Jump to content

PedroDaGr8

S4GRU Sponsor
  • Posts

    689
  • Joined

  • Last visited

  • Days Won

    39

Posts posted by PedroDaGr8

  1. On 12/24/2020 at 11:38 PM, mikejeep said:

    A huge update to SignalCheck Pro has been finished and is rolling out on Google Play right now.. version 4.64 should be available for download sometime over the next several hours. A big thank you to the Beta Crew for providing tons of feedback and helping get this release stable!

    There is a long list of bugfixes and stability improvements, as well as some new features. One significant technical item to note is this is the first build targeting Android 10 since last spring. The previous attempt was rolled back because of issues with stale data on several devices; in testing, this has not been as severe. Please let me know if you experience any problems.

    The full change log is available in the app (popup after updating, or anytime under Help > Change Log), or at https://signalcheck.app/change-log. Besides lots of bugfixes and force closes resolved, here are the highlights:

    Added 5G-NR site notes and logging features. When a 5G cell is identified with a PLMN and NCI, a site note can be created and it will be saved in the Site Log database. If there is no PLMN+NCI (i.e. non-standalone 5G connections), a site note cannot be created and nothing will be logged.

    Improved reliability of Alerts. Some alerts, especially the 5G-NR alerts, were not consistently firing. A new 5G alert for band-specific 5G connections or a custom plain-text match has been added, identical to the existing alerts for LTE. If the 5G band is not able to be identified (i.e. non-standalone connections), the "All Bands" option must be selected or it will not trigger the alert.

    Added "Trail Logger" feature. The existing "Site Logger" records one entry per cell; this new log will record one entry every time the signal strength changes. This can be useful for some users, but should not be enabled when not needed -- in testing, it averaged about 900 entries per hour. If you were to leave this running for awhile, the database could get extremely large in a short period of time. There is likely some battery impact as well, however it is difficult to measure that.

    Added "hits" column to Site Log. This will reflect the number of times the particular cell has been logged.

    Added option to use alternate method to obtain PLMN ID. Android reports the connected PLMN via many different methods; SignalCheck typically the most reliable method, however other methods occasionally report different data. For example, T-Mobile is currently reporting some PLMNs differently; enable this option to use a different method. This is likely needed to see PLMN 312250 in the field. If your device is older or you experience issues seeing connection information with this option enabled, you may need to disable it.

    Improved some system shortcut options on Android 10/11. While improved in this version, not all of these features are working on newer versions of Android due to security and access changes. Hopefully I can continue to work on this.

     

    As always, please feel free to provide any feedback you may have. Thank you to everyone for your support, and Merry Christmas!

    I can at least confirm the alternate PLMN displays 312-250. It triggered today for the first time when I drove by a "keep" tower. Thanks for adding this feature!

    • Like 2
  2. 23 hours ago, ingenium said:

    Generally I don't think it matters which you connect to. I haven't seen a difference between them. It seems 310-120 might be used to regular data, and 312-530 is used for the IMS session for text and VoLTE. Or at least on VoLTE capable devices, some Android APIs report 310-120, and some report 312-530, so they're both likely being used somehow. Signal Check and CellMapper seem to always report 310-120, but Network Signal Guru always reports 312-530. 

    311-490 is only used for T-Mobile or ROAMAHOME. It's likely sticky and persists for a bit, but there should be no practical difference. 312-250 seems to only be accessible with a T-Mobile SIM. I don't know if they get routed any differently, or if it's like broadcasting multiple wifi networks from the same access point (and on the same vlan).

    312-250 makes sense in that it's an easy way to restrict access to a subset of sites. Same for the Magic Box one. It makes it very easy to restrict which sites can be magic box and small cell donors, and they won't even try to connect to one that they're not supposed to. You don't want the UE trying to connect or get handed off to a site it's not supposed to be on and get rejections. You risk getting dropped calls or data interruptions. 

    311-490 was likely the same, as a way to restrict roaming access to some sites. Then they rolled it out to all of them, and rather than reconfigure the roaming side, they just added the PLMN everywhere since the current config was known to work.

    I'm on ROAMAHOME and saw 312-250 in the engineering screen. 

  3. 2 hours ago, mikejeep said:

    I can't remember if I posted it here or elsewhere, but Android reports PLMN IDs several different ways. Right now, SCP uses the one reported by the cell you are connected to. Back in the day, I chose this ID because it proved to be more reliable than the other methods.

    I realize why seeing 312250 would be useful (I have an alert set for it myself and haven't seen it). It definitely seems like T-Mobile has not configured the cell method to report this, for whatever reason. So, I am willing to add an option to use an alternate PLMN method. The app already recognizes 312250 as synonymous with all other Sprint PLMN IDs so site notes should not be negatively impacted.

    Thanks Mike!

    • Like 1
  4. 1 hour ago, Dkoellerwx said:

    Yep sites with that PLMN allow you to connect whenever in range and is treated as the same T-Mobile network. Working well here in Omaha in T-Mobile low spots that have Sprint keep sites.

    I find it interesting that the tower reports AT LEAST two different PLMNs. I noticed if I was connected to a T-Mobile (310-260) tower before connecting to a 312-250 tower, then the tower will report 312-250. On the other hand, if I switch from a Sprint tower which is reporting either 310-120 or 312-530, then the same tower will report 312-530. This makes it really hard to find these towers, since locking my signal to any Sprint band will inherently force me onto 310=120 or 312-530. 

  5. Hey Mike, quick question about the PLMN that SignalCheck reports. I am currently connected to a tower which just got the 312-250 PLMN (see screenshot) but SignalCheck is reporting it as 310-120. I don't know if this is a Samsung and/or T-Mobile thing where they change the PLMN before reporting to the system or what. I just know if I wanted to find the 312-250 PLMNs, not being able to see them or alert on them in SignalCheck makes it a LOT harder. 

  6. On 11/29/2020 at 2:52 PM, RAvirani said:

    It's rolling out slowly.

    You'll be able to catch it at Westwood Village shopping center in West Seattle. The Sprint site in Startup as well as the Sprint site at Snowqualmie Pass are some other examples of 312-250 sites. 

    Two Woodinville towers (SE03XC157 and SE35XC004) just started broadcasting the 312-250 PLMN.

    SE35XC004 (near the Slough in Woodinville) = This one is not surprising. That tower will fill a relative hole for T-Mobile and increase capacity for the parks/industries/etc. nearby. 

    SE03XC157 (Woodinville warehouse district) - I am honestly very surprised (and happy) that this tower was kept considering T-Mobile has two towers within a half mile of this one. While one was clearly going to be kept (the one in downtown Woodinville), the other at 47.76079, -122.13790 is just up the hill from SE03XC157. Based on my experience in the warehouse district area, the second tower was honestly horrible and, despite its location, didn't really cover the warehouse district at all. As a result, indoor speeds were a fraction of what I would get from SE03XC157. So from a coverage perspective, it makes sense to keep SE03XC157 but I didn't expect it to actually happen.

     

     

  7. 1 hour ago, PedroDaGr8 said:


    On an interesting note, my phone (Note 20 Ultra) is back on the Sprint network this morning without me forcing anything. I can force it onto the T-Mobile network (310-260) as there are bands available (though they are around -110db), so I still have access a la ROAMAHOME. I am unsure if it is preferring Sprint again or they changed the preference to actually switch between networks before the data coverage gets REALLY spotty.

    Never mind, the tower I was connecting to just got the 312-250 PLMN. It wasn't showing it earlier this morning but is now. I guess me connecting to that tower this morning was related to that. 

  8. 26 minutes ago, RedSpark said:

    I give the engineering team a lot of credit. They did what they could with what was leftover from the marketing budget.

    As the saying goes: "Bad companies make good and efficient managers"

    On an interesting note, my phone (Note 20 Ultra) is back on the Sprint network this morning without me forcing anything. I can force it onto the T-Mobile network (310-260) as there are bands available (though they are around -110db), so I still have access a la ROAMAHOME. I am unsure if it is preferring Sprint again or they changed the preference to actually switch between networks before the data coverage gets REALLY spotty.

    • Like 1
  9. On 12/14/2020 at 12:45 PM, RAvirani said:

    Getting rid of SE03314B and keeping the other 3 would work out best. Three sites will more than cover the area's capacity needs, and it's not difficult to make use of downtilt and 45° antennas (as opposed to the usual 65° antennas) to keep the SNR from dropping due to inter-site noise. 

    The number one problem with T-Mobile in the Seattle market today is that their signal to noise ratio is consistently terrible (by comparison to the other carriers). This is not due to congestion but due to overlapping sectors. It seems the Sprint team was largely retained (as all my former contacts are staying with T-Mobile) and they're working to clean the network up as much as possible.

    EDIT: New imgur link

    https://imgur.com/a/Op6GVHs

    That is very good to hear! I had noticed that the SNR numbers for T-Mobile look notably worse than what I remembered for Sprint. I thought it was in my head/due to tower distances but clearly it is not. Thanks a bunch for these updates!

  10.  

    26 minutes ago, RAvirani said:

    I saw that permit as well. I honestly think that site should be decommissioned - it does more damage to the SNR of other sites than good. 

    There's another colocation planned in that area on the same building as the AT&T site (47.7095747, -122.1880277). That should be interesting!

    Speaking of the Evergreen Hospital site, looks like THAT one is slated for upgrade as well!

    https://permitsearch.mybuildingpermit.com/PermitDetails/WIR20-00679/Kirkland

    While this application doesn't explicitly state T-Mobile, the applicant "GARY ABRAHAMS" seems to be a T-Mobile contact. Searching for his name brings up only T-Mobile permits. 

  11. 3 minutes ago, RAvirani said:

    I saw that permit as well. I honestly think that site should be decommissioned - it does more damage to the SNR of other sites than good. 

    There's another colocation planned in that area on the same building as the AT&T site (47.7095747, -122.1880277). That should be interesting!

    Agreed, about that site. It is the LEAST optimal site I have ever seen for a tower. It sits so low that it doesn't cover barely any area. Yet now they are adding the highest BW B41 antenna to that location. I am very curious if it is even able to reach the majority of the new Totem Lake development. It would have made FAR more sense to add that antenna to ANY of the other sites.

    I saw that addition, that will make four towers all serving roughly the same exact area. 

    1. The aforementioned Totem Lake site

    2. The new site you mentioned (seems more logical for the B41 antennas). 

    3. The site atop the Evergreen Hospital (maybe even more logical for the B41 antennas).

    4. The site an exit away, by the Cross Kirkland Corridor, which is also a colo with AT&T. Despite being farther away, its reasonable height means  it serve much of the Totem Lake area.

     

     

  12. 13 hours ago, RAvirani said:

    Oops - I totally misread your message. Permits are in to put 2500 on the downtown Kirkland site (SE02100T). That's why my mind went there. Nice find!

    I hadn't been able to find officially what was being upgraded at the Livengood site. Considering B41 and mmWave are the only bands missing from that site (plus downtown Kirkland has no TMo B41 sites), I figured it would be B41. I am still incredibly surprised that TMo did not install B41 when they installed the Kirkland Urban site.

    The building behind the Totem Lake McDonald's (by the 405) is slated for a 2500 upgrade as well. It looks like it will be getting a Nokia AEHC:

    http://permits.kirklandwa.gov/WebDocs/2018064012/3e1ee991-9ca4-469c-b220-898d43f224c9.pdf

  13. 9 hours ago, RAvirani said:

    Interesting. This is the info I got:

    https://imgur.com/a/bVymrPP

    Still live this morning, so not a late night beta. Maybe a staggered roll out which will be complete in two weeks.

    So far, confirmed in Bellevue, Bothell, Kingsgate, Kirkland, Lynnwood, Shoreline, and Woodinville. The user in Shoreline is reporting notably better signal overnight as well (from 1 bar to 3-4 bars).

    • Like 1
  14. On 11/2/2020 at 8:17 AM, mikejeep said:

    New versions of SignalCheck Pro and SignalCheck Lite were pushed out late last night and are now live on Google Play! This release has a lot more changes behind the scenes than new features, but there are some improvements. The most significant changes are improved 5G support for Lite, and GCI pattern matching and offset rules for site notes on PLMN ID 312250, which is popping up on Sprint sites that are expected to stay in place.

    This release was squeezed in minutes ahead of last night's deadline imposed by Google Play requiring app updates to target Android 10 or newer. Many will remember that I tried to do this last spring, but it caused a lot of headaches and it was rolled back. Hopefully I can produce better results with the next attempt!

    I like the change from 0MHz to ? MHz. I think that makes much more sense. 

     

    I've never been able to nail down a pattern, so I'm guessing it's a Samsung issue of some sort. 

    • Like 1
  15. On 11/6/2020 at 8:41 AM, PedroDaGr8 said:

    Some more towers have been moved over to 1910/1990 for B25 this morning. I am starting to wonder if the towers on 1910/1990 are ones which will potentially be kept.  It seems like at this point, the only ones left on 1907.5/1987.5 are towers which will obviously not be kept. 

    And over the weekend, the remaining towers have been converted to 1910/1990. 

    • Like 1
×
×
  • Create New...