Jump to content

mikejeep

Honored Premier Sponsor
  • Posts

    3,284
  • Joined

  • Last visited

  • Days Won

    197

Everything posted by mikejeep

  1. Things went down here in the Boston suburbs around 4am. No LTE at all and very erratic 1X/EVDO. I caught a glimpse of LTE a few minutes ago, we'll see if it holds. Twitterland hints the entire Northeast is affected. -Mike
  2. I did some digging and discovered that network type 18 indicates "IWLAN" network type. It's hidden from the the public Android source, but can be accessed by reflection. Your diagnostic report showed that you were connected to WiFi at the time. I had never heard of IWLAN before, but a quick Google search shows some references to WiFi calling. Do you have an Airave or something similar? It could always just be a Samsung bug since it's such a new device, we know how their enginerrs are.. -Mike
  3. T-Mobile band 2 LTE *usually* shows up as GCIs ending in 0B-0E. Like Trip said, T-Mobile did not implement an entirely consistent pattern across its network, so there are some exceptions. Unfortunately without a standard method to identify the band within Android, SignalCheck is forced to make educated guesses in these situations. However, I'm working on a new feature that will hopefully get the exact frequency on many devices. I'm still testing it but getting closer to a release. Speaking of releases, the beta crew got an update tonight! Unfortunately no frequency display yet, but 1X+LTE bug is fixed along with some other new stuff. Check the website if you want a sneak peek at the changes. -Mike
  4. If I have a chance this weekend, I'll get back over there with a decent camera and see if there's any equipment outside of the building at the base of the pole. I'm intrigued because nearly all of the sites in that area received their equipment upgrades a long time ago. This site and another nearby (mounted on a high-tension powerline support, which is rare around here) were the only exceptions. -Mike
  5. B25 is live as of this morning! I spent quite a bit of time in the area testing and looking for indications of B26 or B41, but it appears to be B25 only at this point. -Mike
  6. I'm terrible at identifying equipment.. does my photo show anything helpful? I wouldn't even try to guess which rack is Sprint. I've studied the "How to identify" posts and I'm still bad at it [emoji1] -Mike
  7. This site is still in progress.. the stealth covers are still set aside, but new antennas are finally visible (see below). A few nights ago I stumbled across some techs doing field testing; all they would share was they were working for Sprint and AT&T. I saw the new LTE PCI appearing as a neighbor cell in SignalCheck, but I couldn't connect to it. Today that PCI is gone. I can no longer get a 1X connection to this site, but eHRPD is still live. Hopefully LTE goes live soon.. -Mike
  8. Since I started testing my fix for this "bug" it's certainly seeming more and more like this is not a bug at all, but intentional behavior by Android 5.0+ devices, most often when connected to a weak LTE signal. The only "bug" is that the current release of SignalCheck does not display the GCI (which, in turn, causes other issues, like band identification and site note entry/display). But my fix does appear to be solid and ready for release. The only thing holding up an update is the new EARFCN features, which I'm close to finishing. -Mike
  9. Guess that means all the bugs are Android's fault, not mine! Does SCP work ok with the new splitscreen features? I thought I needed to add some code based on the early articles I read today, but it's not entirely clear. -Mike
  10. Per the Android docs, there is no network type 18. 1X should be reported as type 7. When you see this, can you send a Diagnostic Report? Not sure if there will be anything useful in it, but I'm curious. Are you sure it should be reporting 1X when you're seeing this? -Mike
  11. Hmm. It could very well actually be a problem with the PLMN causing issues then. A Diagnostic Report while it's happening should show me what I need. Keep an eye out for any of these circumstances if you don't mind! -Mike
  12. Gotcha. Next time you notice it happening, can you send a Diagnostic Report to me and let me know? The only thing that would block a log entry from being recorded is an invalid PLMN, unless there is an sqlite issue going on behind the scenes. Hopefully something will be in the log that reflects it. If you have logcat access, keep an eye on it; within the app you can watch it under the Diagnostics menu tab if you scroll all the way down. -Mike
  13. Thanks. At least I know it's not getting killed off in the background or anything like that. Does the onscreen information keep updating, or does it appear frozen when the logging stops? Just trying to pinpoint exactly what might be going on. If anyone else is experiencing the same issue, feel free to chime in. -Mike
  14. Do you know if it happens if you keep SignalCheck in the foreground the entire time? Do you know if it happens if you leave the app's Location Service disabled? As JWMaloney says below, apparently there's an issue with the CDMA Moto E. Even if I can't figure out what the issue is, if we can identify how to resolve it I might be able to implement a workaround: Thank you for the info, I wasn't aware of this issue. SignalCheck shouldn't have any trouble if the first cell isn't the connected one. Even though that has appeared to be the industry norm to this point, I assumed that could eventually change. -Mike
  15. Yes. Slowly but surely I am quietly squashing bugs, but I have not tackled this specific one yet. I did, however, confirm this morning that I have resolved the issue with limited LTE information being displayed when 1X+LTE appear together on newer devices! -Mike
  16. Do you get any sort of error message when you click the Reset button, from SignalCheck or otherwise? -Mike
  17. Correct. I don't think this new method will work on as many devices as the existing methods do, and as of right now it does require root also. -Mike
  18. Yes, my intention is to use the EARFCN to identify the band instead of taking the educated guess based off of GCI like it does now. As of this moment, it's just displaying it as another piece of data and not interpreting anything based on it. Once I can get it doing that smoothly and reliably, I'll change the app's overall behavior to rely on that. This is huge.. I'm excited. Hopefully it works on a decent swath of devices. I do plan to combine the UL/DL when they are identical and thought an option to only show DL would be useful too. With the neighbor cell display, screen space has become a premium. But first I need to make sure the data is reliably what I expect it to be, then I'll work on the aesthetics. I have a half-finished total redesign of the app with tabs that I'd love to get working, but my developer skills aren't quite there yet. On a somewhat unrelated note, I think I also figured out why the LTE GCI is hidden when 1X appears.. haven't been able to reliably confirm my suspicions yet because I don't see the 1X+LTE situation often. My old apartment was a fantastic testing ground because I could walk into certain rooms and knew what band my phone would connect to, and I my choice of them all (B25/26/41/Clear, as well as TMobile on Fi). My new place is great, but I'm on the fringe of any Sprint service whatsoever. -Mike
  19. Soooo after 20+ hours of coding this weekend, things are looking pretty awesome! It needs a lot of fine-tuning, but the biggest hurdle is out of the way. Yes, this screenshot from my Nexus 5X shows a B25 connection, with B25, B26, and B41 neighbor cells all at once.. wow. This is going to be cool. Stay tuned... -Mike
  20. I hear ya. I wish it was less complicated! The features that SCP needs root for really don't need to be blocked. -Mike
  21. Mine is rooted, yes; I haven't explored if this would be possible without root. I have an idea about a way to do some magic and possibly get it to work, but I think it may get blocked. After I get a rooted method working, I intend to unroot specifically to see what I can accomplish. -Mike
  22. Yes, Cellmapper shows EARFCNs on my 5X. It uses /dev/smd11. I have made good progress with this for SCP! Working almost 40 hours this weekend(!) but hoping to find some time to get something for everyone to test before I go on a road trip Wednesday. -Mike
  23. My hero! I would have killed to have read your posts about 8 hours ago.. I've been working on implementing this all day. I have tackled a majority of the issues, but it literally took all morning. Figuring out the exact modem command had me hung up for a bit (escape characters tripped me up but I figured it out after I finished my Cheerios) and I was struggling to figure out the best way to read the response from the appropriate device without choking the phone. Your code takes a slightly different approach than my implementation, but I think there's enough there to get me over the hump. Shame on me for not looking at S4GRU earlier.. I would be done by now. THANK YOU! -Mike
  24. Hmm I'm not sure why this would be happening. My Nexus 5X is running 6.0.1 and has no issues. Have you tried clearing SCP's permission out of SuperSU and then re-adding it when prompted? -Mike
×
×
  • Create New...