Dolphin, the GameCube and Wii emulator - Forums

Full Version: Benchmarking Dolphin
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Why are you resurrecting a thread that died a full 2 months ago? If your considering joining me and JADS just go on the irc channel and leave it open. Although we haven't really done anything in awhile....we really need sonic to finish that mutex test which he never gave us before we can continue solving the mystery.
You wanted this information correct? Do you or don't you want some of these issues fixed, isn't that why you put this thread together to begin with? Anyways it doesn't matter what the age of the thread is, my post is on topic, it's relevant and it's contributing to the purpose of this thread, it's not like I'm bumping a thread that's over a year old with irrelevant or off-topic information. I know I want to see some of these issues fixed someday, so I contributed my system information to this thread, in hopes some of these issues affecting performance can be narrowed down.
Alright, sorry if I sounded rude, I didn't mean to I was just wondering why you were posting here.

Ok so you want to help. Well unfortunately I really don't have the time right now to keep doing this, that's why it died. But I can give you some "assignments" if you want. Which will give me some more of the info I need.

Quote:Do you or don't you want some of these issues fixed, isn't that why you put this thread together to begin with?

Actually I put this thread together originally to benchmark dolphin since I was so annoyed with everyone complaining about bad performance and wondering what to do about it. I thought it would help us advise people and the data would show us exactly how a variety of different factors affected performance in dolphin. But then when we started testing....we got some weird results. A lot of the results simply didn't make any sense. Their were two main problems with the results. Some people were experiencing massive slow downs in hyrule field while others weren't and some people were experiencing large speedups that seemed to come out of no where, speedups that went away if 2 cores were disabled.

After everyone left it was just me and JADS for awhile. And then I didn't have time anymore so I stopped going on irc. I'm not even sure if JADS still goes on irc, probably not.

What would help me a lot is if someone managed to get that mutex test from sonic.
Thread needs more members pitching in and posting information and running these benchmarks, that's why this thread died. Sadly no one really wants to do this, since most members here just care about playing free games on an emulator and not about helping improve the emulator.

Unfortunately there's not much more that can be done for performance than what is already posted on the forum, unless everyone has a top of the line GPU and CPU to use with Dolphin. It seems that Dolphin lacks support for occlusion culling and/or frustem culling or the implementation of culling is incorrect, thus degrading performance in games with large areas or larger areas with lots of several enemies. I'm still testing this observation with games between the actual hardware and Dolphin, I will provide more information later when I post the results of my testing.

http://forums.dolphin-emu.org/thread-10188.html
I'm new to Dolphin but I'd really like to take part in benchmarking it if you start up again, I may only have 3 posts but I am deadly serious about helping. Dolphin seems to be an extremely popular emulator and there's information about getting most games to work so I thought I'd like to help out and this would be a great way. I have a lot of spare time, I own all these games, I have Metroid Prime 1 and 2 on GC aswell as the Metroid Trilogy since I don't own MP3 on it's own. I own Wind Waker, I have Super Mario Galaxy and I own Twilight Princess on the Wii but I could lend my friends GC copy if it makes a difference. All of my games are PAL since I live in the UK. I use Windows 7 64-bit and Ubuntu 10.04 64-bit with a ATI 3870 which next month will HOPEFULLY be a GTX 280 so my GPU shouldn't be too much of a bottleneck when I get my GTX 280. I'd also be happy to show the difference in performance between Linux and Windows for those curious. I prefer real figures rather than peoples word of mouth.

EDIT: Here's my GPU-Z screenshot since I can't have more than 5 screenies. Also my Page 1 has been updated cause I uploaded a pic while my CPU was idle so it showed 1445MHz ^_^
[Image: thumb_45c1128a.png]
Quote: Twilight Princess on the Wii but I could lend my friends GC copy if it makes a difference

Yes. We've been curious about an enigma in twilight princess (referred to as the hyrule fields slowdown) which is why we've focused so much on that particular game. Although you could use the wii verison it would be better to use the GC version since all of us are using it and it would make your results comparable to ours.
I noticed that with the DX11 plugin Twilight Princess is quite a bit slower than with DX9. Hyrule Field runs at 5 FPS with DX11 as opposed to 13 FPS - 15 FPS with the DX9 plugin without the speedup patch. Other areas in the game that usually run at full speed with DX9, are slower with the DX11 plugin as well.
i would like to help too, but unfortunately i only have SMG out of these games...is that ok?
(06-22-2010, 09:05 AM)NaturalViolence Wrote: [ -> ]
Quote: Twilight Princess on the Wii but I could lend my friends GC copy if it makes a difference

Yes. We've been curious about an enigma in twilight princess (referred to as the hyrule fields slowdown) which is why we've focused so much on that particular game. Although you could use the wii verison it would be better to use the GC version since all of us are using it and it would make your results comparable to ours.

I thought so, that's why I went and done it anyway ^_^ I'm also currently in the process of ripping ALL my GC games so I've got MP1 and MP2 ready, SMG is ready, both zelda's are ready. I just need information on how and when I need to benchmark. TBH The GC version is easier because I currently don't have money this week to buy a bluetooth adapter and a wireless sensor bar to use my wiimotes so I can just use my 360 controller instead.
Quote:i would like to help too, but unfortunately i only have SMG out of these games...is that ok?

Unfortunately no, at least not right now. You might be able to help when we get all of this twilight princess stuff sorted out.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16