Jump to content

Archived

This topic is now archived and is closed to further replies.

DarkstarF16

Comparisons one week on...

Recommended Posts

It's now been a week. A week where every spare moment has been spent playing with sliders and check boxes and visiting many parts of the world in different weather conditions. I've not used any external weather engines until today. In fact, it's all been default out of the box scenery and planes. Only today have I loaded the freeware Norway to give it a spin with what I believed were my best settings.

 

I've done several complete flights all lasting an hour or more today and some with weather injected by Opus and AS2012. I've also used default ATC with these flights and just to try and make the world like my FSX world, I loaded AI and set it 10% higher than what I use in FSX.

 

First flight was in the UK from EGLL-EGCC. I always fly in Eyefinity spread across three screens. So I loaded up OPUS weather and off I went in heavy overcast. It was soon obvious that something was amiss. P3D was starting to get throttled. Now was it the weather or was it the weather that was being injected? To alleviate the strain, I started to lower sliders and managed to land, but my buttery smooth flight had ended in a stuttering mess. As I exited, P3D reported a crash. This was not an OOM. It was due to ntl(?).dll I've got the crash report and will be sending it to LM.

 

Second flight was exactly the same flight plan, only this time set with the default weather at gathering storms. I started the flight back at my preferred settings and again as the flight progressed I had to lower values. The flight ended successfully and on exit, no issues.

 

Third flight was in Norway, but this time with AS2012 injected weather. This turned out exactly like the first flight.

 

What have I learned in my first week? The weather engines on my system behave oddly. They both caused P3D to crash with the same error. This may be because I just let them connect, I don't know. I did upgrade AS2012 to Sp3. I'm not drawing any conclusions because it's far to early in the scheme of things.

 

The second, and main thing I've discovered, is I can fly with everything maxed out and with shadows but not fly a full flight buttery smooth. Although my system probably falls into the recommended category by LM that doesn't mean I can go mad and turn everything up high. In fact the best flights have been done with what the default values were when I first loaded the sim. I have added AI at 30% and cars at 20% and turned up one slider to max. The default settings I'm using give better autogen as proved in the pictures that follow and really smooth flight.

 

If I were to run the main sim on a single monitor at 1920*1020 I CAN turn my sliders up and get away with it. Because I fly at 5780*1020 all the stuff below is set in both sims to give me good performance spread across three screens.

 

The setting I'm using in P3D and FSX are set to allow comparable performance. One gives me more autogen that I can't see popping. Better cockpit shadows, lighting and smoother flight. The horizon in P3D is just so real. Oh and volumetric clouds and fog too. FSX is in DX10 and has REX sky textures and is enhanced with SweetFX. Both are with the same scenery which is default.

 

This is FSX in Norway. Both shots were taken at exactly the same time of day and height etc.

Posted Image

 

This is P3D at the same location.

Posted Image

 

FSX process explorer

Posted Image

 

P3D process explorer

Posted Image

 

FSX KLAX at 1920*1020

Posted Image

 

P3D KLAX

Posted Image

 

From the screen shots, one sim looks more realistic. It's all subjective. :)

 

The settings I've used.

 

FSX

Posted Image

 

P3D

Posted Image

 

 

Share this post


Link to post

Why do I see ORBX lights in one but not the other ?

 

EDIT: NM I am an illiterate idiot, apparently  :Clown:

 

Your findings reflect mine. I also agree the weather products need some attention, but that is a given. Opus is working ok, but not perfectly.

 

I have also uninstalled FSX at this point.

Share this post


Link to post

There is a big flaw in your tests: you are testing with the sim paused. Very bad and skewed results in both sims. Not realistic to real flying scenarios. Better use slew, in that case, the clock is running in the sim, but the aircraft ain't moving - it's basically like flying, but not flying.

Share this post


Link to post

There is a big flaw in your tests: you are testing with the sim paused. Very bad and skewed results in both sims. Not realistic to real flying scenarios. Better use slew, in that case, the clock is running in the sim, but the aircraft ain't moving - it's basically like flying, but not flying.

Does that count if, whilst in slew I press pause then come out of slew and reset the time? If so, then for each shot I first set the time 14:00 in the Norway shots, went to the location entered slew and went to 1000ft and took the shot. It was only while I was doing this that I noticed that the two sims looked very different. It only takes about a minuet to slew so I double checked the time before pressing 'v'

 

It bugs me too so any suggestions and I'd be happy to try again.

Why do I see ORBX lights in one but not the other ?

 

EDIT: NM I am an illiterate idiot, apparently :Clown:

 

Your findings reflect mine. I also agree the weather products need some attention, but that is a given. Opus is working ok, but not perfectly.

 

I have also uninstalled FSX at this point.

Yes the weather injects, but something is amiss.

 

One thing I forgot to mention, in FSX (and it's probably my system) when I got to 27L at Heathrow, their were about 9 planes waiting to take off. On my system whenever this happens and it's my turn, default ATC keeps me hanging on and on and I have to cross the threshold for it too recognise and give me clearance to take off. I got ready to wait and as soon as the previous plane had left the ground my clearance was issued. I've carried out many re-installs of FSX and this is the first time this has happened to me. Maybe I'm just unlucky in FSX and ATC were smiling on me today.

Share this post


Link to post

Does that count if, whilst in slew I press pause then come out of slew and reset the time? If so, then for each shot I first set the time 14:00 in the Norway shots, went to the location entered slew and went to 1000ft and took the shot. It was only while I was doing this that I noticed that the two sims looked very different. It only takes about a minuet to slew so I double checked the time before pressing 'v'

 

Maybe you misunderstood me: I wasn't talking about the exact time in the sim. Just that the "clock" is running, that the sim is rendering the stuff in real time. Just set yourself where you want to be, center the joystick and activate slew. Unpause the sim. Then take your measurements. Those are realistic measurements.

In your shots pause is active and that is bad. Notice how FPS change greatly when you pause and unpause both sims.

I suggest you save the situation in FSX and then load the same situation in P3D2, as it can actually load FSX situations. FSX can't read P3D2 situations though.

That way, when you saved with slew active, it will be active on reload and your position and time will retain.

Share this post


Link to post

Thanks :) I wasn't concentrating too much on FPS (my fault) but on the difference in autogen between the two sims.

 

The slight difference in the time of day between the two was an unexpected surprise.

 

 

Share this post


Link to post

Well, yes, the autogen is much nicer in P3D2 indeed, but I would very much like to know the performance result. Seems to be a point of discussion lately.

Share this post


Link to post

Upon a closer look, the Norway shots appear to show 2 very different types of textures. I am assuming this is FTX Global V1.1 in both sims. Is that correct?

 

Look closely outside of the airfield. IMO the textures in P3D2 look a lot better.

 

Or do I need a new pair of glasses?

Share this post


Link to post

I also had those ntdll.dll crashes. Since Nvidia stuff is not really recognizing P3d V2, I am suspecting it's related to graphic driver. I hope Nvidia and LM quickly solve this problem.

Share this post


Link to post

Upon a closer look, the Norway shots appear to show 2 very different types of textures. I am assuming this is FTX Global V1.1 in both sims. Is that correct?

 

Look closely outside of the airfield. IMO the textures in P3D2 look a lot better.

 

Or do I need a new pair of glasses?

The textures for Norway are the freeware ones available, and are loaded into both FSX and P3D. P3D displays quite differently to FSX at comparatively low settings which I feel are highlighted by these pictures.

Share this post


Link to post

The textures for Norway are the freeware ones available, and are loaded into both FSX and P3D. P3D displays quite differently to FSX at comparatively low settings which I feel are highlighted by these pictures.

Why would the same textures show up so differently? Enlarge them and study them. They are quite different and my obvious question is why, or if they are indeed the same ...why do they look so different?

 

Sent from my GT-N7100 using Tapatalk

 

 

Share this post


Link to post

Why would the same textures show up so differently? Enlarge them and study them. They are quite different and my obvious question is why, or if they are indeed the same ...why do they look so different?

 

Sent from my GT-N7100 using Tapatalk

Initially, I thought it was simply the difference between how the two sims display. Maybe I've not installed the scenery correctly in P3D. I will certainly check as I don't like mysteries.

Share this post


Link to post
Why would the same textures show up so differently? Enlarge them and study them. They are quite different and my obvious question is why, or if they are indeed the same ...why do they look so different?Sent from my GT-N7100 using Tapatalk
Second attempt at Norway. I had indeed not arranged the scenery library in P3D. Files are now in the exact same order in both systems.Both shots taken in free flight so may not match up exactly. Both with todays date and time set to 10am.My personal opinion is that fsx is far too bright for the time of year, as the sun never gets high in the horizon in such a northerly country. Therefore, P3D displays a more realistic view of the world. I did a time preview and indeed the day is very short. It appears that the whole solar physics engine has been given a complete overhaul. FSXfsx_norway_2_zps843ba22c.pngP3Dp3d_norway_2_zps35c74ad1.png

 

There is a big flaw in your tests: you are testing with the sim paused. Very bad and skewed results in both sims. Not realistic to real flying scenarios. Better use slew, in that case, the clock is running in the sim, but the aircraft ain't moving - it's basically like flying, but not flying.
PERFORMANCETwo flights to highlight P3D and FSX. Both same day , same time. I've used Heathrow (UK200 extreme) as this is the worst performing area in FSX for me. FSX is in DX10 mode and has UTX and GSX installed and is running SweatFX. At the end, I've shown the settings used in each sim. These settings only apply to my PC and how I get smooth flight with maximum settings in each and how hugely they differ. I know which one I like and it can only get better :) (Its funny what making a video does for your take offs! Back to school for me.) FSX
P3D
FSXFSXsettings1_zpse3d40361.pngFSXsettings2_zps9167bcac.pngP3DP3Dsettings1_zps7c443ab8.pngP3Dsettings2_zpsb95b70b8.pngSince making the video I noticed that I had turned off some settings. I've added:Airline Traffic = From 20% - 40%General Aviation = 10% (added)Road vehicles = From 20% - 30%Airport Vehicle density - minimum (added)Performance has decreased by about 1fps

Share this post


Link to post
×
×
  • Create New...