Main Restorations Software Audio/Jukebox/MP3 Everything Else Buy/Sell/Trade
Project Announcements Monitor/Video GroovyMAME Merit/JVL Touchscreen Meet Up Retail Vendors
Driving & Racing Woodworking Software Support Forums Consoles Project Arcade Reviews
Automated Projects Artwork Frontend Support Forums Pinball Forum Discussion Old Boards
Raspberry Pi & Dev Board controls.dat Linux Miscellaneous Arcade Wiki Discussion Old Archives
Lightguns --- Bug Reports --- Site News

Unread posts | New Replies | Recent posts | Rules | Chatroom | Wiki | File Repository | RSS | Submit news

  

Author Topic: MAME 163/205 benchmark  (Read 3181 times)

0 Members and 1 Guest are viewing this topic.

haynor666

  • Trade Count: (0)
  • Full Member
  • ***
  • Offline Offline
  • Posts: 1194
  • Last login:April 19, 2019, 09:47:11 am
  • retro maniac
MAME 163/205 benchmark
« on: July 15, 2015, 09:07:45 am »
Recently I've started rebuilind my arcade PC with mame 163. I was curious how version 163 works with certain games but I ended testing most of sets in game (excluding mahjongs, not working (with few exceptions), fruit, mechanical etc.) All tests was done with G3258 overclocked to 4,2 GHz, Asus H81M-E, Kingston 2Gb value and Win 7x64. Test was done using command mame64.exe - bench 299. Maybe it will usefull for people planning upgrading/buying hardware for mame.


EDIT. Added tests for mame 205.
« Last Edit: March 06, 2019, 07:22:30 am by haynor666 »

abispac

  • Trade Count: (0)
  • Full Member
  • ***
  • Offline Offline
  • Posts: 937
  • Last login:Yesterday at 03:26:37 pm
Re: MAME 163 benchmark
« Reply #1 on: July 15, 2015, 03:15:56 pm »
Recently I've started rebuilind my arcade PC with mame 163. I was curious how version 163 works with certain games but I ended testing most of sets in game (excluding mahjongs, not working (with few exceptions), fruit, mechanical etc.) All tests was done with G3258 overclocked to 4,2 GHz, Asus H81M-E, Kingston 2Gb value and Win 7x64. Test was done using command mame64.exe - bench 299. Maybe it will usefull for people planning upgrading/buying hardware for mame.


Results are in excel 2003 format: https://mega.nz/#!5tsUACJJ!DlKnBEmISXXymrKXNEl4B7v2qOvR8rh4mqXdAePVLWA
Why not just posting the results here, instead o making people download a worthles file?

haynor666

  • Trade Count: (0)
  • Full Member
  • ***
  • Offline Offline
  • Posts: 1194
  • Last login:April 19, 2019, 09:47:11 am
  • retro maniac
Re: MAME 163 benchmark
« Reply #2 on: July 15, 2015, 04:09:13 pm »
Because table is too big for this forum. I converted xls to html - [link removed]
« Last Edit: October 15, 2016, 06:58:06 am by haynor666 »

lostboy4480

  • Trade Count: (+1)
  • Full Member
  • ***
  • Offline Offline
  • Posts: 76
  • Last login:Yesterday at 11:38:44 am
  • I'm a llama!
Re: MAME 163 benchmark
« Reply #3 on: July 16, 2015, 09:00:31 am »
Thank you for doing this,

I was looking at putting together a machine that can play GT Fore and the G3258 looks like it should be able to.

haynor666

  • Trade Count: (0)
  • Full Member
  • ***
  • Offline Offline
  • Posts: 1194
  • Last login:April 19, 2019, 09:47:11 am
  • retro maniac
Re: MAME 163 benchmark
« Reply #4 on: July 16, 2015, 09:18:56 am »
Not exactly. Test was done using command "mame64 -bench 299". This means that there wasn't video or sound output (though mame emulates video and sound, just does not output). When You run game normally some PC power will be needed also for video and sound presentation. Right now teoretical min. speed to make game playable is in my opinion 200% (though is not guaranteed - mmpork has 325%, still there occasional short slowdowns).

Remember, this is average speed from 299 seconds starting from machine initalization to attract mode, title screen etc.).

Haze

  • Trade Count: (0)
  • Full Member
  • ***
  • Offline Offline
  • Posts: 1197
  • Last login:Yesterday at 01:36:21 pm
  • I want to build my own arcade controls!
    • MAME Development Blog
Re: MAME 163 benchmark
« Reply #5 on: July 16, 2015, 11:07:25 am »
Thank you for doing this,

I was looking at putting together a machine that can play GT Fore and the G3258 looks like it should be able to.

The GTFore games need the highest clock i7 (4790K, 4ghz) and the low resolution dipswitch set to maintain full speed.

running them on anything less won't give satisfactory results, and if you want to run them in medium res you're going to have to wait because the most demanding parts still dip under 100% even on that.

benchmarks like this are incredibly misleading as many of them don't represent actual gameplay at all, something can run at 500% for some startup checks then 25% ingame and sitll come out as >100, for Golden Tee Fore you don't hit the most demanding parts (ball stuck in heavy trees / bushes) at all during attract mode either.



haynor666

  • Trade Count: (0)
  • Full Member
  • ***
  • Offline Offline
  • Posts: 1194
  • Last login:April 19, 2019, 09:47:11 am
  • retro maniac
Re: MAME 163 benchmark
« Reply #6 on: July 18, 2015, 09:42:07 am »
Benchmarks could a little more accurate if command bench would allow time longer than 299, that way I could test for example 10 min of attract mode.

haynor666

  • Trade Count: (0)
  • Full Member
  • ***
  • Offline Offline
  • Posts: 1194
  • Last login:April 19, 2019, 09:47:11 am
  • retro maniac
Re: MAME 163 benchmark
« Reply #7 on: March 15, 2016, 03:53:41 pm »

haynor666

  • Trade Count: (0)
  • Full Member
  • ***
  • Offline Offline
  • Posts: 1194
  • Last login:April 19, 2019, 09:47:11 am
  • retro maniac
Re: MAME 163 benchmark
« Reply #8 on: March 06, 2019, 07:21:44 am »
More than 2,5 years have passed so I decided to test mame 205 and see if my G3258 4,2 GHz still holds for recent mame versions. Additionally once again I've calculated other cpu speeds based on passmark scores. Results are in xls format as well in standard html/pdf files.

RESULTS