Here are some results I just got from a couple of hours (Jay) trying 4 core 4 jobs vs 5 core 5 jobs, with and without HT enabled. Clearly demonstrates that HT enabled 4 jobs 4 cores is the best result (Mike), HT disabled 5 jobs 5 cores is the worst:

 

 

...how'd it look? Pretty good, smooth sim on Unlimited. I had a repeating scenario that watched the outside view into the side of the plane and beyond. As the AP took the craft into an arc en-route, the sea view becomes dense urban terrain with airport mid ground. Maybe I would say disabling HT introduced some minor snaps in the flow, but was otherwise pretty good. The HT disabled mode induced an overall less zesty performance of the PC and the sim took several seconds longer to load.

My comment about time related only to 12 physical cores. The AM combinations go up exponentially.

Hi Steve,

 

These results certainly support everything you have been saying and I will do as you have asked. I may not be able to conduct these tests until tomorrow as tonight is 'music night!' - a friend and I meet once a week. We both play electric guitar and I play lead to his backing rhythm. After coffee we usually spend an hour or so on the sim so it might be best to avoid disturbing things today. This evening I will be showing off Concorde-X v1.3 (installed yesterday) which is fully compatible with P3D v3.1

 

By the way, when looking at performance you don't want vsync on, just pure unlimited, otherwise you've got a process holding frames.

 

Yes, I did wonder whether VSync and Triple Buffering might be skewing the on-screen observations and masking, to a degree, the actual performance of the CPU.

 

Clearly it would be worth devising a suitable scenario where testing can produce results which can be considered valid and used for proper comparison purposes. Can you suggest a suitable setup combination (situation, weather, vehicle) that you can relate to when you see my results? Not sure how relevant the in-sim scenery settings, etc. are as we all have our individual preferences re. visible eye candy.

 

Regards,

Mike

My comment about time related only to 12 physical cores. The AM combinations go up exponentially.

There are many combinations of unmasking only 4 LPs, but there's no point in testing them all. After a few you'll realise they are all the same. 01,01,01,01,00,00,00,00 works the same as 00,00,01,01,01,01,00,00 works the same as 01,00,01,00,01,00,01,00.

 

 

Clearly it would be worth devising a suitable scenario where testing can produce results which can be considered valid and used for proper comparison purposes.

Don't worry my test setup is pretty fair, precise, no randoms, full-on every sim subsystem, from dynamic weather to Ai injection. Makes no difference what i'm running, full suite of addons or stock sim. Coincidence with addons avoided, the results pan out the same.

Lol! I wasn't referring to your setup, Steve. I'm quite sure you have it nailed down. In fact I was hoping you might suggest a scenario I could use during testing...or does it matter so long as I stick to the same testing scenario for each test run?

 

Mike

Obviously I'm running a totally repeatable scenario Mike, on specially adapted software, with no vsync, no AA, no random weather or Ai, otherwise the tests would be wasted. LOL

 

Lol! I wasn't referring to your setup, Steve.

I know what you meant Mike, you were suggesting to find a way we can both carry out the same test. It doesn't matter. And so I was saying we don't particularly need to, because like me you'll get the same result whatever the test. Start from a saved flight once P3D is up and running, always make several runs before testing. Your fps trace is hard to read but you are basically looking for less spikey behaviour.

Start from a saved flight once P3D is up and running, always make several runs before testing.

 

 

Okay, Steve, will do. Thanks.

 

BTW, downloading IF-10 Pro 22/01/2016 10.015pb08 now as I type :smile: 

 

Regards,

Mike

That's a beast!

 

(BTW, downloading IF-10 Pro 22/01/2016 10.015pb08 now as I type :smile: )

 

 

I use it every flight Mike..A must have for me..

So what is IF 10 Pro ?

Steve

 

What an extremely interesting topic. So much to test, it seems one setting doesn't suit all. Are you able to publish your findings in one document for easy reference. Your research is much appreciated, thanks.

So what is IF 10 Pro ?

 

 

http://www.codelegend.com/idealflight/

There are many combinations of unmasking only 4 LPs, but there's no point in testing them all. After a few you'll realise they are all the same.

That would be true if one didn't also have concurrently test ASN, GSX, TrackIR, STB, the Traffic Optimizer and a few other addons that are executables. A few like Fraps don't enjoy being told what core to use.

I use a I7 4770K running a 4ghz with HT on can anyone tell me if i would be better of with an AF and if so what should it be

 

many thanks

 

Daz try overclocking to 4.5 and setting AF to 116. I achieved a much improved performance with my 4790k and these settings. Use the free Intel xtreme overclocking tool and download various profiles to experiment with.