![]() |
#51
|
|||
|
|||
![]()
Well the rule of "the crapier the system is, the easier one will spot bandits" isn't true only because of the dots and pixels. Contrast plays a big part in telling shapes apart and the graphics filtering and effects work exactly in the other direction.
You probably heard that there is a study on color blind observers being able to spot camouflage easier than the normal observer. I guess we can compare that to AA filtering, the better the filtering, the hardest it is to detect edges - what plays a big part in telling shapes. Also, the visual acuity is dependent on the sharpness of the retinal focus within the eye, which can be greatly improved by a bigger screen (considering a source image of the same resolution) - bigger arc for the same image, better spatial resolution. Quote:
té mais tityus |
#52
|
|||
|
|||
![]()
IRL it is possible to see a small airplane at least 5 miles if contrast is good.
Their is a trade off in game between realistic view distance and finding your "fun". |
#53
|
|||
|
|||
![]()
I wonder (here I go again) if the single pixel goes from grey value 0-255 between say... distance 5.5miles to 4.5 miles (depending on visibility conditions of course).
|
#54
|
|||||
|
|||||
![]() Quote:
the way il2 currently works you have several mechanisms at work to display a progressively smaller aircraft on your screen: a) the aircraft itself that you can view in external view, with all its detail displayed in its full eye candy glory in closeup b) as you zoom away from the object this initial high detail representation of the airplane/ship gets replaced by a LoD model, a transition which occurs in 3 LoD steps right now, displaying progressively smaller LoD models with significantly less detail but keeping the rough shape of the aircraft. since the smaller LoD models are a cluster of pixels approximating the shape of the aircraft, c) for a small fighter aircraft, somewhere between 1200 and 1800 meters this smallest LoD model will be replaced by a "dot" (in 4.08 this was a 4 pixel square dot). d) for a large aircraft like a b17 this small dot will completely disappear somewhere around 5000 meters (depending on the il2 version you are using, the latest 4.09 has extended this visibility to 10.000 meters or even further iirc) So the problems we are dealing with in il2: 1) how well/bad distant "dots" are visible ? at original release in il2 (2001 ?) these dots in their smallest form were a block of 9 black pixels that stood out like the proverbial dog's balls. soon after this size was reduced, but it briefly increased again in patch 3.01 or 3.02 because people complained they could spot them well enough. but on release a small vocal minority whined they were to visible, and in the next patch oleg reduced it to a 4 pixel square of 2 black and 2 grey pixels, which stayed till il2 version 4.08 iirc. around the time oleg changed those dot sizes several times, people were transitioning from crt's to lcd's in the west, and this resulted in a very varied set of reporting of how visible/invisible the dots are. it is possible that from 4.09 this was further reduced to 2 pixels (i havnt tested it in 4.09 or 4.10). the context of these changes is that the earlier pc technology was not powerfull enough to display highly detailed LoD models for multiple aircraft very far out, and as a way to reduce cpu/gpu drain they transitioned to the Dot's we currently have. 2) how well/bad medium distance LoD models are visible (particularly the 2e and 3e smaller ones for more distant aircraft). these pixel clusters are flat little pixels on a 2D display medium (your monitor), the fact the distant object is the right size does NOT mean it is as visible as the same plane would be in real life circumstances. 3) how well/bad the end result situational awareness is of an il2 virtual flyer (sitting behind a small monitor in his living room), compared to a real ww2 pilot in the same situation looking at the same distant objects. a major part of this is indeed how accurately we can "spot" nearby threats and potential targets (by looking at some parts of the sky/ground) and how well we can keep track of them once visually acquired the presumption in a discussion like this is of course that we are trying to reproduce the visibility real pilots had during ww2, both of aircraft in the sky and for seeing ground targets (eg, we are NOT just playing a shoot'em up game of "hard to see and hard to find aircraft" where we believe the fact they are hard to find or nearly invisible when out further then 300 meters means this is realistic and represents the experience of real ww2 fighter pilots) another presumption is that virtual pilots have correctly callibrated their displays, and have setup the correct Field of View (FoV) for their monitor sizes (which is rarely the case). lastly, they need to have reasonably normal vision to start out with, and have compared il2 being displayed on a few different mediums (like crt, cheap TN lcd monitors, normal average quality MVA/PVA or IPS displays, etc..). note: there are now newer oled screens, or people using large flatscreen tv's or projectors, but those are not commonly used yet. lastly people need to be honest and objective about reporting what they see in those comparisons. Quote:
Quote:
Quote:
Quote:
ideally you'd want a dedicated detailed thread that looks at the technical issues behind this visibility problem in il2, so we know exactly what needs to be remedied. i only know part of the answers, but i'll try and make some specific comments in some of the posts in this thread to clarify the parts i do know about Last edited by zapatista; 02-19-2012 at 05:30 AM. |
#55
|
||||
|
||||
![]() Quote:
once that smallest LoD model transitions to a dot however, the dot itself is determined by pixel size, so having larger pixels will still significantly increase visibility. if for fighters this is at 7000 to 10000 meters, it matters much less then at the 1500 meters distance it is at now, because that single dot flying at 400 km/hr towards you can very quickly become a danger if he sees you much earlier then you can spot him. sadly is suspect that for small fighters oleg still will allow the dot-to-LoD transition point to occur somewhere around 1500 to 2000 meters (because that is where in size it becomes around 2 pixels big so it would be a potential pc resource waste to keep trying to draw it in detail if it is actually smaller then 2 pixels). so to the resolution based cheat and visibility problem will likely continue (hope i am wrong there) Last edited by zapatista; 02-11-2011 at 07:49 AM. |
#56
|
||||
|
||||
![]() Quote:
if your system is callibrated correctly and your FoV is setup ok, i suspect you dont have a TN based screen |
#57
|
||||
|
||||
![]() Quote:
its one of the few times in life cheaper is better (for il2). the little grey/black dots of distant objects glitter and shine on your screen, making them stand out more (compared to others with 8 bit monitors seeing a single shade of grey slightly change as the dots glide over green/brown/white/grey backgrounds off the underlying terrain) |
#58
|
||||
|
||||
![]()
foo'bar,
i looked up your monitor model, it is indeed again a 6 bit monitor (see http://www.prad.de/en/index.html ). given how much you know about grafix and video and your liking of il2, and since you use programs like photoshop etc (?) i am a bit surprised you were not aware of that when you selected your monitor. as you probably know "eight bits" also allows 256 shades of grey. The human eye can discern about 85 intensity levels on a good monitor (high dynamic range, which means the black is very black and the white is very bright). these extensive "steps" in grey levels allows a very smooth transition in black/greys and offers much better detail when viewing video and grafix when lower end displays are only 6 bit, the shades of colors available to try and accurately depict the exact shade of color it needs to display is significantly reduced, and they will use "dithering" to approximate the color or grey tone as much as possible. this is a process by which computers approximate the display of colors in an image that are not available, and this is achieved by varying the patterns of dots that make up the image. like this now if this is for a static image being displayed in printed media for example, it might not matter that much because the blended dot of black and white are displayed a static snapshot, and they very small and the detail might be hard to see with the naked eye when reading a paper or looking at a photo. but if the grey/black dot is moving while this dithering is continuously occurring and you have a process like dynamic dithering which is constantly changing (because it cant seem to make up its mind what shade of brey/black to display), then suddenly it makes these dots stand out much more. the 2 images below illustrate the effect that enhances "dot visibility" for 6 bit monitor users in the il2 flightsim series. in the above picture note the smaller square in the bottom right corner, giving a zoomed and enhanced view of what produces the "glittering" effect (as dithering artifact) this 2e illustration shows a slightly different 6 bit dithering technology used on some other monitor pannels, but the overall effect produces a similar result and that is the most likely explanation why people with 6 bit lcd monitors consistently keep reporting much better dot spotting in il2, it is a case of the cheaper and nastier the monitor the more clearly you can see the dot's because of the artifacts the 6 bit display causes when you have a little 4 pixel block of dithering grey sliding over a static green/brown/white/blue background. it is only after over the years noticing that people with 6 bit monitors had significantly less problems with dot spotting that this technological issue about dithering was identified as a likely explanation. when previously polls were done to see who could see dots the best, the 6 bit monitor issue was the most common denominator (presuming all monitors are correctly calibrated, viewers look at the same scenery and object, etc..).the above is a very simplified explanation for a complex issue, and it is further complicated by a varied range in technologies used in modern panels, and the fact some brands use misleading advertising and labeling their products (or dont disclose when a panel in a particular model changes). note: for those interested in finding out what technology their own monitor uses, most manufacturers dont advertise the tech details very openly (partic for the lower end models) . a good site like http://www.prad.de/en/index.html will have most of the required detail listed for many models. the below table gives a rough idea of different technologies and if it is 6 or 8 bit, it helps to describe what we are seeing from most modern panels: All TN Film panels = Dithering. Some are 6-bit with FRC, some are 6-bit extended to 9-bit and figure as 8-bit in the specs (or 16.7 million colours). More modern panels seem to be the latter, look for "16.7 million colours" quoted in specs. PVA Traditional = 8-Bit PVA + Overdrive = Dithering, 6-Bit +FRC S-PVA + Overdrive = Still real 8-bit MVA Traditional = 8-Bit 19" and below MVA + Overdrive = dithering, but not as obvious as with PVA + Overdrive >19" MVA + Overdrive = real 8-bit S-IPS Traditional = 8-Bit New S-IPS panels + Overdrive = No obvious issues AS-IPS = 8 bit, not heard any reports of colour issues on these for those who want a simple quick test to see in living color how good/bad their monitor is in this 6/8 bit debate, look for the "color gradient test" (a little exe) from this website. the article itself also explains this complex issue a little more, and is one of the better one available. http://www.tftcentral.co.uk/articles.../6bit_8bit.htm . anybody using a 6 bit monitor will see significant banding, while most 8 bit monitor users wont. Last edited by zapatista; 02-20-2011 at 02:37 AM. |
#59
|
|||
|
|||
![]()
I use a Samsung 22-inch LCD with a tn panel. Been aware of the 6-bit issue for some time (but unfortunately not before buying the Samsung). Does anyone know if the banding that is visible on sea and sky when running il-2 on nVidia graphics has anything to do with 6-bit panels? Was puzzled by it - as i was running 32-bit colour, never could understand it and put it down to some nVidia glitch?
I've been noticing some of the new cheaper ips panels that are starting to come on the market. Was also keeping an eye on the progression of 120Hz technology but it seems to be painfully slow. Are current ips response times able to run games (flightsims in particular) without noticeable smearing or other effects? Also read something recently about what could be the Holy Grail - ips panels with 120hz refesh rate. Last edited by kendo65; 02-11-2011 at 01:08 PM. |
#60
|
||||
|
||||
![]() Quote:
the first 2 lcd's i ever bought yrs ago i returned the next day for an exchange, they were so bad compared to the professional grade crt i had been using before. it's after that i started looking into the how and why of the technology involved. both subsequent flatscreen monitors i bought in following years i was very happy with Quote:
these are some of the best monitor review sites i know http://www.prad.de/en/monitore/buyers-guide/start.html and their buyers guide is very good http://www.tftcentral.co.uk/articles...chnologies.htm http://www.flatpanelshd.com/focus.ph...&id=1229341472 |
![]() |
|
|