Overclock.net banner

Dolphin 5.0 CPU benchmark - results automatically updated!

17K views 48 replies 20 participants last post by  bmgjet 
#1 · (Edited)
DISCLAIMER: The purpose of this thread is to use Dolphin and luabench as a general CPU benchmark that can used to determine a CPU architecture's relative capabilities in emulation-based workloads; it is not intended to benchmark Dolphin itself and should not be viewed as an example of absolute performance in Dolphin (which typically is several times slower).


Submit your benchmark results here:
https://docs.google.com/forms/d/1gOj1MHvM0BB_ZknxTnO5wrK3D6waqNGnyPP_oRej4f8


Current results:
https://docs.google.com/spreadsheets/d/1k12sv1NXGGuSOY0NhsuONtRCte51GHKdgA7ciL76mBs



Delroth, the creator of the previous Dolphin benchmark, has put together a newer benchmark that should be more comparable to real-world performance in Dolphin. With the release of Dolphin 5.0, I figured that now was an appropriate time to put together a new Dolphin Benchmark thread - but this time, taking a page from Overclock.net's Cinebench 15 benchmark thread, the results will automatically update without any input from myself!


When specifically benchmarking the per-GHz and IPC performance between different CPU architectures, please try to run your CPU at 2.0 GHz, 3.0 GHz, or 4.0GHz. In order to make comparing easier, I have and will also mark all 2.0GHz, 3.0GHz, and 4.0GHz runs with inverted colors (white text with a dark background) on the spreadsheet. Thank you for your cooperation.


WARNING: Do not pause the benchmark at any time - doing this will actually hurt your final score!

Also note: Ignore the error [string ''IUB ambiguity codes''] while running the benchmark - this is NORMAL!
Screenshot of said error (Click to show)


[Windows only] "For dummies" edition - not for paranoid users!
(Paranoid users should instead scroll down and read the instructions under "Windows geeks that know what they're doing")


Download the following EXE (note that it's merely a self-extracting 7z archive, meaning you can browse the file contents via any archive manager that supports 7z):
https://drive.google.com/file/d/0B3hUE5vj6-14NHJ6RlREbVZ6eDA

Run the EXE and click "Extract". Open the newly-created "Dolphin 5.0 CPU Benchmark" folder and double-click "Run Benchmark".

Select "Yes" or "No" if you want to send performance analytics to the Dolphin developers, then wait for the benchmark to complete.

Once "Overall time" is displayed, the benchmark is complete; take a screenshot and submit your results via the form above. If this forum thread hasn't had any new posts in a while, then please consider making a post in this thread as well.




Other platforms + Windows geeks that know what they're doing

Go grab a fresh copy of Dolphin 5.0 (not the newest Dolphin dev build; it must be the Dolphin 5.0 stable release version).

Once you've launched the actual program, go into "Config" and set the "Speed Limit" to 'Unlimited'. (optionally you may also want to go into the "Audio" tab and set the "Audio Backend" to 'No audio output')

Also go into the "Graphics" settings, find the "Hacks" tab, enable the "External Frame Buffer (XFB)" and set it to 'Real'. (Windows users may want to change their graphics "Backend" to 'Direct3D 11' which is necessary for many older GPUs and might also be a bit faster)

Now download the following Wii homebrew application and run it in Dolphin; this is the actual CPU benchmark:
http://delroth.net/luabench.dol ~ Mirror download link

Once "Overall time" is displayed, the benchmark is complete; take a screenshot and submit your results via the form above. If this forum thread hasn't had any new posts in a while, then please consider making a post in this thread as well.
 
See less See more
1
#2 ·
Posted mine, been playing around with bclk clocks (Reason why my RAM speed is a bit wacky)
 
#4 ·
#6 ·
Dam all those 6700k's and their scores....
 
#7 ·
Quote:
Originally Posted by NeoReaper View Post

Dam all those 6700k's and their scores....
Just ran at 4912, tried 5ghz but the dolphin crashed. srore 273 1st for now.
 
#8 ·
#11 ·
Quote:
Originally Posted by Bluescreendeath View Post

Uploaded multiple results for i7 2600k at different speeds, and an OCed G3258 @ 4.0GHz. The scaling is very linear: 2x the clock speeds in i7-2600k and the G3258 directly translates into 2x faster benchmark timing.
Your scores seem lower than one would expect... I mean, a Core 2 Duo shouldn't have better IPC than a Sandy Bridge!
 
#12 ·
Quote:
Originally Posted by Nintendo Maniac 64 View Post

Your scores seem lower than one would expect... I mean, a Core 2 Duo shouldn't have better IPC than a Sandy Bridge!
That's what I thought too. But I just ran the i7-2600K multiple times downclocked to 2.0GHz, and it got around 18 minutes every time.

Here is another run with the 2600k at 2.0GHz. Score is a miserable 18min 21s.

I ran passmark and my cpu got 8670.4 - which is around the average score for the i7 2600k on passmark. So it seems my cpu is running fine in other programs. Is it possible that Dolphin 5.0 nerfed Sandybridge performance and buffed Core2 performance?

 
#13 ·
I added my Dell Latitude E6230 to the results. I have played Super smash bros at 720p on Dolphin fine before (Very few slowdowns depending on the stage).
The laptop had absolutely nothing else running in the background so this is the best result its ever gonna get (Scored 900 seconds).
 
#14 ·
Quote:
Originally Posted by NeoReaper View Post

I added my Dell Latitude E6230 to the results. I have played Super smash bros at 720p on Dolphin fine before (Very few slowdowns depending on the stage).
The laptop had absolutely nothing else running in the background so this is the best result its ever gonna get (Scored 900 seconds).
Weird. Your i3 3110M at 2.4GHz scored 900 seconds, but Nintendo Maniac's Core2Duo T8300 at 2.4GHz scored a better score of 826 seconds. Which means Dolphin 5.0 has a Core2 running at better IPC than an i3 Ivy Bridge at the same clockspeed.
 
#15 ·
Quote:
Originally Posted by Bluescreendeath View Post

Quote:
Originally Posted by NeoReaper View Post

I added my Dell Latitude E6230 to the results. I have played Super smash bros at 720p on Dolphin fine before (Very few slowdowns depending on the stage).
The laptop had absolutely nothing else running in the background so this is the best result its ever gonna get (Scored 900 seconds).
Weird. Your i3 3110M at 2.4GHz scored 900 seconds, but Nintendo Maniac's Core2Duo T8300 at 2.4GHz scored a better score of 826 seconds. Which means Dolphin 5.0 has a Core2 running at better IPC than an i3 Ivy Bridge at the same clockspeed.
Also taking into the fact that the Dell laptop I have has Speedstep and other power states locked. I have changed everything I can in Windows (Power profiles, drivers... I have cleared the internet on ways of fighting it but there is no way of fighting whats been locked in the bios) and these settings can't be changed in the bios (The C states can be changed to performance mode which has been done) but there is no other way of getting this I3 to perform any better. (And modifying the bios is out of the question since Dell used an custom made EEPROM chip and the bios image is heavily protected)
 
#16 ·
That's wacky - it almost makes it seem like it's my Core 2 Duo time(s) that's invalid.
headscratch.gif
Unless...maybe it's something with SMT? The only other thing I can think of is the larger L2 cache on the Penryn Core 2 Duo.

Well I'm hoping to get a run on an i5 Westmere soon (read: dual core with HT) so hopefully that'll bring some clarification to this situation (I hope to do runs with HT enabled and disabled).

If I'm lucky I may just get in some runs on a Merom C2D and Athlon 64 x2 as well, and maybe a Bobcat E-350 to boot. I actually have a Merom Celeron as well, but it's currently not in any system and I don't want to disassemble the two compatible systems...besides, it's only 1.83GHz anyway which is too low for the 2GHz that I've chosen for IPC comparisons (2GHz is a multiple of both 200 and 133 while 1GHz is not, and 3GHz is too high for many mobile & low-powered CPU architectures).

EDIT: If it really is L2 cache, I have both a Brisbane and a Windsor Athlon 64 x2, and one of the biggest differences outside of clockrate, node size, and TDP, is that Brisbane has half the L2 cache that Windsor does - maybe that could be revealing...
 
#17 ·
Quote:
Originally Posted by Nintendo Maniac 64 View Post

That's wacky - it almost makes it seem like it's my Core 2 Duo time(s) that's invalid.
headscratch.gif
Unless...maybe it's something with SMT? The only other thing I can think of is the larger L2 cache on the Penryn Core 2 Duo.

Well I'm hoping to get a run on an i5 Westmere soon (read: dual core with HT) so hopefully that'll bring some clarification to this situation (I hope to do runs with HT enabled and disabled).

If I'm lucky I may just get in some runs on a Merom C2D and Athlon 64 x2 as well, and maybe a Bobcat E-350 to boot. I actually have a Merom Celeron as well, but it's currently not in any system and I don't want to disassemble the two compatible systems...besides, it's only 1.83GHz anyway which is too low for the 2GHz that I've chosen for IPC comparisons (2GHz is a multiple of both 200 and 133 while 1GHz is not, and 3GHz is too high for many mobile & low-powered CPU architectures).

EDIT: If it really is L2 cache, I have both a Brisbane and a Windsor Athlon 64 x2, and one of the biggest differences outside of clockrate, node size, and TDP, is that Brisbane has half the L2 cache that Windsor does - maybe that could be revealing...
Over at Dolphin Forums, I just downloaded from your link and ran the povray benchmark at 2.0GHz and 2.4GHz.

i7 2600k @ 2.0GHz = 22 minutes 49 seconds (1369 seconds)

i7 2600k @ 2.4GHz = 18 minutes 46 seconds (1126 seconds)

Pics attached.


 
#21 ·
5 minutes 38 seconds on signature system.
Quote:
Originally Posted by Nintendo Maniac 64 View Post

Your scores seem lower than one would expect... I mean, a Core 2 Duo shouldn't have better IPC than a Sandy Bridge!
Quote:
Originally Posted by Nintendo Maniac 64 View Post

The only other thing I can think of is the larger L2 cache on the Penryn Core 2 Duo.
Dolphin (and emulation in general) has always been picky about cache performance. Core 2 has an enormous L2 cache compared to later Intel architectures, and clock for clock, Core 2's L2 is much faster than the later part's L3s.

Does the default Dolphin 5.0 build support AVX?
Quote:
Originally Posted by Cakewalk_S View Post

What's the CPU usage during the benchmark? Does this utilize CPU threads properly? How many threads does it support?
The packaged benchmark is only loading a single thread.
 
#23 ·
Quote:
Originally Posted by Nintendo Maniac 64 View Post

But did you run it in Dolphin 5.0?
I think I got it working. I copied and pasted the Dolphin 5.0 files into the POV benchmark folder that had the older Dolphin version. When I started the program, it said "Dolphin 5.0" - and was able to press the play button to run the benchmark (pic attached).

These are the results I got.

i7 2600k @ 2.0GHz - 16min 37 sec (997 seconds)
i7 2600k @ 2.4GHz - 15min 52 sec (952 seconds)



 
#24 ·
Uploaded mine

http://i.imgur.com/x0fDcah.png

However the google doc seems to have duplicated the edit as a new entry and my timestamp and score are placing me in last

5 mins 32 seconds on my 4790k
 
#25 ·
Quote:
Originally Posted by SoloCamo View Post

However the google doc seems to have duplicated the edit as a new entry and my timestamp and score are placing me in last
This is normal (at least as of right now); I am completely incompetent with any sort of scripting so new results always go to the end.

However, you can still manually sort the results yourself via the time in seconds (which is why I make it clear to not include any letters or labels - they screw up the sorting).

Currently what I normally do is I just check in once in a while and manually apply all the current results to be sorted by time in seconds and apply ISO 8601 date formatting as well (the M/D?Y date formatting makes it easy to see which results are new when everything else is Y-M-D).

EDIT: Oh dear, there truly are two results of the same thing. Did you submit twice because you thought that it incorrectly sorted your submission? Maybe I should make a note saying that all new entries are listed at the bottom until I manually sort them...

Currently the easiest way I know how to remove a result is to unlink the spreadsheet, remove the duplicate result from the form responses, and then create a new spreadsheet.

EDIT 2: Wait, are you saying that you made an edit to your first submission via the edit link after you click "Submit"? And that the spreadsheet actually lists both the original submission and the edit? If so, then that's wacky though possibly still the correct behavior...maybe I should disable editing if that's the case.

Though if that is the case, then I must ask what it is that you edited; from what I can tell, both entries are exactly the same save for the timestamp...
 
#26 ·
Quote:
Originally Posted by Nintendo Maniac 64 View Post

This is normal (at least as of right now); I am completely incompetent with any sort of scripting so new results always go to the end.

However, you can still manually sort the results yourself via the time in seconds (which is why I make it clear to not include any letters or labels - they screw up the sorting).

Currently what I normally do is I just check in once in a while and manually apply all the current results to be sorted by time in seconds and apply ISO 8601 date formatting as well (the M/D?Y date formatting makes it easy to see which results are new when everything else is Y-M-D).

EDIT: Oh dear, there truly are two results of the same thing. Did you submit twice because you thought that it incorrectly sorted your submission? Maybe I should make a note saying that all new entries are listed at the bottom until I manually sort them...

Currently the easiest way I know how to remove a result is to unlink the spreadsheet, remove the duplicate result from the form responses, and then create a new spreadsheet.

EDIT 2: Wait, are you saying that you made an edit to your first submission via the edit link after you click "Submit"? And that the spreadsheet actually lists both the original submission and the edit? If so, then that's wacky though possibly still the correct behavior...maybe I should disable editing if that's the case.

Though if that is the case, then I must ask what it is that you edited; from what I can tell, both entries are exactly the same save for the timestamp...
Well the page did hang for quite a bit while submitting so I refreshed it after waiting a while. That may have caused the other submission as well. Either way, my bad on that.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top