Hi Flyer,

 

Apology accepted...obviously it was a misunderstanding. ... It's the risk of any joke due to timing :-)

 

Truth is, these videos.... all of them... rock! ...Rob's efforts much appreciated.

Hi Flyer,

 

Apology accepted...obviously it was a misunderstanding. ... It's the risk of any joke due to timing :-)

 

Truth is, these videos.... all of them... rock! ...Rob's efforts much appreciated.

Thanks for accepting my apology.

 

The last one, the CRJ in SFO was the smoothest P3DV2 video I have seen. This can Be faked but I am certain Rob Is not Recording slow and Speeding up the capture. LOL

 

VAS at 3.2 on those setting flying that kind of addon into a hub like that on any other version FSX, P3D1.4, V2.1 would certainly lead to an OOM on any system.

 

Also it was recorded with FPS at unlimited, Tripple buffing and V-sync. That completely counter intuitive to my strategy for Smoothness in the sim.

 

Can't what to get my hands on V2.2 :lol:

If they do, they haven't told me.  I have my own theories and they're pretty technical, but I believe I posted those ideas some time ago in another thread and it's just my speculation from a programmers perspective.

Rob,

 

Thanks for jumping in and providing useful feedback for LM and the masses.  Much appreciated!  It appears the VAS issue has been very much tamed now and we can move on to other areas while at the same time perhaps they will target those remaining stuck textures that aren't being utilized.

 

On a total side note--ObjectFlow apparently can now work in V2.x and I'm really wanting to see KRDD come alive!  I'm only 90m from there ;o)  

 

Cheers,

Noel

On a total side note--ObjectFlow apparently can now work in V2.x and I'm really wanting to see KRDD come alive!

 

Got a link to that info?  I'd be happy to test it if it's available.

 

 

Do you have any thoughts on why the VAS usage was different on your second flight? You are using high-end add-ons in both flights. Is it linked to the type of scenery you are flying over?

 

The differences are:

 

1.  Not flying in the Orbx PNW region

2.  Not flying in the Orbx NCA region

3.  Using Aerosoft Twin Otter not the Quality Wings RJ100

4.  Using Aerosoft Thessaloniki

 

Something to keep in mind about resource releasing ... it's very much dependent on how 3rd party products achieve their goals as to why their resources aren't being released ... they (3rd party) work around limitations by using creative methods, but those creative methods have a down side.

 

Although LM have taken the brunt of the finger pointing (some valid) it's still a two way street and blame is rarely one sided

 

Cheers, Rob.

 

 

The clincher is....when ou load up the aircraft, go to Windows Task Manager, Processes, and assign all cores to P3D or FSX if running Legacy mode. It is smooth as butter.

 

Interesting Will ... I'll give that a go.  What I've found is that Spot View introduces stutters (not fps issue, but real stutters) however Locked Spot view does not.  This has me really puzzled ... why would one external view introduce stutters and not the other ... this happens on any aircraft but it just more obvious on the Q400.

 

As I understand it, the Q400 slews itself thru the virtual world -- please correct me if this is not accurate.  So my thought was something about slew mode ... this issue actually dates back to FSX, where certain views stutter while others don't.

 

I hadn't thought about this being a "threading" issue but maybe it is and it's worth trying your suggestion on core assignments. So, off to do some more testing.

 

Cheers, Rob.

 

 

exodus1977, on 29 Mar 2014 - 12:12 PM, said:


2. I noticed in the last few seconds of your video some outside camera views. Were these done with v2.2's new flight recording system?



No, that view is the standard "FlyBy". I have done some testing with the new flight recording system but I had some issues with my setup and have reported to LM.

 

Hmm, how did you do that? The first landing was 'in cockpit'. the other two views were out of the cockpit. were these multiple landings you recorded? I'm just trying to understand how you did it as I would like to try the same.

 

 

The first landing was 'in cockpit'. the other two views were out of the cockpit. were these multiple landings you recorded? I'm just trying to understand how you did it as I would like to try the same.

 

Oh I misunderstood, that's the built in Instant Replay feature (this has been around since FSX and earlier) ... if you look in the lower left corner you can see the counter (you'll need to view full screen mode in HD so the YouTube playback bar is not active).

 

Cheers, Rob.

 

 

that's the built in Instant Replay feature

 

Ah, got it. I've been trying to sort that out as I've never used it before.

 

 

Got a link to that info?  I'd be happy to test it if it's available.

 

My bad, I mis-read another comment on the ORBX forum re this issue.

Hey Robert,

 

thanks for your videos about the upcoming 2.2 version of this great P3D.

I'm looking very much forward to this, the VAS improvements as well as

the cloud shadows feature.

 

Looking at your last vid (the Bae RJ100 one) I see your P3D shows the same

issue with extrusion bridges as mine does. At 8:51 you can see that the

bridge lays upon very high poles, which get higher if you would move away

from it.

 

I thought this was related to UTX, but as you don't have UTX installed it

must be something else, FTX Global? or just P3D bug?

 

And could this also be a cause of VAS growth?

 

Regards,

 

 

I see your P3D shows the same
issue with extrusion bridges as mine does. At 8:51 you can see that the
bridge lays upon very high poles, which get higher if you would move away
from it.

 

Interesting, I'll see if I can locate that position ... pretty sure that bridge would be from Orbx Vector 1.1 ... if you have a specific coordinate for me to check, I'd be happy to look at it.

 

Unlikely cause of VAS growth ... from what I've been told (and it's limited information) 3rd party developers have come up with rather creative ways to work around some of the limitations of the SDK ... this creativity, however, could have negative effects on how objects are released, which contributes to the continued increase in VAS.  The VAS issue is a "shared" experience and it looks like LM have got a good handle on it (in my 2.2 beta) from their side of the fence.

 

Like I listed earlier, I did a 4+ hour flight where VAS actually went down for a while then went back up a little on landing (2.4GB) with same rather high graphics settings.  So clearly 3rd party does indeed play part into the VAS growth/release equation, it's not just an LM issue.  I don't really want to suggest VAS "bugs" either ... could be that working around limitations to meet 3rd party desired goals is the "only" option.

 

Cheers, Rob.

Made another quick video showing off some of the cloud shadows ... same graphics settings as before ... for some reason YouTube didn't process this as 360p??  But it's available in HD (720, 1080, 1440).

 

https://www.youtube.com/watch?v=iOHMyMbWKRE

 

Cheers, Rob.

I took a look at the video and I love the additional dimension that the cloud shadows add.  This has been a feature that has been requested since I have been involved in flight simming and it is great to see that it is now part of P3D.

Rob, you are like a rehab nurse for a crack addict to me right now.  I still haven't had time to setup a clean Windows for P3D, and my current installation is borked for reasons outside of P3D.  Your videos are really nice and are giving me a lot to look forward to when I can get a fix this week sometime, just in time for the 2.2 release.  Thanks.

 

I'm running a measly 3770k and EVGA ACX 780, but I've managed to get up to 4.8 stable.  No OC on the GPU yet, but no water there either.   I wonder how that will do, and I imagine I'm in the same hardware boat as most others here.

 

Micro-stuttering doesn't bother me much, just that I do see it.  If LM has managed to clear the longer stuttering overall I'm going to be very happy.  (card stutter vs code stutter?)

 

Even if SLI works a little bit I might go grab another card, if wife and baby turn their backs long enough.

Micro-stuttering doesn't bother me much, just that I do see it.  If LM has managed to clear the longer stuttering overall I'm going to be very happy.  (card stutter vs code stutter?)

 

Hang on to that single card for now ... I just made an interesting discovery tonight that gained me 90% performance increase with just one setting change ... also allowed me to crank up Water to Ultra, and add a virtual "ton" of clouds.

 

I'll post another video tonight of CYPS revisited and I'll be interested to see what folks say and if they notice anything different.

 

Cheers, Rob.

 

EDIT: I should thank Wes on this one, it's actually something he mention to me that got me thinking.

I just made an interesting discovery tonight that gained me 90% performance increase with just one setting change ... also allowed me to crank up Water to Ultra, and add a virtual "ton" of clouds.

 

Hmm, it makes me think something was terribly restricted previously in your setup?  90% increase is, well just a wee bit ... huge!

 

I have very to ultra smooth performance UNLESS my GPU is up to 99+%, then it can get less than perfect, but still very good.  By this I mean freedom from micro stutters.  I have also run water on Ultra and have max dense clouds out to 100m using 2096 bit cloud textures since day 1.  What the hey was the one setting change?

 

Since using PNW & NCA now in those sorts of cloud particularly of late w/ lots of weather in Nor Cal & the PWN, I'm beginning to see some use for a 2nd Titan.   Hate to have to replace this PSU at 850W but I may have to.   I have to wonder how long it will take to get TWO Titans up to 100% utilization.  I'm guessing 2 Titans in SLI would use less total watts than 1+1 Titan, i.e. 2x1 Titan's watts used under load.

 

 

I just made an interesting discovery tonight that gained me 90% performance increase with just one setting change

 

I just wet my pants!

 

Oh wait...  thats ... little one.  again.  never mind.