A Pentium 4 at 4.0 GHz is still much slower than a 3.0 GHz 6700K. Clock speeds aren't really an indicator of processor power anymore. Rather the the instructions per cycle is more important today. The new Intel chips today are able to complete many more instructions each cycle, which makes them quicker today although the number of cycles per second hasn't really moved beyond 4 GHz for a while.

 

Just discovered my several years old 4930k hex seems to be running too hot, maybe 85C max during p3D.. used to be around 70.. i'm assuming this is the result of years of overclocking or maybe the thermal compound needs changed, unsure.. but its had me thinking of swapping to something newer as well.. beyond the whole 980 vs 1080 issue.

 

Apparently late this year Kaby Lake cpu's are due out.. i7-7700k that may have turbos 4.2-4.5ghz..  Or perhaps just jumping to an i7-6700k .. passmark cpumark wise though, i get 16,000 right now on a hex, core speed aside.. the i7-6700 at 4.0ghz is 10,985 not overclocking.  (there is also a broadwell 6850 at 3.6/3.8 with 14,091 cpumarks).. i guess if i switched to a 6700, overall i might see a 4fps drop if its truly only related to the core speed.. i say 4, because in overclocking 3.4 to 4.4 i see 7-9 fps improvement.  

 

It would have been nice to get a cpu and not have to overclock it.. maybe waiting till november for the 7700 is the ticket if I do swap cpus.

 

Back on the GPU.. ran some retests.. 99% gpu on my fsmark07 now.. temps still too hot, 82-83C at the threshold.. 90% fan at that point too.. idle with no fan its 53C, set the fan to 30% and more like 44C.. doesnt matter if i set the fan curve to have more on at idle or not, still reaches critical mass each time in my case with the airflow i have going on.

 

I also ran my barebones tests and my fsmark07 with nothing installed rings in around 45 fps with 3 screens (5760 res).. vs 31 with the sim fully loaded.. ill try a nothing installed comparable on the 980ti to see how that differs, expecting it to be the same though.  I still think there is enough of a jump with typical 3D/VR games over the 980ti to justify keeping the card though, in general.

Just for reference, i had a chance to test two 1080 ftw (hybrid) cards in SLI.. results in this thread.. not a huge difference either, even in VR, mainly advantage being turning on more effects and SGSS without much of a hit.

Maybe your best bet for now would be to pick up another 980Ti for SLI.  It would probably fairly inexpensive now.  I have that setup and gained about 18-20% in FPS over my single 980Ti.

 

Jeff

Maybe your best bet for now would be to pick up another 980Ti for SLI. It would probably fairly inexpensive now. I have that setup and gained about 18-20% in FPS over my single 980Ti.

 

Jeff

I don't have a 980ti anymore, just the 1080 ftw hybrid along with the extra 1080 for the test. Both cards have to be identical for sli

 

Did you change any NI settings to boost things with SLI with p3d?

 I just turned SLI on/off in NCP.  I used the same scenario/location for both SLI on and off and that's all.  I didn't do any more-sophisticated testing, but did try it several times with same results.  I was using default aircraft, ORBX, and MyTraffic 6. You can see my setup in my sig.  My system is running smoothly and I like to hope maybe I have a little extra "juice" if I need it.

 

Jeff

 I just turned SLI on/off in NCP.  I used the same scenario/location for both SLI on and off and that's all.  I didn't do any more-sophisticated testing, but did try it several times with same results.  I was using default aircraft, ORBX, and MyTraffic 6. You can see my setup in my sig.  My system is running smoothly and I like to hope maybe I have a little extra "juice" if I need it.

 

Jeff

Someone in the flyinsidefsx side of things seemed to indicate that VR and flyinsidefsx dont mix.. leading to latency and other issues (and maybe the pan stuttering i'm getting with sli).. are you running VR or just 2D?

 

I disabled sli, it didnt fix the stuttering however.. removing that second card did though.

I just cant hit 90 ATW with 4 SGSS and a single card in VR land.

 

 

Someone in the flyinsidefsx side of things seemed to indicate that VR and flyinsidefsx dont mix.. leading to latency and other issues (and maybe the pan stuttering i'm getting with sli).. are you running VR or just 2D?

 

No VR here.

 

Jeff

Someone in the flyinsidefsx side of things seemed to indicate that VR and flyinsidefsx dont mix.. leading to latency and other issues (and maybe the pan stuttering i'm getting with sli).. are you running VR or just 2D?

 

I disabled sli, it didnt fix the stuttering however.. removing that second card did though.

I just cant hit 90 ATW with 4 SGSS and a single card in VR land.

This should have read, SLI + flyinside not mixing.. however, i came to realize i needed an HB sli adapter to get the full bandwidth, i'm retesting 1080 sli in the next few days (if i can shake a micro stuttering issue i'm not having)

This should have read, SLI + flyinside not mixing.. however, i came to realize i needed an HB sli adapter to get the full bandwidth, i'm retesting 1080 sli in the next few days (if i can shake a micro stuttering issue i'm not having)

I received the SLI HB Evga adapter today.. short story is through my tests, no change, didnt help with head turning/panning stuttering.. neither does frame rate lock at any level.

 

Also.. when you use SLI + 4SGSS + rain (at least rain / clouds), the sim frame rate stays the same, but the ATW drops to 30-40!  yikes.

So as of this writing it appears the 1080 SLI idea is a no go.. will just have to live with no SGSS and semi low frame rate in some areas with 3 layers of clouds + rain (18-25 fps worst case).

I posted similar findings here http://www.avsim.com/topic/489219-1080-gtx-and-p3d/ when the 1080 first came out. I got a set of 2 of them and performance was nearly identical in P3D. Sorry if you were expecting it to change anything for you.