I’ve been in the Android world less time than a lot of you, but I’ve probably spent more time messing with different apps, ROMs, settings and tweaks than most just because it’s what I’m interested in for reasons I can’t quite explain. I’ve followed a lot of Android, custom ROM, network, and app and hardware development over the past four years and here are some things I don’t think matter that many users are obsessed with.

Benchmarks don’t matter (much)

AnTuTu Benchmark showing this ROM runs 6% slower than my last ROMBenchmarks can be a great indicator of how well overall a ROM performs, although a device can have benchmarks in the crapper but outperform something that has a significantly higher score. It all depends on what you do with the device.

An example: my HTC One M8 is about ten times faster than my HTC EVO 4G LTE, and scores roughly six to eight times higher on most benchmarks. For some the things I use phones for (flashing ROMs, making backups, copying data to and from work), the HTC EVO 4G LTE has a significantly lower score but is the faster device.

How? ROMs are smaller usually by 40%, and SD card read and write speed are the same, meaning the One M8 is essentially slower. But that’s just for me.

A ROM that scores 33,000 and a ROM that scores 35,000 on an AnTuTu get claims that “it’s 2000 faster,” which is true… that’s a difference of 6% on a generalized non-battery drain including test. For all we know that ROM could run twice as fast as the 35,000 but be incredibly underperforming in an area that doesn’t matter at all to you. The 33 ROM could last half a month on the battery.

Why do we mention benchmarks? Because although they’re somewhat useless, they do give you an idea of the speed of what general ROM operations are going to perform at and generally the reports back from them are not that a Yugo and a Ferrari are the same because they both can go 90 miles an hour.

Generally.

Updates don’t matter (that much)

Android 4.4.infinityThe Android community is one of the few places where I’ve ever seen the claim that because something was updated frequently that it was well taken care of and developed.

Hold off the attacks here for a second – let me give you an example:

Imagine you got a car, you took your car to the mechanic at the place that sold you the car and you got it back but something wasn’t right and you took it into the shop and something wasn’t right so you took it back into the shop, etc. Every time you get it back something else is wrong with it and you take it back and the mechanic fixes it for free and something else is wrong with it or what you took it in for wasn’t fixed.

You wouldn’t think that mechanic was the greatest, but several ROM developers who spend time patching their mistakes, and several application developers who I’ve caught pushing snake oil, use minor updates and user testimonial to continue making people believe that the ROM or application they’re developing is the best thing ever and they’re devoted to spending all their free time doing this for no recompense so please donate.

Now in this instance, usually the mechanic is doing this for free on the car you got for free, so the example fails. They were under no obligation to give you that car, nor maintain it. The example here is that an update doesn’t mean it was or is being done right, which is what lot of people seem to judge a ROMs greatness by – activity. High activity without an absurd amount of new features is generally useless if the product is solid.

I watched a performance boosting scam application get multiple updates in the market during a month, each with testimonials that people’s devices performed better after the new version, and I’m sure they believed it, but looking at what it did there were no changes other than a version number and some text on how to donate to unlock the pro version.

The latest and greatest update clamor is people running to get Android 4.4.3 then claiming it’s so much better than 4.4.2. It’s not, it’s a version number and some improvements to security.

Now there are good developers who are adding to their ROMs constantly, this isn’t meant to belittle them, it’s just the code kiddies who throw a ton of patches into their FrankenROM, slap up a “please donate to our cause” PayPal link, and release pretty badly mangled ROMs while complaining they’re so overworked or overtaxed and doing this for nothing.

User testimonials on custom ROMs don’t matter (usually)

Titanium Backup's 400 or so appsWhen most root users switch a ROM, they’re coming from a ROM that had all their applications running in the background – their Facebook, Google+, Twitter, Flickr, Instagram, Zynga apps, GPS fixer, Dropbox, modified media scanner, FourSquare, Yelp, and all those apps that sit in memory occasionally popping up to use the processor or internet connection.

They’ve flashed a new ROM and 24 hours into it claim it’s the best, the fastest, the greatest thing ever. They haven’t installed all their bloat yet. A month or two into new ROM use, there are complaints that it’s not as fast as it used to be. A report on how great a ROM is that’s been out for a day or two is not particularly useful.

With how I travel between ROMs, I carry all my bloat and reinstall it before I start trying to feel out the ROM … because that’s how I’m going to be running it in a few days anyway.

If it feels zippier, it’s zippier with everything I installed and ran with on the last ROM and the one before that.

One speed test doesn’t matter

It’s 3:22 a.m. on a Sunday and you see a speed test of someone getting 40+Mbps on carrier X’s LTE. They claim carrier X gets 40+Mbps sustained in their tests. Well yeah, their test was when the network was not in use. My neighborhood gets six Mbps LTE now weather permitting, but 4-6:30pm I’ll be lucky to get dial-up speeds as everyone and their dog is driving by during rush hour with their smartphones with Facebook, Google+, Twitter, Flickr, Instagram, Zynga apps, Dropbox, FourSquare, Yelp, etc. all on the network syncing occasionally.

That’s where apps and maps that show speed tests fail. Then again, attempting to display a 24 hour speed test map would be kind of absurd.

Your ROM developer’s opinion

Much like I don’t base my political beliefs by the company president of the automobile I drive, I generally think if the developer of a ROM I am using is going on a tirade about someone else’s work there’s probably a story behind it that’s not being told, or you’re only seeing one side.

One of the great things in the Android world is the ability to add, lift up, and work together. Unfortunately, what the users see is pre-ban arguments, flame wars, and a lot of fairly hard for the average user to identify technical information involving GIT and commit times.

If you know your ROM developer is a great person, their opinion probably matters. If you’re just sticking behind them because they’re giving you something fun to play with on your phone … meh. Your loyalty is easily purchased.

Display megapixels don’t matter much

While some phone are 1080 and some go higher, there’s a limit to what you’re physically capable of seeing on a little display. If it looks good to you, it looks good. You’re not going to be able to tell the difference between 1080×1920 and anything higher on a 5 inch display.

I mean, I guess you can if your eyes are better than 20/20 or you view your phone at a couple of inch distance, but it’s pretty much a non-issue when one device claims superiority because it has more pixels you can’t distinguish between.

Capture megapixels don’t matter much

My EVO 3D met the floor with me on top of itOne of the things I realized on my now-deceased EVO 3D was that you can have a perfectly decent capture resolution and still come out with a crappy looking picture.

While lower megapixels mean it’s harder to crop and zoom, cameras with higher and low-quality light pixel detection can make a 6MP picture that looks grainy, the wrong colors, blurry, horrible.

It’s absurdly subjective, and while I’d rather have a higher resolution if possible, I’m finding that lower light correct color photos are more my style. Your style will vary.

RIP EVO 3D #1

Paying for as of yet impossible development doesn’t help

There once was a developer in the HTC world who promised a product that could not be delivered yet, and claimed great personal expense was required to make it happen that required recompense.

Basically it was “donate to me and my team, we’ll get you what you want,” and many did. Enough did that people would talk and discuss what they’d actually donated, which amounted to hundreds just from the people who managed to find themselves in another forum on another site wondering why nothing had happened.

The developers said no, this was not the case and that the number was more in the $40 range meaning several people who supposedly supported said developer were liars.

The then-so-far impossible development appeared from another unrelated developer for free, who was immediately accused of having stolen from the team that people had been donating to that was getting nowhere while saying “ask us for an ETA again and we’ll delay release a month”.

After weeks of screaming that was their development, they about-faced and claimed it didn’t matter, which it really didn’t on that device. They went on to pull the same request on other devices, I’m guessing it worked for them as they’re still going with the same routine.

My opinion

My opinion is my own, some share portions of it but you’re free to disagree and not be against me.

There are instances in each one of the above where there are exceptions – e.g. an AnTuTu of 400,000,000 is probably going to beat my 35,000 in any conceivable death match scenario, many users know what they’re doing and how to properly report on how well something performs, and there are developers who bash other developers on a regular basis who that’s just their thing and they’re generally 100% right.

Those are exceptions to what my experience has shown to be the norms. Your opinion and experience of course will vary.