One thing I really miss that was in V1.2 is showing the frequency of a between pulses as well as the timing. I’ve attached the new and old screen shots so you will know what I’m talking about.
I’m glad you are releasing new saleae software. I look at new releases from time to time, and everytime I have a question: why new interface is so huge (buttons, fonts, icons, menu)?
I combined new interface view with old one to show the diff:
You are correct, the new interface needs to move in the direction of VS Code - we actually have been looking to it as a model as well. Tanveer (our designer) has been working on a new ‘sidebar’ that uses the VS Code concept of the ‘vertical icon bar’ shown here:
As it is with the example above, the sidebar will be collapsible as well as re-sizeable (horizontally). As for the size of elements, this is a bit of an open question; I want things to be maximally simple and friendly… yet it also needs to be super powerful (certainly will be over time). I hear you about “capacity”.
Def. a work in progress, thank you for the input!! Please do keep it coming as we iterate this design.
Hi
Nice work guys it starts to look very promising. Some comments on the Alpha 10:
I was able to run it in linux without any issue
The interface is very responsive I like it,
It would be nice to be able to pan vertically in the analog measurements using a modifier key like control, it could allow time and voltage pan when the key is pressed or midle click…
A voltage auto range button can be very useful so the analog measurement fits 80% percent of the vertical limits of what is shown.
I had problems sometimes putting the time markers, nothing happened when the mouse was over the timing display or clicked there, nothing
Time markers are nice and the functionality to go to the marker and highlight it is very appreciated, I however would make the color transition from green to blue when hoovering much faster or instantaneous, you want fast feedback of what is happening on the screen.
It is impossible to put a time marker and snap it to an edge, you have to place it and then move it…
Pair markers will be missed so you can easily get the frequency of some events… a wild idea is to have something like a child timing marker so right click add child marker and then you get delta T and frequency for that pair.
Thank you for the great feedback Luis - and welcome!
Actually you can, using the spacebar as the modifier. The ‘discoverability’ of this is terrible right now, we will fix it. We’ll also be making it possible to edit this modifier, and other shortcuts.
Totally! Is planned.
Noted, thanks. We’re reworking the timing markers now so this should get resolved as part of that effort.
Good point. Will do.
Totally agree, this is not good. Will work on it.
This functionality is super-hidden right now, we’re working on fixing this now.
Yes, I understand where you’re going here. We’re working though some different options for this now.
Question:
Would you say that 90% of your use cases for timing markers are to use them as “pairs”? My assumption is that this is the default use case - and so should more or less be the default behavior. Not sure tho.
I don’t know if your question is for at Luis, or readers in general, but I’ll assume the latter and say I almost always use the timing markers in larger sets than pairs.
Do you mean with the current (non-Alpha) 1.X Saleae software, or different software?
Could you give an example of a particular use case?
It’s def my intention to support essentially arbitrary use cases in the future , e.g. 2.1*(T2-T1)/(T3-T0) but for the next release we were going to focus on getting pairs to be easy. Very interested in uses cases for 3 or more markers. I imagine T1-T0, T2-T0, T3-T0, etc (multiple measurements off of a single starting point) is the main other one?
I downloaded and installed release alpha 10 on Windows, as my MacBook only features USB Type-C ports and I don’t have any adapter yet. Hint for the next generation of Saleae hardware: use USB-C Gen.2 for faster speed!
By default, the installer selects C:\Program Files\Logic as target folder. I may have already this folder with the main version of Logic I’m using for production. I suggest the installer proposes C:\Program Files\Logic2a instead.
Did you manually specify C:\Program Files\Logic when you installed the older software? By default, the “classic” software should install in either C:\Program Files\Saleae Inc or C:\Program Files\Saleae LLC, depending on which version it is. I actually selected C:\Program Files\Logic for the alpha software so it would not conflict with the older software.
Unless you need to keep the original software in C:\Program Files\Logic, I would suggest removing it and re-installing it in the default directory.
You might not need to do anything too - I haven’t tested this yet, but I would not be surprised that if when you update the alpha software in the future, it may default to use the directory you manually specified the first time you installed the alpha.
The “classic” windows installer and the alpha windows installers use different installer IDs so that the two installs should not conflict with each other - you should be able to install & update them independently.
Let me know if you have any other problems with the Windows installer or any feedback. The alpha release here was not digitally signed due to an ongoing problem with our CI provider, but we should have that fixed month when Advanced Installer releases an update for us. I’ve recently gone through the process of configuring a new installer and now is a great time to make improvements.
Thank you for the procedure. Actually, I gave a different name and selected a different folder. So the the alpha release is under C:\Program Files\Logic2a while the production version is under C:\Program Files\Logic.
apologies for the delayed response. I do this with non-Alpha Saleae Logic (using latest beta). My main use for this is looking at timing between and duration of packets on a serial bus, often also correlating that some external signal (whether different comms, discrete, analogue or power). I will try to establish the baseline expected behaviour and then catch the anomalous behaviour to see how the circumstances differ from the baseline.
I think multiple times from a single base (T2-T0, T1-T0 etc) and time between adjacent markers (T2-T1, T1-T0) would both be useful. The latter is what I do now, but perhaps that’s just a consequence of needing multiple marker pairs. I also like the ability to arbitrary expressions.
In terms of UX, I think I’d like to be able to hover a point, hit a keyboard button to drop a marker at that time. Ctrl+drag a marker could result in a new marker with a default calculation of New - Old.
Where with current Saleae software and traditional scopes/logic analysers I’ve used the concept seems to be that of a marker pair, this seems to separate this into two different concepts of ‘markers’ and ‘measurements’ (or ‘expressions’) - and I like that.
Thanks, me too! Has a ways to go. One thing we’re not sure of yet (experimenting with some different ideas) is how to combine the single marker concept with easily creating pairs - without requiring keyboard modifiers, etc. We’re thinking of always showing a “difference” calculation for the most recent 2 markers, that can optionally be “pinned” i.e. made permanent. The next release may appear to ‘regress’ a bit as we try to work this out.
I was trying out the 10 Alpha and noticed a few issues. Not sure how many are addressed or unimplemented.
I’m on MacOS 10.14.6 (18G95) on a MacBook Pro (Retina, 15-inch, Mid 2015) using a Logic Pro 16.
When using the I2C decoder there are no “Bubbles” showing the decoded data. They appear in the search list, but not in the view. I’ve included a screenshot.
When toggling visibility the gap above the waveform changes, so the bubbles appear to simply not be rendering
When editing the I2C encoder I can’t select 7-bit without clicking the third option, clicking elsewhere, then clicking the first option.
The save button on protocol editing will occasionally not respond to clicking and freeze the application.
Clicking on the decoded data will not bring me to the waveform that it correlates to.
Shrinking the font of the decoded data or widening the side panel would be helpful. The data is significantly cutoff.
