Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
phoenix1291 wrote:
I don't know if this is the case with other cores (I didn't have time to test), but with Atari7800hawk core (last dev version), for me it's impossible to make a video rendering by resizing it, I tested with most codecs, by not resizing the video at the output, no worries, the video is good, but by resizing, only the sound is ok, the image is not, there's just a black screen.
woah that's weird. After a cursory glance I don't see any reason why this should be happening. I'll look into it.
EDIT: Seems to be broken in "Allow alpha to matter in screenshots. " here (BitmapBuffer.cs):
public void DiscardAlpha()
{
// HasAlpha = false;
}
The commented out code breaks scaled video capture here (MainForm.cs):
bbin.DiscardAlpha();
var bmpout = new Bitmap(_avwriterResizew, _avwriterResizeh, PixelFormat.Format32bppArgb);
bmpin = bbin.ToSysdrawingBitmap();
That's definitely the cause. not sure what side effects are are uncommenting out that.
EDIT: fixed in master
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
No, streemerz should have the same sync requirements as battletoads but somehow never synced on console in earlier tests despite many tries . So it’s likely that battletoads synced only from lucky error cancelling.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
For the first time in a long time I have some new direction for possible NES console verifications. I have created a movie file with a version of Bizhawk that syncs to Mesen's power up timings.
the movie file won't sync in the dev build of bizhawk since the timings are different, but here is a r08 of the input file I made using my personal build:
https://www.dropbox.com/s/ukn7330nwkjhqxe/streemerz.bk2%20%281%29.frame.r08?dl=0
If anyone out there has the ability to do some console testing please give it a try. Keep in mind that NES power up timings are still quite variable, so it might take numerous tries to get it to sync, as this game is ppu timing sensitive.
If this works out it is likely that I will change NEShawk's timings to match Mesen, then there will be many new areas to investigate, but for now hopefully this test works.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
I practically never watch published tases.
I can recall occasionally downloading an mkv if I didn’t want to run a movie in an unfamiliar emulator , like for comparisons in Dega movies .
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
Game boy link cable support is essentially complete for both GBHawk and Gambatte. Well for 2 players at least. Also GBHawk appears to have slightly better compatibility. Future work might involve 4 player support but it’s not likely in the near term.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
I have started work and made initial commits on Game Gear link cable support (GGHawkLink)
I was able to reuse a large part of the GBHawkLink code which made the initial steps relatively painless.
So far you can use multidisk bundler and load the 2 games but they currently aren't actually linked.
It looks like it should be possible to get all the linking stuff sorted out. It's a bit messy though since the SMS core is pretty far behind in terms of development compared to NESHawk and GBHawk. So if I run into any roadblocks it might delay things quite a bit.
I'm not sure how much linked Game Gear play really offers, but it seemed like low hanging fruit so I went ahead and started it. this also shares a lot with what would be needed for subframe input support, so it's a first step in that direction as well.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
I just made a fairly significant commit to Atari 2600 that improves HMove emulation.
now various newer demos relying on precise Hmove behaviour (like Bang! and Flush) work.
Also behaviour is much closer to Stella in "Hit HMOVE At Various Cycles After WSYNC Test (Bradford W. Mott) (1998) (PD)" although there are a couple remaining differences.
Anyone using A2600 please keep an eye out for regressions.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
dwangoAC wrote:
Thanks for the feedback, keep it coming. A couple of points - I don't think that TAS content at GDQ's has "jumped the shark" in the classic sense as we're not so desperate for views that we're trying anything and everything to retain them. Having said that, we're possibly in a midlife crisis. We're 5 years in to something that I anticipate will be a 50 year journey so perhaps that's not apt either, more like we're having growing pains, but you get the idea. And yes, I plan to be doing this for 50 years. Maybe not the same form, maybe not the same "traditional" consoles that we use now that will be archaic in 50 years when I'm in my 80's, but a continuing, evolving concept of using physical hardware and tools to create art.
I think this is a good explanation of the current situation. There isn't really a clear path forward from here. Even with LibTAS, only very few games are worth presenting in traditional TAS style that could actually be entertaining (Celeste being the current example.)
What comes to mind for me for a good direction to look in is things like 'Luigi wins by doing absolutely nothing' or the similar thing Masterjun did a while back with Kirby's Air Ride. I think it's unintuitive and silly goals like this, yet still easy to understand and follow, that could provide the most entertainment.
Of course they are also much harder to come up with. And you have to be careful they don't become gimmicky and boring like Scribblenauts.
___________
On a completely separate note, there seem to be more and more dead ends lately in terms of getting things to actually work. Obviously I'm not privy to the details, but this seems to have eaten up a rather large amount of time. Maybe it's time to set out a semi-formal roadmap of things that need to be done / researched and what future work this could enable. Some focused R&D seems necessary, and not just for immediate needs, thinking a few years down the road.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
I have two observations here.
First, @spikestuff, assuming you aren't otherwise removed from staff (which seems it won't be the case) maybe it's time to take a brake? To me these are the actions of someone way too invested in something that isn't all that important. Maybe setting it aside for a few months can let you get a fresh perspective on things and give you a fresh sense of balance. I've had to do this at times in the past and it really helped me realize how far out of whack I had gotten.
Second, I have to say, I am almost stunned at the staff response here. I can only describe it as half interested and tired. But even that isn't really the right words. It just strikes me as strange. Did morale fall off a cliff lately?
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
My Goodness.
Thank you for bringing all of this information to everyone's attention BrunoVisnadi. Bringing this kind of shadowy nonsense into the spotlight will hopefully bring a close to it.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
Ah ok now I understand. For some reason I was confusing two ideas about the screen state in my head.
Should be fixed in the most recent commit. Thanks for being persistent.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
SubNESHawk is now available as a core option in BizHawk.
It is a subframe input capable NES core. It offers poll based input as well as ppu-cycle level reset timing.
This enables some long sought after TASes to be made such as SMB3 DMC glitch to credits.
To use SubNESHawk, first select NEShawk as your NES core (Config->Cores->NES->NESHawk), then check the 'SubNESHawk' option in Config->Cores.
If you want to TAS NES games as normal, you can still use ordinary NESHawk just by not having the SubNESHawk box checked.
NOTE: This core is not meant for casual play. Consider it TAS only.
NOTE2: Movies from this are not yet fully supported on TASVideos, and the core is still marked as experimental.
If you have any questions comments or suggestions feel free to post or pm me.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
feos wrote:
Is ppu cycle printed in the trace log or shown in any other way?
Good point, having that info in the trace log would make it much easier to work with.
I just implemented it.
EDIT: Note that other cores using the 6502 tracer will just return 0 for the ppu cycle. I guess I could work on giving them a proper counter as well.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
I added in subframe resets. You can reset at ppu cycle precision.
How it works is when you pick a number to target and then press the reset button. If that cycle number comes up during the frame a reset will occur . If not no reset will occur.
Note this breaks existing movies , sorry about that.
I’m not sure if there are any useful test cases for this, but this is probably the last change that’s happening .
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
Mari0: It was ok, definitely too long though. I agree that some context would have helped. Probably could have replaced the 4 player segment with the kind of introduction Masterjun suggested since it was too hard to follow anyway.
Castlevania: Boring. To echo some earlier points, there is no expectation of failure in TAS, so seeing perfect play isn't impressive. It was also very jumpy which made it confusing for me and not fun to watch. (Also it was on emulator but I think that point is beaten to death by now.)
Scribblenauts: I like the idea of interaction, but I don't like pay to play. This is on GDQs end so nothing you can do about it. I also found it strange that the automation made it slower then just typing in the answers by hand, which at least for me put it more in the realm of a needless gadget. But maybe a general audience liked it I don't really know.
Everything seemed to go smooth though so that's good. Although that guy describing the difficulty of working with the scribblenauts machine makes me think that it was probably a close call again.
Anyway I'm looking forward to seeing more cool stuff come out of LibTAS, so that seems like a fruitful direction to go in.
Too bad the console verification front seems to be hitting a wall and giving you lot's of problems, but that's on emulator authors so you can only do so much.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
Interesting. I think I found a flaw in the link logic, so I'll post a fix. (It will break savestates so you'll have to export to BK2, but it should still sync.)
Anyway that's probably the most intense game of GB Tennis ever played! Pretty cool stuff.
Editor, Experienced Forum User, Published Author, Expert player
(3560)
Joined: 11/30/2014
Posts: 2744
Location: US
The screen doesn't update immediately because I didn't want to put the entire screen state in the savestate. If it's really preventing you from doing something I can add it in, but I'd prefer not to.
Can you clarify the desync? Did it play differently when you were TASing it and then when you replayed it it was different (i.e. savestate de-sync) or is it just the desync between the 2 consoles like in the video (link desync)?
