JP2009116550A - Plotting processor, plotting processing method and plotting processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain an image with little distortion, and to simplify an arithmetic operation. <P>SOLUTION: This plotting processor 1 is provided with a first selection part 2; a second selection part 3; and an arithmetic part 4. The first selection part 2 compares variation (du/dx) with variation (dv/dy), and selects the larger value. The second selection part 3 compares variation (du/dy) with variation (dv/dx), and selects the larger value. The arithmetic part 4 considers the value selected by the first selection part 2 and the value selected by the second selection part 3 as perpendicularly crossing two sides of a right triangle, and substantially calculates the value of the oblique side of the right triangle. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a drawing processing device, a drawing processing method, and a drawing processing program, and more particularly to a drawing processing device, a drawing processing method, and a drawing processing program for acquiring a texture of an optimal size from a texture storage unit that stores a mipmap.

  Texture, which is a technique for improving the expressive power of texture and the like by pasting a two-dimensional image (texture) of a picture or photograph on the surface of a three-dimensional figure to perform realistic drawing in three-dimensional drawing of computer graphics Mapping methods are known (see, for example, Patent Documents 1 to 3 and Non-Patent Document 1).

  In three-dimensional drawing, there is a case where a figure is deformed and drawn in order to draw a small figure to express that the drawn figure is far away or to express that the figure is inclined (see, for example, Patent Document 1). ). In this case, aliasing occurs because the drawing figure and the size of the texture to be pasted do not match. In order to prevent aliasing, the mipmap (Mipum In Parvo) Map (Mipum In Parvo Map) created by repeatedly reducing the original texture size to 1/2 in both length and width is stored in the texture storage unit in advance, and the optimum A LoD (Level of Detail: reduction ratio) calculation for obtaining a texture of a size is performed (for example, see Non-Patent Document 1).

  The LoD of the texture indicates how much the original image has been reduced when texture mapping is performed. The original image is reduced to 1/1, 1/2, 1/4,. The relationship with LoD is as shown in the following equation (1) (see, for example, Patent Document 1).

1/1: LoD = 0
1/2: LoD = 1
1/4: LoD = 2
1/8: LoD = 3
:
1 / ρ: LoD = log ₂ (ρ) (1)
Moreover, ρ is expressed by the following equation (2).

Here, “max []” means “select the larger one of the mathematical expressions delimited by commas in parentheses ([])”, and (du / dx), (dv / dx), (du / dy), and (dv / dy) indicate the amount of change (displacement amount) of the texture coordinates (U, V) with respect to the pixel (X, Y) in the XY coordinates. du / dx) indicates that the edge element dx in the X-axis direction of the pixel of the pixel (X, Y) is changed by u in the U-axis direction in the texture coordinates (U, V). Therefore, the amount of change in dx on the XY coordinates is the left-side mathematical formula delimited by the commas in parentheses in the above formula (2). The amount of change in dy is a mathematical expression on the right side separated by commas in parentheses in the above formula (2).
JP 2000-155850 A JP 2001-195602 A Japanese translation of PCT publication No. 2002-504251 Mark Segal., Kurt Akeley., "The OpenGL Graphics System: A Specification". Version2.1- December1, 2006. P173, Scale Factor and Level of Detail. [Retrieved on 2007/11/5] Retrieved from the Internet: < URL: http://www.opengl.org/registry/doc/glspec21.20061201.pdf>

However, in the formula (2), there is a problem that a normal LoD cannot be obtained near the value in the parenthesis is switched to either, and the drawn image is distorted.
An example in which distortion occurs in an image is shown.

FIG. 9 is a diagram showing a figure by a conventional drawing method.
FIG. 9 shows a figure having a viewpoint at the center of the tunnel image 91. Since the viewpoint is at the center of the tunnel image 91, the side surface of the tunnel image 91 is equidistant from the center. Since it becomes smaller in the perspective as it is closer to the inner side (the center of the image) of the tunnel image 91, LoD becomes larger. That is, the outside of the texture 92 constituting the tunnel image 91 is a texture with LoD = 0, the texture 92 is a texture with LoD = 1, the texture 93 is a texture with LoD = 2, and the texture 94 is The texture is LoD = 3, and the texture 95 is a texture with LoD = 4.

  Since the side of the tunnel is equidistant from the center, the color of the mipmap selected by LoD should draw a concentric circle, but in the vicinity where the value at “max []” shown in Equation (2) switches to either (in FIG. 9, Distortion occurs in the portion surrounded by the thick line.

In addition, in the formulas (1) and (2), there is a problem that processing by hardware becomes enormous because there is a square root calculation.
SUMMARY An advantage of some aspects of the invention is that it provides a drawing processing apparatus, a drawing processing method, and a drawing processing program capable of obtaining an image with less distortion.

  It is another object of the present invention to provide a drawing processing apparatus, a drawing processing method, and a drawing processing program capable of facilitating calculations as other purposes.

In the present invention, in order to solve the above problem, the following drawing processing apparatus is provided.
The drawing processing apparatus according to the present invention is an apparatus that acquires a texture having an optimum size from a texture storage unit that stores a mipmap.

The first selection unit compares the change amount (du / dx) of the texture coordinates (U, V) with respect to the pixel (X, Y) and the change amount (dv / dy), and selects the larger value. .
The second selection unit compares the change amount (du / dy) of the texture coordinate (U, V) with respect to the pixel (X, Y) described above and the change amount (dv / dx), and determines the larger value. select.

  The calculation unit regards the value selected by the first selection unit and the value selected by the second selection unit as two sides that intersect the right angle of the right triangle, and substantially determines the value of the hypotenuse of the right triangle. Calculate.

  According to such a drawing processing device, the first selection unit selects the larger value of the change amount (du / dx) and the change amount (dv / dy). In addition, the second selection unit selects the larger value of the change amount (du / dy) and the change amount (dv / dx). Then, the value selected by the first selection unit and the value selected by the second selection unit are regarded as two sides intersecting at right angles of the right triangle, and the value of the hypotenuse of the right triangle is calculated by the calculation unit. It is virtually calculated.

Further, in order to solve the above-described problem, in a drawing processing method for obtaining a texture of an optimal size from a texture storage unit storing a mipmap, a change amount of texture coordinates (U, V) with respect to a pixel (X, Y) ( du (dx), change amount (dv / dx), change amount (du / dy), change amount (dv / dy), {(du / dx) ² + (du / dy) ² } ^1/2 and { A first step of selecting the larger value of (dv / dx) ² + (dv / dy) ² } ^1/2 and {(du / dx) ² + (dv / dx) ² } ^{1 / 2} and {(du / dy) ² + (dv / dy) ² } ^1/2 , the second step of selecting the larger value, the value selected by the first step and the second The larger one of the values selected in step, the texture pixel size and the rendered image And a third step of setting the ratio to the pixel size of the drawing.

  According to such a drawing processing method, calculation can be facilitated.

  According to the present invention, it is possible to draw a figure with less distortion. Further, the calculation can be facilitated.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
First, an outline of the present invention will be described, and then an embodiment will be described.
FIG. 1 is a diagram showing an outline of the present invention.

The drawing processing apparatus 1 includes a first selection unit 2, a second selection unit 3, and a calculation unit 4.
Hereinafter, the U-axis direction constituting the UV coordinate system of the texture to be pasted to the graphic of the side element dx in the X-axis direction of the unit pixel (X, Y) having the coordinate to which the graphic to be textured is mapped as the XY coordinate The amount of change is the amount of change (du / dx), the amount of change in the V-axis direction is the amount of change (dv / dx), and the amount of change in the Y-axis direction dy of the unit pixel (X, Y) in the U-axis direction Is the amount of change (du / dy), and the amount of change in the V-axis direction is the amount of change (dv / dy).

The first selection unit 2 compares the change amount (du / dx) and the change amount (dv / dy), and selects the larger value.
The second selection unit 3 compares the change amount (du / dy) with the change amount (dv / dx), and selects the larger value.

  The calculation unit 4 regards the value selected by the first selection unit 2 and the value selected by the second selection unit 3 as two sides that intersect each right angle of the right triangle, and determines the value of the hypotenuse of the right triangle. Operate virtually. The value of the hypotenuse may be obtained by calculating the square root, but is preferably obtained using an approximate expression. This makes it easy to calculate.

  According to such a drawing processing apparatus 1, the first selection unit 2 selects the larger value of the change amount (du / dx) and the change amount (dv / dy). Further, the second selection unit 3 selects the larger value of the change amount (du / dy) and the change amount (dv / dx). Then, the calculation unit 4 regards the value selected by the first selection unit 2 and the value selected by the second selection unit 3 as two sides intersecting at right angles of the right triangle, and the hypotenuse of the right triangle Is substantially calculated.

By calculating the LoD using the calculated hypotenuse value by such a calculation means, it is possible to easily obtain the texture of the optimum size.
Embodiments of the present invention will be described below.

FIG. 2 is a block diagram illustrating a configuration of the drawing apparatus.
The drawing processing apparatus 10 includes a CPU (Central Processing Unit) 20, a GPU (Graphics Processing Unit) 30, an external memory 40, and a monitor (screen) 50.

The CPU 20 transmits the vertex coordinates (three-dimensional coordinates) of the three-dimensional object data to be drawn and the corresponding parameters to the GPU 30.
The GPU 30 includes a geometry 31, a rasterization unit 32, a LoD calculation unit 33, a texture acquisition address generation unit 34, and a pixel blender 35.

The geometry 31 receives the vertex coordinates from the CPU 20 and the corresponding parameters.
In the geometry 31, the vertex coordinates of the image on which the object is projected onto the monitor 50 at the viewpoint angle are calculated and the corresponding parameters are transmitted to the rasterizing unit 32.

  The rasterizing unit 32 rasterizes the received vertex coordinates to obtain each coordinate (XY coordinate and UV coordinate), and transmits the XY coordinate and the UV coordinate to the LoD calculation unit 33 among the obtained coordinates. The UV coordinates are transmitted to the texture acquisition address generator 34, and the XY coordinates are transmitted to the pixel blender 35.

The LoD calculation unit 33 obtains LoD from the XY coordinates and the UV coordinates, and transmits the LoD to the texture acquisition address generation unit 34.
The texture acquisition address generation unit 34 generates an address for acquiring a texture based on the UV coordinates and the LoD, and transmits the generated address to the external memory 40.

The external memory 40 includes a frame buffer 41 and a texture storage unit 42.
When the texture storage unit 42 receives the address, the texture storage unit 42 generates a texel constituting the texture and transmits the texel to the pixel blender 35.

  The pixel blender 35 can process (change colors or blend with other colors) texels with various settings, and writes the processed texels to the frame buffer 41 corresponding to the XY coordinates.

The monitor 50 displays the contents stored in the frame buffer 41 as pixels. Thereby, an image is obtained.
With the configuration as described above, the processing function of the present embodiment can be realized.

Next, the configuration of the LoD calculation unit 33 will be described in detail.
The LoD calculation unit 33 calculates the larger of the two values in “max []” in order to compensate for the switching of “max []” in the conventional equation (2) where normal LoD calculation is not performed. Instead of selecting, the value ρ1 of the hypotenuse of a right triangle having two values orthogonal to each other is obtained, and an operation of setting log ₂ (ρ1) to LoD of this embodiment is performed. That is, two orthogonal sides are expressed by the following equations (3), (4),

  The value ρ1 of the hypotenuse of a right triangle is expressed by the following equation (5).

  Further, when Expressions (3) and (4) are substituted into Expression (5), the following Expression (6) is obtained.

  At this time, as illustrated in FIG. 9, the amount of change increases as the depth increases as the figure of the tunnel image 91 increases. On the other hand, the amount of change in the direction along the cylindrical outer peripheral surface is not so large. Therefore, the following equations (7) and (8) are obtained from the equation (6) by ignoring the smaller change amount of u and v.

The larger value of Equation (7) and Equation (8) is selected.
ρuv = max [ρu, ρv] (9)
Formula (2) is changed into Formula (10) below using Formula (3) and Formula (4).

ρxy = max [ρx, ρy] (10)
Conventional distortion is caused by low LoD. As shown in the following expression (11), by selecting the larger value in each of the expressions (9) and (10), the expression (10) is used for an image without distortion, and an image with distortion is obtained. Then, distortion can be corrected by using equation (9).

ρ1 = max [ρxy, ρuv] (11)
However, the magnitude of the change amount (du / dx) and the change amount (dv / dy), the change amount (du / dy), and the change amount (dv / dx) can be obtained without obtaining and directly comparing ρxy and ρuv. )), The largest value among the formulas (3), (4), (7), and (8) can be determined. Therefore, the hypotenuse value ρ1 can be easily calculated. it can.

FIG. 3 is a flowchart showing a calculation algorithm of the LoD calculation unit.
First, it is determined whether or not the change amount (du / dy) is larger than the change amount (dv / dx) (step S1).

  When the amount of change (du / dy) is larger than the amount of change (dv / dx) (Yes in step S1), it is determined whether the amount of change (du / dx) is larger than the amount of change (dv / dy) (step). S2).

When the amount of change (du / dx) is larger than the amount of change (dv / dy) (Yes in step S2), ρ1 = ρu is set (step S3).
If the amount of change (du / dx) is equal to or less than the amount of change (dv / dy) (No in step S2), ρ1 = ρy is set (step S4).

  On the other hand, if the change amount (du / dy) is equal to or less than the change amount (dv / dx) in step S1 (No in step S1), whether or not the change amount (du / dx) is greater than the change amount (dv / dy). Is determined (step S5).

When the amount of change (du / dx) is larger than the amount of change (dv / dy) (Yes in step S5), ρ1 = ρx is set (step S6).
If the amount of change (du / dx) is equal to or less than the amount of change (dv / dy) (No in step S5), ρ1 = ρv is set (step S7).

Next, the configuration of the LoD calculation unit 33 that realizes such a calculation algorithm will be described.
FIG. 4 is a block diagram illustrating a configuration of the LoD calculation unit.
The LoD calculation unit 33 includes a change amount measurement unit 331, comparison units 332 and 333, a magnitude comparison unit 334, a calculation unit 335, and a logarithmic calculation unit 336.

  The change amount measuring unit 331 includes a change amount (du / dx), a change amount (dv / dx), a change amount (du / dy), and a change amount (dv) of the texture coordinates (U, V) with respect to the pixel (X, Y). / Dy).

The comparison unit 332 compares the amount of change (du / dx) and the amount of change (dv / dy), and outputs the larger value (hereinafter referred to as value α).
The comparison unit 333 compares the amount of change (du / dy) and the amount of change (dv / dx), and outputs the larger value (hereinafter referred to as value β).

  As a result, the operations corresponding to steps S1, S2 and S5 shown in FIG. 3 are executed, and the amount of change (du / dx), amount of change (dv / dx), and change necessary to obtain the hypotenuse value ρ1. A value α and a value β corresponding to any two of the amount (du / dy) and the change amount (dv / dy) are obtained.

The magnitude comparison unit 334 compares the value α and the value β, and outputs a larger value (hereinafter referred to as a value γ) and a smaller value (hereinafter referred to as a value δ).
The computing unit 335 obtains the hypotenuse value ρ1 with the values γ and δ determined by the magnitude comparison unit 334 as inputs.

The logarithmic operation unit 336 calculates log ₂ (ρ1) using the hypotenuse value ρ1 obtained by the operation unit 335 to obtain LoD.
Next, the calculation unit 335 will be described in detail. First, the calculation method of the calculation unit 335 will be described, and then the configuration of the calculation unit 335 will be described.

  As described above, the value of the hypotenuse ρ1 is any one of ρu, ρy, ρx, and ρv. However, if any of these values is calculated, a square root operation is performed. Therefore, in order to simplify the calculation, the calculation unit 335 performs a polygonal line approximation calculation including a linear expression in which the value γ is regarded as a fixed value and the value δ is regarded as a variable. Specifically, the following equation (12) is approximated.

FIG. 5 is a diagram for explaining how to obtain the coefficients in the approximate expression.
In the present embodiment, different coefficients 1 and 2 are selected at a ratio of the value δ to the value γ, and are substituted into the equation (12), thereby obtaining the hypotenuse value ρ1. Specifically, an area corresponding to the ratio of the value δ to the value γ is prepared. Next, the region to which the value δ belongs is determined according to the ratio of the value δ to the value γ. Then, coefficient 1 and coefficient 2 for optimal approximation are obtained for each region to which the value δ belongs.

In FIG. 5, as an example, the region is divided into four (2 ² ), and if the value δ is less than ¼ of the value γ, it is determined that the region belongs to region # 1. If the value δ is ¼ or more than the value γ and less than 2/4, it is determined that the value δ belongs to the region # 2. If the value δ is 2/4 or more of the value γ and less than 3/4, it is determined that it belongs to the region # 3. If the value δ is 3/4 or more of the value γ and less than 1, it is determined that it belongs to the region # 4. Then, coefficient 1 and coefficient 2 of each region are obtained by calculation, selection from a table prepared in advance, or the like. Thereby, an optimal approximate expression can be selected for each region.

In this way, by adopting different approximation formulas depending on the ratio of the value δ to the value γ, high-precision approximation can be performed.
Next, the configuration of the calculation unit 335 will be described.

FIG. 6 is a block diagram illustrating a configuration of the calculation unit.
The calculation unit 335 includes a coefficient selection unit 335a, a storage unit 335b, and a ρ1 calculation unit 335c.

Note that the calculation unit 335 shown in FIG. 6 has a configuration corresponding to FIG.
The coefficient selection unit 335a generates an address (selection address) for selecting a coefficient at a ratio of the value δ to the value γ. In the present embodiment, if the value δ is less than ¼ of the value γ, a selection address specifying the region # 1 is generated. If the value δ is equal to or greater than ¼ of the value γ and less than 2/4, a selection address designating the region # 2 is generated. If the value δ is 2/4 or more of the value γ and less than 3/4, a selection address for designating the region # 3 is generated. If the value δ is 3/4 or more of the value γ and less than 1, a selection address for designating the region # 4 is generated.

The storage unit 335b includes a coefficient selection table 3351 and a selector 3352.
The coefficient selection table 3351 has coefficient blocks b1 to b4 each including a combination of coefficient 1 and coefficient 2 having different values for each region. That is, the number of coefficient blocks stored in the coefficient selection table 3351 is equal to the number of divided areas.

  Here, the coefficient blocks b1 to b4 have coefficients 1-1 to 1-4 having different values as the coefficient 1, and coefficients 2-1 to 2-4 having different values as the coefficient 2, respectively. Yes.

  The selector 3352 selects any one of the coefficient blocks b1 to b4 based on the selected address. Specifically, the coefficient block b1 is selected if it is a selection address designating the region # 1. If the selected address designates the area # 2, the coefficient block b2 is selected. If the selected address designates the area # 3, the coefficient block b3 is selected. If the selected address designates the area # 4, the coefficient block b4 is selected.

Then, the coefficient 1 and the coefficient 2 included in the selected coefficient block are output to the ρ1 calculation unit 335c.
The ρ1 calculation unit 335c obtains the hypotenuse value ρ1 by substituting into the equation (12) using the obtained coefficient 1, coefficient 2, value γ, and value δ.

Next, the logarithmic operation unit 336 will be described in detail. First, the calculation method of the logarithmic calculation unit 336 will be described, and then the configuration of the logarithmic calculation unit 336 will be described.
The obtained hypotenuse value ρ1 can be replaced by the following equation (13).

ρ1 = (1 + a) × 2 ^b (13)
Further, as described above, LoD can be obtained by the following equation (14).
LoD = log ₂ (ρ1) (14)
Here, when Expression (13) is substituted into Expression (14), it can be expressed as the following Expression (15).

LoD = log ₂ ((1 + a) × 2 ^b ) = log ₂ (1 + a) + b (15)
In equation (15), “b” is the integer part of LoD, and “log ₂ (1 + a)” is the decimal part of LoD.

The logarithmic calculation unit 336 calculates LoD by obtaining the integer part and the decimal part of the equation (15).
Next, the configuration of the logarithmic operation unit 336 will be described in detail.

FIG. 7 is a block diagram illustrating a configuration of the logarithmic operation unit.
The logarithmic operation unit 336 includes a carry determination unit 336a, a storage unit 336b, and an addition unit 336c.

  The carry determination unit 336a obtains “a” and “b” shown in Expression (13) from the hypotenuse value ρ1. Here, since Expression (13) indicates that “(1 + a)” is carried by “b” in the binary number, the carry judging unit 336a judges how many bits are carried. To do. Then, the obtained “a” is sent to the storage unit 336b, and “b” is sent to the adding unit 336c.

Here, when a fixed decimal point is used for the calculation of the hypotenuse value ρ1 so far, the number of bits of the hypotenuse value ρ1 from the decimal point to the most significant “1” is “b”.
On the other hand, when a floating point is used for the calculation of the hypotenuse value ρ1 so far, the value can be easily obtained from the exponent part of the hypotenuse value ρ1. When “1” is detected, “a” is indicated thereafter.

The storage unit 336b includes a constant storage table 3361 and a selector 3362.
The constant storage table 3361 stores a plurality of values (constants) of “log ₂ (1 + a)” corresponding to the value of “a”. In FIG. 7, as “log ₂ (1 + a)”, four values (“log ₂ (1 + a) # 1”, “log ₂ (1 + a) # 2”, “log ₂ (1 + a) # 3”, “log ₂ ( 1 + a) # 4 ") is stored. Note that the number of constants to be stored can be determined according to the number of bits of “a” and the accuracy to be realized. For example, if “a” is 8 bits, the maximum number of “log ₂ (1 + a)” stored in the constant storage table 3361 is 2 ⁸ = 256.

The selector 3362 selects “log ₂ (1 + a)” stored in the constant storage table 3361 corresponding to “a” sent from the carry determination unit 336 a and sends it to the adding unit 336 c.

The adding unit 336c adds “b” output from the carry determining unit 336a and “log ₂ (1 + a)” output from the storage unit 336b to obtain the LoD, and generates the obtained LoD as a texture acquisition address. To the unit 34.

FIG. 8 is a diagram showing a drawing figure by the drawing apparatus of the present embodiment.
As described above, according to the rendering processing apparatus 10 of the present embodiment, the calculation unit 335 substantially determines the value ρ1 of the hypotenuse of a right triangle having two values γ and δ as two orthogonal sides. Since the calculation is performed, the distortion of the textures 52 to 55 of the drawn tunnel image 51 can be corrected as shown in FIG. In addition, since the computing unit 335 obtains the value ρ1 of the hypotenuse without performing the square root computation, and the logarithmic computation unit 336 obtains the LoD without performing the logarithmic computation, the LoD computation unit 33 can be reduced in size. Can do.

In addition, the comparison units 332 and 333 can compare the amount of change to facilitate calculation.
Further, the storage unit 335b stores the coefficient for each region in a table in advance, so that the ρ calculation unit 335 can easily perform the calculation.

  If the number of coefficient blocks is small, the storage unit 335b can be configured by a small selector 3352 as in this embodiment without using a ROM (Read Only Memory) or the like.

  In addition, although the example which divided | segmented the area | region into four was demonstrated in FIG. 5, this invention is not restricted to this, The precision of an approximate expression can be improved by increasing the number of areas. Increasing the number of areas can be easily handled by changing the configuration of the storage unit 335b.

  The number of regions to be divided is not particularly limited, but is preferably a power of 2. As a result, when the LoD calculation unit 33 is configured by hardware, it can be handled by bit shift, and thus the size of the apparatus can be reduced.

  The drawing processing apparatus, the drawing processing method, and the drawing processing program of the present invention have been described based on the illustrated embodiment. However, the present invention is not limited to this, and the configuration of each unit has the same function. Can be replaced with any structure having Moreover, other arbitrary structures and processes may be added to the present invention.

In addition, the present invention may be a combination of any two or more configurations (features) of the above-described embodiments.
The above processing functions can be realized by a computer. In that case, a program describing the processing contents of the functions that the drawing processing apparatus 10 (in particular, the LoD calculation unit 33) should have is provided. By executing the program on a computer, the above processing functions are realized on the computer. The program describing the processing contents can be recorded on a computer-readable recording medium. Examples of the computer-readable recording medium include a magnetic recording device, an optical disk, a magneto-optical recording medium, and a semiconductor memory. Examples of the magnetic recording device include a hard disk device (HDD), a flexible disk (FD), and a magnetic tape. Examples of the optical disc include a DVD (Digital Versatile Disc), a DVD-RAM (Random Access Memory), a CD-ROM (Compact Disc Read Only Memory), and a CD-R (Recordable) / RW (ReWritable). Examples of the magneto-optical recording medium include MO (Magneto-Optical disk).

  When distributing the program, for example, a portable recording medium such as a DVD or a CD-ROM in which the program is recorded is sold. It is also possible to store the program in a storage device of a server computer and transfer the program from the server computer to another computer via a network.

  A computer that executes a drawing processing program stores, for example, a program recorded on a portable recording medium or a program transferred from a server computer in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. The computer can also read the program directly from the portable recording medium and execute processing according to the program. In addition, each time the program is transferred from the server computer, the computer can sequentially execute processing according to the received program.

It is a figure which shows the outline | summary of this invention. It is a block diagram which shows the structure of a drawing apparatus. It is a flowchart which shows the calculation algorithm of a LoD calculating part. It is a block diagram which shows the structure of a LoD calculating part. It is a figure explaining how to obtain | require the coefficient in an approximate expression. It is a block diagram which shows the structure of a calculating part. It is a block diagram which shows the structure of a logarithm calculating part. It is a figure which shows the drawing figure by the drawing apparatus of this Embodiment. It is a figure which shows the figure by the conventional drawing method.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 10 Drawing processing apparatus 2 1st selection part 3 2nd selection part 4, 335 calculating part 20 CPU
30 GPU
31 Geometry 32 Rasterization Unit 33 LoD Operation Unit 34 Texture Acquisition Address Generation Unit 35 Pixel Blender 40 External Memory 41 Frame Buffer 42 Texture Storage Unit 50 Monitor 331 Change Measurement Unit 332, 333 Comparison Unit 334 Size Comparison Unit 335a Coefficient Selection Unit 335b, 336b Storage unit 335c ρ1 calculation unit 336 Logarithmic calculation unit 336a Carry determination unit 336c Addition unit 3351 Coefficient selection table 3352, 3362 Selector 3361 Constant storage table b1-b4 Coefficient block

Claims (8)

In a drawing processing apparatus that acquires a texture of an optimal size from a texture storage unit that stores a mipmap,
A first selection unit that compares the amount of change (du / dx) of the texture coordinates (U, V) with respect to the pixel (X, Y) and the amount of change (dv / dy), and selects the larger value;
A second selection unit that compares the amount of change (du / dy) of the texture coordinates (U, V) with respect to the pixel (X, Y) and the amount of change (dv / dx), and selects the larger value; ,
The value selected by the first selection unit and the value selected by the second selection unit are regarded as two sides intersecting a right angle of the right triangle, and the value of the hypotenuse of the right triangle is substantially calculated. An arithmetic unit;
A drawing processing apparatus comprising:   The said calculating part has a magnitude comparison part which determines the magnitude relationship between the value selected by the said 1st selection part, and the value selected by the said 2nd selection part, The size comparison part of Claim 1 characterized by the above-mentioned. Drawing processing device.   The calculation unit regards the larger value determined by the magnitude comparison unit as a fixed value, regards the smaller value as a variable value, calculates an approximate expression composed of a linear expression, and calculates the value of the hypotenuse. The drawing processing apparatus according to claim 2, wherein the drawing processing apparatus performs calculation.   The drawing processing apparatus according to claim 3, wherein the calculation unit separately selects a coefficient of the approximate expression according to a ratio of the variable value to the fixed value.   The drawing processing apparatus according to claim 4, further comprising a storage unit that stores the coefficient. When calculating the arithmetic expression for the reduction ratio of the texture expressed by the sum of the decimal part (log ₂ (1 + a)) and the integer part (b), the decimal part (log ₂ (1 + a)) corresponding to a is calculated. A storage unit for storing in advance;
A determination unit that applies the value of the hypotenuse calculated by the calculation unit to ((1 + a) × 2 ^b ) to determine the a and the integer part (b);
A logarithmic operation unit further comprising: an addition unit that adds the integer part (b) determined by the determination unit and the decimal part (log ₂ (1 + a)) stored in the storage unit. The drawing processing apparatus according to claim 1, wherein: In the drawing processing method for acquiring the texture of the optimum size from the texture storage unit storing the mipmap,
In the change amount (du / dx), change amount (dv / dx), change amount (du / dy), change amount (dv / dy) of the texture coordinates (U, V) with respect to the pixel (X, Y),
First to select the larger value of {(du / dx) ² + (du / dy) ² } ^1/2 and {(dv / dx) ² + (dv / dy) ² } ^1/2 And the steps
A ^second value for selecting the larger one of {(du / dx) ² + (dv / dx) ² } ^1/2 and {(du / dy) ² + (dv / dy) ² } ^1/2 And the steps
A third step in which a larger value of the value selected in the first step and the value selected in the second step is a ratio between the pixel size of the texture and the pixel size of the drawn image When,
The drawing processing method characterized by including substantially. In a drawing processing program that acquires a texture of an optimal size from a texture storage unit that stores a mipmap,
Computer
A first selecting means for comparing the amount of change (du / dx) of the texture coordinates (U, V) with respect to the pixel (X, Y) and the amount of change (dv / dy), and selecting the larger value;
A second selecting means for comparing the amount of change (du / dy) of the texture coordinates (U, V) with respect to the pixel (X, Y) and the amount of change (dv / dx), and selecting the larger value;
The value selected by the first selection means and the value selected by the second selection means are regarded as two sides intersecting at right angles of the right triangle, and the value of the hypotenuse of the right triangle is substantially calculated. Computing means,
A drawing processing program characterized by functioning as

Images