What's new
By:

Emulation vs Real CPS2 Hardware Differences

decimata

decimata

Student
Joined
Nov 16, 2015
Messages
92
Reaction score
25
Location
Marlborough, MA
What kind of things have people noticed between playing on an emulator vs real hardware for CPS2?

Has anyone noticed any speed differences, graphical, sound or other differences?

I'd love to see video side by sideanyone has it.

That being said, I prefer real hardware because of its single purpose nature, no distractions, no updating, that being said, I'm really just curious.
 
I notice a bit of input lag on my ipac+groovymame setup vs CPS2. Which is fine for me for every game except SSF2X.
I love being able to pause games in emulators.
 
Yes, lag might be the only issue depending on what monitor you are using.
 
AFAIK, CPS-II is pretty well emulated.
Still, I too do prefer original hardware even it would be 100% the same.
 
I think probably 99% of us prefer the original hardware. I've always used emulation for testing purposes and till this day I still do. I probably would have never gotten into this hobby if it wasn't for emulation.

Playing Mame and NeorageX back in the days sparked so manY great childhood memories. That's when I started to buy the original hardware.
 
There's a weird square box in Super Street Fighter 2's intro (Or Super Turbo, can't recall which at the moment). I also remember being amazed when I was playing Marvel Super Heroes at the arcade a few years back and got hit with Shuma Gorath's Chaos Dimension. The way the screen split up was very different than what FBA/Nebula/Kawaks do (Mame doesn't emulate this properly period). Though I was playing Marvel vs Street Fighter on Saturn more recently and Chaos Dimension was closer to what the emulators do, so I'm not sure if I'm remembering wrong. There's also no more arcades here in Toronto so I don't have either Marvel games to compare it to.
 
shingouki2 said:
There's a weird square box in Super Street Fighter 2's intro (Or Super Turbo, can't recall which at the moment). I also remember being amazed when I was playing Marvel Super Heroes at the arcade a few years back and got hit with Shuma Gorath's Chaos Dimension. The way the screen split up was very different than what FBA/Nebula/Kawaks do (Mame doesn't emulate this properly period). Though I was playing Marvel vs Street Fighter on Saturn more recently and Chaos Dimension was closer to what the emulators do, so I'm not sure if I'm remembering wrong. There's also no more arcades here in Toronto so I don't have either Marvel games to compare it to.
Click to expand...
Raster Effect. Emulators doesn't emulate that at all (they don't even try).

Other issue, which is more noticeable on shooting games involve the way the real machine renders video. The CPU get interrupted by the video processor in a periodic cycle which is synchronous with the screen timing (raster) by the video hardware which pumps the post rendered graphics (using DMA) from the PPU to the video DAC. That reduces the effective CPU speed from 16Mhz to about 12Mhz. That is explained on the MAME source code. And it's not emulated either. That affects how the game slows down when there's a lot of bullets on the screen. Making it behave differently on the real thing vs emulation.
 
I can't say anything from a technical or quantifiable perspective, although I wonder if what leo has just posted has something to do with it, but I find in particular street fighter alpha 2 very 'janky' to play on emulators - the timing of things feels quite off and inputs just don't feel quite right. I had far more experience with emulation of A2 than real hardware from about 2000 until a couple of years ago, and now that I've tried real hardware for a more considerable time, the real deal is much better. I imagine Vampire Saviour may be the same. SSF2T doesn't feel quite so bad, but the gameplay style may not lend itself to noticing quite so much (or perhaps I play charge characters a lot more in ST *because* of it? hrm!)
 
l_oliveira said:
shingouki2 said:
There's a weird square box in Super Street Fighter 2's intro (Or Super Turbo, can't recall which at the moment). I also remember being amazed when I was playing Marvel Super Heroes at the arcade a few years back and got hit with Shuma Gorath's Chaos Dimension. The way the screen split up was very different than what FBA/Nebula/Kawaks do (Mame doesn't emulate this properly period). Though I was playing Marvel vs Street Fighter on Saturn more recently and Chaos Dimension was closer to what the emulators do, so I'm not sure if I'm remembering wrong. There's also no more arcades here in Toronto so I don't have either Marvel games to compare it to.
Click to expand...
Raster Effect. Emulators doesn't emulate that at all (they don't even try).
Other issue, which is more noticeable on shooting games involve the way the real machine renders video. The CPU get interrupted by the video processor in a periodic cycle which is synchronous with the screen timing (raster) by the video hardware which pumps the post rendered graphics (using DMA) from the PPU to the video DAC. That reduces the effective CPU speed from 16Mhz to about 12Mhz. That is explained on the MAME source code. And it's not emulated either. That affects how the game slows down when there's a lot of bullets on the screen. Making it behave differently on the real thing vs emulation.
Click to expand...
That's not entirely true. Rasters are indeed emulated, if not then there'd be a lot of background effects missing in all CPS2 games. Multiple rasters in one screen isn't, this is mentioned with the Shuma Gorath's Chaos Dimension in the Mame source for CPS2. Dedicated emulators do emulate this technique, but it might just be for that particular super, and may not be emulated correctly (Haven't seen the original in a few years so I might be remembering it wrong).
 
fluxcore said:
the timing of things feels quite off and inputs just don't feel quite right.
Click to expand...
I agree, the feeling just isn't the same. Hits seem to connect at inconsistent (ranging from late to really late) rates, never at the times I'm expecting them.
It's most noticeable to me in the two-in-one simple hit combos, often times the 2nd hit don't register with the normal feel/joystick action.
Or a 2nd fireball (SS2T) while the first has just collided, normally this can be done so quickly the game will slow down on the release frame, on emulation I can't get that slow animation consistently.
 
samplehunters said:
I just connect the ipac through the PS/2 port thinking it would avoid the whole USB polling rate issue. I guess I should try it through USB someday.
Click to expand...
There was a lot of debate about this on BYOAC with many claiming that since PS/2 was handled closer to the metal it was faster, but a lot of newer PCs just emulate the legacy ports rather than handling them properly. There was an actual lag test done on BYOAC comparing the USB to PS/2 speeds and on a modern PC and they all seemed to come out dead even or with a slight advantage to USB (and this was before polling increase). Even the guy behind ipac/jpac claims there's no significant difference in speed.

I can't seem to find it at the moment, but like you I used to go for the PS/2 variants thinking they'd be faster until I read that test a couple of years ago and have been buying the USB stuff since.
 
Interesting. I'll have to try the USB version sometime.

For now I just deal with the lag/wonkiness of the controls. It's fine since I don't play the games competitively,
they feel 99.1% accurate, and that's enough for me.

Also, according to Niko, the Nesica versions of HSF2 and 3rd Strike are running on emulators. I'm really interested
in how well they play. Can't be that bad if people are willing to drop coins on them in the arcades, played on LCD
screens no less.
 
Last edited:
You must log in or register to reply here.
Back
Top