https://i.imgur.com/BpIVAyR.png

https://farm5.staticflickr.com/4903/46287175114_786b163a18_o.png

djmcave:

I think the idea is letting the benchmark run for longer and catch some thermal throttle.

Agree. Along with removing any variance that may occur due to the test being to short like single core boost clocks etc. I see no issue having a longer running test it could be a bit more accurate albeit I doubt its going to make a large change in the accuracy.

ViTi95:

https://i.imgur.com/BpIVAyR.png

Is that a dual chip?

https://ibb.co/Kqf39jB 135 cb (2600K) 89 fps (gtx1060/6 GB) no hyperthreading, virtualization enabled, default values of freq. and voltages with HT, CPU score rises to 172 (all cores) ///edit1: OpenGL score ~88 with CPU HT enabled/// Looks like I bought me an excellent card - runs a few frames slower than an average 1080. Thanx Gods I did not haste at the time I was upgrading from 770 (almost opted for a Radeon, since there was a kind of momentary shortage of 1060/6 gigas so I was offered to choose between one with 3 GB and some AMD card, don't remember which model. 😎). Luckilly, I said "No! I'll wait." I was a DAMN fool once - only once in this regard - when i had purchased Powercolor X1600 pro, quite a few years ago. I don't have the nerves to describe my nightmares caused by this card - simply, it was Power-less and Color-less. A mess. Never again an AMD card, unless it is wrapped up in a 1000 Euro banknote and given to me as a present (if it ever happens, I'll buy me a nice KnowVidia card for that money and call the guys who are collecting the electronic waste to take away...you know what). On another note: why was it so hard to make this program detect OS correctly ("win 8", instead of 10)?. I think I'll do some reverse engineering here...Yeah, yeah, I know it isn't crucial, but still leaves a bit of a bitter taste in one's mouth;) edit2: NOW it reports OS correctly: https://ibb.co/Kqf39jB https://i.ibb.co/ncY5Ds4/cinebench.jpg

Hello! I'm new here. Here is my result. https://i.ibb.co/vZ5vmM8/ANN-CINEBENCH-R15-EXTREME-Unofficial-MOD-for-the-community-to-Stress-HEDT-platforms-Page-2-guru3-D-Forums-Google-Chrome.jpg

Dimer81:

Hello! I'm new here. Here is my result. https://i.ibb.co/vZ5vmM8/ANN-CINEBENCH-R15-EXTREME-Unofficial-MOD-for-the-community-to-Stress-HEDT-platforms-Page-2-guru3-D-Forums-Google-Chrome.jpg

For you even this extreme benchmark is not that extreme. 😀

9900k @ 5,0 Ghz all core (with some things running at the background); http://rw-racing.nl/GoT/Cinebench_15_EE_9900k%20at%205Ghz.PNG Single core score is pretty poor? (the MP ratio is higher than i expected) Would be cool for notebooks to have an option to run this test like 20 times to see the differance in score after each test?.

Undying:

For you even this extreme benchmark is not that extreme. 😀

Normal test cinebench r15 scores 7400 points )

I am planing to make Mobile version for it, it will run for few min before the final run with the actual R15 score after the cooling will beg for mercy 🙂 This whey you gonna be able to tell which model will be better for long workloads/has better cooling that won't limit the CPU.

update my score with my ryzen at 4Ghz https://farm5.staticflickr.com/4902/47016178781_7c3b1a47ca_o.png

i9 7900X 507 cb @4.5 GHz 10 threads 659 cb @4.5GHz 20 threads https://i96.photobucket.com/albums/l169/JaxMacFL/cene%20ext_zpsx9dbnzqo.png

AMD Ryzen Threadripper 1950X @ 4.0Ghz Normal run 3417 Modded run 918 https://i.imgur.com/Ip6Xx9W.png

http://towercat.com/Cinebench-extreme 501.PNG Ryzen 7 1700X at 4200Mhz with 1.42Volts, DDR4-3400 16-17-17-17-36. Could do 501 points a few times, but not more.

I have almost finished another MOD for notebook pc's to test if they suffer from thermal throttling: Xiaomi Notebook Pro has great cooling so I need to test it on other laptop;), I will try to test Yoga 730 with i5 8265U. Scores are same as official R15, just CPU test is the same test and the single core test replace with MP core stress with x4 scene ("Performing (Sustained performance) Render Test ... Preparing Project-It may take few Minutes") and after that it will run regular MP test, if cooling is weak it will shown in "CPU (Sustained performance)" score. P.S: before "CPU" test it will cool down for 15 sec: "Performing Render Test ... Preparing Project", so do't worry that the render doesn't start immediately. *Don't pay attention to the other single core scores in the screenshot - I will delete them. Edit: Download Link for CINEBENCH R15 For Mobile if you like to test and report(again you can only copy the "Plugins" folder to mod your own CB R15). https://ufile.io/ysr51 https://i.postimg.cc/L6XTLj04/CB-R15-For-Moble-Xiaomi-Note-Book-Pro.jpg If there will be an Interest I can open another thread for this version.

Damn this is very hard to run...I can't run my usual 4.7ghz for benching without the Temps go outside of my comfort zone..So 4.6ghz is my OC with this bench.. https://i.imgur.com/cTngP7El.jpg

Thanks for the update! https://image.noelshack.com/fichiers/2019/07/2/1549963797-clipboard01.png

This is currently a competition benchmark at Hwbot, check it out here: https://hwbot.org/competition/Pro OC 2019 Round 1/stage/4531_cinebench_r15_extreme

Posted on overclock.net on CINEBENCH R15 EXTREME thread- enjoy ;-).

Linus used this mod on the W-3175X PC un-boxing stream:D -jump to 1:38:00 minute: [youtube=fDkWNgmKCcE] Scored 1152 pts W/O OC. https://i.postimg.cc/2S44n0yC/CB-R15-EXTREME-On-Linus.jpg

Undying:

Is that a dual chip?

Yes, it's a dual socket HP Workstation XW6400