Search Results

Search found 812 results on 33 pages for 'computational geometry'.

Page 19/33 | < Previous Page | 15 16 17 18 19 20 21 22 23 24 25 26  | Next Page >

  • GUnload is null or undefined using Directions Service

    - by user1677756
    I'm getting an error using Google Maps API V3 that I don't understand. My initial map displays just fine, but when I try to get directions, I get the following two errors: Error: The value of the property 'GUnload' is null or undefined, not a Function object Error: Unable to get value of the property 'setDirections': object is null or undefined I'm not using GUnload anywhere, so I don't understand why I'm getting that error. As far as the second error is concerned, it's as if something is wrong with the Directions service. Here is my code: var directionsDisplay; var directionsService = new google.maps.DirectionsService(); var map; function initialize(address) { directionsDisplay = new google.maps.DirectionsRenderer(); var geocoder = new google.maps.Geocoder(); var latlng = new google.maps.LatLng(42.733963, -84.565501); var mapOptions = { center: latlng, zoom: 15, mapTypeId: google.maps.MapTypeId.ROADMAP }; map = new google.maps.Map(document.getElementById("map_canvas"), mapOptions); geocoder.geocode({ 'address': address }, function (results, status) { if (status == google.maps.GeocoderStatus.OK) { map.setCenter(results[0].geometry.location); var marker = new google.maps.Marker({ map: map, position: results[0].geometry.location }); } else { alert("Geocode was not successful for the following reason: " + status); } }); directionsDisplay.setMap(map); } function getDirections(start, end) { var request = { origin:start, destination:end, travelMode: google.maps.TravelMode.DRIVING }; directionsService.route(request, function(result, status) { if (status == google.maps.DirectionsStatus.OK) { directionsDisplay.setDirections(result); } else { alert("Directions cannot be displayed for the following reason: " + status); } }); } I'm not very savvy with javascript, so I could have made some sort of error there. I appreciate any help I can get.

    Read the article

  • Possible to lock attribute write access by Doors User?

    - by Philip Nguyen
    Is it possible to programmatically lock certain attributes based on the user? So certain attributes can be written to by User2 and certain attributes cannot be written to by User2. However, User1 may have write access to all attributes. What is the most efficient way of accomplishing this? I have to worry about not taking up too many computational resources, as I would like this to be able to work on quite large modules.

    Read the article

  • How can I make my Google Maps api v3 address search bar work by hitting the enter button on the keyboard?

    - by Gavin
    I'm developing a webpage and I would just like to make something more user friendly. I have a functional Google Maps api v3 and an address search bar. Currently, I have to use the mouse to select search to initialize the geocoding function. How can I make the map return a placemark by hitting the enter button on my keyboard? I just want to make it as user-friendly as possible. Here is the javascript and div, respectively, I created for the address bar: var geocoder; function initialize() { geocoder = new google.maps.Geocoder (); function codeAddress () { var address = document.getElementById ("address").value; geocoder.geocode ( { 'address': address}, function(results, status) { if (status == google.maps.GeocoderStatus.OK) { map.setCenter(results [0].geometry.location); marker.setPosition(results [0].geometry.location); map.setZoom(14); } else { alert("Geocode was not successful for the following reason: " + status); } }); } <div id="geocoder"> <input id="address" type="textbox" value=""> <input type="button" value="Search" onclick="codeAddress()"> </div> Thank you in advance for your help

    Read the article

  • Is there a shorthand term for O(n log n)?

    - by jemfinch
    We usually have a single-word shorthand for most complexities we encounter in algorithmic analysis: O(1) == "constant" O(log n) == "logarithmic" O(n) == "linear" O(n^2) == "quadratic" O(n^3) == "cubic" O(2^n) == "exponential" We encounter algorithms with O(n log n) complexity with some regularity (think of all the algorithms dominated by sort complexity) but as far as I know, there's no single word we can use in English to refer to that complexity. Is this a gap in my knowledge, or a real gap in our English discourse on computational complexity?

    Read the article

  • Software to aid using camera as a scanner

    - by xxzoid
    I want to digitize some index cards with my camera. I'm looking for a program that would automatically fix geometry on the shots (as you would expect the cards come tilted on the picture). I have an app (droid scan lite) on my android phone that does exactly that, but I would prefer to do it on my pc (the phone camera has poor quality and it's slow and focuses badly while I have a decent slr). If the program is open source it's an advantage, cross platform -- even more so.

    Read the article

  • Automatically start VNC server on startup

    - by Vasu
    I installed the Ubuntu desktop on a Ubuntu 9.10 VPS server and am able to connect to the server using TightVNC. However, the VNC server on this VPS can only be started by logging in through SSH and typing the following command: vncserver :1 -geometry 800x600 -depth 16 -pixelformat rgb565 If I run this command on startup or as a schedule task, it won't start. What are my options? Thanks

    Read the article

  • How can I set a minimum thumbnail size with ImageMagick?

    - by Zilk
    I'm trying to create thumbnails of JPG photos using ImageMagick's convert tool. The thumbnails need to have a defined size (210x159), no blank areas, and the image can be cropped if necessary. Unfortunately, I only have ImageMagick 6.3.7 available, which doesn't support the '^' geometry modifier (added in v6.3.8-3). Is there another way to achieve this in earlier versions of ImageMagick? Thanks in advance.

    Read the article

  • Oracle Expands Sun Blade Portfolio for Cloud and Highly Virtualized Environments

    - by Ferhat Hatay
    Oracle announced the expansion of Sun Blade Portfolio for cloud and highly virtualized environments that deliver powerful performance and simplified management as tightly integrated systems.  Along with the SPARC T3-1B blade server, Oracle VM blade cluster reference configuration and Oracle's optimized solution for Oracle WebLogic Suite, Oracle introduced the dual-node Sun Blade X6275 M2 server module with some impressive benchmark results.   Benchmarks on the Sun Blade X6275 M2 server module demonstrate the outstanding performance characteristics critical for running varied commercial applications used in cloud and highly virtualized environments.  These include best-in-class SPEC CPU2006 results with the Intel Xeon processor 5600 series, six Fluent world records and 1.8 times the price-performance of the IBM Power 755 running NAMD, a prominent bio-informatics workload.   Benchmarks for Sun Blade X6275 M2 server module  SPEC CPU2006  The Sun Blade X6275 M2 server module demonstrated best in class SPECint_rate2006 results for all published results using the Intel Xeon processor 5600 series, with a result of 679.  This result is 97% better than the HP BL460c G7 blade, 80% better than the IBM HS22V blade, and 79% better than the Dell M710 blade.  This result demonstrates the density advantage of the new Oracle's server module for space-constrained data centers.     Sun Blade X6275M2 (2 Nodes, Intel Xeon X5670 2.93GHz) - 679 SPECint_rate2006; HP ProLiant BL460c G7 (2.93 GHz, Intel Xeon X5670) - 347 SPECint_rate2006; IBM BladeCenter HS22V (Intel Xeon X5680)  - 377 SPECint_rate2006; Dell PowerEdge M710 (Intel Xeon X5680, 3.33 GHz) - 380 SPECint_rate2006.  SPEC, SPECint, SPECfp reg tm of Standard Performance Evaluation Corporation. Results from www.spec.org as of 11/24/2010 and this report.    For more specifics about these results, please go to see http://blogs.sun.com/BestPerf   Fluent The Sun Fire X6275 M2 server module produced world-record results on each of the six standard cases in the current "FLUENT 12" benchmark test suite at 8-, 12-, 24-, 32-, 64- and 96-core configurations. These results beat the most recent QLogic score with IBM DX 360 M series platforms and QLogic "Truescale" interconnects.  Results on sedan_4m test case on the Sun Blade X6275 M2 server module are 23% better than the HP C7000 system, and 20% better than the IBM DX 360 M2; Dell has not posted a result for this test case.  Results can be found at the FLUENT website.   ANSYS's FLUENT software solves fluid flow problems, and is based on a numerical technique called computational fluid dynamics (CFD), which is used in the automotive, aerospace, and consumer products industries. The FLUENT 12 benchmark test suite consists of seven models that are well suited for multi-node clustered environments and representative of modern engineering CFD clusters. Vendors benchmark their systems with the principal objective of providing comparative performance information for FLUENT software that, among other things, depends on compilers, optimization, interconnect, and the performance characteristics of the hardware.   FLUENT application performance is representative of other commercial applications that require memory and CPU resources to be available in a scalable cluster-ready format.  FLUENT benchmark has six conventional test cases (eddy_417k, turbo_500k, aircraft_2m, sedan_4m, truck_14m, truck_poly_14m) at various core counts.   All information on the FLUENT website (http://www.fluent.com) is Copyrighted1995-2010 by ANSYS Inc. Results as of November 24, 2010. For more specifics about these results, please go to see http://blogs.sun.com/BestPerf   NAMD Results on the Sun Blade X6275 M2 server module running NAMD (a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems) show up to a 1.8X better price/performance than IBM's Power 7-based system.  For space-constrained environments, the ultra-dense Sun Blade X6275 M2 server module provides a 1.7X better price/performance per rack unit than IBM's system.     IBM Power 755 4-way Cluster (16U). Total price for cluster: $324,212. See IBM United States Hardware Announcement 110-008, dated February 9, 2010, pp. 4, 21 and 39-46.  Sun Blade X6275 M2 8-Blade Cluster (10U). Total price for cluster:  $193,939. Price/performance and performance/RU comparisons based on f1ATPase molecule test results. Sun Blade X6275 M2 cluster: $3,568/step/sec, 5.435 step/sec/RU. IBM Power 755 cluster: $6,355/step/sec, 3.189 step/sec/U. See http://www-03.ibm.com/systems/power/hardware/reports/system_perf.html. See http://www.ks.uiuc.edu/Research/namd/performance.html for more information, results as of 11/24/10.   For more specifics about these results, please go to see http://blogs.sun.com/BestPerf   Reverse Time Migration The Reverse Time Migration is heavily used in geophysical imaging and modeling for Oil & Gas Exploration.  The Sun Blade X6275 M2 server module showed up to a 40% performance improvement over the previous generation server module with super-linear scalability to 16 nodes for the 9-Point Stencil used in this Reverse Time Migration computational kernel.  The balanced combination of Oracle's Sun Storage 7410 system with the Sun Blade X6275 M2 server module cluster showed linear scalability for the total application throughput, including the I/O and MPI communication, to produce a final 3-D seismic depth imaged cube for interpretation. The final image write time from the Sun Blade X6275 M2 server module nodes to Oracle's Sun Storage 7410 system achieved 10GbE line speed of 1.25 GBytes/second or better performance. Between subsequent runs, the effects of I/O buffer caching on the Sun Blade X6275 M2 server module nodes and write optimized caching on the Sun Storage 7410 system gave up to 1.8 GBytes/second effective write performance. The performance results and characterization of this Reverse Time Migration benchmark could serve as a useful measure for many other I/O intensive commercial applications. 3D VTI Reverse Time Migration Seismic Depth Imaging, see http://blogs.sun.com/BestPerf/entry/3d_vti_reverse_time_migration for more information, results as of 11/14/2010.                            

    Read the article

  • How John Got 15x Improvement Without Really Trying

    - by rchrd
    The following article was published on a Sun Microsystems website a number of years ago by John Feo. It is still useful and worth preserving. So I'm republishing it here.  How I Got 15x Improvement Without Really Trying John Feo, Sun Microsystems Taking ten "personal" program codes used in scientific and engineering research, the author was able to get from 2 to 15 times performance improvement easily by applying some simple general optimization techniques. Introduction Scientific research based on computer simulation depends on the simulation for advancement. The research can advance only as fast as the computational codes can execute. The codes' efficiency determines both the rate and quality of results. In the same amount of time, a faster program can generate more results and can carry out a more detailed simulation of physical phenomena than a slower program. Highly optimized programs help science advance quickly and insure that monies supporting scientific research are used as effectively as possible. Scientific computer codes divide into three broad categories: ISV, community, and personal. ISV codes are large, mature production codes developed and sold commercially. The codes improve slowly over time both in methods and capabilities, and they are well tuned for most vendor platforms. Since the codes are mature and complex, there are few opportunities to improve their performance solely through code optimization. Improvements of 10% to 15% are typical. Examples of ISV codes are DYNA3D, Gaussian, and Nastran. Community codes are non-commercial production codes used by a particular research field. Generally, they are developed and distributed by a single academic or research institution with assistance from the community. Most users just run the codes, but some develop new methods and extensions that feed back into the general release. The codes are available on most vendor platforms. Since these codes are younger than ISV codes, there are more opportunities to optimize the source code. Improvements of 50% are not unusual. Examples of community codes are AMBER, CHARM, BLAST, and FASTA. Personal codes are those written by single users or small research groups for their own use. These codes are not distributed, but may be passed from professor-to-student or student-to-student over several years. They form the primordial ocean of applications from which community and ISV codes emerge. Government research grants pay for the development of most personal codes. This paper reports on the nature and performance of this class of codes. Over the last year, I have looked at over two dozen personal codes from more than a dozen research institutions. The codes cover a variety of scientific fields, including astronomy, atmospheric sciences, bioinformatics, biology, chemistry, geology, and physics. The sources range from a few hundred lines to more than ten thousand lines, and are written in Fortran, Fortran 90, C, and C++. For the most part, the codes are modular, documented, and written in a clear, straightforward manner. They do not use complex language features, advanced data structures, programming tricks, or libraries. I had little trouble understanding what the codes did or how data structures were used. Most came with a makefile. Surprisingly, only one of the applications is parallel. All developers have access to parallel machines, so availability is not an issue. Several tried to parallelize their applications, but stopped after encountering difficulties. Lack of education and a perception that parallelism is difficult prevented most from trying. I parallelized several of the codes using OpenMP, and did not judge any of the codes as difficult to parallelize. Even more surprising than the lack of parallelism is the inefficiency of the codes. I was able to get large improvements in performance in a matter of a few days applying simple optimization techniques. Table 1 lists ten representative codes [names and affiliation are omitted to preserve anonymity]. Improvements on one processor range from 2x to 15.5x with a simple average of 4.75x. I did not use sophisticated performance tools or drill deep into the program's execution character as one would do when tuning ISV or community codes. Using only a profiler and source line timers, I identified inefficient sections of code and improved their performance by inspection. The changes were at a high level. I am sure there is another factor of 2 or 3 in each code, and more if the codes are parallelized. The study’s results show that personal scientific codes are running many times slower than they should and that the problem is pervasive. Computational scientists are not sloppy programmers; however, few are trained in the art of computer programming or code optimization. I found that most have a working knowledge of some programming language and standard software engineering practices; but they do not know, or think about, how to make their programs run faster. They simply do not know the standard techniques used to make codes run faster. In fact, they do not even perceive that such techniques exist. The case studies described in this paper show that applying simple, well known techniques can significantly increase the performance of personal codes. It is important that the scientific community and the Government agencies that support scientific research find ways to better educate academic scientific programmers. The inefficiency of their codes is so bad that it is retarding both the quality and progress of scientific research. # cacheperformance redundantoperations loopstructures performanceimprovement 1 x x 15.5 2 x 2.8 3 x x 2.5 4 x 2.1 5 x x 2.0 6 x 5.0 7 x 5.8 8 x 6.3 9 2.2 10 x x 3.3 Table 1 — Area of improvement and performance gains of 10 codes The remainder of the paper is organized as follows: sections 2, 3, and 4 discuss the three most common sources of inefficiencies in the codes studied. These are cache performance, redundant operations, and loop structures. Each section includes several examples. The last section summaries the work and suggests a possible solution to the issues raised. Optimizing cache performance Commodity microprocessor systems use caches to increase memory bandwidth and reduce memory latencies. Typical latencies from processor to L1, L2, local, and remote memory are 3, 10, 50, and 200 cycles, respectively. Moreover, bandwidth falls off dramatically as memory distances increase. Programs that do not use cache effectively run many times slower than programs that do. When optimizing for cache, the biggest performance gains are achieved by accessing data in cache order and reusing data to amortize the overhead of cache misses. Secondary considerations are prefetching, associativity, and replacement; however, the understanding and analysis required to optimize for the latter are probably beyond the capabilities of the non-expert. Much can be gained simply by accessing data in the correct order and maximizing data reuse. 6 out of the 10 codes studied here benefited from such high level optimizations. Array Accesses The most important cache optimization is the most basic: accessing Fortran array elements in column order and C array elements in row order. Four of the ten codes—1, 2, 4, and 10—got it wrong. Compilers will restructure nested loops to optimize cache performance, but may not do so if the loop structure is too complex, or the loop body includes conditionals, complex addressing, or function calls. In code 1, the compiler failed to invert a key loop because of complex addressing do I = 0, 1010, delta_x IM = I - delta_x IP = I + delta_x do J = 5, 995, delta_x JM = J - delta_x JP = J + delta_x T1 = CA1(IP, J) + CA1(I, JP) T2 = CA1(IM, J) + CA1(I, JM) S1 = T1 + T2 - 4 * CA1(I, J) CA(I, J) = CA1(I, J) + D * S1 end do end do In code 2, the culprit is conditionals do I = 1, N do J = 1, N If (IFLAG(I,J) .EQ. 0) then T1 = Value(I, J-1) T2 = Value(I-1, J) T3 = Value(I, J) T4 = Value(I+1, J) T5 = Value(I, J+1) Value(I,J) = 0.25 * (T1 + T2 + T5 + T4) Delta = ABS(T3 - Value(I,J)) If (Delta .GT. MaxDelta) MaxDelta = Delta endif enddo enddo I fixed both programs by inverting the loops by hand. Code 10 has three-dimensional arrays and triply nested loops. The structure of the most computationally intensive loops is too complex to invert automatically or by hand. The only practical solution is to transpose the arrays so that the dimension accessed by the innermost loop is in cache order. The arrays can be transposed at construction or prior to entering a computationally intensive section of code. The former requires all array references to be modified, while the latter is cost effective only if the cost of the transpose is amortized over many accesses. I used the second approach to optimize code 10. Code 5 has four-dimensional arrays and loops are nested four deep. For all of the reasons cited above the compiler is not able to restructure three key loops. Assume C arrays and let the four dimensions of the arrays be i, j, k, and l. In the original code, the index structure of the three loops is L1: for i L2: for i L3: for i for l for l for j for k for j for k for j for k for l So only L3 accesses array elements in cache order. L1 is a very complex loop—much too complex to invert. I brought the loop into cache alignment by transposing the second and fourth dimensions of the arrays. Since the code uses a macro to compute all array indexes, I effected the transpose at construction and changed the macro appropriately. The dimensions of the new arrays are now: i, l, k, and j. L3 is a simple loop and easily inverted. L2 has a loop-carried scalar dependence in k. By promoting the scalar name that carries the dependence to an array, I was able to invert the third and fourth subloops aligning the loop with cache. Code 5 is by far the most difficult of the four codes to optimize for array accesses; but the knowledge required to fix the problems is no more than that required for the other codes. I would judge this code at the limits of, but not beyond, the capabilities of appropriately trained computational scientists. Array Strides When a cache miss occurs, a line (64 bytes) rather than just one word is loaded into the cache. If data is accessed stride 1, than the cost of the miss is amortized over 8 words. Any stride other than one reduces the cost savings. Two of the ten codes studied suffered from non-unit strides. The codes represent two important classes of "strided" codes. Code 1 employs a multi-grid algorithm to reduce time to convergence. The grids are every tenth, fifth, second, and unit element. Since time to convergence is inversely proportional to the distance between elements, coarse grids converge quickly providing good starting values for finer grids. The better starting values further reduce the time to convergence. The downside is that grids of every nth element, n > 1, introduce non-unit strides into the computation. In the original code, much of the savings of the multi-grid algorithm were lost due to this problem. I eliminated the problem by compressing (copying) coarse grids into continuous memory, and rewriting the computation as a function of the compressed grid. On convergence, I copied the final values of the compressed grid back to the original grid. The savings gained from unit stride access of the compressed grid more than paid for the cost of copying. Using compressed grids, the loop from code 1 included in the previous section becomes do j = 1, GZ do i = 1, GZ T1 = CA(i+0, j-1) + CA(i-1, j+0) T4 = CA1(i+1, j+0) + CA1(i+0, j+1) S1 = T1 + T4 - 4 * CA1(i+0, j+0) CA(i+0, j+0) = CA1(i+0, j+0) + DD * S1 enddo enddo where CA and CA1 are compressed arrays of size GZ. Code 7 traverses a list of objects selecting objects for later processing. The labels of the selected objects are stored in an array. The selection step has unit stride, but the processing steps have irregular stride. A fix is to save the parameters of the selected objects in temporary arrays as they are selected, and pass the temporary arrays to the processing functions. The fix is practical if the same parameters are used in selection as in processing, or if processing comprises a series of distinct steps which use overlapping subsets of the parameters. Both conditions are true for code 7, so I achieved significant improvement by copying parameters to temporary arrays during selection. Data reuse In the previous sections, we optimized for spatial locality. It is also important to optimize for temporal locality. Once read, a datum should be used as much as possible before it is forced from cache. Loop fusion and loop unrolling are two techniques that increase temporal locality. Unfortunately, both techniques increase register pressure—as loop bodies become larger, the number of registers required to hold temporary values grows. Once register spilling occurs, any gains evaporate quickly. For multiprocessors with small register sets or small caches, the sweet spot can be very small. In the ten codes presented here, I found no opportunities for loop fusion and only two opportunities for loop unrolling (codes 1 and 3). In code 1, unrolling the outer and inner loop one iteration increases the number of result values computed by the loop body from 1 to 4, do J = 1, GZ-2, 2 do I = 1, GZ-2, 2 T1 = CA1(i+0, j-1) + CA1(i-1, j+0) T2 = CA1(i+1, j-1) + CA1(i+0, j+0) T3 = CA1(i+0, j+0) + CA1(i-1, j+1) T4 = CA1(i+1, j+0) + CA1(i+0, j+1) T5 = CA1(i+2, j+0) + CA1(i+1, j+1) T6 = CA1(i+1, j+1) + CA1(i+0, j+2) T7 = CA1(i+2, j+1) + CA1(i+1, j+2) S1 = T1 + T4 - 4 * CA1(i+0, j+0) S2 = T2 + T5 - 4 * CA1(i+1, j+0) S3 = T3 + T6 - 4 * CA1(i+0, j+1) S4 = T4 + T7 - 4 * CA1(i+1, j+1) CA(i+0, j+0) = CA1(i+0, j+0) + DD * S1 CA(i+1, j+0) = CA1(i+1, j+0) + DD * S2 CA(i+0, j+1) = CA1(i+0, j+1) + DD * S3 CA(i+1, j+1) = CA1(i+1, j+1) + DD * S4 enddo enddo The loop body executes 12 reads, whereas as the rolled loop shown in the previous section executes 20 reads to compute the same four values. In code 3, two loops are unrolled 8 times and one loop is unrolled 4 times. Here is the before for (k = 0; k < NK[u]; k++) { sum = 0.0; for (y = 0; y < NY; y++) { sum += W[y][u][k] * delta[y]; } backprop[i++]=sum; } and after code for (k = 0; k < KK - 8; k+=8) { sum0 = 0.0; sum1 = 0.0; sum2 = 0.0; sum3 = 0.0; sum4 = 0.0; sum5 = 0.0; sum6 = 0.0; sum7 = 0.0; for (y = 0; y < NY; y++) { sum0 += W[y][0][k+0] * delta[y]; sum1 += W[y][0][k+1] * delta[y]; sum2 += W[y][0][k+2] * delta[y]; sum3 += W[y][0][k+3] * delta[y]; sum4 += W[y][0][k+4] * delta[y]; sum5 += W[y][0][k+5] * delta[y]; sum6 += W[y][0][k+6] * delta[y]; sum7 += W[y][0][k+7] * delta[y]; } backprop[k+0] = sum0; backprop[k+1] = sum1; backprop[k+2] = sum2; backprop[k+3] = sum3; backprop[k+4] = sum4; backprop[k+5] = sum5; backprop[k+6] = sum6; backprop[k+7] = sum7; } for one of the loops unrolled 8 times. Optimizing for temporal locality is the most difficult optimization considered in this paper. The concepts are not difficult, but the sweet spot is small. Identifying where the program can benefit from loop unrolling or loop fusion is not trivial. Moreover, it takes some effort to get it right. Still, educating scientific programmers about temporal locality and teaching them how to optimize for it will pay dividends. Reducing instruction count Execution time is a function of instruction count. Reduce the count and you usually reduce the time. The best solution is to use a more efficient algorithm; that is, an algorithm whose order of complexity is smaller, that converges quicker, or is more accurate. Optimizing source code without changing the algorithm yields smaller, but still significant, gains. This paper considers only the latter because the intent is to study how much better codes can run if written by programmers schooled in basic code optimization techniques. The ten codes studied benefited from three types of "instruction reducing" optimizations. The two most prevalent were hoisting invariant memory and data operations out of inner loops. The third was eliminating unnecessary data copying. The nature of these inefficiencies is language dependent. Memory operations The semantics of C make it difficult for the compiler to determine all the invariant memory operations in a loop. The problem is particularly acute for loops in functions since the compiler may not know the values of the function's parameters at every call site when compiling the function. Most compilers support pragmas to help resolve ambiguities; however, these pragmas are not comprehensive and there is no standard syntax. To guarantee that invariant memory operations are not executed repetitively, the user has little choice but to hoist the operations by hand. The problem is not as severe in Fortran programs because in the absence of equivalence statements, it is a violation of the language's semantics for two names to share memory. Codes 3 and 5 are C programs. In both cases, the compiler did not hoist all invariant memory operations from inner loops. Consider the following loop from code 3 for (y = 0; y < NY; y++) { i = 0; for (u = 0; u < NU; u++) { for (k = 0; k < NK[u]; k++) { dW[y][u][k] += delta[y] * I1[i++]; } } } Since dW[y][u] can point to the same memory space as delta for one or more values of y and u, assignment to dW[y][u][k] may change the value of delta[y]. In reality, dW and delta do not overlap in memory, so I rewrote the loop as for (y = 0; y < NY; y++) { i = 0; Dy = delta[y]; for (u = 0; u < NU; u++) { for (k = 0; k < NK[u]; k++) { dW[y][u][k] += Dy * I1[i++]; } } } Failure to hoist invariant memory operations may be due to complex address calculations. If the compiler can not determine that the address calculation is invariant, then it can hoist neither the calculation nor the associated memory operations. As noted above, code 5 uses a macro to address four-dimensional arrays #define MAT4D(a,q,i,j,k) (double *)((a)->data + (q)*(a)->strides[0] + (i)*(a)->strides[3] + (j)*(a)->strides[2] + (k)*(a)->strides[1]) The macro is too complex for the compiler to understand and so, it does not identify any subexpressions as loop invariant. The simplest way to eliminate the address calculation from the innermost loop (over i) is to define a0 = MAT4D(a,q,0,j,k) before the loop and then replace all instances of *MAT4D(a,q,i,j,k) in the loop with a0[i] A similar problem appears in code 6, a Fortran program. The key loop in this program is do n1 = 1, nh nx1 = (n1 - 1) / nz + 1 nz1 = n1 - nz * (nx1 - 1) do n2 = 1, nh nx2 = (n2 - 1) / nz + 1 nz2 = n2 - nz * (nx2 - 1) ndx = nx2 - nx1 ndy = nz2 - nz1 gxx = grn(1,ndx,ndy) gyy = grn(2,ndx,ndy) gxy = grn(3,ndx,ndy) balance(n1,1) = balance(n1,1) + (force(n2,1) * gxx + force(n2,2) * gxy) * h1 balance(n1,2) = balance(n1,2) + (force(n2,1) * gxy + force(n2,2) * gyy)*h1 end do end do The programmer has written this loop well—there are no loop invariant operations with respect to n1 and n2. However, the loop resides within an iterative loop over time and the index calculations are independent with respect to time. Trading space for time, I precomputed the index values prior to the entering the time loop and stored the values in two arrays. I then replaced the index calculations with reads of the arrays. Data operations Ways to reduce data operations can appear in many forms. Implementing a more efficient algorithm produces the biggest gains. The closest I came to an algorithm change was in code 4. This code computes the inner product of K-vectors A(i) and B(j), 0 = i < N, 0 = j < M, for most values of i and j. Since the program computes most of the NM possible inner products, it is more efficient to compute all the inner products in one triply-nested loop rather than one at a time when needed. The savings accrue from reading A(i) once for all B(j) vectors and from loop unrolling. for (i = 0; i < N; i+=8) { for (j = 0; j < M; j++) { sum0 = 0.0; sum1 = 0.0; sum2 = 0.0; sum3 = 0.0; sum4 = 0.0; sum5 = 0.0; sum6 = 0.0; sum7 = 0.0; for (k = 0; k < K; k++) { sum0 += A[i+0][k] * B[j][k]; sum1 += A[i+1][k] * B[j][k]; sum2 += A[i+2][k] * B[j][k]; sum3 += A[i+3][k] * B[j][k]; sum4 += A[i+4][k] * B[j][k]; sum5 += A[i+5][k] * B[j][k]; sum6 += A[i+6][k] * B[j][k]; sum7 += A[i+7][k] * B[j][k]; } C[i+0][j] = sum0; C[i+1][j] = sum1; C[i+2][j] = sum2; C[i+3][j] = sum3; C[i+4][j] = sum4; C[i+5][j] = sum5; C[i+6][j] = sum6; C[i+7][j] = sum7; }} This change requires knowledge of a typical run; i.e., that most inner products are computed. The reasons for the change, however, derive from basic optimization concepts. It is the type of change easily made at development time by a knowledgeable programmer. In code 5, we have the data version of the index optimization in code 6. Here a very expensive computation is a function of the loop indices and so cannot be hoisted out of the loop; however, the computation is invariant with respect to an outer iterative loop over time. We can compute its value for each iteration of the computation loop prior to entering the time loop and save the values in an array. The increase in memory required to store the values is small in comparison to the large savings in time. The main loop in Code 8 is doubly nested. The inner loop includes a series of guarded computations; some are a function of the inner loop index but not the outer loop index while others are a function of the outer loop index but not the inner loop index for (j = 0; j < N; j++) { for (i = 0; i < M; i++) { r = i * hrmax; R = A[j]; temp = (PRM[3] == 0.0) ? 1.0 : pow(r, PRM[3]); high = temp * kcoeff * B[j] * PRM[2] * PRM[4]; low = high * PRM[6] * PRM[6] / (1.0 + pow(PRM[4] * PRM[6], 2.0)); kap = (R > PRM[6]) ? high * R * R / (1.0 + pow(PRM[4]*r, 2.0) : low * pow(R/PRM[6], PRM[5]); < rest of loop omitted > }} Note that the value of temp is invariant to j. Thus, we can hoist the computation for temp out of the loop and save its values in an array. for (i = 0; i < M; i++) { r = i * hrmax; TEMP[i] = pow(r, PRM[3]); } [N.B. – the case for PRM[3] = 0 is omitted and will be reintroduced later.] We now hoist out of the inner loop the computations invariant to i. Since the conditional guarding the value of kap is invariant to i, it behooves us to hoist the computation out of the inner loop, thereby executing the guard once rather than M times. The final version of the code is for (j = 0; j < N; j++) { R = rig[j] / 1000.; tmp1 = kcoeff * par[2] * beta[j] * par[4]; tmp2 = 1.0 + (par[4] * par[4] * par[6] * par[6]); tmp3 = 1.0 + (par[4] * par[4] * R * R); tmp4 = par[6] * par[6] / tmp2; tmp5 = R * R / tmp3; tmp6 = pow(R / par[6], par[5]); if ((par[3] == 0.0) && (R > par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * tmp5; } else if ((par[3] == 0.0) && (R <= par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * tmp4 * tmp6; } else if ((par[3] != 0.0) && (R > par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * TEMP[i] * tmp5; } else if ((par[3] != 0.0) && (R <= par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * TEMP[i] * tmp4 * tmp6; } for (i = 0; i < M; i++) { kap = KAP[i]; r = i * hrmax; < rest of loop omitted > } } Maybe not the prettiest piece of code, but certainly much more efficient than the original loop, Copy operations Several programs unnecessarily copy data from one data structure to another. This problem occurs in both Fortran and C programs, although it manifests itself differently in the two languages. Code 1 declares two arrays—one for old values and one for new values. At the end of each iteration, the array of new values is copied to the array of old values to reset the data structures for the next iteration. This problem occurs in Fortran programs not included in this study and in both Fortran 77 and Fortran 90 code. Introducing pointers to the arrays and swapping pointer values is an obvious way to eliminate the copying; but pointers is not a feature that many Fortran programmers know well or are comfortable using. An easy solution not involving pointers is to extend the dimension of the value array by 1 and use the last dimension to differentiate between arrays at different times. For example, if the data space is N x N, declare the array (N, N, 2). Then store the problem’s initial values in (_, _, 2) and define the scalar names new = 2 and old = 1. At the start of each iteration, swap old and new to reset the arrays. The old–new copy problem did not appear in any C program. In programs that had new and old values, the code swapped pointers to reset data structures. Where unnecessary coping did occur is in structure assignment and parameter passing. Structures in C are handled much like scalars. Assignment causes the data space of the right-hand name to be copied to the data space of the left-hand name. Similarly, when a structure is passed to a function, the data space of the actual parameter is copied to the data space of the formal parameter. If the structure is large and the assignment or function call is in an inner loop, then copying costs can grow quite large. While none of the ten programs considered here manifested this problem, it did occur in programs not included in the study. A simple fix is always to refer to structures via pointers. Optimizing loop structures Since scientific programs spend almost all their time in loops, efficient loops are the key to good performance. Conditionals, function calls, little instruction level parallelism, and large numbers of temporary values make it difficult for the compiler to generate tightly packed, highly efficient code. Conditionals and function calls introduce jumps that disrupt code flow. Users should eliminate or isolate conditionls to their own loops as much as possible. Often logical expressions can be substituted for if-then-else statements. For example, code 2 includes the following snippet MaxDelta = 0.0 do J = 1, N do I = 1, M < code omitted > Delta = abs(OldValue ? NewValue) if (Delta > MaxDelta) MaxDelta = Delta enddo enddo if (MaxDelta .gt. 0.001) goto 200 Since the only use of MaxDelta is to control the jump to 200 and all that matters is whether or not it is greater than 0.001, I made MaxDelta a boolean and rewrote the snippet as MaxDelta = .false. do J = 1, N do I = 1, M < code omitted > Delta = abs(OldValue ? NewValue) MaxDelta = MaxDelta .or. (Delta .gt. 0.001) enddo enddo if (MaxDelta) goto 200 thereby, eliminating the conditional expression from the inner loop. A microprocessor can execute many instructions per instruction cycle. Typically, it can execute one or more memory, floating point, integer, and jump operations. To be executed simultaneously, the operations must be independent. Thick loops tend to have more instruction level parallelism than thin loops. Moreover, they reduce memory traffice by maximizing data reuse. Loop unrolling and loop fusion are two techniques to increase the size of loop bodies. Several of the codes studied benefitted from loop unrolling, but none benefitted from loop fusion. This observation is not too surpising since it is the general tendency of programmers to write thick loops. As loops become thicker, the number of temporary values grows, increasing register pressure. If registers spill, then memory traffic increases and code flow is disrupted. A thick loop with many temporary values may execute slower than an equivalent series of thin loops. The biggest gain will be achieved if the thick loop can be split into a series of independent loops eliminating the need to write and read temporary arrays. I found such an occasion in code 10 where I split the loop do i = 1, n do j = 1, m A24(j,i)= S24(j,i) * T24(j,i) + S25(j,i) * U25(j,i) B24(j,i)= S24(j,i) * T25(j,i) + S25(j,i) * U24(j,i) A25(j,i)= S24(j,i) * C24(j,i) + S25(j,i) * V24(j,i) B25(j,i)= S24(j,i) * U25(j,i) + S25(j,i) * V25(j,i) C24(j,i)= S26(j,i) * T26(j,i) + S27(j,i) * U26(j,i) D24(j,i)= S26(j,i) * T27(j,i) + S27(j,i) * V26(j,i) C25(j,i)= S27(j,i) * S28(j,i) + S26(j,i) * U28(j,i) D25(j,i)= S27(j,i) * T28(j,i) + S26(j,i) * V28(j,i) end do end do into two disjoint loops do i = 1, n do j = 1, m A24(j,i)= S24(j,i) * T24(j,i) + S25(j,i) * U25(j,i) B24(j,i)= S24(j,i) * T25(j,i) + S25(j,i) * U24(j,i) A25(j,i)= S24(j,i) * C24(j,i) + S25(j,i) * V24(j,i) B25(j,i)= S24(j,i) * U25(j,i) + S25(j,i) * V25(j,i) end do end do do i = 1, n do j = 1, m C24(j,i)= S26(j,i) * T26(j,i) + S27(j,i) * U26(j,i) D24(j,i)= S26(j,i) * T27(j,i) + S27(j,i) * V26(j,i) C25(j,i)= S27(j,i) * S28(j,i) + S26(j,i) * U28(j,i) D25(j,i)= S27(j,i) * T28(j,i) + S26(j,i) * V28(j,i) end do end do Conclusions Over the course of the last year, I have had the opportunity to work with over two dozen academic scientific programmers at leading research universities. Their research interests span a broad range of scientific fields. Except for two programs that relied almost exclusively on library routines (matrix multiply and fast Fourier transform), I was able to improve significantly the single processor performance of all codes. Improvements range from 2x to 15.5x with a simple average of 4.75x. Changes to the source code were at a very high level. I did not use sophisticated techniques or programming tools to discover inefficiencies or effect the changes. Only one code was parallel despite the availability of parallel systems to all developers. Clearly, we have a problem—personal scientific research codes are highly inefficient and not running parallel. The developers are unaware of simple optimization techniques to make programs run faster. They lack education in the art of code optimization and parallel programming. I do not believe we can fix the problem by publishing additional books or training manuals. To date, the developers in questions have not studied the books or manual available, and are unlikely to do so in the future. Short courses are a possible solution, but I believe they are too concentrated to be much use. The general concepts can be taught in a three or four day course, but that is not enough time for students to practice what they learn and acquire the experience to apply and extend the concepts to their codes. Practice is the key to becoming proficient at optimization. I recommend that graduate students be required to take a semester length course in optimization and parallel programming. We would never give someone access to state-of-the-art scientific equipment costing hundreds of thousands of dollars without first requiring them to demonstrate that they know how to use the equipment. Yet the criterion for time on state-of-the-art supercomputers is at most an interesting project. Requestors are never asked to demonstrate that they know how to use the system, or can use the system effectively. A semester course would teach them the required skills. Government agencies that fund academic scientific research pay for most of the computer systems supporting scientific research as well as the development of most personal scientific codes. These agencies should require graduate schools to offer a course in optimization and parallel programming as a requirement for funding. About the Author John Feo received his Ph.D. in Computer Science from The University of Texas at Austin in 1986. After graduate school, Dr. Feo worked at Lawrence Livermore National Laboratory where he was the Group Leader of the Computer Research Group and principal investigator of the Sisal Language Project. In 1997, Dr. Feo joined Tera Computer Company where he was project manager for the MTA, and oversaw the programming and evaluation of the MTA at the San Diego Supercomputer Center. In 2000, Dr. Feo joined Sun Microsystems as an HPC application specialist. He works with university research groups to optimize and parallelize scientific codes. Dr. Feo has published over two dozen research articles in the areas of parallel parallel programming, parallel programming languages, and application performance.

    Read the article

  • OpenGL ES 2/3 vs OpenGL 3 (and 4)

    - by Martin Perry
    I have migrated my code from OpenGL ES 2/3 to OpenGL 3 (I added bunch of defines and abstract classes to encapsulate both versions, so I have both in one project and compile only one or another). All I need to change was context initialization and glClearDepth. I dont have any errors. This was kind of strange to me. Even shaders are working correctly (some of them are GL ES 3 - with #version 300 es in their header) Is this a kind of good solution, or should I rewrite something more, before I start adding another functionality like geometry shaders, performance tools etc ?

    Read the article

  • Collision detection with curves

    - by paldepind
    I'm working on a 2D game in which I would like to do collision detection between a moving circle and some kind of static curves (maybe Bezier curves). Currently my game features only straight lines as the static geometry and I'm doing the collision detection by calculating the distance from the circle to the lines, and projecting the circle out of the line in case the distance is less than the circles radius. How can I do this kind of collision detection in a relative straightforward way? I know for instance that Box2D features collision detection with Bezier curves. I don't need a full featured collision detection mechanism, just something that can do what I've described.

    Read the article

  • gpgpu vs. physX for physics simulation

    - by notabene
    Hello First theoretical question. What is better (faster)? Develop your own gpgpu techniques for physics simulation (cloth, fluids, colisions...) or to use PhysX? (If i say develop i mean implement existing algorithms like navier-strokes...) I don't care about what will take more time to develop. What will be faster for end user? As i understand that physx are accelerated through PPU units in gpu, does it mean that physical simulation can run in paralel with rastarization? Are PPUs different units than unified shader units used as vertex/geometry/pixel/gpgpu shader units? And little non-theoretical question: Is physx able to do sofisticated simulation equal to lets say Autodesk's Maya fluid solver? Are there any c++ gpu accelerated physics frameworks to try? (I am interested in both physx and gpgpu, commercial engines are ok too).

    Read the article

  • OpenGL and atlas

    - by user30088
    I'm trying to draw element from a texture atlas with OpenGL ES 2. Currently, I'm drawing my elements using something like that in the shader: uniform mat4 uCamera; uniform mat4 uModel; attribute vec4 aPosition; attribute vec4 aColor; attribute vec2 aTextCoord; uniform vec2 offset; uniform vec2 scale; varying lowp vec4 vColor; varying lowp vec2 vUV; void main() { vUV = offset + aTextCoord * scale; gl_Position = (uCamera * uModel) * aPosition; vColor = aColor; } For each elements to draw I send his offset and scale to the shader. The problem with this method: I can't rotate the element but it's not a problem for now. I would like to know, what is better for performance: Send uniforms like that for each element on every frames Update quad geometry (uvs) for each element Thanks!

    Read the article

  • How to control in the vertex shader where pixel ends up in the renderTarget?

    - by cubrman
    What if I have an arbitrary renderTarget, that is smaller than the screen (say it is 1x1 pixel) and I want to make sure in the VertexShaderFunction that all my pixels end up exactly in that 1 pixel region? No matter what I do, they all seem to get culled at some point, though GraphicDevise.Clear() works OK. Where is the top left corner of the renderTarget Vertex-shader-vise? I tried output.Position = (0,0,0,0)/(0,0,0,1)/(1,1,1,1)/(-0.5,0.5,0,1) NOTHING works! Fullscreen quad is not an option 'cause I actually need to process geometry in the shaders to get the results I need.

    Read the article

  • Direct3D9 application won't write to depth buffer

    - by DeadMG
    I've got an application written in D3D9 which will not write any values to the depth buffer, resulting in incorrect values for the depth test. Things I've checked so far: D3DRS_ZENABLE, set to TRUE D3DRS_ZWRITEENABLE, set to TRUE D3DRS_ZFUNC, set to D3DCMP_LESSEQUAL The depth buffer is definitely bound to the pipeline at the relevant time The depth buffer was correctly cleared before use. I've used PIX to confirm that all of these things occurred as expected. For example, if I clear the depth buffer to 0 instead of 1, then correctly nothing is drawn, and PIX confirms that all the pixels failed the depth test. But I've also used PIX to confirm that my submitted geometry does not write to the depth buffer and so is not correctly rendered. Any other suggestions?

    Read the article

  • SQL Azure maximum database size rises from 10GB to 50GB in June

    - by Eric Nelson
    At Mix we announced that we will be offering a new 50gb size option in June. If you would like to become an early adopter of this new size option before generally available, send an email to [email protected]  and it will auto-reply with instructions to fill out a survey to nominate your application that requires greater than 10gb of storage. Other announcements included: MARS in April: Execute multiple batches in a single connection Spatial Data in June: Geography and geometry types SQL Azure Labs: SQL Azure Labs provides a place where you can access incubations and early preview bits for products and enhancements to SQL Azure. Currently OData Service for SQL Azure. Related Links: SQL Azure Announcements at MIX http://ukazure.ning.com

    Read the article

  • Create edges in Blender

    - by Mikey
    I've worked with 3DS Max in Uni and am trying to learn Blender. My problem is I know a lot of simple techniques from 3DS max that I'm having trouble translating into Blender. So my question is: Say I have a poly in the middle of a mesh and I want to split it in two. Simply adding an edge between two edges. This would cause a two 5gons either side. It's a simple technique I use every now and then when I want to modify geometry. It's called "Edge connect" in 3DS Max. In Blender the only edge connect method I can find is to create edge loops, not helpful when aiming at low poly iPhone games. Is there an equivalent in blender?

    Read the article

  • Omni-directional shadow mapping

    - by gridzbi
    What is a good/the best way to fill a cube map with depth values that are going to give me the least amount of trouble with floating point imprecision? To get up and running I'm just writing the raw depth to the buffer, as you can imagine it's pretty terrible - I need to to improve it, but I'm not sure how. A few tutorials on directional lights divide the depth by W and store the Z/W value in the cube map - How would I perform the depth comparison in my shadow mapping step? The nvidia article here http://http.developer.nvidia.com/GPUGems/gpugems_ch12.html appears to do something completely different and use the dot of the light vector, presumably to counter the depth precision worsening over distance? He also scales the geometry so that it fits into the range -.5 +.5 - The article looks a bit dated, though - is this technique still reasonable? Shader code http://pastebin.com/kNBzX4xU Screenshot http://imgur.com/54wFI

    Read the article

  • FFmpeg not recording audio during screen capture

    - by King
    I'm using the script below to run FFmpeg on Ubuntu 10.10. I followed these instructions to install FFmpeg & x264. While ffmpeg does capture the screen it does not capture the mic audio. I've checked that the mic works via "System Preferences". Anyone have any ideas on what the problem(s) could be and suggestions on how to resolve this issue? Thanks. ffmpeg -f alsa -ac 2 -i hw:0,0 -f x11grab -r 30 -s $(xwininfo -root | grep 'geometry' | awk '{print $2;}') -i :0.0 -acodec pcm_s16le -vcodec libx264 -vpre lossless_ultrafast -threads 0 -y screen-capture.mkv

    Read the article

  • Add autorandr before kdm starts

    - by Serge Tarkovski
    I want to add autorandr before kdm starts. Autorandr works well within KDE, however, in kdm I still have ugly 1024x768 resolution when my external monitor is connected. I tried adding autorandr --change to /etc/kde4/kdm/Xsetup: #! /bin/sh # Xsetup - run as root before the login dialog appears #xconsole -geometry 480x130-0-0 -notify -verbose -fn fixed -exitOnFail -file /dev/xconsole & /sbin/initctl -q emit login-session-start DISPLAY_MANAGER=kdm /usr/local/bin/autorandr --change >> /tmp/autorandr echo "Xsetup finished" >> /tmp/xsetup-finished A debug message in /tmp/xsetup-finished appears correctly. /tmp/autorandr is empty (so it seems autorandr runs without errors). I also tried to move autorandr --change line before /sbin/initctl -q emit login-session-start DISPLAY_MANAGER=kdm with no effect. P.S. Of course, autorandr profiles I created under KDE session, are in my home folder, but Xsetup script runs under root, so I created a symlink from my ~/.autorandr to /root/.autorandr.

    Read the article

  • Ray Tracing Shadows in deferred rendering

    - by Grieverheart
    Recently I have programmed a raytracer for fun and found it beutifully simple how shadows are created compared to a rasterizer. Now, I couldn't help but I think if it would be possible to implement somthing similar for ray tracing of shadows in a deferred renderer. The way I though this could work is after drawing to the gbuffer, in a separate pass and for each pixel to calculate rays to the lights and draw them as lines of unique color together with the geometry (with color 0). The lines will be cut-off if there is occlusion and this fact could be used in a fragment shader to calculate which rays are occluded. I guess there must be something I'm missing, for example I'm not sure how the fragment shader could save the occlusion results for each ray so that they are available for pixel at the ray's origin. Has this method been tried before, is it possible to implement it as I described and if yes what would be the drawbacks in performance of calculating shadows this way?

    Read the article

  • How does an Engine like Source process entities?

    - by Júlio Souza
    [background information] On the Source engine (and it's antecessor, goldsrc, quake's) the game objects are divided on two types, world and entities. The world is the map geometry and the entities are players, particles, sounds, scores, etc (for the Source Engine). Every entity has a think function, which do all the logic for that entity. So, if everything that needs to be processed comes from a base class with the think function, the game engine could store everything on a list and, on every frame, loop through it and call that function. On a first look, this idea is reasonable, but it can take too much resources, if the game has a lot of entities.. [end of background information] So, how does a engine like Source take care (process, update, draw, etc) of the game objects?

    Read the article

  • What's a good data structure solution for a scene manager in XNA?

    - by tunnuz
    Hello, I'm playing with XNA for a game project of myself, I had previous exposure to OpenGL and worked a bit with Ogre, so I'm trying to get the same concepts working on XNA. Specifically I'm trying to add to XNA a scene manager to handle hierarchical transforms, frustum (maybe even occlusion) culling and transparency object sorting. My plan was to build a tree scene manager to handle hierarchical transforms and lighting, and then use an Octree for frustum culling and object sorting. The problem is how to do geometry sorting to support transparencies correctly. I know that sorting is very expensive if done on a per-polygon basis, so expensive that it is not even managed by Ogre. But still images from Ogre look right. Any ideas on how to do it and which data structures to use and their capabilities? I know people around is using: Octrees Kd-trees (someone on GameDev forum said that these are far better than Octrees) BSP (which should handle per-polygon ordering but are very expensive) BVH (but just for frustum and occlusion culling) Thank you Tunnuz

    Read the article

  • Cool examples of procedural pixel shader effects?

    - by Robert Fraser
    What are some good examples of procedural/screen-space pixel shader effects? No code necessary; just looking for inspiration. In particular, I'm looking for effects that are not dependent on geometry or the rest of the scene (would look okay rendered alone on a quad) and are not image processing (don't require a "base image", though they can incorporate textures). Multi-pass or single-pass is fine. Screenshots or videos would be ideal, but ideas work too. Here are a few examples of what I'm looking for (all from the RenderMonkey samples): PS - I'm aware of this question; I'm not asking for a source of actual shader implementations but instead for some inspirational ideas -- and the ones at the NVIDIA Shader Library mostly require a scene or are image processing effects. EDIT: this is an open-ended question and I wish there was a good way to split the bounty. I'll award the rep to the best answer on the last day.

    Read the article

  • Triangle Strips and Tangent Space Normal Mapping

    - by Koarl
    Short: Do triangle strips and Tangent Space Normal mapping go together? According to quite a lot of tutorials on bump mapping, it seems common practice to derive tangent space matrices in a vertex program and transform the light direction vector(s) to tangent space and then pass them on to a fragment program. However, if one was using triangle strips or index buffers, it is a given that the vertex buffer contains vertices that sit at border edges and would thus require more than one normal to derive tangent space matrices to interpolate between in fragment programs. Is there any reasonable way to not have duplicate vertices in your buffer and still use tangent space normal mapping? Which one do you think is better: Having normal and tangent encoded in the assets and just optimize the geometry handling to alleviate the cost of duplicate vertices or using triangle strips and computing normals/tangents completely at run time? Thinking about it, the more reasonable answer seems to be the first one, but why might my professor still be fussing about triangle strips when it seems so obvious?

    Read the article

< Previous Page | 15 16 17 18 19 20 21 22 23 24 25 26  | Next Page >