Sensitive content
This media may contain sensitive content.
Video yükleniyor...
Video Yüklenemedi
testing out some jiggle physics 📚🤓🤭 #nsfw #booty #asian #asiangirl #egirl #butt
0 Yorum
Yorum bulunmuyor
Orijinal gönderinin yorumları burada görünecek
Benzer Videolar
0:23
Sensitive content
Just Testing Some butt Wiggles~ 🧡 Trying out something new with recording :3c #Femboy #Vrchat #VrchatPhotagraphy #VrchatNSFW #Lewd
itzSpookz
177,999 görüntüleme • 2 ay önce
2:41
Sensitive content
Str8 Nigga sucking dick and getting fucked in the Woods at night, Look how the Dick goes in his Booty like it was alive like a snake or worm or some like I’m going in this Niggas Fat ass butt and I’m not coming out 😩😩😩🥵🥵🥵😍😍😍 🍑💦🍆🌳🌲🍃🌱🌌
Billy P
30,709 görüntüleme • 1 yıl önce
!["I have nothing that would suggest that it is of out-of-this-world origin, except that we can't explain at that point the physics that appear to be employed." ~Rounds (Except? That's a pretty big except that points towards anomalous tech. And it doesn't have to be from out of this world in order to belong to a non-human intelligence. Maybe it's human but not part of our society as we know it? Maybe future humans, or another form of human? Maybe something based here? No matter what Ross Coulthart threw at Senator Mike Rounds, it was deflected without a second thought. Keep reading. I went beyond the video clip.) Rounds: "I try to talk to anybody that has firsthand knowledge of actually working on projects. In fact, I told my team that I don't have an interest in talking to people that have second hand knowledge or that will relay information to me without evidence that they have firsthand knowledge of any particular type of materials or projects. I've talked to individuals that have told me their firsthand knowledge of things that they can't explain, but I have never gotten to the point where I could determine what it was as an item or a project or a platform, that what it really is." (So, has he had any firsthand witnesses tell him that why worked hands-on a non-human craft? If so, what did Rounds do to confirm or refute that? Did he get access to the program and see a craft up close? Did he ever find a program that had that type of evidence? I thought Ross did great but I would have loved if he asked Rounds this: Were you ever denied access to a program/location? "I've given [members of Congress and the executive branch] not only the existence of non-human intelligence but the address to go look to see it. And they were denied access." ~Jay Stratton - Jay Stratton ~ Rounds: "And so, are there things that I can't explain, based upon direct information from individuals that have observed information or materials? The answer to that is yes. But I have nothing that would suggest that it is of out of this world origin, except that we can't explain at that point the physics that appear to be employed. So, whether it's from one of our adversaries or whether it's from a unique project here within the United States, I simply don't know and I won't draw conclusions." (He admits that they can't explain the physics and then reverts to it possibly being from an adversary or one of our own USG projects. And doesn't suggest that it may very well be something truly anomalous.) Ross: The WSJ says, in the 1960s, live, thermonuclear weapons were tested against EMPs. Do you find that plausible? I couldn't believe the USAF would do that. Rounds: I know we've done a great deal of research with regard to EMPs. We've had to study EMPs as a weapon system and our adversaries have looked at them as a weapon system. EMPs would do damage to electronic equipment but not to humans or structures. (Welp. He obviously finds it plausible.) Ross: How about enhancing the whistleblower protections? Rounds: "I don't know that the existing protections fail us." Ross: What about AARO? Rounds: I met with AARO leaders and found them to be frank and open, "I don't have any evidence that says that there's in any way a desire to limit their ability to do research or to hide the information that they're doing. " (Bring in Kosloski and ask him if he told James Fox that he had to get approval from the DOD even if he just wanted to part his hair.) Ross: Rubio said we've had repeated instances of something operating over restricted nuclear facilities, and it's not ours. And we don't know whose it is. Any kind of craft over sensitive national security sites is a national security threat. Rounds: I don't disagree. But I don't know if it's ours or if it belongs to one of our adversaries. And if it's neither, then I don't know whose it is. I'm not gonna jump to conclusions and I'm gonna keep my options open. I wanna make sure that if it's something one of our organizations in the US actually is testing or working with that we not disclose it. If it belongs to one of our adversaries, we wanna make sure that we have all of the information we can gather on it and be prepared to defend against it. "And if there's something that is neither ours nor theirs, it belongs some place else, then my curiousity really starts to grow at that state of the game and I wanna be open to those possibilities as well."](https://image.24vids.com/tw-1939901501583069293/amplify_video_thumb/1939886192201494528/img/dyvzXIitMuwqXiaZ.jpg)
![I think I can finally report some success training a quite accurate IDM capable of recovering keystrokes from Minecraft gameplay, even in quite PvP-heavy situations. At this point the model does not only know what keys are pressed to the extent reasonably discernible, it also knows how fast it is moving in 3D space at all times, even when knockback is mixing with the self-move impulse. Now, recovering keystrokes from normal external capture footage is just about impossible. E.g. W/A/S/D does exactly nothing during partial tick frames and jumping mid-air is also equally useless, so asking the model to recover key down states is inherently unreasoanble. Mouse deltas are also completely arbitrary units, as game mouse sensitivity introduces an arbitrary scale factor into the equation. The only good option is to think carefully about your model-environment contract, and only record "logical actions", not raw keystrokes. So here's a few unfortunate lessons I had to learn in roughly this order. - Choose good units. (bad: mouse deltas, good: delta radians [yes, you will need game-internal state]) - Capture from inside the main game loop and read the game fbo to get consistent frame-action pairing. Doing post-mortem pairing is hopeless. - Carefully define when you think keystrokes actually have an effect. (jump only works on ground, when flying or in water etc.) More subtle: The key may already be down, but no tick has happened yet to actually use the value. Hence: ignore Seperate gamestate into "fast and slow-moving" components. E.g. movement is likely tick based, camera rotation is very likely updated every frame in essentially every game ever. - Think about your frame-action correspondance contract (How old is the frame in relation to the inputs you capture? Will double or tripple buffering affect you?) Think about the game loop timeline, where you are sampling, how old the data you are reading is, and where the ticks are happening around you. Language models used to simply not have a model-environment contract, but even now with the model "living" in a designated harness, the contract still boils down to formatting, and tool implementation intrinsics. While also important, it is still quite a bit more obvious because the violations are in some way shape or form reflected as text you can actually see. - ffmpeg dropping frames cummulatively screws the model the further you get into the sequence because your targets are now shifted. If you can't encode the video in real-time, too bad. - Sodium has a frames in flight system different from vanilla Minecraft, which will also offset your targets from your frames. (there goes that data...) - Models are succeptible to latency. If there is too big of a delay between action and on-screen reflection, your performance degrades. At this point I realize ~100hours of gameplay is essentially no longer usable as a dataset. You can train on this data, but all you'll get is a mushy mess. However, some good news: - Making the model predict physics gamestate scalars helps the model generalize. For instantaneous events like jump, it's unreasonable to ask the model emit a short burst of jump=true at exactly the right time, however if you also predict your current y-velocity, the model has supervision signal for the "latent" from which that onground jump becomes apparent. Recovering x/z motion is also somewhat easier than unmixing it into plausible keystrokes for inertia-heavy player controller logic. - Regressing physics gamestate scalars also seems to make your dataset "bigger". While pure keystroke classification will overfit quickly, predicting exact physics gamestate scalars forces the model to generalize more and you can tolerate far more epochs before validation loss starts to stall out. This is the only reason why it was bearable to dump 100h+ of dataset hours and replace it with ~3 hours of gameplay after the 4th revision of the file format (yeah...) and somehow still have better performance. Now, you might be asking, "isn't this brittle?" and the answer is yesn't. Frame-action correspondance matters for training, but not so much during inference. So as long as you are sampling in roughly the same interval as your training data, you aren't violating any hard contract per-se. Somewhere around the frames ticks are happening, and during training you capture various tick-capture offset relations per random chance, so nothing is too obviously wrong here. HOWEVER, you will get screwed by gui scale, shaders, resource packs, "shit that recording is 1920x1040 because somebody doesn't know fullscreen exists" and other unfortunate edge cases of reality. But I suppose this is the role of dataset size. If all those "contract violations" that a youtube video has compared to the training data are addressed, I think this is a way to turn Youtube into a labeled dataset. I could never shake the feeling that VPT is a sound idea in practice, while never having been properly executed, and I think one reason why it hasn't is because that label boostrapping part is just a pain in the butt to get right. Now, what the player is doing is of course not the only label you can extract from video, but it has to be one of the targets predicted during pretraining to "align" the pretraining objective. Some notes on the video here, the colored dots on the analog visualizer are the ground truth, while the gray dot is the model prediction. Green means correct prediction, red means incorrect prediction at that frame. Model P(key) reports how wrong the prediction is from green (0.0) to red (1.0). You will also notice that during periods of rapid slow down, left and right actions become close to irrecoverable, because there is just that little motion. And some jump actions are not predicted correctly because I got the detection condition for jump events wrong... (duh) LMB/RMB for other than sustained events (like item-consume and block break) also seem to be hopelessly irrecoverable for now. Swing was supposed to do the same thing as motion y did for jump, but its too well behaved as an increasing counter. Maybe partial-tick interpolated values work better (v5 file format then... ugh..)](https://image.24vids.com/tw-2042731804780826849/amplify_video_thumb/2042721691844120576/img/yugYQfuXXQmJpdVp.jpg)
