Those who own a MiniPro TL866CS/A EPROM programmer may have come across issues programming vintage 2716/2732 EPROMs for their classic solid state pinballs.

I've modified the original v6.85 MiniPro software to correct some programming times and voltages based on specific EPROM datasheets and also corrected some of the poor English spelling in the interface.

Program times for these old EPROMs is insufficient in the stock software and potentially leads to weak/unreliable writes. These vintage chips spec a program pulse time of approx 50ms per cell. The stock software does it in single digit milliseconds.
The TL866 programmer will often verify 2716 writes as successful, but the EPROMs fail in their intended circuit and other EPROM programmers verify the writes as bad.
Discussion of these issues has occurred in other Pinside threads for this EPROM programmer.

WARNING! Consider this software as experimental. Use at your own risk! Software is based on the original v6.85 from the companies web site. If your TL866CS/TL866A is a counterfeit clone, there's a good chance running genuine software will brick your clone programmer.
There's no guarantee this will improve or perform perfect writes. These EPROMs are decades old now.

Download links:
https://www.mediafire.com/file/7dhcpgye2da75gf/MiniPro_v6-85X1.zip/file
https://www.sendspace.com/file/f5lle9

Details of the modifications are in the download.

The following image shows insufficient program time of a ST 2716 in 7.0 seconds using stock software

M2716_Program_Time_ST-M2716.png

More respectable program time of the same ST 2716 in 89 seconds using this modified software:

M2716_MOD_Program_Time_ST-M2716.png

If your firmware updates have been successful and you're running genuine software from the companies web site, you are probably ok.

The manufacturer has some details showing counterfeit units you can check:
http://www.autoelectric.cn/en/note.html

What version software were you running before trying to update to the newer manufacturers version?
What software version are you running now and what's the firmware version reported?

The programming algorithms are stored in the programmers firmware. Sometimes the manufacturer updates the algorithms to support new types of chips so reflashing the firmware is required with some newer versions of the software. However if your unit is a counterfeit there's a high risk of the official software wiping the programmers firmware.

Did/are you getting any messages about your programmer being a counterfeit?

Your 03.2.82 firmware is old. The latest firmware is ver 03.2.86 that comes with software version 6.85
I believe your firmware is from software version 6.71

Someone has set up a repository of most of the software releases and they're available here under the MiniPro folder:
https://github.com/Kreeblah/XGecu_Software

So if you're concerned about updating the firmware, just download and use software version 6.71

The firmware versions are bundled with and are really specific to the software versions. Their firmware's are not downloaded separately.

If you run the latest 6.85 software and update the firmware to v03.2.86 then there's no reason to be using software 6.71 anymore.

Software 6.71 is bundled with your old version 03.2.82 firmware - it doesn't include 03.2.86 so can't update to it.

Seems to me if your unit isn't a counterfeit then you should be safe updating the firmware. You can then run the modified software here that corrects some programming voltages and pulse timings for vintage EPROMs.

On a side note, I have their new cheap TL866-3G (T48) programmer which supports 25V program EPROMs (actually a quick check I did measured it at just over 23 volts).
Sadly (as of software version 12.15) it still has problems carried over from the original TL866CS/A programmers with some old 2716 EPROMs. I have some perfectly working Intel, National Semiconductor and Mitsubishi 2716 EPROMs that are blank yet these programmers fail blank check and read random data with these chips. Something isn't right with their timing. These programmers are working with ST and Hitachi 2716 EPROMs although the factory program pulse time of 1000us is too short. A single program cycle programs and verifies in these units ok, but take the EPROM to another programmer and you get verification errors on some bits. Plug them in your pinball and they won't work. Using these programmers, vintage EPROMs need to be programmed multiple times for a deeper program. That was the reason I modified the v6.85 MiniPro software to lengthen the program pulses so multiple program cycles weren't needed.