After reading last week's reaction, let me give a detailed update on the CPS3 core status:
The architecture design work is finished. The core will work on MiSTer and SiDi128. It cannot work on Analogue Pocket because the SDRAM is too small for the task. CPS3 games are just too large
Statistical data collected on the games have guided an architecture that should fit well on a standard MiSTer device and most likely on a SiDi128
I am only targeting No-CD games in order to simplify the implementation.
The first implementation is only targeting the games that fit in a 64 MB SDRAM, which are Street Fighter III: New Generation and Red Earth. Extending the core to use 128MB SDRAM and the rest of the games will come after that. This decision is to stage the changes to the JT cores framework in two smaller steps, rather than a big one
The PPU (Picture Processing Unit) is in the works at the moment. On this occasion, I am doing the PPU first and I will have images working before the CPU is connected. I have followed this approach in the past too when new CPUs were used, like in this case.
We will have updates on the PPU implementation with actual images on the next two Fridays. The status as of today is focused on SDRAM and memory handling and it is not doing any actual visible rendering.
If there are no surprises -and you never know with these systems- the core should be released somewhere between the 24th of April and the 8th of May. We are going to stop the work on the NGP until this core gets released
What's Next
Lately we have been very busy with work that doesn't seem to be as appreciated, judging by subscription numbers. The thing is that most arcade cores take a similar amount of effort and games that look simple may hide development pitfalls that take time. Plus, supporting such a large number of games as JTCORES do implies having to reply to a lot of queries and issues from users. A year ago, I thought we could be three developers in the team, but subscription came down instead of up, and we are still two with more work than ever. The team also has a specialist on reverse engineering of PCBs, which is important for preservation and progressing the community knowledge but he doesn't develop new cores or fixes issues per se.
So I wan't to make sure that after CPS3 we are more aligned with patrons. Please vote from the following list. After the initial CPS3 release, we will take on the most voted one next.