Ohhhhhhh :-S is it just a simple case of reinstall and go?

Good morning from the West Coast of NA.  Just spent the last 30 minutes catching up on this tread.

I have to agree with "soundrats".

I  bought this software yesterday.  What I have found is that if I start a flight in a heavy on fps area and use FA then it works as advertised.  However, after I fly off to some undemanding area, no urban, then the frames are considerably lower than I would normally get (IE locked at 30fps I would see 27-29 without FA).

So, I turn off FA.  And flying around this same area for over 20 minutes instead of my frames returning to the close to 30 I would normally get, I have about 1/2 of that 15-20 maybe.  It never returns to what it should.  Also I am getting some microstutters I never had before and some blurries....gees this is frustrating.  I am following the manual to the letter.

My only conclusion can be that there is still something impacted by this program or why would it not return to its normal performance.

I shut down my FSX, off my computer for the night and now I've rebooted from a cold start and restarting FSX from a saved position and I get my better performance values.

 

I just don't know.  This app seems to be more of a pain in the butt than a real overall solution.
 

 

 

So, I turn off FA. And flying around this same area for over 20 minutes instead of my frames returning to the close to 30 I would normally get, I have about 1/2 of that 15-20 maybe.

 

If you'd exited FSX and restarted, your frames would probably have returned to normal right then. 

 

Isn't this kind of an artificial situation anyway?  Under what conditions would you fly with FA for a while, then turn it off and fly without it for the rest of the flight? 

 

While I haven't tried the program, what you described is pretty much what I'd expect to happen.

 

Hook

Isn't this kind of an artificial situation anyway? Under what conditions would you fly with FA for a while, then turn it off and fly without it for the rest of the flight?

 

Bryan, Larry is correct. The only I know of to get good consistent benchmarks with FSX is to 1) have a saved flight complete with weather etc, always starting from the same conditions 2) having a predetermined flight path (with good turns) AP ok 3) rebooting after every test flight and 4) clearing the shader cache before each flight. This is the way I have tuned my rig and test FA.

 

After flying the same route in this way a half dozen times, you will know where the slowdowns, stutters, etc are. FA did not increase my frame rates at every point in my 15 minute test flight. BUT every phase was smoother especially the low FPS parts.

 

I am posted this before but here is again, my rig, my test flight, and my settings: http://uchisworld.wordpress.com/2014/05/26/happy-anniversary-part-2/

 

FA is the real deal but has limitations like any other component of FSX.

 

edit: Note the slightly jerky feel to the video I posted on my site. That is ALL gone with FA.

I'm stunned I flew the PMDG 777 out of FSDT's  KLAX during a major thunderstorm, using Rex 2048 clouds with 110mile draw distance. all sliders to the right with the exception of autogen which was set at dense and water a 2x high and I was seeing between 35-45 FPS and no stutters through the whole flight! When my wallet heals up from my PC build I'm gonna buy this it makes the sim so smooth! Works great in DX10!

Could this possibly change in the future I have 3-4 apps related to FSX that I use when I fly but they will not open now because I removed .NET v4.5 to run the Fibre Accellerator.

You should NOT have "removed .NET v4.5" at all!

 

This has been mentioned at least six times now just in this thread. The .NET packages are independent of one another.

Could this possibly change in the future I have 3-4 apps related to FSX that I use when I fly but they will not open now because I removed .NET v4.5 to run the Fibre Accellerator.

There was no reason to do that! Apparently many users are confused by just what .NET is, and how it works. You can have multiple versions if .NET installed on the same computer at the same time.

 

If you already have .NET version 4.5, but are missing version 3.5 SP1, (that the fibre accelerator requires), then just install the missing .NET version 3.5.

 

Again, you DO NOT have to uninstall version 4.5, first, in order to also have version 3.5 - the two different versions can both exist at the same time - they do not conflict with each other.

I guess my test wasn't by the book benchmarking, that I would agree.

 

I just don't get that if I turn FA off why it would not fully return to my "regular" fps experience in high performance areas such as non-urban, mountain areas.

 

I'll keep plugging away.

An amazing product. I tried and purchased it. It works great for me!

BTW: amazing thread also. Never thought I would see Pete Dowson's own cfg-file!

(page 17, #412:)  B)

http://forum.avsim.net/topic/445641-fsx-fiber-accelerator/page-17#entry3025101

 

During last May and June my FSX-live consisted of experimenting on a complex sim situation varying the usual suspects: fiber frame fractions, buffer pools, cpu affinities and priorities as well as int/ext frame limitations.

Nothing worked out as it should and it was (again) like in the movie "Groundhog Day"; a never ending story. The GPU stalls and stutters remained.

I just told my wife that I had given up tuning my sim, when the "Fiber Frame Accelerator" was published.

After replacing my (relatively o.k.) cfg with a freshly bojote-tuned one (changing inter alia the frame rate limit from 0 to 60) and restoring the Nvidia-FSX-setup to its initial state (no Vsync limitations) this FSPS app worked like a charm. I used 30 as the lower limt and will try the 20 vs. 40 recommendation posted by FSPS.

 

FWIMC: my test scenario that kept me going for two month was the following:

KNPA (Pensacola, FL, USA), April 30th 2014, 11:05Z (dawn), RWY25L, ASN WX (historical weather)

Route:

RWY 25L take off to  KNPA280/12 (artificial waypoint) 2500ft, left turn HDG 160, 1500ft HDG100, SPD170kt, intercept LOC07L, G/S07L (tailwind), go around @50ftAGL, missed approach (FMC routing) right turn inbound SIDNY holding, 2500ft (teardrop entry), exit holding, HDG340, descent 1500 ft, 2nd ILS07L approach, landing, vacate RWY07L at end, re-programme FMC, RWY25L departure and so on...

 

This scenario will be great for testing the PMDG/ASN WXR, as it is full of precipitation, CBs and TS (1500ft base to FL400+ tops), microbursts (fasten your seat belts!) and so on.

Many thanks to FSPS, as I will now be able to enjoy this complex WX scenario with almost "0 frustration" at "30+ framerates"!

 

Greetings,

Claus

I just don't get that if I turn FA off why it would not fully return to my "regular" fps experience in high performance areas such as non-urban, mountain areas.

 I suspect, and just an educated guess really, is that FSX was still running with the settings that it was getting before FA was terminated. Kinda sorta seen this behavior in my testing of it.

Another report of my experience.

 

Please take note that I have what could be referred to as a low end system.  i7920, computer specs in my profile.

 

I have been "tweaking" the settings for FA.  I just completed a flight in horrible weather, if have REX Essentials HD with full HD textures loaded.  A very smooth fight holding at 20 frames, no blurries, no microstutters, nice experience.

 

I found that it appears that the "TARGET" frames setting could be crutial to a good experience.  I was following the manual and using a high setting of 50 and even 60 fps target,  I just lowered the TARGET to only 40 frames, ran with the lower frames at 16.

 

If I raise the TARGET to 50+ I get, blurries, microstutters and a frames that jumps around.

 

I don't know exactly what is happening, I'm just saying that on my lower end system it seems that I should not push the TARGET beyond 40.

 

Off for more testing now in some higher demanding areas.

I just lowered the TARGET to only 40 frames

 

I just did a test on "30 vs. 60" and "20 vs. 40" and got a similar result:

the higher values lead to some blurries (here and there), sometimes "a stutter" (but not periodically), a flow interruption (rate variance) and slightly higher knob/switch/TrackIR/EZCA latency.

So on lower end machines (like mine) the "lower-20 and upper-40"-recommendation as posted by FSPS makes a lot of sense.

 

(Edit: this results in a robust overall performance in terms of "look and feel". Which means that FSX would now even get my mother in law's approval; which is a hard test...    )

 

It is really a great joy to see FSX running smoothly under the worst scenario conditions... Great innovation and product!

 

Greetings,

Claus

Edited July 9, 201411 yr by vr-pilot

Amazing to think that FSPS knows best for it's own software, isn't it?

I think Im just gonna wait awhile and see what the story is when the dust settles.   My PC runs fine.  

Definitely I will buy FA but not now. Nice job FSPS team.