最新版はこちら。 突っ込みは各日付の BBS エントリのほか、 メール (nakano@st.seikei.ac.jp) や フォーム からどうぞ。 なおスパム除けのため、BBS 機能には 緩い認証を入れて います。 検索エンジンから来た方は、エンジンの方のキャッシュを見るか、 下の簡易検索を試してみてください。
|
Namazu for hns による簡易全文検索 詳しくは 詳細指定/ヘルプを参照して下さい |
||||||||||||||||||||||||||||||||||||||||||||||||
あなたがゲーム番組へ出ていて、3つのドアのうち1つを選ぶとします。1つのドアの後ろには車があって、あとの2つのドアの後ろにはヤギがいます。うーん、いまいち納得できん。ほんとに等価か? というわけでちと google ってみた。
あなたはドアを1つ、たとえば1番のドアを選んだとします。番組の司会者は残ったドアの1つ、例えば3番のドアを開けます。司会者は、それぞれのドアの後ろに何があるのかを知っています。3番のドアにはヤギがいました。ここで司会者はあなたに、「2番のドアに変えますか」と聞きます。
さて、2番のドアに変えた方がいいでしょうか。
この問題は、
A君とB君がいます。A君は素直に1番のドアを選びました。しかし、B君は残りの2つのドアを選ぶと言い張ります。揚句の果てに、司会者に1つに決めるから、ヤギの入っているドアを先に教えてくれと頼む始末でした。司会者はいい人で、3番のドアを開けてヤギが入っていることを教えてくれました。B君は2番のドアを選んで、こう言いました。「これでA君と同じように3つの中から1つのドアを選んだんだ。公平だろ」と言いました。
あなたは、A君の立場を選びますか、B君の立場を選びますか。
という話と等価である。
B君の立場の方が有利であることは言うまでもない。
「選択を変えない場合に当たる確率は1/3。選択を変えてハズレるのは最初に当たりを引いていた場合で,それ以外は当たるわけだから選択を当たる確率は1-1/3=2/3となる。したがって選択を変えた方が有利」j
結局,答えを知っている相手(司会者)による選別を経たほうが確率が高いということだ.自分が最初に選んだ一つだけは相手の選別の対象となっていないので,確率が上昇しない.
-N Do not execute a remote command. This is useful for just forwarding ports (protocol version 2 only).
We report on experiments on vertically shaken binary granular mixtures, which separate into their components due to the external excitation. This well-known phenomenon, where large particles rise to the top of the mixture, is called the Brazil-nut effect. Recent theoretical findings predict also a reverse Brazil-nut effect, where large particles sink to the bottom of the container. We choose spherical beads of various diameters and materials in order to observe the transition from Brazil-nut effect to its reverse form. The direction of demixing depends sensitively on the external excitation, so that it is possible to switch between both effects for a given mass density ratio.実験報告。おお。
The scaling exponents α, β, and 1/z in thin films of the organic molecule diindenoperylene deposited on SiO2 under UHV conditions are determined. Atomic-force microscopy, x-ray reflectivity, and diffuse x-ray scattering were employed. The surface width displays power law scaling over more than 2 orders of magnitude in film thickness. We obtained α= 0.684±0.06, β= 0.748±0.05, and 1/z = 0.92±0.20. The derived exponents point to an unusually rapid growth of vertical roughness and lateral correlations. We suggest that they could be related to lateral inhomogeneities arising from the formation of grain boundaries between tilt domains in the early stages of growth.
Optical properties of ultrathin layers (<50 Å) of Au and Pb quench condensed on amorphous germanium (α-Ge) have been measured in situ at 10 K. The development of these films from an insulating state to a metallic state is traced as a function of the film thickness as well as the sheet resistance, R□. Of particular interest is the regime of R□ near 3000 Ω where there is an anomaly in the optical transmission. This anomaly is due to a singularity in the dielectric function when the system undergoes an electronic percolation or insulator-to-metal (I/M) transition.
Nanoindentation of crystalline and amorphous silicon nitride films is studied using 10-million-atom molecular dynamics simulations. A rigid pyramid-shaped indenter tip is used. Load-displacement curves are computed and are used to derive hardness and elastic moduli of the simulated crystalline and amorphous films. Computer images of local pressure distributions and configuration snapshots show that plastic deformation in the film extends to regions far from the actual indent.力技。
Si nanoclusters have been formed by 5 keV Si+ implantation at a fluence of 1×10^16 atoms/cm2 into a 200 Å thin thermally grown SiO2 film on Si (100), followed by thermal treatment at 1000 ℃ with different annealing times. All the annealed samples show a broad photoluminescence spectrum with increasing intensity as function of annealing time. The use of a dual beam time of flight secondary ion mass spectrometry in negative mode with Cs+ ions at low energy for sputtering allows us to observe variations in Si_n- signal due to excess of silicon atoms introduced by implantation. With the high sensitivity achieved using this instrumental configuration it is possible to follow Si signals which give information about the chemical enviroment of the Si atoms. The possibility of studying the time evolution of the nucleation and growth of nanoclusters has been investigated.Sin- クラスターのシグナルがミソ。
The built-in electrical potential of Cu(In,Ga)Se2 (CIGS) solar cells was measured quantitatively and resolved spatially using scanning Kelvin probe microscopy. Profiles of the electrical potential along cross sections of the device demonstrate that the p?n junction is a buried homojunction, and the p/n boundary is located 30?80 nm from the CIGS/CdS interface in the CIGS film. The built-in electric field terminates at the CIGS/CdS interface, indicating that the CdS and ZnO layers of the device structure are inactive for the collection of photoexcited carriers.
The throw distance of Al and W atoms was determined (A) by investigating the pressure-dependent deposition rate and applying the Keller-Simmons one-dimensional model. The power density at the target as a parameter was varied over more than one order of magnitude. Whereas the values determined for Al are not influenced by power, the values for tungsten strongly increase with power. This effect is related to strong gas density reduction related to gas heating for W sputtering, which reduces the probability of collisions between sputtered atoms and the gas. The gas heating was experimentally determined by simultaneous measurements of the pressure in the sputtering chamber and at the end of a tubular probe inserted into the hot plasma region. For tungsten, a strong heating and related gas density reduction was found. An analytical expression describing the spatially resolved temperature profile in a sputtering discharge is developed by means of the Keller-Simmons formulas. The influence of various process parameters on gas heating can be easily studied by the model. There is good agreement between the model calculations with the experiments performed as well as with Monte Carlo simulations and experimental findings from the literature. Forward simulations by the model result in a power-independent throw distance (B) for both elements, which are in agreement with the values (A) resulting from the pressure-dependent deposition rate. Existing differences are explained by the rough assumptions of the model, which is critically discussed.Al だとほとんど効果がないというのがちょっと意外。
By chemisorption of oxygen on an aluminum target, an absolute target voltage increase is noticed which can be attributed to the reduction of the secondary electron emission from the target. By following the target voltage behavior during magnetron sputtering of the aluminum target precovered with a chemisorbed oxygen layer, the cross section for removing the chemisorbed layer could be calculated based on a proposed model that relates the measured target voltage to the target coverage.
Using an energy selective mass spectrometer, ion energy distribution functions (IEDFs) in a low pressure radio frequency (rf) argon discharge have been measured. This experiment has been done in the diffusion chamber of a helicon source. Two kinds of IEDF have been observed depending on the gas pressure: classical "one-peak" spectra at high pressure (>1 μbar) and "two-peak" spectra at low pressure (<1 μbar). These results have been correlated with the plasma potential and density profiles along the axis of the reactor obtained with a Langmuir probe and the IEDF have been fitted using a simple analytical model. In this way, noncollisional diffusion of energetic ions from the source chamber to the diffusion chamber has been shown.