Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[DougW60]

A few weeks ago, we had a post going regarding low frame rates when engaging 8 X Time Compress. I cannot find the post thread so here is the gist of it.

Some of us were experiencing very low frame rates and stuttering when engaging Time Compress and Time Skip.  This begun with and is unique to C.U.P..  Those of us who were experiencing this had older PCs and older GPUs.  One of the discussions surrounded the GPU being the bottleneck but since those reporting this problem had both AMD and nVidea GPUs, I considered the GPU as possible but not absolutely the problem.

I had already planned to upgrade my computer with a new motherboard, CPU, RAM as mine were 6 years old and now I have accomplished this upgrade.

What you see below is a comparison between a CPU, motherboard, RAM from 2009 and a new set up.

FPS recording for IL-2 1946

I7 960 CPU 3.2 ghz – OC 4.1 ghz            I7 6700K CPU stock 4.0 ghz
MSI Eclipse SLI X58 motherboard            ASUS Maximus VIII Hero Z170
SATA II                                            SATA III
Radeon 7970 GPU                                    Radeon 7970 GPU
Acer 27inch 2048 x 1152 monitor            Acer 27inch 2048 x 1152 monitor
Vsync - off                                            Vsync - off
Graphics Settings - Maximum             Graphics Settings - Maximum

Game Load time to Menu: 96 seconds    Game Load time to Menu: 57 seconds

Flight at altitude                                    Flight at altitude

Min – 31                                            Min - 79
Max – 63                                            Max - 187
Avg – 49.4                                            Avg - 102

Carrier Take off                                    Carrier Take off

Min – 17                                            Min - 74
Max – 83                                            Max - 126
Avg – 39.4                                            Avg – 91

Carrier Take off 8x time compression    Carrier Take off 8x time compression

Min – 4                                            Min - 9
Max – 30                                            Max - 79
Avg – 8.1                                            Avg – 13.9
A lot of stuttering (see note below)            Note – there were no stutters

Note: when engaging Time Skip, with the I7 960 there was so much stuttering that it was not worth doing it, going 8x time compress performed better.  With the I7 6700K, we are back to what it used to be / same as to Time Skip under HSFX.

My observations:

With my current monitor, the GPU is able to handle C.U.P. at max settings
Obvious advantage switching from SATA II to SATA III (bandwidth)
Z170 architecture greatly improves data flow

This is just to let others know that C.U.P. does apply greater stress on your CPU compared to other MODs, however, C.U.P. is the best MOD out there (IMHO).  What I observed here is that matching your GPU to your monitor's capability, you have taken care of your needs. You do not need the latest and greatest GPU when your monitor is a 1080p. 

[SAS~Storebror]

Thank you very much Doug, one of the best and most unambiguous comparisons of this topic, much appreciated.
All your results perfectly make sense.
The game load time benefits from bandwidth increase on SATA and RAM, all FPS values benefit from the increased single core performance plus the increased RAM bandwidth.

Some numbers to back up your results:
Single core performance i7-960 vs. i7-6700K@4GHz is about 2500 : 1500 (measured in passmark points, the latter is an estimation of the speed increase to be expected by overclocking, usually this scales at roughly 50% of the overclocking factor on i7-9xx CPUs).
Memory bandwidth (not even Skylake's big point) i7-960 vs. i7-6700K@4GHz is about 26 : 17 (measured in GB/s).
This means that the processing performance of the new CPU is 1.67 times better than the old one and the memory performance is about 1.53 times better.
Both have approximately the same effect, so in total the performance increase to be expected is in the range of factor 2.55.

Your values for comparison are
31 : 79 (factor 2.55),
63 : 187 (factor 2.97),
49.4 : 102 (factor 2.06),
17 : 74 (factor 4.35),
83 : 126 (factor 1.52),
39.4 : 91 (factor 2.31),
4 : 9 (factor 2.25),
30 : 79 (factor 2.63),
13.9 : 8.1 (factor 1.72)

Summarizing all factors, the average factor is 2.48, which fits pretty well.

Best regards - Mike

[DougW60]

Thank you Mike for your reply.

I have seen so many posts where people who are having challenges with their computer's performance are being advised to upgrade their GPUs.  I felt the advice was very narrow and not necessarily right.

Next step is upgrading my monitor and with it will be a new GPU.  Waiting for 4K, IPS panels to increase to 100hz refresh rate which I have been hearing are supposed to come out in a month or two.

By the way, where or how did you get the stats on the CPUs?  I would like be able to compare CPUs like that for future reference.

[DougW60]

Follow up

prior to the upgrade, I was having issues with screen tearing (horizontally) when using TrackIR 5.  Everything that I read indicated it was a combination of my monitor (5ms response time) and the older GPU.

Now the good news - screen tearing completely gone.  Looking left or right is smooth without any flicker or tearing.