6 hours ago, Dirk98 said:

I'm on I7 6700K 4 Cores HT On 8 LPs. Currently for P3Dv4.2 I use AM 11.11.11.01 Everything else related to P3D I put on 10.10.00.00 each. 

So are you saying 11.11.11.11 for Prepar3D.exe may work even better in my case? Did I get it right, Steve?

Thanks

Hi, Yes with four cores we might have the option to run four cores with little addon exe activity or use three cores for the sim with a heavy addon exe setup.

So having decided to go for four when we only have four it is best to keep other activity off the main core - that oasis of uninterrupted processing with 01 on it. Yes we can allow the second task start on core zero with 11 on the right. It's not too bad better than FSX and P3D v3. It does take a tiny shade off the main task but introduces more backend power.

You got it right.

Adding more ones increases the loading speed and eventually decreases the fps and introduces stutter. A bit like increasing FFTF too far.

Another thing to worry about is heat. If we enable all the ones (HT=On, no AM) it will be a lot hotter than with half the ones *during the scenario loading*. Because it's pulling it all in more quickly if there's overhead (see later).

01,01,01,01 uses four cores

11,11,11,11 uses four cores

The backend of P3D, or a video encode will put two tasks per core - halving their effectiveness as individuals, but tops out the core if one does not.

So It is only of benefit if the core is not topping out for very long - the double 11 will intensify that work done - only then and only *intensify*.

 

The fact is that we can't increase fps no matter how careful our setup - but we can easily decrease it or destabilise it without proper care in the setup.

Those looking for more fps need a higher clock rate or lower the sliders to decrease the work done per frame -  the only way to increase fps.

 

With Dirk's 11,11,11,01 he will see a small drop in fps changing to 11,11,11,11 but a small gain in loading.

So if we went from 11,11,11,11 to 11,11,11,01 we get a slight lift in fps because we took some load off the main core - same as reducing the sliders. We only gain fps back from the outright capability.

Phew, from the mail I get I think we are getting there - a general better understanding of this tricky subject is materialising. It's not a big deal we don't double our fps or anything like that we just get a smoother ride.

Looking at the graphs I posted, showing for the first time outside the lab, how changing FFTF works, gives insight into the idea that we can prevent *losing fps* when we don't need the maximised scenery loading speed.

But looking at the 0.9 graph we see a slight rise in fps - don't worry about it - it is because the sim is stuttering, fps is running away periodically with no smooth linkage to that data.

On 3/25/2018 at 6:34 AM, HUSSAR said:

Hi Edward,

Are you installing using full admin rights and not installing from a separate drive?  Did you purchase the version for Prepar3Dv3 by mistake?  Highly unusual for this utility not to work on your system.  

With the help of the developer the problem has been solved. Looks like the culprit was PMDG RAASPRO corrupted dll file. 

I did the flight on PMDG777 from FlyTampa EHAM with ASCA and all other addons running and I did not notice any freezes or stutters. With locked fps to 28 P3D4 was very smooth. I hope it will stay that way. Will try to put more stress on the sim to see how it perform. 

Thanks for help  

This thread seems to have morphed (as often seems to be the case) into a discussion about AM rather than the product posted by the OP. Should we not be listening to the results those who actually have the app rather than just speculating on why it might not work without any personal experience? There are plenty of other threads about AM but only this one, so far, about FFTF Dynamic.

Edited March 28, 20188 yr by vortex681

1 minute ago, vortex681 said:

This thread seems to have morphed (as often seems to be the case) into a discussion about AM rather than the product

It always does.

Can we read too much into the fact that very few users have reported their experiences with the FTF Dynamic utility? Maybe the size of the thread is acting as a deterrent and consequently any useful contribution might get lost amongst the trees. However, to date there seems to be little motivation for starting a new thread.

I tend to disagree about the AM infiltration of this thread. We all recognise the potential value of using the correct system-specific Affinity Mask with Prepar3D and, quite evidently, much misunderstanding still persists how best to go about it. If our installations are not set up optimally then how can we expect the use of the FFTF Dynamic uitility to improve on our experience?

Mike

16 minutes ago, Cruachan said:

Can we read too much into the fact that very few users have reported their experiences with the FTF Dynamic utility? Maybe the size of the thread is acting as a deterrent and consequently any useful contribution might get lost amongst the trees. However, to date there seems to be little motivation for starting a new thread.

I tend to disagree about the AM infiltration of this thread. We all recognise the potential value of using the correct system-specific Affinity Mask with Prepar3D and, quite evidently, much misunderstanding still persists how best to go about it. If our installations are not set up optimally then how can we expect the use of the FFTF Dynamic uitility to improve on our experience?

Mike

Agreed. I have not seen any data suggesting anything does anything about anything.

On 3/25/2018 at 5:47 AM, SteveW said:

Sure enough looks like the 0.33 value is the same as no value. This is in Unlimited, no Vsync, No TB. Misty rainy view cross cockpit to tower, lots of clouds thunder, light buildings and autogen.

Steve,

Am not sure how this works, when anyone sets frames to unlimited the sim is setting FFTF to 0.01 its not drawn or shows in the config file but thats what its using.

I cannot help thinking if the sim is "writing" a FFTF to 0.01 with unlimited then when someone adds a FFTF line in the config of 0.3 or anything to me it seems like you will without seeing it have two entry for for 0.01 that unlimited "adds" and another one!, sounds like a conflict to me.

I've never added a FFTF with FPS set at unlimited only when locked.

You say 0.33 is the same as no value but set FPS from unlimited to 60 and you will lose FPS as it then kicks in 0.30 FFTF that again you cannot see. Anyone can set up a flight run around 50 fps max when unlimited is set and then set FPS to 60 and you see your fps are less than unlimited because of the 0.30.

Setting a FFTF with unlimited to me sounds like having 2 entries fighting against themselves.

But then am old and have a simple mind 

As for AM only for addons

For me I let the sim have 11,11,11,11 and put everything else on 11,10,10,10 or put another way addons are kicked of the main processes 1-2-3.

 

Edited March 28, 20188 yr by Nyxx

16 minutes ago, Nyxx said:

Steve,

Am not sure how this works, when anyone sets frames to unlimited the sim is setting FFTF to 0.01 its not drawn or shows in the config file but thats what its using.

I cannot help thinking if the sim is "writing" a FFTF to 0.01 with unlimited then when someone adds a FFTF line in the config of 0.3 or anything to me it seems like you will without seeing it have two entry for for 0.01 that unlimited "adds" and another one!, sounds like a conflict to me.

I've never added a FFTF with FPS set at unlimited only when locked.

You say 0.33 is the same as no value but set FPS from unlimited to 60 and you will lose FPS as it then kicks in 0.30 FFTF that again you cannot see. Anyone can set up a flight run around 50 fps max when unlimited is set and then set FPS to 60 and you see your fps are less than unlimited because of the 0.30.

Setting a FFTF with unlimited to me sounds like having 2 entries fighting against themselves.

But then am old and have a simple mind 

 

What do the graphs show you? The change in FFTF shows a change in the graph. The 0.33 graph (three of each in the test) is the same as the graph with no FFTF value added into the cfg. Unlimited Vsync=off.

It's very complicated to measure the effect and not measured directly. For example, we don't see the earth encircling the sun in an ellipse. But we can measure the distance as it increases and decreases throughout the year which gives us the ellipse.

 

Edited March 28, 20188 yr by SteveW

Yes Steve I looked at the graphs but its not what I see when I try the unlimited setting and the locked setting. Everytime I do a test with setting that do not let FPS go over 50 I see a drop with FPS locked. It's the very reason people say locking FPS gives a FPS hit. Because it does.It's also why people prefer to run unlimited when they do not know adding FFTF to 0.01 with FPS locked gives them back the FPS they lose by locking FPS.

BTW am not arguing with you since its a written words am just discussing nothing more.

Edited March 28, 20188 yr by Nyxx

What you are seeing is what your system resolves from your complex setup and addons.

You can make the same test and repeat it as I have with available commercial tools and get the same result every time.

As I have repeated many times the affect on your fps is a result of a lot of things playing together. You can only reduce fps or attempt to stop that reduction. Setting Vsync=On trains the output to the monitor refresh and so you won't see the result you are looking for that way.

The graphs speak for themselves.

 

47 minutes ago, Adrian123 said:

Agreed. I have not seen any data suggesting anything does anything about anything.

You can see the FFTF value changes the way the sim loads data - that's pretty clear in the manual but hard to reproduce as a graphic.

FFTF Dynamic is being professionally tested and you guys must wait for the results.

Sorry about the mix up with the source of quotes in this post. I'm not sure how it happened but they should all refer back to Cruachan's last post above.

1 hour ago, Adrian123 said:

Can we read too much into the fact that very few users have reported their experiences with the FTF Dynamic utility? Maybe the size of the thread is acting as a deterrent and consequently any useful contribution might get lost amongst the trees. However, to date there seems to be little motivation for starting a new thread.

You're correct that there don't seem to be that many people who've actually tried it (and therefore have a valuable opinion). There's a lot of discussion about the potential value of changing FFTF but not so much about how successful (or otherwise) the FFTF Dynamic app is at doing so.

1 hour ago, Cruachan said:

I tend to disagree about the AM infiltration of this thread.

Really? I must be looking at a different thread to you! There are seven posts about AM or the use of different cores on the last two pages alone. None of them make any reference to the product in the thread title.

1 hour ago, Adrian123 said:

We all recognise the potential value of using the correct system-specific Affinity Mask with Prepar3D and, quite evidently, much misunderstanding still persists how best to go about it.

I completely agree, but there are many other threads which go into endless detail about AM. That's the place to ask questions about core usage or AM. This thread is supposed to be about FFTF Dynamic!

1 hour ago, Adrian123 said:

If our installations are not set up optimally then how can we expect the use of the FFTF Dynamic uitility to improve on our experience?

Your comment could equally apply to almost every aspect of P3D and all of it's associated apps and add-ons. There are many other threads and guides about how to set up your system for P3D without anyone needing to divert this thread. I don't know about you but I'm genuinely interested to find out how useful the FSPS app is from those who've actually tried it.

Edited March 28, 20188 yr by vortex681

Can we drop this argument please? That's what I call a diversion.