January 5, 20197 yr 12 hours ago, glider1 said: It is a bit strange but in my case trying to overclock three different CPU's over the years (i2600k, i6700k, i7700k), I was never able to get a higher more stable overclock with HT-OFF compared to on. It is just that HT-ON always ran hotter. The CPU clock frequency hit the same limits either way. Agreed. With HT off I see significant slowing of terrain texture loading, so as far as I am concerned, HT on with a properly set affinity mask [253] is a must. Glenn Ryzen 3700X, X570 Pro Wifi, 32GB 3600mhz RAM, Nvidia Titan Xp "Galactic Empire", RM750x PSU, H700 case, 2x NVMe M2 SSD, 1x SATA SSD
January 5, 20197 yr Moderator 2 hours ago, GHarrall said: Agreed. With HT off I see significant slowing of terrain texture loading, so as far as I am concerned, HT on with a properly set affinity mask [253] is a must. I always chuckle everytime this subject comes up because we always come to the same conclusion. Each system is different and we each decide what works best for us. For me HT off and no AM on my 7700 is the way to go. BUT, for me, I also use Project Lasso to assign cores. Vic RIG#1 - I9 14900K MSI Pro z790 RTX 5070Ti 40" 4K Monitor 3840x2160
January 5, 20197 yr 3 hours ago, GHarrall said: Agreed. With HT off I see significant slowing of terrain texture loading, so as far as I am concerned, HT on with a properly set affinity mask [253] is a must. Try 250, just for the heck of it.. Bert
January 5, 20197 yr 3 hours ago, GHarrall said: Agreed. With HT off I see significant slowing of terrain texture loading, so as far as I am concerned, HT on with a properly set affinity mask [253] is a must. That's what works best for me as well. My mask is different than yours, but that can be an elusive thing to find the one that works best, and requires quite a bit of experimentation, at least it has for me. I don't overclock either, and my CPU never gets very hot with HT on, with my Noctua NH-D15 temps never top about 55C, but some of these guys are perfectly happy running their CPU's at 80C, so to each his own I suppose. Scott
January 5, 20197 yr Commercial Member 50 minutes ago, Bert Pieke said: Try 250, just for the heck of it.. Or 243, then 207. 🙂 cheers! Luke Kolin I make simFDR, the most advanced flight data recorder for FSX, Prepar3D and X-Plane.
January 5, 20197 yr For the little people (i5 users)!!!! The little people have been mocked long enough, well no more!!!! A toast!!!!! ..... to quote the great and fabulous MC Hammer; Can't touch this!!! ROG Maximus X Apex Z370 -- 8086 @ 5.3 / NB 5.0 -- GSkill @ 4133 c17-17-32~Cr1 1.42v -- EVGA 1080Ti 6393 -- ROG PG279Q 1440P 150hz -- Corsair H100i V2 --Samsung EVO 850(s) -- Windows7 Pro 64 --Corsair 750X Ken C
January 5, 20197 yr 26 minutes ago, Luke said: Or 243, then 207. 🙂 cheers! Oh jeez... more AM numbers to try... Just what I need.. lol But hey, you never know until you try them... 😀 Scott
January 5, 20197 yr On 1/3/2019 at 2:18 AM, FunknNasty said: Here you go Mike ... big smile! ...and in case were wondering what 4266 looks like: and a bonus 4300 mhz pict: ...might try it later. 🙂 Hi Ken, No argument there 😊 I agree, pretty Impressive by any standards! This is the best I can do with Quad channel DDR memory clocked to 2666Mhz and CPU (all cores) running at 4.6GHz: But, and it's still a big but, what I have now in terms of available performance seems to be all I need and, quite evidently, with some to spare. This without stressing components unduly. I suppose I could push things a little further, but choose not to as it's likely to invite unwelcome problems and risk curtailing the useful life of the system. Unlike previous builds, this one seems to be keeping pace quit comfortably with most, if not all, currently available gaming and flight simming software, thus obviating that recurring compulsion to upgrade prematurely and needlessly. That will change, of course, just not yet awhile 😉Indeed, I do wonder about the continued relevance of those CPU speed demons as we approach the era of GPU domination. Also, if one can ignore the expense, it's so much easier to upgrade the latter. Regards, Mike
January 5, 20197 yr Any of you guys ever tried Unigine's Valley Benchmark..?? Don't know how useful it would be for evaluating your system performance as it applies to P3D, but it's kind of cool, check it out... https://benchmark.unigine.com/valley Scott
January 5, 20197 yr 4 hours ago, FunknNasty said: For the little people (i5 users)!!!! The little people have been mocked long enough, well no more!!!! A toast!!!!! ..... to quote the great and fabulous MC Hammer; Can't touch this!!! Hi Ken nice work my friend ,checked my 8700k in cpuz bench as it runs in the sim , your new goal must be around +660 not show any teaser have some new galax binned mems 4800mhz xmp Edited January 5, 20197 yr by westman http://
January 6, 20197 yr 5 hours ago, Cruachan said: Hi Ken, No argument there 😊 I agree, pretty Impressive by any standards! This is the best I can do with Quad channel DDR memory clocked to 2666Mhz and CPU (all cores) running at 4.6GHz: Regards,Mike Hey Mike ...thanks for taking the time to post some performance shots of your puter. Nice improvement on that image score ...how'd you do it, or what did you change with your setup. 3 hours ago, SunDevil56 said: Any of you guys ever tried Unigine's Valley Benchmark..?? Don't know how useful it would be for evaluating your system performance as it applies to P3D, but it's kind of cool, check it out... https://benchmark.unigine.com/valley Dude! Post it! -your score! .....I don't care what benchy you use, all that matters is we start posting stuff about our systems. The more flight simmers post benchies of their systems the better off the community will be for the purpose of troubleshooting (thank you, Greg) performance topics and evaluating components to suit our individual needs and wants. 3 hours ago, westman said: Hi Ken nice work my friend ,checked my 8700k in cpuz bench as it runs in the sim , your new goal must be around +660 not show any teaser have some new galax binned mems 4800mhz xmp Thanks man! I'm coming for you! rofl .....I'm up against it ....the wall. That +660 looks like 5660 from my vantage point. WTG on securing that 4800 set. Yes! I wanna see it in action and what it takes to get it to run. ROG Maximus X Apex Z370 -- 8086 @ 5.3 / NB 5.0 -- GSkill @ 4133 c17-17-32~Cr1 1.42v -- EVGA 1080Ti 6393 -- ROG PG279Q 1440P 150hz -- Corsair H100i V2 --Samsung EVO 850(s) -- Windows7 Pro 64 --Corsair 750X Ken C
January 6, 20197 yr 10 hours ago, FunknNasty said: Nice improvement on that image score ...how'd you do it, or what did you change with your setup. Hi Ken, Haven’t a clue. Only thing that changed was the version of RealBench (2.43 -> 2.56) so it looks like some optimisations may have been implemented between the two versions. Regards, Mike
January 6, 20197 yr 11 hours ago, FunknNasty said: Dude! Post it! -your score! .....I don't care what benchy you use, all that matters is we start posting stuff about our systems. The more flight simmers post benchies of their systems the better off the community will be for the purpose of troubleshooting (thank you, Greg) performance topics and evaluating components to suit our individual needs and wants. Here's my ROG CPU-Z bench score, stock i7 8700K, not OC'd, HT on, AM 4095. Below that is my bench from Unigine's latest benchmark app called "Superposition", it is way cool... https://benchmark.unigine.com/superposition Edited January 6, 20197 yr by SunDevil56 Scott
January 6, 20197 yr Commercial Member I thought that ProcessLasso can manage HT off? I remember that it has a special option for HT off when you are setting affinity mask. For me it's a kind of annoying to disable HT (even if it helps P3D smoothness) because I use a lot Adobe apps that works much better with HT on, logically. Current system: ASUS PRIME Z690-P D4, Intel 12900k, 32GB RAM @ 3600mhz, Zotac RTX 3090 Trinity, M2 SSD, Oculus Quest 2.
Archived
This topic is now archived and is closed to further replies.