Reliability of unbranded Android phones vs 'originals'

[wildetudor]

Hi everyone,

After my Samsung S2 suffered water damage, I now need to buy a new smartphone, and I had almost decided for the S3, which now costs only GBP40 less than the S2. Until I stumbled upon this: http://www.amazon.co.uk/S7189-MTK6589-Android-Smartphone-UNLOCKED/dp/B00BG8F9G0

I have no problem with the idea of owning an "unbranded" phone, but I'm just worried that it might later turn out to be of substantially lower quality than the original. A colleague of mine owns a previous model of this Star phone (N7100), and he's been happy with it for the past year or so. After playing around with his phone a bit, it seemed fine to me, however I didn't get to test it in more detail (e.g. compare image quality of camera output with that from the S2, or see how stable the GPS fix is).

I'd have really liked to read a proper review for it (with benchmarks etc), or at least see a believable amount of user ratings. The Amazon product page has some good user reviews, but only 22 of them, and you can't really find much information about this model (or about the Star make in general) on the Internet.

What do you guys think? Is there reason to believe this phone would perform (visibly/audibly) worse than its closest Samsung equivalent, or that its life expectency would be shorter? Many thanks in advance for any help, I really appreciate it!

 

[El Presidente]

I'd steer clear TBH.

Whilst I've had no personal experience, from what I've seen around the forums, the "clone" devices generally seem to be more trouble than they're worth.

We've a pretty good summary here: http://androidforums.com/android-lounge/631594-off-brand-phones-tablets-worth-low-cost.html

Although reading reviews of that particular phone, they all seem relatively positive, they're all less than 6 months old so there's no real telling if it will stand the test of time.

 

[wildetudor]

Good point, the oldest review on there is only from May 2013, so hardly any evidence that they stand the test of time.

In terms of performance while the phone lasts (however long that might be), I found an Antutu benchmark score for this phone of about 13500, as compared to a 15500 score for the S3, and a ~7000 score for the S2. This looks decent, but the question is really what is the expected life of such a phone... Also, camera samples can't be found easily

 

[wildetudor]

Not sure I agree with some of the points in the article, for instance

- "being a knock off makes them look just like their counterpart from the brand name manufacturer but act completely different than the actual device it is copying" -- from what could I see on the Star N7100, it behaves and acts just like any other Android phone, such as my late S2 or any of the S3's I've had a play with

- "if you are paying a low price for a device the hardware won't be that good either." -- clearly, Samsung can afford to charge more for their phones even if we assumed they have in fact similar hardware to unbranded ones, just because they have a solid brand behind; that's just marketing, and I don't think that price reflects quality in such a simplistic, linear way

 

[Thom]

The answer to your original question is actually a question. It is what I always ask first ... when are you going to replace the equipment you are about to buy?

Your course of action is then clear ... 6 months = do whatever you want assuming they can actually replace it if the hardware dies ... 2 years = don't even consider it. (I have a Droid Bionic that is just under two years old. The system has been upgraded through two major versions during that period (GB to ICS to JB). Don't expect that in the device you are considering. There are applications that run under ICS and GB that are not supported under GB.)

... Thom

 

[El Presidente]

Not sure I agree with some of the points in the article, for instance

- "being a knock off makes them look just like their counterpart from the brand name manufacturer but act completely different than the actual device it is copying" -- from what could I see on the Star N7100, it behaves and acts just like any other Android phone, such as my late S2 or any of the S3's I've had a play with

It might well do, now. Give it 6 months, a year and then see how it stands up against the S3's etc.

Experience around here shows that these devices never last as long or are as reliable as their equivalents.

- "if you are paying a low price for a device the hardware won't be that good either." -- clearly, Samsung can afford to charge more for their phones even if we assumed they have in fact similar hardware to unbranded ones, just because they have a solid brand behind; that's just marketing, and I don't think that price reflects quality in such a simplistic, linear way

I sort of agree with what you're saying, there's obviously a lot of markup on Samsung/HTC/real brand devices, but there'll also be markup on the clone devices too.

If they can afford to sell the device at

 

[Hadron]

Not sure I agree with some of the points in the article, for instance

- "being a knock off makes them look just like their counterpart from the brand name manufacturer but act completely different than the actual device it is copying" -- from what could I see on the Star N7100, it behaves and acts just like any other Android phone, such as my late S2 or any of the S3's I've had a play with

That depends on what you are looking for. Someone buying a fake Note 2 or S3 should not expect to have the split screen feature, smart stay or indeed Touchwiz (though you may consider that last to be a blessing ). Someone buying a fake HTC One should not plan on taking any Zoe's, or expect good low-light camera performance. And a device with a much less powerful gpu and 1/4 of the RAM may do fine for light use, but some users would certainly notice the difference. You'll often find that screen size matches the original, but resolution is lower. Etc.

It all depends on what your expectations are. I think the main point is that you should not expect to get an equivalent device, even though it looks the same.

- "if you are paying a low price for a device the hardware won't be that good either." -- clearly, Samsung can afford to charge more for their phones even if we assumed they have in fact similar hardware to unbranded ones, just because they have a solid brand behind; that's just marketing, and I don't think that price reflects quality in such a simplistic, linear way

Yes, a brand can charge a premium for the name, but that's only part of the equation. In some respects the hardware isn't similar, and often relies on people not spotting what the specs hide. So as I said above, size is only one property of a screen (resolution, contrast, viewing angles, brightness and efficiency come to mind). "quad core cpu" means little unless you know the manufacturing process and architecture - you can stick in 4 low-powered cores made with an old fab process (higher power consumption) and people who know no better will say "quad core, just like the original". So hardware is typically not broadly similar to the flagships (and it's usually flagships you clone).

And then there's the question of quality control. I'll leave comment on that to those who have experience, though it's clearly the most important factor of all.

 

[wildetudor]

Thank you so much for your very helpful replies, I feel a lot more clear on this now, and I practically have made my decision that it is indeed not worth to go for one of these phones. Just a couple of comments.

Hadron, I see your point about hidden downgrades in hardware, but just to play devil's advocate, both the Star S7189 and the Samsung S3 have 1GB of RAM and quad-core CPU (with only a 0.2 Ghz difference in CPU clock); the GPUs are harder to compare, at least on paper. Still, overall, the comparable Antutu benchmark scores (S3: 15500 and Star: 13500) confirm that the hardware is not *that* much worse.

Nevertheless, my main arguments against the Star phone are:

1) lack of upgrades; it seemed to me like the "Android 4.1.2" spec they put on the box is presented as "hardwired" / written in stone, which seemed dodgy; if you say that these devices will not be able to upgrade their Android version, then this confirms what Thom says, that they should be bought for the short run, rather than being devices you still expect to use after 3 years. OTOH, I do expect that the S3 will be perfectly usable for most tasks in 3 years' time, if certainly outdated from a hardware perspective

2) reject rate. I suspect that if you're unlucky enough to receive a faulty handset, it's not going to be a walk in the park to get it replaced, if considering only the long waiting times involved by China/HK shipping

 

[El Presidente]

Re. point 2, you might find Amazon will replace it no problem, especially if it's faulty out of the box.

They've always been top drawer with me when I've needed to return stuff.

 

[Rxpert83]

Hadron, I see your point about hidden downgrades in hardware, but just to play devil's advocate, both the Star S7189 and the Samsung S3 have 1GB of RAM and quad-core CPU (with only a 0.2 Ghz difference in CPU clock); the GPUs are harder to compare, at least on paper. Still, overall, the comparable Antutu benchmark scores (S3: 15500 and Star: 13500) confirm that the hardware is not *that* much worse.

Nevertheless, my main arguments against the Star phone are:

1) lack of upgrades; it seemed to me like the "Android 4.1.2" spec they put on the box is presented as "hardwired" / written in stone, which seemed dodgy; if you say that these devices will not be able to upgrade their Android version, then this confirms what Thom says, that they should be bought for the short run, rather than being devices you still expect to use after 3 years. OTOH, I do expect that the S3 will be perfectly usable for most tasks in 3 years' time, if certainly outdated from a hardware perspective

2) reject rate. I suspect that if you're unlucky enough to receive a faulty handset, it's not going to be a walk in the park to get it replaced, if considering only the long waiting times involved by China/HK shipping

The hidden spec downgrade still applies, it may still be quad core and have a max clock speed similar to the original, but not all quad core CPUs are the same performance wise. There's a lot more than goes into it than the number of cores and clock speed. You would need the actual name of the processor to do a proper comparison, and you'll probably notice that's absent (the name can even be faked on the device settings, you'd have to open it up to really know).

I suspect overall quality takes a huge hit. It may have the same size screen, battery, MP camera, but those specs don't measure the quality of the hardware involved. They feed off of consumers seeing the same numbers and not fully understanding what they mean or how significant they are.

Also keep in mind benchmarks are not absolute. They can be faked fairly easily. Many knockoffs are rooted out of the box so there could be all sorts of benchmark fooling things added in.

Point:
1) I would not expect any android updates, and it will not have a strong developer community so you likely won't be able to grab an updated rom yourself.

2)

 

[wildetudor]

Re. point 2, you might find Amazon will replace it no problem, especially if it's faulty out of the box.

They've always been top drawer with me when I've needed to return stuff.

You're right, they're normally very good about replacing stuff without quipping, but I believe this only goes for items sold/dispatched by Amazon, rather than - as in this case - by a third-party seller.

 

[wildetudor]

I would not expect any android updates, and it will not have a strong developer community so you likely won't be able to grab an updated rom yourself.

My first thought about the lack of community support and ROMs was "well, if it comes already rooted, why would you still need to replace the ROM"? But I guess community support is useful for a ton of other things as well, and I don't think I'd be prepared to go "on my own" with a phone that is not only officialy unsupported by a community such as this one, but even (understandably) looked down upon

They feed off of consumers seeing the same numbers and not fully understanding what they mean or how significant they are.

Completely agree.

Also keep in mind benchmarks are not absolute. They can be faked fairly easily. Many knockoffs are rooted out of the box so there could be all sorts of benchmark fooling things added in.

I'm sure they can be faked, but in this case I don't think the YT user who did this comparison would have the motive to fake them, so the scores are probably genuine.

 

[chanchan05]

Thank you so much for your very helpful replies, I feel a lot more clear on this now, and I practically have made my decision that it is indeed not worth to go for one of these phones. Just a couple of comments.

Hadron, I see your point about hidden downgrades in hardware, but just to play devil's advocate, both the Star S7189 and the Samsung S3 have 1GB of RAM and quad-core CPU (with only a 0.2 Ghz difference in CPU clock); the GPUs are harder to compare, at least on paper. Still, overall, the comparable Antutu benchmark scores (S3: 15500 and Star: 13500) confirm that the hardware is not *that* much worse.,

Irrelevant. Antutu already holds a disclaimer that chinese fake phones are hardwired to show fake scores on Antutu. A better way to actually determine performance is using a high power game and playing it on the device.

Clockspeed again is also an irrelevant spec as long as you don't know the architecture of the CPU. A dual core A15 is actually comparable in computing power to a quad core A9.

 

[wildetudor]

I see. Am I correct that your prediction would be that, if you were to run a high-power game on the Star N7189 and on the Samsung S3, the Samsung would smoke the Star in terms of performance?

 

[mikedt]

The Guide in China says, don't buy a Star or whatever fake, you'll probably regret it.

 

[chanchan05]

I see. Am I correct that your prediction would be that, if you were to run a high-power game on the Star N7189 and on the Samsung S3, the Samsung would smoke the Star in terms of performance?

Yes. Even Xaomi and Meizu phones from China, which outsource CPUs from Qualcomm and Samsung, while much cheaper than the main brands, are still much more expensive than the Star phones. This goes to show that to get comparable quality, you'd have to spend a bit.

 

[mikedt]

Yes. Even Xaomi and Meizu phones from China, which outsource CPUs from Qualcomm and Samsung, while much cheaper than the main brands, are still much more expensive than the Star phones. This goes to show that to get comparable quality, you'd have to spend a bit.

Thing is Xiaomi and Meizu are both legitimate reputable brands that manufacture their own products, and Meizu does have their own B&M stores as well. Although Meizu does their own distinct fork of Android called Flyme.

Whereas with the Star things, well one thing "Star" is not even a brand or trademark, it's become sort of generic name that various trading companies use for their knock-offs. All made by unknown manufacturers. Most of them seem to be based on MediaTek SoCs, and are probably based around MediaTek reference designs. But it means though one "Star S4" could be quite different to another 'Star S4", in terms of quality, features, etc, etc. Pot luck really. Friend of mine in the UK bought a "Star" something or other, and the front glass broke within two weeks, he didn't do anything to it, it was just extremely thin and rather fragile.

 

[Hadron]

I see. Am I correct that your prediction would be that, if you were to run a high-power game on the Star N7189 and on the Samsung S3, the Samsung would smoke the Star in terms of performance?

I'm not sufficiently au fait with gpus to make a firm prediction. The chipset that one uses has 4 A7 cores (vs 4 A9s in the S3) and the gpu is clocked at about 2/3 the speed. So less powerful, but I'd hesitate to make a quantitative prediction.

It's worth remembering that last year's dual core snapdragon SoCs would at least match and often outperform the Exynos quad core of the S3. Core counting really isn't a good way of predicting performance.