Why a public prerelease:

Before we make a full release, we need to collect as many real-world samples as possible. This is needed in order to calculate statistics and:

• insert accurate baselines
• collect possible incompatibilities of custom KERNAL/DOS variants
• publish comprehensive comparison online
• collect non-compatible operation status messages returned from various devices and account for them

This prerelease can be identified by the "DATA COLLECTION MODE" ("DATENSAMMLUNG MODUS" in German interface) mode description:

What to do:

1. Prepare EMPTY (!) medium. A diskette, HDD directory, CF/SD flash card, etc.
2. Run the program as many rounds as you can afford, on as many real hardware combinations, with as many KERNALs as you can. Collect the result file(s) and send to us. Address is given once all tests are finished.
3. If you encounter a reproducible error, not caused by a faulty device or medium, this may be caused by the fact that your device/ROM combination reports status that is not compatible with what original Commodore hardware and operating system would return. If this happens, and you double check that this is not a problem with e. g. faulty diskette, then please turn "IGNORE ERRORS" option on and re-run the tests.
4. Please note that if the errors mentioned above are caused only by the "Format test", it is not advised to disable this test as we want to collect information about what your device returns as status when given the format command
5. Collect the files written to the test disk/directory or make a disk image and send those to the displayed e-mail address
6. If you encounter any problems, anomalies, or (OMG, no…) bugs, please report them to the same address asap. Alternatively you can also contact me here

IMPORTANT: Real hardware only. No emulators, please!

Obviously I cannot verify how the file was created but I ask you kindly again - please do not submit emulator generated files. This doesn't have anything to do with how much I like or dislike emulators (I find them an invaluable tool and use them myself too!). It is about obtaining as accurate data about real hardware (!) as possible. In the long run this can help improving the emulators too.

NOTE: The program is able to recognise several, popular KERNAL versions. If it doesn't recognise the KERNAL of your machine, then the ROM content will be written to the disk. Unless you have specific reasons for not doing so, we would like to ask you to send that file too. This can help us verify the results and expand the list of popular KERNAL extensions.

Download:

• speed_test_v2a_pre20200425-0.d64.zip – ZIPped disk image with the program
• speed-test_v2a_pre20200425-0.c64.zip – ZIPped standalone program executable

• 4 Sale
• About
• Misc hacks
• Project 64 Reloaded
• Services
• Various projects
  • DolphinDOS2
  • DolphinDOS3
  • ROMs replacements
  • 64'er Speed-Test
    • V2 public prerelease
  • emufiles.qlgenerator
  • SD_QLPlugins
  • SDview
• Contact Form

In preparation

One of the major projects in my pipeline is a new floppy speeder, which aims at beating all currently existing ones on multiple fronts, without bumping clock speeds or altering the controller's architecture in any other non-kosher way. This is of course nice and dandy but how does one reliable prove this? And not only to the world but also to oneself? Of course there are existing tools to measure storage performance, aren't there? Well - there are. One that relatively comprehensively tests speed of various storage/filesystem operations was published by 64'er Magazine and called – yeah – "Floppy-Speed-Test" or "speed-test" for short. This is the best thing I found, but…