The RF performance of a cell phone seems to have become a contentious issue, mostly because there are a number of people out there who didn't agree with my assessment of the Ericsson T39m. They claim they I don't test phones properly, and they lump Howard Chui in with me for the same reasons. I can't fully defend my position, in that I do not use test instruments and laboratory conditions to test this facet of the phone's functionality. Instead, I will outline below how I test the RF performance of a phone, and what sort of things I look for when I do.

The first thing we need to define is RF Performance, at least in terms of what I refer to when I say it. There are actually two independent variables that come to together to form this rating. The first is the phone's ability to pull in a weak signal and thus retain a connection to the network (both when idling, and when involved in a call). The second is the phone's ability to cope with those weak signals, which is its ability to translate weak signals into error-free audio reproduction. While it might seem as though these two things are somehow related, they often aren't.

I have tested phones that have virtually identical abilities to pull in weak signals, yet completely different response to weak signals. The i85 and i90 are good examples. The i85 is no better at pulling in a weak iDEN signal than the i90, but it has a simply marvelous ability to keep the audio sounding virtually flawless right to the bitter end. The i90 on the other hand has considerable difficulty keeping the audio free of disruptions under the same conditions. So, even though they both have the same ability to pull in weak signals, I rate the i85 as having superior RF Performance than the i90.

The next issue is how such comparisons can be made when I don't usually have both phones to compare simultaneously. There is no doubt that this is the biggest problem with making such determinations, and it certainly isn't very scientific to go completely on memory. Even if my memory was perfect, signal propagation can vary from day to day, thus changing the results that a phone might get in a known location.

I fully admit this limitation when I'm testing non-GSM phones, but when it comes to GSM models, I don't have that problem. In this case I can always compare the test phone to a reference phone on the same network, at virtually the same time. Being able to put both phones up against one another in this way allows me to see comparative differences between them.

To judge the signal capturing ability of a phone, I need a location that provides borderline signal strength that tapers off in a relatively linear fashion. Fortunately for me, god created malls for that. Shopping malls provide an ample supply of areas where signals fade to non-existence, and they do so quite predictably as one moves further and further "in" through various halls. While it isn't exactly a laboratory-quality environment, it does provide a way to not only reduce the signals gradually, but it also allows two phones to be compared at known signal conditions.

Regardless of the slight variability of this approach, it can generally be said that one phone has greater RF sensitivity if you can walk further down the hall before the signal fades away. By extension, one can also conclude that one phone is better than another if it can connect to the network and provide faultless audio, while the other phone cannot in the same location.

It is also under these conditions where the phones can be compared for their ability to deal with weak signals. Even if both phones hold on to the signal to the same point in the hall, one might degrade more rapidly than another. A good example of this is the Motorola P280. It has a phenomenal ability to crank out error-free audio almost to the point of the signal's non-existence. Virtually no other GSM phone seems to have this ability, though some do come close. Others start to produce audio problems very early on, and things just get worse until the signal fades away.

Another aspect of RF performance is the phone's ability to hold on to the network while it is idling. I test this ability in "free space" (with the phone held away from my body with the antenna pointing up), and in places on my body where the phone would commonly be carried. This type of testing sometimes reveals severe problems with RF sensitivity when the phone is put in close proximity to the body. Many of today's smaller phones beg to be carried in a shirt pocket, yet some phones suffer incredible signal losses when you do this. Perhaps the worst phones of all for this were the old Nokia 6190 and 5190.

One final aspect of RF performance concerns the phone's ability to cope with real life situations. It has been my experience that under normal usage some phones exhibit problems that don't come to light when testing them indoors. In the case of GSM phones, the way the model deals with handoffs can vary from very mild annoyance to all-out irritation. For CDMA phones, I find that some models seem to produce far more audio anomalies while on the move than others. These sorts of things also form part of my assessment of the phone's overall RF goodness.

But is RF performance really that important? If you use your phone outdoors most of the time, then RF sensitivity or weak signal behavior isn't going to matter much. However, real world behavioral problems most certainly will. However, no matter how good you think a network's coverage is, indoor locations will always be a challenge. The better the phone's ability to cope with weak signals, the more indoor places you'll be able to use your phone. This is especially true when it comes to the phone's ability to hold on to the network while idling.

So you be the judge. Do my methods of testing unfairly treat certain phones? I don't believe they do, but you are free to discount my methods of testing if it makes you feel better, because in the end, it's your money. If you want strong RF performance at the expense of all else, then don't pay attention to those who say I'm harsh on a phone for its poor RF performance while ignoring other positive aspects. What good are all those other positive aspects if the phone doesn't meet your expectations when it comes to RF.

Some Thoughts on the Importance of Good RF

Now having said all that about the testing of RF performance, the next question should come to mind is how important is RF performance anyway? This is a perfectly valid question, and one that needs to be addressed.

When we use our phones, we use them in one of 3 types of coverage areas. Type 1 is an area where the signals are sufficiently strong that even the worst phones work well. Type 2 is an area where the signals are non-existent, and no phone works, no matter how good it is. Finally we have Type 3, which is an area where the signals are iffy, and the performance of the phone will make all the difference in the world.

Some will say that since Type 3 areas are such a tiny percentage of all the areas we potentially use our phones, that we shouldn't concern ourselves with the RF performance at all. This logic is sound, but only if we look at this in terms of each area's size, and ignore how frequently we might be there. An analogy to this is driving through the countryside. Probably less than one tenth of one percent of the land is covered with tarmac, and yet as we drive through the countryside, we are on that tarmac almost 100% of the time. So although this is an extreme example, it demonstrates that other factors can have a huge impact.

So let's go back to Type 3 coverage areas, and let's assume that they represent less than 1% of the all the areas you might find yourself. The important question is, how often you are in such an area? If you spend the bulk of your time driving on the roads, then chances are that you are rarely in a Type 3 area, and in your case the RF performance of a phone really doesn't matter much. On the other hand, if you spend a lot of time at your office or in your home, and you rely on your cell phone at those locations, but you don't get strong coverage there, then you are in a Type 3 area most of the time. Between these two extremes, we find a wide variation that depends in part on where you use your phone the most, and what sort of coverage you get there.

I can't predict how often you'll spend in a Type 3 area, but I can predict that you'll likely feel cheated if you discover that someone else is using their phone while you can't, and yet you're both on the same network. It's really just a matter of peace-of-mind, since knowing that your phone can work as well as anything on the market makes you less concerned about missing important calls, or having to run to the nearest window to avoid dropping the call. No phone can perform super-human feats, but in Type 3 areas, the differences between phones is amplified to the point that some phones seem unless while others seem infinitely capable.

So the answer to our question is really "it depends". You'll have to examine your usage patterns and determine for yourself how often you spend in an area where the performance of your phone could mean the difference between owning a cell phone, and owning a paperweight. Once you've made that determination, only then can you move on to weigh other considerations when purchasing a phone.