mobile photography technology, culture and community
www.dpreview.com

We put the HTC One's ultrapixels to the test

172

Shortly after the launch of its latest flagship smartphone, the One, HTC published a white paper explaining the thinking behind the 4MP 'ultrapixel' camera in its latest device. This covers some key factors affecting image quality, including the lens, sensor and processor, in a fairly non-technical way. The document is light on numbers though, and worryingly high in quotes from journalists so scientists and engineers may find it unsatisfying.

The core of HTC's argument is based around the idea that using fewer, larger pixels on a sensor that's the same size as the 13MP unit found in competitors such as the Sony Xperia Z should offer better image quality. This of course runs counter to conventional marketing, which has always contended precisely the opposite, i.e. that more pixels are better. We're a little dubious about these claims - there’s a pretty strong argument that what really determines image quality when looking at the image as a whole is the total amount of light captured by the sensor, not by each individual pixel.

However, the HTC One's advantages are not all about pixels - HTC is also talking about the (pretty unarguable) low-light advantages offered by the optically-stabilised F2.0 lens, along with the improvements to video footage that image stabilisation brings. The fast sensor read-out means its electronic shutter completes the exposure more quickly than previous sensors - 1/48sec vs 1/30sec - which should reduce the 'rolling shutter' effect in movie recording.

And there is an argument for using fewer pixels on a smartphone, since most images are shared at much lower resolution than 4MP, it's difficult to imagine the main advantage of high resolution sensors, i.e. higher image detail, being especially important to smartphone users. We have now finally had a chance to test HTC's ultrapixel-claims first hand. Andy Westlake at our London outpost got hold of a HTC One review sample and has shot a range of comparison scenes in different light situations with his the new HTC flagship and his iPhone 4s.

The current weather in the UK (only very brief moments of sunshine) meant we could not take as many shots as we would have liked to but the images below should give you a good idea of what to expect from the HTC One. We will follow up with a more detailed test under controlled lighting once we receive a test unit here at headquarters in Seattle.

Daylight, low ISO

The 4MP backside-illuminated CMOS sensor in the HTC One is made by ST Microelectronics and also being used in a number of video devices. This explains the, for a still camera, unusual aspect ratio of 16:9. Switching to 3:2 or 4:3 aspect ratios crops the image sideways. The latter for example reduces the overall pixel count to 3MP.

In the well-illuminated scene below the Apple sticks to its base ISO setting of 50 while the HTC reports a slightly unusual sensitivity of ISO 112. The HTC is doing a decent job in these conditions albeit with a much warmer color response than the iPhone, but the latter is clearly better. In good light more pixels simply means more detail and this is clearly visible in the samples below. Whether this persists in typical social media use is a totally different question though.

 HTC One, ISO 116, ISO 112
 100% crop
 Apple iPhone 4s, ISO 50
 100% crop

Here we've got another shot in good light but more overcast conditions. Both phones struggle rendering the distant low-contrast areas but again the Apple's 8MP sensor captures more detail overall. On the flip-side the HTC produces a cleaner image with visible less shadow noise.

 HTC One, ISO 102
 100% crop
 Apple iPhone 4s, ISO 50
 100% crop

Digital Zoom

Digital zoom is generally awful on smartphones apart from the Nokia 808 with its PureView technology and both the HTC One and iPhone 4s are no exception. That said, with its comparatively low 4MP pixel count the HTC's image quality is deteriorating faster than the digital zoom of higher resolution smartphone cameras as you can see in the samples below.

 HTC One - 4x digital zoom, ISO 116
 100% crop
 Apple iPhone 4s - zoom, ISO 50
 100% crop

Low light, high ISO

In low light the HTC One is clearly better than the iPhone, producing a much cleaner, more noise-free image and giving brighter, more colourful images. It's difficult to isolate the effect the effect of the ultrapixels in this context as both the IS and the fast F2 lens playe a role too but in a way that’s academic, it's the end result that counts.

These images have been shot in Auto ISO mode but the HTC allows you to select ISO manually. However, a lack of information about shutter speed renders this feature pretty much useless.

HTC One, ISO 395, 1/15 sec
 100% crop
Apple iPhone 4s, ISO 800, 1/15 sec
 100% crop

When taking into account that many editing and sharing apps for smartphones reduce the size of images it also makes sense to compare the low-light samples above at roughly the same size. For this purpose we have reduced the iPhone 4s image to the same width as the HTC image (1520 pixels). Due to the Apple's 4:3 aspect ratio you end up with an image just over 3MP but the size of objects in the image is approximately identical in both photos.

As you can see in the 100% crops below the low-light advantage of the HTC is somewhat reduced when looking at images of the same size but the HTC One's image is still visibly cleaner with less luminance and chroma noise. So, when shooting in low light the HTC offers a real advantage over higher resolution sensors, even when the image size is reduced for sharing, social media or editing.

 HTC One, ISO 395, 100% crop
 Apple iPhone 4s, ISO 800, reduced image size

Soft image areas

In quite a few, but not all of our HTC One sample shots we noticed extreme softness on the right side of the frame. We haven't been able to determine the exact reasons for this but we are suspecting the effect could be caused by the image stabilization system decentering one of the lens elements. Once we get a review unit here in Seattle we will check if this is only a problem with the particular copy of the phone that Andy had in London and invesitage further. The problem is very visible in the samples which were taken within a few images. Exposure data for the two images is identical but if you click through to the full version of the image on the left you can see that are large area on the right side of the frame is extremely soft, making the image pretty much unusable.

 This picture shows extreme softness on the right side of the frame.
 This one, taken only minutes later at the same settings, does not exhibit any signs of softness.

First image quality impressions

Overall, from what we can see from these first sample shots, the HTC One with its 4 "ultrapixels" CMOS sensor performs pretty much as we would have expected. In low light and at higher sensitivities the HTC delivers cleaner images than the higher-resolution iPhone 4s. However, this is as much due to the optical image stabilization and fast F2 lens as the sensor. The HTC's advantage decreases if you compare images at the same size but is still clearly visible.

It doesn't come as a surprise that in good light the HTC trails noticeably behind its higher resolution peers in terms of detail capture. In good light more pixels simply mean more detail. The question is by how many users this will be viewed as a disadvantage, given a large proportion of smartphone images is being edited and shared at lower resolutions anyway.

We should also mention the HTC's strong tendency to produce purple flare with light sources that are located just outside the frame, a very similar effect to the one found on the iPhone 5. This is very obvious in the first image of our sample gallery below.

We will analyze the purple flare effect and image quality of the phone in general in much more detail once we receive a reviewable unit at our studio in Seattle. In the meantime we have included a few additional sample images in the preview gallery below for you to examine.

Sample Gallery

There are 12 images including a couple of panoramas in our HTC One preview samples gallery. Please do not reproduce any of these images on a website or any newsletter / magazine without prior permission (see our copyright page). We make the originals available for private users to download to their own machines for personal examination or printing (in conjunction with this review), we do so in good faith, please don't abuse it.

Unless otherwise noted images taken with no particular settings at full resolution. 

HTC One Preview Samples
1 / 12
Image 1
Image 1
Image 2
Image 3
Image 4
Image 5
Image 6


Comments

Total comments: 172
12
hp79
By hp79 (11 months ago)

"In quite a few, but not all of our HTC One sample shots we noticed extreme softness on the right side of the frame."
Do you guys have any update? I have a AT&T HTC One, and I've been experiencing this exact same problem on few of my photos. It becomes totally crap shot when there's this much blur. Basically half the area of the picture becomes a blurry mess. I even tried the latest firmware that's in a custom ROM AndroidRevolutionHD 7.1.

0 upvotes
canon20s
By canon20s (Apr 25, 2013)

Looks like the soft image problem is a Coma issue. Look at the shape of the lights. Maybe as its focusing it get out of alignment, refocusing 'fixes' it?

0 upvotes
Michael Ma
By Michael Ma (Apr 24, 2013)

The 4MP looks much better in low light. But the overall image is still poor. It won't convince me why I should have this instead of double the megapixels on well lit situations.

0 upvotes
Abdullah M
By Abdullah M (Apr 3, 2013)

I think many image problems like noise, reduced when downsize the photo. So, higher pixel density I think is better.

0 upvotes
itsastickup
By itsastickup (Apr 3, 2013)

From the examples given the noise is definately less with the HTC even where the iphone has been resampled.

1 upvote
yabokkie
By yabokkie (Apr 4, 2013)

usually higher pixel density means better SNR but that's more a side effect of sensor design than the pixel pitch itself, which basically has no effect on image quality other than resolution.

this One got a 2/3 stop larger aperture lens than 4S and it has to perform 2/3 stop better for that sake.

btw, 2/3 stops is about the difference between APS-C and 4/3".

Comment edited 2 times, last edit 9 minutes after posting
0 upvotes
Ashuaria Lee
By Ashuaria Lee (Apr 3, 2013)

IMO 3MP(for conventional 4:3 or 3:2 image) is too small for nowdays.
The problem is, even if it is JUST for the facebook or twitter, 3MP is just too small.

0 upvotes
rramjit
By rramjit (Apr 2, 2013)

I am tired of people commenting that pixel count does not mean better image. This is an old and outdated logic.
We are in different times...where 20MP is the old 12MP....all you have to do is look the ratings on image quality for Nikon slew of 20+MP cameras. The D800 at 36MP is oustanding and is rated BEST of Dslrs in terms of image quality overall...DXO Mark scores. what does this tell you?
Sensors today are allowing excellent images at high pixel density so stop arguing with outdated rules. Judge on image quality as you see it!
The Iphone 4S photos are better on color accuracy but that is the end of it. Other than the HTC's photos being too warm for my taste, it appears to be superior for sharpness, noise, light sensitivity.
With that said, I am still amazed at the IPhone 4S photos...that is..until I got my Samsung Note II.:)

3 upvotes
Photato
By Photato (Apr 2, 2013)

A) Bigger pixels=noise free images.
B) More pixels= higher resolution.

If you want A & B you need a larger sensor.
For smaller sensors you need to pick A or B, but can't have both.

I think the HTC's One strike a great balance for total image quality.

Is unrealistic to expect pixel-sharp pictures with a 8+MP hand held smartphone.

0 upvotes
yabokkie
By yabokkie (Apr 2, 2013)

> Bigger pixels=noise free images.

tell me if 5D3 is noise free while D800 is not, and tell me how much the image quality will change along with pixel size. no one can give a theory because there is none.

there is no change in image quality when you change the pixel count dramatically (the pixel quality does change though but who cares).

Comment edited 3 minutes after posting
0 upvotes
Photato
By Photato (Apr 2, 2013)

Hell, this is not even a theory, it is a fact.
Bigger pixels are more sensitive.

0 upvotes
yabokkie
By yabokkie (Apr 2, 2013)

a fact that someone can see but no one can predict?
and what really those people see is the question.

0 upvotes
yabokkie
By yabokkie (Apr 2, 2013)

> This is an old and outdated logic.

never been a logic since the 1990s. even before that, Sony made first sensors with microlenses in 1989, and later Sony and Samsung made small BSI sensors for videos and mobiles that perform same or better than SLRs (per unit area).

it's interesting that we can scale pixel size freely dozens of times with little compromise of image quality though smaller pixels may perform slightly better.

Comment edited 5 minutes after posting
0 upvotes
CNY_AP
By CNY_AP (Apr 2, 2013)

If you read (and think) carefully, this should end this conversation once and for all:

You can downsize a high res image to get effectively "large pixels" (multiple smaller pixels get added together), but you can't do the opposite.

The author (Andy) carefully and correctly stated that the TOTAL amount of light collected by the WHOLE sensor is what matters for noise. Having more pixels affects the TOTAL amount of collected light in that each pixel has "some" area dedicated to electronics (non-light gathering area), therefore reducing the area being utilized to collect the light. So having 2x more pixels, IF all else remains the same, wastes 2x more surface area of the sensor. IF that "non-image gathering" area is small, say 5% of the sensor's total area, then you'll never notice the extra noise from having 2x more pixels.

So the question then becomes: how much of each pixel is "wasted" on the electronics? And for the area that IS collecting light, how efficient is the sensor?

0 upvotes
shaolin95
By shaolin95 (3 months ago)

Obviously you are clueless about photography but good try....

0 upvotes
Clint Dunn
By Clint Dunn (Apr 2, 2013)

16:9????? Really?? Man...I really want this phone but 16:9 for photos is just brutal. As the article mentioned once you crop to 4:3 you are only left with a 3MP image.

16:9 is ok for landscape snaps but useless for most portraits.

0 upvotes
Photato
By Photato (Apr 2, 2013)

Oh!! and why oh why!? Apple decided it was smart for the iPhone 5 to have a 4:3 camera sensor in a 16:9 device?
In that case the HTC sensor aspect ratio makes more sense.

And speaking of sense! I cant bring myself to understand why Apple and others decided to move to a 16:9 display ratio since most photographs, are either 4:3 or 3:2 and ergonomically the thumb finger has more reach in 3:2 format when operating it with one hand.

Comment edited 3 minutes after posting
0 upvotes
Photato
By Photato (Apr 2, 2013)

Were these photos taken with a tripod?
If so, these samples wont say much about the real world quality of these gadgets.
Personally, I prefer the colors on the HTC One though.
In reality, most high megapixel count cameras result in a bland noisy mess when shooting hand held.

0 upvotes
yabokkie
By yabokkie (Apr 2, 2013)

pixel count has nothing to do with camera shake and almost nothing to do (complex but small error) with noise.

you can change pixel count freely and see neither image blur nor image quality changes (they do change proportionally at pixel level).

Comment edited 4 minutes after posting
0 upvotes
Photato
By Photato (Apr 2, 2013)

The higher the pixel count, the less tolerant the whole camera system becomes of bad technique, bad optics, bad stabilization, subject motion, etc
I find ludicrous 8+ megapixel shots taken handheld in bad light. Those shots are not pixel sharp 99% of the time.

0 upvotes
yabokkie
By yabokkie (Apr 2, 2013)

there is no difference if you measure the blur in millimeters or say against picture diagonal.

don't count things in pixels which is not a stable unit to use. pixels should be infinitely small as possible (even to a fraction of the wavelength of light, like the best films we had).

0 upvotes
Photato
By Photato (Apr 2, 2013)

Nope. Even with film "pixel" size matters. Silver salts in high sensitive film are bigger for a reason. Likewise higher resolution film contains finer grain but is less sensitive. To get decent Dynamic Range and Gamut you need to collect discrete samples of photons per area, it just the basis of image forming. With blur you lose contrast and consequently sharpness.

1 upvote
yabokkie
By yabokkie (Apr 2, 2013)

less sensitive for standard development. you can significantly under expose a photo and with the knowledge of that exposure, develop the film correctly using customized procedure (for that frame and may destory all others, that we can do better in digital).

> With blur you lose ...
the blurs are the same as long as you shake the cameras the same. why we don't call low resolution image stablizer.

Comment edited 2 times, last edit 5 minutes after posting
0 upvotes
Photato
By Photato (Apr 2, 2013)

Not the same, the higher ISO film will show better results.

1 upvote
Zlik
By Zlik (Apr 2, 2013)

No comparison against the last generation Smartphones ?

What about the iPhone 5, the lumia 920 and the pureview 808 ?

That's ridiculous.

1 upvote
Roman Korcek
By Roman Korcek (Apr 1, 2013)

Does this test take into account the HTC One bug where when you set the ISO value manually the HTC One does not use it as an actual value but as an upper limit instead, also writing the upper limit into the EXIF data instead of the actual value used?
http://www.heise.de/foto/meldung/Smartphone-Kamera-Bug-HTC-One-schreibt-falsche-EXIF-Daten-1827489.html

Comment edited 12 seconds after posting
0 upvotes
Lars Rehm
By Lars Rehm (Apr 2, 2013)

We did not set the ISO manually for this test.

0 upvotes
Roman Korcek
By Roman Korcek (Apr 2, 2013)

Thank you for the clarification.

0 upvotes
marshim
By marshim (Apr 1, 2013)

The HTC lens seems to be wider than the one on Apple.
This way your 100% crops are kind of irrelevant imho, usually they are supposed to be at the same focal length.
Correct me if I'm wrong.

0 upvotes
Lars Rehm
By Lars Rehm (Apr 2, 2013)

well, firstly you can't have the same focal length if you have to fix-focal-length lenses. Secondly the HTC uses, due to its sensor, a default aspect ratio of 16:9 vs the Apple's 4:3. Vertical framing on these images is roughly the same, so yes, the 100% crops do make sense. With the HTC you just got some more stuff in your images on the left and right.

0 upvotes
yabokkie
By yabokkie (Apr 2, 2013)

it's mission almost impossible because it's difficult to hold a phone and take picts while swimming in the Thames.

to compare images of different aspect ratios, my suggestion is to forget about the frame borders and try to let each subject occupy the same area ratio in the frame.

in practice, calculate the linear ratio of one border so that the subjects will be projected to the same area ratio of each sensor.

Comment edited 6 minutes after posting
0 upvotes
Lars Rehm
By Lars Rehm (Apr 2, 2013)

that's what we've done, vertical framing is identical. That's exactly how we deal with different aspect ratios in our dpreview studio tests.

0 upvotes
yabokkie
By yabokkie (Apr 2, 2013)

for example if the subject's face occupies 12.3% of area in on image (like 24*36*0.123=106.3 mm2) we would want the same face occupy the same 12.3% portion of the area in another (like 17.3*13.0*0.123=27.7mm2).

regardless of the 12.3% portion, setting vertical framing to sqrt(2)*2/3=0.9428 (3:2 is slightly shorter) will guarantee same area ratio of any subject in both image.

0 upvotes
yabokkie
By yabokkie (Apr 2, 2013)

for 4:3 and 16:9 sensors, the ratio of PH should be 1.1547 (shorter PH for 16:9) for fair comparison, then the HTC One will look better than the samples here (it will be even better with a higher resolution sensor).

we did use PH extensively for TVs because we didn't have TVs of different aspect ratios.

Comment edited 3 minutes after posting
0 upvotes
gs85739az
By gs85739az (Apr 1, 2013)

Madison Ave has done well to get the buying public to shell out mucho dinero + MONTHLY cell phone charges to boot, when there are many alternative cell phones for MUCH less and less monthly fees also!

Can anyone here imagine buying a "real" camera and then paying a "monthly" fee to use it?

Wonder how long before we "believe" we "need" a camera that is also a pocket computer?

hint: it won't be long if Madison Avenue has any say....

Comment edited 2 minutes after posting
0 upvotes
Lars Rehm
By Lars Rehm (Apr 1, 2013)

Nothing stops you from buying a phone outright and not buy a voice/data plan for it. You don't pay any monthly fees and can do most things on it when you're connected to WiFi. I'd personally prefer to make a phone call with my phone though...

1 upvote
zodiacfml
By zodiacfml (Apr 1, 2013)

@gurmusic then get the 4.7MP previous Foveon sensor cameras before they get too worn and torn for your use. I love my Sigma DP2 for its sensor. Everything else about it sucks but I put up with it because the sensor is perfect for my needs and expectations.

1 upvote
gurmusic
By gurmusic (Apr 1, 2013)

Bacause of 8MP I still keep my EOS30D and I will only renew with a Faveon Camera or none...I Think 2MP (or 4MP incase of crop some) but good IQ best for Mobiles for 300 DPI prints standard 4"x6"

1 upvote
ThePhilips
By ThePhilips (Mar 31, 2013)

In a while, somebody made a totally rational decision about MP count. Congrats HTC!

3 upvotes
Dan Tong
By Dan Tong (Mar 31, 2013)

You need to check your article and correct the 2 upside down iPhone images.

1 upvote
Lars Rehm
By Lars Rehm (Mar 31, 2013)

this is how they were taken, we like to leave the originals untouched, so there is absolutely no doubt about the IQ being altered by some editing action. Just download and rotate yourself.

2 upvotes
The A-Team
By The A-Team (Apr 1, 2013)

Lars, I don't think we're going to question your decision to rotate an image right side up.

3 upvotes
mattkiefer
By mattkiefer (Apr 1, 2013)

April fools?

1 upvote
Stu 5
By Stu 5 (Apr 1, 2013)

The A-Team even rotating a j-peg and re-saving it will alter the quality.

1 upvote
zodiacfml
By zodiacfml (Apr 1, 2013)

lol. they weren't like that before.

0 upvotes
yabokkie
By yabokkie (Apr 1, 2013)

think it this way. the image quality of the thumbnails won't change and won't matter so put corrected ones there. let people download the upside-down ones and get lost.

0 upvotes
Lars Rehm
By Lars Rehm (Apr 1, 2013)

Ok, apologies everyone. Something really weird happened in our content management system. The thumbnails were the right way up when I put this piece together on Friday and at some point during the weekend they were rotated. We are investigating, but the thumbnails are corrected now.

1 upvote
tinternaut
By tinternaut (Mar 31, 2013)

HTC's solution is fine for photos straight out of the camera. Ultimately, I think I'd rather take the extra pixels and decide what to do with them myself.

0 upvotes
dpLarry
By dpLarry (Mar 31, 2013)

You don't get to decide when you got high pixel density and high noise.

3 upvotes
ZhanMInG12
By ZhanMInG12 (Apr 1, 2013)

But you can downsample and then apply a bit of NR. Also, there is good NR and there is bad NR, the in-camera type is certainly worse than Photoshop

0 upvotes
yabokkie
By yabokkie (Apr 1, 2013)

> downsample and then apply a bit of NR
should apply NR only which will consume/destroy details. downsampling is the worst case and last resort. that's what's wrong with Ultrapixel, the wrong order, wrong choice.

actually there may well be an 8MP sensor in the phone and they only give you 4MP output. this sounds stupid but at least it will perform better than a physical 4MP sensor.

Comment edited 1 minute after posting
0 upvotes
SLove
By SLove (Apr 2, 2013)

It appears to me that if there really is a 4 MP sensor in the HTC One, it caters shamelessly to people with the 'larger pixels are better' fallacy. It's still surprisingly prevalent even after proven wrong many times. Phil Askey founded a great site in Dpreview, but it seems that unfortunately one of his most enduring legacies is the unscientific belief in superiority of large pixels.

0 upvotes
Jefftan
By Jefftan (Mar 31, 2013)

no clear benefit

1 upvote
ThePhilips
By ThePhilips (Mar 31, 2013)

Smaller size of output images? Less battery drained when scaling down for web - because that's where most of mobile phone pictures end up? Less space is taken on the memory card and images are witten faster to it?

As mobile computing goes, there are plenty of advantages.

1 upvote
yabokkie
By yabokkie (Apr 1, 2013)

phones get very powerful these days and I don't think HTC have to tell that theirs are challenged.

0 upvotes
dpLarry
By dpLarry (Mar 31, 2013)

http://connect.dpreview.com/files/p/cms_posts/1916293861/htc_IMAG0040.jpg
http://connect.dpreview.com/files/p/cms_posts/1916293861/apple_low.JPG
http://connect.dpreview.com/files/p/cms_posts/1916293861/htc_IMAG0058.jpg
http://connect.dpreview.com/files/p/cms_posts/1916293861/apple_IMG_0694.JPG
http://connect.dpreview.com/files/p/cms_posts/1916293861/htc_IMAG0065.jpg
http://connect.dpreview.com/files/p/cms_posts/1916293861/apple_IMG_0711.JPG

1 upvote
clicstudio
By clicstudio (Mar 31, 2013)

NoKia might have a better camera on the phone, but I rather keep my iPhone.
Nokia's operating system has been obsolete for years and And they are not selling as Many phones as they used to. . If you're going to get a nokia phone just because of the better camera, might as well just get a real camera instead

1 upvote
Samuel Dilworth
By Samuel Dilworth (Apr 1, 2013)

Nokia Lumia phones have a new operating system, Windows Phone 8, which is fresh, beautiful, and efficient. It lacks some features (notably a good music player), and it suffers from a dearth of third-party apps, but it is the exact opposite of obsolete: in some ways it’s the most modern mobile operating system available.

2 upvotes
T3
By T3 (Apr 1, 2013)

I would seriously consider a new Nokia Lumia phone, if only it didn't run Windows Phone 8! I'd get one if it ran Android. I think Nokia has made a serious mistake in putting all their eggs into the Windows Phone basket. Too bad they don't also offer an Android version of their phones. Those versions would easily outsell the Windows versions.

1 upvote
clicstudio
By clicstudio (Mar 31, 2013)

It doesn't make sense. Why compare the HTC to a two-year old iPhone 4s?
The iPhone 5 is 10 times better.

0 upvotes
yabokkie
By yabokkie (Mar 31, 2013)

my impression is that iPhone 5 may be 1.2 times better than 4S, and this One 1.5 times better than 5.

about larger pixel size, though it doesn't contribute anything good to image quality, it doesn't seem to lose much in low light. it is, however, inferior otherwise because more pixels mean better NR (by making use of extra resolution info, which is hard when it's extremely dark, then binning/down-sampling is the way to go).

Comment edited 2 times, last edit 5 minutes after posting
0 upvotes
bobbarber
By bobbarber (Mar 31, 2013)

Yabokkie,

That's a wonderful explanation, and we've all heard it a million times from people who claim that more pixels does not damage image quality, but unfortunately, you are contradicted by the evidence. Did you even read the article and look at the sample shots?

On the other hand, it is a good tactic for you to stake out your turf and explain away, educating the rest of us, without any reference whatsoever to data or facts. That way, you get to maintain your self-image as an expert.

5 upvotes
yabokkie
By yabokkie (Mar 31, 2013)

> contradicted by the evidence

what evidence? you don't have.

0 upvotes
bobbarber
By bobbarber (Mar 31, 2013)

Confirmed--You did not read the article, or look at the sample photos. If you had, you would have seen that clearly a high-resolution image downsampled does NOT perform better than this camera in low light. Therefore, your authoritative-sounding statement--"it is, however, inferior otherwise because more pixels mean better NR "--is pure B.S.

I hope this helps.

4 upvotes
yabokkie
By yabokkie (Apr 1, 2013)

the new One does have better image quality but you cannot say that comes from larger pixels. it got a larger aperture lens than iPhone 4S by about 2/3 stops.

to offset this, comparisons should be done with different shutter speeds so that we can have same exposure on the sensors (think both of them 1/3.2").

Comment edited 2 times, last edit 8 minutes after posting
0 upvotes
Anepo
By Anepo (Apr 1, 2013)

Are you high? Typical ignorant fanboy, you must think giant purple spots make a better camera.... Because the iphone 5 camera is garbage, the lens was made with the worst material ever put into a camera and destroys the images, i would rather get an 4s, beside the point 5-10 times? Talk about hyping it up.

Also @yabokkie what the hell kind of drug are you on? Just stop talking, you have proven your ignorance and ZERO knowledge when it comes to photography.

1 upvote
Timmbits
By Timmbits (Mar 31, 2013)

So a phone has less noise in low light.
Let me change carriers and phones right away!

@DPR: I don't need this inconsequential information on the main DPR site. I understand that you are trying to launch this connect thing, and maybe spin it off, but annoying a photographers audience with putting connect into their feed may not be the best way to grow readership in your new venture. And BTW, the settings do NOT stick! If I tell it to not show "connect", it eventually comes back, even with "make default". I'm sure this is within the scope of the competence of amazon's programmers...

2 upvotes
dpLarry
By dpLarry (Mar 31, 2013)

Other photographers like myself don't mind seeing new developments in phone cameras because they are another photographic tool.

The best camera is the camera you have on you.

6 upvotes
Dan W
By Dan W (Mar 31, 2013)

Don't read it if it doesn't interest you. Simple.

3 upvotes
Anepo
By Anepo (Apr 1, 2013)

Welcome stupid drama queen nr 1 million.

You are not the first attention seeking prostitute to come here and complain and neither the last, dont like what they are doing? Then why spend time on it? Go find another website for your dumb ass.

1 upvote
IcyPepsi
By IcyPepsi (Apr 1, 2013)

> The best camera is the camera you have on you
This has been said a thousand times before and I disagree. There are spy pen cameras that you can run out with just as easily as a phone cam. So does that also make the best camera there is? Also, not everybody's style is photo-journalistic. There are people who want to use real flash(s). let alone off-camera.

0 upvotes
dpLarry
By dpLarry (Mar 31, 2013)

Some of the pics are upside down, did someone hack the website?

0 upvotes
Peter_H_77
By Peter_H_77 (Mar 30, 2013)

The low light shots from both phones are unimpressive. 4mp is a comparatively low resolution these days, and pixel size alone is not going to shift these phones in volume. HTC are in trouble and I really can't see the mass market appeal at all.

3 upvotes
dpLarry
By dpLarry (Mar 31, 2013)

What are you expecting? Full frame sensor low light shots?

If you click on the pyramid building shots of the iphone and HTC you'd see the difference.

1 upvote
Peter_H_77
By Peter_H_77 (Mar 31, 2013)

Full frame sensor low light shots? Sure - that's what I was expecting....... From a smart phone - yeah. That's a stupid question mate. My point is that these images are not exceptional by any means, and this phone will not turn around HTC's fortunes.

0 upvotes
yabokkie
By yabokkie (Mar 31, 2013)

> pixel size alone is not going to shift these phones in volume

pixel size doesn't mean good image quality (unless low resolution does). the lens is about f/15+ equiv. on 35mm full-frame, which is better than f/18+ of iPhone 4S. a full-frame sensor won't help with such a tiny lens.

2 upvotes
dpLarry
By dpLarry (Mar 31, 2013)

If you can't see the improvements from the 12 preview samples and the pyramid building night shot pics (click and viewed full screen), then you're pretty dumb.

Comment edited 1 minute after posting
1 upvote
dpLarry
By dpLarry (Mar 31, 2013)

yabokkie clearly you don't understand photography I don't even want to explain it to you.

3 upvotes
Steven Noyes
By Steven Noyes (Apr 3, 2013)

The sad thing is, I think the balance of design question really points to HTC having the correct decision. Most people would greatly benefit by their design choices in the new HTC One. Lower MP and really good IQ in low light conditions. In bright conditions, you still end up with an photo you can print at 8X10 if you want to and in low-light, you end up with a photo you can print and share that is higher quality than the competition.

The sad thing is, the consumer won't buy into this logic. Apple did a similar thing on the 4S/5. They increased the MP but did not decrease the pixel size. Smart.

0 upvotes
bobbarber
By bobbarber (Mar 30, 2013)

Why can't somebody do this with a "regular" camera?

I realized long ago (as did many others) that more megapixels was counterproductive in many situations. I got sharp prints at 8x10 from my 2 megapixel Kodak.

What I do now with higher megapixel cameras, is shoot them at reduced resolution. But these images seem to show that that is not as good a tactic as just having fewer, and better, megapixels on the sensor in the first place.

Please, for me, 7 megapixels was more than enough for everything, and didn't unnecessarily clog up my hard drive. I can't be the only one that feels this way.

5 upvotes
kmarsh
By kmarsh (4 months ago)

The Canon G series has always done this, used larger, lower MP count, more sensitive sensors with reasonably sized lenses in a bridge camera format.

0 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

> 1/15 sec

I noticed that images are compared at the same shutter speed instead of advertised ISOs, which is great.

my experience is that shutter speeds may not be the same, either. for comparisons between cameras, firing a flash for everyone is a good idea but I have no idea how to sync the phones.

return to shutter speed, it can be calibrated easily using a CRT display (anything that's stable and fast moving, the scan beam here). please find some for your lab. the last one I had broke several years ago.

0 upvotes
Lars Rehm
By Lars Rehm (Mar 31, 2013)

This is coincidental. Both camera phones were shot in Auto mode as they do give very little controls of exposure (the iPhone even less than most Android phones).

1 upvote
yabokkie
By yabokkie (Mar 30, 2013)

> 100% crop

comparison of 100% crops is effectively the same as comparison of pixels and should have exactly the same result, be it one vs one or a million vs a million.

if it's the image quality that you are interested, you should compare the same (portion of) images, eye for eye or tooth for tooth (not tooth for teeth, or lips for face).

0 upvotes
dpLarry
By dpLarry (Mar 30, 2013)

Just click on the pics above and download them and view them full screen.

I find that the HTC is better in every case. Better color, dynamic range. Just the way it captures light is more pleasing and dare I say, more cinematic.

And 4mp, it is actually ... Sharper!

7 upvotes
ajendus
By ajendus (Mar 30, 2013)

Nice results from the HTC (and even from the iPhone 4S). I think it would have been more appropriate to have compared the iPhone 5 rather than the 4S, however. The iPhone 5 has significantly better results than the 4S.

0 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

not big deal but I think this is a 1/3.2" sensor of near 16 mm2. not 1/3" as HTC call it.

a larger pixel sensor won't deliver better image quality but it does contribute to lower costs of the sensor itself, the processing chip and memory, as well as more tolerance to lower quality lens and harsh assembly.

2 upvotes
dpLarry
By dpLarry (Mar 30, 2013)

Where did you get the idea that it contributes to lower costs? I've already stated at the bottom of this page that sensor costs has everything to do with size and practically nothing to do with pixel count. Most people would imagine... geez, 8 million pixels vs 4 million pixels on a tiny sensor, of course the 4 million pixel sensor will be cheaper. But that's not the case.

When I looked at sensor manufacturer's prices 1/3" chip, it's the same price whether it is 4 MP or 8MP or 13MP. It's only more expensive it it is a newer chip. Or larger chip.

The lens is almost certainly the same. It's a tiny simple fixed lens.

Yabokkie, I've noticed from many posts you are very good at making stuff up.

5 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

> practically nothing to do with pixel count.

large pixels can be fabbed with old generation equipment, that you need huge investment to update. isn't that simple enough to understand?

2 upvotes
micahmedia
By micahmedia (Mar 30, 2013)

Yeah, it's unfortunate that DPR doesn't give you a way to ignore the comments of certain users. I find myself reading the comments sometimes and thinking wtf?! before I realize who it is. Yabokkie is top of my list of people to ignore.

6 upvotes
johnmcpherson
By johnmcpherson (Mar 30, 2013)

It does however, improve low light performance and contrast.This is simply the nature of the bestie; more area in the pixel sensor itself equates to more photons for the sensor to sample, i.e. more photons means a larger sampling for the idividual pixel sensor.

4 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

if we think that the photons arrive at the sensor through the lens aperture are the same, changing pixel size/count merely means how we divide them differently and thus has no effect on the image quality.

or think that you have a million dollars. you have a million no matter it's all Washingtons or Hamiltons, you won't get any richer to have all Franklins, and you won't lose a penny changing to all one cent pieces.

Comment edited 3 times, last edit 13 minutes after posting
2 upvotes
canon20s
By canon20s (Apr 25, 2013)

Except there is a loss when you divide it. Imagine changing all million dollars to pennies, but having to count EACH penny. You will inherently loose more money miscounting pennies than $10,000 bills. Each pixel has to count photons, but also has its own read noise. Less photons PER pixel (smaller pixels) + the SAME read noise per pixel means LESS Signal to Noise.

0 upvotes
danroso
By danroso (Mar 30, 2013)

Sensor size and aperture are more important than pixel size. Nokia 808 still the King!

8 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

it's the aperture (size in area or diameter).

0 upvotes
Essai
By Essai (Mar 30, 2013)

the king of the ghost phones. I have never seen a new Nokia phone yet (2011+)

2 upvotes
SLove
By SLove (Mar 31, 2013)

Depends on where you live. HTC is not exactly a popular brand where I live, but it has nothing to do with the quality of their phones; the brand just isn't that well known here. In any case, this is not a high school popularity contest, is it?

1 upvote
yabokkie
By yabokkie (Mar 31, 2013)

I think HTC make very good phones and their new One looks good. only I don't like this "ultrapixel" junk. it's against simple maths (not even physics, people believe in large pixels because they cannot even count numbers).

1 upvote
SLove
By SLove (Mar 30, 2013)

The closest technological peer of the HTC One is the Nokia Lumia 920 with its optical IS and f/2.0 lens. While I don't own neither camera, judging from the various online examples I'm pretty sure the HTC One would show no real advantage over the Lumia 920 in low light shots with the latter scaled back to 4 MP. Of course the Lumia 920 would still show better resolution at full resolution.

4 upvotes
Essai
By Essai (Mar 30, 2013)

byt who wants a Nokia ? Nobody...

2 upvotes
dpLarry
By dpLarry (Apr 1, 2013)

Yes Essai, people wanted a Nokia... 10 years ago!

0 upvotes
SLove
By SLove (Apr 1, 2013)

Nokia is still the leader of phone camera technology. Was already 10 years ago and still is. As for the popularity contest: Nokia N and E series smartphones were extremely popular everywhere except the US just five years ago. The iPhone 3GS was the first iPhone that sold anywhere near the numbers of bestselling Nokia smartphones globally.

Comment edited 36 seconds after posting
1 upvote
CAClark
By CAClark (Mar 30, 2013)

It always seems to me that photos from phones are never fantastic, and only look decent when down res'd to go on the interweb.

It seems this kind of in-depth camera analysis of a phone is slightly tragic. Sure it's harmless throw away info to band around in conversation, but every photo ever from a phone looks great until you see it at 100%, and then it looks like hell.

Anyone sufficiently interested in this level of technicality of a phone camera should really go and buy an actual camera. They'll have some fighting chance of actually good 100% photographs.

3 upvotes
dpLarry
By dpLarry (Mar 31, 2013)

"Anyone sufficiently interested in this level of technicality of a phone camera should really go and buy an actual camera."

We all already have 'an actual camera', most likely several cameras.

3 upvotes
CAClark
By CAClark (Mar 31, 2013)

DpLarry

That's my point, I don't really know who this pointless pixel peeping of a phone camera is aimed at :-) People who pixel peep like this are unlikely to care much as long as a photo us presentable on a phone screen or in a web browser, on account of their 1 at least actual camera.

0 upvotes
KoKo the Talking Ape
By KoKo the Talking Ape (Mar 31, 2013)

Well, I guess I am the kind of person who this article is aimed at. I don't always have a camera on me, so the phone, while a distant second to a real camera, is better than nothing. Knowing that, what is wrong with finding out what phones make better cameras than others? I have to choose a phone, why not make it one that takes better pictures than the others?

2 upvotes
MaRcIu
By MaRcIu (Mar 30, 2013)

Is this test with the updated firmware? because I saw some samples before and after update in daylight and the details are sharper.

3 upvotes
ciresob
By ciresob (Mar 30, 2013)

Seems clear to me the HTC has better DR - no clipped highlights clouds, less noise in shadows under bridge, even though higher ISO was used. Seems to work as expected.

1 upvote
HowaboutRAW
By HowaboutRAW (Mar 30, 2013)

Raw would help too.

0 upvotes
zodiacfml
By zodiacfml (Mar 30, 2013)

They look alright in web sizes but still awful at 100%, inefficient strong noise reduction and high sharpening. Whatever, I set my phone to take 1600x1200 images. Smartphone cameras just don't deserve to take high resolution photos.

DPR, why not make an article comparing different camera formats reduced to 4-8MP resolution from their native resolution? That would be interesting. That would be easy too if you were to use the images in the studio comparison tool.

0 upvotes
coudet
By coudet (Mar 30, 2013)

Maybe HTC should try "hyperpixel" next - even larger pixels for even lower performance than the HTC One?

0 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

no one can beat a "single pixel camera" if the pixel quality is important.

0 upvotes
GatanoII
By GatanoII (Mar 30, 2013)

What's the point of comparing an Android phone camera to an iOS phone, no one is going to switch system based on a camera.

Redo the test with the best selling Android phone as a reference and then it's a fair comparison, Galasy S3 vs HTC One that's what could be interesting, even if the One must be compared to the S4 as this are the phones that are going to battle in the stores ... and do extended video comparison as many people care more about video on a phone than photos now, since memory space is not a problem any more (especially the phones with big microSD cards)

4 upvotes
JadedGamer
By JadedGamer (Mar 30, 2013)

It's not about "battle in the stores" - though thanks for pointing out indirectly why so few Android-using manufacturers actually make money - but comparing to the currently most popular camera phone, which remains the iPhone line. Presumably, they did not have an iPhone 5 available, but the main difference between that and the 4S is in the sturdiness of the lens.

0 upvotes
TimT999
By TimT999 (Mar 30, 2013)

Actually the iPhone 5 is significantly better than the 4S in low light -- a feature that several of the tech sites have pointed out. Here's the Gizmodo review comparing the iPhone 5 with the 4S:

"Apple's greatly improved the image processing on the iPhone 5 [over the 4S] such that the camera can now shoot up to a sensitivity of ISO 3200 while still ending up with less noise. This is a sea change."

So a test between the HTC and the 5 would have been a fairer test in terms of capabilities. And of course why compare phones using an older model anyway unless you are trying to give the new phone a little extra advantage. Regardless of the reasoning, using a 4S instead of the 5 was the wrong choice.

Comment edited 2 minutes after posting
0 upvotes
T. L. Rutter
By T. L. Rutter (Mar 30, 2013)

What's the point of comparing an Android phone camera to an iOS phone, no one is going to switch system based on a camera.

Actually, when I consider a smartphone, I also consider the camera within the smartphone. The camera is the center of all social networking, when we can snap beautiful, attractive well-framed pictures showing where we are or what we are doing and share it with the world. For those of us who are avid photographers, having a smartphone that has a decent camera is a must and represents our work. And yes, 98% of those who are avid photographers would jump ship if the "other side" produces a better camera for their smartphone that is on the wireless carrier that works best for the end-user. I would have bought the Nokia 920 over the galaxy note 2 had the camera been better and the wireless carrier wasn't AT&T.

0 upvotes
ThePhilips
By ThePhilips (Mar 31, 2013)

"no one is going to switch system based on a camera."

DPR is read also outside of US, you know. There are such places where you do not need to sell your organs to change the carriers or phones.

0 upvotes
tkbslc
By tkbslc (Mar 30, 2013)

Looks good in low light, but in daylight it is a wreck.

0 upvotes
ThePhilips
By ThePhilips (Mar 31, 2013)

"[...] but in daylight it is a wreck."

At 100%. And nobody - especially on a mobile phone - looks at images at 100%.

0 upvotes
CyberAngel
By CyberAngel (Apr 3, 2013)

ThePhilips:
My mom takes paper prints
and she uses her phone
I know others who go to pictures on a PC
(including myself, who also zoom on phone)
so "nobody" is replaced by "people seldom", ok?

Comment edited 31 seconds after posting
0 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

larger pixels deliver better pixel quality,
smaller pixels deliver better image quality.

1 upvote
dpLarry
By dpLarry (Mar 30, 2013)

No, the opposite.
Larger pixels deliver better image quality.
Smaller pixels deliver better pixel peeping quality.

9 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

> larger pixels deliver better image quality

not true unless what you mean by image quality is really pixel quality, or there is now way you can prove it.

look at the comparisons above. they all compare a portion of HTC to a different portion of half area from compitition. this is unfair. under this condition, we can claim that HTC is better only if it can deliver more than one stop (1.4 times) better SNR.

in physics, pixel size does not affect image quality (if the same portion is compared). in reality smaller pixels perform better because higher gains (higher ISOs) can beat readout noise better.

a good example is D800 vs D600/5D3, all of them state of art cameras. though D600/5D3 got nearly twice as large pixels they don't provide better image quality.

Comment edited 3 minutes after posting
1 upvote
Digital Suicide
By Digital Suicide (Mar 30, 2013)

In my opinion, phone camera must perform best in daylight. Personally I try to avoid use it at dark times, because results are very bad. I have dedicated cameras for that.
So this phone has no advantage over any other, even if it performs a little bit better in low light situations, but no better in daylight.

Comment edited 28 seconds after posting
2 upvotes
dpLarry
By dpLarry (Mar 30, 2013)

Most people, esp girls, use their phone cams extensively at night, at parties, concerts etc.

So your opinion is just your opinion.

8 upvotes
xarcex
By xarcex (Mar 30, 2013)

the iPhone has always had a great camera, shame on Apple and its closed system though :(

2 upvotes
JadedGamer
By JadedGamer (Mar 30, 2013)

Why? The popularity of the platform among developers means these is a vast selection of photo apps and image editing opportunities there. No shame in that at all.

You see, openness alone does not bring benefits. You cannot run what you want on an Android, only what someone have bothered with writing for it. This is also why the Ouya will struggle against the "closed" console platforms.

1 upvote
SLove
By SLove (Mar 30, 2013)

The iPhones prior to 4 had a very mediocre camera. Many mid-range Nokia phones from 2009 (e.g. 6720 and 6710s) had a much better camera than the 3GS. In fact their cameras were about as good as the iPhone 4 camera. Only with the iPhone 4s Apple finally managed to clearly beat the Nokia Carl Zeiss 5 MP camera module in IQ, which was introduced in 2007 with the Nokia N95.

3 upvotes
nniicckk
By nniicckk (Mar 30, 2013)

I am sorry, but this entire test/preview comes across as badly orchestrated.

A phone camera is not a telephoto. The zoomed in daylight picture of Big Ben is a test of resolution, not of the benefits of larger pixel size. In this case, the camera with more pixels to spare will show a sharper picture when you zoom in.

The same goes for the low light building picture. You are cropping away a large part of the image, so it isn't just a test of sensor noise, but also one of resolution.

If you really want to test the sensor, compare it in a scenario where camera phones are commonly used and currently suck. Take a badly lit scene in a pub and see which sensor handles it better. Focus on low light urban portraits, for instance, the ubiquitous peace sign drunk pics people take, and see which phone camera does better.

Just to clarify, I have no skin in this review. I see HTC bucking a meaningless megapixel trend and would really like to know if it gives me anything.

7 upvotes
Lars Rehm
By Lars Rehm (Mar 30, 2013)

I am not quite sure what you are trying to get at but I can confirm that is not meant to be a test of just sensor noise. These images tell you something about resolution too and some readers might even be interested in what happens when you use a digital zoom on a 4MP camera. Btw the low light images have only been cropped for layout purposes, if you click on them you still get to see the full version.

8 upvotes
Oddrain
By Oddrain (Mar 30, 2013)

I agree a few typical social media examples compared at the same size would be good. The question for some smart phone users will be whether the trade off of lower detail in good light is worth the better performance in some of the typical low light social situations. In any case it would be a good excuse to spend an afternoon in the pub whilst actually "working" !!

0 upvotes
nniicckk
By nniicckk (Mar 30, 2013)

A test is only meaningful if it tests something valid. The HTC claim is that by sacrificing megapixels and using a larger pixel size, they can offer better picture quality in many common usage scenarios.

The premise is that resolution is being sacrificed. By definition, it is not a test metric. The use case envisioned by the manufacturer is that people do not want to take hires pics with a smart phone and then crop them for use with social media.

As an analogy, if a car manufacturer claims that to make a car go really fast, they are going to sacrifice fuel efficiency. Should someone even consider fuel efficiency or the size of the gas tank as a test metric?

In your case, 3 of the 4 image tests focus on resolution. There isn't a single test that shows how well the system reproduces color or retains detail in low light. The tests don't help me if it is worth making the resolution tradeoff to gain something else.

Perhaps you can cover that in a future review?

5 upvotes
dpLarry
By dpLarry (Mar 30, 2013)

nniicckk, I couldn't have said it better.

The way most people are using phone cams, if you take a hundred pics with the HTC and iphone and Samsung. And the way most people use the pics, looking at them on the computer screen or posting to social media sites, my guess is that the HTC will come out pretty good, if not on top.

And it's got an optical stabilizer to boot!

I downloaded the pics on this page and looked at them full screen and the HTC beats the iphone in both daylight and night shot.

So here's a manufacturer putting a sensible camera design into a phone, 'bucking a senseless megapixel trend'. However the casual mass consumer coming to an influential site like this one look at this review, look at the first few pixel peeping crops and will say... this phone is just not as good.

0 upvotes
Lars Rehm
By Lars Rehm (Mar 30, 2013)

@nniicck don't worry, our full review will of course look into image quality in much more detail than this.

3 upvotes
phr0g
By phr0g (Mar 30, 2013)

I must echo some of the above comments. Cropping in this test is irrelevant. What matters is how good the scaled image looks on a screen.
On my 17" laptop I prefer the HTC images in each case.
(I say this as an owner of both an iPhone 5 and an HTC One X - And in the case of these 2 cameras, the iPhone wins with a similar testing methodology).
In the end, I use the camera on my phone for snaps and social media, and as such, a 4 MP picture with higher quality is very useful. Who needs storage-space-hogging 13 MP images on a phone, which at best will mostly be a waste of pixels?
If I want truly decent images then I'll use one of my real cameras.

1 upvote
LensBeginner
By LensBeginner (Mar 30, 2013)

I'm with the reviewer on this one: low light performance comes in the HTC One at the cost of resolution, so you have to appraise the tradeoff (i.e. test both parameters) in order to find out if it was something worth trying or not.
In this case, IMHO and based on what I see I'd say no.

0 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

> The premise is that resolution is being sacrificed.

at no profit of the user ... maybe saving of the storage?

Comment edited 4 minutes after posting
0 upvotes
zodiacfml
By zodiacfml (Mar 30, 2013)

If I could just have some extra cash, I would but this immediately just because of the sensible sensor with image quality comparable to a point and shoot, besides the build quality of the phone is the best I have seen yet from an Android.
In fairness to iPhone, it still has better white balance or colors.

0 upvotes
Sam Carriere
By Sam Carriere (Mar 30, 2013)

Oh, if only DPreview were as quick to review cameras as it is phones.

1 upvote
Lars Rehm
By Lars Rehm (Mar 30, 2013)

If you look closely you will see that this is not a review (although I would have liked to do one by now).

2 upvotes
ErikH
By ErikH (Mar 30, 2013)

I tend to agree with Sam. I appreciate this article but I am very interested in some of the (real) camera reviews which have only been previewed many months ago.

I'm not trying to be disagreeably, just honest. So many cameras, not enough time. You all need more money. :)

0 upvotes
yabokkie
By yabokkie (Mar 30, 2013)

> you will see that this is not a review

haha, definitely not.

0 upvotes
rfsIII
By rfsIII (Mar 30, 2013)

The paradox of photography is that any sensor can produce usable pix in midday sunlight, but that's the worst time to take pictures because there are no shadows to give form to objects, and people squint like moles. The best photos are at the edges of the day or after the sun has down so it's wonderful that someone has finally created a cellphone cam that can shoot in a dim interior, or under outdoor streetlight. Bless you HTC. And Seattle, in addition to your normal tests, please shoot this camera in typical social situations such as art openings, jazz clubs, and all-night rave parties.

Comment edited 1 minute after posting
2 upvotes
Flee808
By Flee808 (Mar 30, 2013)

I have always enjoyed reading your interesting reviews. Your review helped me decide to buy the 808. Just a request- when you compare image quality between 2 phones, is it possible to do it by ISO, say A versus B at ISO 400? This will give a better picture of the phone's performance at a specific ISO. How does the HTC One compare with the Samsung Galaxy S2 at ISO 400? Thank you.

2 upvotes
Lars Rehm
By Lars Rehm (Mar 30, 2013)

The iPhone actually doesn't give you any control over ISO, it automatically picks the ISO for a given scene, which makes it virtually impossible to compare both phones at the same ISO. The HTC can use a lower ISO in the same scene because it's got a faster lens and optical IS, so I think it makes a lot of sense to compare images with different ISO, at the end of the day it's the end result for a given scene that counts.

4 upvotes
kessler
By kessler (Mar 30, 2013)

It struck me that comparison at the same shutter speed was a great idea.

0 upvotes
Lars Rehm
By Lars Rehm (Mar 30, 2013)

Not sure how serious your comment is, but just in case, you cannot select shutter speeds manually either :-)

0 upvotes
zonoskar
By zonoskar (Mar 30, 2013)

Of course you can compare ISO for ISO with an iPhone. Just take the picture with an iPhone and replicate the chosen ISO using manual selection on the other device. Just because the iPhone doesn't give you control doesn't mean you can't compare.

0 upvotes
yabokkie
By yabokkie (Apr 1, 2013)

> which makes it virtually impossible to compare both phones at the same ISO.

you never have to do it. actually nobody should compare cameras at the same ISO. basically we should compare them at the same exposure, the same shutter speed, etc. ... ISO is only a tool to do that and it's a bad reference for digital cameras.

Comment edited 33 seconds after posting
0 upvotes
dpLarry
By dpLarry (Mar 30, 2013)

DPreview is one of the main culprits of pixel peeping. I look at my pictures full screen . Not take a 100x100pixel square.

0 upvotes
Lars Rehm
By Lars Rehm (Mar 30, 2013)

What are you trying to say?

0 upvotes
Total comments: 172
12
About us
Sitemap
Connect