Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Thanks for the continued testing!
Lady Bug reminds me that the STIC does output a border that is part of the screen, I'll have to make sure to add that.
Some of the music sounds like it pops a bit, maybe I can look into that too.
For pacman, I posted in AtariAge so maybe I can get a response there, but what do you mean by z-order? All the collision detection seems to be working. PacMan and the ghosts do collide in the sense that their respective bits in the collision registers are set, but nothing happens. I must be missing some detail somewhere.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
There have been lots of changes to lots of cores recently. First thing I would try is using the same BizHawk version on both computers just to limit variables. If that still doesn't work then maybe try copying your config.ini file from the first computer to the second, could be something in there that is messing things up. If that still doesn't work, might be a version difference. What versions are you using?
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Alright! I added in the last bits of code and cleaned up some stuff and now TAS tools are available in Intellivision!
You should be able to play and TAS any game you want by now without bugs (except for Pac Man where I don't have any idea what is wrong yet.) In my testing so far everything seems to work without desyncs.
Note that for TASing purposes you'll want to pick the standard controller from the Intellivision menu.
There is still a long way to go to clean up the core and get things in proper shape and improve accuracy on edge cases, but as a whole its basically a working intellivision emulator. Have fun!
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Thanks for the testing!
Some games require mappers that aren't implemented yet. I just implemented Pac Man now as the first test of mapper support and it worked fine, so in a couple more days I should have all the mappers in there.
Controls and video dump will need adelikat to fix them.
Intellivoice is not implemented currently, and I don't know if it will be.
EDIT: ok, I fixed Pole Position and added many games to the gamedb. At this point all games should be playable.
I also noticed that pacman passes right through ghosts, so that's a bit of a problem. :)
I'll try to sort that out soon.
With mapping done I now have access to the test cartridge too which should help.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
I looked at some trace logs to see where differences came from that makes this run de-sync in most recent versions of bizhawk.
AS far as I can tell, it comes from right here:
817D: ED LD A, R AF:0040 BC:1400 DE:50E3 HL:733C IX:C300 IY:DD03 SP:DFE4 Cy:204661845 cnp-h-Zs
The R register is counter that increments on instruction calls, and and the timing fixes to the z80 means it has a different value at this point (2704 instead of 2C04 in the previous versions.)
This seems to change the boss pattern.
So the desync is unrelated to interrupts or SRAM, it appears purely cpu timing related.
I don't know how/if this effects judging, the current cycle timings are indeed more accurate in the current version, I just thought I'd point this out so its documented.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Challenger wrote:
My Bizhawk 1.11.9 desncys this TAS during the Boss 1 battle.
Anybody up for encode?
Some changes have occured to SMS inbetween 1.11.4 and now.
Quite a few timing fixes to the Z80 core happened.
I also fixed SRAM and interrupt handling, which was pretty far off before and worked largely due to luck in how games used interrupts conservatively.
I don't know which of these is effecting sync on this game, maybe I can look into it tomorrow.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Alright, so I got sound working and fixed numerous bugs, so the core is in a pretty good state now.
Any 8kb rom should work and play/sound correctly.
Next steps are to get TAS tools working and do banking to get the rest of the games working.
Any play testing is welcome at this point to dig up bugs.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Aww that's a shame, I was looking forward to seeing what craziness was going to arise from DKC. Maybe with a bigger time horizon it could be possible in the future? Or is it just too impractical?
Good luck with everything else though!
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
nrg_zam wrote:
I'm getting an immediate unhandled exception when opening TAStudio on the latest interim build, but not on 1.11.9.1. TAStudio still works, but without markers or a scrollbar on the piano roll.
System.NullReferenceException: Object reference not set to an instance of an object.
at BizHawk.Client.EmuHawk.MarkerControl.MarkerView_QueryItemBkColor(Int32 index, RollColumn column, Color& color)
at BizHawk.Client.EmuHawk.InputRoll.DoBackGroundCallback(PaintEventArgs e, List`1 visibleColumns)
at BizHawk.Client.EmuHawk.InputRoll.DrawBg(PaintEventArgs e, List`1 visibleColumns)
at BizHawk.Client.EmuHawk.InputRoll.OnPaint(PaintEventArgs e)
at System.Windows.Forms.Control.PaintWithErrorHandling(PaintEventArgs e, Int16 layer)
at System.Windows.Forms.Control.WmPaint(Message& m)
at System.Windows.Forms.Control.WndProc(Message& m)
at System.Windows.Forms.Control.ControlNativeWindow.OnMessage(Message& m)
at System.Windows.Forms.Control.ControlNativeWindow.WndProc(Message& m)
at System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)
Hi,
So I had this same bug when I tried to use TAStudio.
I'm not sure how this could have worked for feos since it looks like the code is taking an instance of Emulator before one exists to call Frame from?
I put in a check to make sure it exists and defaults to zero otherwise and now markers are back agian.
I must not be understanding something about how everything works together since this seemed like a pretty serious exception that everyone would have gotten, but hopefully it works now for you too, so maybe give it a try.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
I'm getting this exception in the most recent Dev Builds whenveer I open the emulator or load a game (any core):
System.InvalidOperationException: Sync mode is required
at BizHawk.Client.EmuHawk.SoundOutputProvider.set_BaseSoundProvider(ISoundProvider value)
at BizHawk.Client.EmuHawk.Sound.StartSound()
at BizHawk.Client.EmuHawk.WinFormExtensions.FormExtensions.ShowHawkDialog(CommonDialog form)
at BizHawk.Client.EmuHawk.MainForm.OpenRom()
at BizHawk.Client.EmuHawk.MainForm.OpenRomMenuItem_Click(Object sender, EventArgs e)
at System.Windows.Forms.ToolStripItem.RaiseEvent(Object key, EventArgs e)
at System.Windows.Forms.ToolStripMenuItem.OnClick(EventArgs e)
at System.Windows.Forms.ToolStripItem.HandleClick(EventArgs e)
at System.Windows.Forms.ToolStripItem.HandleMouseUp(MouseEventArgs e)
at System.Windows.Forms.ToolStripItem.FireEventInteractive(EventArgs e, ToolStripItemEventType met)
at System.Windows.Forms.ToolStripItem.FireEvent(EventArgs e, ToolStripItemEventType met)
at System.Windows.Forms.ToolStrip.OnMouseUp(MouseEventArgs mea)
at System.Windows.Forms.ToolStripDropDown.OnMouseUp(MouseEventArgs mea)
at System.Windows.Forms.Control.WmMouseUp(Message& m, MouseButtons button, Int32 clicks)
at System.Windows.Forms.Control.WndProc(Message& m)
at System.Windows.Forms.ScrollableControl.WndProc(Message& m)
at System.Windows.Forms.ToolStrip.WndProc(Message& m)
at System.Windows.Forms.ToolStripDropDown.WndProc(Message& m)
at System.Windows.Forms.Control.ControlNativeWindow.OnMessage(Message& m)
at System.Windows.Forms.Control.ControlNativeWindow.WndProc(Message& m)
at System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)
EDIT: Things seem to work if I comment out the exception
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
So, anything happening here?
If still looking for ideas, I think Gunstar Heroes would be a good choice to replace warpless walkathon. Exchanging 25 minutes of walking with 31 minutes of action seems like a good trade to me.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Alright back to work on this!
With the help of adelikat savestates are now functioning. So far in my testing I haven't found any problems but if anyone finds any bugs just let me know.
Be aware that currently there is no mapper support, so only games with normal rom size (8kb) will funciton for the time being. But if you find any 8kb roms that don't work, please post as those are valuable for debugging.
I am working on sound right now. I'm hoping to have it done in the next couple of weeks.
Also adelikat really improved the controllers too , so things are really coming together!
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Woah 0_0
This is some seriously impressive stuff. That first person mode is great. I've seen it done in other ways, but not derived from the original rom, amazing! Of course most NES games aren't meant to be done first person, but it's still cool to see, and that mini map is a great bonus!
How do you render the enemies in such detail?
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Huh? I'm not quite following your response here. Your questions were:
We had a run with good votes, but due to bad ratings it went into the Vault. Now we have a slightly better run with good votes and mostly unchanged entertainment value. How would you predict are the ratings going to be? How would you argue against over 30 ratings?
My response is that the current audience is voting on the current run, and their perception of entertainment value can be different. That there is no need to predict what ratings it might get, since it will be rated anyway after acceptance. I'm not seeing where a previous audience's ratings are overcoming a current audience's feedback. Ratings are also silent.
puts the player above the audience.
I'm also not getting this part, where is the player involved? Are you saying that two similar runs shouldn't be rated differently because it looks like bias against the player? If so then I would say that since these are two seperate audiences, nevermind being seperated by a decade, then getting two different ratings is very possible without any kind of bias.
In particular: Too many bad publications turn the audience away.
The run is getting published one way or another, what is this part referencing?
I understand (I think) some of what you are saying here. A previous run of a game that is not that different from its succesor would seem to warrant it getting a similar standing. But I disagree that this should be taken into consideration to an extent that a current, new audience cannot overcome it.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Masterjun wrote:
We had a run with good votes, but due to bad ratings it went into the Vault. Now we have a slightly better run with good votes and mostly unchanged entertainment value. How would you predict are the ratings going to be?
For everyone arguing that a judge should just look at the votes and decide from there, or that a 15-0 ratio automatically assures a place in Moons, just look at the previous run and the ratings then.
How would you argue against over 30 ratings?
Well, I would argue that people are asked to vote on the current run and whether they found it entertaining, and if you have a system that asks people to vote, then the votes should have some relevence in the outcome.
Based on your arguments above, you could have arrived at the same judgement right when the run was put on the workbench, without any votes one way or the other, in which case, what was the point of voting?
I'm just saying that this doesn't encourage participation or confidence in the voting system. And I don't see the harm in saying 'well the previous run went to vault but people seem to like this one so let's put it in moons and see how it does in ratings.' I think this is better then pre-supposing an outcome.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Alright finally by the hurdles of level 4, go Tompa!
I'm really excited to see the finished project. Maybe even a console verification if possible? That would be really cool.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Amaraticando wrote:
As of version 1.11.9, this still happens:
Amaraticando wrote:
event.oninputpoll is not working in version 1.11.8.2 for the SNES.
The following script, for instance, doesn't print anything.
Language: lua
local poll_count = 0
local function foo()
poll_count = poll_count + 1
print("poll count: ", poll_count)
end
event.oninputpoll(foo)
Just so you know, InputCallbacks.Call() is currently commented out in LibsnesCore.cs (line 511), so this isn't a bug, just not implemented. Maybe onmemoryread/write events will work as a proxy until it is fixed?
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
^ I looked at this bug about a week ago and you are right that this is definitely a video mode issue. I spent a few hours trying to resolve it expecting it to be fairly simple, but it turned out to be not that simple afterall. Right now I shelved it until after I'm done with Intellivision, but I'm open to suggstions if you can point to something specific. (I did try to force different video timings, and got some interrupts to fire at least, but that didn't fix anything.)
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
I generally don't mind tiering decisions since I consider the system irrelevent, and I also understand that voting isn't the only factor in tiering, but I have to say that if you are going to meta-game the voting to the point where a run can go 15-0 at acceptance and be placed in the vault, then perhaps it's a sign that something has gone awry.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Oh it turns out I actually ran into some similar issues with NesHawk's tile viewer a while ago and then forgot. The window deisgner seems to make some inconsistencies sometimes.
In this case the 'Markers' and 'Branches' boxes have no minimum size and somehow get corrupted by whatever magic windows is doing to create the window.
I set a minimum size to 50x50 and the error went away completely, and the window now draws correctly after a fresh unzip.
I can commit it if you are happy with that feos.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2752
Location: US
Yes all this was from fresh unzip.
Ah ok, the problem is with default TAStudio window size, which on my laptop looks like this:
When I try to make a marker in this state the crash happens.
When I expand the columns to look normal I can make markers without issue:
I guess it's trying to make a scroll bar in the 'marker' window and failing because it doesn't fit in the default space.
When I try to make a marker in this state the crash happens.
When I expand the columns to look normal I can make markers without issue:
I guess it's trying to make a scroll bar in the 'marker' window and failing because it doesn't fit in the default space.