A_radek Posted September 3, 2017 Posted September 3, 2017 (edited) Fenris, your probably more knowledgable in this area than me. I only read up on what I need to know. Just checked and your absolutely right. With an Avx offset off 2, BoX (or steam VR, like you mentioned) will bring me down to 4,6ghz. Makes me wonder what Avx offset I had set when I did the benchmark as I remember tweaking it back and forth back then. If anything, your voltage in my amateur eyes seems a little high. I'm running 1.25 core voltage in the mode Asus calls "adaptive". Meaning it will use 1.25 at full load and fall down to what ever is the minimum pre programmed (in the CPU) voltage for any given ratio, when idle. According to an Asus article I remember reading (authors claimed to have tested hundreds of i7 7700k's), they came to the conclusion that good 7700k samples could reach a cache frequency within 300mhz of the core clock. If that is true I don't have a good sample. With the cache at 4,5 my system froze repeatedly and sometimes already during Windows boot. Went back to 4,2 and I haven't had the energy to test in betweens. Also haven't tested what effect cache frequency would have in BoX, so don't know if it would be worth it to increase voltage for a faster cache. My current settings are: a cache of 4.2, CPU multi 48, core voltage of 1.25 and Avx offset 0. Again according to that same Asus article 80% of the 7700k samples they tested could be run at those settings. Haven't had a single crash since I settled for this a few weeks ago. Temps during an il2 session are usually around 58c on all cores , with shorter spikes of up to 65c. This is that article I mentioned before: http://edgeup.asus.com/2017/kaby-lake-overclocking-guide/ Asus bios specific, but lot's of good info regarding 7700k's in general. I'm also a multiplayer kind of guy. Only. Mostly fly on Wol and some taw when it's up and running a campaign. Know exactly what you mean about the load on the CPU on a busy multiplayer server. Wish they could find a way to split that main thread. For us VR users that would be bliss. Edited September 3, 2017 by a_radek
SCG_Fenris_Wolf Posted September 6, 2017 Posted September 6, 2017 Hey Radek, thanks for the information as well, very interesting. I assumed that probably more than half of all test results we gathered with chiliwili are factually wrong/incomplete, as they display a core clock that the game does not run on in VR. For example, see at Panther's results - reaching lower results than me with a better GPU, much higher cpu clock. He had 5.0GHz while I think I was running on unknowingly at that time, at 4.7 or 4.8 (not 4.9), and still got better results! That makes no sense to me, unless his cpu was clocking at 4.6 instead of 5.0. I'd recommend chiliwili to actually delete all results and start over, with a requirement to also include AVX offset setting kept at 0. I didn't expect SteamVR + the IL-2 game to pull the AVX throttle, but hey, who did expect that. These things are usually not pulled for games.
TG-55Panthercules Posted September 6, 2017 Posted September 6, 2017 Hey Radek, thanks for the information as well, very interesting. I assumed that probably more than half of all test results we gathered with chiliwili are factually wrong/incomplete, as they display a core clock that the game does not run on in VR. For example, see at Panther's results - reaching lower results than me with a better GPU, much higher cpu clock. He had 5.0GHz while I think I was running on unknowingly at that time, at 4.7 or 4.8 (not 4.9), and still got better results! That makes no sense to me, unless his cpu was clocking at 4.6 instead of 5.0. I've been wondering about that myself recently. I've always had the AVX offset at 0, but because of some fan spiking I was getting at 5.0 and your comments about the 7700k not liking 5.0 in AVX for extended periods I backed off to 4.6 a few days ago to see if it made a difference. It did seem to help with the fan spiking issue. Then in running a few tests regarding OVRDrop yesterday I started seeing results that were as good or better at 4.6 than I had been seeing with 5.0 last week. Unfortunately, due to some RL issues I won't be able to run any more tests for a few days, but I am beginning to think that my earlier 5.0 test results may have been somewhat suspect.
Lensman Posted October 12, 2017 Posted October 12, 2017 I'll start by saying that I can run IL-2 in a RIFT with Ultra settings and 100% smooth frames. To achieve that I use: An ASUS Rampage II mainboard (very old and SATA 2 only). 24 Gb of RAM. Twin Samsung SSDs, Win 10 64 on one, IL-2 on the other. A first generation Titan-X GPU at stock speeds. An Intel i7 975 Extreme CPU overclocked to 4.2 Ghz - this is a VERY OLD CPU!!! Twin USB-3 pci-e boards plugged in. MS Sidewinder Ff 2 MFG Crosswind pedals CH-Products Pro throttle QED I have an ancient CPU and mainboard and a good but elderly GPU and still I'm getting great results.
TUS_Samuel Posted October 13, 2017 Posted October 13, 2017 Lensman, it will be interesting to put your CPU into this list https://docs.google.com/spreadsheets/d/1gJmnz_nVxI6_dG_UYNCCpZVK2-f8NBy-y1gia77Hu_k/edit#gid=1870565807 More info here: https://forum.il2sturmovik.com/topic/29322-measuring-rig-performance-common-baseline/
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now