Root [MAJOR SOLUTION] Fix your slow 3g speeds:EVO4g

[Napegator]

Does this fix also work the the HTC Hero? I get crappy speeds of 150-230 speed on the hero.

I saw this: Rooted ASOP: results may vary but some have had success restoring a sense backup, and then going back to ASOP. Don't wipe.

Dont wipe anything? Just restore the backup over the sense one?

Thx

anyone out there that might shine some light on this?

 

[sisspc]

Ummm...

before (up/down):
126/045
126/049
122/047

after (up/down):
060/025
089/041
106/039

I'm on Myn's Warm 2.2, if that makes a difference and I was already at PRL 60675.

 

[MizzouBrent]

I haven't tried this yet, but here is what I am getting (on a side note, how do I test 3g signal strength?)

3G - D/U
874/546
537/695
1170/656
1102/590

4G
1869/879
3964/895
1451/962
1895/961
1312/648

And just for kicks, here are my wifi readings
3319/694
3828/683
3631/687
3955/690

What is a reasonable expectation of Spring 3G speeds?

 

[ovrrdrive]

Do you happen to know what your PRL version was at the start? I'm wondering if it makes a difference if you actually ended up updating to a new PRL version.

It was already at 60675 before I changed anything...

 

[scudder]

Hello!!

I'm posting a data set for everyone to look at. HOPE this might be of some use, or useful for comments/suggestions. Also hope my presentation isn't too confusing

Last night I tried Mr. Ed's suggestions. I did speed tests...all utilizing ONLY the speedtest.net app so that the data is "consistently inconsistent" with the speed testing server and app as I present the data. I did tests BEFORE implementing changes and then you'll see a whooooole bunch of tests after!

I made a few conclusions at the end but by no means scientific or conclusive. However, NOTICE my tests at "the restaurant" during lunch were I was in the SAME SPOT testing various conditions. SORRY it is a long post but I'll summarize what I think is THE MOST IMPORTANT INFO right now. I tested, while sitting in the SAME SPOT (restaurant for lunch) the following:

(1) the PRECISE PROXY CHANGES suggested in this post;
(2) changes in this post but FORCE-UPDATING THE PRL; and
(3) REVERTED BACK TO ORIGINAL PROXIES

And here are those results!!

Speeds AFTER IMPLEMENTING CHANGES:
D: 316 U: 93
D: 410 U: 53
D: 271 U: 193

Speed after changes to proxy (zeros) but PURPOSEFULLY UPDATING PRL (a "No-no!")
D: 249 U: 148
D: 325 U: 136
D: 374 U: 169

Speed after REVERTING BACK TO ORIGINAL PROXY SETTINGS
D: 259 U: 139
D: 440 U: 219
D: 363 U: 225

SCOTT'S CONCLUSION: not much of any difference in download speed comparing any of the settings. I had a DECREASE in UPLOAD when I followed this post and seem to have had the BEST UPLOAD speed with the ORIGINAL proxy settings...



And here's the rest of the longggg data set...with what/how I did and where/when...

Last night I speed-tested my phone in 3G "AS WAS"...with NO MR. ED CHANGES from 1047pm to 1130p. Data includes one WIFI test at 1132p. I would throw out the 1133p test because it is much worse than previous 3G tests.

I then FOLLOWED THE POST and implemented the changes...

1145p and 1150p were first tests AFTER implementing the changes.

I continue testing after 1145p and 1150p, including a couple of WIFI tests, then back to 3G at 1224a, 1225a and one more at 209a.

I then PASS OUT from EXHAUSTION



After beddy-bye time, I wake up and STILL HAVE THE CHANGES from this post on my phone. From 827a to 844a, I test while driving at random areas on the road during my commute to work (about 11 miles). Again, all in 3G. I then test while AT MY OFFICE 1206p, 144p, 145p, 150p, 153p.

I then leave to go to get some lunch. NO LONGER at my office, I now do three tests AT THE RESTAURANT...again, with the MR. ED CHANGES...from 205p to 206p. Here are the images of the speed tests from my commute and my time at work:

I then ON PURPOSE force-update my PRL and test the speed again 3 TIMES from 215p to 219p. I then SCRAP MR. ED's CHANGES in the original post...STILL sitting in same spot at the restaurant....

And THEN...and HERE IS THE KICKER....while AT THE RESTAURANT, I RESTORE MY PHONE back to what it was WITHOUT ANY CHANGES from this post. I redo the profile, redo the PRL. I then do the ##3282# and change the first proxy to 8085, change the second proxy BACK to pd.vog.sprintpcs.com, exit and reboot the phone, allow Hands Free Activation AS WELL AS PRL UPDATE and whatever else might have happened. I then do THREE MORE tests from 229p to 231p.

NOTICE that there is NO REAL download increase once I went back to the ORIGINAL 8085/sprintpcs.com proxy settings but a NOTICEABLE INCREASE in upload speed!!

From 259 on, I tested while ON THE ROAD, during my 7 or so mile commute home. I would THROW OUT 304p as an anomaly, as would I with 338p.

From 338p on I am now IN MY HOME testing 3G and WIFI. Again, I have SCRAPED the MR ED settings, I am BACK to the 8085/sprintpcs.com proxy settings. NOTICE how much better my 3G speeds are at home THIS AFTERNOON compared to my original speeds LAST NIGHT before I did anything...NOTHING to do with the settings, or going back to my original settings, I don't think, but weird!!

FEW CONCLUSIONS I have:

1. 3G connectivity and strength varies WIDELY depending upon location. While I was able to conduct a test while "standing still" at the restaurant, I wish I could have...but it was impossible...to test Mr. Ed's settings and the ORIGINAL proxy settings WHILE ON THE MOVE.

2. It does appear that I have NO NOTICEABLE INCREASE in speeds once this post's changes were made I think the TELL-TALE DATA is when I was at the restaurant, sitting in one spot...tested the changes in the post, updated PRL as warned NOT to do, then totally reverted back to the original proxy settings.

3. Lastly, on a personal note, my WIFI in my house SUCKS!! 1400-2800!!?!? I gotta call U-VERSE...I'm supposed to be TEN TIMES THAT (which I have had in the past)

Again, hope this info might be helpful to someone...even if just applicable to my personal experience and location!

 

[Mr. Ed]

I have never said this was going to work for everyone nor that speed test should be used for accuracy or consistency haha ..I do recomend two different tests that I have used in the post.

I am curious of what the mileage of travel is. Connection quality will change from tower to tower. Currently I am trying to find a pattern, or at least map of tower locations in my area.

This is an ongoing project of mine...and though it has increased the speeds at several locations during my 45 minute commute...my speeds at work and at several other spots were piss poor.

No matter what I did while at work...in any given order, had much of an effect on the speeds or connection quality. While I am at work I can't even get webpages to load.

My settings intact I left and sporadically tested on the commute home ...I found several locations that I previously had low speeds were now around 1 Mbps...and others remained the same super low speeds. Got home and speeds were right as they were last night. 1.0 + mbps

Conclusion...there are some tower issues. With that not everyone is going to see increases.

For the record the http proxy settings have very little to do with this. The proxy settings are a fix suggested by Sprint to fix streaming errors in music streaming services like Pandora, Slacker, Sirius/XM, and Rhapsody. Some users noticed a slight increase in speeds when applying that fix so I figured you might as well do that while in the EPST menu.

The main factor is updating the PRL after updating the Data profile..in my continued testing today it still holds ture on my phone, if you update the PRL after the data the speeds suffer.

 

[Pioneer74]

This slowed my dl speeds to a crawl. Had to restore the original lines to get the speed back.

 

[Mr. Ed]

The lines have nothing to do with the fix. That is a streaming audio fix that is currently being used by sprint. I think I have mentioned 5 or 6 times now lol.

Wouldn't this be awesome

 

[scudder]

I have never said this was going to work for everyone nor that speed test should be used for accuracy or consistency haha ..I do recomend two different tests that I have used in the post.

Understood!! I was one of your VERY early readers and "tryers" of the information in the original post...so early, in fact, that I dialed your "##3283# but realized before the correction that ##DATA# was what you meant !!

While I never thought that you meant this to be THE fix I would be dishonest if I didn't have GREAT HOPE that this might be the "holy grail" of my slow 3G issues when I read the thread title:

[MAJOR SOLUTION] FIX YOUR 3G SPEEDS: EVO 4G

Not for me...oh well...life goes on... But i DO appreciate the recent addition of yours with "Your Results May Vary" in the OP!!

Anyway, my intention with the post of my speeds was NOT AT ALL too poo-poo your efforts. QUITE THE CONTRARY...I have FOUR other EVOs in my family/close circle of friends and I INSISTED they try this fix...and did so after my results!! I simply wanted to provide a data set with data collected while certain variables were constant (e.g., my location at a restaurant during lunch, or my settings) while some variable changed (my location).

My hope was that either YOU could use the data in further investigation into what works, what doesn't, where and why...or that other people with similar disappointing results might see that it similarly didn't work for not just them

This is an ongoing project of mine...and though it has increased the speeds at several locations during my 45 minute commute...my speeds at work and at several other spots were piss poor.

No matter what I did while at work...in any given order, had much of an effect on the speeds or connection quality. While I am at work I can't even get webpages to load.

Interesting....might explain why I had such varying results on my commute...?? I'm still trying to figure out while I had a FOUR-FOLD increase in my speeds here at home from last night to this afternoon...??

For the record the http proxy settings have very little to do with this. The proxy settings are a fix suggested by Sprint to fix streaming errors in music streaming services like Pandora, Slacker, Sirius/XM, and Rhapsody. Some users noticed a slight increase in speeds when applying that fix so I figured you might as well do that while in the EPST menu.

The main factor is updating the PRL after updating the Data profile..in my continued testing today it still holds ture on my phone, if you update the PRL after the data the speeds suffer.

Forgive my naivete on the subject...actually UTTER IGNORANCE with regard to the technical aspects of the fixes you suggested ...but perhaps...if what you say above is true regarding the possibility that the results might not have much to do with the proxies and/or making sure you do NOT update the PRL after confirming changes to the phone data...the same increase in speeds my occur if one simply "goes through the motions" of your ##3282# fix but INSTEAD of changing the proxies to "0" (zeros) just "saves the data"....instead KEEP the 8085/sprintpcs.com, save the data AS IS "go through the motions", reboot the phone and let it boot back up. If the proxy settings have NOTHING to do with the speed increases but, in the same way that the PRL update process LOWERS the 3G speeds for some, perhaps the mere re-saving of the data, confirming the change and rebooting the phone process might be the key to INCREASING the data speeds.

Again, apologize...an UNtrained mind thinking out loud.

If what I am saying above won't work because of my general lack of knowledge about the technical aspects of the changes (sorry!!) then I was wondering why the need for the 0's (zeros) in the proxy settings you suggest??

I'm just trying to understand a bit more than I do....THANKS for your time...I'm on YOUR SIDE!!

 

[Mr. Ed]

I am unclear on the exact reason why the proxy settings fix streaming issues as well.

In theory..yes one could omit that and get the better speed results too. The updating of the PRL after the forced data restore seems to be the culprit and canceling that appears to be the solution. THAT IS THE IMPORTANT STEP (on most phones lol)

Better streaming is just an added benefit while you are in the epst. In the cases/threads that I pulled the HTTP info from, streaming services either skipped horribly or did not work at all, applying the 0 in each spot corrected it. Another reminder...changing the http settings is not new info...it has been out for weeks...and is being pushed to phones by sprint.

Your information is indeed helpful, and is very similar to the multitude of tests that I did prior to posting "the fix"

stationary locations, and then stationary at different tower locations and then while commuting. Exact tests run by me...

performed steps as typed got faster speeds

performed steps and updated prl slower speeds

performed steps as typed got faster speeds

performed steps and updated prl slower speeds

then wiped everything and flashed a different rom and did the same, got the same results

then wiped everything and flashed back to my original and repeated everything, got the same results ..I did this for hours last night from my driveway sitting in the car, then from inside the house, then again in my garage.


as for commute
It is important for folks to realize that you travel from tower coverage to tower coverage daily....not sure that some realize that. I most likely pass through 25 towers on my way to work. ( I would really like to know where they all are and correlate that to speeds)

it's also important to note that speedtest and test my speed are merely tools for comparison...not exact speeds.

In your case your results varied drastically. I would try clearing data for speedtest, even consider exiting and re entering the app between each test (or swap servers back and forth) and maybe even deleting and re downloading the app. though as you can see in my results above the tests will vary ...but it's obvious when sitting in the same spot and the speeds go from well below 1 Mbps to 1 + Mbps that something is working differently. They are not going to vary that drastically.

The only changed variable being the PRL update after data being done, or not. In some of my tests http settings were original others they were 0

 

[scudder]

It is important for folks to realize that you travel from tower coverage to tower coverage daily....not sure that some realize that. I most likely pass through 25 towers on my way to work. ( I would really like to know where they all are and correlate that to speeds)

I fooled around with the following app months ago...just to "take a look around" but it had some limitations with 3G/CMDA as stated in the app description. Perhaps this...or something similar might of use to you??

https://market.android.com/details?id=com.technolatry.antennas

Thanks for the thoughts about speedtest and clearing the data/quitting/reinstalling. I TRIED to be consistent with using the same app in the same location and didn't really care if the results indicated 20 kbs or 24,367 Mbs. There numbers were of no real concern as much as the fact that, for me, the numbers really didn't change significantly.

 

[Mr. Ed]

Scudder what kernel are you using?

Thanks for the app I will check it out.

More testing and I have added a variable....kernel. so far java kernels decrease speeds ie no change. Barely able to keep my eyes open so no more tests tonight.

If this no workie for you...try a different mkernel

 

[autorail1]

I did the steps to a T. My PRL is still at 60675. Let me try again.

 

[scudder]

Scudder what kernel are you using?

Thanks for the app I will check it out.

More testing and I have added a variable....kernel. so far java kernels decrease speeds ie no change. Barely able to keep my eyes open so no more tests tonight.

If this no workie for you...try a different mkernel

I don't even know the answer to that question...see how advanced I am?? I am rooted but have NO custom ROMs...all stock...never flashed anything except some Font Packs.

Here is my kernal info from my Software Info:

2.6.32.15-ge2gb08e
htc-kernal@and18-2 #11

Any kernal suggestions...if I happen to get bold??

 

[Mr. Ed]

I did the steps to a T. My PRL is still at 60675. Let me try again.

omgka this is not going to change your prl. It should still be 60675 when you are done. Please someone tell me what to add so that people quit asking why their prl did not change.

I don't even know the answer to that question...see how advanced I am?? I am rooted but have NO custom ROMs...all stock...never flashed anything except some Font Packs.

Here is my kernal info from my Software Info:

2.6.32.15-ge2gb08e
htc-kernal@and18-2 #11

Any kernal suggestions...if I happen to get bold??

Ok# 11 is an older kernel. What software version are you on? 3.29?

Regardless try out netarchy 4.1.9. I will provide a link when I get to my pc at work in a few.

 

[autorail1]

LOL, I read it later in red. DAMN!!

omgka this is not going to change your prl. It should still be 60675 when you are done. Please someone tell me what to add so that people quit asking why their prl did not change.



Ok# 11 is an older kernel. What software version are you on? 3.29?

Regardless try out netarchy 4.1.9. I will provide a link when I get to my pc at work in a few.

 

[Hunterbeav]

omgka this is not going to change your prl. It should still be 60675 when you are done. Please someone tell me what to add so that people quit asking why their prl did not change.



Ok# 11 is an older kernel. What software version are you on? 3.29?

Regardless try out netarchy 4.1.9. I will provide a link when I get to my pc at work in a few.

Ok so let me get this straight my prl should change or sould not change?............................................lol just kidding

 

[Mr. Ed]

[GPL][Kernel]netarchy-toastmod, Stable: 4.3.1 Beta: 4.3.2 - xda-developers

Kernels... I uses sbc 4.1.9 no havs.

Really curious to see if some that got no results were to change their kernel to a different kernel (no havs even) and try again what the new results would be.

 

[scudder]

[GPL][Kernel]netarchy-toastmod, Stable: 4.3.1 Beta: 4.3.2 - xda-developers

Kernels... I uses sbc 4.1.9 no havs.

Really curious to see if some that got no results were to change their kernel to a different kernel (no havs even) and try again what the new results would be.

Thanks!! Never flashed a Kernal before but, hell, I'll try! I just need to find some time. Not quite on the priority list at home here but want to try at some point soon.

If someone tries this before me, please POST RESULTS!!

 

[jblazea50]

Thanks for the tutorial. This seems to be working. I was getting around 0.06-0.10MB download. Now, it is 2-3x faster.

The verizon prl works the best, but not sure about additional charges for roaming that might occur. So will keep this and see if the speed remains constant.

Thanks

 

[mdizzle99]

Change HTTP PD Proxy port from 8085 to 0
Change HTTP PD Proxy port from pd.vog.sprintpcs.com to 0
This isn't new info...we are just doing this while we are in there. Some have reported a small increase in speeds after doing it...but nothing as significant as what NOT UPDATING the PRL after Data has shown.

Small note to start off...this is very minor but figured I'd still mention it. The part I bolded should read "Change HTTP PD Proxy Address from..." and not "Change HTTP PD port from..."

Like I said, no biggie, but there it is. Now onto my real reason for the post. Did anyone else have Pandora stop working after doing this fix? Pandora would start up, even show what song was "playing," but it would never actually play anything. When I went back and changed the proxy address to 0.0.0.0 instead of simply 0, Pandora started working again.

Second reason for this post. Unfortunately this fix did not help my 3G speeds. It made my 3G slow down quite a bit and become less stable. After updating my prl, it seems my speeds go back to where they were.

So for the past few weeks I have been running with proxy port set to "0" and proxy address set to "0.0.0.0"
I do believe doing these changes and restoring data did help me a bit, but I also needed to let the prl update.

 

[Mr. Ed]

thanks!

some have reported that changing to 0.0.0.0 works better, I have been using 0 since that fix first surfaces in Jan with no issues yrmv

you are the second or third to mention that you had to update the prl
like i mention in the first post...results are going to vary...for the most part the method works for a number of phones. it isn't going to be the case for all phones, as there are too many variables...the most notable being the different towers. I am not sure how much of the thread you have read but on my phone specifically...it has consistently shown slower speeds EVERY time the PRL update is allowed to complete, and 1.0 Mbps plus EVERY time I skip that part. (now totaling hundreds of times ;P)

so like i said haha yrmv

thanks for pointing out the port thing!

 

[jbsg02]

So far this has been a success for me. I just got a 1629 kbps download from speedtest.net, probably the fastest speedtest I've ever had on 3G with my EVO. I'll test this over a few days.

 

[Mr. Ed]

thought I would give a quick update...as some of you know I have issues getting good speeds at work.. always have.

Called sprint today (2 hours on the phone geeeeez)

verified via four sources that multiple 3g outages exist across North Texas, all thanks to the winter storms we had a few weeks ago.

other than that I couldn't really get anything useful out of them, other than that they are working on a new update but they gave me no details at all

It was pretty cool..they let me know what towers I should be connected to here at work...even gave me their location ID's. Unfortunately I haven't found a way to use that info lol.

 

[evomattnc]

just wanted to share my odd results...I did this process as described and ran a test afterward. I saw marked improvement. I have horrible signal at work, as a matter of fact i roam while inside, Anyhow, i went outside and ran a test after this and got speeds around 1.5 immediately after the profile update. It was amazing.however, i just went and stood in the same spot and tested again without having done anything since the update and got .5-.6 every time i ran the test...odd