EPT[Game] Osu! - Page 252
| Forum Index > General Games |
|
Bobbias
Canada1373 Posts
| ||
|
Cyro
United Kingdom20278 Posts
Now for the important stuff, how this all affects osu: Microsutter: There is no microsutter in osu if your pc isn't total dogshit. No vsync/buffering at all, will always have the lowest latency. Doubly so when considering the way osu handles inputs latency. Tearing is irrelevant: Objects in osu don't move, tearing is a non-issue. And tearing is the only reason to ever use vsync/triple buffering. Its a tradeoff of tearing vs latency. "screen based motion blur" is irrelevant: Nothing moves in osu, only disappears/reappears. Your mouse doesn't overlap between frames either. Backlight strobing: It is essentially making "old" frames disappear faster at the cost of delaying the appearance of new frames. Move your cursor around really fast in a circle, how many can you see, more than 1? Backlight strobing or not won't change how your eyes work. You are mistaken on screen based motion blur being irrelevant. If your eyes are tracking a cursor that is moving at 1000 pixels per second, on a screen with constant backlight, even if pixels transition instantly, it'll blur by ~17 pixels on a 60hz monitor. That's not a small amount of pixels. This has NOTHING to do with the refresh of the screen, merely the fact that our eyes are moving, expecting a constantly moving object, while recieving light from an object that near instantly changes positions and then sits on the same place for an extended period of time. It's a very minor concern for OSU (i mostly checked it out for laughs) but it's visible. If you have not seen close to zero motion blur, you can't really compare - it's really a black and white thing. This stuff is LOVED among people who like motion quality and especially in the FPS scene. No vsync/buffering at all, will always have the lowest latency. Doubly so when considering the way osu handles inputs latency. Yes, constant rendering = lowest latency. The point is, in some situations you can sync with refresh (especially @120hz) with extremely small latency penalties, especially when you're capable of rendering a game (like osu or CS) at 500+fps Backlight strobing: It is essentially making "old" frames disappear faster at the cost of delaying the appearance of new frames. Not really. Lets say we have constant backlight, you get something like.. 11111111111222222222 vs strobe: 110000000022000000 There is a delay of literally a few milliseconds before backlight is turned on to allow for pixel transitions. g2g on many of these monitors is rated at 1ms. If you'd like to continue conversation, it's probably better to do so on OCN in the 120hz/lightboost thread, where there are plenty of people who can explain things better than me and answer your questions. http://testufo.com/#test=photo&photo=toronto-map.png&pps=240&pursuit=0&height=0 ^Increase the speed until you can't read text any more, how fast is it? If you're on a bad 60hz panel, it can be below 400. A great one, maybe 800 with quite a lot of blur. If you're strobing you can increase speed to 1920+ and be able to read no problem, motion blur is reduced by a factor of 12. In sc2 for example, it's extremely visible. If you have a pack of zerglings running on creep, w/ strobe, you have solid, green, easily readable healthbars. Without, it's a blurry blob. You really need to compare before and after to see this type of thing - i had no idea that it was such a big deal and did not primarily buy monitor for strobe capability, only for smoothness of motion, cutting another ~8ms from my mouse to screen latency, etc. On blur being a good thing: I think that's an opinion and i disagree with it. You really have to see stuff like the zergling healthbars, enhanced motion quality in general and things like ability to identify a target while moving the camera in an FPS game. It's the biggest unfair advantage that i have ever taken from a component upgrade. No post processing looks like a bunch of pictures, not natural flowing movement. It looks almost unnaturally smooth @120fps/120hz compared to "regular" lcd's - you don't >need< motion blur for the illusion of motion at such framerates. That is hardly so at youtube's 30fps. It's part of the reason there's typically less motion blur on a PC game played at 60fps vs a 24fps movie - though there's still display/eye based motion blur if you remove all of it from frames, and increasing framerate further allows you to remove that very comfortably.Low persistence is not such a bad thing - for example it's being used heavily in the development of VR technology - it's a BIG focus of the oculus rift - 120hz low persistence (strobed) to simulate realism. I'd suggest you look into that a bit if you are interested~ For osu, the only stuff it really affects is cursor and hitcircles moving in. I was just having fun messing with it~ | ||
|
maru~
2345 Posts
On March 19 2014 14:33 OdinOfPergo wrote: + Show Spoiler + Be prepared to grab your brain in despair, as I fail to clear even the most simplest of patterns! I'm honestly not sure how to check spectators though.. So if you catch me online throw me a pm so I can add you myself ;p. I've been skimming through this thread today for tips.. So far the big things I need to work on - Posture: I can't be that bad, because I never have pain while playing. I've been a long time guitar/piano enthusiast.., so maby that's helping me. I'll sometimes end up with a sensitive wrist at the end of my 3-11 hour gaming spree. But it's usually just a tight feeling. Not pain, but I can tell my muscles have apostrophized and they no longer really like to support my gaming sprees. I expect some good practice hours will remedy this.. A big nono I've developed since I'm breezing through right now is I don't single tap. Like ever.. when I started I alternated fingers. I pretty much switch between index/middle finger for x/z since as long as I've been playing. That didn't change when I came back either. As for perspective on that, I don't really think it's helping me any and the past 2-3 days I've been playing solely single tap (with middle finger on z) My acc and max combos have gone done a lot but I feel like my speed and ability to keep rhythm are going up slowly. Point in cause is that previously I could only clear the really simple hard maps. Like 2.0-2.6 star maps. The super ezpz stuff. Since I've been trying to learn single taps and better alternating only when I had to I can clear higher star ratings.. Like 2.4-3.0 maps (it's not pretty but I don't die most of the time at least.) That being said I would love to have some links to some good training maps! I have seen and dled a few of the maps that I've seen put up in this thread.., but I'll be honest, I have a really hectic schedule and not so much time to research things. So I'd appreciate any recent recommendations for maps to help me progress. On that note, I'll jump back to my initial statement. I'm really bad at the triangular patterns that start getting introduced in hard/insane map sets. But I do need to learn the basic patterns eventually or I'm going to hit a ceiling (Which is approaching fast :[ ) I fail even the easiest maps over and over again because I fail to be able to read these progressions. So I'd really like some.. er... how to say.. not easy.. but weaker hp drain? maps that I can practice these on without insta failing half way through the progression. I really don't like to run the no fail mod/HT, hints my request for lower hp drain maps. Anyway, I could probably write a book on the thing I need to improve but hopefully this is enough for you guys to get a general idea for now.. I'm usually on late night/early morning U.S. time for anyone who wishes to spectate. But that can fluctuate due to my work schedule.. First of all, if you haven't done so already you should adjust your dpi and mouse settings. I can't be that bad, because I never have pain while playing. Doesn't say much if you mostly play normal and hard maps I think. As Yorbon said, you don't really need to worry about improving certain aspects at this point. I would focus on hard maps for now until you feel comfortable for the most part (as in you can pass most hards on first try and usually get a decent combo/acc), then do the same with "easy" insanes etc. Some beatmaps that might or might not be interesting for you: https://osu.ppy.sh/b/108146 (qinche) https://osu.ppy.sh/b/96382 (Hard) https://osu.ppy.sh/b/110835 (Hard; this one is tricky) https://osu.ppy.sh/b/178588 (cRyo's Insane) https://osu.ppy.sh/b/250577 (Hard, quite fast and fun to play) https://osu.ppy.sh/b/357381 (both insanes, fast) https://osu.ppy.sh/b/184680 (Ra's Hard) https://osu.ppy.sh/s/42854 (might be too difficult) https://osu.ppy.sh/b/346804 (try both) Have fun. | ||
|
Cyro
United Kingdom20278 Posts
Doesn't say much if you play mostly play normal and hard maps I think. Agree, no pain or problems doing easy stuff says nothing. I've had to constantly refine mouse settings and posture increasing skill and still an ongoing process. The harder stuff is, the higher standards are and the more likely you are to run into issues - it's why we have millions of people for example playing LoL with mouse settings like 3000dpi that "works fine for them" that are completely unable to do more refined tasks Improving posture and mouse/keyboard control in general is good for everything, i found it helping a lot of stuff that i wouldn't even imagine it to help | ||
|
Xafnia
Canada874 Posts
| ||
|
maru~
2345 Posts
On March 20 2014 01:55 Xafnia wrote: Userpage now lists all your scores by pp in descending order, and how much pp they give. Is this how much pp the score would give to a player with 0 pp or how does it work? | ||
|
Yorbon
Netherlands4272 Posts
On March 20 2014 01:57 maru~ wrote: I believe the highest x (?) scores add up to your total pp amount. So that means the amount of pp a score gives is the amount it gives minus the amount the lowest score in the top x gives, the latter of which will be kicked out of the top x.Holy shit, I just saw the new Top Ranks in the profile. I don't get the pp value though. Is this how much pp the score would give to a player with 0 pp or how does it work? edit: ignore what i said. It's kind of similar to it, but the weights are listed at top ranks as well. They should be giving some insight. | ||
|
Yorbon
Netherlands4272 Posts
On March 20 2014 00:06 Bobbias wrote: I'm far more relaxed at singletapping than alternating (as in, state of mind). Also, changes in tempo are really stressfull. I do practice alternating songs (for example this with dt or something) to improve my aim at high speeds and the tension on my left hand fingers are pretty great at quarter notes into eighth note patterns. Also, tension on my right hand rises, because i need it to keep track of 2 seperate entities (at jumpy parts, that's why spaced streams with fast changes in direction are so hard for me to play).Alternating is a HUGE advantage. By reducing the speed that you have to play any single finger at it's easier to keep relaxed, which is what you should always be trying to do. The tenser you are, the more you'll slow yourself down. I do agree that alternating is better, but i don't see the advantage being huge, and it being easier te relax might be a more personal issue than you think, because my experiences are contrary. | ||
|
Cyro
United Kingdom20278 Posts
![[image loading]](http://d116qauid5ktok.cloudfront.net/cdn/3.4.10/wp-content/themes/oculus/img/product/three.jpg) http://blogs.valvesoftware.com/abrash/down-the-vr-rabbit-hole-fixing-judder/ ^scroll down to persistence | ||
|
Xafnia
Canada874 Posts
On March 20 2014 08:44 Cyro wrote: Xafnia, here's a shot from oculus dev kit 2.0 advertising low persistence (strobed OLED) display http://blogs.valvesoftware.com/abrash/down-the-vr-rabbit-hole-fixing-judder/ ^scroll down to persistence You'd want the right side image, on a normal display. None of that is relevant to a game like osu. The original point of this discussion is about vsync/triple buffering etc, in osu. I'm not arguing in any way the relevance of backlight strobing, vsync, triple buffering, in other games. I didn't want to go too in detail in my previous post (I mean, ultimately, I could discuss this for hours). That whole blog is based on HMDs, not monitors, which are not really the same, they have a lot more problems inherent with their use than monitors do. Previous blog post: http://blogs.valvesoftware.com/abrash/why-virtual-isnt-real-to-your-brain-judder/ Why isn’t judder a big problem for movies, TV, and computer displays? I mentioned in the last post that HMDs are very different from other types of displays, and one aspect of that is that judder is a more serious problem for HMDs. Why isn’t judder a major problem for movies, TVs, and computer displays? Actually, judder is a significant problem for TV and movies, or at least it would be except that cinematographers go to great lengths to avoid it. For example, you will rarely see a rapid pan in a movie, and when you do, you won’t be able to see much of anything other than blur indicating the direction of motion. Slow LCD switching times, like those in the Rift development kit HMDs, result in per-frame pixel updates that are quite different from the near-instantaneous modification of the pixel state that you’d see with OLEDs or scanning lasers; with LCD panels, pixel updates follow a ramped curve. This produces blurring that exaggerates smearing, making it longer and smoother, and masks strobing. While that does mostly solve the strobing problem, it is not exactly a win, because the loss of detail is even greater than what would result from full-persistence, rapid-pixel-switching judder alone. | ||
|
Cyro
United Kingdom20278 Posts
You'd want the right side image, on a normal display. You're pretty much the only person i've heard of who thinks that, lol - the example on the right makes it simply impossible to identify moving targets, read moving text, in general left just allows motion without completely borking up the sharpness of >everything on the display<. You really have to see it to get the point - and by 120fps/120hz, there is no concerning loss of realism from not blurring things wildly. It's an exaggerated example because of stuff you quoted but it's a real example so i thought it relevant The original point of this discussion is about vsync/triple buffering etc I thought there were two discussions~ We have different views on how triple buffering works. This is probably because it doesn't always work the same way :D In computer graphics, triple buffering is similar to double buffering but provides a speed improvement. In double buffering the program must wait until the finished drawing is copied or swapped before starting the next drawing. This waiting period could be several milliseconds during which neither buffer can be touched. In triple buffering the program has two back buffers and can immediately start drawing in the one that is not involved in such copying. The third buffer, the front buffer, is read by the graphics card to display the image on the monitor. Once the monitor has been drawn, the front buffer is flipped with (or copied from) the back buffer holding the last complete screen. Since one of the back buffers is always complete, the graphics card never has to wait for the software to complete. Consequently, the software and the graphics card are completely independent, and can run at their own pace. Finally, the displayed image was started without waiting for synchronization and thus with minimum lag.[1] Due to the software algorithm not having to poll the graphics hardware for monitor refresh events, the algorithm is free to run as fast as possible. This can mean that several drawings that are never displayed are written to the back buffers. This is not the only method of triple buffering available, but is the most prevalent on the PC architecture where the speed of the target machine is highly variable. Another method of triple buffering involves synchronizing with the monitor frame rate. Drawing is not done if both back buffers contain finished images that have not been displayed yet. This avoids wasting CPU drawing undisplayed images and also results in a more constant frame rate (smoother movement of moving objects), but with increased latency.[1] This is the case when using triple buffering in DirectX, where a chain of 3 buffers are rendered and always displayed. The first method described looks at double buffer's latency and says >how can we fix this?< By flipping buffers 2 and 3 and constantly rendering onto them between screen refresh periods (instead of filling buffer 2, locking it and waiting a long time for the screen refresh) then that means buffer 2 or 3, whichever is not being worked on, likely has a more recent frame rendered and ready to be displayed than if you had simply rendered once and locked the buffer. Instead of rendering once, as soon and as quickly as possible, then holding that old frame until the screen is ready - this method uses the GPU and constantly renders to another buffer, flipping it onto the "ready to be displayed" buffer every time it has a new, complete frame. This means if you are capable of running the game at 600fps, with double buffered 60hz v-sync, your GPU would be at 10% average load. It would throw out a frame in ~1.67ms and then your input would be sat in a locked buffer for ~15ms with the GPU idle. With triple buffering using this method, the GPU would stay at 100% load, throwing out that 600fps - 10 frames per 60hz refresh - into alternating back buffers. The screen would update and grab a new frame, at which point input would be a couple milliseconds old on the last frame which was rendered in ~1.67ms, hitting the buffer with a couple milliseconds between input and screen refresh cycle start as opposed to ~15ms. Do you understand that concept? I'm not sure if i'm explaining well enough, or how else to explain it -- The second method described just adds another frame in rendering buffer - The goal of it is >not< to reduce latency, it is to increase stability of performance when meeting every refresh is not guaranteed, by allowing the GPU to start work on a third frame before the second one is displayed, etc. They are filled and locked in sequence. Input lag is equal at best, often worse. This seems to be what you were thinking of, though i would never want to use this~ | ||
|
OdinOfPergo
United States840 Posts
I'll start by avoiding normals like the plague (for the most part). I'm not really sure I'd understand the right scenarios for practicing them except for maby doing dt on higher bpm maps to work on speed/acc. So I feel like just not playing them and focusing on trying to work towards the same cause on hard maps will probably be better. I recognize the point for nf/ht and I guess I can force myself to use ht for now. My main qualm with it is the way it distorts the music. Think like a really terrible recording quality on a mp3. But I can diffidently see your point about how using it to give myself extra time to read/recognize patterns would be immensely useful while I'm trying to learn them. About mouse settings; I read before that 6/11, 800 dpi and 1:1 ratio through osu is the best setting to use. I actually did try this out but I find it completely unplayable. I feel like I have to "throw" my mouse to get it to move just about anywhere. I'm sure a good part of it is I've been using a much higher dpi setting for years now.. I bumped my dpi up to 1300 which is still really sluggish feeling to me but it's at least playable.. So how much of this is preference and what's actually going to hurt my play here though? | ||
|
Cyro
United Kingdom20278 Posts
Meanwhile for FPS players, using 5 inches to turn 180 degrees is considered quite a high sensitivity. It's all about mindset and what you are used to. To be honest if you're coming from high DPI, i would put in the extra effort to get used to lower, so you can switch around as you like. I went 3200 to 450 and then increased as i wanted to, afterwards. Take a look at this vid, skip a minute in: Not saying to use this sens (~400dpi) but 2x is very very hard to argue against. 3x that, max, i'd say is about where you want to reasonably stop | ||
|
Yorbon
Netherlands4272 Posts
Regarding dpi and such: aside from dpi, another relevant thing is your resolution. I play at 500 dpi on 1280 (width) resolution. This means i have to move my mouse 1280/500=2.56 inches to cross the screen. Assuming the same resolution, you'd need 1280/1300=around 1 inch to cross screen. To be very honest i personally think around 2 inches is near the minimum of what you want to have. My experience with high sensitivity (800/1000=0.8 inches) is that it's okay with small amounts of decent jumps, but becomes unplayable with very jumpheavy maps as well as small circles and spaced streams with directional variety. Unfortunately, going down in dpi was for me very hard compared to going up. Again assuming a resolution of 1280, i'd say 1300 is too high in the long term. But the most important is the amount of movement needed to cross screen. Play around with that for a bit. | ||
|
OdinOfPergo
United States840 Posts
My res is 1920x1080. I think I wouldn't have as much of an issue with lower dpi settings if I had a bigger mousepad (it's seriously tiny x.x). When I was playing around with settings last night I found that even in the 11, 12, and 1300 range I still tend to run off the edge on accident from time to time. But I'm sure that will change as I get use to it. Anyway I'll watch the vid above now but ye I'll keep messing around with it. Ninja edit: playing without a mousepad is impossible btw. I have a glass desk.. | ||
|
maru~
2345 Posts
Also, I have never used HT mode so far. NF I found quite useful (getting used to speed, practicing streams) and it doesn't distort the song. I still rarelly used it though. | ||
|
OdinOfPergo
United States840 Posts
So with that I guess I'll just widdle myself backwards on sens. And ye I suppose I'll stop by somewhere tomorrow and buy a new mousepad. I'll really need one if I change my mouse sens back down. Because 5x7 just isn't going to cut it if I have to move my mouse over half that (assuming that I always start in the middle) to clear the screen. (also the dude in that vid is nuts. I can barely play osu using a keyboard let alone solo mouse x.x!!!!) But thanks for the tips again guys! I'll definitely work towards it. | ||
|
Cyro
United Kingdom20278 Posts
I grabbed a 45x40cm pad a few years ago for about equivalent of $10-15~ If you're describing yours in inches, that's about ~8x as much area | ||
|
OdinOfPergo
United States840 Posts
my previous house actually burnt so this was a free hand me down from my grandma's thrift shop. I just need to buy a better one already lol. | ||
|
maru~
2345 Posts
On March 20 2014 06:58 Yorbon wrote: I believe the highest x (?) scores add up to your total pp amount. So that means the amount of pp a score gives is the amount it gives minus the amount the lowest score in the top x gives, the latter of which will be kicked out of the top x. edit: ignore what i said. It's kind of similar to it, but the weights are listed at top ranks as well. They should be giving some insight. Yeah, there were no weights when I first checked. It makes sense now. | ||
| ||
