The "ISOless" sensor

dosdan · 03-19-2011, 08:49 PM

Originally posted by Smeggypants

Have you tried shooting ISOLess already? if not? why not?

No, I've got a K20D & a K100D Super, and money with 4 kids, is tight. Eventually, I plan to get a K-5 or its replacement.

I'll start shooting young soccer again next month and the focus, burst speed, high-ISO noise & DR issues (particularly when shooting with the sun on the other side of the players - I can't always relocate - and then dodging the shadowed faces in PP) I've been facing for the last 2 years will resurface. This may well force me to upgrade.

Dan.

Last edited by dosdan; 03-19-2011 at 08:56 PM.

les3547 · 03-20-2011, 10:44 AM

I wonder if someone would mind converting the concepts discussed in this thread to a set of exposure principles for the K-5? For those not totally familiar with all the underlying concepts, it can be a little bit tricky extracting practical methods from theory.

dosdan · 03-20-2011, 02:51 PM

For you K-5 owners, how about a simple test?

Looking at the picture in Noise and DR comparision: Canon 5D vs 5D2 vs 7D vs Pentax K5 where a shot of a wide DR scene was taken with a K-5 at ISO100 and then 4 stops EV boost applied in PP, compared to the equivalent exposure of shooting at ISO1600, you can see how the highlights are not blown in the desk lamp when using the superior DR of base ISO. (DR decreases with increasing ISO, as already discussed.)

Consider a tricky exposure situation like blowing out candles in a dark room, where you're trying to get decent illumination on a person's face and the cake and not have the candles completely overexposed. It would be interesting to compare the K-5's matrix metering performance (if using Auto-ISO, set the max. ISO to ISO1600, since above that digital "amplification"is used), against shooting in raw using the same shutter speed & aperture at ISO80 in Manual exposure mode and boosting in PP to see if you can get a better capture. (If you've got any candles in the house, this test is easy to stage.)

The other consideration in a candle-lit scene is the WB. It will normally be very reddish from the colour temperature of the candles. Do you leave in that way or, if you're shooting raw, do you adjust the WB afterwards to fix up the skin colour, or do you set it part-way between these two settings?

Dan.

dosdan · 03-20-2011, 03:15 PM

Originally posted by les3547

I wonder if someone would mind converting the concepts discussed in this thread to a set of exposure principles for the K-5?

I'm not a K-5 owner so I don't know how well the matrix metering performs, but in general terms with this camera sensor's characteristics, if it's a difficult lighting situation and you're shooting raw, it seems to be a better strategy to use some negative EV comp (-0.3 to -1 EV, or even more if operating at base ISO), than to ETTR and risk blowing a highlight.

Dan.

Last edited by dosdan; 03-20-2011 at 03:26 PM.

les3547 · 03-20-2011, 03:43 PM

Originally posted by dosdan

I'm not a K-5 owner so I don't know how well the matrix metering performs, but in general terms with this camera sensor's characteristics, if it's a difficult lighting situation and you're shooting raw, it seems to be a better strategy to use some negative EV comp (-0.3 to -1 EV, or even more if operating at base ISO), than to ETTR and risk blowing a highlight.

Dan.

Thanks. Fortunately, that's how I shoot all the time anyway (with a K-7).

Smeggypants · 03-20-2011, 08:21 PM

Originally posted by dosdan

For you K-5 owners, how about a simple test?

Looking at the picture in Noise and DR comparision: Canon 5D vs 5D2 vs 7D vs Pentax K5 where a shot of a wide DR scene was taken with a K-5 at ISO100 and then 4 stops EV boost applied in PP, compared to the equivalent exposure of shooting at ISO1600, you can see how the highlights are not blown in the desk lamp when using the superior DR of base ISO. (DR decreases with increasing ISO, as already discussed.)

The lower lights are better at 1600. I'm not bothered abut the slightly blown lamp section at 1600. the quality of the rest pof the scene is more important. If I really wanted to capture the detail of the filament of the lamp I would under expose accordingly.

Don't get me wrong Dan, I understand your argument and if the technology allowed a much more detailed image capture, say 24bit, against the current 14bit of the K-5 I would be a lot more supportive of the ISOless concept.

Class A · 03-21-2011, 04:28 AM

Originally posted by Smeggypants

The lower lights are better at 1600.

What you are seeing as "better" might be just in-camera noise reduction. Pentax is known to apply noise reduction on RAW files for higher ISO shots. Maybe not, IIRC, RAW NR only sets in after ISO 1600 with the K-5.

BTW, have you asked the creator of the image whether you are allowed to re-post it here?

Last edited by Class A; 03-21-2011 at 04:57 AM.

dosdan · 03-22-2011, 03:59 AM

Since I haven't got a K-5 yet, I have to reply on others' shots. Here is an interesting 4-stops-plus boost from ISO80, from bazz (I've received his permission to post the links). His comments:

Testing out my new Opteka mini softbox which arrived in todays mail and fired off a shot before the flash had time to fully recharge. Straight out of the camera: Exposure Failure

And after some Photoshop ACR adjustments, (+4 EV and +120 Brightness).
And then a crop, NR, levels and USM in PS CS5: Result after PP

Dan.

les3547 · 03-22-2011, 08:58 AM

Originally posted by dosdan

Since I haven't got a K-5 yet, I have to reply on others' shots. Here is an interesting 4-stops-plus boost from ISO80, from bazz (I've received his permission to post the links). His comments:

Testing out my new Opteka mini softbox which arrived in todays mail and fired off a shot before the flash had time to fully recharge. Straight out of the camera: Exposure Failure

And after some Photoshop ACR adjustments, (+4 EV and +120 Brightness).
And then a crop, NR, levels and USM in PS CS5: Result after PP

Dan.

Dan, this isn't addressed specifically to you, but to anyone who can answer.

I bought my first serious DSLR last summer (the K-7), and then shortly after the K-5 came out. I was a little upset by that, I wish I'd known it was coming. I mention this so it's clear that I am questioning the advantages of the K-5 over the K-7 (since I am considering whether to invest in the K-5, or wait for the next Pentax and buy other equipment I need now).

I am specifically interested in the K-5's advantage at ISO 100, which is the only setting I use. So I naturally want to know how the K-7 sensor would have performed in the same circumstances above . . . i.e., would one be able to similarly recover that bug if the K-7 had been used?

Smeggypants · 03-22-2011, 10:00 AM

Originally posted by les3547

Dan, this isn't addressed specifically to you, but to anyone who can answer.

I bought my first serious DSLR last summer (the K-7), and then shortly after the K-5 came out. I was a little upset by that, I wish I'd known it was coming. I mention this so it's clear that I am questioning the advantages of the K-5 over the K-7 (since I am considering whether to invest in the K-5, or wait for the next Pentax and buy other equipment I need now).

I am specifically interested in the K-5's advantage at ISO 100, which is the only setting I use. So I naturally want to know how the K-7 sensor would have performed in the same circumstances above . . . i.e., would one be able to similarly recover that bug if the K-7 had been used?

I would doubt it as the K-7 has less dynamic range. You can try it. use manual mode and under expose by 4 stops and see how much can boost and compare the noise levels.

Smeggypants · 03-22-2011, 10:05 AM

Originally posted by dosdan

Since I haven't got a K-5 yet, I have to reply on others' shots. Here is an interesting 4-stops-plus boost from ISO80, from bazz (I've received his permission to post the links). His comments:

Testing out my new Opteka mini softbox which arrived in todays mail and fired off a shot before the flash had time to fully recharge. Straight out of the camera: Exposure Failure

And after some Photoshop ACR adjustments, (+4 EV and +120 Brightness).
And then a crop, NR, levels and USM in PS CS5: Result after PP

Dan.

Yup this is will known on the K-5. I've managed several recoveries like that. even at 5 stops or more. Amazing really compared to previous generation DSLRs. You'd end up with a mass of oddcoloured noise doing that on a K10/K20/K7 under exposure

The caveat is that once you get to a properly exposed pic from the original almost dark frame you've lost much footroom for post pro.

This is why disregarding ISO ( sensor gain ) simply shooting at base ISO and boosting later is not a good idea.

Class A · 03-22-2011, 06:03 PM

Originally posted by Smeggypants

This is why disregarding ISO ( sensor gain ) simply shooting at base ISO and boosting later is not a good idea.

You are confounding shooting at base ISO with reducing the dynamic range. Of course, if you choose shutter speed speed and aperture settings for a ISO 12800 shot with a conservative exposure (i.e., staying several stops under the potential exposure that still doesn't clip highlights) but shoot at ISO 100 then chances are that you will reduce the DR by clipping many shadow details.

However, shooting at ISO 100 with appropriate shutter speed and aperture settings so that you don't crunch the DR has little disadvantages with the K-5. There are small differences up to ISO 1600 but from then on the camera uses the same ("base" ISO) strategy anyhow, i.e., underexpose and push in post.

ChrisJ · 03-22-2011, 07:40 PM

I shoot bracketed shots all the time for HDR, so I thought I'd try this by using the Exposure slider in ACR on the 2 stops under exposed image (all images were shot at 200 ISO and a constant aperture, only the shutter speed altered on my new K5), when I did noise appeared and lots of it.

Just for variety I took the same shots and altered the exposure by altering the ISO leaving shutter speed and aperture constant, when I opened up the under exposed image in ACR little if any noise.

From this simple test it appears that altering ISO does keep noise down on a K5, at least up to 1600. Shooting at base ISO then opening the exposure for me seems to increases noise.

Now I'm baffled. I always understood that there was an 'optimum' ISO for any sensor and that has certainly been true for me until now. My Oly E-P1 for instance 'optomises' at an ISO of 360, in other words it is noticably less noisy at 360 ISO than it is at 100 ISO.

I understand that the Sony sensor on the K5 is a paradime shift in sensor technology, my result so far bear that out, but shooting at base doesn't seem to work, may be shooting at 'optimum' might. Further tests follow.

Chris

Smeggypants · 03-22-2011, 09:28 PM

Originally posted by Class A

You are confounding shooting at base ISO with reducing the dynamic range. Of course, if you choose shutter speed speed and aperture settings for a ISO 12800 shot with a conservative exposure (i.e., staying several stops under the potential exposure that still doesn't clip highlights) but shoot at ISO 100 then chances are that you will reduce the DR by clipping many shadow details.

However, shooting at ISO 100 with appropriate shutter speed and aperture settings so that you don't crunch the DR has little disadvantages with the K-5. There are small differences up to ISO 1600 but from then on the camera uses the same ("base" ISO) strategy anyhow, i.e., underexpose and push in post.

I work on what I find in practise

Quote:

There are small differences up to ISO 1600

100 -1600, i.e 5 stops, is NOT a small difference. Like I say you can certainly recover a 5 stop underexposure with the K-5. I've done it. but the IMPORTANT issue to note is that you've lost all your leeway for post pro.

There is NO confounding. Shooting at base ISO does reduce dynamic range when it causes underexposure for the chosen aperture/shutter settings. 5 stops is 30db dynamic range lost. Which make a big difference in PP.

Further ... you can't 'clip' shadow details. Clipping is when the signal exceeds the maximum level allowable by the analog/digital signal chain. Shsdaw details are lost by nit enough bits to digitise these low levels ( or by noise masking )

Quote:

However, shooting at ISO 100 with appropriate shutter speed and aperture settings so that you don't crunch the DR

This is irrelevant to the discussion. if shooting at ISO100 is appropriate to get proper expose with chosen shutter and aperture settings then that is normal use and jsu part of traditional photography.

Smeggypants · 03-22-2011, 09:29 PM

Originally posted by ChrisJ

I shoot bracketed shots all the time for HDR, so I thought I'd try this by using the Exposure slider in ACR on the 2 stops under exposed image (all images were shot at 200 ISO and a constant aperture, only the shutter speed altered on my new K5), when I did noise appeared and lots of it.

Just for variety I took the same shots and altered the exposure by altering the ISO leaving shutter speed and aperture constant, when I opened up the under exposed image in ACR little if any noise.

From this simple test it appears that altering ISO does keep noise down on a K5, at least up to 1600. Shooting at base ISO then opening the exposure for me seems to increases noise.

Now I'm baffled. I always understood that there was an 'optimum' ISO for any sensor and that has certainly been true for me until now. My Oly E-P1 for instance 'optomises' at an ISO of 360, in other words it is noticably less noisy at 360 ISO than it is at 100 ISO.

I understand that the Sony sensor on the K5 is a paradime shift in sensor technology, my result so far bear that out, but shooting at base doesn't seem to work, may be shooting at 'optimum' might. Further tests follow.

Chris

Yes exactly. We're not at a point where an ISOless DSLR is sensible, despite the claims of the theorists.

03-20-2011, 02:51 PM	#63
dosdan Pentaxian Join Date: Dec 2007 Location: Brisbane, Australia Photos: Gallery \| Albums Posts: 1,741 Original Poster	For you K-5 owners, how about a simple test? Looking at the picture in Noise and DR comparision: Canon 5D vs 5D2 vs 7D vs Pentax K5 where a shot of a wide DR scene was taken with a K-5 at ISO100 and then 4 stops EV boost applied in PP, compared to the equivalent exposure of shooting at ISO1600, you can see how the highlights are not blown in the desk lamp when using the superior DR of base ISO. (DR decreases with increasing ISO, as already discussed.) Consider a tricky exposure situation like blowing out candles in a dark room, where you're trying to get decent illumination on a person's face and the cake and not have the candles completely overexposed. It would be interesting to compare the K-5's matrix metering performance (if using Auto-ISO, set the max. ISO to ISO1600, since above that digital "amplification"is used), against shooting in raw using the same shutter speed & aperture at ISO80 in Manual exposure mode and boosting in PP to see if you can get a better capture. (If you've got any candles in the house, this test is easy to stage.) The other consideration in a candle-lit scene is the WB. It will normally be very reddish from the colour temperature of the candles. Do you leave in that way or, if you're shooting raw, do you adjust the WB afterwards to fix up the skin colour, or do you set it part-way between these two settings? Dan. Last edited by dosdan; 03-20-2011 at 08:10 PM.

03-22-2011, 03:59 AM	#68
dosdan Pentaxian Join Date: Dec 2007 Location: Brisbane, Australia Photos: Gallery \| Albums Posts: 1,741 Original Poster	Since I haven't got a K-5 yet, I have to reply on others' shots. Here is an interesting 4-stops-plus boost from ISO80, from bazz (I've received his permission to post the links). His comments: *Testing out my new Opteka mini softbox which arrived in todays mail and fired off a shot before the flash had time to fully recharge.* Straight out of the camera: Exposure Failure And after some Photoshop ACR adjustments, (+4 EV and +120 Brightness). And then a crop, NR, levels and USM in PS CS5: Result after PP Dan. Last edited by dosdan; 03-22-2011 at 04:05 AM.

03-22-2011, 07:40 PM	#73
ChrisJ Veteran Member Join Date: Nov 2008 Location: Eckington, Derbyshire UK Photos: Gallery \| Albums Posts: 316	I shoot bracketed shots all the time for HDR, so I thought I'd try this by using the Exposure slider in ACR on the 2 stops under exposed image (all images were shot at 200 ISO and a constant aperture, only the shutter speed altered on my new K5), when I did noise appeared and lots of it. Just for variety I took the same shots and altered the exposure by altering the ISO leaving shutter speed and aperture constant, when I opened up the under exposed image in ACR little if any noise. From this simple test it appears that altering ISO does keep noise down on a K5, at least up to 1600. Shooting at base ISO then opening the exposure for me seems to increases noise. Now I'm baffled. I always understood that there was an 'optimum' ISO for any sensor and that has certainly been true for me until now. My Oly E-P1 for instance 'optomises' at an ISO of 360, in other words it is noticably less noisy at 360 ISO than it is at 100 ISO. I understand that the Sony sensor on the K5 is a paradime shift in sensor technology, my result so far bear that out, but shooting at base doesn't seem to work, may be shooting at 'optimum' might. Further tests follow. Chris Last edited by ChrisJ; 03-22-2011 at 07:42 PM. Reason: spelling

03-22-2011, 09:28 PM	#74
Smeggypants Veteran Member Join Date: Sep 2010 Posts: 1,536	Originally posted by Class A You are confounding shooting at base ISO with reducing the dynamic range. Of course, if you choose shutter speed speed and aperture settings for a ISO 12800 shot with a conservative exposure (i.e., staying several stops under the potential exposure that still doesn't clip highlights) but shoot at ISO 100 then chances are that you will reduce the DR by clipping many shadow details. However, shooting at ISO 100 with appropriate shutter speed and aperture settings so that you don't crunch the DR has little disadvantages with the K-5. There are small differences up to ISO 1600 but from then on the camera uses the same ("base" ISO) strategy anyhow, i.e., underexpose and push in post. I work on what I find in practise Quote: There are small differences up to ISO 1600 100 -1600, i.e 5 stops, is NOT a small difference. Like I say you can certainly recover a 5 stop underexposure with the K-5. I've done it. but the IMPORTANT issue to note is that you've lost all your leeway for post pro. There is NO confounding. Shooting at base ISO does reduce dynamic range when it causes underexposure for the chosen aperture/shutter settings. 5 stops is 30db dynamic range lost. Which make a big difference in PP. Further ... you can't 'clip' shadow details. Clipping is when the signal exceeds the maximum level allowable by the analog/digital signal chain. Shsdaw details are lost by nit enough bits to digitise these low levels ( or by noise masking ) Quote: However, shooting at ISO 100 with appropriate shutter speed and aperture settings so that you don't crunch the DR This is irrelevant to the discussion. if shooting at ISO100 is appropriate to get proper expose with chosen shutter and aperture settings then that is normal use and jsu part of traditional photography. Last edited by Smeggypants; 03-22-2011 at 09:39 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Post your "Park Bench" "or "Picnic Table" images	tessfully	Mini-Challenges, Games, and Photo Stories	2201	2 Days Ago 03:25 PM
Please explain "Sensor is Outresolving Some Lenses"	Parallax	Pentax K-5 & K-5 II	21	11-24-2010 05:05 PM
Sensor shift during "long" exposures	eclipsed450	Pentax DSLR Discussion	11	12-18-2009 12:51 PM
K10D "Sensor moving" possible problem?	jaitas	Troubleshooting and Beginner Help	6	05-07-2009 02:08 PM
"Green Clean" Sensor Cleaning System.	NLAlston	Pentax Camera and Field Accessories	2	05-16-2007 07:35 PM

03-19-2011, 08:49 PM	#61
dosdan Pentaxian Join Date: Dec 2007 Location: Brisbane, Australia Photos: Gallery \| Albums Posts: 1,741 Original Poster	Originally posted by Smeggypants Have you tried shooting ISOLess already? if not? why not? No, I've got a K20D & a K100D Super, and money with 4 kids, is tight. Eventually, I plan to get a K-5 or its replacement. I'll start shooting young soccer again next month and the focus, burst speed, high-ISO noise & DR issues (particularly when shooting with the sun on the other side of the players - I can't always relocate - and then dodging the shadowed faces in PP) I've been facing for the last 2 years will resurface. This may well force me to upgrade. Dan. Last edited by dosdan; 03-19-2011 at 08:56 PM.

03-20-2011, 10:44 AM	#62
les3547 Veteran Member Join Date: Jun 2010 Location: Sebastopol, California Photos: Albums Posts: 2,020	I wonder if someone would mind converting the concepts discussed in this thread to a set of exposure principles for the K-5? For those not totally familiar with all the underlying concepts, it can be a little bit tricky extracting practical methods from theory.

03-20-2011, 03:15 PM	#64
dosdan Pentaxian Join Date: Dec 2007 Location: Brisbane, Australia Photos: Gallery \| Albums Posts: 1,741 Original Poster	Originally posted by les3547 I wonder if someone would mind converting the concepts discussed in this thread to a set of exposure principles for the K-5? I'm not a K-5 owner so I don't know how well the matrix metering performs, but in general terms with this camera sensor's characteristics, if it's a difficult lighting situation and you're shooting raw, it seems to be a better strategy to use some negative EV comp (-0.3 to -1 EV, or even more if operating at base ISO), than to ETTR and risk blowing a highlight. Dan. Last edited by dosdan; 03-20-2011 at 03:26 PM.

03-20-2011, 03:43 PM	#65
les3547 Veteran Member Join Date: Jun 2010 Location: Sebastopol, California Photos: Albums Posts: 2,020	Originally posted by dosdan I'm not a K-5 owner so I don't know how well the matrix metering performs, but in general terms with this camera sensor's characteristics, if it's a difficult lighting situation and you're shooting raw, it seems to be a better strategy to use some negative EV comp (-0.3 to -1 EV, or even more if operating at base ISO), than to ETTR and risk blowing a highlight. Dan. Thanks. Fortunately, that's how I shoot all the time anyway (with a K-7).

03-20-2011, 08:21 PM	#66
Smeggypants Veteran Member Join Date: Sep 2010 Posts: 1,536	Originally posted by dosdan For you K-5 owners, how about a simple test? Looking at the picture in Noise and DR comparision: Canon 5D vs 5D2 vs 7D vs Pentax K5 where a shot of a wide DR scene was taken with a K-5 at ISO100 and then 4 stops EV boost applied in PP, compared to the equivalent exposure of shooting at ISO1600, you can see how the highlights are not blown in the desk lamp when using the superior DR of base ISO. (DR decreases with increasing ISO, as already discussed.) The lower lights are better at 1600. I'm not bothered abut the slightly blown lamp section at 1600. the quality of the rest pof the scene is more important. If I really wanted to capture the detail of the filament of the lamp I would under expose accordingly. Don't get me wrong Dan, I understand your argument and if the technology allowed a much more detailed image capture, say 24bit, against the current 14bit of the K-5 I would be a lot more supportive of the ISOless concept.

03-21-2011, 04:28 AM	#67
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Originally posted by Smeggypants The lower lights are better at 1600. What you are seeing as "better" might be just in-camera noise reduction. Pentax is known to apply noise reduction on RAW files for higher ISO shots. Maybe not, IIRC, RAW NR only sets in after ISO 1600 with the K-5. BTW, have you asked the creator of the image whether you are allowed to re-post it here? Last edited by Class A; 03-21-2011 at 04:57 AM.

03-22-2011, 08:58 AM	#69
les3547 Veteran Member Join Date: Jun 2010 Location: Sebastopol, California Photos: Albums Posts: 2,020	Originally posted by dosdan Since I haven't got a K-5 yet, I have to reply on others' shots. Here is an interesting 4-stops-plus boost from ISO80, from bazz (I've received his permission to post the links). His comments: *Testing out my new Opteka mini softbox which arrived in todays mail and fired off a shot before the flash had time to fully recharge.* Straight out of the camera: Exposure Failure And after some Photoshop ACR adjustments, (+4 EV and +120 Brightness). And then a crop, NR, levels and USM in PS CS5: Result after PP Dan. Dan, this isn't addressed specifically to you, but to anyone who can answer. I bought my first serious DSLR last summer (the K-7), and then shortly after the K-5 came out. I was a little upset by that, I wish I'd known it was coming. I mention this so it's clear that I am questioning the advantages of the K-5 over the K-7 (since I am considering whether to invest in the K-5, or wait for the next Pentax and buy other equipment I need now). I am specifically interested in the K-5's advantage at ISO 100, which is the only setting I use. So I naturally want to know how the K-7 sensor would have performed in the same circumstances above . . . i.e., would one be able to similarly recover that bug if the K-7 had been used?

03-22-2011, 10:00 AM	#70
Smeggypants Veteran Member Join Date: Sep 2010 Posts: 1,536	Originally posted by les3547 Dan, this isn't addressed specifically to you, but to anyone who can answer. I bought my first serious DSLR last summer (the K-7), and then shortly after the K-5 came out. I was a little upset by that, I wish I'd known it was coming. I mention this so it's clear that I am questioning the advantages of the K-5 over the K-7 (since I am considering whether to invest in the K-5, or wait for the next Pentax and buy other equipment I need now). I am specifically interested in the K-5's advantage at ISO 100, which is the only setting I use. So I naturally want to know how the K-7 sensor would have performed in the same circumstances above . . . i.e., would one be able to similarly recover that bug if the K-7 had been used? I would doubt it as the K-7 has less dynamic range. You can try it. use manual mode and under expose by 4 stops and see how much can boost and compare the noise levels.

03-22-2011, 10:05 AM	#71
Smeggypants Veteran Member Join Date: Sep 2010 Posts: 1,536	Originally posted by dosdan Since I haven't got a K-5 yet, I have to reply on others' shots. Here is an interesting 4-stops-plus boost from ISO80, from bazz (I've received his permission to post the links). His comments: *Testing out my new Opteka mini softbox which arrived in todays mail and fired off a shot before the flash had time to fully recharge.* Straight out of the camera: Exposure Failure And after some Photoshop ACR adjustments, (+4 EV and +120 Brightness). And then a crop, NR, levels and USM in PS CS5: Result after PP Dan. Yup this is will known on the K-5. I've managed several recoveries like that. even at 5 stops or more. Amazing really compared to previous generation DSLRs. You'd end up with a mass of oddcoloured noise doing that on a K10/K20/K7 under exposure The caveat is that once you get to a properly exposed pic from the original almost dark frame you've lost much footroom for post pro. This is why disregarding ISO ( sensor gain ) simply shooting at base ISO and boosting later is not a good idea.

03-22-2011, 06:03 PM	#72
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Originally posted by Smeggypants This is why disregarding ISO ( sensor gain ) simply shooting at base ISO and boosting later is not a good idea. You are confounding shooting at base ISO with reducing the dynamic range. Of course, if you choose shutter speed speed and aperture settings for a ISO 12800 shot with a conservative exposure (i.e., staying several stops under the potential exposure that still doesn't clip highlights) but shoot at ISO 100 then chances are that you will reduce the DR by clipping many shadow details. However, shooting at ISO 100 with appropriate shutter speed and aperture settings so that you don't crunch the DR has little disadvantages with the K-5. There are small differences up to ISO 1600 but from then on the camera uses the same ("base" ISO) strategy anyhow, i.e., underexpose and push in post.

03-22-2011, 09:29 PM	#75
Smeggypants Veteran Member Join Date: Sep 2010 Posts: 1,536	Originally posted by ChrisJ I shoot bracketed shots all the time for HDR, so I thought I'd try this by using the Exposure slider in ACR on the 2 stops under exposed image (all images were shot at 200 ISO and a constant aperture, only the shutter speed altered on my new K5), when I did noise appeared and lots of it. Just for variety I took the same shots and altered the exposure by altering the ISO leaving shutter speed and aperture constant, when I opened up the under exposed image in ACR little if any noise. From this simple test it appears that altering ISO does keep noise down on a K5, at least up to 1600. Shooting at base ISO then opening the exposure for me seems to increases noise. Now I'm baffled. I always understood that there was an 'optimum' ISO for any sensor and that has certainly been true for me until now. My Oly E-P1 for instance 'optomises' at an ISO of 360, in other words it is noticably less noisy at 360 ISO than it is at 100 ISO. I understand that the Sony sensor on the K5 is a paradime shift in sensor technology, my result so far bear that out, but shooting at base doesn't seem to work, may be shooting at 'optimum' might. Further tests follow. Chris Yes exactly. We're not at a point where an ISOless DSLR is sensible, despite the claims of the theorists.