Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
I've looked over the sound generation code several times, and as far as I can tell samples are being generated correctly in the PSG. Jzintv and BizHawk seem to be doing a fair bit of post processing beyond this point though (I'm not really sure of the details) so it's possible the relative dullness of BizHawk is due to that. I won't be looking much more into it, if someone else wants to try I'll be happy to help though.
@PikachuMan: I won't be doing intellivoice, mainly out of disinterest.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
Here are the two runs, both the old one and the new one. They sync on the latest Dev build of BizHawk.
Old:
http://tasvideos.org/userfiles/info/37881733414401898
New:
http://tasvideos.org/userfiles/info/37881743719985076
There are no differences in input or gameplay, the only change is in how lag is flagged between the two emulators.
I'm at least somewhat hopeful/confident that these runs will sync. But I think this is also a very rare situation that won't really offer much insight into other games, but hey its better then nothing!
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
For anyone who may be interested, I added support for the ColecoVision Super Action Controller in the latest Dev Build. The thumb wheel is broken though and I'm having a hard time finding enough info about it to make it work, but the rest of the controller works for anyone who wants to try it out.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
RobinKeatonTAS wrote:
Alyosha wrote:
RobinKeatonTAS wrote:
After testing these methods out I found out that I have no idea how the macro tool works, and couldn't find any info on it. And I don't know what tool to use to "paste in" the inputs. Help please?..
I don't know how the macro tool works either! 8D
Did you find Input Log.txt? Extract it and post it somewhere and post a link here and I should be able to help out.
Yes I got the .txt and posted the inputs here http://pastebin.com/TuNeqCtm
It's an input list for the jump rope minigame to earn Hee Beans in Mario & Luigi Superstar Saga (in case that was important to note)
Thanks again for putting up with me uwu
http://pastebin.com/pUMaiCRi
Here is the input log with all the buttons repeated. Each row like this:
| 0, 0, 0, 100,...........|
is 1 frame of input.
You had ~3500 frames of input in your original file, now you have ~7000 in the new paste, just copying and pasting them all.
Now copy that file and put it back in your .bk2 and ttry playing it, it should run twice as long and play through that mini game twice if it syncs up right.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
RobinKeatonTAS wrote:
After testing these methods out I found out that I have no idea how the macro tool works, and couldn't find any info on it. And I don't know what tool to use to "paste in" the inputs. Help please?..
I don't know how the macro tool works either! 8D
Did you find Input Log.txt? Extract it and post it somewhere and post a link here and I should be able to help out.
There is also IRC if you want to use it, someone there migt be able to help walk you through the steps real time:
#tasvideos at irc.freenode.net
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
feos wrote:
Alyosha wrote:
I am having this same issue with the dev builds.
When I try to run BizHawk, nothing happens.
But, if I run DiscoHawk, exit it, and then open BizHawk it runs just fine.
I don't know what magic is happening to make this work, but hope it helps in debugging.
Please let zeromus teamview you! This sounds really weird.
Both bizhawk and tastudio start up perfectly for me with the appveyor dev build and local release build.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
Hmm, nope there doesn't seem to be any nearby poll. Back to the drawing board I guess.
True, are you still able to try this on console and record the results? I tried many combinations of starting states and NMI hits but they all sync. Maybe if I can see exactly what's happening on screen I can try t orecreate it.
EDIT: Actually, I see where it would desync in level 6 due to windowing input latches exactly where True describes in the console verification tests page. I don't see where it would desync in level 9, though as described in the most recent published movie's submission page.
Both movies encounter a similar problem in area 6. What happens is an input is latched right before NMI. Only a couple of the subsequent 8 polls are done before the NMI, with the rest of the them done only after about 2200 cycles. I'm not sure how exactly the bot behaves in this situation though, but it's my best guess.
EDIT2: I also notice that FCEUX has different lag frames in those two movies in the respective areas where Trus said they desynced, so that might be the cause as well, maybe it will work dumping it from BizHawk? (Maybe I should have checked that first.)
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
I am having this same issue with the dev builds.
When I try to run BizHawk, nothing happens.
But, if I run DiscoHawk, exit it, and then open BizHawk it runs just fine.
I don't know what magic is happening to make this work, but hope it helps in debugging.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
I'm not sure there is a formal way to do this in the emulator, replaying the movie is not meant to repeat your actions, it's meant to start back at the save state so you can change them, but there is a way to copy your inputs repeatedly.
After you save your movie, you'll have a .bk2 file.
Open this file with something like 7-zip.
Inside you will see a file called Input Log.txt
This file contains all of your button presses in the movie, one line per frame.
It sounds like what you want to do is just repeat these inputs multiple times. So just open Input Log.txt in notepad and copy paste all the inputs as many times as you would like.
As long as these exact same inputs repeat what you want in the exact same way (no RNG in game etc) then this procedure should work.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
Brief update here. It seems like I'm running into a wall again.
I tested Bionic Commando as True reports that desyncs on console, but it syncs just fine in BizHawk. Looking at the trace logger it seems like there are 2 possibilities.
1. Emulation is wrong. It desyncs at a lag frame where only one or two extra instructions could change the outcome, but so far I haven't found a way to change this result.
2. The bot isn't picking up an input poll. On this particular frame input is polled at the very end of the frame, only two dozen insturctions from vblank, maybe this is throwing something off in the replay?
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
Was able to watch the segment live, overall I think it went very well. Choosing a general approach to commentary on MK64 worked out great I think. It was noticable how comfortable you were talking about tasing in general rather then trying to point out everything happening in the run, a lot of it was pretty obvious just from watching anyway.
SM64 commentary though was not so good in my opinion. That one definitely could have used 5 minutes before the run started to explain some details and some history of how significant the moat door skip is.
Overall I liked the segment, and I'm still amazed these runs work on a real N64, it's just plain impressive to watch.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
Cool! Will there be a youtube video or something similar compilation of the complete run?
Not sure what's been done on Meat Boy but I'll be interested to see the results.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
I gave an initial attempt at looking into this, it's certainly more complicated then I initially thought.
So far what I have is 2 savestates on the same frame in slightly different position (0x05B0). One savstate leads to a fuel drop if you go forward, the other doesn't. So far I haven't found any other variable that relates to this drop.
I also have a savestate a little further back, and it seems the order in which you kill enemies also plays some role.
Looks like a more complete RAM map of this game will be needed to sort this out. I'll have some more time later in the week to dive into it a little deeper.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
Warepire wrote:
This is crazy, I am deeply impressed by your progress here. Well done.
Thanks! But so far pretty much all of this is stuff that is already known. Putting it all together into something complete and internally consistent is the challenge now.
Thanks to True's console testing I've got a lot of new data to work with. Battletoads one player warpless made it furtehr then ever, but only once, so that needs investigating. A new sync record for Streemerz was also set, but it desycned at a room with tons of clowns running around, and in several different ways at that, so lot's to look at there.
Perhaps the most important results were from the read2004 test rom. As suspected 4 seems to be a typical alignment at power on, so that is good to know, 7 is as well. Unexpectedly, values rangnig from 0-12 of initial 'FF' were observed that needs to be studied.
Overall I think I need to pick an easier game then the hard core demanding Streemerz and Battletoads, so the hunt is on for that. Unpredicatability of NES power on makes sensitive games difficult to test and identify problems.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
I've made some progress lately in analyzing why and how games sync. I improved NMI behaviour in the emulator and now have managed to see a behaviour that has never occured before in emulator but happened about 50% of the time in True's console testing.
Unfortunately, in doing so I lost sync in Battletoads 2p warps, which did sync on console , so must in some way have represented a valid console state.
The complicating factor here is that ppu/cpu alignment allows for some pretty diificult to capture changes in emulation (without emulating at another level lower accuracy.) So the current build must be emulating a different alignment then the release nuild.
Also, NMI timings are not thoroughly tested by test roms. But if this streemerz run in my test build syncs it should mean its close.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
Thanks to the efforts of Meerkov another game gets crossed off the list, Cobra Traingle!
I never expected such a big improvement to this game, so nice work there.
12 left!
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
Micro Machines depends heavily on PPU revision. It reads from $2004 to perform some of its graphics routines. However, on early NES's this address wasn't readable, so this would result in heavy graphical glitches.
Incidently, I also rented this game as a kid and don't remember any graphical glitchs so I guess you just got unlucky, too bad since its such a fun game.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
link_7777 wrote:
Yes, I'm still planning to submit, and obviously I need to lock everything down this week. I see that Shadow of the Ninja was suggested, but in case we end up on console I'd like to stick to titles that I have in my (limited) collection. With that limit in mind I'm thinking about submitting:
Battletoads (2p warps seems like a good suggestion)
Bionic Commando
River City Ransom (playaround)
None of these uses the DMC channel so I should be able to work with them all pretty easily. (Battletoads being already known to work.)
Personally I'd still want to see runs playing on console if at all possible, even if its along side the emulator with hitboxes and such as feos suggested, it just doesn't feel real without it.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
I've made a number of accuracy improvements to NESHawk lately, including getting Huge Insect to work. This game relies on an obscure hardware glitch in the PPU that I overlooked until now.
Also Micro Machines now runs without graphical glitches, but as always with this game this is tentative until proven totally accurate.
I've made numeroud changes recently, so if anyone notices any bugs please let me know. I tested pretty thoroughly with all the most difficult cases I know so I don't think anything is wrong but can't be certain.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
TheWinslinator wrote:
Ok that makes pretty good sense. I went ahead and booted up my Intellivision with a CRT, and the image dimensions seem to match up perfectly, so you've got me convinced!
I just want to make sure - have all the sound bugs been ironed out yet? I can still detect many sounds being slightly different than how they are on console. Since jzintv sounds nearly identical to how the sound is on console, I decided to do a short comparison video of some sounds between Bizhawk (current dev build) and jzintv, so it's easier to show what differences I'm hearing:
https://www.youtube.com/watch?v=SI6IcnOjqXs
Hope this helps!
Ah thank you that helps immensely.
I guess I don't have a good ear for audio so having them side by side like that really makes it easier to pick out the differences.
Hopefully it will help me track down the remaining bugs.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
^ That is the playfield output but not the complete video output.
This is what is programmed into the core:
public int BufferWidth { get { return 176; } }
public int BufferHeight { get { return 208; } }
public int VirtualWidth { get { return 277; } }
public int VirtualHeight { get { return 208; } }
The 'Buffer' is the physical video signal output, it is 176x208 after accounting for the display border (this corresponds to the 176x104 that Winslinator had with scanline doubling that the console does.)
The Aspect ratio (the 'virtual' part) depends on the TV but I stretched it to be 4x3 which is mostly what was around at the time.
Editor, Experienced Forum User, Published Author, Expert player
(3572)
Joined: 11/30/2014
Posts: 2744
Location: US
I recently stumbled across a test rom that tests some bizarre NES sprite evaluation glitches that NES Hawk wasn't even close to passing. (Mesen and punes aren't close either.)
So I completely overhauled sprite evaluation and now I get all the correct reads and sprite data and all pathological behaviour accounted for.
The only difference between this and the results posted by Quietust in NESDEV is one extra 'FF' at the beginning. Quietust mentioned that there might be a variable number of those bytes at the beginning (which represents the tests starting a different pixel) but I currently don't understand why this would be the case or what the expected outcome should be.
This is actually good news since I didn't have any new good leads lately as to why some runs were failing to sync, so I hope this leads to some new information.
But all the behaviour is now emulated correctly, so this is a pretty big accuracy improvement. Let's hope there are more to come!
See this thread for more info:
http://forums.nesdev.com/viewtopic.php?t=2295
I've made some progress lately in analyzing why and how games sync. I improved NMI behaviour in the emulator and now have managed to see a behaviour that has never occured before in emulator but happened about 50% of the time in True's console testing.
Unfortunately, in doing so I lost sync in Battletoads 2p warps, which did sync on console , so must in some way have represented a valid console state.
The complicating factor here is that ppu/cpu alignment allows for some pretty diificult to capture changes in emulation (without emulating at another level lower accuracy.) So the current build must be emulating a different alignment then the release nuild.
Also, NMI timings are not thoroughly tested by test roms. But if this streemerz run in my test build syncs it should mean its close.
I've made a number of accuracy improvements to NESHawk lately, including getting Huge Insect to work. This game relies on an obscure hardware glitch in the PPU that I overlooked until now.
Also Micro Machines now runs without graphical glitches, but as always with this game this is tentative until proven totally accurate.
I've made numeroud changes recently, so if anyone notices any bugs please let me know. I tested pretty thoroughly with all the most difficult cases I know so I don't think anything is wrong but can't be certain.
The only difference between this and the results posted by Quietust in NESDEV is one extra 'FF' at the beginning. Quietust mentioned that there might be a variable number of those bytes at the beginning (which represents the tests starting a different pixel) but I currently don't understand why this would be the case or what the expected outcome should be.
This is actually good news since I didn't have any new good leads lately as to why some runs were failing to sync, so I hope this leads to some new information.
But all the behaviour is now emulated correctly, so this is a pretty big accuracy improvement. Let's hope there are more to come!
See this thread for more info:
http://forums.nesdev.com/viewtopic.php?t=2295