Canon g1 x mark 2 review

If ever DPR gets to reviewing the G1XII, we will not be able to compare IQ with the G1X due to the changed test scene chart.
You might be able to.... if Dpreview redoes their studio test shots for the G1X. They actually have done this for several recent models that were reviewed right before they changed their test charts.

Personally, I don't think it's likely, since the G1X is a niche product, but it could happen.
 
If ever DPR gets to reviewing the G1XII, we will not be able to compare IQ with the G1X due to the changed test scene chart.
I would think that dpreview would do direct comparison as they did in the past. But we still have imagingresource.com as well for direct comparisons once they tested it.

Looking at resolution charts it's also quite soft at full zoom and large aperture. Will be interesting to see what dpreview will find although they may not test all combinations (certainly studio comparison is typically at best settings).
 
If ever DPR gets to reviewing the G1XII, we will not be able to compare IQ with the G1X due to the changed test scene chart.
You might be able to.... if Dpreview redoes their studio test shots for the G1X. They actually have done this for several recent models that were reviewed right before they changed their test charts.

Personally, I don't think it's likely, since the G1X is a niche product, but it could happen.
The G1 X is already there.
 
Same issue my Sony RX100M2 had.. except it was at 28mm F1.8 and only became useable at F2.8

http://www.dpreview.com/forums/thread/3593706

When you make a lens that covers such a wide range, there will be compromises.

Not unusual really.
yes, it seems like this is what you get when making a bright lens for larger sensor too compact. I'm really starting to doubt the value of that bright lens, which IMHO is the main selling point of the G1X II. I think I would prefer a slower lens (e.g. f/3.5-5.6) with the same focal length range that is sharp wide open (and smaller / lighter and possibly cheaper).

I hope the Nikon P8000 gives better results near wide open; the smaller sensor and experience with some previous Coolpix and Nikon 1 lenses should help.
Same was said about S100 few years ago when going from 28 to 24mm and apparently also with Sony (and tons of other camera's). This is typical especially with f2.0 and wide 24mm (unless you are buying a $1000+ lens). Checkout some of the top lens reviews on dpreview just for reference.
I don't think this is related to lens price, it is first of all related to lens construction (too compact). The S100/110 (and some other bright/WA 1/1.7" compacts) has this problem to a minor degree. A bit soft wide open, but very good one stop down; and less sharp corners at WA due to 'stretching'.

In RX100 the problem is already more obvious, visible sharpness loss in borders/corners at WA unless stopped down several stops.
Note however don't expect it in daylight conditions to choose f2. It would only choose that with low light conditions and in that case you'll lose some detail anyway (certainly if alternative is higher ISO). Also remember you can add more sharpening but not all pictures will look like a resolution chart...
Some people want to use f/2 in medium to good light levels, where the sharpness loss will be painfully obvious (judging from the first results). For landscape / architecture style images I try to keep ISO as low as possible for better sharpness and DR, and for this type of images sharpness loss in the corners is often undesirable (soft corners will be far more obvious that e.g. in people or low light shots). I can accept softness full open if it is really sharp by f/4 in the WA range, but even that is questionable.
Until we see dpreview studio comparison still bit hard to do a good comparison between previous model and other cameras. Don't expect huge difference at lowest ISO compared to camera's with smaller sensors. Biggest difference would be in low light in theory given much larger sensor.
As mentioned before, I'm not interested in high ISO performance; btw, any improvement will be the result of in-camera processing which you can also realise with better processing from RAW.

I would not judge lens quality from the DPR studio shot - very unrepresentative conditions for my type of photography (relatively close distance, low contrast, uniform WB, no flare / specular highlight isssues etc.) and VERY sensitive to alignment and other issues.
 
Same issue my Sony RX100M2 had.. except it was at 28mm F1.8 and only became useable at F2.8

http://www.dpreview.com/forums/thread/3593706

When you make a lens that covers such a wide range, there will be compromises.

Not unusual really.
yes, it seems like this is what you get when making a bright lens for larger sensor too compact. I'm really starting to doubt the value of that bright lens, which IMHO is the main selling point of the G1X II. I think I would prefer a slower lens (e.g. f/3.5-5.6) with the same focal length range that is sharp wide open (and smaller / lighter and possibly cheaper).

I hope the Nikon P8000 gives better results near wide open; the smaller sensor and experience with some previous Coolpix and Nikon 1 lenses should help.
Same was said about S100 few years ago when going from 28 to 24mm and apparently also with Sony (and tons of other camera's). This is typical especially with f2.0 and wide 24mm (unless you are buying a $1000+ lens). Checkout some of the top lens reviews on dpreview just for reference.
I don't think this is related to lens price, it is first of all related to lens construction (too compact). The S100/110 (and some other bright/WA 1/1.7" compacts) has this problem to a minor degree. A bit soft wide open, but very good one stop down; and less sharp corners at WA due to 'stretching'.

In RX100 the problem is already more obvious, visible sharpness loss in borders/corners at WA unless stopped down several stops.
You might be right that S100/S110 is better at f2 but I can't quantify it yet until we see direct comparison (and reviewer said wasn't fully sure about results). Imaging-resource has dedicated tests for these so will be interesting to check it out there.
Note however don't expect it in daylight conditions to choose f2. It would only choose that with low light conditions and in that case you'll lose some detail anyway (certainly if alternative is higher ISO). Also remember you can add more sharpening but not all pictures will look like a resolution chart...
Some people want to use f/2 in medium to good light levels, where the sharpness loss will be painfully obvious (judging from the first results). For landscape / architecture style images I try to keep ISO as low as possible for better sharpness and DR, and for this type of images sharpness loss in the corners is often undesirable (soft corners will be far more obvious that e.g. in people or low light shots). I can accept softness full open if it is really sharp by f/4 in the WA range, but even that is questionable.
Why do you want to shoot f2 if low-light isn't a concern? Shallow DOF?
Until we see dpreview studio comparison still bit hard to do a good comparison between previous model and other cameras. Don't expect huge difference at lowest ISO compared to camera's with smaller sensors. Biggest difference would be in low light in theory given much larger sensor.
As mentioned before, I'm not interested in high ISO performance; btw, any improvement will be the result of in-camera processing which you can also realise with better processing from RAW.

I would not judge lens quality from the DPR studio shot - very unrepresentative conditions for my type of photography (relatively close distance, low contrast, uniform WB, no flare / specular highlight isssues etc.) and VERY sensitive to alignment and other issues.
We all have different needs obviously but some people want less noisy high ISO performance for handheld low-light photography (RAW is not doing any miracles). If you don't care about high ISO you might as well consider G16 or other camera's in this class since dynamic range and resolution is very similar at a much lower price.

If 'sharpness' is important you could try to apply more unsharp mark but in terms of resolution the G1X v1 wasn't really very high. For that you're better off with something like A7/A7R/RX10.
 
Last edited:
Here's btw imaging resource corner test with S110 (I do have an S100 as well):

http://www.imaging-resource.com/PRODS/canon-s110/canon-s110A4.HTM

Not that great either but again no idea how to quantify it yet until we see better test:



ZS110hVFAWB-250UL.JPG


S110 F2 upper left corner
 
That's my question as well,

If you're using F2, you either want more bokeh or are shooting in very low light. For this lens, the bokeh (from marco's explanations) isn't great at F2 anyways, and the bokeh at F4 looks good.

If you need sharpness, then go to F4

The camera will most likely not pick anything before F4 in good lighting conditions

I just returned a S120 because how darn soft it was at F1.8 @ 24mm, and how bad the wifi function worked.

but the IQ of the G1X Mark II seem to be a lot better, so I'll manage.
 
That's my question as well,

If you're using F2, you either want more bokeh or are shooting in very low light. For this lens, the bokeh (from marco's explanations) isn't great at F2 anyways, and the bokeh at F4 looks good.
sometimes I want limited DOF, but quite often I use a large aperture because it enables me to use decent ISO handheld. Many of my shots are with subjects near infinity, where a limited DOF (e.g. from 10 meters to infinity) is all I need.

Increasing ISO beyond a certain limit will negatively impact DR and noise. With my 450D camera, when taking landscape type shots around golden hour being able to work at f/2.8-f/4 instead of f/8 is very valuable. My experience is that 'better High ISO performance' is not magic, but really quite limited when you work from the RAW file; so I prefer not to depend on High ISO performance, but have a usable large aperture instead.
If you need sharpness, then go to F4
as mentioned, if it is really sharp at f/4 it is probably acceptable to me; but in that case I'm still wondering why I'm paying (in price/size/weight) for such a bright lens...
The camera will most likely not pick anything before F4 in good lighting conditions

I just returned a S120 because how darn soft it was at F1.8 @ 24mm, and how bad the wifi function worked.
I tried an S110 for a week and did a lot of testing, also comparing it to my 450D with 15-85IS lens. Based on internet reviews I was expecting dubious corner sharpness in the WA range, but this proved very acceptable especially when using RAW. The slightly 'soft' rendering full open but was no problem for me, it was acceptable IMHO and completely gone by f/2.8. Only at 120mm equiv. the rendering looked very soft at any aperture (note: it only looked soft due to low micro-contrast, there actually was a surprising amount of small detail).

My main problem with the S110 was that due to all the in-camera processing (for vignetting, distortion etc.) there are obvious artefacts and detail loss in the corners, especially in the jpeg. Even at low ISO the borders can turn into 'pea soup' compared to my 450D (while the center of the image looks really good). Working from RAW and finetuning all processing the results were just acceptable, but with such a second camera I would want to use jpeg output most of the time and not spend too much time on PP. Apart from that the fixed LCD (no EVF or tilt screen) proved a serious limitation for me.
 
Here's btw imaging resource corner test with S110 (I do have an S100 as well):

http://www.imaging-resource.com/PRODS/canon-s110/canon-s110A4.HTM

Not that great either but again no idea how to quantify it yet until we see better test:

ZS110hVFAWB-250UL.JPG


S110 F2 upper left corner
I do most of my shooting at large subject distance (landscape, architecture, cityscapes etc.) and have found this type of studio tests VERY unreliable to judge lens performance at long distance. I have used many different digital cameras professionally and the difference between such tests and real life results are sometimes staggering. The corners of my S110 looked pretty good, better than I was expecting based on internet tests like the example above. Compared to my 450D with 15-85IS zoom (both have 12 MPixel sensor) the corners of the S110 were less sharp in the WA range, but the difference wasn't 'huge'.

One of the problems with testing is decentered lenses. The Canon S series had plenty of problems with that and maybe I was lucky to have a well-centered copy. The image corners of my camera had similar sharpness, while some review sites obviously had strongly decentered lenses; I remember both DPR and IR had such problems but not sure if that was the S100 or S110).

Btw, this type of 5x standard zooms often seem to suffer from decentering, the 15-85IS has its share of this as well while e.g. with the very cheap 18-55IS I have never seen this issue. And this problem seems to get worse when the lens is made more compact (probably due to mechanical tolerances).
 
It was S100 I believe but it was fixed later on and dpreview retested it. I do have S100 and there's some corner softness almost to full focal length (sharp otherwise) but no surprise really with very wide 24mm and compact zoom lens. Similar experience I had while ago with one of Sony 4/3 kit lenses where best resolution was visibly higher at F8 compared to F4 (arguably losing some of the superior ISO perf) so this problem is not unique to P&S camera's.

That said remember we all like to check at pixel level when new camera comes along but in real world very few really need to zoom 1:1 or crop 1:1 (with corners included) or need a huge print. Most will just look at images downsized on a 2-3MP screen or so. If you can see softness even zoomed out then that's definitely issue though.

Some studio tests can be misleading because perceived sharpness depends a lot on subject matter (the ISO resolution charts help though for good relative and maybe worse case comparison). Then there's also huge combinations of focal length and apertures and you just see 1 combination in the test. So always good to check lots of user samples but I guess nothing is certain how it will work out for you until you try it yourself.

If you look at imaging resource measures resolution you notice that there's not huge difference in resolution or dynamic range between advanced Canon P&S and G1X (assuming Mark II is similar). The biggest strength of G1X is really at high ISO. It's several stops better than an S series (not sure yet about Mark II). If you need lots of resolution I'd go for larger sensor (if you can afford it A7 has excellent resolution but obviously not as portable).
 
Last edited:
Lucien... please, forget about the s100 and RX100, this thread is about the G1X2. This issue with the G1X2 is that the lens is flawed, period. If you look at the 12 resolution charts half way down this page: http://www.monox.jp/digitalcamera-sp-canon-powershotg1xmk2-04.html

Half of them are sharp in the center and blurry at the corners, the other half are sharp at the corners and not so sharp in the center. Therefore, if the lens will not give you sharp images from corner to corner and in the center across the entire focal range and f stops, WHAT GOOD IS IT. I would not buy camera for $800 (or even $100) with a 2 year old, outdated 13 MP sensor and a lens like that. IMO, Canon really dropped the ball on this one, even worse than the original G1X. Can you not see an issue with this lens and its sharpness? Do you think it is a quality lens?

http://www.monox.jp/img/canon_powershotg1xmk2_125mm_f020.jpg

For people who were excited about the specs, starting at 24mm with f2.0, (which was/is one of the biggest selling points) it seems that its nothing to be so excited about after all. And, getting very soft edges across the focal range is also nothing to be proud of either :-|

-- Growing old is inevitable, Growing up is optional!
 
I have the G1X Mark II and I confirm that 24mm at F2 is soft/blurry.

24mm is good at F4.
Kinglau711... thank you for the update. Can you also confirm that it is equally 'good' at the corners as it is in the center of the image?

This image was shot at 24mm at F4, and as can see, it is pretty sharp in the center...

http://www.monox.jp/img/canon_powershotg1xmk2_125mm_f040.jpg

... but the corners are very blurry.

--
Growing old is inevitable, Growing up is optional!
 
Last edited:
Lucien... please, forget about the s100 and RX100, this thread is about the G1X2. This issue with the G1X2 is that the lens is flawed, period. If you look at the 12 resolution charts half way down this page: http://www.monox.jp/digitalcamera-sp-canon-powershotg1xmk2-04.html

Half of them are sharp in the center and blurry at the corners, the other half are sharp at the corners and not so sharp in the center. Therefore, if the lens will not give you sharp images from corner to corner and in the center across the entire focal range and f stops, WHAT GOOD IS IT. I would not buy camera for $800 (or even $100) with a 2 year old, outdated 13 MP sensor and a lens like that. IMO, Canon really dropped the ball on this one, even worse than the original G1X. Can you not see an issue with this lens and its sharpness? Do you think it is a quality lens?

http://www.monox.jp/img/canon_powershotg1xmk2_125mm_f020.jpg

For people who were excited about the specs, starting at 24mm with f2.0, (which was/is one of the biggest selling points) it seems that its nothing to be so excited about after all. And, getting very soft edges across the focal range is also nothing to be proud of either :-|

-- Growing old is inevitable, Growing up is optional!
It's strange that at the 24mm focal length, the center sharpens up, but the corners become softer as you go from f/2 to f/4. This effect does not happen at the other focal lengths, though.
 
Lucien... please, forget about the s100 and RX100, this thread is about the G1X2. This issue with the G1X2 is that the lens is flawed, period. If you look at the 12 resolution charts half way down this page: http://www.monox.jp/digitalcamera-sp-canon-powershotg1xmk2-04.html

Half of them are sharp in the center and blurry at the corners, the other half are sharp at the corners and not so sharp in the center.
that would suggest field curvature (which may or may not be a problem at other distance setting for outside shots) or a misaligned lens.
Therefore, if the lens will not give you sharp images from corner to corner and in the center across the entire focal range and f stops, WHAT GOOD IS IT. I would not buy camera for $800 (or even $100) with a 2 year old, outdated 13 MP sensor and a lens like that.
Agree. Just looking at the testcard images it doesn't look much better than the 5x cheaper S110 IMHO. But I don't pay much attention to such testcard shots as they are not representative for my type of photography. It does look like the ambitious bright lens is mostly for bragging rights, and not very useful for general photography (with the lens near wide open). I will be comparing images with my 450D + 15-85IS lens; if the G1X Mk II is better at f/4 than my current combo I might still buy it because of size/weight and tilt screen - if no better option comes along in the next months.
 
I have the G1X Mark II and I confirm that 24mm at F2 is soft/blurry.

24mm is good at F4.
That would negate much of the advantage of a high speed lens at that focal length..
Yes, with a DSLR lens one could say it helps to get a brighter viewfinder but that doesn't apply here (except that it probably helps for viewfinder visibility with low light shots).

My somewhat comparable 15-85IS DSLR lens is f/3.5 at the 24mm equiv. position and pretty good at f/4. Will be interesting to compare the image quality of these two at 24mm equiv. Judging from the samples I have seen the large aperture lens is only interesting for those who don't need sharp corners (or best image quality in general).
 

Keyboard shortcuts

Back
Top