[Rondec]

What is the point? I can post iso 1600 photos that look fine at web sizes with a K5 or K3. It is clear that the 6D is better by a stop with regard to high iso -- iso 1600 on K3 should be about iso 3200 on a 6D. It does drive me crazy when people take random photos from different cameras and at the least imply that iso 400 on a K3 is the equivalent of iso 1600 on a 6D. These random photos just tell me that there are a wide range of settings on all of these cameras that allow for decent images without too intrusive noise at web sizes. But we already know that, don't we?

This is iso 6400 on K5 II



But I don't think this says too much except that with a decently exposed photo, not printed too big iso 6400 is OK on APS-C.

[rullrich]

What is the point? I can post iso 1600 photos that look fine at web sizes with a K5 or K3. It is clear that the 6D is better by a stop with regard to high iso -- iso 1600 on K3 should be about iso 3200 on a 6D. It does drive me crazy when people take random photos from different cameras and at the least imply that iso 400 on a K3 is the equivalent of iso 1600 on a 6D. These random photos just tell me that there are a wide range of settings on all of these cameras that allow for decent images without too intrusive noise at web sizes. But we already know that, don't we?

This is iso 6400 on K5 II



But I don't think this says too much except that with a decently exposed photo, not printed too big iso 6400 is OK on APS-C.

Is this picture denoised or out of camera?

[Rondec]

Is this picture denoised or out of camera?

RAW developed with default settings in Lightroom. I think Lightroom has some default color noise reduction and I didn't change that, but didn't run it through any other program for noise reduction.

[Dan Rentea]

Whatever the advantage is I'm not seeing it. The K-3 image is a better image, but that's probably because of the difference in ISO.

If you look better you will see that Pentax image has some noise (it's more visible if I zoom the image). Don't forget that on Canon 6D is not ISO1600, is ISO2500.

SO, while this example shows an advantage for a K-3 image, at least for me, there are so many extenuating concerns, that it's hard to say we are comparing the cameras. We are evaluating technique cameras, focus and ISO all at the same time. It's hard to learn a thing from that. And it's quite possible that someone's technique favours one camera over another.

Look at the EXIF on both:

Canon 6D: ISO 2500, f5, 1/60, 200mm
Pentax K-3 II: ISO800, f3.2, 1/125, 50mm

It's not about technique, focus and ISO at the same time, at least not in those 2 images. It's all about what image looks better. If you guys see that Pentax Image look better, ok than. Case closed!

If I zoom both images even with 20% you might see it properly (or different) which image looks better. Those are images scaled down to 640 x 430 pixels.

---------- Post added 11-06-15 at 03:23 PM ----------

FWIW: On my monitor, the K-3 II image appears to be the best (OTH... I've always had a thing for redheads - even fake ones... to the point a real one actually married me .)

Cheers... M

On the image with the redhead I agree, but I used a flash and ISO 400 for that image. But Canon at iso 2500 and no flash came quite close in terms of image quality.

---

Last edited by Dan Rentea; 11-06-2015 at 08:45 AM.

[Ratcheteer]

Again, on my hi-rez calibrated monitor, the K-3II image appears more natural and more attractive

so let me get this right, you have calibrated your monitor? how and what colour space cause to be honest the canon images look fine, and the first pentax shot looks over cranked (most likely due to the flash and white balance making some colours pop and others subdued) also noise is showing in the dark area of the photo taken at iso 800

[Nicolas06]

I prefer the Canon image, but this is because I like the subject more.

Is this picture denoised or out of camera?

Let's be honest, if you use a proper denoising software, you gain easily 2-3EV, so much more than the APSC/FF difference. As we advence in time, the jpeg camera get better and better so the difference is more arround 1EV,

If you care of high iso performance and don't make denoising part of your workflow you have a problem. I would expect at least that you'd let standard LR/DxO do the job. Theses softwares are less expensive than getting a better camera adn they don't add weight to the bag. And if you are really into it and use say DxO Prime, you'll get one more EV in addition!

[biz-engineer]

Again, on my hi-rez calibrated monitor, the K-3II image appears more natural and more attractive ( ).

I did find an image from the same event if you want to see some differences.

Look here a iso 800 shoot from the K3... It doesn't look noisy and doesn't use a flash:

Generally Canon colors are a bit more washed out, less contrast than Pentax. Canon tends to expose to the right, Pentax exposure is more conservative (although the K-3 deliver more optimized exposure vs previous Pentax cameras). In previous photos posted, the white balances are not the same. You show pictures at a size of 800x600 or so, at that enlargement size a point and shot if sufficient. From features and user interface, Canon cameras are not necessarily better than equivalent Pentax cameras, but Canon don't seem to mind about sensor and details as much as Pentax, however, the strength of Canon (and Nikon) is more in their broad portfolio of products, commercial presence, after sale services and availability of second hand equipment.

Regarding the comparison of image quality between a K-3 and 6D, basic photography says that you can boil camera settings down to a triangle: ISO, Aperture, Shutter speed and place your camera settings in somewhere inside this triangle depending on the available light, whatever the format. Use the same settings to compare different formats. If you move inside the triangle to demonstrate your point, chances are that you're are going to run in circle while biting your tail. You are trapped in the triangle, you have no chance to get out of this triangle, except by multi-shot techniques, flash etc, but then you don't compare the same things.

---

Last edited by biz-engineer; 11-06-2015 at 01:29 PM.

[Dan Rentea]

Generally Canon colors are a bit more washed out, less contrast than Pentax. Canon tends to expose to the right, Pentax exposure is more conservative (although the K-3 deliver more optimized exposure vs previous Pentax cameras). In previous photos posted, the white balances are not the same.

Totally agree with you. Between Pentax, Canon and Nikon I will say that colors from Pentax are more balanced than Canon or Nikon. Colors from Canon are a little washed out and colors from Nikon are more contrasty than both Pentax and Canon.

From features and user interface, Canon cameras are not necessarily better than equivalent Pentax cameras, but Canon don't seem to mind about sensor and details as much as Pentax, however, the strength of Canon (and Nikon) is more in their broad portfolio of products, commercial presence, after sale services and availability of second hand equipment.

Again, totally agree with you. I can say that K-3 II image quality is better than 7D Mark II at any ISO, but the Af from Canon 7D Mark II is mind blowing fast. Even when compare the image quality from K-3 II with Nikon D7200 I can say that up to ISO 800 they are on par (colors from K-3 II are more balanced), after ISO 800 Nikon starts to have a little advantage. But keep in mind that those thoughts are based strictly on the tests I've done with those cameras. So it's just my opinion.

Regarding the comparison of image quality between a K-3 and 6D, basic photography says that you can boil camera settings down to a triangle: ISO, Aperture, Shutter speed and place your camera settings in somewhere inside this triangle depending on the available light, whatever the format. Use the same settings to compare different formats. If you move inside the triangle to demonstrate your point, chances are that you're are going to run in circle while biting your tail. You are trapped in the triangle, you have no chance to get out of this triangle, except by multi-shot techniques, flash etc, but then you don't compare the same things.

I did compared at the same event 2 images with the same settings and I think that you can imagine the result. I don't even bother to look for those images because the difference in image quality is quite big after ISO 800.

If I wanted to be a Pentax forum trouble maker, I'd start a thread "Pentax K-3 stacking and stitching better than 645z" and then everyone would argue that the 645z is better anyway (the opposite kind of argument of that thread, just because the 645z is from Ricoh and not from Canon). [... actually, I know that with stitching I can beat the IQ of a 645z.... but shutttt, don't tell that to anyone or you gonna piss-off people who spend a massive amount of money into a 645z] The funny thing about die hard Pentaxian is they are able to tell you things like "cows can fly". I'd rather prefer thread when people explain in specific conditions how to work around some limitations to get the results that you desire, instead of defending fallacies.

I know that people tend to be more subjective when someone else say that another camera from another brand can deliver better image quality than Pentax does (especially that we are on Pentax forum), but who wants to see the difference between 2 cameras MUST test both cameras. I personally don't argue with someone else based on what I read on the internet. DXO or DPreview can't decide for me what camera is better. For example A7R II has on paper more af points than 3 Canon 1Dx combined, but is not even close to 1Dx or Nikon D4s or 7D Mark II when comes to tracking subjects.

[Rondec]

Even with pixel shift, Imaging Resource found K3 trailed D810 and 645z. However, it was clearly better than the 6D and D750/D610 straight up.

I do think it is odd that high iso comes up so frequently in this thread, since few good landscape photos will be shot at high iso. I shoot landscape a lot and I would far rather have a good iso 25 than a relatively noiseless iso 6400 on a camera.

[Dan Rentea]

Even with pixel shift, Imaging Resource found K3 trailed D810 and 645z. However, it was clearly better than the 6D and D750/D610 straight up.

I do think it is odd that high iso comes up so frequently in this thread, since few good landscape photos will be shot at high iso. I shoot landscape a lot and I would far rather have a good iso 25 than a relatively noiseless iso 6400 on a camera.

I don't have ISO 25 on Canon 6D, but I have ISO 50.

Agree with you, pixel shift images looks impressive. But when you go home and see a lot of artefacts that you can not see in the first place on your camera LCD...it's annoying.

I don't do landscape, but I did played with pixel shift outdoor on different days and the results were horrible in 99 photos out of 100. If you resize an image for Facebook you may not see the artefacts from moving leafs for example, but on a 27'' monitor, at full resolution... On the other hand, on product photography I really loved the details from pixel shift images.

[Nicolas06]

Regarding the comparison of image quality between a K-3 and 6D, basic photography says that you can boil camera settings down to a triangle: ISO, Aperture, Shutter speed and place your camera settings in somewhere inside this triangle depending on the available light, whatever the format. Use the same settings to compare different formats. If you move inside the triangle to demonstrate your point, chances are that you're are going to run in circle while biting your tail. You are trapped in the triangle, you have no chance to get out of this triangle, except by multi-shot techniques, flash etc, but then you don't compare the same things.

This is not really true as typically from an artistic point of view, you are after some framing, some type of perspective distorsion, some deph of field control. This imply that to get the same results, you'll not choose the same focal length neither the same apperture. Basically the FF need to be closed one step down more and use a longer focal length to keep same framing and perspective compression/distortion. The added high iso performance really is lost in this closed one step down more. The benefit come into play when finally it is interresting to trade lower noise for less deph of field... Exactly as if you decide to use a wider apperture on the APSC camera. If this is possible on the APSC camera, then you didn't gain any low light gathering with your FF.

It is only if basically you are already at the widest apperture available for the APSC lense you use and the loss of deph of field is acceptable and that the corresponding FF lens you would use can support at least the same apperture you'll gain something.


So basically if you are thinking to buy for your FF a 70-200 f/4, a 100-300 f/4 or a 150-450 f/4-5.6 or even a 300mm f/4 as well as a 135mm f/2, the corresponding matching lenses in framing are available in APSC with a faster apperture so you don't get more low light capabilities, even by shooting wide open with the FF. What you really gain if that the lens FF lens perform better wide open than the matching APSC lens, even if that may no be fully visible at iso 3200...

But of course if you are investing in a 70-200 f/2.8, 100-300 f/2.8, 200mm f/2 or a 50mm f/1.4 then there no fast enough APSC lens available that provide you the same framing and you gain an edge.

This is of course at the expense of the lens price difference (that is sometime huge), size and weight... and also of course deph of field. An APSC shooter will have to spend quite some money to get a 24mm f/1.4 lens to try to match a 35mm f/1.4 on FF, but I'am not sure the price would be much different, except the FF keep the advantage for low light and dof control. To match 50mm f/1.4 you go 30mm f/1.4 and get less dof control and less low light capabilities. Price wise the 30mm f/1.4 is a bit more expensive that true. Usually still the matching APSC lens (even if that happen to be an FF compatible one) will be same price or less expensive as well as smaller and lighter.

[biz-engineer]

This is not really true as typically from an artistic point of view, you are after some framing, some type of perspective distorsion, some deph of field control. This imply that to get the same results, you'll not choose the same focal length neither the same apperture. Basically the FF need to be closed one step down more and use a longer focal length to keep same framing and perspective compression/distortion. The added high iso performance really is lost in this closed one step down more. The benefit come into play when finally it is interresting to trade lower noise for less deph of field... Exactly as if you decide to use a wider apperture on the APSC camera. If this is possible on the APSC camera, then you didn't gain any low light gathering with your FF.

So basically if you are thinking to buy for your FF a 70-200 f/4, a 100-300 f/4 or a 150-450 f/4-5.6 or even a 300mm f/4 as well as a 135mm f/2, the corresponding matching lenses in framing are available in APSC with a faster apperture so you don't get more low light capabilities, even by shooting wide open with the FF. What you really gain if that the lens FF lens perform better wide open than the matching APSC lens, even if that may no be fully visible at iso 3200...

That's why most 645 lenses are about f4 / f4.5. Do you realize that these people buying a 645z are spending loads of money for nothing since they would get the same results with a shorter f2 lens mounted on a K-S2 for ten times cheaper. I agree, on a FF camera, lenses are slower than APSC lenses, for instance the Tamron 70-200 f2.8 lens is one stop slower than the DA*50-135 f2 (ouuppps it's also f2.8 ?), there's also a FA50 f1.4 for full frame, and you can easily get the cheapo DAL35 f1.0 for APSC (ouupppss it's actually f2.4 ?). I like to agree that FF lenses are one stop slower than APSC lenses (especially in France ;-)). I agree, for beginner, the DoF of APSC makes it easier to use; with a full frame you have to know what you are doing, and with a 645z , that's even more true. On my Canon PS A710IS, I did not have any issue with DoF, everything was in focus, and I can confirm that after I purchased a K200D, the shallower DoF of the K200D really annoyed me, I especially did not like the blurred backgrounds :-). I agree, you could sell your K-3 + lenses and get yourself a Panasonic DMC-FZ1000 with the Leica 25-400 to cover to cover all your needs and have no DoF problem. I still can't understand why Pentax did not put a u4/3 sensor in their K mount camera bodies and offer aset of f0.7 lenses in order to provide the same DoF and image quality as one would get with a large APSC sensor. I'm aware that the iPhone use an f0.5 lens in order to get its DoF and SNR the same as a K-3.

[Nicolas06]

That's why most 645 lenses are about f4 / f4.5. Do you realize that these people buying a 645z are spending loads of money for nothing since they would get the same results with a shorter f2 lens mounted on a K-S2 for ten times cheaper. I agree, on a FF camera, lenses are slower than APSC lenses, for instance the Tamron 70-200 f2.8 lens is one stop slower than the DA*50-135 f2 (ouuppps it's also f2.8 ?), there's also a FA50 f1.4 for full frame, and you can easily get the cheapo DAL35 f1.0 for APSC (ouupppss it's actually f2.4 ?). I like to agree that FF lenses are one stop slower than APSC lenses (especially in France ;-)). I agree, for beginner, the DoF of APSC makes it easier to use; with a full frame you have to know what you are doing, and with a 645z , that's even more true. On my Canon PS A710IS, I did not have any issue with DoF, everything was in focus, and I can confirm that after I purchased a K200D, the shallower DoF of the K200D really annoyed me, I especially did not like the blurred backgrounds :-). I agree, you could sell your K-3 + lenses and get yourself a Panasonic DMC-FZ1000 with the Leica 25-400 to cover to cover all your needs and have no DoF problem. I still can't understand why Pentax did not put a u4/3 sensor in their K mount camera bodies and offer aset of f0.7 lenses in order to provide the same DoF and image quality as one would get with a large APSC sensor. I'm aware that the iPhone use an f0.5 lens in order to get its DoF and SNR the same as a K-3.

To me you are in the right when you say that most 645 lenses are slow, this show that bigger sensor are not necessarily about low light performance and less dof. Because no a 645Z has no thinner dof or better low light performance than an FF or APSC. Sure the sensor is better but there no lenses that match. But people still buy it, even through it is more expensive.

Oh maybe that's because they guy can't master what true great photographer can do with an f/1.4 lense on an FF and they need to hide that with very expensive equipement while they should be better with a bridge.

That's your hypothesis...

Doesn't sound really credible to me.

Photography is not about ultimate low light performance. It is not about razor thin deph of field.

[biz-engineer]

That's your hypothesis...

That your hypothesis. You sort of mentioned that FF @ f4 = APSC @ f2.8. So, a FF @ f2.8 is the same as a 645z @ f4. But the FF is a lot cheaper. I follow you and extrapolate across formats. But now you seem to say, no it's not the same. I'm confused.

[Rondec]

Is the whole goal of photography narrow depth of field and clear high iso? Then what is the point. Clearly full frame is better in both of those situations, but those aren't landscape situations. The reason why the 645z does well is because it is a beast in low iso situations and the glass for it is pretty amazing with regard to its rendering.

There are plenty of times that I would like a little more dynamic range, but a 6D was never the camera that came to mind in those situations as my K5 II has more low iso dynamic range than a 6D. A D810 sensor on the other hand...