#Sony PlayStation
1 messages · Page 48 of 1
I start getting phantom inputs after a while too, with DualShocks.
Did something change in the core? How difficult is it to roll back?
I may go hook up my second, several months outdated MiSTer, just to see if I can reproduce the behavior.
Have not tried my GunCon in a while but now I am curious as well.
@marsh star have you posted about this over on the MisterAddons discord at all? I have, in both Product Support and SNAX64 channels, but haven’t gotten many responses yet.
Thank you for validating that I am not crazy.
Make sure your DualShock is not haunted
@still palm do you have any idea what might be causing these problems with snax and dualshock? Nothing has been changed in the core for years.
I’d be interested to know if anyone else is running into that issue
@marsh star @sinful thicket
Whilst I don't have a solution, if it's an issue that doesn't take long to reproduce. I wonder if it's possible to try and older PSX core
Just in case if something recent knocked something out of balance.
The snac module was not touched at all.
https://github.com/MiSTer-devel/PSX_MiSTer/tree/main/releases there is the old releases if you want to try those but I don't think that's it.
@shut escarp checked with latest stable core and everything works fine for him.
i used to have this problem, but i mounted a virtual memory card in slot 2 at all times and it doesnt occur for me anymore with snax64
my issue was a bit different tho, my inputs on my ps1 controller would autofire instead of completley stop working
I wouldn't test with Final Fantasy 8 since it known to have something weird going on, that I can correct if I have both memcards mounted or setting to real MCs
https://github.com/ShendoXT/padtest I'd try this to see if the problem shows there
Maybe the cable has an issue?
Worth noting I have a SNAX (2p) not a SNAX64 (4p but only for n64). Don’t THINK it should affect PSX but just clarifying
A really long extension cable, or maybe alot of other things plugged in could do something
yeah maybe a bad connection in the cable or a broken solder joint somewhere but I think they said they tried multiple things.
we may need to see a pic from @sinful thicket to see if they're doing anything weird
@still palm
I don’t see any obvious signs of cable strain. I have it mounted via Velcro specifically to reduce the cable strain, since it’s a thick cable and the I/O port is mounted rather high.
(nice case)
I guess if that is enough to have broken the cable somehow, I have a support issue to open with MisterAddons.
no psx extension cables right?
Do you have the sega gensis snax adapter? it uses all the pins and I think you would see something weird there if you had something wrong with the snax itself.
And did you have more than 1 psx snax cable adapter? just to rule that one out
On second thought if the problem was the regulator on snax that knocks the 5v down to 3v3 that the psx uses, sega adapter wouldn't catch it but the n64 one might. the Sega one would verify no problems with the io though
I’ll give that all a shot this weekend and report back; thanks. No extensions. Multiple controllers, multiple adapters (2) tested.
I tried to fix something that doesn't work on real hardware anyway @lyric pagoda. Donald Duck Going Quackers works the same on a real PSX when two controllers are connected. The intro is skipped. When you disconnect the second DualShock, the intro plays correctly. When you turn the second controller off in Core and the first one is set to DualShock, the game works correctly. It works the same way.
I'll come back to Crash Bandicoot 3 topic later, but there's no error in the core of the Donald Duck game, it's 1:1 to the real PSX.
Ahahahaha that’s crazy, @ivory verge you’re a badass for finding that!
pal version
For me, the matter is closed. In core, all you need to do is disable pad no. 2. (it's enough to set it to digital)
This is a lesson for me to “get off my butt and pull out the PS1” instead of wasting hours looking for something that doesn't exist 🙃
…so what you’re really saying is…
MiSTer wins again!!!

Using an Xbox Series controller in DS mode last night to play Wipeout 3 SE, I ran into a problem with the controller losing the connection while wired. I haven’t really experimented to see if the controller has the same problem with other games or even cores, but I’m curious if this is something anyone else has seen.
the issue could be closed as real behavior 🙂
Yesssss!!!!
So, no issues with N64 via SNAX64 that I can tell. I played a few games, including Sin & Punishment. Maybe I'll try using a rumble pack as well!
Genesis, however, was completely different. In Xeno Crisis in 6 button mode, my retro-bit 6 button clone acts as if X and Y are both being pressed all of the time. I reseated the controller and everything, no luck.
I also confirmed that said Genesis controller does not exhibit the same issue, in the same game, on my Analogue Mega Sg.
Xeno crisis isn't a good test, it's a homebrew I think there was some problem with it.
Hmm, in that case can you recommend another game? Is there some difference between 6 and 3 button modes and support?
street fighter 2 would test all of the six buttons of the controller. but even the 3 button controller uses all 7 io. so any game really, just not xeno crisis
we were talking about retrobit controller over in Help
I coulden't get my Genesis SNAC adapter to work 🙁
what behavior?
Newer retro bit and other clone clone controllers have issues with Forgotten Worlds, which works fine on an actual 6 button with mode held. (forcing 3 button is recommended). Ms. Pac-Man needs mode to be held all the way to the splash screen to force 3 button mode, but it even works with problematic controllers when that is done.
I need to try Xenocrisis with snac, don't remember having issues on real hardware.
or be like me and find out you need to buy another hardware revision to debug an error because throwing in the dark is getting no where...
god that one pissed me off because i didnt need a 5th unit....
retrobits genesis repros don't work on snac afaik
no-ones quite sure why, something to do with the cable length
The Retrobit Saturn controller didn’t work on SNAC for me too. That’s weird lol.
i wonder if any of the magic blue1 did for the psx cores snac support ccould help with this
if you have an older io board for mister, you might have same issue, i did on my OLD one where there was 1? less pin connected in it, you can do a minor bodge wire to fix if so
id have to dig out old board so see what all i had to bodge
Honestly that worked out for the best for me, I had @humble thorn’s REFLEX Saturn board and installed it in the Retrobit shell so now I have a super badass controller.
Any luck testing with an OEM Genesis/MD pad? I guess my Retro-bit isn’t a good test case.
FYI I can use my N64 controller with a rumble pak and SNAC without issue.
Any one have any issues with the Army Men games not getting in game anymore? All of them seem to be getting stuck on loading screens for me except Army Men 3D, Army Men Air Attack 1 and Army Men Air Attack 2. Tested a couple of the ones not working on a core from July 10th 2025 and they get in game fine there
I see it too. I can get sarge's heroes 1 to load, but not 2
where is kuba's test core. let me look at that
happens on Kuba's 12/26 test core too
Ok, it works on the 12/04 core
so it's something after that
doesn't work on 12/24 either
@ivory verge
My SNAC adapter had some weirdness where ir refused to be controller 1 for some reason, I think it also only partially even functioned as controller 2
All my other snac adapters work fine
Oops, nevermind, I thought this was the Genesis channel
@spark ermine @chilly kraken thanks for checking!
The only change after 12/04 was for Trasher/Dave Mirra. I tested nearly 200 games, but it's impossible to test them all. Army Men - these games must do something unique. I'll check it out. If I can't fix it, we can always add it as an additional option for Dave Mirra, not globally.
army men doing something weird rings a bell
cant really find anything, but maybe it was on PSIO
maybe this is something we can implement a PS1 database.txt for
have these fixes apply per game
this is not the case of 'you need this thing turned on for the game to work' this is a 'you turned on something clearly labeled as unsafe and now a game doesn't work'
Auto overclocking option left the chat
Can someone @ me when MiSTer wins again plz?
I won’t be able to sleep at night until this is shown to be true
the original reporter hasn't said if they were using increased disc speeds - hopefully not
I'll try to solve this problem; these games really require a different approach. I can't do it right away. If I can't fix it, we'll decide whether to permanently revert the change or add an additional option for Dave Mirra (it actually works with most games without any issues). We'll see.
I'll add Army Men to the test games when changes are made 🙂
This repair wasn't easy for Dave Mirra, I knew there might be some edge cases
sorry to rain on your parade, kuba
No 3DO 😂
i havent emulated PSX in like a million years, so i am a bit out of touch!
is there a compatiblity list like for N64? or does everything just work nowadays for PSX?
any reason the hblank/vblank setting is not always enabled, what are the downsides of enabling that?
fixed hblank/vblank mostly just helps with keeping the image synced. i don't think it's necessary to fudge with it unless you're noticing dropouts or other problems in your video signal
as for compatibility, assume it's just as good as real hardware minus whatever issues are listed in the github repository https://github.com/MiSTer-devel/PSX_MiSTer/issues
hblank/vblank is off by default for crt reasons iirc
wipeout2097 is good for testing it, screen shakes when you hit mines (which can cause hdmi to drop out)
enjoy the core, it's one of the best!
fixed blanks off(default): will have sync drops for few games that rumble the screen
fixed blanks on: no sync drops, but black bars around the whole image for many games
Sync drops for hdmi? analog? both?
HDMI for digital out. Don't know how DirectVideo analog will react.
VGA out is not influenced by this setting at all
yeah on analog the screen will shake just like it would with the original psx
cool, thank you guys! 🙂
I think it was fixed but maybe the GitHub wasn’t updated.
https://github.com/MiSTer-devel/PSX_MiSTer/issues/270
Was MGS fixed?
i believe @ivory verge fixed that yeah
That sounds right 😎
Maybe there are other issues on the GitHub page that need to be closed too then
Yes, I attempted to fix it, but I only uploaded the core here for testing; I never made a PR. I need to come back to it sometime and do it properly in the code.
Guys
On superstation one, I’m confused how the SNAC ps1 ports/controllers will/do work?
Can you use the ps1 controller in menu and in psx core? Ie on my mister I use a usb controller then switch psx core to SNAC.
But I would like a superstation, but want to use it just like a console and ideally not have two controllers to get into core then play - is this possible?
The way that snac works is that it speaks directly to the core, and since the core doesn't know about the menu, it can't send commands to it
Hmm
Makes sense but contradicts what I think I’ve seen in clips
Yeah, Taki is working on a forked core/main (we aren't sure) that allows for snac to control the menu, but you would need to ask him more about that as it's not part of the official mister project at the moment
certainly very non-standard
Ah gotcha
So forgive my ignorance but people who have received their devices are simply using SNAC like we do as mister users
Keyboard / usb / Bluetooth device for menu then using the SNAC toggle to play psx via SNAC
yeah no-one really knows how that works yet
Seems like little value to the super if that’s the case
Unless you can tell the super to cold boot into psx core only and therefore the SNAC controller works (but then no game selection would be possible 🤔)
No apology necessary - I wish taki had prepared new folks better for this stuff.
Yeah, if you want to use PSX controllers, they will only work with the PSX core, and you will need something else (usb keyboard/controller/remote/etc) to control the menu.
In all other cores, you will need to use a usb controller/keyboard
though we've seen clips from takki of the psx controller doing menu stuff too now so
yeah, he said it was a forked core (or main?)
Possibly that will probably going to change on some forked cores too, to have the PS1 SNAC working on some other cores, like the GB, GBA...
yeah, that also was in the video I think
as @proven seal said, he has reinvented LLAPI
So the hardware ps1 controller port is referred to a ps1 SNAC yeah?
yes, exactly
and in a vanilla mister set up, that only works with PSX and only with games, not the menu
super stations are out. !N! has one
Lucky one.
software seems way behind hardware here
I just wish taki would ship something. seems wild that a product that is not marketed as a mister at all just ships as a mister in a nice shell
This is the way
This
I have mister so get it but genuinely the product must confuse a lot of people
yeah, it was marketed as a PSX clone that also happens to be a MiSTer
I assumed using Robert’s mister core BUT skinned into a ps1 (with the option to side load cores etc as it’s basically a mister)
It will be cool if he gets the snac stuff working. It will be a nightmare to keep updated for folks who care about that. But I suspect a lot of the folks who bought super stations are new to mister and looking for an analogue-type device
which, again, wild that it's just a mister
from what i've seen its just the usual mister software with psx set as default core
N can confirm, but I think it literally just boots into the mister menu with standard options
Yeah exactly
They’re obviously very transparent by giving more
But maybe isolating the psx core should have been the default
oh its not even defaulting into psx? lol
Mister menu
this is the most baffling part to me. having the front controller ports be SNAC but booting into the mister menu is rather poor foresight, in my opinion
i mean, you'd be equally stuck unable to pick a cd-image even if it did
Yeah, I feel bad because they obvbiously needed to get the hardware out the door because people had become restless (and they were already late)
But there’s a console mode coming that hopefully will fix some of these things
but man, not having the software ready seems wiiiiild
yeah i kinda get the impression this is stunt shipping just to say they did it in 2025
The hardware was finished before the software, but it's not like there are 1000 SuperStation in the wild so far.
i dont get why they couldn't of sorted the software side out ages ago though they shouldn't need final hardware to do that or anything
@stone ore Sorry that you get all the early adopter benefits. I know you have a mister but if you need help with stuff, we can totally help with anything that is part of the mister project. Things get a little sticky when it comes to stuff outside of that
I realised I didn’t correct anyone
I do not have my superstation yet
It defo does boot into mister however currently
i ordered one too, hopefully they sort things out before too long
I love mister but genuinely would just love the superstation to basically be an analogue type product that supports the mister community if that makes sense
I hope they make an option to lock it down into a console mode for example
Did you go for Grey?
Looks like blue are being shipped
And did you go for the dock for cd loading
Or standard operating procedure for modern consoles these days...
yeah, its obvious that it's not the experience he wanted to give folks
but you live long enough to become the villain etc etc
Genuinely, if the mass public receive theirs and can’t open the psx core it’ll be a bad look
Hopefully there’s a software fix before then
The mister already supports auto-booting to a core on poweron right?
It's just a setting you would do in an ini file?
If it does I didn’t know this
Or at the very least, to a game (and thus the core too)? Could just auto-boot that attract screen then
"mister-boot.chd" I mean
Not sure if there's any copyright issues with that one
put bootcore=PSX in the mister ini
Ah that’s cool
So… I’d still need more than just my SNAC ps1 controller tho to load the game right?
I can see a modification to get PSX SNAC to work in the OSD... problem is if anyone uses the SNAC for anything else...
I am assuming there would be no way to detect if it's not a PSX controller and disable the SNAC OSD feature
No buttons on the console itself ot navigate the menu like the M64?
🤷 #fpga-platforms message
How would that work if you used the dock to connect some other SNAC device though
Could it cause damage?
we don't know. you know as much as we know
I do? I assure you I likely know a lot less 
that tweet is the only thing we have seen. we dunno how he is doing it, so we don't know how it affects anything else.
So you really do know as much as we do 🙂
I don't like the fact that the snac port is paywalled behind buying the CD drive. That annoys me. Could've just made the front USB port a SNAC port but no, apparently that's feature creep but by all means, let's put an NFC reader in it for tap to cuz so many more people use that than they do SNAC 🙄
And on top of that, you can't get the attachment drive/dock whatever in clear blue. Just solidv grey or black. So if you were someone who wanted to buy one, but have the blue console, it's not a cohesive and matching design
(unless there is a clear blue dock and I missed that somehow)
The Dock will be the same color as the console.
The black version shown is just for final mold testing.
It looks like there is an option for clear blue now. I didn't see one back when I pre ordered the SS
So at least there's that
Taki was clear on that multiple times on social medias.
it was always to color match the main unit you ordered
yeah, it's part of the preorder page IIRC
I can't keep up with anything posted on Twitter because I don't see an option to filter by newest post on that platform
people feel strongly about this device 🙂
I just feel strongly about the main unit not having SNAC that's all. But that's because I want to use my existing snac adapters and controls for things like N64.
I would bet if you tested users, way more preference for usb in the front for generic usb controller input than a snac port (not the psx one, but the one you mention)
Stuff I already paid money for
I don't need anything other than so it's put someone like me into a frustrating position of, do I spend another $45 plus tax and shipping just for the port, or do I take my chances with cheap USB adapters
I view the unit a little differently perhaps
It has snac for the console it’s built for / around
Fully hear your opinion but surely if you care a lot about other cores you really want a mister etc
Just my thoughts
yeah, I think this really is supposed to be a PSX FPGA. The MiSTer side is less of the focus
we know it's a mister, but I really don't think it's being sold as that
I didn’t know it was using mister at first just thought it used the psx core that is on the mister if that makes sense
yeah, for sure. I thought it was going to boot into the PSX core by default. then I thought it was going to boot into console mode by default.
Yeah agree
The value to me is that is will basically be a psx that I can plug and play
Love to get away from the more tech side of the mister
So if they can make it boot like a psx then throw a disc in! Perfect
Yeah, I can understand extra USB ports and the M.2 port being in the dock, but the SNAC port is odd. Would have been better to have the NFC in the dock and SNAC port in the main unit
Oh by the way, is there a built in level shifter for the SNAC in the dock? IIRC, you don't need a level shifter for PSX SNAC right?
I wouldn't know.
it is what it is at the end of the day. I'll figure things out for myself when the time comes. I'll most likely end up just getting a Mayflash N64 to USB adapter for my controllers on eBay for cheap and calling it a day.
after being dissatisfied for years with all of the available mister case designs, i should really buckle down and learn to create my own
by all means, if you have the 3D printer necessary to do it yourself. If I had a 3D printer that's what I would've done already
i don't currently own one but it's something i intend to buy once i move. perhaps until then i could brainstorm ideas for a case
Well, you end up being limited somewhat by is inherent shape as a stack, but if you want to go the route of getting connectors that'll let you spread the stack out but still connect everything, you can open up your design options by a lot
There are places you can order something 3d printed for yourself, in some cases even local libraries have 3d printing services provided what you want to print won't take too long
Hi mate
This doesn’t seem to work, the mister is overwriting what I’m changing when I re insert the sd back into the mister
🤯
Remove the first ; symbol on the same line.
This could be embarrassing - I’ll try now 😊
And you have a 20 secs timeout before autobooting the core, on the next options - at bootcore_timeout.
If you need to, you can reduce it.
Thank you
Does this look better
It doesn’t auto boot on these settings unfortunately
It just re wrote my saved changes too
I don't know if you can use some values <10 for the timeout.
sorry, there was a bug fixed for bootcore but not sure if released in stable yet
Sorry, what does the timeout do? Hold the mister in the main menu for that period of time?
if not, if you put the whole filename it'll work (while fix isn't stable): bootcore=PSX_20251204.rbf
Awesome I’ll do it now
I'll let you with Zet-sensei who is the expert here, just helping on this as I looked into the bug. Going offline for 2h now
Thank you Nelson
It worked? Not sure what core version you have
No it didn’t work
I used my core version title and saved it but auto boot not activating
I’m running update all
Then will change the core auto boot to the latest rbf on the off chance that fixes things
What’s the full name and location of this ini file?
It’s in my mister at present so will feedback soon
But it’s on the root and Mister.ini
I’ve checked that my mister isn’t set to mister_alt for example
Vaguely remember someone with issues with bootcore_timeout, I’d try commenting/removing it.
Cool
I’ll delete that block of txt thank you
Would it make sense tho that my changes are no longer showing when I re insert the sd card back into my pc? It’s like the mister is re writing the ini info on boot up
Yep defo won’t work
And it reset my ini file too
Just deleted the ini
Let mister re create it on the sd card
Edited it again and nothing changed
Still doesn’t auto boot etc
edits to your ini file are not sticking?
this is odd, not sure if it's the mister, never heard of this behaviour. are you taking the SD card back and forth from the Mister to a PC or something to make file edits?
And you're unmounting the SD card after saving the file and before removing it? I'd run some tests on that SD card maybe
btw the easiest way to interact with the mister filesystem, is either via ssh or mounting the SD card via samba https://mister-devel.github.io/MkDocs_MiSTer/setup/games/?h=samba#transfer-roms-over-the-network-with-sambasmb
Yes I’m removing the sd itself
I have two sd cards for mister and they’re both performing the same way
This highlighted file is the config file I’m amending and it’s then auto changing when in the mister
Moved it here in case the smart folks there can help #help message
Hi
Issue is seemingly my side
I can save the ini changes
But when the sd is removed from pc then re inserted the changes haven’t actually saved
Tried on my MacBook and when I try and save the ini changes it tells me it cannot save!
Happening on both SD cards too
I can add / delete and amend games etc fine 🤯
I’m done lol
What is the name of the ini file you’re changing
And try just changing the ini file on the mister itself
There’s a script for it
It’s this
Can I type to change the Ini on the mister? I thought just toggle options
Yeah, there’s a script called ini_settings
Cool
I just thought it would only toggle resolutions etc not allow me to add auto boot cores etc
I’ll give it a whirl
I think the advanced option is what you want for text entry
If you still have that test core, I’ll test it out this weekend 😁
It's because they made a modification to the board wiring (my suspicion is that it was done to prevent it from working with Genesis to USB adapters, but I'm not sure on this).
Seems like one should avoid buying any repro controllers from Retrobit
they were never that great anyway imo
I didn’t know! I bought a Saturn one and it didn’t work with SNAC and surprise surprise, that controller had issues working with real hardware too lol
The saving grace was being able to use @humble thorn Reflex CTRL board in it and it’s a pretty badass USB controller now imo
The reflex adapt is really versatile
The Saturn controllers had issues with the pad breaking or buttons getting stuck too. On the retro-bit wireless Saturn pad, directions on the d-pad just stopped working. Reading Amazon reviews, it seemed to be a common issue.
It's almost like the cheap ali Express controllers somehow do it better for a fraction of the price
8bitdo M30 Genesis works with SNAC
i think 8bitdo have really done themselves proud over the years. now they make licensed controllers that no one questions the quality of, and are discussed as upgrades to the console stock controller.
oops, I meant Reflex CTRL. My bad.
IIRC the issues with the Retro-Bit 2.4g Saturn controllers has been resolved in their second revision
The issues were the pad breaking where with the wired, though I had buttons sticking on both wired and 2.4G. With the BT controllers, only the Saturn ones had the pad issue. The Switch compatible Genesis USB pads added a home short cut that interferes with controller operation. The Genesis 9 pin controller compatibility issues were actually introduced in later batches.
Originally there was no snac on the dock either, some of us were studying the possibility of making custom converter for the ps1 ports. They underestimated the snac need from the community and did not design for it.
Because enough of us complained about snac it was eventually added to the dock.
Of course it would be better to have on the console, but I don't feel like I'm being paywalled considering all the other extra features from the dock and the console itself compared to the prices of adding them to a normal mister stack where any extra accessory/cable is expensive as well.
- they're working on making ps1 snac compatible with all cores.
they could've easily consolidated all of the features on the main console and the dock into one unit, and their custom SNAC cores are essentially a different flavor of the LLAPI cores that we've had for years now
... Why would you want your repro of a decades old controller to intentionally not work with USB adapters?
because they want to sell their 2.4g version which comes with a usb dongle for PC
and they also might sell a USB wired version
Hi. do we know if the COP0 is supported by the core?
https://gbatemp.net/threads/i-tricked-the-ps1-hardware-to-ignore-spyro-3s-uncrackable-security.678736/
I went on a journey to bring the legendary double-jump glitch from Spyro 2: Ripto’s Rage back into Spyro 3: Year of the Dragon. Not on an emulator. On real PlayStation 1 hardware first.
Here’s the problem.
Spyro 3 did not merely remove the glitch. It actively fights you if you try to restore...
COP0 is there of course, but maybe/probably not implemented to extent explored by this hack
That was quite a bit of drama in that story.
From what I see it is a simple instruction breakpoint triggering some interrupt.
It's fully described here: https://psx-spx.consoledev.net/cpuspecifications/#cop0r7-dcic-debug-and-cache-invalidate-control-rw
The core does currently not implement it.
In general I didn't implement anything where haven't found any example use, because I almost certainly do some mistake and cannot test.
Could certainly be added. If someone is interested in it, I can give some hints on where it most likely needs to be done.
What does that do and why is it important?
Would it take up a considerable amount of space of the FPGA to add?
So people would buy their USB repro instead.
They made both? As seperate controllers? I see, thats annoying
Any obvious reason that mister doesn’t fill my vga connected crt monitor? Thank you
I don’t want to ruin the image by trying to stretch it using the monitor settings etc
I think you need fixed h blanks
You don’t ruin the image that way, that sort of scaling on a CRT will just alter the waveforms controlling the beam; no image processing is performed
Mostly interrupt/exception control, and some debug related control registers that the hack in question uses
Thanks for answering my question!
The first part of your explanation sounds important but not the second part. Obviously the goal is to always fully replicate hardware so I’m just trying to identify what’s mission critical for the general public.
Has anyone figured out the regression with the 3DO games yet
or do I need to go back to an older core
army men ?
- Added CD Backward seek filter option (U) to the misc. section,
fix for Dave Mirra / Trasher games is now optional
snac to control the menu would be a serious game changer for me
Does having a 2nd controller or mini keyboard in arms reach really mess up your misters feng shui that much? I see that complaint a lot and it’s never really been a problem for me and I use SNAC a lot
unironically yes lol, at least for me. being able to control everything with a single pad goes a long way to keeping things feel organized and "console-like"
Then use a DS4 and be done with it unless you’re playing something that warrants SNAC like a light gun/steering wheel/n64 game is how I look at it
Or m30 or whatever your insert-catch-all controller of preference is
that's what puts me in an odd position though; i prefer to use original controllers where applicable. i don't even own any modern controllers, actually
for now, i just use USB adapters for everything, which is mostly fine enough. i would love the lower latency offered by SNAC, but i can't justify investing into it when it can't navigate the menus
the difference between SNAC and a quality USB adapter is on the order of 1-1.5ms, It's completely not worth worrying about
yeah, that's why i'm not super torn up over lacking a SNAC setup. been using the bliss-box adapters for about a year now and they've been great, 1ms of latency i believe
You could also use mister remote to control it
That even lets you load a game directly instead of loading the core first then the game
No, it's just some additional trigger for the exception and one additional target.
In worst case it would make the 2x core a little harder to build.
The main purpose of this was to use some external debugger, but as we have no way to even add it, it's pretty pointless besides such hacks.
By the way: our cheat engine could very likely do the same, but they have chosen to do it that way so that it also works on original hardware.
I am guessing the Duckstation code posted at the end is not just a simple Gameshark style code or compatible with the mister cheat engine?
It's not described, but it likely is.
However, instead of overwriting the required values in RAM directly, it instead implements a routine in RAM that is triggered using a CPU breakpoint to jump into it.
This second part is not available in the mister core.
You could very likely make a much shorter cheat code to just overwrite the value in RAM directly without the routine.
It would work in Duckstation or Mister, but not on real hardware.
However, this code also doesn't work on hardware, it would require a patched CD image.
Just simply changing the RAM values on mister would not trigger the copy protection?
Wait, wasn't the copy protection only on the PAL version now that I think about it?
From what i understand it only happens when you modify the "normal code" outside of breakpoint exception.
The game cannot observe the whole ram and usually you wouldn't be able to modify ram on the real ps1 without modifying the code.
Do certain cores, including PSX, not support VGA scan-doubler (ie, 31khz PC CRTs)?
Or do I just have something grossly misconfigured
PS1, saturn and N64 don't support the scandoubler. you'll have to come up with some custom modelines unfortunately
Did anyone ever figure out what was causing Driver 2 to be slow?
(Destruction Derby can lag a fair bit too on cactus creek, but I suspect that’s probably the same on actual hardware)
I own both those games, I can check against real hardware
Destruction Derby I’m almost certain is accurate to original hardware
I played it a tonne as a kid. I don’t remember slowdown but I’m also not surprised. They pushed the boat out with 20 cars in a race
Driver 2 I think @ivory verge confirmed is slower
Third mission specifically or something like that
why did you add an extra n and e
🇬🇧
I like your fancy words
Actually I think this one is French
Like about half of our words 😅
(The fancy ones)
Yes, I checked it on old and new cores. There are drops in some places to 9 fps. The only improvement is with 2x CPU + turbo high.
Thank you
It would be interesting to know why that is - because I guess it’s not the only game that suffers
the only game with such drops is Re-Volt
If you are sure it runs to slow at a specific point compared to real hardware, please upload a savestate and I can analyze what the bottleneck is
i think they said it was just the start of the very first level
the place they mentioned is the beginning of mission 2 or 3 (i can look up which one specifically) under the L train in Chicago where you have to slalom between some dirt piles at the beginning
it is the very beginning of a level, but not the first level
i did confirm along with kuba-j that it is indeed much slower than hardware there
ah yeah i just found it "It's immediately noticeable on the 3rd mission where you have to keep up with an elevated train. Right at the beginning of the mission, you have to slalom around a few piles of dirt, and go through some small pieces of trash on the street. The slowdown is extreme."
that is where it is very noticeably divergent from hardware. it seems to maybe run slow relative to hw in general on MiSTer overall though, that part just gets extreme
would definitely be a good spot to see what is going on
have you got a save at mission 3?
no, but there might be one around somewhere, or a cheat. i have to go to bed due to work tomorrow at the moment
looks like there are complete saves on gamefaqs, i bet you could load into the mission directly from one of those but i haven't checked
This can be reproduced in every mission; there are points where the game slows down significantly. You can easily notice this in the first mission just by driving through the city.
If nobody beats me to it, I can try to create a savestate at that point later today
Ah wow, just like real Chicago
For light gun games
Will my setup accept a light gun?
VGA connected monitor and this adapter
try it out
what kind of monitor is it?
VGA crt
it might work
I don’t know if the yellow composite video cable will need to be connected to a composite input on a crt tv?
I think it will?
the yellow cable from the guncon goes into the one on the snac adapter
Yep, then the other end is female, so I’m wondering if I then need to feed to a crt tv (not monitor)
no
it just has to go into the snac
guncon on a snac works on my VGA monitor, but it doesnt work for everyone
Hmm strange
But good thank you
Hopefully it doesn’t need that video cable
I always assumed they included the female for people who don’t have scart etc
I used to have the light gun and can’t recall ever using composite cables
I dont believe that will work with a guncon, That is a 480p monitor and guncon only works with 240p
Ah okay
Any way to scale it back to 240p
I don't -think- so but some VGA monitors have wacky settings / features.
Generally, you'll need a regular 480i/240p consumer television (or PVM monitor) for light gun stuff and they wont play nice with PC CRTs which are 480p minimum
This is the reported resolution on time crisis if it means anything
Hmmmm, That's more promising then
Just got a gun to try
If I set the psx core setting to snac the trigger does make a sound but nothing on the crt
If I set to guncon in the core then I get nothing
Any ideas incase its user error
jumper needs set on analog io board v6.1,it's next to user port (set it to io6) or the yellow plug won't work and you'd need to hook it up elsewhere
or it's just the monitor anf GC not seeing the flashes
Kuba strikes again! 
Will try and find some time to fire up Driver 2 now and get a save state for the slowdown
Let's hope the tutorial isn't as brutal as in the first game 😄
There is no tutorial
I wanted all the games that came out for the PlayStation launch in Japan (1994) to be fully playable 😄
A. IV Evolution
Crime Crackers
Gokujō Parodius Da! Deluxe Pack
Mahjong Goku Sky: Atsushi
Mahjong Station Mazin
Nekketsu Oyako
Ridge Racer
TAMA
all are now working correctly
Amazing, thank you!
what problem ? 🙂
yeah , it's the launch list 😄
only Parodius and Crime Crackers had a problem
i don't see any slowdowns , but yes there are bug with screen adjust
but it's nothing serious
@tropic stratus , @ivory verge – Here's a save state just before the slowdown in Driver 2. (There's also a regular save that should be at the start of the mission, in case I did the save state too late to get the problem).
thanks!
Lemme know if it's borked in any way and I can redo it 🙂
(I remember that mission well 😅)
This is how it’s currently set - looks okay?
yes looks correct.
Thanks
Then in core set the core to snac
The side a and b buttons work to progress the intro in point blank but then the calibration doesn’t work in terms or the cursor - trigger button work or makes a sound tho
if trigger works and screen flashes or it makes a sound you'll probably have to try it on a crt and not that pc monitor (if you have one)
Oh sad times
I assume it’s a monitor issue then
Screen does flash and gun shot sounds but it doesn’t show the cursor on calibration screen which I THINK it does on og hardware at the calibration screen
Thanks for the input tho @still palm
I was able to get PSX working pretty well on my NEC FE700 using this custom modeline:
forced_scandoubler=1
vga_scaler=1
video_mode=640,16,96,48,480,10,2,33,25175,-hsync,-vsync```
This works well on the N64 and Saturn cores for me as well
@ivory verge THANK YOU for what you're doing with this core! ❤️
PSX core is definitely one of the top 5 mister cores imo, and it's really awesome to see kuba come in and button up a lot of the longstanding issues. this is shaping up to be a perfect core 
JT says that the NGPC is getting saving either this week or next...
🎶 🎶 🎶 So you're about to change your tune 🎶 🎶 🎶
this song sucks
Can he take it out of beta plz
When it’s finished….
Fine 
I have been guilting him on his server. don't worry, I got u
Doesn’t it still have some major issues?
yeah
Currently, it can't play Pac-Man
even though it was on a list of working games early on
Core is getting renamed to Neo Geo Pocket Color Forever 
Thank you!
the game will slow down starting at around timer 1:57:5x and run fast again at around 1:56:00.
This has nothing to do with where you are in the world or where you are looking at. Instead it's because of the CD loading.
You can easily see this with the LBA overlay in the debug menu: the game only runs slow while the CD is reading.
If I remove the CD completly and then load the savestate, the game will slow down forever after 1:57:xx
If i "open the lid" then load the savestate, the game will freeze at the same spot it would slow down and continue when the lid is closed again.
If I use fast seek and 8x CD speed, the slowdown is even worse.
Turbo has no effect on low and medium, but doubled the speed on high, which means that CPU speedup alone doesn't help, but DMA and GTE speedup helps.
Looked into my emulator and what happens:
- game sets new position on the CD to read
- reads it
- issues pause (causing a long wait time)
- setmode (drive speed set to same value, so has no effect?)
There is an interesting comment about pause in the code:
"todo: should this be swapped between single speed and double speed? DuckStation has double speed longer and psx spx doc has single speed being longer"
Duckstation however swapped this end of 2024:
from
const TickCount pause_time = IsReadingOrPlaying() ? (s_state.mode.double_speed ? 2000000 : 1000000) : 7000;
to
const TickCount min_ticks = (s_state.mode.double_speed ? 1000000 : 2000000);
New build with that change: gameplay now drops from ~20fps to ~14 instead of ~9 fps before while the CD reading is going on.
Not sure what the framerate on a ps1 is, but I would be very surprised if that wasn't the root cause.
Wow, thank you for looking into this, fixing it so quickly, and providing a detailed explanation on the cause of the issue. Can’t wait to test the new core!
Is there any other game you think that might be affected positively/negatively so that we can focus our testing to be most helpful?
I'm not sure if this can be of any use to Robert or not for the PSX core, but found it very interesting, and a good read about Spyro 3's anti-tamper mechanisms.
we discussed about it the last days already, tldr: it's nothing new PSX documenation tells everything about it
#1046940919607345272 message
a whole lot of games, if not all could be influenced. It's a quite common command, however most of the time the game would not busy-wait for it to finish.
I would like to wait for someone to compare against ps1 if framerate more or less match now.
Thank you. Agree, that would be great to know
Oh no, I missed all the driver 2 action. 
Good news! Now you have another excuse to play it! 😉
Hey guys, is there a hot key to turn guncon on or off? I think I hit I pushed into the trackpad on my dualsense and it said something like guncon off and then on. Could this be the issue people are having? Kind of like how you can change from digital to analog mode in game but with guncon on/off instead? I couldn’t get guncon to work for weeks until I changed some setting and just worked as intended
It's a great version of a great game! Better than driver 3 for sure. 
when you push the trackpad it says Analog on and off not guncon
This talk about a dormant debug command reminds me, I might have misunderstood but I think I recall hearing that the PSX has basically a terminal for debug/dev purposes that is still running internally even though it's not really hooked up to anything? And that many retail games still output to this since it's not like they needed to remove it if the user would never see it. Is that true or did I misunderstand?
It's actively sending it through the serial port? What about games that had link support?
Also by "comms link" do you mean those ISA cards with the parallel port that were meant to connect to those action replays from the '90s?
Nice, I have one of those parallel cards still in my Pentium 3 computer, I used to use it to back up my Saturn saves
Is there a way to view this terminal output on the core or is it not implemented?
AFAIK it dosen't, I meant if there is a way to view that data somehow or if the core dosen't process it
Hey
I hope there is a simple fix but not looking hopeful for me
I don’t really understand why the cursor / tracking doesn’t work
Did you have to config the guncon somehow?
Naturally I’ve just set it to snac and hoped it would work
It does just work when set to snac and used with a 240p TV instead of a 480p monitor
I was having issues for a while as well. I even tried a reflex adapt and installed updated mapping config and did update all again. Then I decided to try the SNAC adapter again and just started working so unsure how o can help
Didn’t you have the composite sync plugged in to the wrong place initially?
Mines connected just here
I’m sure it’s the issue on vga monitor and not crt tv disappointingly
Just a reminder that if we can get some eyes on this, maybe Kuba and Robert can coordinate a new release #1046940919607345272 message
I don’t know if Robert’s build contains Kuba’s changes too, or if it’s a different branch?
So it might be good to get them both in one build at some point. Though we can still test independently
yeah, I think after both are verified as good, we can merge them. Might be better to have them separate for now in the event that there is a regression in one of them
I have prepared a draft PR, but I am waiting to see if anyone finds any problems. There shouldn't be any, but you never know.
I have noticed that if you are touching (not pressing) the trackpad on a DS4 or Dualsense at all while pressing the Menu button it will switch to “Lightgun mode” @shut escarp
for the longest time i thought there was a bug or something that was turning on Lightgun mode when pressing Menu and then I finally realized that I was just baaaarely grazing the touchpad when I pressed it 
Oh I see. That’s what happened to me. Ok. Good to know.
Tested Robert’s latest build - seems much better on Driver 2 generally
Not seeing as much random slowdown - now the slowdown seems to correspond to things actually happening on the screen, which makes sense
Before it seemed to occasionally randomly slow
Which makes sense given his message
But that scene in particular is still bad - and I suspect it was on real hardware. Just a bit of a perfect storm - very slow scene combined with the additional slowdown from the CD stuff
Also tried a few other games and nothing seems broken or worse
I will check Driver 2 on real hardware this weekend
It’s a great game 
Yep. I used to play it loads back in the day
Driving around and being able to get out of the car was mind blowing back in the day
Then GTA3 came out
I never managed to link two consoles up 😭
Oh, I did with the Xbox 🤔
oh neat i didnt know R4 had 4 player mode like that
ive done a bunch of split screen R4 with jogcon and negcon
it’s so cool that some of these longstanding issues are still getting fixed…amazing!
Omg this makes me so happy. I beat Driver 1 last year on MiSTer and was bummed that Driver 2 wasn’t working. Thank you for looking into this issue!
the Bushido Blade system link was very cool to experience
i modded 2x ps1 with the Xstation and imported the cable
I'm thinking about selling mine.
market was flooded with xstation modded consoles
i was quite tired of ebayers trying to knock my price down to the competition despite my listings having a full re-cap, region free bios and other QoL treatments
Dude with both players in first person? It was so awesome.
Dosen't the xstation take care of regions anyway?
Is that your collection?
Also those cases are ugly compared with the ones in Europe 😅
I think the European ones are ugly so there!!!!!!!!
The people are beautiful though
😠
You really don’t like them? 🥺
I dunno, I just assume it’s nicer
I forgot how they looked like
Our Dreamcast cases were nice too, but flimsy as hell so they all broke
What about the old style PS1 cases?
Ours were fatter and had a wider edge - they looked unique
Although tbf that one probably isn’t original because I think they all said “PlayStation” on the black edge
Were they everywhere, or just in US/NTSC?
I think they were just in the US, and only around the launch era, they switched to steandard Jewel cases pretty quickly
same with saturn, I have a few games with those long style cases like that
I think my copy of NiGHTS is the long case version, like this
That’s pretty cool
Saturn took longer I think to switch to standard cases
Although the Tekken one kinda sucks since the artwork is empty at the bottom 😅
Yeah those are great
Actually, wait, did Saturn even switch? I think they mght have stuck with it in US
And just used standard cases in Japan
I have the one PS1 game with a purple spine, THE ONLY GAME
bkueish purle
Like Mario’s Sky
It generally allowed for a bigger manual, but they had to put in a piece of foam to protect the CD:
IIRC, some of the PS1 long cases were also all a brittle black plastic, no clear parts
Here's an example:
Yeah I have a collection of longboxes
The middle and right one are the older style of the tall boxes, that were all opaque black plastic instead of having a clear part
Oh, the edge was ridged originally
It was the fake printed one that I hated
Must.. Not.. Make.. Obvious.. Joke...

^
What was the plot?
Was it like Conker’s bad fur day?
No it’s like some collection of really shitty jigsaw puzzles for kids
Yeah you’ve got read ridges and fake ridges
Fake ridges smh
i've only got one longbox - doom
lol early Namco manual art
Hell yeah
you mean with right?
The only problem is the D ain’t ridged
No ridged D for thee
King's Field 
it's slow , it;s ugly but i love it 😄
Devs will never make it big
D?
Quiz dreams, eh, Roberto? 
i'm just going to assume its about helping that blue lady find her lost clothes
That would be my assumption
does not sound like a pleasurable experience then
That’s looks awesome
That looks way cooler than my bullshit
You guys should make a list of your hidden gems 
can you stop making me want to go unbox my physical games and get shelves for them again
i just packed them all away
Googled it and that game looks unplayable unless you know Japanese, sadly 🙁
Unless there’s a patch out there
you can get a lot done pointing your phone at games to translate these days
yeah depending how reading heavy it is
Maybe with the DE-32.5 Maxi we can use AI to translate everything
Looks like a hybrid game with some RPG-ish elements
So might be quite reading heavy imo
Want to play an FMV game where you work with Adam West (original Batman) to find who stole the Chaos Chip, a microchip capable of advanced problem-solving from your ex-girlfriend?
@serene nebula do you have a list/catalogue of all your games?
Of course

Hope you’re good at poker and other casino games.
Sadly I don’t know the rules of poker
my collection is too small (thankfully tbh) to need a database for it or anything
It’s way too confusing for me
doesn’t matter, Adam West helps you
only my absolute favorites
Yeah I went buck wild unfortunately
now I have mister and don’t touch those games lol
All my old games are in my parents’ attic 😅
Left them there when I moved out for uni. Think they’re still there…
quite alot of my collection is actually loose discs too so it fits in a cd folder
But no core can be perfect. Why don’t you play on original hardware? 
honestly? Laziness lol
Do you want to test Driver 2 for us and compare to Robert’s new core? 
you're way more likely to get issues from your real PSX's aging laser than from the cores remaining issues
I was just trolling him based on an argument we had earlier 😄
But we kissed and made up, and then he showed me his ridgeless D
i've got a real PSX with xstation and hdmi mod and i still use the core 9/10 times
Yeah, I have no reason to use original hardware for any of the cores MiSTer supports
I won’t sell the consoles/games, but I probably should because they’ll just stay in storage until I die
yeah mines just for verifying bugs at this point heh
Sadly I’m a PAL gamer soooo 
we get pal bugs too
As if any of you ever play those games 
pal is less of an afterthought on roberts cores than others
as he likes to play in german
Most of my PSX game collection on MiSTer is PAL
For real? Nice
There’s the odd game that I’m tempted to go back to the PAL version. The accurate speed NTSC ones are much faster than I remember 😅
Like the WWF Smackdown games
I feel like 95% of my PS1 game playing time was spent playing the 5 WWF games I had 😅
Ever since the first PSX test cores appeared, I've been testing PAL versions, not many people wanted to do that 🙂
i prefer PAL wipeout - it was made in europe
I have a few PAL games where I heard those versions were better. Wipeout and Colin McRae are the ones that come to mind
I don't know, it's also a matter of sentiment. That's how I remember those games, that's how they worked and that's how they looked to me. I don't have that with any other system, and PS1 is the most important console ever for me
I can relate to that
Most of the NTSC versions seem the same to me, just with the exception of the Smackdown games that are too fast, and a couple of others that have different names (rapid racer)
Oh, and Gran Turismo PAL had a much better soundtrack iirc
Only thing I do not like about PAL is the pal covers for Resident Evil
Resident Evil 2 looks grim
But the Japanese covers, pure gold
Edward Scissorhands Tag Tournament
I think RE1 had the same cover as in Japan
So it was basically as Miyamoto intended
Nope, I am wrong. Sorry
Playing Driver 1 to test out the recent core builds. Really enjoying it
Forgot how good it was
I know I have a Darkstalkers longbox, I think that's my only one
I think some softwar emulators alrady can do that, never tried it so no idea how well it works
It’s pretty easy to do on a PC tbh
they were some quirks and edge cases.
i also preferred certain bios screens to other and removal of any trace of PAL50
Just compared the smackdown games and they seem to be the same speed across PAL and NTSC. Just seem faster than I remember now I’m old
Oh, I might have found a bug, but it’s a bit tricky to figure out…
WWF War Zone PAL seems to freeze on the intro FMV (video freezes, sound keeps playing). It seems like it’s fine 9/10 times on the latest stable core (2025-12-04), but on the two recent test builds (pause Swap and 2026-01-13) it seems to freeze more often than not. I thought it was a bug introduced between the stable and recent test builds, but if I take a save state and reload it, occasionally the stable build will freeze too.
It might be that I made the save state “too late” on a potentially "bad" core (i.e. one with an error) which causes something to mess up occasionally even on the stable core - but it’s interesting… NTSC version seems completely fine though
bin/cue redump images running from a usb ssd (never had any problems before)
It's in the opening FMV, so should be easy enough to reproduce in case someone wants to try – but here's a savestate as well
I'll dig in again tomorrow to see if I can figure out any consistent conditions. It's definitely much less frequent on the stable core – unless I'm just imagining things
i tried your save state , it's totaly random, even with old cores it could be happen
hangs on second run, on core 20250825
test bulid 20260113 , sixth run hangs
The problem exists, but it is completely random regardless of the core version.
Thanks for looking into it ❤️
Maybe it also happens on original hardware. I don’t have my PS1 nearby so I can’t test to confirm 🙁
strange that it happens randomly. There is not much randomness in the PSX core other than VRAM (due to being in DDR3) and slightly async Video clock.
I understand if it's in a game, but during boot time is really strange
only happens during FMV
it's not always even the same point
And honestly, if you start the game and watch the intro once, you'll probably never even encounter a moment where it freezes. I didn't.
can you skip it?
yes
ok, so at least some very easy workaround
When I test games, I often leave them to play FMV/gamplay continuously.
but WWF games are not my favorite 😄
If it's random could that be an issue with memory timings of the FPGA competing with the CPU for the DDR3?
Without these random timings of DDR3, the only other randomness is the video clock, but that level of randomness is minimal. So it's very likely the DDR3 as source of this
Some very old core had synchronous video out, but it's probably too old to be worth testing.
it's not completly removed, but requires a new build: https://github.com/MiSTer-devel/PSX_MiSTer/blob/main/PSX.sv#L262-L263C6
But to be honest, I wouldn't spend the time unless you really love that intro
(It pretty constantly froze for me - not freezing was rare)
If you really love the intro you can see it on the NTSC version. That one didn’t seem to crash for me
It never freezes for me on first playback on any core
And it was always at the same point for me
Feels like we might be getting some different behaviour 😅
I wonder if the release of the SuperStation One is going to make these bug reports even murkier, since it uses an open bios
I mean, I don’t care and it’s not important. I was just sharing in case it was useful/relevant for development
If I can stomach the process of cold rebooting and reproducing, I might try and record some stats - just because it’s interesting that we see something different 😄
I had a weird issue like that with a saturn game, if I left the intro running it eventually freezes, nobody else could replicate it
In the grand scheme of things it’s not important
I will check the core with the video sync out option enabled
It’s only one of the greatest games of all time 😭
(It’s not)
Maybe the game itself is broken 😄
NiGHTS? 😛
Me too, but I don’t recall watching the FMV too much 😅
I don’t remember an issue though, so it’s provably a core thing
Probably*
I suspect Kuba would have noticed by now 😅
Or Robert
I mainly test PAL versions, I have no problems with FMV in my set of games
so it's definitely not a common occurrence
Personally, I don't want to spend too much time looking for the cause of this wwf bug.
its very minor
FF8+9 both have very long FMVs and i completed both without any crashes with the german version.
When testing FMVs the fear effect games gotta be some of the best games to test, they run FMVs in the background a lot of the time.
Yes, I played Fear Effect, it works fine
I thought my PSX core was bugged the other day because of IQ's weird soundtrack 🤣
It completely does not fit the game and I thought it had somehow cached the tracklist from a previous image I had loaded that session before I switched to IQ
That's maybe my favorite OST on the system. For PSX, I don't think there's a single OST with so high production values, except maybe Heart of Darkness.
Man, how I love both of those games, in part because of their music...
It feels like I am exploring a forest or the world map in a JRPG
Not... doing cube-based puzzles in a dark void
What the game is trying to "tell" is an sci-fi tale, akin to the original TRON film.
The delicate classical/symphonic music fits the 70's/80's notion of high culture and reflexive sci-fi, as opposed to today's only "fi" based around mindless action with no philosophical themes.
IMHO, sci-fi became dangerous for capitalism, so it was swapped for stupid StarWars-like stuff so people don't dream about utopias, as that would hurdle the enslaving process we're in
I can't tell if you're being sarcastic or not
it's also hilarious given what Lucas has said about Star Wars
IQ's music is not very different from, let's say, Robert Wise's Star Trek 1979 film or Star Trek:TNG, but these are huge marxist utopias so current terminal capitalism sci-fi has moved away from the themes and the culture on those on purpose
Not at all. I am very serious. But I will leave it here.
You mean that Jar Jar is the key to all of it?
Jar Jar is a stupid character in a stupid saga that was fun in the 70s and 80s, nothing more
Jar Jar is from like 00
Also that was an actual quote from Lucas IIRC
That many have memed to hell
But as I said, I will leave it here. Too many conditioned people around I don't want to offend.
Simply understand that, at a time, actual good music like the one in IQ was to be expected on a videogame with an sci-fi setting, because that was how culture worked when there was such thing as culture.
And I explained why you feel like that about the music in that game.
For me, it fits the game perfectly. And I wish videogames had evolved in that line, instead of the insufferable POS they are today.
Please don't try to tell me why I believe the things I do, that's pretty rude
You’re not cultured enough to appreciate it
.... how do you not see that as being rude?
I’m kidding
ok, ok, just ignore everything I said, sorry I made you feel like that! 🙂
open your third eye
No need to complicate our lives with such discussions. You are perfectly fine!
Ah no, I stay away from spiritual junk, so no 😄
I can't, it has Astigmatism
You’re not cultured enough to appreciate it
If you mean retro-cultured enough as people in some parts of the world where religion is on a high revival, yes, then I am proudly uncultured 😄
Well, hopefully you become cultured enough some day to realize how silly you're being 😛
Oh, direct insult. No better arguments huh?
I will let time and history make it's job, there's nothing I can change.
Have a nice day! (And don't take it personal!)

Anyway, of the games I was trying out I finally managed to put a decent amount of time into Vagrant Story, I game I remember seeing commercials for all the time and thuoght it looked neat but never had the chance to play. Was very impressed by it, felt more like one of Square's PS2 era cinematic RPGs than a PS1 era RPG
Though I miss getting EXP from fights
does vagrant story ever move to a setting other than brown caves/dungeons?
i've bounced off it so many times in the first hour
I only played for about 4 hours, but I was underground that whole time
... plenty of non-underground settings in cutscenes though, and technically the intro fight is in a castle/cathedrial
Sidney was so cool. It was just pure luck that I didn't end up with his tattoo
There’s some variation to the environments. You’re on the village streets for a while. But mostly dungeons, yep.
Yes
There's also the forest
@tropic stratus Question, I made someone do a test for me on the PS1 core to see if cop0 is fully implemented, it is often neglected in emulation. So I had high hopes for MiSTer since it should be accurate to the real deal. Sadly, according to the test for my specific use case, it didn't work. So perhaps it can be improved. I helped debug it for Duckstation as it also didn't properly implement it. Unlike what is commonly believed, it does actually get used, for Legacy of kain for example using it as secret storage for libcrypt keys. What I did with it however is completely different, I used it to change EPC in real time to bypass Spyro 3's security without leaving a footprint. I did this on retail hardware by re-enabling the debug exception vector at 80000040. So far only 3 emulators support it + the real Playstation. I can offer a patch/mod for Spyro 3 that re-enables the double jump from Spyro 2 to prove this, only when it succesfully triggers, cop0 should be properly implemented
Question being, was cop0 taken into account for the core?
#1046940919607345272 message
Lol that is my article (dramatic by design I might add to also entertain the non technical reader)
So the case is already known
It is indeed using execution breakpoints, the use for them are re-enabled at boot in my mod. As the exception vector for it is nopped out by Sony. But some games still use it as hidden storage
Was it only used as storage for libcrypt keys, or did any games use it for anything else?
@ivory verge
Hello, how are you?
Could you please let me know if you plan to release a build including the c-blend option for the most recent updated versions of the PSX and N64 cores?
Thank you very much for your attention and for all the great work you’ve been doing
Don't think it was used for much else. It gets set up at boot, you can see it when stepping through at the bios startup. Probably as an init value. So besides lybcript, I think Sony told all the devs to not touch the debug exception vector, unless a debug unit was being used. So a debug unit bios probably has it enabled by default, on retail ps1 the vector is nopped out, but it is usable, just gotta redirect it
Either way, it works on a real retail PS1. Which to me is always the golden standard
a test rom for it would certainly be helpful
I agree that 100% parity with the original hrdware should be the gold standard, but just a few years ago just about nobody thought PS1 would even be possible on MiSTer, including the person who wrote the PS1 core. I would love it to be 100% but I don't think it's possible on the DE-10.
rob already said this is possible
its just not in since we didn;t know how to test it
Well whomever I need to send the rom to, let me know @abstract drift if an explanation is needed on how to do a double jump that is fine too, or a noob friendly version that just adds a life when jumping
ideally the testrom isn't a whole game, it can take hours to simulate a game up to gameplay
I just make simple tests since I don't own debugging hardware, the games are my debuggers along with 3 emulators. I know how to program it in ASM and know the expect outcome, that's how I get away with it
I said I don't think it's possible to make the core 100%, not specifically this feature. Also IIRC didn't Robert say that basically the more you add the more it makes stuff like the turbo core less viable?
But yeah, if there is a test rom that can help see if the feature works as intended when added. I wonder if this would impact libcrypt games though
The current implementation of them I mean
i'm kind of curious how legacy of kain even works on the core
if it really needs that
Don't libcrypt games had subchannel files to get past the copy protection? Also IIRC it's only PAL region games that ever used it
yeah we've got the subchannel stuff
Was the infamous Spyro 3 DRM only in the PAL version too?
it doesn't bypass libcrypt though it kinda just makes it work naturally
The method I use with the BPC/BPCM/DCIC allows one execution breakpoint at a time, but I cycle them and know the naturally flow and path the game takes. Spyro 3 for example will stop you at boot on a shitty chipped system, but that can be bypassed invisible by changing the corresponding conditional jumps to always take the branch by trapping it, then changing the EPC, then in the exception vector you use K0 and K1 to do the compares to the EPC, knowing on which branch it happened. Thus defeating the AP/AT while all integrity checks pass. Did the same thing for my double jump
That initial stop you are talking about, that's for non-stealth modchips right? Surprised it had that, games started to remove that check since it goes off regardless if your copy is legit or not, and just a simple gameshark can bypass it
Yeah exactly, but those gameshark codes can not target assembly when it comes to Spyro 3, normally, it'll trigger AP later on. Spyro is extremely aggressive. A ram code that checks the 14 loops works, but the one that is out there is redundant being 28 lines, it could have been 2
But the method I now use can do assembly changes, without being spotted. So, Sony left us a really nice tool
This is also how some anti-cheat can be bypassed on more modern systems, simmilar concept
It's full on assembly, the first line reads when the jmp to the game becomes active, but goes to 8000B0BC instead where it installs all custom exception handlers before it boots into the actual game. 90 (4 byte) code types can be split up into the gameshark 80 (2 byte) style except for the first line, it's a 4 byte conditional IF equal. I helped conjure up most of the ideas for the Duckstation code types, I'm one of their in-house cheat code providers
I know the game shark would not have been able to get past Spyro 3's DRM, I meant the first one that's just the modchip check that other games had too
Legacy of cain works, as storing data in COP0 works perfectly, it's just the breakpoint exception that doesn't trigger as so far no one needed it and I had no way to test it even works without writing tests myself.
If you have a test, feel free to just post it here, tests should be small enough.
In theory this should be pretty simple. The main question for such things is always in which cycle to trigger, so it throws the exection at the correct program counter.
This is more complex than in duckstation, as the core replicates the full 5 stage CPU pipeline.
But if your test does check for the saved PC in the exception handler, it should be straight forward.
(for your code it might not matter if the exception and return point is really exact, but to be exact it makes a huge difference)
Speaking of legacy of kain, is there a way to speed up the loading time when opening the menu on mister? Great game but those loading times are really a bit much.
maybe try changing the CD speed to 4/6/8x
Afaik it's something with data not being stored where it's supposed to, with the devs having to rush the port on psx.
Or just play the game as Miyamoto intended 
My testing so far is game-driven, real hardware first. I step through the flow in a debugger, then validate on a retail PS1 by reflecting state in-game (for example, mapping a value like EPC after a debug exception to something visible such as score/lives), so it is not a synthetic “test ROM” yet.
If that level of evidence is useful for MiSTer, I can help, 100%.
Two key points from what I have observed:
The debug exception must trigger at the exact address. DuckStation initially fired it 4 bytes early, and that difference mattered, because it would cause conditional branches to fail, that got fixed. In other words, if the breakpoint is set for instruction at X, the exception needs to be taken at X, not X-4.
EPC behavior: the general rule is EPC is effectively read-only in the sense that writing back to the EPC register itself does not behave like writing to normal state. It is commonly treated as “you don’t patch EPC by storing to it”. In my tests I did it anyway, as it does no harm.
I still want to verify edge cases, but the consensus is that “write EPC back” does nothing. What does matter is what the CPU loads into EPC on exception, and what you return to.
Fastest reproducible test I can offer is Spyro-based: I have a very simple pass/fail signal.
If COP0 debug exception behavior matches real hardware, a jump adds a life (or similar visible flag).
If it does not, the game either crashes or the life never increments.
If you want something cleaner than a full game dependency, I can do a minimal asm-only test too, but it will take more time.
If you are fine with “modify anything for the test case”, then I can adapt the Spyro test quickly and provide exact expected behavior, addresses, and what constitutes pass/fail.
And yes, my test does check the PC/EPC and uses it for a conditional branch. That's how I've done my testing, so it is exact
Wow, thanks @dull wedge for the extensive research into this! Glad to see more people getting involved in improving the core. 👍
It benefits everyone, and I'm a huge PS1 enthusiast.
Eventhough I don't own a MiSTer FPGA system yet, it'd be cool when I do perhaps get the SuperStationOne, that I can rely on it to be as accurate as possible and retire my poor Playstation haha
PS1 is for sure a very important system on MiSTer for a lot of people. The game library is arguably the best of all systems on it.
I still keep my PSX with Xstation too, but to be honest havent found reasons to boot it up haveing the current MiSTer core. Though I used it for comparisons early on in the core development. I was the one checking Misadventures of Tron Bonne for every new core release until it got fixed 😊 . Those were the times!
Anyway we appreciate your help
unfortunatly Spyro doesn't work well:
- it doesn't show me the status of registers after the exception and I cannot validate against PS1 (or duckstation if that is accurate enough)
- simulation of a game takes ages
Not sure if you have some framework already, but if not you might look into this repo: https://github.com/RobertPeip/PSX
I already added a lot of tests there. Those are easy compiled, easy executed and can provide simple text outputs without much effort.
If you add something there that captures the state after the debug exception and prints it to the screen, I would be very willing to implement it.
Any other test environment is of course also fine.
Ill see what I can do when I have some spare time, I'm sure we'll be able to figure something out 😁
The SuperStationOne looks so cool, but not sure I could click fast enough to get one 😄
wdym? You can click add to cart right now?
really??
Ships Q1 2026 or Earlier // Limit 1 Per Order The world's first affordable FPGA gaming console. Inspired by the PSone. Supports PS1 games, memory cards, and original controllers. Region Free. Tons of I\O. Plug & Play. Fully compatible with MiSTer FPGA cores. Learn More Package Contents: SuperStation One 64GB Micro
Hi, it is possible to support MemcardPro over snac? @tropic stratus
The card works but the game ID is not currently being detected
I know this doesn't answer the question but it's very easy to export/import the memory card data between mister and mcp. I migrated all my memory cards by changing the file extension and copying it.
(Rest of the file name also has to be what the mister/mcp expecte, obviously)
Yes, indeed it's not the same or as convenient as simply inserting the same memcardpro I use on PSX or PS2 into Mister via SNAC and playing the same saved games I already have seamlessly from one place to another.
i have not tried it, but there seems to be a solution: #1046940919607345272 message
IIRC there is no way for a stock PSX to send that information to the memcard port, you would need some kind of modification such as a PS-IO, xStation, or a modded BIOS, so not surprising that the MiSTer core does not support that without a modified bios either
(From what I understand the PS2 does, but only for PS2 games)
So, I did some work today, will do some more work later. Made an executable that will turn red on wrong EPC behaviour, Blue on correct EPC behaviour. Since I've not made an executable like this before, it's been a nice little experiment. So later I'll set up the dcic at boot, and have the normal exception vector re-arm it
Should be plug an play after that
Actually, might maybe only need to set it once, so no need to trigger an additional exception
It's always a question of how far you want to go.
I can imagine all kind of weird things one could test:
- breakpoint exception at some instruction that would throw an exception
- breakpoint exception when interrupt happens
- breakpoint exception at branch instruction
- breakpoint exception in branch delay slot
But maybe let's keep it simple first and just put it somewhere safe and check for COP0 registers to be correct after it triggered 😅
Yes, it worked, only in card slot1, but works
Yeah, lets keep it simple, build from there. I've built a small list of possible dcic values for execution breakpoint, reading from an address, value change I believe. There's a lot that can be done, I'm probably just scratching the surface. I already figured out ways it could be use for anti-cheat. But that's besides the point for testing purposes, just a cool side effect of the findings
There's a dcic value as well to only catch branches. Could be cool for a trace function when logged to easily find no clips and other stuff. I have something like that for Duckstation (Pugsy made it for me, he does a lot for Mame)
Curious what SNAC adapter you’re using. I’m only able to make it work with the Crown Arcade adapter.
But it works reliably and I use my MCP2 with my PSX. Haven’t tried slot 2 but it’s reliable in slot 1 with the hacked BIOS
Do we still just have the same old choices for PSX SNAC adapters? Only ones they were able to find that did everything I wanted were like $80 or more
retrocastle one might do everything, but let someone that has one confirm
That's the closest, but the memorycards are a bit of a mess in their placement. Also it was designed to work with their mister stack and while it works with others, it's a bit of a mess to use it with others
there might be different versions of MCP2, I'm not sure of the differences though
The MCP1 had differences, earliest versions had a FPGA in them, but this apparently made them have all sorts of compatibility issues
Newer versions just have the EMP32 and work a lot better
oh maybe you're right and that's what it was, my memory is a little fuzzy on that
Debug exception setup for 0x80010064 & bootup starts at 0x8000B100, then jmps to 80010000
When 0x80010064 is trapped, it brings us the debug exception handler at 0x80000040
0x80000040 has a jmp straight to 0x8000B184, here the DCIC gets de-armed, the epc gets increased by 24
This brings us at 80010088 where we turn the color blue
scenario 1, If non of this ever happens, the screen should turn red.
scenario 2, debug exception never happens so EPC never gets changed, should turn screen red.
scenario 3, a good ol crash (Unlikely)
scenario 4, blue screen but not re-arming or looping back to 8000B100
It eventually all loops back to 8000B100 & starts it all over
My tv, or rather my cheapass upscaler for some reason does not like this shit, so it just shows no input (actually might need to add vblank, else my code is too fast, still fine for a test though).
Maybe for that use case scenario I should try a one shot instead
If I disable the cop0 emulation in the debugger on pcsx-redux it will turn red 🙂
Full cop0 emulation only gets emulated on pcsx-redux when the debugging is enabled, so this is a great test
@tropic stratus So, I tried and I did notice that pcsx-redux does allow writing to the EPC, on the PS1 it does no harm either
Might ask Nicolas Noble about that
any chance to compile this into an exe?
With CD, i have to go through the BIOS, which makes simulation impractical. Same effort as debugging with Spyro.
If not, maybe I can create a savestate shortly before the breakpoint...
It was an exe before I made it bin/cue, so yes
@tropic stratus
If it doesn't display at all, I might need to add a vblank later. Having some people over to fix the heating, so if so it'll be later today
Has anyone had the opportunity to test this core #test-builds message for longer? I haven't found any bugs.
No issues that I’ve found, but I’ve been dividing my time between your core and Robert’s pause swap
I have merged core with pause changes
Happy to test that also 🙂
Thanks. Will copy that over and continue playing games testing
It's included my changes from PR + Robert's pause change
So that’s basically everything to now?
Aside from anything you’re currently cooking
Yes
@serene nebula pin request please 🙂
I'd been jumping between the two cores too and I haven't noticed anything either
Thank You ❤️
Freezing intro FMV in WWF War Zone (PAL) update: following @ivory verge’s suggestion, it seems like Turbo low is sufficient to prevent the freezing 🙂
(Thanks kuba)
tried to run it in my emulator.
As it doesn't output anything, I can't tell what is going on.
I zipped an export of the first million CPU instructions executed, maybe you can give me a hint what is going on and what to look at.
I can find the 0x8000B100 entry point, but it never gets to 0x80010000
Like, never?
After the routine at 8000B100 there is a straight jump to 80001000
When I get back ill check the log
I just see that when it jumps to 8000b100, there is just no data at this point, so NOP is executed until it hits 8000b940 where some weird data is, causing an exception, going to 80000080
looked at the exe header:
- initial PC is 0x8000b100[offset 0x10]
- load address is 0x80010000[offset 0x18]
- file size is 0x200
So it copies 512 words from the exe to 0x80010000, but starts executing from 0x8000b100
when i look at other .exe i have, they typically have both initial PC and load address set to 0x80010000
Ok, I suppose it is the bios, as an exception should not occur otherwise
Yes, but it isn't neccesary. My Spyro 3 patch works the same way. However, I am curious so I will check the log when I am back home.
8000B100 should have data there, if the BIOS is correct (SCPH1001) hle or modded bios could probably set it around 8000A1000, which is wrong. I could potentially isolate it to the 80010000 range completely so BIOS version wouldn't matter as much, as the exception for debug will always happen at 80000040 anyway.
Don't know, I used the exe handler code duckstation had in 2021 for my emulator and the core.
Well, lets say I reimplemented it, so I could have made mistakes back then or it was changed in the meanwhile
Duckstation was updated to properly emulate cop0 now
Else it'd just break
I helped Stenzek debug it
I run the scph1001 bios, which alligns with allowing to boot at 8000B100
But I guess as an universal ALL bios, 80010000 could be the new start
do it worth a gun con 3 for about 60€ for this core? Including sensors
I really want to play point black
you could save money and get a wiimote
if you want a high end lightgun experience pay a little more for a gun4ir/openfire
This should disregard any bios version
it's weird that I have to put in my social security number to get it to run but whatever
?
sorry, just a joke 🙂
finally, someone will keep it safe for me
I certainly shouldn't be trusted with it
Foh sho homie, foh shooo
Might need to add a dummy file for the disc version to satisfy the PS1, gonna test that right quick
started with a guncon3. just finished building out a gun4ir with a guncon1. it's night and day. guncon3 felt like a laser pointer presentation because you spent more time looking at the onscreen cross hair. gun4ir felt like what I remembered playing on a CRT aiming down the sights, but on a 65" LCD
ok, now I get red screen, cop0_BPC and cop0_DCIC get loaded, but exception does not yet trigger. Can probably start with that.
There is some issue you might want to look into: there are endless writes to GP0 which overflow the GPU Fifo.
Neat that the emulator is still useful at this point
Is there a way to enable rumble when using a Dualsense? Or is that not supported?
Any recommended snac ps1 adapter to buy in EU? I have a JogCon etc i wanna try
It should work, you can test in the main menu by holding L and pushing X or Y
I am aware, it also goes in a constant loop but at least it serves its purpose for now. I never made an exe like this before, so anything graphical is new to me. I usually do things under the hood, invisible. Never had to setup the GPU for anything before. So aside from the COP0 assembly routine, you're witnessing a new experiment from my side haha
But I'll further work on it. It's a great test. Tried it on a bunch of emulators just to see which ones can be trusted. It's a very small list
On a keyboard?
No sorry, the x and y button positions, so Square and Triangle on the actual Dualsense controller
while holding L1 trigger
Thanks I’ll try this when I get back home.
By the main menu do you mean the start menu of the mister or the pause menu within the core?
pause menu (OSD) in the core
Ooo conflicting stories
i know it works for sure in the main startup menu, never tried in-core
Got it. Thanks guys
oh does it only work from the startup screen? I don’t really use autofire, but it seems odd that it wouldn’t just work using the OSD in general
not autofire, rumble
ohhh
Ok so startup menu for sure
ok maybe listen to @forest trench 😆
that's where I always test
New auto fire feature is awesome btw
I’ve only done that rumble test a couple times a long time ago and can’t remember where I did it
Can finally use a proper controller for ps1 shmups
I'm going to look into this then. thank you
guncon1 or guncon2? I have never had one, so not sure what is better to look into my local marketplaces
Guncon1 with SNAC is -flawless- with a CRT on PS1 core
Guncon2 over USB is -very so so- with a CRT but works on all cores and doesnt need any extra adapters
oh but you're talking about scrapping a controller to gun4ir it, rip
I used a guncon1 that was sold as not working. I did a basic install, nothing fancy like a solenoid or rumble motor. it's not bad if you have basic soldering skills
@tropic stratus So, I've further improved the program. I gotta do some more testing later. Would be cool to eventually have it as a standard for emulators, console copies etc as test rom
Here we go, confirmed to work on real console and all the accurate emulators! @tropic stratus
It's a one shot process, so start the break at 80010000 for the whole thing or 800100D4 one step before the custom exception handler should kick in, if behaviour is accurate
Will take a look the next days to implement it. Thank you
exactly
You're welcome, glad to be able to provide some form of help
Fixed
The worst case for me
I think I need a drink🤣
http://redump.org/disc/125702/ This disc also has an issue with software emulators except recent versions of Duckstation. According to VGPC discord: the demo.vb track is loaded and meant to be tran...
It was my worst nightmare; you have no idea how many times I returned to this topic.
The other mirumiru demos also work correctly now, without freezing during movie playback.
I remember that MiruMiru demo when I was helping Stenzek on DS.
But never found a solution before I left back then.
Anyone in here actually already owns a Super Station One?
So, how is it?
Great so far - a Mister PI with a better form factor.
It just lacks some new menus - for new comers.
You got the fan upgrade as well? I heard it ran a little hot, not sure if that is truly the case
That doesn't seem too bad
just excited for when mine eventually arrives...
I brought it to a classic gaming association earlier, they were enthusiast.
My major criticism is the lack of a SNAC bypass without the Dock.
That's fair
Kinda agree the snac bypass probably should have been on the front USB
Doesn’t seem to be working for me on my Dualsense
Oh wait, it’s not working on my Dualsense edge but it is working on my regular Dualsense
Maybe the edge was never given rumble support on mister?
The problem is that the controller recognition stuff is tied to Linux updates on MiSTer
Because MiSTer Linux updates happen infrequently, support formore recent controllers tends to lag by many months
There is a way to force it by SSHing into the MiSTer and adding a rule for the controller VID:PID
Ah gotcha. Sorry I should have been more specific that I was using the Edge the whole time. I figured the 2 controllers were interchangeable from a connectivity standpoint but I guess not
the model id (PID) would differ between the 2, so even if they function the same it's just a matter of not recognizing it
I appreciate the help!
Is it worth submitting a ticket to someone to request edge support or is that just a waiting thing?
Edge rumble support I should say. The controller works fine otherwise
It's most likely already added, just waiting for the next MiSTer linux update
I only checked MIruMiru 1 & 2 completely, no issue at all found with the test core from today.
I just finished spider man on core from yesterday, gonna update to today's core and play spider man 2 now. Not expecting to find any issues but just in case
Or maybe not lol, I thought I saw new core posted here today
are you sure? the edge has already been out for years
@mild forge Ok I just checked, and hid_playstation.c hasn't been touched in 5 years
So before the Dualsense Edge was released
It's possible the VID:PID mapping for Sony controllers is done somewhere else, but in any case if the Edge is not being recognized as a Dualsense it may warrant an issue to be raised.
@manic citrus had updated the xpad driver some months back and added a bunch of new xinput controllers, which is what I was thinking of.
Dualsense VID:054C PID:0CE6
Dualsense Edge VID:054C PID:0DF2
interesting, you can see how it got added here https://github.com/torvalds/linux/commit/b8a968efab301743fd659b5649c5d7d3e30e63a6
yeah, our hid_ids and hid_playstation don't contain those changes for the Edge
Can't check right now due to energy crisis, but I just wanna know
Does the core support this dual analog flight stick as a control method mappable to a modern pad (ds4, dualsense)?
So they repurposed these into dual analog gamepads (pre-dualshocks)?
So you can play Descent on these?
yeah they have 2 modes, green light is like the sticks, red light is (more like) a dualshock
I have that, it’s awesome
