Author Topic: CX4 Timing  (Read 3382 times)

0 Members and 1 Guest are viewing this topic.

Offline James-F

  • Sr. Member
  • ****
  • Posts: 340
  • Karma: +29/-0
    • View Profile
CX4 Timing
« on: April 14, 2018, 01:13 PM »
Apparently the timing of the CX4 implementation changes if "in-game hookk" is enabled or disabled.

This can be easily seen on Megaman X2 demo where X almost wins the big robot boss on real cartridge and when "in-game hook" is enabled with CX4 speed=normal.
When "in-game hook" is disabled megaman dies in the battle against the big robot in the demo sequence, so CX4 is more accurate with in-game hook ON speed=normal, which is counter intuitive.

In short:
hooks:on, cx4 speed:normal  = x almost wins (like cartridge Jap/US).
hooks:on, cx4 speed:fast = x dies.
hooks:off, cx4 speed:normal = x dies.
hooks:off, cx4 speed:fast = x dies.

How it should end:
https://streamable.com/dced3

Tested on 2chip, 1chip, Super NT.


PS.
On HiGan and Snes9x, megaman always dies.
European Megaman X2 and PAL SNES, megaman always dies (all sd2snes configurations).
« Last Edit: April 14, 2018, 01:33 PM by James-F »
Mega Everdrive x5, Everdrive 64 v3, Everdrive N8, N8 Pro, SD2SNES, Joyzz.

Offline polyh3dron

  • Newbie
  • *
  • Posts: 38
  • Karma: +1/-0
    • View Profile
Re: CX4 Timing
« Reply #1 on: April 14, 2018, 02:40 PM »
Can confirm all of this.

Offline nesfreak

  • Newbie
  • *
  • Posts: 9
  • Karma: +0/-0
    • View Profile
Re: CX4 Timing
« Reply #2 on: April 15, 2018, 01:35 AM »
Just tested on Super NT with sd2snes, only tested with hooks:on, same result as James-F.

hooks:on, cx4 speed:normal  = x almost wins
hooks:on, cx4 speed:fast = x dies.

Offline James-F

  • Sr. Member
  • ****
  • Posts: 340
  • Karma: +29/-0
    • View Profile
Re: CX4 Timing
« Reply #3 on: April 15, 2018, 03:08 PM »
I think the base line should be Hooks:Off, Speed:Normal, then calibrate CX4 speed to match the real cart, which we know it can do already.
Hopefully ikari sees this post, and acknowledges.
Mega Everdrive x5, Everdrive 64 v3, Everdrive N8, N8 Pro, SD2SNES, Joyzz.

Offline OneBagTravel

  • Sony KV-20FV300
  • Full Member
  • ***
  • Posts: 153
  • Karma: +17/-0
  • Has the Perrier gone straight to my head?
    • View Profile
    • OneBagTravel.com
Re: CX4 Timing
« Reply #4 on: April 15, 2018, 05:50 PM »
Here's an old thread where he talks about it and some others went into more detailed testing

http://krikzz.com/forum/index.php?topic=5486.msg42254#msg42254
Travel light & smart. SFC using HD Retrovision component cables.

Offline Great Hierophant

  • Sr. Member
  • ****
  • Posts: 292
  • Karma: +19/-0
    • View Profile
Re: CX4 Timing
« Reply #5 on: April 18, 2018, 06:30 PM »
Demonstration of in-hook on and in-hook off behavior :

https://www.youtube.com/watch?v=2aN9Qu5u6Hc

I would be interested to know if X dies with a real PAL X2 cart.  It could be that the speed differences in the console cause death, although I think it unlikely.
« Last Edit: April 18, 2018, 06:35 PM by Great Hierophant »
Check Out My Vintage Video Game & Computing Blog : http://nerdlypleasures.blogspot.com/

Offline animeloverxX93

  • Newbie
  • *
  • Posts: 16
  • Karma: +2/-0
    • View Profile
Re: CX4 Timing
« Reply #6 on: April 20, 2018, 05:23 PM »
As someone already mentioned before, ikari is well aware of that.
I remember when I commented on his blog about it some time ago.
Finger's crossed we get to see his new CX4 timing implemented in the next release; honestly, I don't really mind waiting for this to happen.

What I do ponder about is the limited space in the FPGA due to multiple implementations of various features and how the SD2SNES could still haave feature-complete SNES support...