We are happy to announce the availability of Dew Debugger Visualizer v2. The debugger visualizer is a part of the MtxVec product and is available from Rad Studio XE3 forward, if TeeChart support is checked during the installation. Its features and capabilities are best described and demonstrated by the new "Users Manual". Check out new features and bug fixes.
Some of those features are available also for VS.NET users, if they have installed Dew Lab Studio 2024 SDK for .NET
We are happy to announce the release of Dew Lab Studio 2024 R2 for Delphi and C++Builder. This latest update enhances support for Linux, most notably delivers updated shared libs for Red Hat Linux v8 and v9, provides native support for .csv and .txt files via LoadFromFile/SaveToFile for all vector and matrix types and fixes a number of issues, most notably a fix to Spline1D interpolation routine. Below you can find two screens from RHEL made with Embarcadero's Fmx for Linux, which shows MtxVec, DSP and Stats Demos compiled and running on Linux:
We are happy announce the release of Dew Lab Studio 2024 for .NET Framework and .NET Core. This latest release delivers builds for .NET Core v7 and v8. Among many new features are the greatly enhanced visualizers for Visual Studio IDE for the typical types Vector, Matrix, VectorInt, MatrixInt, support for XML and JSON serialization, first introduction of fair critial sections on .NET and much more. A more comprehensive list of changes can be found here.
We are happy to announce availability of Dew Lab Studio 2024 with support for Embarcadero Rad Studio Athens 12.0. This latest version delivers latest performance optimizations and a number of new features and bug fixes.
We are happy to announce the availability of Dew Lab Studio 2023 for Delphi and C++ Builder. This latest release rounds up a series of enhancements introduced over the course of the last year, that were affecting multi-threading. This is the first release that introduces the use of a fair critical section, which is required to achieve linear scaling on computers with large (8+) core count. This release is complemented by two videos released in december 2022 at CodeRage 2022:
We are happy to announce the availability of Dew Lab Studio 2022 for Embarcadero Delphi and C++Builder. Special care was given to the multi-threading library and the thread-pool. Substantial improvements in speed are to be expected in case of high-thread count (16+). Brute-force exact K-NN with exceptional performance has also been added to the library. Another set of functions which received treatment were ARIMA statistical forecasting routines from the Stats Master.
We are happy announce comprehensive support for hardware acceleration across all key features of MtxVec when deploying to Linux 64bit OS. Additionally, support for native ARM compiler when deploying to Apple M1 CPUs has been added. This update concludes two major milestones reached in 2021:
Release of 100% native C# version of MtxVec for .NET Core v5.0 and v6.0.
Availability of comprehensive hardware acceleration libraries on 64bit Linux. They are available for both Embarcadero Delphi and .NET Core distributions.
For the month of Januar of 2022 we plan to add a number of videos to our YouTube channel. The videos are aimed at new and existing customers. Drop by to explore features that you might like or did not know that they are there. Subscribe to our channel and get notified when new videos are added.
We are happy to announce the availability of Dew Lab Studio 2021 R2, which adds support for Embarcadero RAD Studio 11. This latest release of Delphi features long anticipated support for High DPI applications.
We are happy to announce the availability of the trial version of the Dew Lab Studio for .NET Framework and .NET Core. All products have been rewritten in C# language and the evaluation versions can be downloaded from the public nuget.org package repository (Search for "Dew.") . With this release Dew Lab Studio occupies a unique position of having a nearly 100% coherent API across multiple languages and platforms: Delphi, C#, C++, FireMonkey, Windows, Linux, iOS, OSX, MacOS and Android, .NET Framework and .NET Core. Regardless of your toolset, you can always find the familiar tools and performance. We also got our sample projects uploaded to github.
We are happy to announce the availability of Dew Lab Studio 2021 for Embarcadero RAD Studio. This latest release features the most comprehensive support of all features across all supported platforms (Windows, Linux, iOS, MacOS, Android). Most notably, shared high-performance libraries based on Intel OneAPI for Linux 64bit (x86) are now included in to the distribution for registered users. For the most part of the past year, we have been working to implement programming language mirroring of our software. The first result of this effort will be the release of Dew Lab Studio for .NET Core, which is due shortly and will feature 100% source code in C#. Other languages like C/C++ and Java are planned to be supported in the future.
We are happy to announce the availability for Dew Lab Studio 2020. The latest version brings support for Delphi 10.4 Sydney. For a comprehensive list of changes and enhancements please check the release histories for MtxVec, DSP Master and Stats Master.
Dew Lab Studio has been updated to support Visual Studio 2019.NET. This update includes all enhancements and improvements that have gathered in the last year. Most importantly, updated dlls with improved performance. Support for newer version of TeeChart.NET trial is now also included.
We are happy to announce the availability for Dew Lab Studio 2019. The latest version brings support for Delphi 10.3 Rio and a number of refinements to support FireMonkey and cross-platform development.
We took an Intel Core i7-7820X for a spin and compared the speed-up for scientific computations to Intel Core i5-4670. In the table below you can see some results, which are very typical across a large range of different scientific algorithms. The test run is from our "Efficient multithreading" example in the MtxVec demo. The code computes DFT using vectorized sin, cos, add, multiply and sum of vector.
i5-4670, 32bit, 4cores,
i7-7820X, 32bit, 4cores
i7-7820X, 64bit, 4cores
i7-7820X, 64bit, 8cores
Pascal, one core (not vectorized)
40.24s
34.59s
35.62s
35.19
One CPU core (vectorized)
7.12s
5.86s
3.72s
3.77s
With blocks, one CPU core
6.80s
4.67s
2.44s
2.40s
With hand-written blocks
5.75s
4.25s
1.75s
1.76s
Threaded (naive)
9.12s
7.22s
5.96s
5.52s
Threaded, with blocks
1.77s
1.22s
0.55s
0.34s
Threaded, blocks, Annonymous
1.78s
1.18s
0.57s
0.33s
Tthreaded, hand written, DoForLoop
1.54s
1.11s
0.43s
0.27s
Threaded, blocks, TParallel.For
2.93s
2.27s
1.20s
0.97s
The code executed with MtxVec takes full advantage of all instruction set features. This includes AVX-512 included with i7 7820X. Note that "turbo" frequencies between both CPUs are different. When using AVX, the CPU will also not "turbo boost" up to the highest frequency. i7-7820X was mostly boosting up to 4.0GHz and the i5-4670 remained at 3.4GHz. The test was run with "default" optimized motherboard configuration and without overclocking.
Best results are in bold separately for single core (1.76s) and multi-core (0.27s) in the rightmost column. It appears that Intel software tools (compiler + libs) only optimize for AVX-512 for 64bit apps. In this (64bit) case the performance improvement per core is about 1.11/0.43 = 2.5x between both CPUs. In case of 32bit apps, the gain is only about 1.3x.The ratio of the fastest code path on 7820x against non-optimized code reaches a factor of 35/0.27 = 130x when all 8 cores are used with AVX-512. The fastest code path running on one core gives a gain 35/1.76 = 19.8x
Interestingly enough, the dgemm on which linear algebra (LAPACK) mostly depends on remains at only 30% gain even in 64bit mode. Possibly related to missing AVX-512 instructions available only on 7900X-series CPUs and some XEON CPUs. More AVX-512 capable CPUs are scheduled to be released in 2018 and 2019.
AVX-512 largely delivers on the promise on increasing the performance per clock by about 2x even in heaviliy multithreaded scenarios. This fact however is largely absent from various benchmarks that can be found on internet. Either the tested applications are not 64bit or they are not yet properly optimized for AVX-512 (instructions + memory bandwidth). When compared to i7-8700K, the multimedia and scientific benchmarks should be showing an advantage of about 1.8x per one core for i7-7280X.
The most recent release of Dew Lab Studio for .NET delivers a cumulative update of all the new features added to our products Math, Stats and DSP Master in 2017.
The first release of Dew Lab Studio in 2018 brings support for Linux to MtxVec, DSP Master and Stats Master. The support is for now limited to those units, which do not require GUI and to the Core Edition with the latest Embarcadero Rad Studio Tokyo 10.2 (Update 2).
The latest update to Dew Lab Studio brings comprehensive support for Accelerate framework on Apple devices running iOS, iPad and OS X. The DSP Master has been complemented with cross-platform enabled components for audio playback and recording thus greatly simplifying development and deployment of audio processing/analysis applications to mobile platforms. Additionally mtxVec received a major upgrade for its expression parser/scripting engine bringing it much closer to Matlab/Scilab like capability. It is now possible to write while-loops, for-loops, if-else clauses and make use of the concatenation operator: a = [1,2 ; 3, 4];
The latest update brings support for Rad Studio 10.2 Tokyo and the .NET version delivers support for Visual Studio 2017.NET. Major new features include introduction of integer matrix type and extensive new integer math optimizations. Among other things, the expression parser received a major upgrade and now includes support for integers, integer vectors, integer matrices and boolean vectors and matrices.