Extermination Schematics & Pang! for the Pocket
I had planned to have Out Run for today but those super scalers sprites are asking for a bit more love. We’ll go on a honeymoon over the weekend.
Work didn’t stop in other fronts. We spent two days improving the open source tool for simulation coverage that we are using for the NeoGeo Pocket. It isn’t our plan to become contributors to the open source EDA tool scene; but sometimes you need to improve atool to get the work done.
Esperanza finished Taito’s Extermination schematics (ERC clean, for those electronics engineers out there). Find them attached. She has moved to the CPS3 schematics after the poll results. We are not sure how much those schematics will tell us. There are so many large custom chips that just with the schematics will not be enough at all. But it is the first step. CPS3 will be a long development as it will have to move through several research stages. Unless you want us to just port bindly whatever MAME does. In that case we can have it done in a breeze.
About the Pocket, most users seem ok with sending the chip ID of a device that isn’t connected to internet. The chip ID is more convenient than having a key, like MiSTer users do. Because you don’t need to worry about updating the key or copying it over. A lot of users want to have the cores eventually released without DRM. That’s no problem. Early access in MiSTer cores have normally been 1-2 months. We can do the same if Pocket patrons want.
Find attached the arcade FPGA core JTPANG (compatible with the arcade Pang!) for the A. Pocket. This one is an open compilation, like JTGNG. The MRA files required are already in the folder where you need to place the .rom files, to make things easier. You can also use this core to check your device ID when you press the Analogue button.
Wish me good luck with that red Ferrari on our weekend honeymoon.
Español
Había planeado tener Out Run para hoy, pero esos sprites de súper escaladores están pidiendo un poco más de amor. Iremos de luna de miel el fin de semana.
El trabajo no se detuvo en otros frentes. Pasamos dos días mejorando la herramienta de código abierto para la cobertura de simulación que estamos usando para NeoGeo Pocket. No es nuestro plan convertirnos en contribuyentes a la escena de herramientas EDA de código abierto; pero a veces es necesario mejorar una herramienta para realizar el trabajo.
Esperanza terminó los esquemas de exterminio de Taito (ERC limpio, para esos ingenieros electrónicos). Encuéntrelos adjuntos. Se ha mudado a los esquemas de CPS3 después de los resultados de la encuesta. No estamos seguros de cuánto nos dirán esos esquemas. Hay tantos chips personalizados grandes que solo con los esquemas no será suficiente. Pero es el primer paso. CPS3 será un desarrollo largo, ya que tendrá que pasar por varias etapas de investigación. A menos que quieras que simplemente portemos bindly lo que sea que haga MAME. En ese caso, podemos tenerlo hecho en un abrir y cerrar de ojos.
Acerca de Pocket, la mayoría de los usuarios parecen estar de acuerdo con enviar la identificación del chip de un dispositivo que no está conectado a Internet. La identificación del chip es más conveniente que tener una clave, como lo hacen los usuarios de MiSTer. Porque no necesita preocuparse por actualizar la clave o copiarla. Muchos usuarios quieren que los núcleos finalmente se lancen sin DRM. Eso no es un problema. El acceso temprano en los núcleos de MiSTer normalmente ha sido de 1 a 2 meses. Podemos hacer lo mismo si los usuarios de Pocket lo desean.
Encuentre adjunto el núcleo arcade FPGA JTPANG (compatible con Pang!) para A. Pocket. Esta es una compilación abierta, como JTGNG. Los archivos MRA necesarios ya están en la carpeta donde debe colocar los archivos .rom, para facilitar las cosas. También puede usar este núcleo para verificar la identificación de su dispositivo cuando presiona el botón Analógico.
Deséame buena suerte con ese Ferrari rojo en nuestra luna de miel de fin de semana.
Comments (27)
Now that Extermination schematic is ready, how long before work on the core commences? Thank you!
Hi, just for curiosity, how may i save the setting inside the service menu in arcade core pang? They have default values every time i load the core on my mister
After using the service menu, you also need to select save settings in the MiSTer OSD to trigger the saving to the SD card.
One thing to be aware of when generating schematics for CPS3 is that the first revision of the hardware included some unused battery-backed SRAM inside the white plastic module on the main PCB. The theory is that Capcom could load game patches into the SRAM as part of location testing. The SRAM and battery was never used for the final released games, and later production runs of the PCB omitted it from the hardware. On the boards that have it, the battery can leak and/or explode and damage a substantial section of the PCB. Whether you include the SRAM and battery in the schematics depends on if you want the schematics to be as thorough as possible or only include the hardware that is actually used for the games.
I vote to go for more accuracy for CPS3. That way the FPGA implementation is a substitute for the original hardware which is expensive and difficult to maintain. Great work as always from your team! Really looking forward to Outrun as well.
I would appreciate it if someone or Jotego could give us a comparison of how accurate mame is for cps 3 vs original hardware. If it’s already in the 90% then perhaps a port is fine. I know Jotego is very obsessive about perfection though. Perhaps a mame port while we wait for the more accurate version is a good compromise ^_^
MAME isn’t inaccurate because those guys don’t know what they’re doing, it’s inaccurate because software emulation works fundamentally differently to the way hardware works. The MAME logic should still be functionally equivalent. It’s just better from a hardware preservation perspective to study and implement each custom chip as its own module and then ‘wiring them up’ in the FPGA (similar to how it is on the real PCB). If you only care about running the title, it basically achieves the same end result. Of course, by having someone else study the chips, something could be discovered which wasn’t implemented correctly in MAME, but generally, the MAME logic should be correct.
Hi Jose, I'm very happy and excited for OutRun, will the title be available next Friday or sooner?
It will be the next release. Hopefully next Friday. It's really close to completion now.
I tend to think that it's worth the time doing a full analysis of CPS3, even if it slows down the rate at which playable cores are produced. MAME is an excellent resource and often has highly accurate or perhaps in some cases perfect documentation, but the rigour of re-analysing it helps identify minor errors etc. as we have seen before. Some of these minor errors might not impact gameplay but they improve the accuracy of the documentation which can then feed back into MAME also, and the skill and tenacity of team Jotego in this reagrd is why I support you on patreon. I tend to see my support broadly as supporting you in using your skills and the mister platform to analyse, document and reimplement these old systems in parallel to things like MAME, giving us even more certainty in the accuracy of the documentation (like whether a starfield is rendering incorrectly). The fact that we get to play the games on MiSTer (or I guess also Pocket now too) is a bonus. I guess the TL;DR is I don't see that the playable game is a product I'm purchasing from you, so I don't need something which is playable to be delivered quickly at the expense of the underlying mission to analyse, preserve and reimplement video gaming history.
I agree with this sentiment. I view my support as contributing to preservation which also produces amazingly accurate documentation and cores that can then be run on as many platforms as needed. I really like the trajectory this project is on in that regard.
I absolutely love Street Fighter III Third Strike and did vote for the CPS3 core, but at the same time my thought is that CPS3 only has 6 games (3 if you don't count revisions, then it's basically Street Fighter III, JoJo and Red Earth). Looking at it pragmatically, it makes sense to have the core blindly following what MAME does as I don't think it's worth spending a huge amount of time doing research on a core that doesn't offer a lot of titles. The counter argument to that is that CPS3 boards are extremely expensive and hard to find these days, so from a hardware preservation standpoint it would be good to do it properly, I just don't believe it's worth the effort. A lot of people forget that MAME exists primarily to document arcade hardware, the software emulator just happens to be a side-effect of that work. I'd be comfortable trusting their work and having a hardware implementation that basically does what MAME does.
That's right. MAME's goal is to document. Even if we don't get to know CPS3 to the last bit, I'd like to be able to say that we pushed accuracy one level up as part of making the core.
> Wish me good luck with that red Ferrari on our weekend honeymoon. I have no doubt about you. Fantastic work as always. I have my ignition keys ready :)
Yeah, I think it would be very nice to do public releases for Pocket 1 - 2 months after Patron release like you do for MiSTer.
There will always be two types of supporters: ones that have their noses pressed against the glass hoping for faster results and those who trust the development process. Hopefully this explanation helps convert more of the former into the latter. Always good to learn more about development in an attempt to understand the challenges, and I'd be among those who'd appreciate similar release process on the Pocket as what's currently available on MiSTer. For support, I'll bump up my pledge going forward to accommodate both platforms. -Thanks
Please take your time with cps3. It is much appreciated!
Enjoy your honeymoon
Been waiting for my favourite game OutRun since 1986! I can only hang on a few more days! lol Thanx Jotego for making this happen & for all your hard work !!! :)
Thanks for your work! Is there any chance you can change the category from "Arcade compatible" to just "Arcade" to match other cores going forward?
It’s a legal matter. It’s the same reason why Analogue calls the ROM folder “assets”
Thanks for Pang and all your hard work so far, I can't wait for Out Run! Also want to voice my support for releasing Pocket cores publicly after a month (or two)
Nice! Great job everyone and thank you for the hard work!
Thanks Jose! Have a great weekend!!
Good luck on that red Ferrari on honeymoon weekend! A release like Out Run deserves the extra time. Thank you for your hard work!
I definitely support public releases for the Pocket not tied to a Chip ID.
Same here!