SWMich4G

That wouldn't make any sense. The difference between 8GB and 16GB costs the phone maker what, $2 or $3? They don't make smaller versions so they can price them down - they make bigger versions (relative to the smaller offering) so they can price those up. Releasing a 8GB model instead of a 16GB model would not even factor into the price point they choose for the base model, the cost is so insignificant. The decision between 16 and 32 would probably be based almost solely on what the competitors are offering.

Several of the sites are marked live lte on that route. Unless you know all of the purple appeared at once it could be from several trips...so not testing at the same time. I think the 3g theory is the most likely if it's not actually 4g but like I said there are problems with that theory. It's also a possibility some sites aren't publicly broadcasting. Regardless most of that will be live in a few months so I wouldn't worry about the map at this point. 800lte will make the maps very inaccurate for those with 1900 only phones, and with the iphone out that's probably not too far away.

Believe what you want.

The phone's size has nothing to do with the size of the sensor or the lens. The 5s has a 1/3" sensor, as do the G2 and the S4. The Nexus 5 has a 1/3.2" sensor. The Z1 and S4 Zoom have bigger sensors at 1/2.3", which is almost twice the area of the Nexus 5's sensor. The Lumia 1020 more than doubles that at 2/3". All of these would fit into a tiny flip-phone if the manufacturer wanted to do that, so phone size doesn't matter. But as Digiblur said they all suck if you want a photo you can actually print. My camera's sensor is 50 times the size of the iphone's, so it captures 50 times as much light. You'll definitely notice the difference between the 1020 and the Nexus 5 in less-than-ideal lighting, but a phone camera will never have a decent sensor because of the lens size and flange distance requirements of a larger sensor.

I haven't had time to catch up on the whole thread to see if this has been brought up, but I noticed something odd in the status bar when I'm picking up a signal stronger than -40 dBm: Is the #?*! a glitch or is it just a joke?

Or someone just drove by when they were testing towers that weren't up when you drove by. Based on the gaps, I think you can rule out it being Verizon since they don't currently and probably haven't had big gaps there for quite a while like the Sprint map shows. Verizon also appears to have a weak spot around I-95 and 462, where the map shows dark purple on Sprint. Sprint has a tower right there (probably testing or not publicly broadcasting 4G), and I assume Verizon does not. For the most part it follows the tower placement of Sprint. The gap around Walterboro means it wouldn't be 3G unless the mapper turned off sensorly for just that stretch since Sprint has strong 3G around there.

You're joking, right? Not that it's the designers' fault. You just can't do much with a microscopic sensor and lens.

You made me realize I had no idea what thread I was in after reading the last page. I had to check. So...who else has a note 3?

I think I am in the wrong iphone thread.

If some people are complaining about not seeing the coverage overlay (not me - I see it just fine), give them the option of decreasing the transparency of the overlay. With the new maps, you can't see the MAP, which is kind of important. Even without the coverage overlay, those new maps are hard to use and apparently incomplete/inaccurate.

I just checked OpenSignal out and in my area it's basically backwards. There's absolutely nothing mapped around all the 4G towers near me, yet it's showing strong 4G in farmland and highway areas where the towers aren't broadcasting LTE at all. Half of the mapped 4G is actually in Sprint dead zones for 3G and voice. EDIT: I should say 1900 MHz dead zones since I haven't been to those areas since 800 went live.

If that's the case, there are definitely ways around that. As it is right now, google grants exceptions for some non-profits or for maps deemed in the public's interest. At least in the US, sensorly was the best crowd-sourced coverage application, which serves the public quite a bit when you consider how accurate the carriers' own coverage maps are. If sensorly is not a non-profit, or cannot get an exception from google, they could definitely start taking donations or charging for premium features. As for the update, I'm not willing to try it because it's visually just not nearly as good and I'm unsure if I'd be able to roll back. I really don't like the new non-google map and how difficult it makes it to see the map. Also based on the comments about the UI (accessing menus) some of the "features" (auto-centering on your gps location), and its performance (slow), it sounds like it would be a lot more difficult to use while driving when you only have brief moments at stoplights/etc to safely interact with the app. I both like and dislike the addition of the speed tests: I think it's a good way to get more people to start mapping, so they can show off their network's e-peen, yet on the other hand it puts unnecessary strain on networks and uses up non-Sprint users' bandwidth. Since this is a crowd-sourced app, you really have to look at the desires of the masses, and at least from what I've read here and elsewhere the update takes a step backwards in some of its aspects while improving in others. I'd be interested in the developers chiming in here or elsewhere about the reasons behind the google maps change and what their thoughts are on the initial reactions we've shown about the various changes.

What does this mean for reselling? If someone has a legitimate freedompop mifi on their account and then sell it, can the new buyer activate it on his account? It would have been originally purchased from freedompop, but obviously freedompop wouldn't make any money from the 2nd sale. I think I asked this somewhere else but can't remember if anyone knew the answer.

I forgot to say it looks much nicer without the "no coverage" gray all over the maps now. It's much easier to see where there is LTE mapped now. It also made making the monthly comparison maps easier since I didn't have to remove it myself.

NV can't give sprint the best coverage. They don't have nearly enough towers compared to Verizon. If they ever announce NV 3.0 and plan to add 20,000+ towers, maybe then they'd have a chance.

One. Most of them follow the same pattern when converted to HEX as your 1x800 towers but with a few not fitting. There are plenty of gaps but I'm still missing a lot of towers and I'm wondering if they'll change the ones that don't fit the pattern as well.

Unfortunately this update seems a bit underwhelming compared to the last one. Partially because it's only been 4 weeks as opposed to 6, and partially because there have not been many areas exploding with new coverage. The month Ohio starts really kicking in should look nice. Also, I'm not sure what kind of images people are looking for so I made a couple new ones now that I have a few months of progression. I hope imageshack doesn't mind the bandwidth September exclusive coverage: September overlapping coverage: September animated comparison: July-September exclusive progression: July-September Sprint coverage progression: Uploaded with ImageShack.us I didn't bother posting a coverage overlap progression or an AT&T progression because I already have too many images here. If anyone wants those I'll make them as well.

But will there be a pattern? They can't just randomly assign them or they'll interfere, but I dunno if it's a complex assignment method that will seem arbitrary or if I collect enough of them I'll be able to extrapolate a pattern/algorithm. Is there any Samsung market that is well-documented? I just noticed there's some info on Columbus in the sponsor section, but there the PNs always appear to differ by 3. :/

The sensorly maps display purple wherever anyone ever connected to 4G while mapping. Since it is light purple for a small stretch of McCarter Hwy, that means 1 person got barely usable LTE at least 1 time on that road. If 99 people get no LTE and 1 person gets LTE, it maps the LTE on the map. If you're in an area with 3 or 4 bars on the map and getting nothing, then there may be something wrong. Anything showing 0 or 1 bar should be ignored as fringe or unusable. I've found 2 bars sometimes means consistent, decent LTE, but also shows up in areas where I can't always connect. It could also be your phone. If your phone isn't actively searching for LTE for the couple mile stretch, it's obviously not going to connect. If you force it to search you may have better luck.

I have a question about the shades of purple. I have some theories but am looking for a definitive answer. What is it that causes carriers like AT&T to be so much darker than Sprint? I initially thought most of it was the density of live sites, but zooming in you can see that even in markets where Sprint is denser than AT&T the AT&T map is darker purple. I also noticed the AT&T mapping of a single tower quickly goes from 3 bars down to nothing near the edge of service. On the Sprint map, the signal has very little 4 bars, then gradually fades from 3 to 2 to 1, etc. This leads me to a few theories: First, it's possible AT&T phones "give up" on LTE much sooner than Sprint and switch to 3G before LTE is unusable. That would eliminate the lighter purple on the AT&T map and overall everything would average much darker. It also may be a property of the difference in band. With a lower frequency like 700 MHz does the signal just fade faster at the edge, but stay stronger up until that point? Difference in band also means Sprint users are mapping much weaker indoor signals compared to AT&T, but I doubt that matters since the majority of mapping is outside. Alternately, it could also be affected by mapping activity. Since there are more Sprint mappers that are insanely active on sensorly, they are going to be mapping much more for each tower as opposed to just driving by it. Since most of a tower's coverage area will be lighter purple (geometry of concentric circles), the more you map two dimensionally instead of just driving by, the lighter the area averages out to. Finally, there's the possibility that it could use something like RSSI to determine shade. This would give AT&T an advantage because of their bandwidth compared to Sprint's. I doubt this is the case since Sensorly itself displays RSRP, but I have no way of really knowing. Most likely it's a combination of factors, but I'm curious as to what other people think since I'm not as knowledgeable as some of you on wireless stuff. EDIT: Looks like bruce just asked this. I disagree with the only answer someone gave him though.

I don't have much knowledge about wireless technology so bear with me if I'm missing something. I understand the basics of PNs and LTE IDs. I'll be making another post asking for more information on all sector identifiers so look out for that. But first I was wondering what's going on in Michigan. I just recently started tracking PNs and SCIDs, although I had already been tracking BIDs. I can't track the HEX cell identities cause I don't have the right phone and I don't know of a database with them. From what I've read and what I've seen in other markets, PN offsets tend to be sequential and differ by either 3 or 6 between the sectors of a site. I also noticed LTE physical identities tend to differ by 169 between the sectors of a site. For the sites around me I've been able to confirm the numbers for I get the following values: PN offset (N, SE, SW) Site 1: 114, 270, 348 Site 2: 18, 333, 444 Site 3: 495, 93, 315 (sectors are actually N, N, SW) Site 4: 54, 267, 438 LTE physical identity/3 digit SCID (N, SE, SW) Site 1: 21, 403, 26 Site 2: 406, 15, 11 Site 3: 203, 123, 220 (sectors are actually N, N, SW) Site 4: 233, 223, 138 Is this normal? I'm not seeing much of a pattern here, and these things have to be planned ahead since they are non-unique but can't overlap. It's making it a serious pain to track them, since I have to be certain of what sector I'm connected to AND have to get all 3 for every site. I'd much rather there be a pattern so I could drive by, grab one of them, and figure out the rest based on the one. Thoughts?

Please re-read my post comparing it monthly. I'm not wrong at all - I factored in everything significant other than unexpected leaving, which has costs for BOTH plans. I posted all assumptions before the numbers. I mention the unmeasurable cost of selling your phone. There are countless other immeasurable things such as the opportunity cost of your money. I mention staying for 2 years and staying for 10 years. Staying for 50 years will result in the same scenario - One Up costs about $7/month more. Let me try it this way: ONE UP GENERALLY COSTS MORE PER MONTH THAN A NORMAL PLAN ASSUMING YOU CHANGE PLANS AT THE END OF A TWO YEAR CYCLE. If you're gonna say I'm wrong again, quote my text and highlight where I am wrong. I'm not posting every single possible scenario because that's stupid. I plan ahead and use up my contracts. I understand that year-old iPhones sell for $350+. And I know people comparing to One Up want to change phones every year, or they shouldn't be considering it. If you think it's a great plan, that's great! If the time and energy you spend posting a phone on ebay, throwing it in a box, and mailing it is worth well over $100 that's definitely something to consider (not being sarcastic; at times in my life it's definitely worth more than that). If you have a volatile living situation and can't avoid ETFs then you might avoid traditional plans, and probably shouldn't be with Sprint in that case. I was comparing from an economic perspective the dollar amounts for a frugal individual that still wants a yearly new phone.

You can't compare outstanding balance to ETFs. There are no ETFs when you leave after a contract. There is remaining balance on One Up regardless of when you leave if you are someone who upgrades every year. If you want to compare not upgrading the last year to not upgrading the last year (the only way to get no outstanding One Up balance), you can do that. Then with One Up, you don't pay that $325 at once, but you're paying the extra $12/mo to be on One Up -> $181 better off for not upgrading. The traditional way, this means you don't pay the $650 for your last year's phone, but don't make the $350 from selling the year old phone when you would have bought that -> $300 better off for not upgrading. So again, that makes One Up even worse than the previous comparison because you aren't taking advantage of the upgrade you are paying for. As for ETFs, you always have the option of staying on your contract to avoid them, which was what both of our calculations assumed (ending at the end of a 2-year cycle). Not to say there's no risk or cost involved since you can't always plan ahead (e.g. job moving overseas) and avoid them, but there is also risk with One Up since you have even more outstanding balance if you don't leave at the end of a year. So the One Up plan has the advantage of a 1 year commitment before minimizing exit fees (vs the traditional 2), but it still carries some risk for unplanned leaving.

Here's a quick third way: Take Koi's math but add in the outstanding balance he forgot to add for One Up at the end of two years gives you $2280 + $27.08 x 12 = $2605 for One Up for two years. That's $149 more than he had for the traditional way. $149/24 = $6.20 more per month. This is the closest the prices will be. Since you make money every time you resell and renew your traditional contract, that $6.20/mo ends up over $7/mo the longer the time frame you use. Expand his example to 10 years instead of 2: Traditional monthly payments + buy & activate every year you get a subsidy + full price phone every year you don't - selling your phone every time you get a new phone: 80*120 +236*5 + 650*5 - 350*9 = $10,880 One Up monthly payments + phone payments - bill credit + yearly activation plus outstanding balance at the end of 10 years: 80*120 +27.08*120 -15*120 + 36*10 + 27.08*12 = $11,734.56 Traditional method savings = $854.56. Divided over the 10 years: 854.56 / 120 = $7.12/mo. In his example he had you keep your last phone but sell or give back every other phone, so that is what I did for this math. If you want it like my other examples where you sold your last phone, that profit is the same both ways and cancels out.

No, my calculations are correct. Traditionally you buy a phone every year, and you sell a phone every year, so you always have 1 phone. 350/12 is what you make (per month) selling your phone at the end of every year (when you buy a new phone). $236/24 is what you spend (per month) buying your new phone + activating with the subsidy, which you only get every 2 years. Since you only get that subsidized price half the time, every other year you're on your own so you have to spend full retail price (but they don't charge you activation), which comes out to $650/24 when broken down per month. 1 phone every 24 months plus 1 phone every 24 months = 1 phone every 12 months. With the traditional way you don't ever have to give Sprint back your phone. If you didn't sell one every year, you'd start accumulating extra phones. Looking at it another way, with One Up you lose your phone subsidy ($450 every 24 months), but you gain the $15/mo bill credit and you spend $36 more every other year since activation is charged twice as often. You also lose the money you'd make selling your phone ($350 every year), but you save $27/month in payments when you turn in your old phone (Sprint waives the due balance so you're not paying off 2 phones at once). $450/24 - $15 +$36/24 + $350/12 - $27.08 = $7.34 per month you are losing compared to the old way. There's the math in 2 different ways, coming out to the exact same amount of savings by doing it the old way. Either way, One Up is about $7 more expensive per month.