Canon 1D-X: "Full" 2-Stops ISO Improvment...only in JPEG??

[Mt Spokane Photography]

The exact same story came out when the 1D MK IV was announced.  Chuck Westfall said the 1D MK IV was 1-1/2 to 2 stops better than the 1D MK III in JPEG, but about 1/2 stop better shooting raw.  It was due to the more sophisticated in camera processing that could be done using the Digic IV processor.  Processing has improved, and Digic V is more powerful than Digic IV.  Jpegs will improve over the 1DS MK III and 1D MK IV. 

The question I have, is how much improvement in raw over my 5D MK II?  1/2 stop or 1 stop?  If I can get clean raw images at ISO 12800, thats about 1/2 stop or slightly more than my 5D MK II, but if they are clean at ISO 25,600 then there is a significant advance, which I somehow doubt.

When I bought my 5D MK II in 2008, using LR 2 allowed me to shoot at ISO 3200, but results were noisy.  Now, with LR 4, I shoot ISO 6400 images and with no NR at all, they are remarkably clean.  Processing software has improved noticibly in the last three years, so I've gained about 2 stops when using raw, just by spending a few dollars on software upgrades.

[Mt Spokane Photography]

Don't a number of pro sports photographers (football, hockey) shoot jpeg, due to the large number of images they take during games and the push to get them out to the newspapers immediately (especially for late games)?

Time is the critical factor, so PJ's do use RAW + JPEG.  images are ready for online and newspaper publication immediately using jpeg, and, if a large magazine spread or other ultra high quality image is needed later, the raw can be post processed.

[wockawocka]

This is quite funny. The camera shoots in Raw, JPG is a conversion.

If the extras stops comes from Noise reduction in camera then Canon will get shafted sideways.

I'm a professional, 95% of the time I only shoot Raw. I will use Jpeg if I need to see things instantly on a Macbook on location.

In all of my cameras noise reduction is turned off, always, and I never use Noise reduction in post either. As a buyer of two models of the 1Dx I expect to see a significant improvement on the Raw files. If not I will seriously consider the smaller form factor of the 5D3 of which the only issue I had with the current model is the autofocus.

The weight of the 1Dx isn't practical for me, but it's necessary that I have a camera that focuses properly.

[NoWii]

The question I have, is how much improvement in raw over my 5D MK II?  1/2 stop or 1 stop?  If I can get clean raw images at ISO 12800, thats about 1/2 stop or slightly more than my 5D MK II, but if they are clean at ISO 25,600 then there is a significant advance, which I somehow doubt.

Why do you doubt? Between 5DmkII and 1D X, there has been nearly 4years of development. I think, that's enough for 1/2 stop. And don't forget the megapixel difference. 21mp vs 18mp, means bigger pixels and better performance. Thoose two factors togeder, could easily give us 1 stop better iso performance, so why not even more? I wouldn't be suprised if there's 3/2stops of difference.

[briansquibb]

The 1DX is a pro camera designed for pros. Jpeg shooting is important for them which would explain the emphasis on the extra stops.

What makes you think pros don't shoot with RAW just as much as JPEG? I think most of the pros I know use RAW+JPEG, rather than just JPEG, and when asked their reason is often simply so they can copy off the jpegs onto a small memory card and deliver them directly to a customer at end of job for previewing and pick selection, allowing them some more time to work the RAWs.

Sports pros just send their photos straight to the editors to get on the paper/web in real time - that is why they use the wireless adaptors. They dont have the time to process or even look at the pictures they are taking. That is exactly what they will be doing at the Olympics which is why Canon are pushing the 1DX so all the lens you see will be white.

[elflord]

Why do you doubt? Between 5DmkII and 1D X, there has been nearly 4years of development. I think, that's enough for 1/2 stop. And don't forget the megapixel difference. 21mp vs 18mp, means bigger pixels and better performance. Thoose two factors togeder, could easily give us 1 stop better iso performance, so why not even more? I wouldn't be suprised if there's 3/2stops of difference.

I would doubt it because ISO performance hasn't been moving very fast. For example, according to DxOMark's website, there is less than one stop between the 1Ds (original) which is about 6 years older than the 5D Mark II. The difference in megapixels is a factor of about 2 (11mp vs 21mp).

Whenever a new camera is released (even among cameras on a more frequent release cycle), there is talk of 1-2 stop improvements in ISO performance, but the reality is very different.

If it's as good as the Nikon D3s, that would be just under a 1 stop improvement over the 5D Mark II.

[wockawocka]

Until some fantastic new tech comes along the laws of physics will stop improvements. Just like lenses haven't got any smaller in 30 years. a 35mm sensor will always be a 35mm sensor with the same floorspace for pixels.

That's why this DSLR may be the last one before pros move to mirrorless. I never thought would happen but the damn Leica rangefinder has a full frame CMOS in it.

[JR]

When I bought my 5D MK II in 2008, using LR 2 allowed me to shoot at ISO 3200, but results were noisy.  Now, with LR 4, I shoot ISO 6400 images and with no NR at all, they are remarkably clean.  Processing software has improved noticibly in the last three years, so I've gained about 2 stops when using raw, just by spending a few dollars on software upgrades.

@Mt Spokane, i have yet to install the beta version of LR 4 (was waiting for the full verison) so I am currently using LR3.  Are you saying you noticed a significant improvement in RAW processing with LR4 to the point your ISO 6400 is like your ISO 3200 with LR2?  And with no NR applied?  Sorry I am just very intrigued and interested by your comments and not sure I understood...

If true then I will for sure get the LR4 beta!  Thanks for letting me know if you have a chance.

Jacques

[jrista]

When I bought my 5D MK II in 2008, using LR 2 allowed me to shoot at ISO 3200, but results were noisy.  Now, with LR 4, I shoot ISO 6400 images and with no NR at all, they are remarkably clean.  Processing software has improved noticibly in the last three years, so I've gained about 2 stops when using raw, just by spending a few dollars on software upgrades.

@Mt Spokane, i have yet to install the beta version of LR 4 (was waiting for the full verison) so I am currently using LR3.  Are you saying you noticed a significant improvement in RAW processing with LR4 to the point your ISO 6400 is like your ISO 3200 with LR2?  And with no NR applied?  Sorry I am just very intrigued and interested by your comments and not sure I understood...

If true then I will for sure get the LR4 beta!  Thanks for letting me know if you have a chance.

Jacques

I've also been working with both LR3 and LR4 with my 7D's RAW photos. So far, it seems to be a bit of a toss-up regarding NR. The initial imports seem to be a better than with LR3, and noise starts out a little better (along with pretty much everything else). However when it comes time to actually adjust the luminance or color noise sliders, I've had mixed results. Sometimes LR4 seems better by a small margin, however sometimes it seems to be a tad worse than LR3. I'm seriously hoping they improve the algorithms for Canon cameras. LR3 was a pretty big improvement over LR2, but so far, I'm not seeing the same degree of improvement with LR4. I've also been hearing more and more lately that Canon's DPP has far better noise removal for Canon RAW than LR ever has, and their code is publicly published, so Adobe could use it at will if they so desired (which they don't seem to.)

Noise aside, LR4 does seem to have some pretty nice improvements over LR3 when it comes to exposure adjustment. Whatever they changed, boosting darks and shadows seems to have much less of an impact on noise than LR3...so there is a bonus in that respect.

[JR]

@jrista, then Canon must have improved or changed DPP over the past 12 months because the verison i got with my 5D 12-18months ago was really bad compared to LR3.  I never bothered to check if Canon made any sort of improvement to DPP, maybe I will look it up based on your comments.  Interesting.

J

[Mt Spokane Photography]

When I bought my 5D MK II in 2008, using LR 2 allowed me to shoot at ISO 3200, but results were noisy.  Now, with LR 4, I shoot ISO 6400 images and with no NR at all, they are remarkably clean.  Processing software has improved noticibly in the last three years, so I've gained about 2 stops when using raw, just by spending a few dollars on software upgrades.

@Mt Spokane, i have yet to install the beta version of LR 4 (was waiting for the full verison) so I am currently using LR3.  Are you saying you noticed a significant improvement in RAW processing with LR4 to the point your ISO 6400 is like your ISO 3200 with LR2?  And with no NR applied?  Sorry I am just very intrigued and interested by your comments and not sure I understood...

If true then I will for sure get the LR4 beta!  Thanks for letting me know if you have a chance.

Jacques

My point is that software, all the raw software including in camera software in newer camera bodies has had a lot of improvements over the last 3+ years, so even without sensor changes, you can get better images.  I do not use DXO, but Neuro says its excellent, and I believe him.  I might try it on the same image to see if I see any differences.

LR2 was the current version when I bought my 5D2.  LR3 was a huge improvement in processing.  I took some 2500 images at a theater event last week, half with my assistant using the 7?d, and half with the 5D MK2.  Light was so poor, that most of the 7D images were at ISO 3200, and noisy.  However, a few hundred of my 5D MK 2 images were taken at ISO 6400 and I was suprised.  i went back to LR4 and verified no NR.

There is chroma noise, of course when you view at 1:1, its obvious.  It is much better, however, than I could do in 2008, and LR3 is also much better.

Here is a image, and a 1:1 with no NR so you can see the noise.



100%crop.  you can see the chroma noise, but just a little NR and its gone.

[Mt Spokane Photography]

I did process the image in DPP 3.1.1 just now, turned noise sliders to zero, noise is about the same, however, the image was exposed to the right, so pulling down the highlights probably reduces the visible noise.

DPP does not have the easy controls that LR has for adjusting the tone, and I'm not all that good at working just with curves.

As said earlier, its not about LR3 or LR4, its just that all the software out there has improved and you get a better image without upgrading your equipment, so a comparison needs to be done with todays software on both cameras, and that means comparing RAW.

[JR]

As said earlier, its not about LR3 or LR4, its just that all the software out there has improved and you get a better image without upgrading your equipment, so a comparison needs to be done with todays software on both cameras, and that means comparing RAW.

Got it...I see your point.  I know personally I saw a big difference between DPP and LR3.  One software I have not tried yet but want to is Topaz De-noise.  I heard from a few poeple that it is very good but dont know how it stacks against LR3 or LR4...

Jacques

[Edwin Herdman]

RAW vs. RAW, and JPEG vs. JPEG?  I imagine that must be what Chuck means here.

[motorhead]

There seems to be a serious misunderstanding of where the 1Dx sits in Canons line up. Its a next generation 1D, intended for PJ,s on a tight deadline. Hence the Jpeg thing.

Any pro looking to replace an aging 1Ds mkIII should be very careful if they seriously think the 1Dx is a possible replacement.

While not admitting it (yet?), Canon will eventually be forced to reexamine their thinking because the 1Dx fails to deliver on so many fronts for the landscapist/studio/fine art pro and serious amateur. Either that or loose customers to the Dark side.