for a bunch of unknowns: the effect of a number
of cloud layers in ActiveSky, adding RealityXP gauges, using "lite" vs.
"full VC" cockpits etc.
quote:
> Good test.. what would you say is the main reason for the B200 being
> the biggest `resource challenge' ?
See my post above at 5:30pm
Dallas
| |
| Dallas 2005-10-31, 2:36 am |
| "Not4wood"
quote:
> thanks, thats pretty interesting. Have you done the same testing with
> adding more textures to the sliders, and or traffic?
Not yet, but I plan to test the effects of several different sliders.
quote:
> Or how bout maybe
> certain planes that we all use like the Posky, MelJet, Pad and so on.
Posky?... I don't do freeware... [ nose in the air like Thurston J.
Howell ]
Come along Lovie....
:-)
Dallas
| |
| Quilljar 2005-10-31, 5:33 am |
| Not4wood wrote:
quote:
> thanks, thats pretty interesting. Have you done the same testing with
> adding more textures to the sliders, and or traffic? Or how bout
> maybe certain planes that we all use like the Posky, MelJet, Pad and
We ALL use? I've never even heard of any of them and I have 150 aircraft LOL
:-)
Sincerely,
Quilljar
|
| |
|
|