Official shmupmame Super Turbo thread


I love you Papasi, no joke :slight_smile: lets see if it works…


Suddenly very excited about ST again…


Papasi this fucking ROCKS! Any chance to get a link for the modified build you mentioned on the second post? THANKS!




it probably would be better to have separate folders for the older ROM versions you want to play. So you can use both the latest MAME but also the older Finalburn.
Because now CPS2 ROMs will also change due to Qsound decapping.

Latest MAME does not play those games because it requires qsound.bin


I want the modified build for the easy input menu, i want to check if it saves multiple button configs.

Played today with a friend, this emu ROCKS, thanks again, Papasi!


Got shmupmame 4.2 downloaded at the Arkadeum studio and we played pretty much the whole day. Felt buttery smooth aside from the minor input delay from using a projector. Yet again, receives my seal of approval for best form of emulated ST.


papasi, have you looked into testing USB lag?


i got it working for regular mame but shmupmame’s author is not replying to the support thread on shmup forum. i couldn’t get shmupmame to compile because he’s not providing full source code, only modified files.



Anyone knows how to map those “Custom” buttons? I want to play with my bro, but he plays Blanka and maps the hop (3K) to a button.



u can by editing the txt document that saves ur inputs, this is how I do it in damdai’s supercade to use 3P n 3k, idk what is the actual name of the file in shmupmame but shouldnt be hard to find it


papasi, any luck with the 60 fps camera / lagless display and doing the nki input lag test?


not sure what i missed. all the tests i did above (except stage timer) were done using a 60fps camera (aka nki test)


I guess I’m thinking about it more in an isolation of the lag problems. I.e., what’s the USB controller lag vs. the inherent lag of shmupmame. Maybe its not possible to isolate the various lag contributors with the nki test…not sure.

The nki test just (basically) measures the input lag as a whole, correct?



it’s the total lag, including input / processing / display.


Ok, cool. So, really just input and processing (for the CRT tests, anyway) since that shouldn’t be relevant from a lag perspective.


whenever I change any property of the game (game speed, enable cheats, etc) it doesn’t commit. If I go back to the properties menu they are set as the initial default ones.


Whenever you change a property, click on another box/option to ensure that your change is highlighted in “black.” I usually do this on the properties for the actual game and CPS2 options to ensure that the settings are intact. Once both the game and CPS2 properties match, then click “OK” and it should be good to go. You can also manually edit, the configuration .ini file to a particular option as well.

Did some more testing that yielded some very interesting results.

  1. The USB poll rate program does indeed actually work (using Win7) for most USB sticks (I haven’t been able to get my PS360 sticks to work, but all others worked). Basically, you can select the poll rate to 1000ms = 1ms as opposed to the standard Windows default USB poll rate of 8ms. There is a DEFINITE improvement.

  2. Disabling bi-linear filtering on Direct 3D seems to increase the responsiveness aka reduce the input lag as well.

Papasi’s test pretty much confirmed how close this emulator is to actually matching the cps2 board; these new adjustments should close the gap even more :wink:

Small reminder ** The best way to utilize this emulator is to output the video signal from your PC/Mac/laptop to an outside monitor that has 0ms delay (VGA CRT/IIyama LCD/Euro brand LCDs) or one of the low input lag LCDs mentioned in the tech talk section. **


Thanks DG. The highlight trick you mentioned let me change the settings. Btw, the cheats section has A LOT of things you can change. It’s really awesome.

Now I am curious about the minimum requirements to run shmupmame and still perform so close to the CPS2. I think the raspberry pi is not powerful enough (yet). Otherwise, it would be a very cool, cheap and portable device for ST.


Very interesting…which poll rate program did you use? Were you able to verify the polling rate actually changed via another program…if so, which one?

Also, which win7…x32 or x64?