Jump to content

mikejeep

Honored Premier Sponsor
 View Profile  See their activity

  • Posts

    3,284

  • Joined

  • Last visited

  • Days Won

    197

 Content Type 

Everything posted by mikejeep

  1. mikejeep
    Got it. Confirmed the OS is undoubtedly reporting EARFCN 950. Based on the math and everything posted here and elsewhere, it seems as if there's an issue with Android or Qualcomm radio firmware when the EARFCN is in the higher ranges (above 47000 at least). I reported it to Android, I'll figure out how to nag HTC and Qualcomm too. -Mike
  2. mikejeep
    Sorry! Complain to the Nova Launcher people.. that has been the most common offender based on those who have reached out to me.. do all of the different SCP icons do it? Or only the multi-layered ones (i.e. band + signal strength)? That's the only thing I can think of that might differ from other apps. -Mike
  3. mikejeep
    I've had a couple e-mails about this.. the light/dark color switching is completely controlled by the OS. There are no options on the developer side to configure how that works. Others who reported this issue had custom ROMs or launchers installed, so it's likely related to some sort of customization like that. It adjusts properly on stock ROMs. -Mike
  4. mikejeep
    Sending diagnostic reports is partially fixed.. still failing about 25% of the time but nobody can figure out why. I installed a completely different script on the server and it is still occasionally kicking back that "Temporary redirect" error. I will keep working on it.. EDIT: It seems to be working consistently again. Sorry for the inconvenience. -Mike
  5. mikejeep
    Please share if you find anything! I've already adjusted 06/07/08 to only show 10x10 in Samsung markets. I could have 07/08 indicate B25 second carrier oDAS in ALU/Ericsson markets. However, 09 (and 0A/0B) already represents B25 second carrier there. -Mike
  6. mikejeep
    Flompholph nailed it.. it's not a bug, that is how the app has always functioned. I do feel your pain, because when I'm trying to ID a new 1X connection it's annoying. LTE sites are easier to correlate because there is a clear, consistent pattern between GCIs at the same site. On the CDMA side, a character in the hex BID value does indicate the sector, but as stated, it varies depending on which market you are in. I tried coming up with something to match these up, but it proved to be sloppy and unreliable at the time. If I could determine a way to identify which BID configuration is in play (I'm guessing there might be a range of SIDs for each vendor?), I might be able to revisit it. -Mike
  7. mikejeep
    Thanks @RyanThaDude, you beat me to it! @davidtm, go to Logs > Backup Log Database on your old device. That will save a file (/SignalCheck/timestamp_signalcheck_log.db) on your device. Copy that file onto your new device and go to Logs > Import Log Database on that device. Navigate to wherever you saved the file, select it, and you should be all set. Right now, importing a log will erase any existing log entries on that device. The app will warn you of this. -Mike
  8. mikejeep
    Mini-Macro.. Nokia equipment. See here: http://s4gru.com/index.php?/topic/7688-how-to-spot-mini-macro-on-macro-setups-nokia/ Peek at the Help screen within the app to see explanations for all of the acronyms. -Mike
  9. mikejeep
    Thanks guys, my hosting provider is having issues. I got the website back online this morning, still working on the reporting gadget.. -Mike
  10. mikejeep
    I would like to make the diagnostics more robust, right now it just crudely tries to send it once and bails if it doesn't work. I was mostly concerned with better handling of short-term network interruptions, and hadn't considered that some devices might only be able to send data on Wi-Fi. Definitely something I could improve. -Mike
  11. mikejeep
    Nevermind.. that was a waste of time. Pretty sure whoever reached out was just some sort of overseas script-reader who cleans out tweets. I was told in broken English that my feedback was "appreciated" and "interesting" but until the phone programming is changed, that feature will not be available. I asked to have it escalated or passed on to an appropriate team, and got a generic "Thanks for being our customer!" back. I didn't expect much from their generic social media support, but I expected more than that.. -Mike
  12. mikejeep
    Hmm, same here apparently. Something happened on my hosting provider's end, nobody will be able to send any. Will update this post when I get it resolved.. thanks for the heads up! EDIT: It's partially working again.. still fails about 25% of the time, nobody can figure out why. I will keep on it. -Mike
  13. mikejeep
    I received a tweet and some DMs from Samsung tonight. They want specifics about which devices are experiencing this issue. If you (and anyone else with a Samsung on Android 7 that does not see any LTE EARFCNs displayed) could go to About > Send Diagnostics and send me a report, that will give me all the information I need. Please include a mention about why you are sending the report so it doesn't get lost in the pile. (EDIT: I already found recent reports from @nowerlater and @Mr.Nuke. Anyone else, feel free to send.) Thanks! -Mike
  14. mikejeep
    Bingo.. right now any B25 06/07/08 comes up 10x10 because (until now) I didn't have any conflicts with those sectors. I can change it to only show 10x10 on 06/07/08 if it's a Samsung site. @Dkoellerwx, does this oDAS broadcast on any other sectors? If anyone has similar info, I'd love to correlate this enough to add an oDAS indicator in non-Samsung markets.. -Mike
  15. mikejeep
    Nothing since my initial back-and-forth with them. Sent a follow-up status check last week and haven't heard back. Shot a tweet and developer forum post at them too. I was surprised to get any reply initially (I've tried reaching out to them in the past) but it gave me some optimism. I will keep on them. IIRC, back when GCIs were added in Android 4.2, they were the last OEM to properly report those too.. -Mike
  16. mikejeep
    I think I might have accidentally figured this out. Simple math: NSG EARFCN (66486) - SCP EARFCN (950) = 65536 65536 is 2^16, aka the max 16-bit integer. The API specifies it is a 18-bit integer, but I'm guessing something buried inside the OS got messed up along the way. Not sure if it's an HTC glitch or an Android glitch, but I found a similar report discussing issues with EARFCNs above 47000, so I added to it. If anyone wants to 'star' this on the Android Issue Tracker, it might help get the Android developer team's attention. -Mike
  17. mikejeep
    Sorry this took so long to compile, but here are some more details about the changes in the latest version of SignalCheck.. alphabetical for simplicity: Added automatic site note correlation between Verizon B2/B4/B13 LTE sites. (Pro) Much like Sprint B25/26/41 site notes are shared, Verizon site notes are now shared between bands too. Added Change Log display on first app launch. Self-explanatory.. also accessible from the Help screen. Will be updated each release. Added indicators for many AT&T, Sprint, T-Mobile, and Verizon LTE bands/cells/carriers. Mostly self-explanatory.. these are all PLMN+GCI-based. Acronyms are all listed in the glossary on the Help screen. Small cells, DAS sites, Magic Box, Mini-Macro, etc.. good stuff. Added LTE band display (and option to include EARFCN) to neighbor cells. (Pro) Implemented native LTE EARFCN display and band identification on Android 7+. (Pro) Android 7+ devices will at last display LTE EARFCN information in SCP without root, hard-coded patterns, or any other special magic -- as long as the manufacturer has properly implemented the routines. Unfortunately, Samsung has not done this on most of their devices; hopefully they get their act together. Added visual notification when any alerts are triggered on Android 6+. (Pro) Audio/vibrate Alerts will not trigger in newer versions of Android without a visual notification to the user. For now, this means a dismissable notification will appear in the status bar. I am working on gutting the Alerts code and hopefully will have a better solution soon. Implemented Android runtime permissions requirements. On Android 6+, users will now be prompted to explicitly grant Location, Phone, and Storage permissions to the app due to new Android security requirements. These are not new permission requirements -- each has been in place since nearly the beginning of development. It is simply a new method of requesting they be granted. Failure to grant all of these permissions will prevent the app from functioning properly. There are (crude) protections in place to shut the app down gracefully, but force closes may result if permissions are denied but users still access the app. I am working on improving this handling. Please ask if you have any concerns or questions about permissions! Implemented basic privacy policy. SignalCheck does not share any data beyond typical crash analytics, however a basic policy was implemented to satisfy Google Play requirements. Removed calculated indicators for T-Mobile LTE bands in some markets due to deployment inconsistencies. I received detailed information about T-Mobile's US network, and learned why some users saw invalid band information calculated from the PLMN+GCI. In the interest of accuracy, the app will no longer attempt to calculate the LTE band when connected to T-Mobile sites on the East Coast or in most of the West. Compatible Android 7+ devices will show any band/EARFCN information available. Click here for full details. Resolved issue with device freezing while Site Logs were exported. (Pro) Resolved issue with GSM Site Log export failing. (Pro) The export function will now run in the background instead of locking up the display. And GSM logs are finally fixed! Resolved issue with incorrect LTE site notes displaying if database had been manually edited. (Pro) Advanced users that make manual edits to site notes in the SCP database should not have future issues. Resolved issue with Location Service restarting when power is connected. (Pro) In certain situations, connecting a charging cord to the device would start (or restart) the app's Location Service even if the user did not have this option enabled. Updated launcher icon to material and round designs. Yay, new icons!! Changed AT&T LTE DAS indicators to “iDAS” (indoor) or “oDAS” (outdoor). Changed PLMN 311940 to Clearwire. Improved identification of Sprint LTE band 25 second carriers in certain regions. Improved identification of Sprint LTE connections for prepaid users. Resolved force closes with Location Service on devices using non-Latin characters. (Pro) Resolved issue with LTE PCI 0 not displaying. Resolved issue with provider display showing PLMN instead of name in certain situations. Resolved issue with some LTE band 30 notification icons not displaying. (Pro) Significant behind-the-scenes code optimizations. Updated help screen. These should all be self-explanatory, but feel free to ask questions. Back to working on the next update.. thanks to everyone for their support!! -Mike
  18. mikejeep
    Question for the masses.. I am working on adding more notification icons for international LTE bands, and I noticed that when choosing to display "LTE [freq]" with the RSRP value, the text is very small. The font is as large as possible to fit within the area provided. Nobody has asked for it, but would anyone prefer the "LTE" text be omitted if displaying like this? That would allow me to match the font used when showing "LTE [band]", which is more visible at small resolutions. This would only impact the particular notification icon configuration that displays frequency and signal strength at the same time. To summarize, we'll call these Option A (keep things as is) and Option B (get rid of "LTE" part). See below for examples. Option A: Option B: Any thoughts? Doesn't matter to me (I prefer to show the band), just wanted to toss it out there. -Mike
  19. mikejeep
    A few of us have seen odd blips of data like that since I added the native EARFCN routines to SCP. The PLMN is most likely wrong, but the rest of the data is probably correct. jefbal99 reported some "Sprint" sightings on the wrong bands in Michigan that pointed to T-Mobile, and I've seen "Sprint" connections around town that I tracked to AT&T. Our devices are probably catching quick glimpses of other networks while scanning for signals. I know in my case it happens when I lose all native (Sprint) service. Sometimes not all of the data from the OS updates at the exact same moment, so you see some stale info with some new info like that. When you sent the diagnostics, you were connected to PCI 71. I reached out to someone more familiar with the AT&T network that will hopefully be able to share some insight as to what you are seeing. -Mike
  20. mikejeep
    Your diagnostic report didn't show anything out of the ordinary at all.. this was what was displayed on the screen at the time: Provider: AT&T B2 (310410) GCI: 011C0B08, PCI: 71, TAC: 4633 UL EARFCN: 18850 (1875.0 MHz) DL EARFCN: 850 (1955.0 MHz) All of the background data matches up with the display. EARFCN 850 is what the OS was reporting at the time; that is band 2, and the GCI matches the band 2 pattern. I am confident that SCP is showing reliable data, at least according to your device. I don't know where NSG is getting its information. -Mike
  21. mikejeep
    Yes.. the diagnostic report (About > Send Diagnostics) takes a snapshot of the data present on your device at the moment you submit it. I would like to see it though. NSG accesses the modem directly with proprietary methods, while SCP relies on what the Android OS is reporting. -Mike
  22. mikejeep
    And all the neighbor cells in the SCP screenshot are reporting Band 2, leading me to believe that is most likely what the true connection was at that moment, since Nougat doesn't seem to show neighbor cells on other bands. But definitely weird across the board there. SCP shows B4 because the app prioritizes hard-coded band patterns over EARFCN calculations (mostly so that "special" indicators like additional carriers or small cells can be shown). -Mike
  23. mikejeep
    1) The app tries two different methods to get GSM-based neighbor cells.. one returns a PLMN+LAC+CID, and the other returns only a PSC. PSC is not logged or labeled because no PLMN is returned with that method, so there's no way to narrow it down enough. 2) I will check that app out, thanks! Makes sense; as Flompolph mentioned, it's following the GCI pattern for that. I need a diagnostic report from Tengen31 while connected to these sites to see how I can address this. -Mike
  24. mikejeep
    Sprint tends to get the most goodies because the S4GRU crowd gives me the most information. I'm more than happy to add anything I can, but I need to know where to find the info. Feel free to let me know if you have anything for whoever your current provider is! I had never heard of Patreon, that's interesting. But that highlights one of my biggest concerns.. I already feel that I have an obligation to continue improving the app indefinitely because so many people have purchased it. Establishing a voluntary fee structure or soliciting donations increases my anxiety/guilt when I don't have enough time to get timely updates out. I understand where you are all coming from but I get more angry e-mails than you would expect from people expecting instant 24/7/365 personalized support and daily app updates because I got $1.39 from them three years ago. imex99, you get me that Airave 3 and I'll buy YOU a beer.. Could a Mini-Macro fit in the enclosures you believe are small cells? The GCI you shared matches the MM pattern that has been identified, which is why SCP handles it that way. But that doesn't necessarily mean something hasn't changed. Things are constantly evolving. -Mike
  25. mikejeep
    I saw Tengen31's post in the main SignalCheck thread first and already replied there. The app should report 311870 as Clearwire unless the device is registered on Boost Mobile. Is Tengen31 a Boost subscriber? If I had that noted somewhere, it fell through the cracks.. -Mike
×
×
  • Create New...