Performance stats plus tips to help you attain optimal VDP speed!
This page serves to provide a benchmark set of test results that can be used to evaluate DesignMerge output speeds for PDF/VT. In addition, this page also provides some very useful tips for maximizing overall performance for your DesignMerge system. In addition, we have provided the complete list of the hardware and software that we used for our benchmark systems.
First off, it is important to understand that the overall performance for any particular VDP job will be relative to the complexity of the design. Many things affect overall performance, including the use of large and/or variable graphics, complex design elements, or advanced Rules (see the Performance Tips section below for more details).
To demonstrate this, and also to create a performance benchmark for DesignMerge, we put together a simple VDP job, which is a single-page personalized letter from a bank (click for preview). The job consists of a background PDF image, and includes basic variable text information overlayed on top. All of the PDF output files for this test were produced with PDF/VT caching enabled, and the PDF Preset utilized was the Adobe InDesign standard [High Quality Print]. The table below presents the overall output times for producing PDF/VT files.
Record Count | Mac (Apple Silicon) | Mac (Intel) | Windows 10 |
---|---|---|---|
500 | 3 sec | 6 sec | 6 sec |
1000 | 4 sec | 9 sec | 8 sec |
5,000 | 16 sec | 42 sec | 40 sec |
7,500 | 25 sec | 59 sec | 57 sec |
10,000 | 34 sec | 1 min 23 sec | 1 min 17 sec |
15,000 | 52 sec | 2 min 8 sec | 1 min 50 sec |
20,000 | 1 min 8 sec | 2 min 51 sec | 2 min 35 sec |
25,000 | 1 min 29 sec | 3 min 43 sec | 3 min 25 sec |
50,000 | 2 min 52 sec | 6 min 3 sec | 5 min 55 sec |
These times represent the best possible output speed achievable at the time of testing because the systems utilized had sufficient memory, fast CPU’s, and solid state hard disk drives, all of which contribute to maximizing the overall performance of any VDP system.
All PDF files were saved to the local hard disk drive on the benchmark systems. We do not recommend saving files directly to server volumes or printer hot folders as this can decrease overall performance. If you want to print to a hot folder, we recommend using the PDF Driver Post-Processing option called “Move to Specified Folder”. With this option, you save the PDF file on your local disk in a temporary folder that you create. Then when it is done, DesignMerge automatically moves it to the hot folder you selected. This ensures maximum overall performance. For more details, please see page 10 of the DesignMerge PDF Driver manual, which can be downloaded here.
A full copy of the benchmark job can be downloaded here if you would like to run these tests on your own system. When you run the test, please do the following:
You can use any of the provided sample data files. The total output time will be presented in the summary window at the end of the DesignMerge VDP session. Compare the times for your system to the benchmarks provided above. If your times are much slower, you may need to consider a hardware upgrade.
Please see the Performance Tips Page in our online knowledgebase for a number of helplful tips related to document setup and workflow modifications that can really improve overall processing times.
Following are a few suggestions for your computer system that will be sure to maximize the performance of your DesignMerge software. The suggestions below are listed in order of importance (e.g. your hard disk speed is more important than the amount of memory you have installed in your system). Please click here to view the specs for our own benchmark systems.
Printed output from DesignMerge can be very disk-intensive, and therefore the faster the hard disk that you can provide, the faster your overall output speeds will be. We recommend using a solid-state hard disk (SSD) for the best possible performance, however, if SSD is not an option, please be sure to select a drive that has a speed of at least 7,200 RPM.
The print processes for DesignMerge are very disk intensive. When creating VDP output files, we recommend not printing directly to server volumes (such as a printer hot folder) as this can slow the process down dramatically. If you want to print to a hot folder, we recommend using the Post-Processing option “Move to Specified Folder” that was described earlier. In addition, you should try to ensure that all files (InDesign document, fixed graphics, variable graphics and database files) are run from the local hard disk drive as well.
As DesignMerge performs merging operations inside of InDesign, it takes advantage of the speed of the computer’s CPU. The faster the CPU, the faster DesignMerge will merge data. Because the CPU speeds on the market change so frequently, our only recommendation is to purchase a computer with the fastest CPU that you can find for your budget.
DesignMerge by itself does not require a significant amount of extra memory in order to achieve good performance. In fact, if you have sufficient memory to run Adobe InDesign, and to open and edit your documents without noticeable latency issues, then your system should work well with DesignMerge. However, more memory allows you to run other applications, and may even improve overall system performance, so we always recommend using as much memory on your system as your budget will allow.
So you have a job, and you are wondering which DesignMerge print driver format to use to obtain the best possible performance (both on the desktop, and at the printer). The first thing to do is to determine which VDP file formats your digital print controller can accept (because this may limit your choices). The VDP formats that can be created by DesignMerge include PDF and PDF/VT, Adobe PostScript, and PPML.
Most modern production printers support PDF/VT, and for 99% of your variable data jobs that will be the preferred output format. For older systems that may not support PDF/VT, the PPML format would be the next best choice, and if you cannot print PPML then PostScript may be a good fall-back solution (but be prepared for some slow output because PostScript is not optimized for variable data).
Once you know what your options are, we recommend the following basic guidelines: