Dolphin, the GameCube and Wii emulator - Forums

(06-30-2016, 07:46 AM)Admentus Wrote: [ -> ]Ahh yes, "grab a fresh copy of Dolphin 5.0". I believe it is confusing because it may either refer to the stable or development builds. A fresh copy does not specifically which one it is.

I was not aware of that; my impression of "fresh copy" just meant one that hasn't been modified or "tainted", not something that may be newer or more recent.

I've edited the statement; is it better now?

Ahh yes, much better. We humans tend to stick to well-approved and often-used terms for our understanding, otherwise we seem to get lost what it is all about. Details and specifics are key in describing a discussion. If you only knew how many questions are asked when a game does not run. Specifics such as hardware, the game itself and the version of Dolphin are often forgotten to mention (hence why it is in the global rules to state them if help is needed). Their intent to ask for help may be good, but none will get any wiser from poor explainations. Don't take this as a critic aimed at you, it is just in general all around the world. You see thus why proper wording is important to avoid frustration among the userbase. On the Dolphin forums the terms stable and development builds are well known and fresh builds... well not so much.

Can we get some Dolphin 5.0 benchmarks with some older CPUs and non-overclocked CPUs? A stock Haswell or an overclocked older CPU like the Nehalem/Sandy Bridge/AMD FX is more useful than seeing a bunch of overclocked Intel i5/i7 4000s/5000s. We all know those higher end Haswells/Broadwells/Skylakes will run Dolphin fine. I'm interested in cpus that just meet the requirement, or can be overclocked to push it over the edge to meeting Dolphin's requirements.

I have access to a Westmere, Penryn, and Merom; will those do?

Also I could underclock my own Pentium, but it doesn't exactly seem that useful when you can just use a calculator to figure out how such a CPU would perform - at 2.3GHz it should score something like 658 seconds, or 10m 58s.

(06-30-2016, 01:17 PM)Nintendo Maniac 64 Wrote: [ -> ]I have access to a Westmere, Penryn, and Merom; will those do?

Also I could underclock my own Pentium, but it doesn't exactly seem that useful when you can just use a calculator to figure out how such a CPU would perform - at 2.3GHz it should score something like 658 seconds, or 10m 58s.

I have a G3258 and a Sandy Bridge 2600k myself, so I'll try to benchmark it at various speeds and upload my results. If you could submit some results from older cpus like Westmere and Conroe/Penryn/Wolfdales at stock and OCed speeds, that would be great. Thanks!

I can't overclock them because all 3 are laptop parts.

(07-01-2016, 03:22 PM)Nintendo Maniac 64 Wrote: [ -> ]I can't overclock them because all 3 are laptop parts.

Ah, right. The benchmark scales linearly in accordance with CPU frequency right? (or does it make a huge jump once you reach a certain level of performance?)
If it's linear, I guess we can just calculate the corresponding scores in their OCed desktop counterparts.

(07-01-2016, 03:32 PM)Bluescreendeath Wrote: [ -> ]Ah, right. The benchmark scales linearly in accordance with CPU frequency right?

Honestly I haven't come accross a single CPU-limited workload that hasn't, but they do apparently exist so it'd probably be good if I did a non-overclocked G3258 run...

I'll probably use a flat 2.0GHz and possibly do a 2.0GHz run on the Westmere and Penryn as well (the Merom is already 2.0GHz).

With my 2600k Sandy Bridge @ 4.0Ghz, I got a total time of 7 minutes and 6 seconds (426 seconds) with the Dolphin 5.0 Benchmark. Is this right? This seems a bit too fast. Did v5.0 improve benchmarks/performance that much over previous versions?

I'll try this again at stock...

I think I screwed up somewhere. Did I download the wrong benchmark? My final results screen seems to be different from that of other people's.