Download the Black Hole benchmark (Beta). It tests the pure calculation speed of your cpu in double and integer (floating and integer). The bench has 2 tests, multithreaded and singlethreaded. Multith... Black Hole benchmark Download
Thanks! Results: Spoiler Don't know if that's good or not. Just have to wait and see. **Added** Oops, I see there's already a thread for results. Sorry about that.
Nah if anything I'm doing it wrong. I've noticed the same thing on the 4.1 benchmark thread. I think I just need to take a look in the BIOS and see if there's any throttling or power management junk or etc I've overlooked. Either that or this CPU isn't as good as I originally thought. I got it for practically nothing so it's a possibility. **Added** After playing around in the BIOS I managed this: Spoiler Not much better but oh well. To be honest it's been almost a decade since I last overclocked anything or really messed around with the BIOS, especially with this Rampage IV. There's just a staggering amount of options to change when compared to any mobo I've ever owned in the past. Technology is already so fast nowadays and I've got a steady job so I haven't needed to squeeze every last drop of performance out of my electronics like I used to. I just do the basics of updating the BIOS when needed, turning off any throttling or whatnot, and making sure my clock speeds and timings are where they should be and then I leave it alone. And on that note, if anybody else has a Rampage IV extreme and has any pointers I'd appreciate any help I can get. **Added** After googling your CPU I see you've got quite a nice OC on it. 2.83GHz to 3.8? **Added** @4.2GHz Spoiler
Clock for clock the old intel series aren't that much behind in pure calc speed. If you'd test an i5 2500k at the same speed of the c2q, you'll only see a difference of 200-300 points. Anyway, the new series are cooler and can be clocked much higher (w/o ln2) The reason why the difference quad threaded cpu vs 8threads is minimal, is because singlethreaded and quad threads will give around the same result. That said, overclocking does help VERY much in this bench, especially multithreaded will scale up. Here will all 8threaded cpus gain much points, because the scale of the multithreaded test is much bigger . I'll try 5.2ghz tomorrow. Edit: People with overclocked phenoms are very welcome atm!
Really, it's an okey score compared to some others I saw already. Some scores: Athlon 64 3700+ stock I7 3770 stock (3.4ghz, for singlethreaded 3.9ghz, turbo always on 1 thread) So it seems there IS still a big difference c2q if they were on same clock. Could you test that c2q on 3.5ghz? I5 2500k @4.3ghz I7 920 stock (2.63ghz) I7 2600k @5ghz (will do 5.2 or more tomorrow) I7 3930k @4.9ghz I7 2630QM stock (2ghz) For people who want to get the best score out of their systems, lowering RAM LATENCY (speed not really), makes pretty much sense in the score and will give it a 200-500 points boost.
Wow, we have the creator himself here. Thank you for explaining that to me. I'm surprised there's so little difference between a lot of these CPUs. I wish I still had my Phenom II 965 rig in one piece so I could test that for you.
Np, could you plz experiment with your ram latency? Lower it a bit (1-2 cycles down or something) and test again. You'll have better scores. Also, really close ALL programs. Even the smallest backgroundprograms like skype,... can affect your score pretty much.
@3.8GHz For some reason I got a better score running the memory at 1600MHz than 1866. I'm assuming it's because the 3820 is only supposed to run up to 1600MHz when in quad channel? I tried tightening my RAM timings but I couldn't manage anything lower than 8-9-9-23-1 @1.55v (I was at 9-9-9-24-1 @1.5v originally) and I didn't feel safe going much higher even though I've heard of people going up to 1.6 with this specific RAM. Nonetheless, your advice has improved my score quite a bit. Thank you. It also got me back into OCing again a bit, although I've got a lot of catching up and re-educating to do before I feel as comfortable doing it as I once did. I feel like a junkie who's had his fix after 10 years of sobriety....that wasn't the best analogy.
Going over 1.6v with your ram isn't dangerous, I run my ram at 1.59v all the time. Because I've 2 kingston hyperX sticks and 2 ripjaws, but the kingstons need more voltage to get these 1600mhz...
That's a relief. I'll try upping the voltage a tad then and see if I can tighten them up a bit more, although if I can't I'm happy with 8-9-9-24. Thanks again for all your help!