Jump to content

EVO LTE on CM10 nightlies - LTE Data session issue


aparish21

Recommended Posts

I have just recently got both the CM10 rom on my device along with LTE coverage near me and have an odd issue. I generally have wifi coverage so data isn't a problem, but I noticed once I picked up LTE coverage, if I turn off my wifi, my device says it is still connected to LTE but cannot establish a data session. even if i cycle though connection settings, reboot, etc....

 

Anyone else seeing this or similar issues

 

 

Also, is there anyway to update PRL on a mod?

Link to comment
Share on other sites

What I would do is backup the current ROM you are using. Then flash a stock rooted ROM back to your device, to see if its still having the same issue. Right now I'm running Decks CM10 and since LTE is sparse right now I'm wondering how we know if LTE is working on these custom ROMs the way it should..

Link to comment
Share on other sites

What PRL are you running and what is your need to change it?

PRL is 25006, the need to change was to see if another version would change result...

 

I somehow jinked up my nandroid of stock ICS and the phone won't boot to it. I ended up factory resetting and the re-flashing the CM rom I had on it (CM10 nightly 10_03) and that seems to have resolved for the moment. I am able to get a data session on LTE now, or at least the device says I do... speeds are avg 3-6 DL and 2UL so whatever connection I have isn't great. -108dBm on LTE

Link to comment
Share on other sites

Glad I read this because i was gonna start another post. Over the weekend I went to a site that was know to have NV upgrades done. I origianlly was going to flash back to a stock ROM but didnt have enough time. All day while there I could see that I had eHRPD but my data was totally jacked up the network state either reading disconnecting or disconnected. I didn't know this though until I got home later in the day and looked at my screenshots. What made this more frustrating is that at some point all towers I came across that day went eHRPD and I was trying to connect tyo veriy. This was all on Saturday.

 

That night I flashed by to a rooted stock Sense ROM to test again. This time no issues at all connected to eHRPD pretty easily, at night it seems to want to stay in 1x but that may be something else. This morning on the drive never disconnected and my data was fine. The problem with the CM10 and likely any AOSP ROM is that LTE is not saturated enough for devs to know it if works correctly.

Link to comment
Share on other sites

CM10 has an issue with the radio interface layer and will not connect to ehrpd for any sustainable amount of time. Lte works, but it is worse than a sense rom when it comes to connecting to it. I've been lurking around the irc channel and it seems like that's what they are concentrating on fixing. Even if they fix the bugs it's not going to connect to lte any faster until htc fixes the radios on our phones. I'm starting to believe that no threshold issue exists and the radio programming is horrible and doesn't know that lte is available.

Link to comment
Share on other sites

Apparently there is an issue with LTE -> eHRPD handoff associated with Qualcomm chips. I've heard this is due to how Sprint handles LTE authentication and that the issue isn't happening on VZW devices. The OEMs have come up with temp fixes which is why LTE seems to work better on stock roms but isn't up to par on AOSP builds.

 

This is also happening on the Sprint Samsung GS3.

Link to comment
Share on other sites

Apparently there is an issue with LTE -> eHRPD handoff associated with Qualcomm chips. I've heard this is due to how Sprint handles LTE authentication and that the issue isn't happening on VZW devices. The OEMs have come up with temp fixes which is why LTE seems to work better on stock roms but isn't up to par on AOSP builds.

 

This is also happening on the Sprint Samsung GS3.

 

Interesting. Can you explain more about how Sprint authentication works? Does this problem also affect the acquisition of LTE signal in the first place or after losing it? Does the OEM workaround involve the mysterious "LTE available" files we have been viewing through the hidden menus on both handsets?

Link to comment
Share on other sites

 

Interesting. Can you explain more about how Sprint authentication works? Does this problem also affect the acquisition of LTE signal in the first place or after losing it? Does the OEM workaround involve the mysterious "LTE available" files we have been viewing through the hidden menus on both handsets?

 

I believe those are used for scanning. Basically the phone remembers where LTE was. HTC didn't implement theirs correctly for some reason.

 

 

Sent from my C64 w/Epyx FastLoad cartridge

 

 

Link to comment
Share on other sites

I believe those are used for scanning. Basically the phone remembers where LTE was. HTC didn't implement theirs correctly for some reason.

 

 

Sent from my C64 w/Epyx FastLoad cartridge

 

But we know that the list in the "LTE_available" file is not a list of LTE towers.

 

And as far as the GS3 goes, it does not function perfectly, either. By all reports it outperforms the EVO 4G LTE in holding and/or acquiring an LTE signal. But as someome who has used a GS3 extensively for 4G testing, there also are many times when it will fail to make an LTE conneciton without toggling.

 

So if the root cause of all this is some general weakness in Sprint's authentication scheme, and autoprime knows more detail about this, I would like to hear it and where the information comes from.

Link to comment
Share on other sites

 

But we know that the list in the "LTE_available" file is not a list of LTE towers.

 

And as far as the GS3 goes, it does not function perfectly, either. By all reports it outperforms the EVO 4G LTE in holding and/or acquiring an LTE signal. But as someome who has used a GS3 extensively for 4G testing, there also are many times when it will fail to make an LTE conneciton without toggling.

 

So if the root cause of all this is some general weakness in Sprint's authentication scheme, and autoprime knows more detail about this, I would like to hear it and where the information comes from.

 

Correct, it is a list of 1x sectors as LTE doesn't use a SID. Just like Evdo doesn't.

 

Sent from my C64 w/Epyx FastLoad cartridge

 

 

Link to comment
Share on other sites

But we know that the list in the "LTE_available" file is not a list of LTE towers.

 

And as far as the GS3 goes, it does not function perfectly, either. By all reports it outperforms the EVO 4G LTE in holding and/or acquiring an LTE signal. But as someome who has used a GS3 extensively for 4G testing, there also are many times when it will fail to make an LTE conneciton without toggling.

 

So if the root cause of all this is some general weakness in Sprint's authentication scheme, and autoprime knows more detail about this, I would like to hear it and where the information comes from.

 

To be 100% honest, I am not sure of what the exact issue is with the LTE auth. The info I'm relaying comes from a friend on a dev team who works with both devices. I was told that Verizon isn't having these issues but Sprint has a "different setup".. and that there will be LTE connection issues.

 

This topic was only lightly discussed as it came up during another LTE discussion. I had assumed it was due to Sprint still relying on CDMA.. where as Verizon is using a more "standard" WCDMA/LTE setup on their LTE devices. This is just my guess.. and I do not know the true reason for this issue.

 

Adding a little more info... I was told this is something Qualcomm knows about and does not care to fix. So that tells me that the reason something "isn't working" is due to a non-standard setup.. so obviously thats not something Qcom wants to go out of their way to fix. But again.. that's just more guessing from me.

 

The only thing I had hoped to achieve in my original post in this thread was to let others know that Qualcomm, OEMs, carriers and XDA Devs are aware of this issue. How it will be handled is anyones guess. I'm just hoping the CM team finds a "fix" to make LTE/eHRPD a little more usable on AOSP roms. Sorry for the lack of advanced detail.. I know we all like that stuff. Of course, if I hear anything else I will update.

Link to comment
Share on other sites

To be 100% honest, I am not sure of what the exact issue is with the LTE auth. The info I'm relaying comes from a friend on a dev team who works with both devices. I was told that Verizon isn't having these issues but Sprint has a "different setup".. and that there will be LTE connection issues.

 

This topic was only lightly discussed as it came up during another LTE discussion. I had assumed it was due to Sprint still relying on CDMA.. where as Verizon is using a more "standard" WCDMA/LTE setup on their LTE devices. This is just my guess.. and I do not know the true reason for this issue.

 

Adding a little more info... I was told this is something Qualcomm knows about and does not care to fix. So that tells me that the reason something "isn't working" is due to a non-standard setup.. so obviously thats not something Qcom wants to go out of their way to fix. But again.. that's just more guessing from me.

 

The only thing I had hoped to achieve in my original post in this thread was to let others know that Qualcomm, OEMs, carriers and XDA Devs are aware of this issue. How it will be handled is anyones guess. I'm just hoping the CM team finds a "fix" to make LTE/eHRPD a little more usable on AOSP roms. Sorry for the lack of advanced detail.. I know we all like that stuff. Of course, if I hear anything else I will update.

 

Thanks. It is disconcerting if true that Sprint's scheme for LTE authentication is inherently flawed and relies on finding some workaround. In any event, from field performance so far -- included tests of the Samsung GS3 -- it is apparent that there is not yet a solid solution.

Link to comment
Share on other sites

But what about the Viper and iPhone 5 are they having issues?

 

Sent from my EVO using Tapatalk 2

 

The iPhone certainly doesn't have this issue. I'm also not aware of any other data issues, aside from some users who have reported the phone has the wrong time.

 

 

Sent from my iPad using Tapatalk HD

Link to comment
Share on other sites

But what about the Viper and iPhone 5 are they having issues?

 

Sent from my EVO using Tapatalk 2

 

According to Robert when he was in Waco the Viper was connecting to LTE better than the Evo and the GS3.

Link to comment
Share on other sites

 

According to Robert when he was in Waco the Viper was connecting to LTE better than the Evo and the GS3.

 

The Viper and GNex auto connected to LTE fastest. The GS3 a little slower. And the EVO LTE rarely connected to LTE on its own.

 

Robert via Nexus 7 using Forum Runner

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • large.unreadcontent.png.6ef00db54e758d06

  • gallery_1_23_9202.png

  • Similar Content

  • Posts

    • Since this is kind of the general chat thread, I have to share this humorous story (at least it is to me): Since around February/March of this year, my S22U has been an absolute pain to charge. USB-C cables would immediately fall out and it progressively got worse and worse until it often took me a number of minutes to get the angle of the cable juuuussst right to get charging to occur at all (not exaggerating). The connection was so weak that even walking heavily could cause the cable to disconnect. I tried cleaning out the port with a stable, a paperclip, etc. Some dust/lint/dirt came out but the connection didn't improve one bit. Needless to say, this was a MONSTER headache and had me hating this phone. I just didn't have the finances right now for a replacement.  Which brings us to the night before last. I am angry as hell because I had spent five minutes trying to get this phone to charge and failed. I am looking in the port and I notice it doesn't look right. The walls look rough and, using a staple, the back and walls feel REALLY rough and very hard. I get some lint/dust out with the staple and it improves charging in the sense I can get it to charge but it doesn't remove any of the hard stuff. It's late and it's charging, so that's enough for now. I decide it's time to see if that hard stuff is part of the connector or not. More aggressive methods are needed! I work in a biochem lab and we have a lot of different sizes of disposable needles available. So, yesterday morning, while in the lab I grab a few different sizes of needles between 26AWG and 31 AWG. When I got home, I got to work and start probing the connector with the 26 AWG and 31 AWG needle. The stuff feels extremely hard, almost like it was part of the connector, but a bit does break off. Under examination of the bit, it's almost sandy with dust/lint embedded in it. It's not part of the connector but instead some sort of rock-hard crap! That's when I remember that I had done some rock hounding at the end of last year and in January. This involved lots of digging in very sandy/dusty soils; soils which bare more than a passing resemblance to the crap in the connector. We have our answer, this debris is basically compacted/cemented rock dust. Over time, moisture in the area combined with the compression from inserting the USB-C connector had turned it into cement. I start going nuts chiseling away at it with the 26 AWG needle. After about 5-10 minutes of constant chiseling and scraping with the 26AWG and 31AWG needles, I see the first signs of metal at the back of the connector. So it is metal around the outsides! Another 5 minutes of work and I have scraped away pretty much all of the crap in the connector. A few finishing passes with the 31AWG needle, a blast of compressed air, and it is time to see if this helped any. I plug my regular USB-C cable and holy crap it clicks into place; it hasn't done that since February! I pick up the phone and the cable has actually latched! The connector works pretty much like it did over a year ago, it's almost like having a brand new phone!
    • That's odd, they are usually almost lock step with TMO. I forgot to mention this also includes the September Security Update.
    • 417.55 MB September security update just downloaded here for S24+ unlocked   Edit:  after Sept security update install, checked and found a 13MB GP System update as well.  Still showing August 1st there however. 
    • T-Mobile is selling the rest of the 3.45GHz spectrum to Columbia Capital.  
    • Still nothing for my AT&T and Visible phones.
  • Recently Browsing

    • No registered users viewing this page.
×
×
  • Create New...