MiSTer FPGA Forum

Jotego just updated everyone that at the companys current pace, the CPS3 core is going to take another 15 months to trace all the custom chips. then the added time for circuit extraction, and then getting it to a point to start putting it together to test/demo.
this news was sad to hear, but i understand this arcade board has a bunch of custom chips that will take lots of time to trace and recreate in FPGA.

Jotego has had the bubble wrapped CPS3 board since 2021

and then everyone voted to start work on the arcade board for schematics tracing in late 2022

reading his patreon update for the current progress, 15 months for chip tracing is a long time. Jotego has advised this will be too much of an undertaking with his current staff of employees and resources, and has a couple suggestions. try out some sort of machine learning (autoCkt ?) to help with tracing schematics - https://github.com/ksettaluri6/AutoCkt, or to use what MAME has for the schematics and update the CPS3 core as the custom chips get traced.

im in the boat of lets get something "as good as" current software emulation (with the added benefit of no input lag, original refresh rate, and no sound latency) and use what MAME has, and then make updates with the real chips as they get traced

I feel like this is an April Fools joke. Is it tomorrow where Jotego lives?

PikWik wrote: ↑Sat Apr 01, 2023 1:41 am

im in the boat of lets get something "as good as" current software emulation (with the added benefit of no input lag, original refresh rate, and no sound latency) and use what MAME has, and then make updates with the real chips as they get traced

Getting the CPS3 was a stretch anyway. I'm more in the boat of "leave it on MAME until the MiSTer can be something different" (i.e., as close to 1:1 as possible). To me, the MiSTer project is about preservation so, despite how anxious I am to play the very pretty Street Fighter III on the MiSTer, until it's properly preserved and 'digitally recreated,' my RPi will do just fine. Take your time, Jotego!

mercuryshadow09 wrote: ↑Sat Apr 01, 2023 2:46 am

I feel like this is an April Fools joke. Is it tomorrow where Jotego lives?

In Spain where he lives, is December 28th, but he has a lot of international culture as well.

CPS3 = more fighting games

So I won't lose any sleep over it

fair enough

however, the CPS3 core did get picked by Jotego's patreon voters to devote development time to finish, and is a core that a bunch of people would gladly welcome. i understand most people reading this already know, but SFIII is still being played competitively. heck, even Justin Wong is playing on a MiSTer

and now its looking like development is a long ways away or getting shelved for the foreseeable future. 15 months was only the approximate time for tracing the chips, non-stop for 1 person. that doesnt include everything that would go into the production of the CPS3 core

ive always thought the CPS3 core would happen on the MiSTer, but it will be Jotego's swan song to the community. like, yes he will finish it, but it will be one of the last cores he makes for the MiSTer. which is totally understandable and he can take as long as he needs. everyone will be happy whenever it finally reaches our precious FPGA box

No so sure about that: if they go with the tracing he is not the one doing it so he should have time to work on something else and he is working on something else.

Konami CPU

We are testing the CPU now and we are honestly shocked at how well it is going. If you were here last Autumn you would remember that we were working on the NGP CPU and that progress was slow. Eventually, the freshman engineer who was working on it parted ways with us and I had to look for another person. I found a computer scientist from Venezuela, who had struggled around Europe to find an opportunity and hadn’t been able to find a job where she could grow her skills. She accepted to work with us even though we couldn’t pay much because she wanted to learn. Three months later, she is mesmerized looking at how the CPU she is worked on is moving data up and down. In her own words, when she started she couldn’t understand half of what I was talking about. But now, she has found a new passion and we have found a team member to rely on. We look positively at future CPU developments.

First FPGA synthesis with the new JTKCPU (compatible with Konami CPU).
FPGA resource usage is slightly below the Z80. Compared with the closest CPU, the M6809, we have traded off BRAM for logic cells. That makes the design simpler and faster.
Fmax ≈ 51MHz

From atrac17:
Just a few of the Konami boards I've loaned for development that utilize this CPU.

GX775 - Super Contra
GX999 - Rollergames
GX955 - Parodius
GX875 - Aliens

List of the games using this CPU:

That's an exciting list of games! I'm especially happy by the prospect of The Simpsons and Parodius DA!

The opportunity cost of CPS3 is just massive. It's 15 months of tracing work for the sake of one game (Let's be honest, Third Strike is the only CPS3 game that people really want). And that's all work that isn't even really transferrable to any other core either.

PGM, or even the Psikyo SH2 is a far more sensible platform for him to focus his efforts on. Even the Konami boards with their various, bitty slapdash selection of chips is a better use of his time than CPS3... especially considering that Third Strike is already available on many other platforms.

CPS3 should be one of the last platforms that he works on for MiSTer. The people who voted for it really have no idea what cost CPS3 really comes with.

The fact is : Jotego has submitted a poll and cps3 won. One game can be worth it anyway and address a lot of people.

Now information come that it will take maybe longer than expected because custom chips (maybe another challenge(s) coming in the next step of the dev can be expected)...I would say here nothing special and maybe PGM would and will follow the same path.

I like the transparancy of information shared by Jotego without impossible promise/roadmap.

dcubed wrote: ↑Wed Apr 05, 2023 12:27 am

The opportunity cost of CPS3 is just massive. It's 15 months of tracing work for the sake of one game (Let's be honest, Third Strike is the only CPS3 game that people really want). And that's all work that isn't even really transferrable to any other core either.

PGM, or even the Psikyo SH2 is a far more sensible platform for him to focus his efforts on. Even the Konami boards with their various, bitty slapdash selection of chips is a better use of his time than CPS3... especially considering that Third Strike is already available on many other platforms.

CPS3 should be one of the last platforms that he works on for MiSTer. The people who voted for it really have no idea what cost CPS3 really comes with.

It is possible that single CPS3 game, would accumulate more hours of play than all the other games you mentioned combined. At least in my case, it sure would. Let's be honest, most of the new arcade games that come for mister, I test them for around 5 minutes and never play them again. Some others, I would play once to the end. But a fighting game that is considered the pinnacle of the old generation of 2D fighters, which I already put thousands of hours in emulators and ports, that is still played in competitive tournaments today, would be much more than that. The "available on many other platforms" can be said about each and every game the mister can play.

I'm a Jotego patreon, and I voted for CPS3. The question was about what system I would like to see first, and that's what I wanted. If we have to factor in their amount of work (an unknown variable at that time), and CPS3 is not feasible, then yes, I have no problem in him spending his resources in something else I will also enjoy.

It looks like Haunted Castle will come before CPS3:
From Jotego.
The waveform comes from a simulation of the Haunted Castle core. We have started running tests with the new CPU hooked up.
As the game shares video chips with Contra, if we get the CPU to work, I think the rest will pretty much work.
Although there is a new PCM chip in this board too. PCM chips are usually easy to implement, so I am not worried about it.

Yes, it's not like Jotego and his team only works in a system at a time, plus it looks like the Konami ones have other developers involved. Many other cores will come before CPS3, and updates to other ones. This reminds me we also have the Neo Geo Pocket one pending, which is one I don't really care about, to be honest.

A PGM core, on the other hand, I would be very interested in because of the 3 CAVE games. I still prefer the CPS3 one even if it's for only one game.

Hey! Don't y'all be sleepin' on Jojo!

AmanoJacu wrote: ↑Wed Apr 05, 2023 9:38 am

dcubed wrote: ↑Wed Apr 05, 2023 12:27 am

The opportunity cost of CPS3 is just massive. It's 15 months of tracing work for the sake of one game (Let's be honest, Third Strike is the only CPS3 game that people really want). And that's all work that isn't even really transferrable to any other core either.

PGM, or even the Psikyo SH2 is a far more sensible platform for him to focus his efforts on. Even the Konami boards with their various, bitty slapdash selection of chips is a better use of his time than CPS3... especially considering that Third Strike is already available on many other platforms.

CPS3 should be one of the last platforms that he works on for MiSTer. The people who voted for it really have no idea what cost CPS3 really comes with.

It is possible that single CPS3 game, would accumulate more hours of play than all the other games you mentioned combined. At least in my case, it sure would. Let's be honest, most of the new arcade games that come for mister, I test them for around 5 minutes and never play them again. Some others, I would play once to the end. But a fighting game that is considered the pinnacle of the old generation of 2D fighters, which I already put thousands of hours in emulators and ports, that is still played in competitive tournaments today, would be much more than that. The "available on many other platforms" can be said about each and every game the mister can play.

I'm a Jotego patreon, and I voted for CPS3. The question was about what system I would like to see first, and that's what I wanted. If we have to factor in their amount of work (an unknown variable at that time), and CPS3 is not feasible, then yes, I have no problem in him spending his resources in something else I will also enjoy.

SF3.3 is what everybody wants. It is super expensive to own, very fragile board and is the pinnacle of 2D fighter and most ports were not arcade perfect. So, yes, it is a single game, but it is worth more than 50 cores of arcade games from the 70s that only boomers play.

madmax wrote: ↑Fri Jun 23, 2023 2:51 pm

AmanoJacu wrote: ↑Wed Apr 05, 2023 9:38 am

dcubed wrote: ↑Wed Apr 05, 2023 12:27 am

The opportunity cost of CPS3 is just massive. It's 15 months of tracing work for the sake of one game (Let's be honest, Third Strike is the only CPS3 game that people really want). And that's all work that isn't even really transferrable to any other core either.

PGM, or even the Psikyo SH2 is a far more sensible platform for him to focus his efforts on. Even the Konami boards with their various, bitty slapdash selection of chips is a better use of his time than CPS3... especially considering that Third Strike is already available on many other platforms.

CPS3 should be one of the last platforms that he works on for MiSTer. The people who voted for it really have no idea what cost CPS3 really comes with.

It is possible that single CPS3 game, would accumulate more hours of play than all the other games you mentioned combined. At least in my case, it sure would. Let's be honest, most of the new arcade games that come for mister, I test them for around 5 minutes and never play them again. Some others, I would play once to the end. But a fighting game that is considered the pinnacle of the old generation of 2D fighters, which I already put thousands of hours in emulators and ports, that is still played in competitive tournaments today, would be much more than that. The "available on many other platforms" can be said about each and every game the mister can play.

I'm a Jotego patreon, and I voted for CPS3. The question was about what system I would like to see first, and that's what I wanted. If we have to factor in their amount of work (an unknown variable at that time), and CPS3 is not feasible, then yes, I have no problem in him spending his resources in something else I will also enjoy.

SF3.3 is what everybody wants. It is super expensive to own, very fragile board and is the pinnacle of 2D fighter and most ports were not arcade perfect. So, yes, it is a single game, but it is worth more than 50 cores of arcade games from the 70s that only boomers play.

in your opinion. I would prefer 50+ 70s arcade cores to one generic fighter.

I don’t like fighter game … i prefer 50+ 70s arcade cores too

Third Strike is an amazing fighter & cannot wait to see if jotego can pull it off. That said, I'd also love to see some bronze era stuff come across - Atari Basketball, Night Driver, Stunt Cycle, Ozma Wars and any of those other early Taito/Atari/Kee/Exidy/etc stuff that doesn't already have a core. Either way I can't program these things, so will happily have anything/all we get.

madmax wrote: ↑Fri Jun 23, 2023 2:51 pm

SF3.3 is what everybody wants. It is super expensive to own, very fragile board and is the pinnacle of 2D fighter and most ports were not arcade perfect. So, yes, it is a single game, but it is worth more than 50 cores of arcade games from the 70s that only boomers play.

quite curious how many boomers are playing these games, I'd have said it was more Gen X'ers

My cab has 3 buttons per player so I’m not worried about fighters at all.

bazza_12 wrote: ↑Sat Jun 24, 2023 7:42 am

madmax wrote: ↑Fri Jun 23, 2023 2:51 pm

SF3.3 is what everybody wants. It is super expensive to own, very fragile board and is the pinnacle of 2D fighter and most ports were not arcade perfect. So, yes, it is a single game, but it is worth more than 50 cores of arcade games from the 70s that only boomers play.

quite curious how many boomers are playing these games, I'd have said it was more Gen X'ers

I thought that was a strange comment as well.

Boomers may have created these games, and the general public did bemusedly try out Space Invaders, Pac-Man, and Donkey Kong back in '79-'81 when they were a novelty, but retro-gaming in this day and age is pretty much 100% a Gen-X (or younger) thing.

i agree the demographic for retro video games is ₃₅ or younger, which would not technically be a boomer

but i wanna say boomer is used more as an older crowd/dad vibes context. thats been how ive heard it used

and its already been stated by Jotego, CPS3 is possible but is years away from being released for the MiSTer. unless Jotego can find a way to automate some of the chip tracing, or assume the MAME implementation is somewhat accurate, we will be waiting a long long time

Errrmmmmmmmmmm...strongly, strongly disagree with that age range. Folks that age literally weren't alive for the golden age of video games.

More like 55 and younger.

LamerDeluxe wrote: ↑Mon Jun 26, 2023 11:19 am

More like 55 and younger.

agreed

The Silent Generation: Born 1928–1945.
Baby Boom Generation: Born 1946–1964.
Generation X: Born 1965–1980.
Millennial Generation or Generation Y: Born 1981–1996.
Generation Z or iGen: Born 1997–2010.
Generation Alpha: Born After 2000.

So, according to wiki (I know, I know) 'The first arcade game, Computer Space, was created by Nolan Bushnell and Ted Dabney, the founders of Atari, Inc., and released in 1971; the company followed on its success the next year with Pong.' This puts Bushnell (b.1943) and Dabney (b.1937) both born in the silent generation.. even the creator of Space Invaders (released 1978), Tomohiro Nishikado (b. 1944) belongs to that generation. I'd suggest arcade gaming didn't achieve international success (by that I mean, you were unlikely to see mass produced arcade cabinets all over the world) until the late 70's. Gen X'ers (born in 1965) where approaching their teenage years at the same time Space Invaders was released.

maybe I should write a uni dissertation on this.. or get chatgtp to do it for me..

Any progress on this?

Someone asked last month in the jotego discord and the response was that there's no news.

Jotego and his team are busy with audio balancing across cores, Sega System 18 and Neo Geo Pocket development, and probably other things I'm forgetting.

while i respect Jotego and hold him in great regard, i just dont see this core happening on the MiSTer.
he has tackled some iconic arcades in the past and brought some very cool stuff to the MiSTer, but the CPS3 is something a large group of developers would need to dedicate all of their efforts to. with the current workload Jotego has, i just dont see CPS3 happening on the MiSTer. i wont say ever, but if i had to put a number on it, id say end of 2025 is pushing for a Christmas miracle. most likely Q2 of 2027.

i think the soonest well see CPS3 in FPGA is if the other FPGA team of developers doing a CPS3 core for their FPGA device convert the MAME version into an FPGA core. while that isnt ideal, i think it may be the option that is both the path of least resistance and the one that will ultimately be the mainline core people will use. porting that MAME FPGA core to the MiSTer will be a whole extra endeavor (that may or may not be possible), and even that MAME version of CPS3 in FPGA is a long way off from being finished.

in the meantime, we have groovy_mister, and i think that is as close as we will get. using a properly configured groovy_mister setup, we can expect to see around 4ms of input lag. i think that is perfectly acceptable and is the version of CPS3 that most people who use fightcade are familiar with. even if the other team get their version of MAME CPS3 ported, the overall experience will be virtually identical (just without the need for a computer to feed in the game over ethernet). Plus you will immediately have all the benefits of groovy_mister and several other cores that arent in FPGA format.

my comments arent meant to dismiss anyone or the work they do, FPGA is one of the greatest things to happen to retro gaming, but were really pushing the limits of whats possible, and the CPS3 is right there at the top. not impossible, but will take a lot of work to get over the finish line

PikWik wrote: ↑Thu Jun 06, 2024 3:23 pm

[...] FPGA is one of the greatest things to happen to retro gaming, but were really pushing the limits of whats possible, and the CPS3 is right there at the top. not impossible, but will take a lot of work to get over the finish line

Yes, more or less correct, but good things need good times. And he (they) are perfectionists. Our money at work. We get get the best what we have paid for...see Robert and his efforts on AO486, PSX and N64.
That said: Are you (the reader) already a Patreon of Jotego? If not, go for him!