K-3 high-ISO test shots — FANTASTIC!

Started 5 months ago | Discussions
ragmanjin
Regular MemberPosts: 455Gear list
Like?
K-3 high-ISO test shots — FANTASTIC!
5 months ago

Check out what Pentax's latest can really do. All DNG RAW, processed in Lightroom 5 only. One at ISO 3200, then 5000, followed by 8000 and finally a shot at 40,000:

3200: Lots of DR, great colour depth and no shortage of shadow or highlight detail. All in all it handled three light sources of varying angles, temperatures and intensities like a champ.

At ISO 5000, to me this looks like things are starting off on a good note.

8000. The auto white balance was leaning a little on the green side, but that's pretty easy to fix in post.

Again, these are just test shots, but I have to say I'm thoroughly impressed with what the K-3 can do at ISO 40,000. Sure some of the shadow detail is a little smudgy in certain spots — and there are plenty of shadows to be had in this one — but my Canon T2i (RIP) looked like this at 3200! Note the lack of noise in the well-lit areas of this photo. Just fantastic.

I've said it before and I'll say it again: Pentax is the only brand I would ever trust with auto ISO. With every other camera, it's full manual with the shutter speed as the last variable I set on the fly, but TAv on the K-3, for personal use, is a HUGE boon to me. Very impressed, even compared to the K-5IIs.

Thoughts?

-Raj

P.S. I'll be actually outdoors with the K-3 more over the next few days, so expect more tests soon now that the city's starting to thaw out after more than a week of this ≥-20ºC weather. Video reviews (and a battle of the Nifty 50s) are in the making as we speak.

-- hide signature --

< / sarcasm>

 ragmanjin's gear list:ragmanjin's gear list
Samsung Pro815 Canon PowerShot G11 Pentax K-3 Pentax smc DA 18-55mm F3.5-5.6 AL WR Pentax smc DA 35mm F2.4 AL +7 more
Canon EOS 550D (EOS Rebel T2i / EOS Kiss X4) Pentax K-3 Pentax K-5
If you believe there are incorrect tags, please send us this post using our feedback form.
ragmanjin
Regular MemberPosts: 455Gear list
Like?
Re: And ISO 40,000 compared to 400 ASA film:
In reply to ragmanjin, 5 months ago

ragmanjin wrote:

Again, these are just test shots, but I have to say I'm thoroughly impressed with what the K-3 can do at ISO 40,000. Sure some of the shadow detail is a little smudgy in certain spots — and there are plenty of shadows to be had in this one — but my Canon T2i (RIP) looked like this at 3200! Note the lack of noise in the well-lit areas of this photo. Just fantastic.

When ISO 40,000 looks better than 400-ASA film, I think that's a HUGE win.

Pentax K1000 with SMC-M 50mm f/2 lens on 400 ASA C-41 film, I'd guess near f/4.5 and 1/60" shutter speed; manual scan from the negative. Though I like the contrast, colour rendering and almost harsh style of the film, the level of detail and fine grain in the K-3 shot is simply miles above what this colour negative film can offer or attain. And again, the more detail you have in the original digital file, the more freedom you have to recreate styles like this film in post.

-Raj

-- hide signature --

< / sarcasm>

 ragmanjin's gear list:ragmanjin's gear list
Samsung Pro815 Canon PowerShot G11 Pentax K-3 Pentax smc DA 18-55mm F3.5-5.6 AL WR Pentax smc DA 35mm F2.4 AL +7 more
Reply   Reply with quote   Complain
John_A_G
Veteran MemberPosts: 6,225
Like?
Re: K-3 high-ISO test shots — FANTASTIC!
In reply to ragmanjin, 5 months ago

Raj - I think the results look great.  I am not a pixel peeper.  I prefer to judge images based upon size you would normally view/print them.  Obviously if you're used to printing large landscapes things are different.  But, it really is amazing how far sensor technology has come.

You also bring up a good point regarding the auto-ISO.  I think what gets lost in the "which camera is better" debates is concentrating on the features important to YOU.  I personally have no interest in auto-ISO.  It doesn't fit with the way I shoot much.  But you and I are two different people.  I think too many people get caught up in the notion that a camera has to be the best at EVERYTHING.  It doesn't.  It just has to be the best at what the buyer needs it to be, given financial resources.

Very nice shots!  Thanks for sharing.

Reply   Reply with quote   Complain
stanic042
Regular MemberPosts: 464Gear list
Like?
Re: K-3 high-ISO test shots — FANTASTIC!
In reply to ragmanjin, 5 months ago

all I say is..WOW

-- hide signature --

Rastislav

 stanic042's gear list:stanic042's gear list
Pentax K-5 IIs Pentax smc DA 15mm F4 ED AL Limited Pentax smc D-FA 100mm F2.8 Macro WR Pentax smc DA 35mm F2.8 Macro Limited Pentax smc DA 18-55mm F3.5-5.6 AL WR +8 more
Reply   Reply with quote   Complain
candgpics
Junior MemberPosts: 33
Like?
Re: K-3 high-ISO test shots — FANTASTIC!
In reply to ragmanjin, 5 months ago

Hi.

These high ISOs are excellent, even when viewed on a decent size monitor at 100% (or 100% of the uploaded DPREVIEW image).

I briefly had the K3. I purchased it new and it arrived clearly having been used. The white lens/camera cap was missing and the camera’s settings had been adjusted and the plastic appeared to have been handled a bit. I was a bit taken aback that the white cap was missing, but I tried the camera and thought about keeping t. I took a bit less than 100 pictures and still have them. All shot RAW. Ultimately, I returned the K3.

I have a K5 and two K5II bodies, plus several good lenses (all FA Lmtds and a few DA*).

I found the K3 to be superior (or at least sharper) at ISO 100 and 200, but at 400 noise started to creep in, though, image quality was still arguably bit better than my other Pentax bodies. By ISO 800 the image quality was about even given real world viewing/printing.

Above ISO 800 the K5/K5ii image quality was better and cleaner, had more depth to the shadows, and I preferred the rendering (the K3 I found to be a bit on the cool/blue side—but that was correctable in post).

I use Photoshop CS6 and have a copy of DXO 9.1. I used the Prime NR on a K3 image and compared it to the same image run through ACR and I preferred the ACR. Admittedly, my sampling of DXO with the K3 was limited, though, I still have the K3 images and may experiment a bit more with DXO.

In any event, though I was concerned that the K3 I received might have been returned due to poor image quality, I found overall image quality and the mid and upper level ISOs to be consistent with most of the reports I have seen online. I also found there to be more color noise in the K3 images than with the K5/K5ii. And, the better shadow detail in the K5/K5ii was noticeable, especially as the ISO went up.

Mid and upper level ISOs are important to me since I regularly shoot in these ranges (500-3200) shooting indoor sporting events and of musicians playing locally and print up to 16x20 on a regular basis (occasionally larger). I tend to post process a lot so a bit of noise does not bother me and often adds to the look I use for my images, but the K3 seemed noisier than I had hoped.

I have seen, however, a few notable individuals post very impressive high ISO images that have retained significant detail despite the high ISO and noise reduction.

Of course, the high ISO images might have been the best of the bunch, perfectly exposed, under low but even lighting, and the subject of the picture may have been just perfect. Nonetheless, I have seen several high ISO images that appear to exceed what the general consensus had seemed to be (overall improvements in handling, speed, ergonomics, and improved sharpness at lower ISOs, but inferior high ISO images and some noticeable loss of shadow detail).

Am I missing something?

Thank you in advance for your comments.

Reply   Reply with quote   Complain
lock
Veteran MemberPosts: 5,777
Like?
Thoughts.
In reply to ragmanjin, 5 months ago

Impressive. How much noise reduction did you gain from the downsampling ?

lock

Reply   Reply with quote   Complain
MightyMike
Forum ProPosts: 30,304
Like?
Re: Thoughts.
In reply to lock, 5 months ago

lock wrote:

Impressive. How much noise reduction did you gain from the downsampling ?

lock

the effective noise (ISO) reduces by the square root of the downsize ratio

you take a 6016x4000 image and reduce it to 3008x2000 and that should give you 1/2 a stop, reduce to 1504x1000 and you've got 1 stop... the reason its the square root is that takes into account the randomness of noise, if it were perfectly uniform then the reduction would be equal to the ratio, but if it were uniform we'd also have perfect NR algorithms.

-- hide signature --

Mike from Canada
"I am not a great photographer! God is a great creator! All I do is capture His creation with the tools He has provided me."
'I like to think so far outside the box that it would require a telephoto lens just to see the box!' ~ 'My Quote :)'
http://www.michaelfastphotography.com/galleries/VP-BDI_3a.jpg
http://www.airliners.net/search/photo.search?sort_order=views%20DESC&first_this_page=0&page_limit=180&&emailsearch=mighty_mike88%40hotmail.com&thumbnails=

Reply   Reply with quote   Complain
Leandros S
Contributing MemberPosts: 921Gear list
Like?
But is it hardware or software amplification for intermediate ISOs?
In reply to ragmanjin, 5 months ago

Apparently some sensors can use hardware amplification even for intermediate ISO values such as 5000 or 40000, but I don't know what sensors these are, and whether the K-3 qualifies.

-- hide signature --

No amount of perceived entitlement can replace actual expertise.

Reply   Reply with quote   Complain
ragmanjin
Regular MemberPosts: 455Gear list
Like?
Re: test shot resolution
In reply to John_A_G, 5 months ago

John_A_G wrote:

Raj - I think the results look great. I am not a pixel peeper. I prefer to judge images based upon size you would normally view/print them. Obviously if you're used to printing large landscapes things are different. But, it really is amazing how far sensor technology has come.

You also bring up a good point regarding the auto-ISO. I think what gets lost in the "which camera is better" debates is concentrating on the features important to YOU. I personally have no interest in auto-ISO. It doesn't fit with the way I shoot much. But you and I are two different people. I think too many people get caught up in the notion that a camera has to be the best at EVERYTHING. It doesn't. It just has to be the best at what the buyer needs it to be, given financial resources.

Very nice shots! Thanks for sharing.

Absolutely, John, thank you. I posted the photos at HDTV resolution, which is more than enough for a 4x6 print, since those are the two largest media a lot of K-3 owners are going to use for their high-ISO photos. While this does help the noise performance as Mike mentioned further down the thread, I think it's fair to assume people generally aren't going to be printing their ISO 40000 shots at 20x30 inches.

-Raj

-- hide signature --

< / sarcasm>

 ragmanjin's gear list:ragmanjin's gear list
Samsung Pro815 Canon PowerShot G11 Pentax K-3 Pentax smc DA 18-55mm F3.5-5.6 AL WR Pentax smc DA 35mm F2.4 AL +7 more
Reply   Reply with quote   Complain
ragmanjin
Regular MemberPosts: 455Gear list
Like?
RIGHT!? (nt)
In reply to stanic042, 5 months ago

stanic042 wrote:

all I say is..WOW

-- hide signature --

Rastislav

-- hide signature --

< / sarcasm>

 ragmanjin's gear list:ragmanjin's gear list
Samsung Pro815 Canon PowerShot G11 Pentax K-3 Pentax smc DA 18-55mm F3.5-5.6 AL WR Pentax smc DA 35mm F2.4 AL +7 more
Reply   Reply with quote   Complain
ragmanjin
Regular MemberPosts: 455Gear list
Like?
Re: K-3 high-ISO test shots — FANTASTIC!
In reply to candgpics, 5 months ago

candgpics wrote:

Hi.

These high ISOs are excellent, even when viewed on a decent size monitor at 100% (or 100% of the uploaded DPREVIEW image).

I briefly had the K3. I purchased it new and it arrived clearly having been used. The white lens/camera cap was missing and the camera’s settings had been adjusted and the plastic appeared to have been handled a bit. I was a bit taken aback that the white cap was missing, but I tried the camera and thought about keeping t. I took a bit less than 100 pictures and still have them. All shot RAW. Ultimately, I returned the K3.

I have a K5 and two K5II bodies, plus several good lenses (all FA Lmtds and a few DA*).

I found the K3 to be superior (or at least sharper) at ISO 100 and 200, but at 400 noise started to creep in, though, image quality was still arguably bit better than my other Pentax bodies. By ISO 800 the image quality was about even given real world viewing/printing.

Above ISO 800 the K5/K5ii image quality was better and cleaner, had more depth to the shadows, and I preferred the rendering (the K3 I found to be a bit on the cool/blue side—but that was correctable in post).

I use Photoshop CS6 and have a copy of DXO 9.1. I used the Prime NR on a K3 image and compared it to the same image run through ACR and I preferred the ACR. Admittedly, my sampling of DXO with the K3 was limited, though, I still have the K3 images and may experiment a bit more with DXO.

In any event, though I was concerned that the K3 I received might have been returned due to poor image quality, I found overall image quality and the mid and upper level ISOs to be consistent with most of the reports I have seen online. I also found there to be more color noise in the K3 images than with the K5/K5ii. And, the better shadow detail in the K5/K5ii was noticeable, especially as the ISO went up.

Mid and upper level ISOs are important to me since I regularly shoot in these ranges (500-3200) shooting indoor sporting events and of musicians playing locally and print up to 16x20 on a regular basis (occasionally larger). I tend to post process a lot so a bit of noise does not bother me and often adds to the look I use for my images, but the K3 seemed noisier than I had hoped.

I have seen, however, a few notable individuals post very impressive high ISO images that have retained significant detail despite the high ISO and noise reduction.

Of course, the high ISO images might have been the best of the bunch, perfectly exposed, under low but even lighting, and the subject of the picture may have been just perfect. Nonetheless, I have seen several high ISO images that appear to exceed what the general consensus had seemed to be (overall improvements in handling, speed, ergonomics, and improved sharpness at lower ISOs, but inferior high ISO images and some noticeable loss of shadow detail).

Am I missing something?

Thank you in advance for your comments.

At times I'm finding the images look best if you shoot a stop higher ISO than you think you need, so that you can expose as close to the right of the histogram as possible without clipping any necessary highlights. When you have to pull up shadows, that's really when things start to look pretty splotchy, but even Lightroom allows certain types of selective noise reduction via brush layers and selective colour saturation sliders and all that.

An example from my Canon days, sometimes I would have greenish or magenta clouds of noise in high-ISO photos but no green or magenta subject or content, so desaturating those two colours a little would make a huge improvement to the look of the photos without doing any real harm to anything.

Processing and NR is always a bit of a balancing act, but it's been so much quicker and simpler with the K-3 than with pretty much any other camera I've owned or worked with. The photos I've posted here basically used only the Luminance and Colour noise sliders for NR, I didn't even have to get into the brushes or tricks I would have needed with my other bodies. A large part of that is just knowing the camera, doing your own testing and finding how to expose each shot for how your camera best handles its ISO ranges.

I hope that helps, or at least partially answers your question.

-Raj

-- hide signature --

< / sarcasm>

 ragmanjin's gear list:ragmanjin's gear list
Samsung Pro815 Canon PowerShot G11 Pentax K-3 Pentax smc DA 18-55mm F3.5-5.6 AL WR Pentax smc DA 35mm F2.4 AL +7 more
Reply   Reply with quote   Complain
ozdean
Forum ProPosts: 20,258Gear list
Like?
Re: K-3 high-ISO test shots — FANTASTIC!
In reply to candgpics, 5 months ago

You are mirroring my thoughts and concerns of the K3 and 24 m sensor but these shots of Raj's are excellent.

-- hide signature --

Regards Dean - Capturing Creation
N.B. All my Images are Protected by Copyright

 ozdean's gear list:ozdean's gear list
Pentax MX-1 Pentax smc DA 15mm F4 ED AL Limited Pentax smc DA* 60-250mm F4.0 ED (IF) SDM Pentax smc DA 70mm F2.4 AL Limited Pentax smc D-FA 100mm F2.8 Macro WR +7 more
Reply   Reply with quote   Complain
ragmanjin
Regular MemberPosts: 455Gear list
Like?
Re: K-3 high-ISO test shots — FANTASTIC!
In reply to ozdean, 5 months ago

ozdean wrote:

You are mirroring my thoughts and concerns of the K3 and 24 m sensor but these shots of Raj's are excellent.

-- hide signature --

Regards Dean - Capturing Creation
N.B. All my Images are Protected by Copyright

Hey, thanks Dean.

-Raj

-- hide signature --

< / sarcasm>

 ragmanjin's gear list:ragmanjin's gear list
Samsung Pro815 Canon PowerShot G11 Pentax K-3 Pentax smc DA 18-55mm F3.5-5.6 AL WR Pentax smc DA 35mm F2.4 AL +7 more
Reply   Reply with quote   Complain
Petroglyph
Senior MemberPosts: 2,596Gear list
Like?
Re: K-3 high-ISO test shots — FANTASTIC!
In reply to ragmanjin, 5 months ago

Glad to see the good high ISO results.  And a better noise profile usually means a de-noiser, such as, Topaz can take you several stops even higher and still be able to get great results with shadow noise.

Cheers.

 Petroglyph's gear list:Petroglyph's gear list
Canon EOS 6D Pentax K-3 Pentax smc DA* 60-250mm F4.0 ED (IF) SDM Sigma 35mm F1.4 DG HSM Canon EF 16-35mm f/4L IS USM
Reply   Reply with quote   Complain
ragmanjin
Regular MemberPosts: 455Gear list
Like?
Re: K-3 high-ISO test shots — FANTASTIC!
In reply to Petroglyph, 5 months ago

Petroglyph wrote:

Glad to see the good high ISO results. And a better noise profile usually means a de-noiser, such as, Topaz can take you several stops even higher and still be able to get great results with shadow noise.

Cheers.

I'm sure I would have been able to get even better results had I used actual noise reduction software, but so far I've been content enough with what I can squeeze out of the files in Lightroom alone

-Raj

-- hide signature --

< / sarcasm>

 ragmanjin's gear list:ragmanjin's gear list
Samsung Pro815 Canon PowerShot G11 Pentax K-3 Pentax smc DA 18-55mm F3.5-5.6 AL WR Pentax smc DA 35mm F2.4 AL +7 more
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads