[BBFan010101]

I can quite-conclusively report that it is different.

Still no comments on my crops and which is better - and why you believe it is so?

Alright, I've been lurking this thread for awhile and decided to make an account to chime in.

It's evident, and quite obvious, that the first picture produces a much higher quality product. I'm a big BB fan like everyone else, but most of us here need to stop being delusional and accept the fact that the Motion is producing unsatisfactory photos.

[Bay 13]

Do you have a link? In the video comparison he said he asked BB Mobile and they didn't give him a direct answer which led him to believe the sensors might be different.

Check out latest Crackberry podcast. He was on with Blaize and CBk. He read a message he got from a insider on the podcast

[anon(10268214)]

Including Mobile Nations own Mr. Mobile? Neither bla1ze or CBK have come out and disputed his results (the former suspects similar with his own tests).

If the Motion is using an IMX 378 sensor like the Keyone (which didn't have problems like this btw) and the issues users are experiencing was solely due to a manufacturing problem, I'm sure we'd be hearing more from Sony, BB Mobile, and other OEM's about this.

It's either a software problem that they'll need to improve in future updates or the sensor is different from what the Keyone is using. Not everyone that's reporting issues with the camera is bias or inexperienced with phone cameras.

"Mr Mobile" isn't particularly fond of the Motion. So it comes as no surprise he wouldn't be impressed with the camera either. And obviously neither CBK nor bla1ze are going to stick their necks out. They just keep repeating that they still prefer the KEYone...because they like the PKB. Hint hint.

As others have pointed out, Mr. M did reach out to 'someone' at BlackBerry Mobile and was informed it is not the same sensor as the KEYone. By confirming this he basically justified his own conclusion about the camera...something he clearly did before he even published his review. Is any of this true? Who knows. The only consistent thread in the argument is that he doesn't particularly care for the device, so let's not grasp at straws and pretend at least some aspects of the review may not have been partially biased.

[tickerguy]

Ok, so here's the deal on this post: https://forums.crackberry.com/showth...1#post13085217

The top clip is from the Motion picture posted earlier in the thread of the Bicycle in the window. The bottom clip is from the KeyONE.

As I noted both cameras focused on the basket. The sensors are different; they are both uncropped shots, but reporting different image sizes, so they're definitely different sensors. The artifacting bears this out as well.

The KeyONE fired at 1/33rd second with an ISO of 466. The Motion fired at 1/25th second at ISO 640. Both sensors are f/2.0, so that's common between them.

Of note is that the Motion's sensor has superior image noise despite operating at a 37% HIGHER ISO, and not by a little either. With both sensors being f/2.0 neither has an advantage in terms of admitting light to the sensor itself; they're equal in that regard. The Motion is shooting at a lower shutter speed and would be expected to produce more camera shake, but that's not evident in this case; the camera was held steady enough for that to not be a factor (both were in the area of 1/fl for the shutter speed, so it's reasonable to expect that with moderate care you can successfully hand-hold and not get notable camera shake.)

Now with this difference what I'd EXPECT due to higher ISO and SLOWER shutter would be that the Motion would have more blown highlights and fewer or no good blacks on a comparative basis, since the shots were taken in identical lighting and essentially at the same time. In other words the error should be toward overexposure on a comparative basis for the Motion .vs. KeyONE.

Let's see what we actually got in the 400% crops.

The IMX sensor in the KeyONE is notably deficient at ISO 466; I've seen plenty of ISO100 shots that looked just fine, but I'd consider this borderline unusable even for a 4x6 print, and definitely unusable for anything bigger. JPEG artifacting (looks like oversharpening, likely on-sensor and thus impossible to get around) on top of the sensor noise literally destroys the image quality beyond any reasonable recovery attempt as the water spots on the window wind up buried into the image to a degree that wrecks the ability to recover anything usable. The basket wave detail is basically non-existent in a number of areas but no part of the image escapes this damage; the frame tube detail is flat-out awful. Image dynamic range appears to be materially compressed with double-digit "blacks" on an RGB scale in one or more channels in a number of places; if this is how the IMX sensor is doing it's ISO enhancement (by boosting black levels) it's a big fat fail. The highlights on the chrome are blown out (255/255/255) as expected. What looks to be a "sharper" image is in fact quite-material over-sharpening with destruction of detail as a consequence of sensor noise and layered sharpening artifacts. Attempting to recover the image by smoothing sensor noise is going to do basically nothing for you here. You might be able to if you had a raw image but since there's no raw option on BlackBerry cameras you are screwed.

The Motion's sensor, on the other hand, did MUCH better. The detail in the basket weave was quite-well maintained. The smudges on the glass in front of the bike (water spots?) are clearly resolved rather than being buried by sensor noise and then amplified by JPEG compression and as a result the detail in the basket behind the waterspots on the glass is not destroyed. The frame tubes are cleanly resolved with only a bit of edge destruction on the lower left hand side due to noise and artifact interaction. Similarly, the edges of the cable look VERY good; the compound effects there are well-controlled as well. The basket, likewise, exhibits materially better edge definition as well. There are a couple of places where sensor noise and JPEG artifacting do combine in destructive ways but it's MUCH less-pronounced. Dynamic range is notably wider with this sensor despite being shot at a higher ISO and slower shutter; blacks go VERY close to 0/0/0 on an RGB scale; there are plenty of places with single digits on all three channels. Interestingly enough while the chrome has SOME blown highlights nowhere near all of it is; there is a fair bit of dynamic range in that flare. In other words despite EXPECTING to see plenty of evidence of overexposure on a comparative basis in the Motion shot it's not there. Finding only a FEW pixels where the RGB values are blown out (255/255/255) was a shock; I expected a hell of a lot more of that, and I ALSO expected from a comparative exposure basis essentially NO good, deep blacks (all three channels under 10 on a 0-255 scale) but I've got plenty on that cable. The DR advantage, while not huge, is material -- maybe a quarter to a half stop.

If these are my choices given this as data the KeyONE has, on a comparative basis, the inferior camera -- it's not a close call at all.

I did quite a bit of examination of the KeyONE's low-light performance when I did my review on the unit. What I see here is characteristic; nothing suggests in the data here that this unit is defective or other than what you'd expect, as it squares well with my reviewed samples.

If my comparison point was the camera ONLY and your choices are the KeyONE or Motion IMHO you'd be nuts to choose the KeyONE.

[Invictus0]

"Mr Mobile" isn't particularly fond of the Motion. So it comes as no surprise he wouldn't be impressed with the camera either. And obviously neither CBK nor bla1ze are going to stick their necks out. They just keep repeating that they still prefer the KEYone...because they like the PKB. Hint hint.

Why not? The Motion doing well would naturally mean good things for CrackBerry, I'm sure touchscreen users make up part of their userbase (remember, BlackBerry has been making touchscreen devices since 2008).

As others have pointed out, Mr. M did reach out to 'someone' at BlackBerry Mobile and was informed it is not the same sensor as the KEYone. By confirming this he basically justified his own conclusion about the camera...something he clearly did before he even published his review. Is any of this true? Who knows. The only consistent thread in the argument is that he doesn't particularly care for the device, so let's not grasp at straws and pretend at least some aspects of the review may not have been partially biased.

Those are two separate events, at ~3:30 in the video he specifically mentions reaching out to BlackBerry, the insider tip the other poster is mentioning happened on the podcast.



Even if none of them are fond of the Motion, what proof do you have that they're intentionally making the Motion's camera look bad compared to the Keyone? The problems Mr Mobile brought up in his comparison have also been highlighted by other users and reviewers.

[BBFan010101]

Ticker,

Great analysis and post. The hardware is capable of taking decent photos, so would it be accurate to say it's a software issue?

Still quite a disappointment that the Motion takes photos like the one posted earlier below. I'd be interested to see a comparison with the KeyOne and Motion in similar light settings using your technique.

MotionAttachment 431948

[Invictus0]

Ticker,

Great analysis and post. The hardware is capable of taking decent photos, so it would it be accurate to say it's a software issue?

Still quite a disappointment that the Motion takes photos like the one posted earlier below. I'd be interested to see a comparison with the KeyOne and Motion in similar light settings using your technique.

The sensors could be different but reviewers have posted some great shots in good lighting so it's likely just a software issue.

https://recombu.com/mobile/article/b...-camera-review

[Invictus0]

Check out latest Crackberry podcast. He was on with Blaize and CBk. He read a message he got from a insider on the podcast

Just listened to it, link for others that are interested (around the 52:30 minute mark),

https://crackberry.com/crackberry-podcast-motion-notion

The email he quotes says the sensor is the OV12870 which is what the Idol 5S uses according to Alcatel's spec sheet,

https://alcatelonetouch.us/idol-5s-u...id-smartphone/

[Bay 13]

Just listened to it, link for others that are interested (around the 52:30 minute mark),

https://crackberry.com/crackberry-podcast-motion-notion

The email he quotes says the sensor is the OV12870 which is what the Idol 5S uses according to Alcatel's spec sheet,

https://alcatelonetouch.us/idol-5s-u...id-smartphone/

Thanks. To save money it seems TCL will share some components across phone divisions. I like that Mr Mobile was impartial with his assessment of Motion 's hardware and how it falls short in some areas...

[tickerguy]

The sensors ARE different -- I'm sure of it.

Between the two the KeyONE has the inferior sensor in low light. Artificially over-sharpened images combined with a poorer dynamic range do not produce objectively better results. If the user holds the camera sufficiently-steady for its shutter speed and the camera is able to obtain a reasonable focus lock then the results in low light are superior to the KeyONE by quite a fair margin.

What I find most-interesting is that the low-light results under those conditions appear to show that you MIGHT be able to approach an OIS-enhanced camera's capability with the Motion with a relatively-minor recalibration of the camera software. Since it shoots at perfectly-acceptable noise levels with ISO 640, even under heavy magnification (I was VERY surprised at how little sensor noise there was at that ISO), this implies you MIGHT be able to shoot around ISO 1200 or even 2400 (!) and not have the sensor noise get too obtrusive. Moving the camera software ISO ramp-cap up a stop looks entirely reasonable, in short, which would produce a very-material improvement for non-moving subjects in low light. Moving it up TWO stops is likely a bridge too far, but I could tell you that in about 2 minutes if I had a Motion here -- sadly, I do not.

There is nothing you can do about low-light shots with MOVING subjects, OIS or no, in a cell camera. There the only answer is a higher shutter speed and in low light that's just not going to happen without a much, much larger sensor, aperture or both, and both wind up costing you in depth-of-field which in many situations is unacceptable too.

[conite]

Just listened to it, link for others that are interested (around the 52:30 minute mark),

https://crackberry.com/crackberry-podcast-motion-notion

The email he quotes says the sensor is the OV12870 which is what the Idol 5S uses according to Alcatel's spec sheet,

https://alcatelonetouch.us/idol-5s-u...id-smartphone/

Interesting that Xiaomi has the same two sensors on its Mi 5.

The Mi 5c uses the OmniVision OV12870, and the Mi 5s uses the Sony IMX378.

[Invictus0]

Thanks. To save money it seems TCL will share some components across phone divisions. I like that Mr Mobile was impartial with his assessment of Motion 's hardware and how it falls short in some areas...

Yeah it's a bummer that it's likely using something other than the "Pixel sensor" but who knows, with the right optimizations the Motion's sensor could standout in its own right.

The sensors ARE different -- I'm sure of it.

Between the two the KeyONE has the inferior sensor in low light. Artificially over-sharpened images combined with a poorer dynamic range do not produce objectively better results. If the user holds the camera sufficiently-steady for its shutter speed and the camera is able to obtain a reasonable focus lock then the results in low light are superior to the KeyONE by quite a fair margin.

What I find most-interesting is that the low-light results under those conditions appear to show that you MIGHT be able to approach an OIS-enhanced camera's capability with the Motion with a relatively-minor recalibration of the camera software. Since it shoots at perfectly-acceptable noise levels with ISO 640, even under heavy magnification (I was VERY surprised at how little sensor noise there was at that ISO), this implies you MIGHT be able to shoot around ISO 1200 or even 2400 (!) and not have the sensor noise get too obtrusive. Moving the camera software ISO ramp-cap up a stop looks entirely reasonable, in short, which would produce a very-material improvement for non-moving subjects in low light. Moving it up TWO stops is likely a bridge too far, but I could tell you that in about 2 minutes if I had a Motion here -- sadly, I do not.

There is nothing you can do about low-light shots with MOVING subjects, OIS or no, in a cell camera. There the only answer is a higher shutter speed and in low light that's just not going to happen without a much, much larger sensor, aperture or both, and both wind up costing you in depth-of-field which in many situations is unacceptable too.

It's all down to how well BB Mobile can optimize the "auto settings" IMO, as that's likely what most users will use. Some of the shots we've seen so far show that there's a lot of potential there with the sensor but it'll go to waste if it ends up being hidden behind menus. Between this and the Keyone, if they can improve it before the Motion's wider rollout they could do a complete 180 on public perception of BlackBerry cameras (going back to the BBOS days).

Interesting that Xiaomi has the same two sensors on its Mi 5.

The Mi 5c uses the OmniVision OV12870, and the Mi 5s uses the Sony IMX378.

I wonder how picture quality on those two devices compare against one another and also against what we're currently seeing on the Motion.

[donnation]

The sensors ARE different -- I'm sure of it.

Between the two the KeyONE has the inferior sensor in low light. Artificially over-sharpened images combined with a poorer dynamic range do not produce objectively better results. If the user holds the camera sufficiently-steady for its shutter speed and the camera is able to obtain a reasonable focus lock then the results in low light are superior to the KeyONE by quite a fair margin.

What I find most-interesting is that the low-light results under those conditions appear to show that you MIGHT be able to approach an OIS-enhanced camera's capability with the Motion with a relatively-minor recalibration of the camera software. Since it shoots at perfectly-acceptable noise levels with ISO 640, even under heavy magnification (I was VERY surprised at how little sensor noise there was at that ISO), this implies you MIGHT be able to shoot around ISO 1200 or even 2400 (!) and not have the sensor noise get too obtrusive. Moving the camera software ISO ramp-cap up a stop looks entirely reasonable, in short, which would produce a very-material improvement for non-moving subjects in low light. Moving it up TWO stops is likely a bridge too far, but I could tell you that in about 2 minutes if I had a Motion here -- sadly, I do not.

There is nothing you can do about low-light shots with MOVING subjects, OIS or no, in a cell camera. There the only answer is a higher shutter speed and in low light that's just not going to happen without a much, much larger sensor, aperture or both, and both wind up costing you in depth-of-field which in many situations is unacceptable too.

I have a sincere question. Why do you keep referring people to what you've written here any time they complain about the camera? I get that you are trying to state that the Motion has a better sensor than the KEYone. Great. What does any of that matter if the camera can't pass the eye test? If multiple people are stating that when they take a photo that the pictures look like garbage what does any of what you've written above matter? Is it supposed to make them feel better that you think that the Motion has a better sensor than the KEYone? I'm just curious because most people are going to judge how a camera works by taking a photo and looking at the picture. You're saying that a software update might fix the issue but I'm not sure how that's supposed to make people feel better when there isn't definitive proof that that's the case.

[Robert22]

I have a sincere question. Why do you keep referring people to what you've written here any time they complain about the camera? I get that you are trying to state that the Motion has a better sensor than the KEYone. Great. What does any of that matter if the camera can't pass the eye test? If multiple people are stating that when they take a photo that the pictures look like garbage what does any of what you've written above matter? Is it supposed to make them feel better that you think that the Motion has a better sensor than the KEYone? I'm just curious because most people are going to judge how a camera works by taking a photo and looking at the picture. You're saying that a software update might fix the issue but I'm not sure how that's supposed to make people feel better when there isn't definitive proof that that's the case.

Very well written. Theoretical technical analysis is for nothing, if the reality shows a very different result. And btw, the most people take pictures in automatic mode with point and shoot. There is no time to adjust manual settings or to hold for seconds a stable hand etc. Just normal amateur taking picturec. All other phones can handle it, I really dont know, why should BlackBerry ask of the customer something very special.

[tickerguy]

I have a sincere question. Why do you keep referring people to what you've written here any time they complain about the camera? I get that you are trying to state that the Motion has a better sensor than the KEYone. Great. What does any of that matter if the camera can't pass the eye test? If multiple people are stating that when they take a photo that the pictures look like garbage what does any of what you've written above matter? Is it supposed to make them feel better that you think that the Motion has a better sensor than the KEYone?

I didn't take the picture pair that I analyzed. A forum member did. I do not have a Motion to test (but I'll be happy to do so if someone wants to send me one; I cannot buy one at present in the US since BlackBerry Mobile has not seen fit to put them on sale here at this time.)

Said forum member didn't change anything from defaults; they pointed the camera at a bike display and fired, in low light, with both a KeyONE and Motion.

I note that they took TWO such image pairs and sent them both to me. One was not comparable at all and I explained why, although people tried to slam the Motion's result in that case. Well, yes, that shot was terrible. The KeyONE would have produced a terrible result if the shooter had waited the few seconds or minute and shot the KeyONE into the same condition (high movement-rate and high-contrast moving things in a low-light scene without an explicit focus point being specified.) The KeyONE's image in THAT CASE was superior because there was no moving CAR nor multiple moving people on which the camera could, and DID, lock, being unable to determine what the shooter intended. However, even though the KeyONE's image was superior in that case the sensor noise problem was still present -- and annoyingly so. It's just that Motion camera had locked focus on a MOVING object. This is quite clear.

The KeyONE, in that case, had both a STILL person (sitting on a pole) and a STILL vehicle in the frame (brake lights clearly illuminated.) It was able to select a focus point. HOWEVER, while it did so accurately the sensor noise is AWFUL (see attachment) and so is the resulting artifacting. This output is barely usable on social media.

The Motion clearly DID NOT lock on the same depth from the shooter; the pole and pedestal (same location as the KeyONE shot) are out-of-focus; either because that's not what the camera tried to lock on or it was shaken at the moment of exposure. The people and vehicles are MOVING. I'm not sure what it tried to focus on or if it was able to obtain a lock at all. My first thought was the sign on the building, but that's not it; a close examination shows that wasn't in focus either. It's entirely possible it never got a lock at all, as I can't find anything in the frame that is (1) in focus and (2) not moving (that is, not subject to motion blur.) That's shot is trash but IMHO either the camera was given an impossible job and it failed to guess "correctly" or it was shaken when the picture was being exposed.

In the OTHER case no such confounding issues were present and both cameras were able to select a focus point, and did, with approximately equal accuracy (equal as far as I can determine by inspecting the results.)

The outcome is what it is; the results most-certainly did NOT look like "garbage."

The issue people are claiming exists appears to be the nut behind the button; UNMOLESTED, with the software as it is NOW, the Motion produces BETTER low-light images than the KeyONE and that's not with me trying to optimize the experience by shooting in manual or anything of the sort.

I'm just curious because most people are going to judge how a camera works by taking a photo and looking at the picture. You're saying that a software update might fix the issue but I'm not sure how that's supposed to make people feel better when there isn't definitive proof that that's the case.

There is no issue other than, from what I can determine, may well be intentionally or unintentionally-biased people who are slamming something they either didn't actually use, used incorrectly (is it the camera's fault if you shake it in low light when you press the shutter?) or gave it an impossible task for the conditions in question (e.g. shooting at a scene with multiple moving objects in low light, without selecting something NOT moving to focus on, and expecting the camera to perform well.) In fact SOME people have said that's EXACTLY what they expect -- the impossible -- in that they EXPECT the camera to select a focus point in low light WITH MOVING OBJECTS IN THE FRAME. That's not going to happen reliably and thus you're going to get a very high percentage (maybe 100%!) of trash. The problem in that case is not the camera, it's the user. Expecting the camera to intelligently guess in such a case is beyond silly and into the realm of crazy, especially when a single touch on the screen under those conditions will TELL IT what to focus on.

That the Motion produced that bicycle shot without using manual mode which shows very-clearly MUCH better sensor noise and resolving power under low light conditions, along with superior dynamic range, is what it is.

My best guess is "stop shaking the camera when shooting in low light"; that is, you're holding the phone in a way that is unsteady when trying to shoot in those conditions, which destroys any capacity to get a decent shot. This is in a few cases being compounded by trying to force the camera to guess what you intend it to focus on in difficult conditions -- low light with moving objects in the frame, which ANY phone camera is going to blur because the shutter speed is inadequate to stop their motion. Software changes won't fix that, but I'd like to see them anyway as it appears that low-light performance can be improved over what is, right now, which is clearly very good and already materially superior to the KeyONE.

[donnation]

I didn't take the picture pair that I analyzed. A forum member did. I do not have a Motion to test (but I'll be happy to do so if someone wants to send me one; I cannot buy one at present in the US since BlackBerry Mobile has not seen fit to put them on sale here at this time.)

Said forum member didn't change anything from defaults; they pointed the camera at a bike display and fired, in low light, with both a KeyONE and Motion.

I note that they took TWO such image pairs and sent them both to me. One was not comparable at all and I explained why, although people tried to slam the Motion's result in that case. Well, yes, that shot was terrible. The KeyONE would have produced a terrible result if the shooter had waited the few seconds or minute and shot the KeyONE into the same condition (high movement-rate and high-contrast moving things in a low-light scene without an explicit focus point being specified.) The KeyONE's image in THAT CASE was superior because there was no moving CAR nor multiple moving people on which the camera could, and DID, lock, being unable to determine what the shooter intended. However, even though the KeyONE's image was superior in that case the sensor noise problem was still present -- and annoyingly so. It's just that Motion camera had locked focus on a MOVING object. This is quite clear.

In the OTHER case no such confounding issues were present and both cameras were able to select a focus point, and did, with approximately equal accuracy (equal as far as I can determine by inspecting the results.)

The outcome is what it is; the results most-certainly did NOT look like "garbage."

The issue people are claiming exists appears to be the nut behind the button; UNMOLESTED, with the software as it is NOW, the Motion produces BETTER low-light images than the KeyONE and that's not with me trying to optimize the experience by shooting in manual or anything of the sort.


There is no issue other than, from what I can determine, may well be intentionally or unintentionally-biased people who are slamming something they either didn't actually use, used incorrectly (is it the camera's fault if you shake it in low light when you press the shutter?) or gave it an impossible task for the conditions in question (e.g. shooting at a scene with multiple moving objects in low light, without selecting something NOT moving to focus on, and expecting the camera to perform well.) In fact SOME people have said that's EXACTLY what they expect -- the impossible -- in that they EXPECT the camera to select a focus point in low light WITH MOVING OBJECTS IN THE FRAME. That's not going to happen reliably and thus you're going to get a very high percentage (maybe 100%!) of trash. The problem in that case is not the camera, it's the user. Expecting the camera to intelligently guess in such a case is beyond silly and into the realm of crazy, especially when a single touch on the screen under those conditions will TELL IT what to focus on.

That the Motion produced that bicycle shot without using manual mode which shows very-clearly MUCH better sensor noise and resolving power under low light conditions, along with superior dynamic range, is what it is.

My best guess is "stop shaking the camera when shooting in low light"; that is, you're holding the phone in a way that is unsteady when trying to shoot in those conditions, which destroys any capacity to get a decent shot. This is in a few cases being compounded by trying to force the camera to guess what you intend it to focus on in difficult conditions -- low light with moving objects in the frame, which ANY phone camera is going to blur because the shutter speed is inadequate to stop their motion. Software changes won't fix that, but I'd like to see them anyway as it appears that low-light performance can be improved over what is, right now, which is clearly very good and already materially superior to the KeyONE.

What are you talking about? The people taking these photos aren't new to taking a photo with a phone in low light. You're making a statement like all of these people who are complaining about the camera have never taken a photo with a phone and don't have anything to compare the picture to. If they are taking the photo and think the image looks bad then they are comparing it to some other mobile phone camera they have used before. Getting all technical means nothing if someone takes a photo like they have done a million times in the past and thinks the result is weaker than something they've used before or weak for a phone in 2017. That's really all that matters.

[tickerguy]

What I'm talking about is facts.

You're free to take whatever you wish from whatever you wish. I'm telling you what was sent to me, why you're seeing what you're seeing, and on "automatic" in both cases.

While I can not shake a camera when I'm using it, I can't make someone ELSE not shake said camera, nor can I control someone ELSE'S unreasonable (outrageously so in the case of a low-light shot with moving objects in-frame!) expectations that are NOT comparable conditions, under which they then claim one camera sucks and the other (which they DID NOT subject to the same unreasonable conditions) does not suck.

If you believe it's unreasonable to be expected to hold the camera steady when shooting in low light then buy something that has OIS in it. That gives you ~2-3 stops of camera shake protection, but does zippo for moving subjects and in fact will blur moving subjects MORE in low light all things being equal, since the camera will be programmed to shoot at a lower shutter speed in low light (rather than boosting ISO.) It also makes the camera larger (means it sticks out more and is easier to damage the lens cover), it's more expensive (of course) and it's easier to damage via mechanical shock, since OIS requires a three-axis voicecoil on the lens instead of a single-axis (which is necessary for focus.)

[donnation]

What I'm talking about is facts.

You're free to take whatever you wish from whatever you wish. I'm telling you what was sent to me, why you're seeing what you're seeing, and on "automatic" in both cases.

While I can not shake a camera when I'm using it, I can't make someone ELSE not shake said camera, nor can I control someone ELSE'S unreasonable (outrageously so in the case of a low-light shot with moving objects in-frame!) expectations that are NOT comparable conditions, under which they then claim one camera sucks and the other (which they DID NOT subject to the same unreasonable conditions) does not suck.

What you're doing is trying to make a technical analysis of different photos when the analysis itself doesn't matter. If person A takes a photo and the photo looks bad (or comparatively worse compared to a phone he's used before) then he is going to think the camera is bad. If person B takes a photo and thinks the photo is bad for a camera phone in 2017 he is going to think the camera on the phone is bad. You coming in and saying "Well I've analyzed both photos and its clear to me that there is user error here and that the sensor is superior..." doesn't matter. What matters is that people are taking photos with the Motion camera and they think the photos are sub par. You coming in and telling them that they aren't sub par because you've analyzed the photos isn't really accomplishing anything.

[tickerguy]

What you're doing is trying to make a technical analysis of different photos when the analysis itself doesn't matter. If person A takes a photo and the photo looks bad (or comparatively worse compared to a phone he's used before) then he is going to think the camera is bad. If person B takes a photo and thinks the photo is bad for a camera phone in 2017 he is going to think the camera on the phone is bad. You coming in and saying "Well I've analyzed both photos and its clear to me that there is user error here and that the sensor is superior..." doesn't matter. What matters is that people are taking photos with the Motion camera and they think the photos are sub par. You coming in and telling them that they aren't sub par because you've analyzed the photos isn't really accomplishing anything.

I'm telling you that if you think you should be able to mash a shutter button and shake the camera in low light, whether it's in 2017, 2052 or 1993 then you need a camera that has OIS in it and you need to deal with all that comes with same (more cost, more mechanical complexity, more mechanical fragility AND more motion blur on moving subjects in low light.)

If that's the case then stop reading and don't buy any device that doesn't have OIS in the camera. Period.

The facts are that BlackBerry Mobile put a superior sensor into the Motion compared against the KeyONE. Both are ~12MP sensors, both are f/2.0, and thus comparing their quality is not all that hard. The KeyONE simply cannot handle shooting at the ISOs necessary for good low-light shots without destroying the image quality, while the Motion CAN.

This means that where the choices are between non-OIS sensors the Motion wins, and not by a little bit either. Since I've compared low-light performance between the DTEK60 and the KeyONE (the KeyONE was materially superior) in terms of lower-light photography the order is clearly Motion, KeyONE, DTEK60. In GOOD light the additional resolution of the DTEK60 comes into play (it's a ~21MP camera, so it has roughly double the pixel count) but like all things there's a trade-off, and the trade-off is that the smaller pixels mean each gathers less light, and it simply runs out of dynamic range on the low-light end before any of the 12MP sensors out there.

Choose the right tool for the job you intend to do; if you can't manage to hand-hold the device in a reasonably-steady fashion in low light then OIS is your only means of partial rescue, but with it you must accept that only non-moving subjects are helped and it comes with a cost both in money and mechanical robustness.

[conite]

A single tap on the screen in low light situations goes a long, long way to improving the quality of the shot. Hopefully this will mitigate many of the complaints. If some think this is too onerous, or that other cameras have better auto-settings, then so be it. It is what it is. Maybe future updates will help in this regard.
@ticketguy, despite your analysis, it is curious that Xiaomi uses the OmniVision sensor in the Mi 5c, but put the Sony in the slightly higher-end Mi 5s. Clearly they believe that the Sony is a modest upgrade.

[tickerguy]

Well, they may believe whatever they wish but the facts say otherwise, especially in low-light when it comes to sensor noise.

[donnation]

I'm telling you that if you think you should be able to mash a shutter button and shake the camera in low light, whether it's in 2017, 2052 or 1993 then you need a camera that has OIS in it and you need to deal with all that comes with same (more cost, more mechanical complexity, more mechanical fragility AND more motion blur on moving subjects in low light.)

If that's the case then stop reading and don't buy any device that doesn't have OIS in the camera. Period.

The facts are that BlackBerry Mobile put a superior sensor into the Motion compared against the KeyONE. Both are ~12MP sensors, both are f/2.0, and thus comparing their quality is not all that hard. The KeyONE simply cannot handle shooting at the ISOs necessary for good low-light shots without destroying the image quality, while the Motion CAN.

This means that where the choices are between non-OIS sensors the Motion wins, and not by a little bit either. Since I've compared low-light performance between the DTEK60 and the KeyONE (the KeyONE was materially superior) in terms of lower-light photography the order is clearly Motion, KeyONE, DTEK60. In GOOD light the additional resolution of the DTEK60 comes into play (it's a ~21MP camera, so it has roughly double the pixel count) but like all things there's a trade-off, and the trade-off is that the smaller pixels mean each gathers less light, and it simply runs out of dynamic range on the low-light end before any of the 12MP sensors out there.

Choose the right tool for the job you intend to do; if you can't manage to hand-hold the device in a reasonably-steady fashion in low light then OIS is your only means of partial rescue, but with it you must accept that only non-moving subjects are helped and it comes with a cost both in money and mechanical robustness.

Where are you getting this information from? Has TCL come out and said that the sensor in the Motion camera is superior to the KEYone? They went on and on when the KEYone launched about how the sensor used in the KEYone was the same sensor used in the Google pixel. I haven't heard BBMobile say anything about the sensor in this camera, not even that it has the same sensor as the KEYone in it so I'm not sure where you are getting your information from. I would think that if they put a superior sensor in the camera then they would make sure to highlight that with the phone, but they haven't.

[conite]

Where are you getting this information from? Has TCL come out and said that the sensor in the Motion camera is superior to the KEYone? They went on and on when the KEYone launched about how the sensor used in the KEYone was the same sensor used in the Google pixel. I haven't heard BBMobile say anything about the sensor in this camera, not even that it has the same sensor as the KEYone in it so I'm not sure where you are getting your information from. I would think that if they put a superior sensor in the camera then they would make sure to highlight that with the phone, but they haven't.

Like Xiaomi, I imagine that BlackBerry Mobile also believes the Sony sensor (in the KEYᵒⁿᵉ) is a modest upgrade.

[Robert22]

Christmas party of our company in a restaurant, not dark, but not full light, just typical restaurant, last friday. Everybody was taking pictures with a mobile phone. Just so, point and shoot. Today we looked at all the pictures from different people, unfortunatelly, mine, shot with the Motion, were definitely the worst of all. So please, stop defending the sensor of the Motion or explainig some technical details, the camera is simply very bad. There is nothing more to discuss.

[conite]

the camera is simply very bad. There is nothing more to discuss.

This goes too far the opposite way though. If a software update can improve things quite a bit, I would not describe the camera as "very bad". I think there WILL be more to discuss.

The Alcatel 5s, Xiaomi Mi 5c, and Freedom 2 also use the same sensor, and the reviews are similar - "capable mid-range shooter".