Tmpgenc plus 2.54.37.135 Chinese Version

Tmpgenc plus 2.54.37.135 Chinese Version

Tmpgenc plus 2.54.37.135 Chinese Version Tmpgenc is an MPEG Coding tool software that supports various formats such as VCD, SVCD, and DVD. It can compress and convert common video files and even JPG images into various disc formats (such as VCD, SVCD, and DVD ), it is often used to convert AVI files to mpeg1 and MPEG2 formats (VCD and DVD formats). In addition, it is easy to use, making it one of the most popular video compression software. This software is the best software for converting RM and rmvb into VCD and DVD. After conversion, the image quality is the best, and the speed is acceptable. during conversion, you can choose between speed and image quality, however, it cannot directly identify RM and rmvb. Your system needs to install storm videos and other things first. If you can use Windows Media Player, you can play RM and rmvb. (That is, the system can play RM and rmvb to convert) 　　Tip: Tmpgenc plus can be found online 2.52 2.53 2.54 2.58 3.0 and other English versions, Chinese versions, and simplified Chinese versions, but only 2.53.35.130, 2.54.37.135 version conversion RM, rmvb files can be normal conversion sound, other versions can also be converted normally. Sometimes, the converted MPG file uses RM and rmvb files for audio input sources without sound! The software can be used without installation. SN: TE25-PKNB-NPGA-OOJV-65EY ◆ Even non-professionals can easily perform the "Project Assistant" function of coding settings. ◆ Equipped with a "new 2-time processing variable bit rate encoding engine" for higher image quality" ◆ Create the "DVD, SVCD, and VCD source code" function of the original DVD/VCD MPEG-1/2 soft coding for movie files such as Avi ◆ Adjust the image size and Aspect Ratio During encoding ◆ Video/audio code stream synthesis/decomposition ◆ You can re-adjust the audio in the video to set the sound effect. Tmpgenc plus 2.54.37.135 Chinese version (including all kvcd templates) Tmpgenc plus tutorial Reference contentDetailed description of software parameter settings 　　I. Video This section sets the types and parameters of the output video streams. Most of the parameters are fixed in the template. 01. basic types: mpeg1/MPEG2, mpeg1 for VCD, and MPEG2 for SVCD/DVD. 02. Size: The pal vcd standard is 352x288, the pal SVCD standard is 480X576, And the pal dvd standard is 720x576. 03. Image aspect ratio: Generally, use the 4-3 625 line pal, which is the ratio of the TV screen. 04. Recovery rate: The PAL standard is 25fps. 05. Bit Rate Control: Bit Rate Control Algorithms are the key factors that cause different encoding efficiency and quality of encoder. The MPEG standard does not specify the specific implementation of the secondary algorithm, which is usually the intellectual property content of the commercial version. CBR (fixed bit rate): Keep the bit rate at the average bit rate. The implementation is simple. For complex scenarios, mosaic may occur due to insufficient bit rates. For simple scenarios, the coding space cannot be fully utilized. (The complex scenario here refers to the scenario with rich details/edges and rapid changes ). VBR (2-Pass VBR): "secondary processing VBR ". This means that the encoding efficiency is highest by processing the entire video source twice: The first time is used to determine where the video is in a complex scenario or a simple scenario, and the second time is based on the upper and lower limits of the bit rate, the bit rate is more distributed to complex scenarios. It can be seen in the experiment that tmpgenc indicates that the image is not previewed before 50% during this encoding, And the progress is 0. Therefore, we recommend that Wei long translate it into "secondary processing ". This code stream control method should obtain the best quality under the given bit rate, but it is closely related to the two allocation algorithms. It takes the longest time at the same time. Some other encoders even have three bit rate optimizations. MVBR (manual variable bit rate), sets the maximum bit rate and sets different information loss values for different frame types to achieve local bit rate optimization. You can manually specify a complex scenario as an I frame for finer encoding. See the GOP parameter settings section. CQ-VBR (automatic variable bit rate), set the subjective quality value and the lower limit of the bit rate, and control the quantitative part of the encoder according to the subjective quality standard, the encoder tries its best to save the bit rate while achieving this quality standard. The key lies in the Evaluation Method of the encoder's subjective quality. This is a combination of CQ and VBR. It can also be viewed as automatic. One of the optional values of MVBR weilong Chinese Edition 5 is "filling data without damaging the minimum bit rate". In my opinion, if the bit rate is too low, the meaningless code will be filled (a waste of time, but it may be because of compatibility). The English version is not translated or Japanese. CQ (Fixed Quality) is a set of subjective quality values than MVBR. I don't know how to control it. RT-CBR (Real-time fixed bit rate): Even the GOP-level bit rate optimization is not done CBR, a little faster, the quality is not high. RT-CQ (Real-time fixed quality): Even the GOP layer bit rate optimization is not done CQ, faster, the quality is not high. 06. Bit Rate: the average bit rate in CBR mode. 07. vbv buffer size: If the buffer is large, the encoding optimization will be better, but the decoding requires a larger buffer. Therefore, the parameters in the VCD/SVCD standard are fixed. Otherwise, the machine may not be able to play the video. 08. pofile & level (category and level): this parameter is not available in mpeg1. In the SVCD/DVD application, it should be MP &; mL, and the template is automatically selected. MP &; HL is defined for HDTV. 09. Standard: it seems that this parameter is also related to mpeg2. We should use pal. 10. line-by-line scanning: mpeg1 only supports line-by-line scanning (25 frame/sec), and MPEG2 supports line-by-line scanning (50 field/sec ). If the video is played on the TV, we recommend that you select the barrier to smooth the movement. However, the barrier video on the PC may have a glitch, which is particularly evident in the horizontal motion. 11. Implement the drop-down during playback: This is required in film/NTSC format conversion, that is, the frame rate remains unchanged during encoding without the drop-down. See section B. advanced. Thank you for your correction. 12. YUV format: the bitwise allocated to the brightness, red, and blue. For human eyes, the brightness signal is the most sensitive. Therefore, a large amount of coding space is allocated for precision, while the color difference is rough. Usually. (In fact, the code bit for the blue difference is not 0 in the scheme. I don't know why) 13. DC component accuracy: DCT (discrete cosine transformation) is required for 8x8 image blocks in MPEG encoding. The significance of DC component is basically the average value of 8x8 blocks, generally, you need to allocate a relatively large coding space for the mosaic. Otherwise, the edge effect of the mosaic is obvious. (8 bit is not small, and each bit of oil and water in Image Compression needs to be drained .) 14. motion detection accuracy: MPEG is an intra-frame encoding method for I frames and prediction error code for P frames. It is to find the corresponding part of the P frame graph block in the I frame, and then encode the different parts of the two graph blocks, which can greatly save the bit rate. The higher the motion detection accuracy, the larger the range of image block search matching, the higher the encoding efficiency, and the slower the encoding speed. These algorithms are also not defined in the MPEG standard, and the implementation level of each vendor varies greatly. In general, set it to normal in tmpgenc. 　　Ii. Advanced (image source) This section sets parameters related to the video source and preprocessing the video source Before encoding. 01. Video Source Type: line-by-line or line-by-line scan. This parameter is automatically set when the video source file is opened. Tmpgenc12 cannot automatically recognize type 1 DV, which has been fixed in version 12a. See the coding test page. 02. Field sequence: This is the most chaotic part of the entire tmpgenc and even the entire desktop video field. Tmpgenc12a does not seem to be able to automatically set this parameter based on the video source. It took me a long time to understand this problem. This parameter is very important. Setting the inverse will cause the image to flash obviously. For example, the sequence of motion positions of an object is originally 1-2-3-4-5-6 -......, After the configuration is reversed, it will be 2-1-4-3-6-5-8 ...... For a simulated video source, the field sequence is determined by the capture card type. For DV, field order a is defined. There is no trouble at this point, but the trouble is that although there are only two types of Field orders, there are three types of names for them: field order A/B (in the Ulead software ), even/odd line first (called by tmpgenc), field top/bottom first (called by bitrate viewer), the correspondence between these three terms is the most troublesome. In the English version of tmpgenc12a, the default value is "even line first (field a)", but in weilong Chinese, the default value is "odd field-> even field ", wei long said that this is the original Japanese version. Be careful not to confuse it literally. In short, the relationship between the three statements is as follows: field a = even line first (odd->; even-field) = field bottom first. The most reliable method is to use different settings to generate a video for each high-speed motion scenario, and observe the video on the TV, you should be able to see the difference. 03. aspect ratio of the Video Source: tmpgenc can automatically identify the setting. Generally, it should be 4: 3 625 line pal. 04. Image Display ratio and position: Generally, "Full Screen Display and aspect ratio unchanged" is selected ", the so-called "full screen display and the aspect ratio remains unchanged 2" option may cause some images to be invisible, but this option has never been tried. There may be no difference in the 4-3 video source, but for the 9 wide screen image output on the 4-3 screen, "Full Screen Display and aspect ratio unchanged" is left blank up and down, "Full Screen Display and aspect ratio unchanged 2" will cut off the left and right sides of the screen. I have never tried this. I just want to guess it. If it is not correct, ask your friends to correct it. 05. filter groups: This set can pre-process the video source to improve the image quality. In general, lazy is implementing these functions in NLE software. In addition, the use of filters should be moderate, because objectively any use of filters introduces information loss, which is the cost of improving the subjective quality of low-quality video sources. 06. Image Source range: select some image sources for compression. 07. 24fps: 24fps is a standard for movies. It is not recommended. 08. Eliminate Ghosting: A Ghosting is a shadow that appears when the video source is poor. Never encountered in DV. 09. Noise Elimination: obvious particle noise is returned during shooting in low light conditions. This filter can be used to eliminate it. However, the side effect is to smooth the image. For example, the human face will be as smooth as a rubber doll, but there is no texture. 10. Sharpen edges: You can set parameters for horizontal and vertical edges for enhancement. 11. Simple color correction: Adjust brightness, contrast, gamma, and color. 12. Advanced color correction: color correction can be performed based on different color spaces such as RGB/YUV. 13. De-interlace (de-interlace): Convert the video source scanned by the line into a video scanned by line, if the output video is set to An interlace scan (for example, SVCD/DVD to be played on the TV), do not select. Lazy believes that you may not need to select this option when performing VCD (scanning mpeg1 line by line), depending on the size of the video source. For example, when the DV 576 line is used up, only 288 lines are required for VCD. simply discard one field and deinterlace is not required. On the contrary, if the video source is 352x288, You need to perform de-interlace. 14. Cutting the sword screen: Because the video on the TV stops playing the video around the edge, you can use this to encode only the visible part, which can speed up the encoding, in addition, the reduced bit rate can be used in the area not cropped to improve the image quality. In general, it is safe to crop 5% of the top and bottom lines. 15, drop-down Transformation: the adjustments to be made in the standard conversion are due to the different frame rates of film 24fps and NTSC 30fps. Generally, this is not required. 16. The frame rate remains unchanged. 17. Sound processing: You can increase/decrease the volume and fade in/out the sound. 　　Iii. GOP structure GOP = group of picture. In MPEG, a GOP is a set of consecutive time frames. There are three types of pictures in MPEG: I, P, and B. I are internal encoding frames, basically in JPEG format. P is the forward prediction frame, and the encoding method is the motion detection error code. For more information, see section A's description of motion detection. B is a bidirectional interpolation frame, interpolation is performed based on the front and back I/P frames to encode the interpolation error. We recommend that you do not modify the GOP structure to achieve the best balance between the compression ratio and image quality. In the extreme example, only I frame is used, and the image quality is guaranteed, but the bitstream is very large. 1. Output code stream for editing: This option will cancel the last B frame of GOP. Because B frames are bidirectional interpolation, encoding/decoding requires not only the previous I/P, but also the subsequent I/P frames. Canceling the last B frame removes the dependency between GOP and facilitates editing. 2. Detect scene changes: for fast-changing scenes, force set to I frame to ensure the image quality. 3. manually set the Frame Type: manually set the image frame to I frame that requires fine encoding. Combined with MVBR code stream control, you can fully control code stream distribution. 　　Iv. quantization Matrix The Quantization in MPEG is the quantization of the coefficients after the 8x8 YUV signal block is transformed by DCT. By using a relatively large quantization order for the high-frequency components, the coding space of the high-frequency components is reduced and the compression is achieved. The cost is the loss of image details, blurred edges, and so on. 1. Intra-frame encoding quantization matrix: this refers to the intra-frame encoding quantization matrix used for I frames. 2. Inter-frame encoding quantization matrix: the quantization matrix used to encode the inter-frame prediction error of non-I frames. The weilong Chinese version is called an external matrix. 3. Matrix Template Selection: We recommend that you use the MPEG standard for videos. We can see that the encoding matrix between frames is 16, because the inter-frame error has offset the low frequency component, the high-frequency component is rich, so it is different from the intra-frame encoding matrix. For computer animation, especially 2D line-based animation, we recommend that you use the CG template. We can see that because the c g itself has rich high-frequency signals, the encoding matrix within the frame is also unified to 32. In addition, some friends try to reduce the values of the quantization matrix. Lazy thinks this is of little significance. Because the quantization matrix is not the only factor in quantification, the degree of quantification must be adjusted according to the feedback of the code stream control part. In this way, even if the quantization order is reduced, the quantization coefficient will increase even if the bit rate is limited. 4. YUV output is ycrcb: The ycrcb color space is allocated to the Y Brightness Signal for greater encoding space. If the video source is in ycrcb format, this option can increase the image quality. However, YUV (ccir601) is generally used, such as DV, so do not select this option, otherwise the bit rate is wasted. 5. Floating Point discrete cosine transformation: the speed of the whole point operation is much faster than that of the floating point operation, but the accuracy is not as good as that of the floating point operation. The floating point here is actually to increase the coefficient of DCT transformation from 8 bit to 16 bit, and do not need the floating point generator unit to participate in the transformation, otherwise the speed is intolerable. 6. do not perform semi-pixel motion detection on the static part: Because the video source is separated, a field may appear in a field for the refined static edge line (one pixel width), such as a static subtitle, the other field does not flash. If this option is selected, the flashing will be eliminated. However, it seems that this is irrelevant to the quantization matrix. 7. Soft mosaic: Nothing to say. It is to do some special processing on the edge of the 8x8 graph block. It is best to use sufficient bit rate or bit rate control measures to solve mosaic, because although the softening here only performs low-frequency filtering on the edge, after all, the image will be blurred. 　　V. Audio Most of these templates do not need to be modified. There is no major impact. Not much discussion. 　　Vi. System The MPEG system refers to video + audio. VCD, SVCD, and DVD templates have been set. There is nothing to talk about.