Sorry for the trouble! We’re trying to track down the source of the bug with calibration.
Quick note - the 1.x software displays the device ID in HEX, while the 2.x software shows it in base 10. We need to fix that.
Right now we don’t have a way to manually install cal files in the 2.x software like you could with the 1.x software, which we still need to add too.
I’d like to look up the error reports from your computer, if you have had error reporting enabled. Could you send us your machineID? Just open the main menu -> Switch devices. In that window, click the “?” button, which will copy the details to your clipboard. We can use the machineId to look up any error reports that have been sent from your application, if any.
In the meantime, I’ve generated the alpha software calibration file for your device, you can download it here:
Place that *.cal file in %APPDATA%/Logic/calibrations, then open the software. Let me know if you have any trouble with it.
I love the new “scroll to middle” behavior using the skip to next / previous edge popup buttons. I’d like it even more if it skipped 90% across the trace area (instead of 50%) to show 90% new material .
How easy would it be to space analyzer output to match the trace, especially for async. For example:
shows the full text of the burst of characters, but obviously not aligned with their position in the trace. Personally I’m less interested in the character count, but I’m often interested in the time when characters are sent (so getting their position"right" is important to me).
I don’t understand why there are dots between characters at higher compressions. Also, I don’t see the need for spaces to be quoted, especially as that effectively takes up three character spaces to represent one character.
Error reports are not automatically sent due to the proxy server. I looked at the available error logs, but they all refer to version 1.2.29
I tried your file, but the result is the same. Cannot load calibration file again.
I tried to put the file in “Calibration” and “calibration”, there is no difference.
The Logic 1.X software stores application data in %APPDATA%\Saleae LLC\Logic, while the 2.X software stores data in %APPDATA%\Logic
The 1.x software used %APPDATA%\Saleae LLC\Logic\Calibration for the original cal file format, usually named something like “8114234469533283449.cal”
The 2.x software uses %APPDATA%\Logic\calibrations, and the cal files are usually named something like “2184971446578857821-1564542649389-1.cal”
You mentioned the folder names “Calibration” and “calibration” which I suspect you tried in the older 1.x app data directory. Please try the “calibrations” (plural) folder in the new 2.x app data directory and see if that works.
Also, I’ve suspected that the calibration download problem was caused by proxies, but I haven’t been able to really figure out why it doesn’t just work with automatic proxy configuration.
Do you know if you need to manually configure proxy settings for any application that tries to access the internet? Our web request library, axios, should support automatic proxy detection, but it also supports manual proxy configuration - something we haven’t implemented but could easily.
Also, do you know what proxy software your network uses? (e.g. squid proxy). At some point I’d like to see if I can replicate the setup with a virtual machine, so we can manually test here.
It’s inconsistent on purpose, but might be confusing anyway - We only show the time on the left if there is no major tick on the left half of the screen (the 0 in the second example).
Yes, I was asking if there was a setting somewhere that I had not noticed that would allow me to set the pre-trigger size like V1 had.
We’ll change it to “Pre-trigger time” and “Post-trigger time”, once you select the trigger mode.
Also regarding my #3 when I get time I will start to try to debug this myself.
If you can share the capture with us (here or via private email), that would be super helpful!
The markers appear to use a different time reference
We’re wondering about that tbh. I was thinking of removing the trigger time shift to zero and instead adding a marker on the graph (and navigation button to jump to it quickly). What do you think?
p.s.
We still owe you a response on the custom analyzer issue. We haven’t forgotten
Logic 2.2.5 hung after being left running but not sampling over the weekend. Clicking the close box eventually brought up a new main window with a popup dialog showing something like “loading links”. The new main window could be closed, but not the original main window.
It’s inconsistent on purpose, but might be confusing anyway - We only show the time on the left if there is no major tick on the left half of the screen (the 0 in the second example).
That certainly clears up some of the confusion I had with how it behaved. I assumed it was trying to use some fixed value on the actual timescale for determining when to show it, not the plot width.
Below we can see the -2s appear even before it should at least that is my opinion. I assume this has to do with some unique handling of the last major marking in the negative time direction.
If you can share the capture with us (here or via private email), that would be super helpful!
Sorry I might not have been clear with my original post. The *.sal file I attached in my first post when used with the linked plugin (linked in my original post) will produce the behavior. This is why I question if the plugin may be doing something that perhaps is not valid/supported…or there could be some other problem.
We’re wondering about that tbh. I was thinking of removing the trigger time shift to zero and instead adding a marker on the graph (and navigation button to jump to it quickly). What do you think?
I would be ok with it; though when discussing specific event’s in a timing diagram / logic analyzer capture I feel it may be more natural to refer to the “event” as time 0 (as you guys have it currently). For instance if you have two events ‘A’ and ‘B’ where the capture triggered on ‘A’. The user may want to know the time difference between ‘A’ and ‘B’. With the current method this time difference is built-in to the time scale. If we use the beginning of the time capture as time 0 then we need to do an additional math operation (markers pairs and measurement functions certainly help here, but it is an extra step); this assumes a pre-trigger time greater than 0 exists in the capture. But if there is some benefit to gain from not having to change the marker time reference to match the trigger time reference then it may be worth considering…but it’s hard for me to say right now.
We still owe you a response on the custom analyzer issue. We haven’t forgotten
As a user I want to store my captures. Some of mine locations are read only on my ext4 File system at Ubuntu. Trying to save to those locations results in a error message. This message is displayed for a short time after it vanish into the void. I as a user who takes a sip at my tee would assume that my capture was successfully saved.
As a user I would like the Software to have the following features:
Have a error message to click away when a save-attempt fails.
Tell the user that the file system is read only!
EDIT: I’ve expirienced a blinking red light on my LogicPro. I was switching in the device. I assume overvoltage peak or something? How ever I con only assume. An Errorp popped up, but vanished before I could read the data.
period representation unit mismatch between main display and measurements tab
haven’t seen this issue in a quick skim of the topic:
when I mark a measurements, the time width on the measurements screen appears as expected (showing time in us - 1 in the photo), but on the measurement tab (2 in the photo) the time difference appears in s and not us.
So, I tried to put the file manually in the directory you specified and still got an error.
After that, I used a mobile access point (bypassing the working proxy server) and … the calibration file was applied.
After that, just in case, I deleted the folder and restarted the application. The file downloaded automatically.
Again, we connect to the Internet through a proxy server and … again, the calibration file cannot be used again.
It turns out that checking for a file is only performed if there is an active connection to the Internet?
This message is displayed for a short time after it vanish into the void
Good point. We want you to enjoy your tea
We’ll keep the notification around until you dismiss it. In the future, we’ll also have a notification panel with all previous messages.
We’re planning on fixing it, however, there is no perfect solution here
For example, when you look at -1 sec, would you like the values on the right to be +0.1s or -0.9s?
We’re thinking of removing the zero shift of the trigger and instead mark its position on the timing display. What do you think?
.
