My wife just got an iPad – the first tablet in our house. It wasn’t easy to get it away from her, but last night I managed to take our new iPad to bed. I was expecting an insanely great experience, but my first impression was that network access was slow and flaky.
To test my impressions, I ran it up against a Dell Precision laptop.
First, I pinged UCLA, a nearby university, 100 times each with the laptop and the iPad.
The average ping time for iPad was 63.7 milliseconds, over twice that of the Dell, and the standard deviation four times as great. No wonder it seemed flaky.
| Dell | iPad | |
|---|---|---|
| Minimum | 18 | 22.7 |
| Maximum | 120 | 231.9 |
| Mean | 30.2 | 63.7 |
| Standard deviation | 14.9 | 64.8 |
Next I tested file transfer times (Mb/s), and the laptop was faster.
| Dell | iPad | |
|---|---|---|
| Upload | 6.29 | 1.77 |
| Download | 1.83 | .17 |
That was no surprise given the variability in ping times. It doesn’t look like I’ll be making a lot of Skype calls or watching movies in bed.
My bedroom is at the back of the house, so I checked the signal strength. As shown below, it dropped to around -65 db as I walked from the office, where the WiFi access point is located, to the bedroom.
The laptop is not ideal in my home’s back bedroom, either. It does better when it is near the access point, but its radio is clearly more sensitive than that of the iPad. iPad Wifi is more like that of a netbook than a laptop.
So far, Apple reps have not returned calls for content. Have you seen this issue with your iPad?
Notes:
I measured iPad ping time using Typhuun System Scope Lite, the transfer rate using Speedtest.net and the signal strength with Metageek Inssider.
Wouldn’t a better PING test be to PING things on your local network instead of somewhere on the Internet?
PING’ing things on the Internet is not reliable because of the way web sites use a single domain name, but branch it out to different physical machines. It is very possible that each device PINGs a different physical server, and not to mention that the Internet is pretty volatile.
Could you do this again by PINGing local hosts? I’d be interested to see how it goes.
PS. I can’t say I’ve had Wi-Fi issues on my iPads when in good coverage in my house and I have done many Skype and FaceTime calls on my iPads, and watched many videos with no issue.
> Wouldn’t a better PING test be to PING things on your local network
You are right — I pinged UCLA without thinking. (Go Bruins). That being said, I did the Dell and iPad tests within a minute or two of each other, so I would not expect it to change the relative outcome.
> Could you do this again by PINGing local hosts?
Here are the results this morning pinging within my LAN:
Dell: minimum 1, maximum 90, mean 5.96, standard deviation 12.6 with zero dropped packets (1 sec timeout)
iPad: minimum 5.04, maximum 202.02, mean 18.56, standard deviation 32.49 with 8 dropped packets (1 sec timeout)
The relative story remains the same. That being said, the iPad streams movies and does Skype calls just fine when closer to the access point.
I agree. My one complaint about the iPad (both 1 and 2) is that the YouTube app is very slow! I always watch videos through Safari on the mobile version of YouTube.
My coverage is too good to test this, but not everyone is in Silicon Valley : ) gs
I could use an iPad, slow or not, if you really dislike it and plan to throw it away. :D
Good luck getting it away from my wife! Lar
Maybe I’m coloring too far outside the lines here but I’m getting a mental picture of the iPad as a device with weak wireless capabilities, coupled with the expensive data plans offered by mobile service providers.
> data plans offered by mobile service providers.
My tests were run on the WiFi, not the cellular network. Our iPad is WiFi only, so I can’t say anything about that. As you say, I felt the data plan was too expensive and not all that useful, and that has turned out to be the case — there seems to be WiFi every place we want to use it. WiFi is getting to be serious infrastructure.
I guess point I was trying to convey – albeit not too well – that it seems the weak wireless capability with the iPad might be a way of trying to manipulate us to signing up a data plan. Your thoughts?
I never thought of that — perhaps I’m naive.
Only folks inside Apple could tell us for sure if they are so cynical. I just checked, and they charge an extra $130 for the 3G radio — they have to be making a marginal profit on that. Looking over their accessories, there is no doubt that they make a fat margin on them. I guess they might even get a kickback from the carriers.
On the other hand, it is hard to imagine Steve Jobs letting them make something intentionally bad — he seemed to be such a perfectionist.
Also — I did not consider that the WiFi might be lame when I was making the purchase decision and I bet few people do. I’m skeptical of all appliances with built-in WiFi now.
I envision landfills miles high of tablet computers. At least they stack easily. Most of them will be archos.
Let me preface this by saying I like a lot of Apple products and have spent the $$$ to back up this claim. Having said that, I’ve experienced a lot of WiFi problems with Apple products. The iPad (1 & 2) and iPhone (3G & 4) don’t work in spots where other non-Apple devices work fine. The 2010 MacBook Air has been a particular sore point. I do not, for example, record podcasts over Skype using the Air because WiFi drops out too often. And, Netflix could not stream on it until a relatively recent firmware update.
Todd,
The only Apple product I’ve tried WiFi with was the iPad, but it sounds like Apple is cutting WiFi corners with other products as well.
This issue will become increasingly important as we distribute data throughout our homes using WiFi. I worried that my Roku box would not be sensitive enough to pick up video in the same bedroom as the iPad failed in.
It turned out that it works in that bedroom, but the only way I could find that out was to buy one and try it. I asked Roku if they could supply radio specs or sensitivity data, but got no where.
Consumers need that sort of data, and its lack might present an opportunity for AND. Can we figure out a way to do decentralized testing and comparison of WiFi radios embedded in devices like Roku boxes?
You can see the signal strength of individual devices with an Apple AirPort Extreme. Go to the Advanced tool bar, Logging & Statistics tab, click the Logs and Statistics button. In the Wireless Clients tab, you can see the signal/noise/rate and type for each connected device.
Between 2 identical iPad 1s, one iPad wifi only and the other 3G, and a airport extreme approximately 15 feet away, I found a consistant (over several weeks) 3dBm difference in signal strength. This was awhile ago, so I can’t speak for the iPad2.
James,
I wonder if the difference you noticed was due to manufacturing variability or a difference in design of the WiFi radio/antenna in a 3G iPad.
(My test was with a single iPad 2).