There are currently two options available for downloading the data that we have gathered. The first option is a .zip file download of all of the frequency, sample number, and oscillator data that are stored as comma-separated value (csv) files for easy export/import across many different systems. There are 193 .csv files, one for each of the FPGAs that were tested. The filenames have the format of <BoardID>.csv, where BoardID is the unique serial number of each board tested.
In each file, there are 512 lines for 512 ROs. Each of the line contains 100 RO frequencies separated by comma. These are 100 samples of the corresponding RO. The frequency values are expressed in MHz.
It is important to describe the spatial location of the ROs. The 2-D RO array has 32 rows and 16 columns. The first RO (first line in the csv file) is located at the CLB location X10Y8 (SLICE_X18Y14:SLICE_X19Y15), and the 16th RO (16th line in the csv file) is located at CLB X25Y8 (SLICE_X48Y14:SLICE_X49Y15). The intermediate 14 ROs are located across CLB location X11Y8 to CLB location X24Y8.
Similarly, the second row starts with the 17th RO (17th line in the csv file) at CLB location X10Y9 and ends with the 32nd RO (32nd line in the csv file) at the CLB location X25Y9. Likewise, the 512th RO (the last line in the csv file) is located at the CLB location X25Y39.
The above location information can also be found in a readme file included in the datafile repository.
The second option available is for those who wish to look at a particular subset of the data and have no desire to work with the entire dataset.
The resulting file from filling out the form below will have all data that matches the filters entered in a single .csv file. Each data point will be on a separate row of the .csv file. This will result in a file that very closely resembles the table shown on the Standard Dataset page.
Select your preferences to download the dataset. You may enter information to narrow the results as well as select the information you are interested in.
|   || |
This work is sponsored by National Science Foundation (NSF) grant no 0855095 and grant no 0964680.