Nikon D3200...really really good sensor (not Sony)?

Started Apr 29, 2012 | Discussions
chlamchowder
Senior MemberPosts: 1,982Gear list
Like?
Nikon D3200...really really good sensor (not Sony)?
Apr 29, 2012

I've heard rumors about the D3200 sensor being manufactured by Sony, and plenty of rumors to the contrary. Out of curiosity, I decided to check the dpreview studio comparisons. As I moved up the ISO range (in raw), the D3200 images seemed noticeably cleaner than the a65 images. In addition to having less color noise, the D3200's grain structure was much finer.

I was about to attribute the difference to the translucent mirror when I checked the EXIF data for the studio shots, and was amazed to see that the D3200 received 2/3 stops less light than the a65. In the ISO 3200 image, the D3200 was set to f/8, 1/1250s, while the a65 was set to f/8, 1/800s. That more than makes up for the SLT light loss, but the D3200 still looks better.

Then, I compared the D3200 shots to studio shots from the a580, and it was pretty much a wash. At the pixel level, the a580 might be slightly cleaner in some areas, but resizing would easily make up for the difference. The a580 also got 2/3 stops more light than the D3200, but was still unable to beat it.

On the other hand, at ISO 100, the a580 is cleaner than the D3200 in the deep shadows, but the D3200 also got less light. I think I'll call it a wash there as well, since the higher contrast on the a580 image hides some shadow noise, resizing will reduce the difference, and matching exposures will make the D3200 image look better in comparison.

So anyways, I think the D3200 sensor definitely isn't from Sony. I see several possible conclusions:

1. The D3200's sensor is 2/3 stops better than the current 16 MP sensor at high ISO. That would make for a DxO ISO score of > 1500, which is ridiculous for a crop sensor.
2. The D3200 might be marginally noisier in the shadows at base ISO.

3. The D3200 might be using noise reduction at high ISOs - but it'd have to be extremely good noise reduction because it's holding plenty of detail.

4. The DPR studio shots aren't done very well. ISO performance comparisons are nearly impossible because exposures aren't matched. Detail comparisons are also impossible because cameras are focused on different areas.

Any thoughts?

Also, has anyone noticed that some forum posts have negative times (like "-20 minutes ago")? It looks like dpreview's programmers have written some miracle code to fetch data from the future. I'm impressed.

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Nikon D3200 Sony SLT-A58 Sony SLT-A65
If you believe there are incorrect tags, please send us this post using our feedback form.
AdrianVB
New MemberPosts: 6
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to chlamchowder, Apr 29, 2012

You have to realize that noise and the way the camera handles noise is largely software based. Just like anti-noise systems in programs like Lightroom, your camera has similar software inside of it. Sony and Nikon simply use different software for their devices, as you would expect them to. Sensors just collect light information, it's up the software to dictate how it is interpreted.

Reply   Reply with quote   Complain
Archer66
Senior MemberPosts: 2,257
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to chlamchowder, Apr 29, 2012
Reply   Reply with quote   Complain
TrojMacReady
Senior MemberPosts: 8,364
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to chlamchowder, Apr 29, 2012

I think we can safely cross out scenario 1. 75% QE doesn't sound plausible at all, seeing as no large sensor has broken the 57% barrier the past 3 years.

Reply   Reply with quote   Complain
chlamchowder
Senior MemberPosts: 1,982Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to AdrianVB, Apr 29, 2012

Yes, that's true - that's why I included noise reduction on raw files as a possibility. However, I really think that's unlikely, because the D3200 raw files retain detail very well at high ISOs without showing any evidence of NR blurring. Also, if lightroom applied any additional noise reduction to the D3200 files, we'd see detail loss too.

You have to realize that noise and the way the camera handles noise is largely software based. Just like anti-noise systems in programs like Lightroom, your camera has similar software inside of it.

That's why I was comparing the raw files. JPG files are the ones that are heavily affected by in camera software processing. Unless the manufacturer tampers with the raw files by applying noise reduction to those files as well (Sony did that a while ago and people got pretty mad), using the same software to process the same raw files with all NR settings set to zero (which is what DPR does) should give a good comparison of sensor performance. Theoretically, Raw files = raw data from the sensor.

Sony and Nikon simply use different software for their devices, as you would expect them to. Sensors just collect light information, it's up the software to dictate how it is interpreted.

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Reply   Reply with quote   Complain
chlamchowder
Senior MemberPosts: 1,982Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to TrojMacReady, Apr 29, 2012

Perhaps, but how else can we explain the D3200's ability to pretty much match the a580's high ISO noise performance even with a 2/3 stop disadvantage?

I think we can safely cross out scenario 1. 75% QE doesn't sound plausible at all, seeing as no large sensor has broken the 57% barrier the past 3 years.

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Reply   Reply with quote   Complain
TrojMacReady
Senior MemberPosts: 8,364
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to chlamchowder, Apr 29, 2012

chlamchowder wrote:

Perhaps, but how else can we explain the D3200's ability to pretty much match the a580's high ISO noise performance even with a 2/3 stop disadvantage?

I think we can safely cross out scenario 1. 75% QE doesn't sound plausible at all, seeing as no large sensor has broken the 57% barrier the past 3 years.

Horrible lighting consistency is all I can think of.

The main reason I prefer the Imaging Resource setup. Still not perfect, but usually much closer matched and almost always in line with DXO measurements.

Reply   Reply with quote   Complain
panos_m
Senior MemberPosts: 1,199
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to chlamchowder, Apr 29, 2012

Keep in mind that DPR are varying the light in the studio setup for every camera to achieve a certain middle grey value based on jpeg output at standard (out of the box) settings.

EDIT: What I wrote above means that when a camera tested by DPR with 1/500 and another one with 1/1000 doesn't mean that the first camera necessarily received double the light because we don't know if the light was the same between the two shots.

chlamchowder wrote:

I've heard rumors about the D3200 sensor being manufactured by Sony, and plenty of rumors to the contrary. Out of curiosity, I decided to check the dpreview studio comparisons. As I moved up the ISO range (in raw), the D3200 images seemed noticeably cleaner than the a65 images. In addition to having less color noise, the D3200's grain structure was much finer.

I was about to attribute the difference to the translucent mirror when I checked the EXIF data for the studio shots, and was amazed to see that the D3200 received 2/3 stops less light than the a65. In the ISO 3200 image, the D3200 was set to f/8, 1/1250s, while the a65 was set to f/8, 1/800s. That more than makes up for the SLT light loss, but the D3200 still looks better.

Then, I compared the D3200 shots to studio shots from the a580, and it was pretty much a wash. At the pixel level, the a580 might be slightly cleaner in some areas, but resizing would easily make up for the difference. The a580 also got 2/3 stops more light than the D3200, but was still unable to beat it.

On the other hand, at ISO 100, the a580 is cleaner than the D3200 in the deep shadows, but the D3200 also got less light. I think I'll call it a wash there as well, since the higher contrast on the a580 image hides some shadow noise, resizing will reduce the difference, and matching exposures will make the D3200 image look better in comparison.

So anyways, I think the D3200 sensor definitely isn't from Sony. I see several possible conclusions:

1. The D3200's sensor is 2/3 stops better than the current 16 MP sensor at high ISO. That would make for a DxO ISO score of > 1500, which is ridiculous for a crop sensor.
2. The D3200 might be marginally noisier in the shadows at base ISO.

3. The D3200 might be using noise reduction at high ISOs - but it'd have to be extremely good noise reduction because it's holding plenty of detail.

4. The DPR studio shots aren't done very well. ISO performance comparisons are nearly impossible because exposures aren't matched. Detail comparisons are also impossible because cameras are focused on different areas.

Any thoughts?

Also, has anyone noticed that some forum posts have negative times (like "-20 minutes ago")? It looks like dpreview's programmers have written some miracle code to fetch data from the future. I'm impressed.

-- hide signature --

Panagiotis

Reply   Reply with quote   Complain
AdrianVB
New MemberPosts: 6
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to chlamchowder, Apr 29, 2012

chlamchowder wrote:

Yes, that's true - that's why I included noise reduction on raw files as a possibility. However, I really think that's unlikely, because the D3200 raw files retain detail very well at high ISOs without showing any evidence of NR blurring. Also, if lightroom applied any additional noise reduction to the D3200 files, we'd see detail loss too.

You have to realize that noise and the way the camera handles noise is largely software based. Just like anti-noise systems in programs like Lightroom, your camera has similar software inside of it.

That's why I was comparing the raw files. JPG files are the ones that are heavily affected by in camera software processing. Unless the manufacturer tampers with the raw files by applying noise reduction to those files as well (Sony did that a while ago and people got pretty mad), using the same software to process the same raw files with all NR settings set to zero (which is what DPR does) should give a good comparison of sensor performance. Theoretically, Raw files = raw data from the sensor.

Sony and Nikon simply use different software for their devices, as you would expect them to. Sensors just collect light information, it's up the software to dictate how it is interpreted.

I don't think you're understanding me here. The camera sensor collects light from the lens and sends the information it receives to the cameras processor to be processed. The 3200 and 77 may have the same sensor but they have different processors which run different software onboard. Fixing the noise problem could be just as simple as a firmware update from Sony.

Reply   Reply with quote   Complain
chlamchowder
Senior MemberPosts: 1,982Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to Archer66, Apr 29, 2012

Thanks - it looks like someone else has noticed the difference as well. Several comments:

1. The D3200 was given 2/3 stops less light for most of the shots (except the ones at ISO 6400 and above), yet still manages to beat the 24 MP Sony models.

2. The a65 seems slightly less noisy than the a77, but both models are noticeably behind the D3200.

3. "two stops ahead" would clearly be an exaggeration. However, I think it's safe to say that the D3200's sensor is outperforming the Sony sensor at high ISOs. The real question is how far ahead the D3200 is.

4. Shadow noise at low ISOs is difficult to compare. The Sony models may have an advantage there, but I don't feel comfortable drawing conclusions about that yet.

http://forums.dpreview.com/forums/readflat.asp?forum=1037&thread=41324631&message=latest

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Reply   Reply with quote   Complain
danny006
Contributing MemberPosts: 764Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to panos_m, Apr 29, 2012

I'm sorry, but the JPEG engine from the D3200 sucks, A65 is much better. Less noise and more detail.

http://www.dpreview.com/reviews/studio-compare#baseDir=%2Freviews_data&cameraDataSubdir=boxshot&indexFileName=boxshotindex.xml&presetsFileName=boxshotpresets.xml&showDescriptions=false&headerTitle=Studio%20scene&headerSubTitle=Standard%20studio%20scene%20comparison&masterCamera=nikon_d3200&masterSample=dsc_0020&slotsCount=4&slot0Camera=nikon_d3200&slot0Sample=dsc_0020&slot0DisableCameraSelection=true&slot0DisableSampleSelection=true&slot0LinkWithMaster=true&slot1Camera=sony_slta65&slot1Sample=dsc02913&slot2Camera=nikon_d3100&slot2Sample=dsc_0013&slot3Camera=sony_nex7&slot3Sample=dsc03033&x=-0.34146417249940847&y=0.2464542196642967

And the RAW files from D3200 have the same amount of noise as the Nex-7, and a little (1/3 of a stop) cleaner than A65.

http://www.dpreview.com/reviews/studio-compare#baseDir=%2Freviews_data&cameraDataSubdir=boxshot&indexFileName=boxshotindex.xml&presetsFileName=boxshotpresets.xml&showDescriptions=false&headerTitle=Studio%20scene&headerSubTitle=Standard%20studio%20scene%20comparison&masterCamera=nikon_d3200&masterSample=dsc_0020.acr&slotsCount=4&slot0Camera=nikon_d3200&slot0Sample=dsc_0020.acr&slot0DisableCameraSelection=true&slot0DisableSampleSelection=true&slot0LinkWithMaster=true&slot1Camera=sony_slta65&slot1Sample=dsc02913.acr&slot2Camera=nikon_d3100&slot2Sample=dsc_0013.acr&slot3Camera=sony_nex7&slot3Sample=dsc03033.acr&x=-0.33111084464402635&y=1.6448065119404947

 danny006's gear list:danny006's gear list
Sony RX100 Sony SLT-A77 Tamron SP AF 200-500mm F/5-6.3 Di LD (IF) Sigma 105mm F2.8 EX DG Macro
Reply   Reply with quote   Complain
chlamchowder
Senior MemberPosts: 1,982Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to danny006, Apr 29, 2012

I would agree to a certain extent, but I wanted to discuss raw image quality and sensor performance. JPEG performance depends quite a bit on in camera software, which is why I didn't mention JPEG performance in my original post. Can we stay on topic?

I'm sorry, but the JPEG engine from the D3200 sucks, A65 is much better. Less noise and more detail.

To my eye at least, the D3200 is slightly cleaner than the NEX-7. Especially in the shadow areas, the D3200's grain seems less heavy, and color noise is less apparent with the D3200. But yes, the difference between the NEX-7 and D3200 is very slight in most areas.

And the RAW files from D3200 have the same amount of noise as the Nex-7, and a little (1/3 of a stop) cleaner than A65.

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Reply   Reply with quote   Complain
mike_2008
Veteran MemberPosts: 5,635Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to danny006, Apr 29, 2012

danny006 wrote:

I'm sorry, but the JPEG engine from the D3200 sucks, A65 is much better. Less noise and more detail.

http://www.dpreview.com/reviews/studio-compare#baseDir=%2Freviews_data&cameraDataSubdir=boxshot&indexFileName=boxshotindex.xml&presetsFileName=boxshotpresets.xml&showDescriptions=false&headerTitle=Studio%20scene&headerSubTitle=Standard%20studio%20scene%20comparison&masterCamera=nikon_d3200&masterSample=dsc_0020&slotsCount=4&slot0Camera=nikon_d3200&slot0Sample=dsc_0020&slot0DisableCameraSelection=true&slot0DisableSampleSelection=true&slot0LinkWithMaster=true&slot1Camera=sony_slta65&slot1Sample=dsc02913&slot2Camera=nikon_d3100&slot2Sample=dsc_0013&slot3Camera=sony_nex7&slot3Sample=dsc03033&x=-0.34146417249940847&y=0.2464542196642967

And the RAW files from D3200 have the same amount of noise as the Nex-7, and a little (1/3 of a stop) cleaner than A65.

http://www.dpreview.com/reviews/studio-compare#baseDir=%2Freviews_data&cameraDataSubdir=boxshot&indexFileName=boxshotindex.xml&presetsFileName=boxshotpresets.xml&showDescriptions=false&headerTitle=Studio%20scene&headerSubTitle=Standard%20studio%20scene%20comparison&masterCamera=nikon_d3200&masterSample=dsc_0020.acr&slotsCount=4&slot0Camera=nikon_d3200&slot0Sample=dsc_0020.acr&slot0DisableCameraSelection=true&slot0DisableSampleSelection=true&slot0LinkWithMaster=true&slot1Camera=sony_slta65&slot1Sample=dsc02913.acr&slot2Camera=nikon_d3100&slot2Sample=dsc_0013.acr&slot3Camera=sony_nex7&slot3Sample=dsc03033.acr&x=-0.33111084464402635&y=1.6448065119404947

Yep, that was my take too. I'm sure that the slt and sony haters will choose to ignore the fact that sony has a better jpeg engine than nikon finally, but I for one am glad to see this.

BTW, it's very well known that ACR has different default NR reduction settings for different cameras, and that dpr uses the default settings in their raw tests. It's quite possible any slight differences in the raw comparisons at high iso come about through these modest differences in default NR from ACR. Saying that, the differences I see can hardly be claimed to be significant, imo, notwithstanding the OP's rather breathless claims to the contrary.

-- hide signature --

Hell hath no fury like an ovf-shooter scorned.

 mike_2008's gear list:mike_2008's gear list
Canon EOS 5D Sony SLT-A55 Sony SLT-A77 Canon EF 50mm f/1.8 II Canon EF 17-40mm f/4.0L USM +5 more
Reply   Reply with quote   Complain
chlamchowder
Senior MemberPosts: 1,982Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to TrojMacReady, Apr 29, 2012

That's terrible. This DPR tool is really starting to look bad. Focus variations make detail comparisons difficult, and lighting inconsistency (which isn't mentioned in EXIF data at all) would make noise comparisons impossible as well.

All I can say is, assuming that exposures were indeed matched for the D3200 and other cameras, the D3200 has a slight edge over the a65/77, and is pretty much identical to the a580.

Horrible lighting consistency is all I can think of.

The main reason I prefer the Imaging Resource setup. Still not perfect, but usually much closer matched and almost always in line with DXO measurements.

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Reply   Reply with quote   Complain
mike_2008
Veteran MemberPosts: 5,635Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to chlamchowder, Apr 29, 2012

chlamchowder wrote:

That's terrible. This DPR tool is really starting to look bad. Focus variations make detail comparisons difficult, and lighting inconsistency (which isn't mentioned in EXIF data at all) would make noise comparisons impossible as well.

All I can say is, assuming that exposures were indeed matched for the D3200 and other cameras, the D3200 has a slight edge over the a65/77, and is pretty much identical to the a580.

It's kind of amusing that there is much more interest in the 3200 here than on the nikon forum, driven mostly, it seems, by a few posters like the OP desperately trying to exaggerate how good it is compared with the A65 and A77. Another funny thing is that these same posters don't actually own the A65/A77. It's almost as if they have an agenda.
--
Hell hath no fury like an ovf-shooter scorned.

 mike_2008's gear list:mike_2008's gear list
Canon EOS 5D Sony SLT-A55 Sony SLT-A77 Canon EF 50mm f/1.8 II Canon EF 17-40mm f/4.0L USM +5 more
Reply   Reply with quote   Complain
chlamchowder
Senior MemberPosts: 1,982Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to mike_2008, Apr 29, 2012

I agree - the JPG results from Sony do look better. However (as I stated previously), I wanted to look at sensor performance. Also, I personally shoot everything in raw, so I typically don't pay much attention to JPG results.

Yep, that was my take too. I'm sure that the slt and sony haters will choose to ignore the fact that sony has a better jpeg engine than nikon finally, but I for one am glad to see this.

That could be true. However, noise reduction typically comes at the cost of detail. Given how well the detail in the D3200 raw shots holds up compared to the a65's raw shots, I doubt that 'under the hood' noise reduction at either the camera end or ACR end is responsible. Adobe and Nikon could have pulled a miracle, delivering under-the-hood NR that keeps a fine grain structure and preserves ridiculous amounts of detail, but I doubt it.

I somewhat agree that the differences are insignificant. For anything practical, the noise differences between any recent large sensor camera are irrelevant. However, when looking at buying a new camera, I do like to get the best image quality I can for as little money as possible.

BTW, it's very well known that ACR has different default NR reduction settings for different cameras, and that dpr uses the default settings in their raw tests. It's quite possible any slight differences in the raw comparisons at high iso come about through these modest differences in default NR from ACR. Saying that, the differences I see can hardly be claimed to be significant, imo, notwithstanding the OP's rather breathless claims to the contrary.

Finally, I have several comments and suggestions for your writing style.

The phrase "slt and sony haters will choose to ignore the fact..." implies that looking at raw data, where Sony currently has a disadvantage, is something only done by those who hate Sony and SLTs. That sentence could be better thought out. Here are some points to consider:

1. It is often desirable to have a focused discussion on one issue, so that it's easier to stay on topic. Introducing other issues, like JPEG performance, opens up another can of worms that makes it difficult to concentrate. Some people (like me) also shoot mainly in raw, making JPEG performance irrelevant. Separate issues should be discussed on separate threads.

2. Discussing an area in which Sony's current performance is weak does not imply hate towards Sony. Criticism doesn't imply hate - it's possible to criticize and love a product at the same time. Criticism also drives progress by telling companies where they have to pay attention.

3. How do you know that others "choose to ignore" something? Assuming that you know what others think is generally a bad idea (unless you really can read their minds). Honestly, I haven't paid any attention to JPG performance because I shoot everything in raw. In the past, when Sony's JPG engine was considered weak, I didn't care and bought a Sony camera anyways (an a580) for the raw image quality.

4. Thoughts on image quality are subjective, especially with JPG files where so much processing is involved - I would be cautious about claiming that the statement "sony's JPGs are better" as a fact (although I do agree with you on that point).

Try to avoid language like "breathless claims to the contrary". Such language is unnecessarily rude. A much better way to conclude would be "I respectfully disagree with the OP's stance." If I ended my post with "Despite Mike's breathless claims to the contrary, the D3200's raws are better", how would you feel? What if I left out the insult, and simply wrote, "I still think the D3200 holds an edge in raw image quality"? Being polite helps to avoid arguments and makes discussions more productive. I don't mind the language, but others might, and once a fight starts, it's hard to end.

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Reply   Reply with quote   Complain
mike_2008
Veteran MemberPosts: 5,635Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to chlamchowder, Apr 29, 2012

I have some advice for you.

It's unwise to start a thread titled "Nikon D3200... really really good sensor (not Sony)?" and then later in the thread admit that there are no significant differences with the sony sensor, and that the jpeg processing from the a65 is better than the 3200.

You can hardly complain about the charge of "breathless" with such an OP title. Try to be a little objective.

-- hide signature --

Hell hath no fury like an ovf-shooter scorned.

 mike_2008's gear list:mike_2008's gear list
Canon EOS 5D Sony SLT-A55 Sony SLT-A77 Canon EF 50mm f/1.8 II Canon EF 17-40mm f/4.0L USM +5 more
Reply   Reply with quote   Complain
Amateur Sony Shooter
Senior MemberPosts: 5,303Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to chlamchowder, Apr 29, 2012

You just pointed out the obvious fact that so many people try to defy and defend otherwise. I hope your thread develop to a civilized one. Good luck chlamchowder.

 Amateur Sony Shooter's gear list:Amateur Sony Shooter's gear list
Sony RX100 Canon EOS 7D Sony SLT-A99 Sony 70-200mm F2.8 G Sony 16-35mm F2.8 ZA SSM Carl Zeiss Vario-Sonnar T* +11 more
Reply   Reply with quote   Complain
chlamchowder
Senior MemberPosts: 1,982Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to AdrianVB, Apr 29, 2012

Here's how I understand the sequence of events with a raw file:
1. Sensor is exposed to light, charge builds up in each cell on the sensor.

2. The charges are read off the sensor, and analog-to-digital converters convert those charges to digital values (numbers, if you will).

3. Those numbers reach the processor, which packages them directly into a raw file format without modifying those numbers. Then, the raw file is written to the card. Additional parameters may be written to the raw file, but in an ideal raw file, the actual numbers from the ADC units aren't changed by the processor.

The only change a firmware update could make in this sequence is to modify the numbers coming off the ADC before they are written to the raw file. We've seen this before, in the form of noise reduced raw files on the a700/900. Noise reduction is usually accompanied with a loss of detail. The high ISO files from the D3200 (and Sony cameras as well) are very detailed, which leads me to think that noise reduction isn't being applied to the raw files (unless Nikon pulled a miracle and is reducing noise without losing detail, but I really doubt that).

Unless the firmware update is accompanied by a hardware update (sensor change), I don't think noise performance can be improved without sacrificing detail.

I don't think you're understanding me here. The camera sensor collects light from the lens and sends the information it receives to the cameras processor to be processed. The 3200 and 77 may have the same sensor but they have different processors which run different software onboard. Fixing the noise problem could be just as simple as a firmware update from Sony.

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Reply   Reply with quote   Complain
chlamchowder
Senior MemberPosts: 1,982Gear list
Like?
Re: Nikon D3200...really really good sensor (not Sony)?
In reply to mike_2008, Apr 29, 2012

Later in the thread, someone else brought up the excellent (if true) point that dpreview varies the brightness of their studio lighting, making it impossible to compare noise performance with the studio samples. I did not know that when I wrote my original post - I only observed that the D3200 was somehow outperforming or matching current Sony cameras (including the a580) when EXIF data showed that the D3200's shutter speed was 2/3 stops shorter. It seemed like the D3200 had a roughly 2/3 stop advantage over the a580, which really impressed me, and would represent a significant leap in image quality. But with the additional variable of studio light variations, I really can't make any conclusions.

Also, JPG performance is really irrelevant when discussing sensor performance, simply because so much processing is done to JPG files, and every brand seems to have a different JPG processing philosophy.

Perhaps the title was a little too much, but then I would suggest that it wouldn't be a problem if people stayed on topic, didn't treat the title as an insult (it really wasn't intended to be), and simply put their points down politely. It almost seems as if people are taking criticism against Sony as personal insults, and are acting aggressively to defend Sony. Why is that necessary??? Criticism encourages companies to develop better products. Don't people realize that?

It's unwise to start a thread titled "Nikon D3200... really really good sensor (not Sony)?" and then later in the thread admit that there are no significant differences with the sony sensor, and that the jpeg processing from the a65 is better than the 3200.

I'm not complaining about the fact that you disagree - I was merely pointing out that you could have been more polite, because I thought that would help you in the future (and make the forum a better place).

You can hardly complain about the charge of "breathless" with such an OP title. Try to be a little objective.

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads