Username:
B
I
U
S
"
url
img
#
code
sup
sub
font
size
color
smiley
embarassed
thumbsup
happy
Huh?
Angry
Roll Eyes
Undecided
Lips Sealed
Kiss
Cry
Grin
Wink
Tongue
Shocked
Cheesy
Smiley
Sad
<- 1,  2,  3 ->
--
--
List results:
Search options:
Use \ before commas in usernames
moozooh
moozooh
Edit history:
moozooh: 2011-02-22 04:36:07 am
2011-02-22 04:32:47 am
Hail Discordia!
TASVideos has launched an ongoing list of TASes verified on consoles:
http://tasvideos.org/ConsoleVerified.html

If the idea catches on enough for more NESbots to be produced and distributed, the process will definitely accelerate. I don't expect even a quarter of our movies to be verified this way if only because how much of a pain and financial expense it is to locate the required carts, but if it's done for at least the more popular and/or glitched titles, it's already good enough I suppose.

Those of you with large NES collections could probably help by sending carts for verification to those having a NESbot (currently only its author has one), or getting/building a NESbot yourself. I'm guessing this could be useful for runners that would like to try more advanced glitches, like in Mega Man series or whatnot.

As said earlier, the mere existence of such device does invoke some more headache for SDA verifiers, but there are good sides to it, too. Such as at least knowing about it. It may call for more scrutiny or additional footage in analyzing highly competed runs, but at the end of the day such measures will give more confidence in purity of the product.

Quote from VorpalEdge:
The problem is that x2 uses a special chip.  Its logic is fully emulated but its lag is not.

Correct. It's actually one of the very few SNES games to not have a complete low-level emulation yet. But it's getting there!
Randil
Randil
Edit history:
Randil: 2011-03-01 11:09:26 pm
2011-03-01 11:07:32 pm
I'm just a friendly TASer visiting
Just a small note: The link to the list of console verified games is actually:
http://tasvideos.org/ConsoleVerifiedMovies.html
moozooh
2011-03-02 12:52:51 am
Hail Discordia!
Ah, it was changed since I had written that post. Thanks.
Scepheo
2011-03-02 07:37:55 am
Also, the ability to verify that any glitches seen in a TAS are actual game glitches and not emulator glitches could save some runners a lot of time trying to pull them off.
andrewg
2011-03-02 10:03:17 am
As far as legitimacy of runs. if a run is "too good", SDA could always ask for video proof showing them play.
pojr
pojr
Edit history:
pojr: 2011-03-02 10:19:59 am
2011-03-02 10:19:32 am
Quote from andrewg:
As far as legitimacy of runs. if a run is "too good", SDA could always ask for video proof showing them play.


Therefore, you'd have to record the game footage and you playing the game. That's a lot of space.

Also, seeing the Gradius TAS on actual NES hardware would blow my mind.
dballin
2011-03-02 04:33:13 pm
Balls jerky


Click with caution.
ZenicReverie
ZenicReverie
Edit history:
ZenicReverie: 2011-03-02 05:46:27 pm
2011-03-02 05:44:57 pm
Waiting hurts my soul...
"Oh no!" *mind blown*

I still want to see a game done like this that is known for a lot of random elements, and if it does sync, it'd be nice to know with what amount of consistency does it work.
Harry9397
2011-03-19 03:20:10 pm
hello
This makes me sad, there is always an element of trust in the speedrunning I do - xbox live leaderboards but SDA's records always feel like you can trust them.
How can we trust records now? The only way to be 100% is to have someone verify the hardware and sit besides someone while they run, I am sure it won't come to this, I think if anything changes it will be just looking out for "too good" runs or asking to film the hardware and the TV a little (not that this can't be faked)
I am sure it wouldn't be too hard to do something like this on say SMB and make the run look "real" but a little faster than andrew's run, maybe with a little build up (maybe a year of posting about your ever increasingly good runs until you finally decide to go ahead and submit it, and nobody would believe you were lying)
Just an extreme example there, I am sure there is already a way to get a run past SDA's verification process but whatever, just thought I would share.
dballin
2011-03-19 03:41:37 pm
Balls jerky
im so totally gonna do that. LOOK OUT ANDREW!!
dballin
2011-04-02 05:07:29 pm
Balls jerky
[18:09] <dballin> congrats nes runners
[18:09] <dballin> your runs are safe FOR NOW
[18:10] <dballin> out of the 14 games I loaned to the nesbot guy, 0 of them worked
Zeo
2011-04-03 03:39:10 am
It appears its mainly first party games.  Aka SMB series + Graduis and MM.

I agree, if it becomes an issue, maybe from then on all SMB 1-3 +2J runs should be Video cameraed, and that you need to play a "loss leader" game before you start proving the inputs are coming from the controller.

moozooh
moozooh
Edit history:
moozooh: 2011-04-03 03:54:37 am
2011-04-03 03:54:23 am
Hail Discordia!
Marble Madness is Rare, Mega Man is Capcom, and Gradius is Konami, neither of which is first-party, assuming you mean Nintendo. Those are all big studios, but there's no dependency to trace as to which studios' games would sync and which wouldn't. It has more to do with the fact that imperfect emulators have mainly been tested on popular games, so if they include some hacks or high-level optimizations, more popular games are less likely to show them.
Zeo
2011-04-03 04:03:44 am
Megaman hasn't been verified yet.  ESP 4-6 would mean a huge improvement.

but of the verified games, 3 of 5 are 1rst party.  Also, Nintendo tended to use less hardware tweaks and such, and as you said, have been tested more thoroughly.  Hence I see definite likely correlation between Nintendo published games, and NES bot compatibility.

I'm willing to bet Zelda 1 and 2 are being tested right now.  heads or tails as to which one breaks first.
moozooh
2011-04-03 04:43:39 am
Hail Discordia!
Oh huh, I was sure I remembered a post that said the recent MM (or was it MM2?) TAS syncs, but I can't find it. My bad.
dballin
dballin
Edit history:
dballin: 2011-04-03 09:56:22 am
2011-04-03 09:55:56 am
Balls jerky
6 of the 14 games were Megaman and they didn't work. Actually MM6 played for about 15 minutes before desycing.

Quote:
Either way, I should have all the other videos up by the end of next week. Expect a hugely long post... I only have 39videos (6GB) to upload and link to!


You're just gonna have to wait and see!
error1
2011-04-04 12:50:43 am
Not a Zelda 2 refrence
It's not like people haven't been able to do this for computer speed runs since the early days of speed running. But that's what the verifiers are for. I don't see why this would require any other proof then just convincing the verifiers, it's just something they need to keep in mind when verifying the next smb run.
Zeo
2011-04-04 04:03:34 am
Quote from error1:
It's not like people haven't been able to do this for computer speed runs since the early days of speed running. But that's what the verifiers are for. I don't see why this would require any other proof then just convincing the verifiers, it's just something they need to keep in mind when verifying the next smb run.


Becuase before this, recording off a real unit cut cheating probability down to next to 0, excluding famiclones.
moozooh
moozooh
Edit history:
moozooh: 2011-04-04 04:21:20 am
2011-04-04 04:20:17 am
Hail Discordia!
It's not that simple. If you take an accurate enough emulator and wire it through a TV (or a similar source of analog distortion), you'll get a signal nearly identical to a signal you would get off an actual console. Then, if you record a cheated run at 100% speed using savestates, you will get a run nearly identical to what could be expected from a normal human performance. Things like this put high responsibility on verifiers, so those with experience have to take many things into account while also understanding that no evidence of cheating typically translates to no cheating.

Besides, it's not that the technology behind this didn't exist throughout the past few years. It just wasn't publicized. Being publicized makes it harder to use for cheating purposes now.
andrewg
2011-04-04 08:42:02 am
"Being publicized makes it harder to use for cheating purposes now."

I'd say the opposite. Now that people know about it, they will be more likely to think to cheat by using it. :/ I think I might start recording everything I do by camera. I would hate to do that though >:|
dballin
2011-04-04 10:24:41 am
better watch out. he's working on a snesbot and wants to build a n64bot too. OH NO!!
moozooh
2011-04-04 12:06:50 pm
Hail Discordia!
Quote from andrewg:
I'd say the opposite. Now that people know about it, they will be more likely to think to cheat by using it.


Now that people know about it, they'll also know verifiers will be looking out for it and demand further proof in case they consider the run suspiciously good. Don't be such a Debbie Downer. Smiley
Zeo
2011-04-04 02:26:30 pm
Quote from andrewg:
"Being publicized makes it harder to use for cheating purposes now."

I'd say the opposite. Now that people know about it, they will be more likely to think to cheat by using it. :/ I think I might start recording everything I do by camera. I would hate to do that though >:|


Honestly andrew, I'd really only worry about the three minimalist runs for now.  Since they are such  perfect runs, and would be easily  tagged.  the rest of you're stuff is long enough that there is enough time for enough "human Character" to show.
dballin
dballin
Edit history:
dballin: 2011-04-05 03:16:23 pm
2011-04-05 03:15:06 pm
Balls jerky
Quote from micro500:
I borrowed a bunch of games (thanks dballin!), and none of them synced. Some definitely needed the Japanese cart (Lunar Pool, MM2 and 3), some needed two players. I tried as many movies for each cartridge as I could without having to deal with Famtasia (which I doubt would work anyway). Some didn't work in the emulator after converting the fcm. Any that did work in the emulator though I tried on the console, and below are the results.

I did get my own Tetris cartridge to sync movies. Interestingly though one of the movies does not sync, and I'm not too sure why that is

Console Verified:
Youtube - [movie]1596[/movie]
Youtube - [movie]924[/movie]
Youtube - [movie]399[/movie]
Youtube - [movie]239[/movie]
Youtube - [movie]1502[/movie]
Youtube - [movie]925[/movie]
Youtube - [movie]920[/movie]

Desyncs:
Youtube - [movie]386[/movie]
Youtube - [movie]365[/movie]
Youtube - [movie]1348[/movie]
Youtube - [movie]1070[/movie]
Youtube - [movie]517[/movie]
Youtube - [movie]297[/movie]
Youtube - [movie]1164[/movie]
Youtube - [movie]1444[/movie]
Youtube - [movie]545[/movie]
Youtube - [movie]442[/movie]
Youtube - [movie]1549[/movie]
Youtube - [movie]1241[/movie]
Youtube - [movie]1686[/movie]
Youtube - [movie]726[/movie]
Youtube - [movie]715[/movie]
Youtube - [movie]515[/movie]
Youtube - [movie]1695[/movie]
Youtube - [movie]749[/movie]
Youtube - [movie]1103[/movie]

I get the feeling Excitebike is not emulated correctly, since any of the movies that tapped the throttle to match the game's 4 cycle read loop didn't work correctly. I tried shifting the input on those movies one frame and the game finally saw it and the player moved, but other things didn't work (opponents weren't in the right place, crashing after jumps, etc). It seems like the race starts one frame too early in the emulator compared to the console.

I also borrowed Donkey Kong and Donkey Kong Jr., both of which desynced repeatedly. After returning the games we realized it may have been the cartridge's fault (oxidized contacts). I shot some videos of them, but I feel like they're not worth uploading. Let me know if you guys think differently though!


SAFE

oh.. yeah im not reformatting that [movie] tag thing so yeah go here to see that crap
andrewg
2011-04-06 02:18:19 pm
Quote from Zeo:
Quote from andrewg:
"Being publicized makes it harder to use for cheating purposes now."

I'd say the opposite. Now that people know about it, they will be more likely to think to cheat by using it. :/ I think I might start recording everything I do by camera. I would hate to do that though >:|


Honestly andrew, I'd really only worry about the three minimalist runs for now.  Since they are such  perfect runs, and would be easily  tagged.  the rest of you're stuff is long enough that there is enough time for enough "human Character" to show.


Mario 2 is a horrible run Wink

If someone were to submit some amazing run for a game I'm good at, I'd be tempted to improve it anyway, despite if it were done by an NESbot. I think you would be able to tell if a run were too perfect, but at the same time, some of the runs I've seen are just insanely perfect because of the amount of effort that's gone into them: superc, smb, smb3 are probably some of the best NES examples.


It's like, if some unknown runner submits a time that beats the smb3 run by like 5 seconds, I'd be pretty skeptical, but if Freddy submitted an improvement to his own run, I really wouldn't have many doubts about it. Not to say that well known runners wouldn't stoop to cheating, but I'd say it's less likely.