PC Engine core

[ArinoCX]

After implementing these features in my CPU/VDC this test passes successfully.

Wow, that's great news, issue a pull request !
I'll bet that solves a lot of outstanding issues.

Actually, I just got a Saleae 16 Pro logic analyzer and spare PC Engine to run some tests for comparison between the core and real hardware. I'm still working on setting it up though.

I'll be trying to write some test cases this weekend.

I dont know what you guys are talking about but it sounds important

Thank you for your work!

[srg320]

Wow, that's great news, issue a pull request !

Well, this is a bit different code.

I posted the source here if you are interested.

[dshadoff]

Wow, this is a big change.
I see that our paths diverged around February 16.

How would you like to proceed ? It might be simpler for me to reapply my changes to your updated source code, compared to the other way around.
Is it currently stable, or are you still working on something ?


EDIT:
I grabbed a copy and used the cpu-vdc branch to compile; I needed to rename the main directory's Huc6280.vhd to .bak (since it was duplicated), and I got it to compile, although with a "timing not met" warning.

Wow, you've really got the video section nearly perfect (only a couple small things left). A few new bugs showed up impacting gameplay though (but you must already know that)... but I congratulate you on its current state, you've made a huge advance !

[srg320]

Wow, this is a big change.
I see that our paths diverged around February 16.

How would you like to proceed ? It might be simpler for me to reapply my changes to your updated source code, compared to the other way around.
Is it currently stable, or are you still working on something ?


EDIT:
I grabbed a copy and used the cpu-vdc branch to compile; I needed to rename the main directory's Huc6280.vhd to .bak (since it was duplicated), and I got it to compile, although with a "timing not met" warning.

Wow, you've really got the video section nearly perfect (only a couple small things left). A few new bugs showed up impacting gameplay though (but you must already know that)... but I congratulate you on its current state, you've made a huge advance !

The CPU is now stable, but I have been working on the VDC for less than two weeks, so there’s still a lot of research to do in real hardware (and DMA on this list too). I'm trying to make test roms, but it's not easy.

[breiztiger]

@srg320

i just compil your build and street fighter ii' have missing 2nd player in demo mode

edit : just to report

[srg320]

@srg320

i just compil your build and street fighter ii' have missing 2nd player in demo mode

It's for developers only. DO NOT USE it. Use official release.

[trashuncle]

@srg320

i just compil your build and street fighter ii' have missing 2nd player in demo mode

It's for developers only. DO NOT USE it. Use official release.

I suggest to all to use 19.09.28 build until all current changes are made, as this is last known good version before reprogramming for closer to real hardware accuracy. (ie: Less Glitches... I've played this one for hours upon hours and it works like a charm)

https://github.com/MiSTer-devel/TurboGr ... 190928.rbf

[Newsdee]

This progress is really exciting. Keep up the good work!

[ElDoctor]

@srg320

i just compil your build and street fighter ii' have missing 2nd player in demo mode

It's for developers only. DO NOT USE it. Use official release.

Incredible how people don't understand what Work In Progress means (dshadoff already mentioned weeks ago things will be shaky for a while), and are wasting everyone's time by reporting issues on something that we all know is UNDER DEVELOPMENT and UNSTABLE.

You want to play street fighter ii' without any bugs? Then let the devs focus on fixing bugs instead of replying to your pointless report...

[trashuncle]

Why be negative? This is why I posted something constructive instead (Link to Known-Good Core.) The pointless bug posts have been addressed a couple times now. Forums are to be helpful, not bash everyone who contributes feedback.

[dshadoff]

The CPU is now stable, but I have been working on the VDC for less than two weeks, so there’s still a lot of research to do in real hardware (and DMA on this list too). I'm trying to make test roms, but it's not easy.

How can I help ? You've already handled several of the situations which were concerning me.

[ElDoctor]

Why be negative? This is why I posted something constructive instead (Link to Known-Good Core.) The pointless bug posts have been addressed a couple times now. Forums are to be helpful, not bash everyone who contributes feedback.

Check my other humble posts, I am the first one to praise the amazing and revolutionary work done by the incredible devs.
Was that positive enough for your taste?

Joke aside; really excited to see what dshadoff, srg320 and paulnb are working on now.
On top of Furrtek decapping everything, especially the HuC6280. Great times ahead for accurate good stuff.

[dshadoff]

By the way, SNAC - as the boards are currently designed - will not work 100% properly with PC Engine controllers. A different design will be required.

The reason for this is that the rise time on positive pulses is quite long (>1uS), when 1uS is roughly the duration of the RESET pulse during joypad read, and the PC Engine reads values back pretty quickly in both the positive and negative positions for SEL.

There is additionally a ringing in the upward slope of the outbound signals, depending on how many switches are closed on the joypad. This is particularly notable for 6-button joypads, which identify themselves (and the joypad read phase) by bringing low multiple lines one one phase which aren't really possible on the other (i.e. same lines as up + down).

With the goal of improving the slope of the upward swing (and shifting the ringing earlier in the read cycle), I assembled a version of the SNAC with 4.7K ohm resistors instead of 10K ohm, and there is certainly some visible improvement - but it still doesn't yet work properly. I will try a couple more values for these resistors, but I don't really want to go much below 4K.

[breiztiger]

@trashuncle

sorry for publishing an unnecessary report, I wouldn't bother you anymore

[trashuncle]

@trashuncle

sorry for publishing an unnecessary report, I wouldn't bother you anymore

Didn't bother me, its the negativity that I was commenting on. Anyone taking time out to test things is good for the community. It does also help to review the forum you are posting in though, to make sure that the feedback is needed at the present time. This particular core has a lot of active development going on. Once the next public release goes out, more testing will be required/expected. Compiling an early build is really only useful to test if you can comment on the technical specifics at this phase. I personally do not have the tools or the hardware background to help at all at this level. I am just an avid Turbo fan (and MiSTer obsessed.)

[BitsNStuff]

It's so exciting to see so much work being done on this core now, I know it doesn't have the largest library of games but it was always very much a dream sustem for me when I was a kid. Thanks, really appreciated.

[dshadoff]

I regard all of the current work as just groundwork... once the base system is stable, CDROM has a lot more (and better) games.

[BitsNStuff]

Absolutely. I think of the systems on MiSTer that also have CD add-ons, the PC Engine is probably the best of them.

[Televicious]

Getting an out of range message on my dell 1908fp when i load the 2-20 version of the core. tried all sorts of configs. The 2-12 version works fine however.

A lot of code changed in that release, but it was mostly just interrupt timing (i.e. what pixel to display, rather than the structure of the display). The only thing I see about the structure of the display is the removal of one excess clock per scanline (commit e2d8703).

Is that a type of monitor ?
Yes, a dell 1908fp lcd monitor.

Does it indicate what type of range is being violated ?
No, just that it is not syncing to 1280x1024 60hz, a box floats around the screen displaying that it is out of range.

Do you have more information on what you were trying to do at the time (i.e. at menu ? not at menu, but game not yet selected ? if playing a game, which one ?) ? Resolution mode is important information.
I set the whole mister to 1280x1024 60hz or specifically video mode 4 , that is the native resolution for this monitor. The 2-20 release is the only build that doesn't sync. Have gotten all other cores working great. Have reverted to the 2-12 and been using that perfectly well. Wish I had more info to help.

[trashuncle]

I regard all of the current work as just groundwork... once the base system is stable, CDROM has a lot more (and better) games.

When we get there, I will be very excited! There are still so many PC CDROM games I have never experienced.

[SegaSnatcher]

Am I right to assume Srg320 has implemented his own TG-16 CPU and VDP from scratch?

[dshadoff]

Am I right to assume Srg320 has implemented his own TG-16 CPU and VDP from scratch?

From what I can see, he has made dramatic changes to the VDC (but started from the existing one).
And he may have replaced the CPU (or dramatically rewrote it). I wasn't aware that it had big enough problems to make such invasive changes.
Much of his code is more readable, but comments were removed.

So, I agree with half of your statement - "implemented his own", but I wouldn't say "from scratch"... at least not exactly.

In any case, the changes are large enough that I am wary of trying to integrate it, and would rather spend my time to work with him to fix the remaining issues so it can be integrated.

[SegaSnatcher]

Am I right to assume Srg320 has implemented his own TG-16 CPU and VDP from scratch?

From what I can see, he has made dramatic changes to the VDC (but started from the existing one).
And he may have replaced the CPU (or dramatically rewrote it). I wasn't aware that it had big enough problems to make such invasive changes.
Much of his code is more readable, but comments were removed.

So, I agree with half of your statement - "implemented his own", but I wouldn't say "from scratch"... at least not exactly.

In any case, the changes are large enough that I am wary of trying to integrate it, and would rather spend my time to work with him to fix the remaining issues so it can be integrated.

Thanks for the info, the only major bug I still encounter is with Street Fighter 2, especially Honda's stage. Lots of random glitches the effect the whole frame. It happens for a split second but always been noticebale.

[trashuncle]

Am I right to assume Srg320 has implemented his own TG-16 CPU and VDP from scratch?

From what I can see, he has made dramatic changes to the VDC (but started from the existing one).
And he may have replaced the CPU (or dramatically rewrote it). I wasn't aware that it had big enough problems to make such invasive changes.
Much of his code is more readable, but comments were removed.

So, I agree with half of your statement - "implemented his own", but I wouldn't say "from scratch"... at least not exactly.

In any case, the changes are large enough that I am wary of trying to integrate it, and would rather spend my time to work with him to fix the remaining issues so it can be integrated.

Thanks for the info, the only major bug I still encounter is with Street Fighter 2, especially Honda's stage. Lots of random glitches the effect the whole frame. It happens for a split second but always been noticebale.

Those bugs are present before the core rewrite (build from 19.09.28). I just did some testing. So you might find a real TG-16/ PC Engine and see if it happens on those.

EDIT: And guess what, I went and tested on real hardware, and those glitches are present. I get the random glitches on various stages.

[SegaSnatcher]

From what I can see, he has made dramatic changes to the VDC (but started from the existing one).
And he may have replaced the CPU (or dramatically rewrote it). I wasn't aware that it had big enough problems to make such invasive changes.
Much of his code is more readable, but comments were removed.

So, I agree with half of your statement - "implemented his own", but I wouldn't say "from scratch"... at least not exactly.

In any case, the changes are large enough that I am wary of trying to integrate it, and would rather spend my time to work with him to fix the remaining issues so it can be integrated.

Thanks for the info, the only major bug I still encounter is with Street Fighter 2, especially Honda's stage. Lots of random glitches the effect the whole frame. It happens for a split second but always been noticebale.

Those bugs are present before the core rewrite (build from 19.09.28). I just did some testing. So you might find a real TG-16/ PC Engine and see if it happens on those.

EDIT: And guess what, I went and tested on real hardware, and those glitches are present. I get the random glitches on various stages.

The screen doesn't fully glitch out on real hardware via Honda's stage like on the PCE core. The PCE core has a lot more glitches in Street Fighter 2 than real hardware. I don't know what you are talking about to be honest, but your experience is not what I'm seeing at all.