Multiple RAW image resolutions coming to the Canon EOS R5 Mark II?

[JDEE]

I’m really looking forward to a 60MP-ish R5 body. I’d buy it tomorrow.

1.6x Crop mode should be more like 25MP (instead of current 45>17) and turn my 15-35/2.8 into more of a 24-60mm equivalent but still at a r6mk2 / r3 level of resolution output) which negates my use of a 24-70/2.8 almost entirely and better for event photos or parts of a wedding day where I don’t need so much resolution.

I just wish we could map one button to cycle between crop modes without bringing up the menu. Add a menu item to enable / disable which modes you can cycle between.

For me that would be only 1 and 1.6x. DOF preview button is the perfect place for it and where my current r5 is mapped.

 

[scyrene]

I shoot events and landscapes. 60MP fine for landscapes (although I am happy with 45MP), but 20 MP is ideal for events, so I had an R5 and R6. However, for events, I need two bodies, and can't justify 3 bodies. The ability to shoot a higher MP body at around 20MP would be great. Its not clear to me why that would compromise those who just want 60MP.
And yes, I shoot CRaw.

The other way around that would be to use crop mode with a wider lens, potentially?

 

[dbcooper]

One of the endless complaints I have about the R5 is the raw files are too big for my work and cost me significant time later in the day to transfer and start editing. I often revert back to the 5D MkIV.

 

[AlanF]

One of the endless complaints I have about the R5 is the raw files are too big for my work and cost me significant time later in the day to transfer and start editing. I often revert back to the 5D MkIV.

The RAW files on my 5DIV were 35-40 Mb and the cRAW files on my R5 10-15 Mb less at 20-28 Mb. Have you tried cRAW? Most seem to find cRAW has little if any loss of IQ.

 

[Rudolf_Appel]

I would prefer eliminating rolling shutter and mechanical shutter over 60 MPixels. Both will provide sharper and less distorted images, thus higher quality than 60 MP has to offer.

 

[Frodo]

The other way around that would be to use crop mode with a wider lens, potentially?

Good one! At the end of our last summer, I went on a 200km hike, taking the R5, RF 24/1.8 and RF 85/2. For snapshots, I used the RF 24 in crop mode, giving me an effective 35mm; using the two lenses in FF for "more serious" landscape photos.

 

[Quarkcharmed]

I can see why a smaller raw can be required in professional photojournalism, but even then those look like edge cases.
I don't see why I would use downsampled raws except they're made special, e.g. with a higher dynamic range.

 

[Quarkcharmed]

Good one! At the end of our last summer, I went on a 200km hike, taking the R5, RF 24/1.8 and RF 85/2. For snapshots, I used the RF 24 in crop mode, giving me an effective 35mm; using the two lenses in FF for "more serious" landscape photos.

But in terms of hiking, one RF 24-105 would take less space in the backpack and provide a wider focal range with no need for cropping.

 

[sanj]

I suspect this means that lower-resolution files will be available without the IQ loss of C-RAW. Is that possible?

 

[Del Paso]

Its called pixel binning - where by say 2 adjacent pixels are averaged together across, and then down, to make a smaller image. Depending on how fancy they get, this can be interpolated to average 1.5 pixels together with software tools like drizzle that interpolate between pixels. But it works pretty well. Its used extensively in astrophotography, so not really breaking new ground.

About shutter speed - I suppose its always best to plan for the highest resolution. But how much cushion the lower res modes buy you will have to be tested once its on the market. A lot will depend on implementation.

Brian

Thank you! Now I've understood!

 

[SwissFrank]

This really isn't that hard if you just pay attention.

Why be rude, man? And CR45 if it's "not that hard" why have a CR243223 for this at all? What's wrong with CR234124123413?

 

[SwissFrank]

Why be rude, man? And CR45 if it's "not that hard" why have a CR243223 for this at all? What's wrong with CR234124123413?

Oh, you probably forgot:

CR45=even
CR243223=special secret code
CR234124123413 = just saying the thing you mean in words

"It's not that hard, just pay attention," you say.

 

[shadowsports]

Variable resolution settings for still images is something I've thought about for years. I would take 60MP capability, but might use 45 or so on a day to day basis. I wouldn't want to go backwards unless the body was R1 level. I went from 26.2 to 45MP for a reason, but would be quite happy with this feature on a new body. I have no problems with larger files (transferring, storing) or working with them in post, so bring it on.

 

[koenkooi]

This reads like the return of MRAW and SRAW of the old days where you could shoot in RAW with reduced resolution, e.g for the 1D Mark 4:

• Raw 16M 4,896 x 3,264
• MRaw 9.0M 3,672 x 2,448
• SRaw 4.0M 2,448 x 1,632

I feel compelled to point out that m/sRAW weren't RAW: the were demosaiced, downscaled linear TIFFs. Still a lot better than JPEG, but doing things like adjusting the white balance gives results closer to JPEG than proper RAW.

 

[koenkooi]

It's a DNG file according to the metadata, the 18 MP files as well.
Quote from the German website: "wählen Sie zwischen 60, 36 oder 18 Megapixeln Auflösung für DNG und JPG"
"Choose between 60, 36 or 18 MP resolution for DNG and JPG"

DNG as a container format has a few options on how to store the data, one of them being 'linear DNG', which is demosaiced. DNG-the-format also has options for lossy compression, so the file being a DNG doesn't tell us what it actually contains.

I suppose those DNGs are much bettter than JPEGs and good enough for Leica users. So practically speaking it doesn't really matter whether it's proper RAW or not

 

[koenkooi]

I've noticed that a lot of the articles on about the R5II keep mentioning 'pixel binning' for 8k-from-60MP, but isn't regular oversampling a much superior solution? I get that pixel binning can be done at the hardware level or very efficiently at the start of the pipeline, but you're throwing away a lot of sharpness data.

 

[bbasiaga]

I've noticed that a lot of the articles on about the R5II keep mentioning 'pixel binning' for 8k-from-60MP, but isn't regular oversampling a much superior solution? I get that pixel binning can be done at the hardware level or very efficiently at the start of the pipeline, but you're throwing away a lot of sharpness data.

I'm not sure for this, but in the astro world, hardware binning is a thing of the past. Its all software, and therefore quite possible it really is just oversampling. The term binning lives on though, absorbing the current software practice.

Brian

 

[dolina]

Less than 4 months before EOS R5 Mark II will be released.

 

[neuroanatomist]

Less than 4 months before EOS R5 Mark II will be released.

@SwissFrank would call that a CR74.29478569283745 rumor.

 

[CampanellaFoto]

I obviously know and have used cRAW but more even smaller options would be welcome.