Well I actually listed it as "Turbos to 3.4 on all cores as soon as game starts with occasional spikes to 3.5." How that got interpreted and put in the chart is out of my hands, though I do think 3.4 is the more accurate representation.
The Legend of Zelda: The Wind waker CPU Benchmark
|
@NV - Do you happen to know if the i5-2500K turbo boosts all of its cores to 3.7GHz, or does it follow the sort of pattern seen in laptops, e.g you'd only get 3.7GHz using one core, 3.5GHz using 2 cores, 3.4GHz using all 4 cores, or something to that general effect? I haven't played much with my BIOS in regards to turbo boosting, except to disable it until I felt I needed it.
03-18-2013, 07:51 AM
You can set the turbo values as you wish in bios but default is: 3.4 4 cores, 3.5 3 cores, 3.6 2 cores, and 3.7 1 core. So if you haven't changed anything yourself Dolphin should never see 3.7 Ghz on a stock 2500k. At best you would see 3.6. I am curious as to why I wasn't boosting to 3.5 though. I usually run at 4.5 ghz though so I'm not going to invest any time in trying to solve it. Probably some power saving option.
03-18-2013, 07:53 AM
That's odd. My 3570K does not behave like that and both sandy and ivy bridge use the same turbo algorithms. I can boost to 3.7GHz on all cores as long as the IGP is idle.
"Normally if given a choice between doing something and nothing, I’d choose to do nothing. But I would do something if it helps someone else do nothing. I’d work all night if it meant nothing got done."
-Ron Swanson "I shall be a good politician, even if it kills me. Or if it kills anyone else for that matter. " -Mark Antony
Yo, just did some testing and I get similar result to Xalphenos. With TB enabled for all four cores, I get 3.4GHz consistently running Dolphin, haven't seen it go up to 3.5 though. Haven't tried doing just two cores yet. Did some googling about it, and some people confirm TB is working like Xalphenos described. Any clues about this?
03-18-2013, 08:36 AM
*shrugs*
"Normally if given a choice between doing something and nothing, I’d choose to do nothing. But I would do something if it helps someone else do nothing. I’d work all night if it meant nothing got done."
-Ron Swanson "I shall be a good politician, even if it kills me. Or if it kills anyone else for that matter. " -Mark Antony 03-19-2013, 02:19 AM
Of course both ivy and sandy will boost based on how much cores are used, its 100 mhz more for one less core used
Well according to this Ivy performs similarly to Sandy. A 3570k will turbo to 3.6 on 4 cores, 3.7 on 3 cores and 3.8 on 1 or 2 cores. So it gets a bigger boost, 200 mhz, on all for cores and can do its max turbo on 2 cores instead of just 1. So it makes since that NV is getting 3.7 in dolphin.
03-19-2013, 07:01 AM
Xalphenos Wrote:Well according to this Ivy performs similarly to Sandy. In synthetic tests which is the only type of test that they ran (and they didn't even run a wide variety of synthetic applications). If you look at any of the reviews from major hardware reviews sites you will see that the average performance increase over equivalent sandy bridge models is 10%. Sometimes it's as low as 5%, sometimes it's as high as 15%, but in most cases it hovers around 10%. In dolphins case it seems to be on the low end of the scale.
"Normally if given a choice between doing something and nothing, I’d choose to do nothing. But I would do something if it helps someone else do nothing. I’d work all night if it meant nothing got done."
-Ron Swanson "I shall be a good politician, even if it kills me. Or if it kills anyone else for that matter. " -Mark Antony |
« Next Oldest | Next Newest »
|
Users browsing this thread: 4 Guest(s)