May 12, 20169 yr Ya I'm not seeing the problem here, a few days back Schmuckley did a "WR" exposing this flaw again. Did they not see it .? Or as long as it doesn't affect Skylake, we don't care. Sent from my Nexus 5X using Tapatalk
May 13, 20169 yr So since only Skylake is immune, let's disable or freeze XTU points for old gen. Oh wait... Sent from my Nexus 5X using Tapatalk
May 18, 20169 yr no off against ericwei: but is this the prob we are talking about? its nearly impossible to look trough the scores-takes ages http://url.hwbot.org/23VV51a why no mandatory screenshots for xtu Edited May 18, 20169 yr by basco
May 18, 20169 yr Author I posted how to do it and that it doesn't affect skylake.... So let's try it on skylake smh
May 18, 20169 yr Administrators Seems legit - I see skylake xtu scores quite weird lately as well, especially with B-die used. We will see how this topic develops...
May 25, 20169 yr FYI, "Thanks for your feedback, We will try to reproduce and will let you know the status."
May 25, 20169 yr Christian forwarded me some files from @Digg_de which I forwarded to the XTU guys. Do you have a file of a bugged X99 system I could forward as well?
May 25, 20169 yr Author Christian forwarded me some files from @Digg_de which I forwarded to the XTU guys. Do you have a file of a bugged X99 system I could forward as well? I'll get you one sure The more cores there are the harder it is to bug. I dont have bwe yet to try.
May 26, 20169 yr Good to see hwbot working on solution,think xtu is used way to much to give up on this benchmark,for the record I think it's really great tweak able benchmark that responds to memory tweaking on most platforms,I know we have seen couple lucky runs on skylake.I myself have found xtu to be consistent on haswell and skylake never no lucky runs like say hwbp Same thing applies tho you can never stop cheating,you can only hope most people are honest
August 7, 20177 yr How did the OP really get over 10GB out of a i7-4770K? My PC powered by a i7-4790K crashes when adjusting the ratio to x47 & am lucky to break the 1,200 mark at times, believe my personal record is around 1221 (approximate). Came close today, within 10 points, yet cannot reproduce whatever made my system run at 4.8Ghz, which would had required tuning to x48 (all cores synced). The 4790K & other Haswell Refresh CPU's (reportedly) has 'better' TIM than the long list prior to the Refresh editions. Cannot remember how I achieved that, if running the Windows based ASUS Ai Tuner, or playing around in the UEFI. Although I believe (IMO), had I bumped the voltage up just a hair (runs at 1.3086V with new setting, up from 1.275 with x44). could had beat my personal record. Speccy specs provided for anyone who wants to take a peek. http://speccy.piriform.com/results/Fr2tn7UgXJAINSczWlbCkru Cooling via Noctua NH-D15, which dropped temps huge under load & allowed me to run the XTU app w/out thermal throttling. Am considering delidding, although should I do, will grab a low cost Pentium unlocked version on eBay, many sold new for $59.99 on promo, so should be able to get one for under $40. Better to damage that than a (at the time) $329 CPU. Also have it's little brother in the i5-4690K, has never BSOD'd, even set at 4.4GHz (x44 ratio tuning). Makes me wonder if H/T is really needed. BTW, while I'm just learning how to OC & performed my first build with all new components in late 2015, the 10th place ranking in the league I'm currently in (within the US) & in the Top 25% of Hardware Masters shows that I do know how to do something right.;-) Cat Edited August 7, 20177 yr by cat1092 Added Info to Post
August 9, 20177 yr @cat1092 this was a bug in the benchmark this is not a real frequency NO one have reach close to that speed. Maybe @Splave could be so kind and teach you some of his magic.
August 10, 20177 yr People scoring over 1800 consistently on 6700k's while still under 5ghz and not running any CRAZY memory settings makes me question if there isn't something going on for skylake too.
August 10, 20177 yr Bugged versions of xtu give you the scores over 1800. I saw even with 4000 12-19-19 score of 1816 ). Sent from my iPhone using Tapatalk Pro
August 19, 20177 yr People scoring over 1800 consistently on 6700k's while still under 5ghz and not running any CRAZY memory settings makes me question if there isn't something going on for skylake too. Yeah, and to think, even with DDR3 RAM, the i7-4790K still edges the 6700K which runs DDR4 according to the Passmark site. Must be something about the RAM, I mean one can go with 64GB & use no more modules in doing so, that would leave one for a huge RAMDisk, if that could factor into a benchmark. I could had probably squeezed 5.0Ghz out of my 4790K, however didn't want to go with liquid cooling. Too much cash invested to lose all due to leakage, or worse, set the place on fire. Liquid & electricity doesn't play well together, and I don't aim to be the next to find out. Cat
February 22, 20187 yr @Splave I have a question for you mate. I recently saw your score with 7820X previously was only like 2800 now you are at 4100. So you must have found some nice tweak. I really have a problem though. With 7920X I managed 4272 in XTU, no special tweaks or anything. With 7960X clocked similarly I can only manage to get 4070 in XTU. Wtf is going on there? Anything I need to look out for on the big chips like software or bios settings?? Other programms like Cinebench show no throttling everything is fine. AVX offset is 0 and so on. I really run out of ideas what causes this. I think I should be more in the region of 4700 XTU with the 16 cores. @ 4.8 and 3300 Mesh. I see others sometime also have bad efficiency but I can't find any solution anywhere now that I'm also affected. Edited February 22, 20187 yr by CSN7
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.