• On MovieTome: The 10 worst movies of 2009 so far!
October 22, 2008 9:22 AM PDT

3G speed test: iPhone 3G vs. T-Mobile G1

by Bonnie Cha
  • Font size
  • Print
  • 56 comments

The T-Mobile G1 is officially on sale now, but maybe you're still on the fence about it. What's Google Android all about? What's up with the design? Is it better than the iPhone 3G? Perhaps you're wondering which of these 3G smartphones is faster? Well, glad you asked.

In this quick Prizefight, CNET TV's Brian Tong and I pit the two against each other in a 3G speed test, clocking the time it takes for each device to load CNET News from start to finish. Now, there are a couple of things to remember. Both smartphones use different Web browsers, and there are a number of factors that might affect 3G speeds, such as the area you live in and how many people are on the network at one time. However, in the spirit of friendly competition and out of pure curiosity, we decided to go for it. Check out the video to see who comes out on top.

Also, be sure to check back next week when we'll have a full Prizefight between the iPhone 3G and T-Mobile G1--five furious rounds of battle, judging everything from navigation to multimedia to call quality.

Bonnie Cha is a senior editor for CNET, covering smartphones and GPS. When she's not testing the latest gadgets, you can find her chasing after her crazy lab or surfing in the chilly waters of Northern California. E-mail Bonnie.
Topics:

Phones,

PDAs

Tags:

T-Mobile G1,

iPhone,

Google Android,

T-Mobile,

AT&T,

3G,

smartphones,

cell phones

Share:
Digg
Del.icio.us
Reddit
Recent posts from Crave
Thanksgiving poll: What are you most grateful for?
Note to hospitals: The pen is mightier than the data entry worker
Crave giveaway of the day: LG 32-inch LCD TV
Gadgettes Podcast 86: The Body Episode (a FLASHBACK episode!)
Windows Starter Kit refreshed for 2010
Sit your 'tude in the Mood Chair
Big changes in Security Starter Kit 2010
Passive-aggressive gifts for chronically late geeks
Add a Comment (Log in or register) Showing 1 of 2 pages (56 Comments)
by jacou21 October 22, 2008 9:41 AM PDT
Wow - definitely the lamest speed test comparison ever. How can you come to any conclusion with that kind of test? lame...
Reply to this comment
by arpotu October 22, 2008 10:09 AM PDT
Hehe... do i detect a little jealousy from an iPhone user? True, it's not a complete comparison, but the ultimate goal is to get information the fastest, and the G1 won here, by being almost 33% faster :)
Reply to this comment
by random truth October 22, 2008 1:06 PM PDT
Not if you can count. 20% faster. Do you know what, I would be disappointed if the G1 did not win. I am sure that google has far more experience with the internet than apple.
by thurston24 October 23, 2008 11:22 AM PDT
Uhm. I can count. and its 36% faster. take 22 +33% on the windows calc. or, A 30 percent increase in a number would be 1.30 (or 1.3) times the original number so do 22 * 1.36= 29.92

http://www.notrain-nogain.org/train/Res/Num/numbers.asp

Random truth is not smarter than a 5th grader.

Yeah, the test is really bad. Its as bad as our education system.
by thurston24 October 23, 2008 11:30 AM PDT
This is better for all you to use.

http://www.csgnetwork.com/percentchangecalc.html
by iConquered October 23, 2008 10:29 PM PDT
It is approximately 24% faster. So you are both incorrect. Multiply 30 by 0.24 (24% in decimal form). You should come up with 7.2. Now add 7.2 to 22 (the number of seconds it took the G1 to load the page) and you get 29.2 (which can be rounded up to 30).
by dco888 October 24, 2008 1:19 AM PDT
Read Thurston's advice. He's correct. Remember also your frame of reference (faster than, slower than).

In this case, the two correct statements are:

The G1's 22 seconds is 27% FASTER than the iPhone's 30 seconds. (27% discount of 30 seconds is 22).
The iPhone is 36% SLOWER than the G1 (a 36% increase on 22 seconds is 30 seconds).
by quake21 October 22, 2008 10:30 AM PDT
haha I agree! while this isn't the most accurate method of testing, the end results are clear.
Reply to this comment
by kevinm6 October 22, 2008 10:30 AM PDT
His point was that the T-mobile 3G network is new and there isn't many users on it. So this will be different when T-mobile's 3G network gets hit by over 1 million G1s. I would use Wifi as an indicator of performance to eliminate that variable. Being that 3G speed depends on where you live and how many people are on the network, mileage will vary.
Reply to this comment
by October 23, 2008 9:53 PM PDT
It doesn't matter how many people are on the network. The ONLY thing that matters is the speed you get.

Let's compare it to something else...website hosting speed.

Service A is faster than Service B, but Service A has 50 times more customers, making the effective speed slower.

You don't say, "Well just wait until Service B has more customers!"

No...you say, "Service A is oversold..I'll go with Service B"
by scaught78 October 22, 2008 10:32 AM PDT
That test was so non-scientific, but seriously, it was lots of fun to watch!!

Brian and Bonnie rule!!
Reply to this comment
by menehune9 October 22, 2008 10:46 AM PDT
As other have stated, the T-Mobile 3G network is still sparsely populated. As time progresses the results may change (maybe not), we'll just have to wait and see. I agree with kevinm6, a better indicator of processing speed will be to test using WiFi.

Another idea, instead of testing for speed, they could've tested for web page compatibility. I mean there are some pages out there that work on only a small range of web browsers. Which browser will allow access to lets say X% of the corporate web portals out there or available to access those multimedia contents of some of the web pages out there now? Just brainstorming...
Reply to this comment
by radmoose October 22, 2008 10:58 AM PDT
Does the G1 support FLASH in the browser?
Reply to this comment
by Rawnchie14 October 22, 2008 11:01 AM PDT
I do not think so, I could be wrong though.
by jamielynnfox October 23, 2008 12:38 AM PDT
G1 definitely does not support Flash.
by glennw53 October 23, 2008 7:54 AM PDT
I think utube is flash driven and utube is preloaded on the G1 so yes it has flash and full web pages like ur PC.
by Seaspray0 October 24, 2008 10:33 AM PDT
If it doesn't, then someone can write the app that will make it do so and it can be installed by anyone. It's been reported that Adobe has already written the app for the iphone to support flash, but since you can only get apps through apple and they haven't approved it, then you don't get flash on the iphone. That's the difference between a closed developer network and an open one.
by Rawnchie14 October 22, 2008 11:01 AM PDT
Inconclusive test... maybe they should try using the wireless to eliminate the edges of different carriers.

Regardless, I can tell you right now, even with the Beta of Android, the browser that Google whipped up is damn fast (tried on Sprint's network, on an HTC touch - put all other browsers to shame.).
Reply to this comment
by zpappa October 22, 2008 11:01 AM PDT
You know why this test is truly garbage?

Because the iPhone is rendering AND displaying the entire page as is visible within a standard PC/Mac web browser, the G1 is rendering (we think) but NOT displaying the entire page, for those of you familiar with software development, this may make a lot of sense to you.

Also, two different networks, the comparison is like putting two vehicles on two different tracks and comparing their lap times.

What kind of kickbacks do you guys get for being tMobile fanboys? (I'm not knocking, I'd take money too :) ).
Reply to this comment
by Nick_DG October 23, 2008 4:53 PM PDT
You have it backwards. The G1 browser renders the entire page. The iPhone only renders what you see on the screen. That's why when you browse on the iPhone, you get the checkered pattern when you scroll, then it renders that spot on the page.
by totallytexan October 22, 2008 11:03 AM PDT
take the 33% (questionable) slower speed, and leave behind the heartache of depending on Verizon's customer service.
Reply to this comment
by chrisfrary October 24, 2008 10:53 AM PDT
Well yeah that makes sense....except neither of these phones are on the Verizon Network (unless roaming, G1) T-Mobile is the company you look for
by chrisfrary October 24, 2008 11:00 AM PDT
Nevermind my last comment apparently I wasn't thinking. T-Mobile and AT&T both use GSM neither can roam on verizon's CDMA network. But verizon is set to switch to LTE-GSM in a couple (many) years.
by bobpenn October 22, 2008 12:22 PM PDT
Uhhh. If there was a "hiccup" on the iPhone, shouldn't they have redone the test? Are things so tight that you can't do a take 2?
Reply to this comment
by nb2000nb October 23, 2008 10:48 AM PDT
Apple IS the hiccup
by ragedizzer October 22, 2008 12:28 PM PDT
I did this test last month in my lab. (phones are what I do) I found the G1 to be faster most of the time. But my network is Bellevue/ Seattle so I am curious what the results are. As I wrote in my report the G1 my not be an Iphone killer, but Android might be the Cell phone OS industry killer. From my tests its reach could become nearly limitless. Much like DOS and the PC in the 1980's
Reply to this comment
by jaffreywali October 22, 2008 12:29 PM PDT
I have the ATT Tilt. Did a test using IE Mobile on their 3G network in DC and I got around 25 secs to load cnet.com.

Just looking at the screen shots it looks like while the iphone loads the entire page as it would look like on a computer display the tmobile G1 needs you to scroll to read the content, like my IE mobile browser does.

Overall, while I don't have an iphone I think it's best features are great styling (a phone is about styling) and superb software in terms of use and graphics. I see the Android software as Windows Mobile Lite.

The T Mobile G1 is also ugly and bulky, in my opinion.
Reply to this comment
by stennex October 22, 2008 12:31 PM PDT
You guys should use Wi-fi next time to run the speed test. It would be more even since the 3G speeds differ a lot more for Att compared to T-mobile new 3G service which has far more less people right now.
Reply to this comment
by Motyoj October 23, 2008 8:32 AM PDT
That's exactly what I thought. Networks differ with loads but if there are only two phones sucking off a wi-fi, that should give a better idea in what the hardware can do. I have an iPhone and like it but think that being chained to AT&T is silly.
by minmax007 July 25, 2009 9:59 AM PDT
if you have wifi you will probably use a computer, not the phone. If you are outside, somewhere where you don't have wifi, you would care about 3g speed, trust me
by dizzy0113 October 22, 2008 1:02 PM PDT
Well this was quite a bad test, but I guess it wasn't supposed to be taken that seriously. One thing to point out was the iPhone "hiccup" was safari saying it couldn't load the page and it didn't start until 6 seconds in, so I would say the iPhone took 24 seconds to render the whole page vs 22 secs on the other phone with all the variables is pretty even.

And why not try a test on testmyiphone.com, my3gspeed.com, i.dslr.net/tinyspeedtest.html, or inetworktest.com (and any others I've left out?)
Reply to this comment
by jeitzen October 22, 2008 1:29 PM PDT
mobilespeedtest.com
by bigmc6000 October 22, 2008 1:49 PM PDT
I've always wondered about that - these speed tests are solid jokes on here and they never to intelligent thing of using one of the sites you mentioned. If you're testing the speed of the network and not the speed of the browser you should use one of those sites. It's amazing how ignorant CNet can be sometimes when they claim to be a technology site. Maybe it's because they are spending so much time reviewing cars... (rolls eyes...)
by minmax007 July 25, 2009 10:06 AM PDT
what matters most? what happens in background (the real network speed) or the result you get (network speed, rendering speed, etc)? I think the result is the most important when you really need to browse internet on your phone.
by dirtyqwerty October 22, 2008 1:25 PM PDT
Wifi wouldn't really help, it would take you from a near infinite number of variables to near infinite -1. What are you trying to test here? DL speed? Render speed? The JS engine? How well your proxy gateway caches pages locally?
Reply to this comment
by PizzleDizzle October 22, 2008 2:21 PM PDT
Hmmm. My two year old Sprint 6700 on EVDO loaded the page up quicker than the I-Phone in the test did. This included reformatting the html, as I have it set.

What do the arguments concerning the relative absence of T-Mobile 3G say about this? Does Sprint's 3G network have significantly less traffic than AT&T's?
Reply to this comment
by strad13 October 22, 2008 9:13 PM PDT
Um...yeah. How many customers have they lost in the last couple years?
by dirty55409 October 22, 2008 2:50 PM PDT
terrible speed test kids. besides, how do we know if the cnet page was cached on either phone. The iphone probably had it cached in memory and loaded it faster. way bogus. cnet loves the iphone so much it's kinda sick.
Reply to this comment
by Farjamed October 22, 2008 4:34 PM PDT
people keep saying they should have done a wifi comparison because that is more accurate, but they were testing the 3G speed of the phones. The iphone is an ATT phone and the G1 is a T-mobile phone, so the network IS part of the test. Obviously this wasnt a very scientific test. side note: I have tmobile, and i must say though, their 1700mhz 3g sucks because they are the only ones using 1700mhz and the only phones that can be used on their 3g network will be the ones made specifically for them. BOOO T-Mobile!
Reply to this comment
by Constable Odo October 22, 2008 8:50 PM PDT
Best two out of three or three out of five next time.
Reply to this comment
Showing 1 of 2 pages (56 Comments)

About Crave

The name says it all. Crave is our blog about gorgeous gadgets and other crushworthy stuff. If you would like to contact Crave with a tip or comment, please write to: crave@cnet.com

Add this feed to your online news reader

Crave topics

The browser battles go on and on

roundup From Firefox to IE and from Chrome to Opera and Safari, there's no sitting still for browser makers looking to keep their products fresh and competitive.

3G wireless still holds promise

The next generation of 4G wireless may get all the headlines, but advanced 3G technology will likely dominate services for the next few years.

News
News site map
Latest headlines
Contact News
News staff
Corrections
CNET blogs
Popular topics
Apple iPhone
Apple iPod
Cell phones
Dell
GPS
LCD TV
CNET sites
CNET Site map
CNET TV
Downloads
News
Reviews
Shopper.com
More information
Newsletters
CNET Mobile
Customer Help Center
RSS
CNET Widgets
About CNET