Skip to main content

Show Posts

This section allows you to view all Show Posts made by this member. Note that you can only see Show Posts made in areas you currently have access to.

Messages - JohnCC

Client software / Comboboxes in 0.9.7 don't work with mouse
First of all, it took me while to find this subforum, my apologies for
contacting the author directly. Here goes the write-up:

In version 0.9.7, the dropdown (combobox) widgets in the capture
dialogs don't work anymore. I mean, they _do_ drop down, but it is
not possible to change the selection. This happens both in the sample
rate setting and in the recording size setting.

I resurrected version 0.9.6, and things are normal there. I can
even set values for both (sample rate and recording size), and
they will then be read in 0.9.7 (but are unchangeable).

After focusing the combobox with a mouse click, it is possible to select
the sample rate (or capture size) with the keyboard. <enter> then
correctly selects the item.

I just noticed that the Color scheme selection combobox _does_
work with the mouse. Not entirely normally though: The dropdown
menu appears at the other end of the OLS window, completely
separate from where the actual Preferences dialog is (8 cm away).


Note: I'm using JRE on Slackware Linux
Open Bench Logic Sniffer / Re: Trigger method question

Seems I missed you on IRC. I had to leave for a while.

I think I have everything working. I could upload the firmware to the PIC, and though I had a Segmentation Fault at the end, this was after the 'DONE' message.

Then, I could upload the bit stream without problems, and Jawi's OLS client worked fine. RLE worked too.

The Java client version included in the package did NOT work though, it produced a large amount of run-time messages.

Thank you, Tayken!
Open Bench Logic Sniffer / Re: Trigger method question
Hello Tayken,

I'm terribly confused. I have found _a lot_ of versions of the firmware and bit streams to upload and I'm not sure which is the correct one. I found at least as many pages with instructions, all a little different.
And the instructions tell to use -vid 0x04D8 -pid 0xFC90, while the board enumerated as -pid 0xFC92.

Bus 002 Device 025: ID 04d8:fc92 Microchip Technology, Inc.

I get no reply when using FC90, and I get (using fw_update -vid 0x04d8 -pid 0xfc92 -ver):

$ fw_update -vid 0x04d8 -pid 0xfc92 -ver                 
fw_update Version: 0.2.0
U2IO BootLoader Version reading: FAILED.
Failed to send HID buffer.
Operation aborted.

All this was with the ZIP downloaded from GtiHub. Also tried with ols-0308-expert.tgz, but I believe the fw_update executables are the same.

Open Bench Logic Sniffer / Re: Trigger method question
Ok, Tayken... I tried to use RLE for a while, but I can't seem to understand it. I mean, I do understand the algorithm of course, but the data coming in from the sniffer seems almost random. These are some observations:

  • The client (the standard Java client 0.9.6) shows the data, but doesn't seem to insert the correct time in between changes. Each 'time slot' on the graph has a change, so it's as if the time isn't taken into account
  • The actual data seems very suspicious. Channels are repeated randomly, and I cannot seem to relate the data to what is really happening.

When in 'normal', non-RLE mode, I can get samples which do make sense though, so I suspect the communication is working correctly.

I'd appreciate suggestions! Maybe I'm just doing something wrong here...
Open Bench Logic Sniffer / Trigger method question
Hi all,

This is my first post, and I just got the Sniffer working! Works nicely...

Given the reduced size of the storage, I was thinking it might be interesting to have trigger mode which only stores if there is a change on any of the lines (or on selected lines). Particularly for a project which I'm debugging, this would be very useful, as I am not really interested in the lime scale per se, but I am interested in
the sequence of  changes.

Would that be possible with a reasonable effort?