Fiber Accelerator by FSPS

[Navy Chief]

Anyone know anything about this? Comments, pro or con? NC

http://store.thefsps.com/product_info.php?products_id=54

[Landman]

I was about to post the same question but you beat me to it. You must have gotten their email too. I would like to know about this too. Does it actually work. I bet that it will work differently on different systems but time will tell. So far I have gotten good and steady frames without such software as long as I reduce traffic to 50% in very crowded area like New York, LA, etc. but it would be interesting to see what this product does.

[Meshman]

Rather than be cruel, I would just recommend that you ask the selling party if there is a 100% money-back guarantee. If there isn't, then might I recommend a doubling of your money, which occurs when you fold the bill in half and you slide it back into your pocket.

[Christoph_T]

You can download a Demo as mentioned on the site.
Sounds interesting anyway.

[Pepere]

I received an email form FSPS with a link to a free try before you buy demo but I cannot copy the link...... Have not tried it yet but it sounds very interesting.


http://www.simshack.net/product_info.php?products_id=810



David

[roger-wilco-66]

Rather than be cruel, I would just recommend that you ask the selling party if there is a 100% money-back guarantee. If there isn't, then might I recommend a doubling of your money, which occurs when you fold the bill in half and you slide it back into your pocket.


ROFL!!!

We've seen our share of "wonder-tools", but the approach they use sounds interesting.

Cheers,
Mark

[robcarrich]

Rather than be cruel, I would just recommend that you ask the selling party if there is a 100% money-back guarantee. If there isn't, then might I recommend a doubling of your money, which occurs when you fold the bill in half and you slide it back into your pocket.

Now that is advice I will definitely be following!

[Daube]

There is one member on the French forum that provided some positive feedback.
As advised above, he gave a try to the demo, which is limited to 5 times for 5 minutes each if I understood correctly. And he was very satisfied, so he bought the full product.
From what he said, this tool does not actually improve the performance. It rather adjust dynamically your settings in order to keep the performance as stable as possible.

So it looks like it's definitely worth to try the demo.

[IanP]

He's also just extended the length that the trial can be used for, to a more sensible level...

http://forum.simflight.com/topic/77231-fsx-fiber-accelerator-out-now/

"Please read the new instructions, 5 x 60 minutes demo period extended."

Ian P.

[Bjoern]

It rather adjust dynamically your settings in order to keep the performance as stable as possible.

Approaches into JFK, LGA or EWR would never be the same again. And by "never be the same again" I mean "look like FS98". :dizzy:

[Corky]

Hello,
I'm very interested and watching this topic on other forums too. I have to pass trying this until I can find time to revert my .net Framework to 3.5SP1. From what I've seen and read, it works pretty good to gain around 3fps which is all I gain by overclocking my CPU and vid card anyway. Maybe I can get +6 after this. Did I say I'm very interested?

[Navy Chief]

It will be about three weeks before I get back to simming, but am definitely following these replies! NC

[ak416]

By – in other words – rewritting the FSX graphics engine, FSX now gives to Simmer what he need , when he needs it !
Thats quite a claim, and one that I know for a fact is not possible.

This thing just screams "snake oil" to me. I would avoid this at all costs. I'm not even going to try the demo because who knows what they heck they are actually doing under the hood.

[Daube]

Thats quite a claim, and one that I know for a fact is not possible.

This thing just screams "snake oil" to me. I would avoid this at all costs. I'm not even going to try the demo because who knows what they heck they are actually doing under the hood.

They explain quite well what is done under the hood.
What they wrote is a bit exagerated, but basically they dynamically tweak the main variables of the FSX.cfg in flight. The texture_bandwith_mult, texturemaxload, etc... are adjusted depending on the available ressource. You basically set a lower FPS limit, and the soft reacts as soon as the FPS in game get too close to it to try to prevent them from dropping.

The feedback on the French forum is still positive so far. Some people have downloaded the newest version of the demo, which is extended to 5 x 60 minutes, so we can expect some additionnal feedback soon, I guess.

Also, the guy who bought it has found that if you just launch this soft AFTER you launched FSX, and kill it BEFORE you kill FSX, then your FSX.cfg remain untouched.

[Ferry_vO]

So if I want to fly to some elaborate scenery I just bought it is possible it will (Partially) disappear when I approach..?

[Navy Chief]

Also, the guy who bought it has found that if you just launch this soft AFTER you launched FSX, and kill it BEFORE you kill FSX, then your FSX.cfg remain untouched.


I will make note of this bit of advice! Thanks. NC

[n4gix]

In reality, absolutely no changes are made to any files under any conditions... :encouragement:

[Paul Anderson]

Saw Bill Leaming (n4gix) posted this youtube video link on FlightSim, very interesting.

http://www.youtube.com/watch?feature=player_embedded&v=p6QjTpxe_0E

[Roger]

Interesting idea. It manages certain cfg entries while FsX is running but on closure the FsX.cfg is untouched. The question is for us Dx10ers, does it work with a Dx10 modified cfg with the Fixer applied?

[Navy Chief]

I will definitely be trying the demo, and if it works that good, I will buy it! NC

[TARPSBird]

This sounds like a good FSX improvement for those of us who are too lazy and/or too cheap to upgrade our systems to where the sim can run with maxed-out sliders and top-end FPS. :encouragement:

[Naismith]

I have just been tinkering with the demo and got some quite positive results. I followed the instructions in the read me. I chose my worst performing airport - CYVR (because of all my addons, FSGlobal Mesh, Van+, UT2, FTX) Worst performing Aircraft - the PMDG 737-800. ANd the FSX Thundestorm WX preset.
Normally under these conditions I would avoid YVR like the plague, certainly in that aircraft. I would get fps in single figures and stutters which made it totally unflyable.

My PC for comparison is a 2600K Intel with 16GB RAM and a GTX660 video card. I run FSX in windowed mode (Pseudo full screen) highly tweaked cfg and in DX10 mode. I have the FPS locked at 31.

OK so I fired up FSX, set the parking brake and waited whilst everything settles down and was reading fps of 12.4 or thereabouts in the VC. I took a couple of exterior shots (one with AI in background) and got 14.7 and 16.1 respectively.

I then ran the fiber program and left all the settings at default. The same pics were now. = 13.3, 20.0 & 18.1.
I then pushed the slider down to 15 on the Fiber prog and now the fps were = 14.0, 17.2, 16.1
I then tried the slider on max and FSX froze.

So Cockpit shots gained 1 - 2 frames
Exterior shots without any AI in the picture gained quite between 3 & 6 fps
Exterior with AI showing - no great gain.

Not very scientific I know but for me it was marginal whether or not I would consider buying it.

[Naismith]

Update - Just had a session flying around the Fraser Valley and YVR. FPS are improved in a 2-6 fps region within the VC. But it still stutters badly in the PMDG 737 especially flying west toward YVR, not so bad (micro stutters) flying away from it. Useable more so than without it but I'm not convinced the program is for me. If I were to rate it I would give it 5/10.

[Rimshot]

Interesting idea. It manages certain cfg entries while FsX is running but on closure the FsX.cfg is untouched.

As I understand it the fsx.cfg file is not touched at all! I've tried the demo and managed to get 3-5 frames increase with the Flight1 C182T at KSEA in heavy thunderstorms. My pc is just to darn old, because I still couldn't reach the 10 fps limit in that situation...

[Navy Chief]

This sounds like a good FSX improvement for those of us who are too lazy and/or too cheap to upgrade our systems to where the sim can run with maxed-out sliders and top-end FPS. :encouragement:


Roger that! NC

[Roger]

As I understand it the fsx.cfg file is not touched at all! I've tried the demo and managed to get 3-5 frames increase with the Flight1 C182T at KSEA in heavy thunderstorms. My pc is just to darn old, because I still couldn't reach the 10 fps limit in that situation...

Hi Bert,
Interesting findings. I've just read through the entire Avsim thread (most posts complaining about the demo!) and it seems that the best results come from limiting frames in FsX to a value that you can't normally achieve very often and then Fiber Acc' will dynamically alter cfg entries to suit the loading. Now as I understand it, the vsync tearing fix is only achieved when frames are limited to half the refresh rate, which in my case is 30 because my monitor refresh is at 60Hz. So if I start changing my limit in FsX then I will see tearing as the NVI setting will not be valid. Can anyone confirm this?

[Roger]

Hi Bert,
Interesting findings. I've just read through the entire Avsim thread (most posts complaining about the demo!) and it seems that the best results come from limiting frames in FsX to a value that you can't normally achieve very often and then Fiber Acc' will dynamically alter cfg entries to suit the loading. Now as I understand it, the vsync tearing fix is only achieved when frames are limited to half the refresh rate, which in my case is 30 because my monitor refresh is at 60Hz. So if I start changing my limit in FsX then I will see tearing as the NVI setting will not be valid. Can anyone confirm this?

Well I asked the same question on the Avsim thread and although I feel this issue requires answers I got no reply, so I suppose this (to me at least) important question will remain unanswered...

[MarkH]

Well I asked the same question on the Avsim thread and although I feel this issue requires answers I got no reply, so I suppose this (to me at least) important question will remain unanswered...

Roger, I am not very knowledgeable on this but I'm pretty sure you will not see tearing. I force vsync on and don't ever see any tearing even though I don't use the 1/2 refresh-rate option and I rarely achieve 30fps.

I think the impact of using vsync is that your actual frame rate will in some circumstances fall to a factor (half or a quarter) of the monitor's refresh rate. (This is regardless of what FSX is reporting its frame rate to be.) My understanding of why this is is that you're instructing your GPU to wait for the monitor, not the other way around. When your frame rate significantly exceeds the refresh rate, the actual frame rate will be limited limited to refresh rate. When your frame rate falls just below the refresh rate, the actual frame rate may fall back to a factor of the refresh rate.

(I stand to be corrected on most of that, except the part where I don't ever see tearing ;) )

[ak416]

They explain quite well what is done under the hood.
What they wrote is a bit exagerated, but basically they dynamically tweak the main variables of the FSX.cfg in flight. The texture_bandwith_mult, texturemaxload, etc... are adjusted depending on the available ressource. You basically set a lower FPS limit, and the soft reacts as soon as the FPS in game get too close to it to try to prevent them from dropping.
They don't explain anything that they do. It says in the product description it is not used, "we directly, in real-time, dynamically inserting FSX tweaks without touching fsx.cfg". The developer also mentioned in another post "We don't touch fsx.cfg - neither read or write . It is completely ignored." He then goes on to mention how they "rewrote the FSX graphics engine".

[Roger]

Roger, I am not very knowledgeable on this but I'm pretty sure you will not see tearing. I force vsync on and don't ever see any tearing even though I don't use the 1/2 refresh-rate option and I rarely achieve 30fps.

I think the impact of using vsync is that your actual frame rate will in some circumstances fall to a factor (half or a quarter) of the monitor's refresh rate. (This is regardless of what FSX is reporting its frame rate to be.) My understanding of why this is is that you're instructing your GPU to wait for the monitor, not the other way around. When your frame rate significantly exceeds the refresh rate, the actual frame rate will be limited limited to refresh rate. When your frame rate falls just below the refresh rate, the actual frame rate may fall back to a factor of the refresh rate.

(I stand to be corrected on most of that, except the part where I don't ever see tearing ;) )

Hi Mark,
It would seem that force vsync doesn't work in Dx10.

[Dangerous Beans]

Frooglesim has done a first impressions video on this

https://www.youtube.com/watch?v=pGalckew8cw
I doubt I'll get it as I fly almost exclusively in P3D now but his results are very interesting.

[DarrenL]

Saw Bill Leaming (n4gix) posted this youtube video link on FlightSim, very interesting.

http://www.youtube.com/watch?feature=player_embedded&v=p6QjTpxe_0E

I'm shocked more by the very low frame rates he has on 4.2Ghz i7 with or without it.

[txnetcop]

After looking at all the information I can't see the value in this software. I have an older Q9650, Nvidia GTX 285, 4GB Corsair Dominator RAM, water-cooled clocked at 4.2GHz that will run circles around his newer computer apparently. My frame rates are consistently above 30-32 everywhere I go...why is his computer so slow and why does he need this software? I run Orbx and REX4 with FSX. I honestly cannot believe this software is needed for any of the newer i7 based computers that are clocked at 3.2 or better.
Ted

[MarkH]

My frame rates are consistently above 30-32 everywhere I go...why is his computer so slow and why does he need this software?

Did you read the part where he says he runs three 27" displays? (And an extra one.) If you are saying your computer will hold 30fps with a PMDG aircraft at KSEA in a thunderstorm on a 6Mp display (possibly 11Mp, he isn't specific) I am going to be sceptical.

[txnetcop]

No I don't use jetliners like PMDG in a Thunderstorm and I never saw a need for three monitors, but I can average 30 fps with A2A aircraft with Accu-Feel on a dual monitor at KSEA in bad wx to include the old Boeing B377.
By the way...there are others in here and in other forums I have built units for than can attest to above avg frame-rates and excellent graphics with the units I build.

[MarkH]

No I don't use jetliners like PMDG in a Thunderstorm and I never saw a need for three monitors

I won't labour the point.

[txnetcop]

I won't labour the point.

Mark I guess this looks like I am glad handing myself...that is not the point. There has been so much SNAKE OIL software out there for FSX that really if you know your system capabilities and don't try to overload them you should be able to enjoy a really nice experience. I still don't see this latest piece of software as the end all-do all for FSX. It is an old engine that is finally coming into it's own with the i5 and i7 series motherboards and CPUs. Still it has some crazy quirks that drive the best hardware and software geeks NUTS! I am so glad Lockheed Martin took the reins and made P3D what it is and it is only getting better.
Ted

[henrystreet]

I fly Iris PC-21, VRS Superbug, Ant's T-28C/D, well you get the picture. Almost all below angels 10 with OpusFSX for weather. Done extensive testing with my rig http://uchisworld.wordpress.com/2014/05/26/happy-anniversary-part-2/ been working with computers professionally since the late 1970's yadayadayada.... bottom line: my results with the demo (at KSEA, as depicted in my website, and at Hobart Tasmania by Orbx, are almost identical to Pete Dowson's http://forum.avsim.net/topic/445641-fsx-fiber-accelerator/

My frames did not improve more than 10% but OMG the smoothness factor was much better. And, even before reading Pete's comments, I was able to increase the detail sliders with no loss of FPS or smoothness.

There is a demo you know and based on how it worked, I am purchasing.

[MarkH]

Pete Dowson's http://forum.avsim.net/topic/445641-fsx-fiber-accelerator/ [results]

I bet it's going to sell now!

[n4gix]

I'm shocked more by the very low frame rates he has on 4.2Ghz i7 with or without it.

Just to be clear, this is NOT "my video." I simply posted a link to it to help 'spread the word' so to speak.

[Pepere]

I used the demo at JFK and Seattle it helped to get up to 20 FPS, but very low ground detail. Also at low end airport I only now get low end FPS where as before would have had 30 - 40 FPS. It also effects the response time of the GPS and other instruments. Will not buy it! Don't care for FSPS all there addons I've purchased are all uninstalled.

David

[Pepere]

I bet it's going to sell now! Pete Dowson's http://forum.avsim.net/topic/445641-...r-accelerator/ (http://forum.avsim.net/topic/445641-fsx-fiber-accelerator/) [results


Not if you read the full forum on the subject!

David

[MarkH]

Not if you read the full forum on the subject!

The point I was making is that some voices are worth paying more attention to.

[wombat666]

I bet it's going to sell now!

Maybe yes, maybe no, the price is more than the Sim and I can smell Vasolene in the air!
I'll observe before I'll even consider purchasing.
:173go1:

[Roger]

The program needs FSUIPC to be installed. Pete Dawson will have sold a licence to them. Just good business.

[heywooood]

maybe it should be called Fibber Accelerator....

[henrystreet]

Anyone know anything about this? Comments, pro or con? NC

Chief, I have about 3 hours of testing with the product under my belt and have added to my rig as a carefully selected component. I have researched the way it works and have monitored its activities with processmon and other tools. It is the real deal but perhaps not for every computer or evidently not every user.

PM me if you would like to hear more of my experience but I must warn you, I have no conspiracies to hide.

[Navy Chief]

Chief, I have about 3 hours of testing with the product under my belt and have added to my rig as a carefully selected component. I have researched the way it works and have monitored its activities with processmon and other tools. It is the real deal but perhaps not for every computer or evidently not every user.

PM me if you would like to hear more of my experience but I must warn you, I have no conspiracies to hide.

Henry, hopefully by the end of next week I will have some time to get my system up and running again. Possibly the week after. We will have a LOT of boxes to unpack, following the move. But I definitely want to check this out. Sounds promising! Thanks. Chief.

[henrystreet]

We will have a LOT of boxes to unpack, following the move.

Good luck with getting settled in!

[Bjoern]

There has been so much SNAKE OIL software out there for FSX that really if you know your system capabilities and don't try to overload them you should be able to enjoy a really nice experience.

Word.


If you get a consistent 10+ FPS in the worst of situations with maximum permittable detail for your system, you're good to go.

[n4gix]

The program needs FSUIPC to be installed. Pete Dawson will have sold a licence to them. Just good business.
That's borderline libel. Only the FREE version of FSUIPC is required. Pete makes nothing at all from FREE... :173go1:

[MarkH]

That's borderline libel. Only the FREE version of FSUIPC is required. Pete makes nothing at all from FREE... :173go1:

I understood Roger to be saying the developer buys a licence to use FSUIPC in its app. I don't agree with his analysis of Pete's test report, of course.

[Roger]

That's borderline libel. Only the FREE version of FSUIPC is required. Pete makes nothing at all from FREE... :173go1:

To associate the Fiber program with Pete's program didn't cost anything? Maybe not but I could understand if it did. As I said before, it would be just good business...and I don't like that hair pull smiley...don't use it again in a post replying to me please!

[txnetcop]

Chief, I have about 3 hours of testing with the product under my belt and have added to my rig as a carefully selected component. I have researched the way it works and have monitored its activities with processmon and other tools. It is the real deal but perhaps not for every computer or evidently not every user.

PM me if you would like to hear more of my experience but I must warn you, I have no conspiracies to hide.

Well I used the demo to the limit of my time on two separate systems. I do understand the concept and it isn't a bad premise to work from since FSX is so CPU intensive. However I used it on my old 775 Q9650 4.2GHz which was handling FSX with all kinds of slop for weather just fine...it actually got worse I am glad I save my cfg file! I also ran it on an i7 920 4.5GHz, MSI GTX 660Ti system I just built for some kids at the Rec Center, better but I didn't see much difference, at least not enough to warrant paying the asking price. I did approach it with the same professional demeanor as I had while testing at TechCor(now defunct). No sale for me...at least not with the current version.
Ted

[stansdds]

Well I used the demo to the limit of my time on two separate systems. I do understand the concept and it isn't a bad premise to work from since FSX is so CPU intensive. However I used it on my old 775 Q9650 4.2GHz which was handling FSX with all kinds of slop for weather just fine...it actually got worse I am glad I save my cfg file! I also ran it on an i7 920 4.5GHz, MSI GTX 660Ti system I just built for some kids at the Rec Center, better but I didn't see much difference, at least not enough to warrant paying the asking price. I did approach it with the same professional demeanor as I had while testing at TechCor(now defunct). No sale for me...at least not with the current version.
Ted

I am glad you posted. My rig is a Q9650 overclocked to 3.8GHz. For me, cooling becomes an issue when I push past 3.8GHz and run Orthos. I've considered Fiber Accelerator, but the reviews seem very mixed, many reporting blurry scenery in areas of heavy traffic and scenery. I'm thinking I'll not buy this one.

[n4gix]

To associate the Fiber program with Pete's program didn't cost anything? Maybe not but I could understand if it did. As I said before, it would be just good business...and I don't like that hair pull smiley...don't use it again in a post replying to me please!
FSUIPC is not included in the installer, so no payment to Pete was required for a single use license AFAIK. As for the 'hair pull smiley' I have no way of knowing in advance what anyone might like or dislike, but I am sorry to have hit your dislike trigger.

As an "Administrator" perhaps you could remove that one and replace it with one you do like, or at least don't dislike? <no smiley>

[n4gix]

...it actually got worse I am glad I save my cfg file!

Since it never even reads the FSX.cfg file, much less makes any changes to it, what possible difference did saving your cfg file make? Just curious as I've not even tried the Fiber Accelerator myself., I pretty much keep my FSX/A installation as "stock" as possible for development reasons, so with average fps well in excess of 80 there's no need for it.

[txnetcop]

Since it never even reads the FSX.cfg file, much less makes any changes to it, what possible difference did saving your cfg file make? Just curious as I've not even tried the Fiber Accelerator myself., I pretty much keep my FSX/A installation as "stock" as possible for development reasons, so with average fps well in excess of 80 there's no need for it.

Since I was not sure whether or not it made changes to the cfg I saved it anyway. After I erased the program everything went back to normal. I guess I should have added that...OK?
Ted

[henrystreet]

I do understand the concept and it isn't a bad premise to work from since FSX is so CPU intensive.
Ted

The developer, while simultaneously trying to protect his proprietary information and marketing the product, has given several very specific hints as to how the program works. You may be that one user who understands the concept, but most folks that are trying the demo are not even reading what the developer has said about the product. The software in no way modifies or even references fsx.cfg. I have confirmed this. The factors that it changes dynamically can however be influenced by the initial settings of FSX startup which is determined by the fsx.cfg. The software does not appear to use a wide range of tweaks, only a few.

Perhaps, those trying the demo are only seeing the promises and not, again, the very specific hints as to how the software does what it does. Also the developer gives no good guidelines (but I repeat, several good hints) about how to set the FPS sliders in the application.

I would have predicted that your higher end system would benefit less from FA. And, would have made some suggestions for the FPS sliders based on whatever your current performance might be. Based on my reading of the developer's words (and ignoring his marketing), my testing, and what others, like you, have posted about no improvements, there are also certain configurations, such as photoscenery that I would predict benefit very little from FA.

Perhaps the devs are "over-marketing" or perhaps they are using their support system to help tune customers FA config. These factors have not been subject to my research or testing.

[wombat666]

Long story short.
Some people will purchase this program and some people won't.
The believers will be happy and claim it is brilliant and those who are disappointed with it will cry foul.
It is a free market, and choice is the operative term here.
Me, myself, I'll wait and see how it goes but my solution is continuing to build my next rig.
And I never ever buy the first release of anything, be it software, hardware or a new car.
:lemo:

[henrystreet]

For those still wanting to try Fiber Accelerator, version 1.1 is out. Also the Fiber Accelerator thread at Avsim has some good discussions of settings and conditions that agree with all my testing. Specifically how the sliders and FSX starting conditions affect your experience. Skip to last 3 pages or so of the discussion for the most succinct posts.

[n4gix]

The developer, while simultaneously trying to protect his proprietary information and marketing the product, has given several very specific hints as to how the program works. You may be that one user who understands the concept, but most folks that are trying the demo are not even reading what the developer has said about the product. The software in no way modifies or even references fsx.cfg. I have confirmed this. The factors that it changes dynamically can however be influenced by the initial settings of FSX startup which is determined by the fsx.cfg. The software does not appear to use a wide range of tweaks, only a few.
That was very well put, Henry. The fact is that the fsx.cfg file is read precisely once during the initial load of FSX. The parameters in that configuration file are loaded into memory locations.

Any old-school BASIC programmer will remember how one might peek and poke specific volatile memory addresses to read and modify the contents. These days of course such brute force methods won't work, but there are plethora methods one might use (or possibly abuse?) various APIs to accomplish the same results...

In any case the critical take-away from the above is that the memory is volatile, meaning that once unloaded the changes made while running vanish like a poof of smoke.

[henrystreet]

That was very well put, Henry. The fact is that the fsx.cfg file is read precisely once during the initial load of FSX.


Yes Bill thanks for the correction. My filemon logs for FA do show reading fsx.cfg at the first launch...but only one log entry.

[Adamski_NZ]

As a frequent contributor to that thread myself, I'm appalled (but not surprised) at the average level of comprehension. Even allowing for native English language difficulties, people just don't seem to be READ what's there and *digesting it* before adding still more blather. The total confusion about .NET is a case in point.

From numerous questions aimed directly at the devs (who finally responded) - it appears that F/A doesn't write back to fsx.cfg itself (indeed it simply doesn't use it) - BUT - FSX itself will write/set the target framerate to "UNLIMITED" if you hit the performance buffers. I got the impression that if you quit F/A before FSX, then your original fsx.cfg value is retained.

The problem is people just aren't testing/comparing it properly. My procedure would be:

1) Back up your original fsx.cfg and NVI profile.
2) Create a sample flight (with a variety of scenery areas flown over) and SAVE it. Maybe even use FSRecorder.
3) Install/try F/A with that saved flight (or recording) and compare.
4) Don't just quit FA and see what happens with FSX still running - it will be left in limbo, with a possibly totally inappropriate [temporary] configuration.
5) Quit BOTH and restart FSX and compare.**
6) Start FA and compare.

** You may want to rebuild the textures cache before this step (using a new/random value for SHADER_CACHE_VERSION in fsx.cfg).

Note that DX10 users will need to disable/re-enable their NVI-set framerate limiter (if enabled) and check their vsync settings.

If uninstalling FA, do it properly from the Control Panel and then use a regcleaner to spot any leftovers. I found it uninstalled cleanly.

My gut feeling is that the program "mimics" the way FSX behaves according to your various fsx.cfg settings (and your target framerate in FSX) - almost as if it were creating a new fsx.cfg in memory "on the fly" and using that. Considering the HUGE variation of results obtained by users tweaking fsx.cfg, it's not surprising that there's a similar variation in what FA may (or may not achieve) in any particular FSX setup. Also - I don't think it achieves anything that careful tweaking of fsx.cfg will produce - but how many people have the knowledge to do that (or put in the hours)?

On my quite mediochre system, I found FA didn't improve overall performance much (if at all), but it did allow my to fly at high settings for when I wanted good scenery and then allowed me to knock it back for when approaching a busy airport for landing. Previous to that, I used to have to load one of two fsx.cfg's I have and restart FSX.

It *is* possible (just) to be able to test FA in the limited time available - but you really need to prepare your FSX properly in advance, so as to not blow any FA sessions.

Adam.

[n4gix]

Yes Bill thanks for the correction. My filemon logs for FA do show reading fsx.cfg at the first launch...but only one log entry.
Um, that wasn't a correction so much as a confirmation. When I re-read my previous remarks just now, it seems that I was not clear enough.

It is FSX that reads the fsx.cfg file once. Fiber Accelerator neither reads nor writes to the fsx.cfg file according to the programmer(s). FA won't work with P3Dv2 for the same reason that Pete has to compile a specific version of FSUIPC for it; the memory offsets are different. :bump:

[YoYo]

The new version is out 1.2 now.

+ new demo 5x60 min with new functions.

http://forum.simflight.com/topic/77231-fsx-fiber-accelerator-out-now/#entry469945