The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

It is Finally Here, 907x 100c

tenmangu81

Well-known member
Sorry, I didn't make myself clear. I prefer real colors as a starting point. In Capture One, I start every edit with a linear profile. I've never used the Fuji color profiles in C1 for example, it's like canned food to me.
I don't know exactly what are real colors, but I've worked for about 10 years with Leica M series (M9 and M-P/240) and I was not satisfied with the colours, even when using a colour target calibrated input profile and a linear TRC in Capture One. But I guess Leica S3 is far better than M and comparable with HB, as far as colours are concerned, and as shown by Matt.
That said, I would be very happy if Capture One could open HB files, if, and any if, input profiles are at the expected level. I still prefer Capture One upon Lightroom. Capture One looks far more user-friendly and powerful to me than Phocus, and still even than Lightroom.
 
Last edited:

tcdeveau

Well-known member
I wasn't aware that CI had actually documented this. It was my impression that it isn't supposed to work and I have no clue why it does, but I routinely use flash with ES on the IQ4. There is some flexibility around the times that Brad gives (e.g. sometimes a half second will work) but that doesn't matter in most cases. If it works with the IQ, I'll bet someone just needs to figure out the exposure timing on the 907. Note, this type of flash has limited application in that any ambient light will affect the long exposure.

By the way, I get the feeling that many of us ignore the most important aspect of frame averaging. It is not a smoothing function. Sure, it can be used to make creamy waterfalls but that is incidental. After all, a very small subset of photographers make smooth water or light streak shots. In contrast, everyone can benefit from frame averaging. It is a noise reduction technique. If I am on a tripod and dealing with static subjects, I want frame averaging. It makes a huge difference to pixel granularity (your images are less noisy) and it greatly enhances your ability to retrieve shadow detail. That is why I have so much trouble understanding the clumsy way averaging is implemented on the few cameras that offer it. It should be a simple thing you turn on and off, not a special mode of operation.
IMO if making waterfalls creamy is incidental, then so is noise reduction.

As the name says, it’s just averaging images, so any parameter of the shot that could benefit from the averaging of images (motion, noise, etc) will benefit. I do agree it’s a mistake to only think of it as a smoothing technique, just as I’d say it’s a mistake to think of it only as a noise reduction technique.

There’s a lot of benefit to frame averaging but what it is to a given user just depends on what they’re using it for. For me, I used it primarily for the long exposure effect, and noise reduction was an incidental benefit for my uses. Might be the other way around for someone else.
 

cunim

Well-known member
IMO if making waterfalls creamy is incidental, then so is noise reduction.
Perhaps incidental was the wrong word and I should have said "uncommon". What I mean is that frame averaging, used for noise reduction, is as basic to scientific and remote sensing imaging as adjusting ISO is to photographic imaging. The spatial smoothing effects are considered to be "smearing" artefacts. The situation is different in photography, I agree, and it is good to have backs that let us take advantage of that.
 

tcdeveau

Well-known member
Perhaps incidental was the wrong word and I should have said "uncommon". What I mean is that frame averaging, used for noise reduction, is as basic to scientific and remote sensing imaging as adjusting ISO is to photographic imaging. The spatial smoothing effects are considered to be "smearing" artefacts. The situation is different in photography, I agree, and it is good to have backs that let us take advantage of that.
Fair enough.

There's more to the advantages of frame averaging for the smooting/long exposure effect beyond just smoothing too. Not only were the images much cleaner because of the noise improvements you mention, but you don't have to worry about using a dark frame like you would in traditional long exposure photography, and in many situations I could leave ND filters and filter holders home. To me and other landscape photographers those are pretty big advantages in the field that go beyond just smoothing and noise reduction too.

Frame averaging is super useful tool though, however one uses it, and it'd be great to see it on more cameras like the CFV100c.
 

Steve Hendrix

Well-known member
The wonderful colors of Hasselblad are often mentioned. I have no experience with them.
When I'm shooting, I usually need accurate colors, which is usually possible with my Nikon, Sony, Fuji + Phase One devices.
If there are problems, color correction in Capture One or x-rite helps.
Are the "wonderful" colors from Hasselblad more accurate than others?
I have never seen a convincing example.

The colors out of the box (processed through Phocus) I see with the Hasselblad files from recent years are as good as any color I've ever seen from any capture device (from my own perspective). By good, I don't mean so much that they are more accurate (though they may be). It is the relatively gentle handling of the colors and the tones, the lack of over saturation in certain colors that I see with almost every other digital capture device and then have to tamp down a bit. I would say the colors are more realistic, rather than more accurate (though ... that kind of sounds the same, I know, but it's not). I suppose for some who prefer contrasty, punchy looking files, this could be a negative. But I like the idea of a modest and realistic starting point, and then amping up if you feel the need or the desire versus starting amped up and trying to tone down. Everyone's eyes are different, but ... I have seen files from many many many many different capture devices with my eyes.


Steve Hendrix/CI
 

Steve Hendrix

Well-known member
Fair enough.

There's more to the advantages of frame averaging for the smooting/long exposure effect beyond just smoothing too. Not only were the images much cleaner because of the noise improvements you mention, but you don't have to worry about using a dark frame like you would in traditional long exposure photography, and in many situations I could leave ND filters and filter holders home. To me and other landscape photographers those are pretty big advantages in the field that go beyond just smoothing and noise reduction too.

Frame averaging is super useful tool though, however one uses it, and it'd be great to see it on more cameras like the CFV100c.

The Frame Averaging tool of the IQ4 does a wonderful job of turning harsh shadows into lovely smooth gradients. Often, all things being equal, it just produces a prettier file (in certain conditions).


Steve Hendrix/CI
 

buildbot

Well-known member
I think creating a jpeg is an order of magnitude or two more difficult, as it requires demosaicing. Frame averaging is massively parallel, since each pixel has its own computation that doesn't depend on its neighbors. I don't know what numerical tools ar built in to the cameras processor, but if there are *any* full image numerical routines, and there must be for LCD/EVF display, then frame averaging has to be trivial (I've written it - it takes one line of code). Granted, that's the kind of statement that real coders view as a kiss of death, so I'm hesitant to be 100% certain. I'd love to talk with someone who actually does the low-level coding on any digital camera.
I’m in total agreement, it’s certainly lighter weight than JPEG creation. The only trick is you’d need to convert from integer numbers to floating point to make sure you don’t overflow. (unless they sum into a 32bit int I guess then rescale at the end? Several ways to do it).

The camera probably does more math to make the EVF/LCD look good.
 

corvus

Active member
...
What I like about the product is that, in addition to the many configurations you can use it in, specifically for tech cameras, it helps relieve some of the high cost for those wanting to get into a premium tech camera system with a BSI sensor. And thank you for having a tilt screen.

Steve Hendrix/CI
In principle, I might be a candidate ... I have some experience with TS lenses and since last year with an Arca-Swiss F-Line. The biggest advantage for me would be the non-existent FFD of a CFV digital back for wide-angle in architecture (I'm an architect myself and also shoot for other offices) With DSLM / DSLR, of course, a certain willingness to compromise is always required, as we have already discussed in detail here in the 35mm lens thread, for example.
Not so much the resolution, but BSI is of course an argument against a CFV 50. However, if I follow the posts here, the solution is not yet clear to me, as the bigger investment will be the lenses. And all in all, I don't know if that would be a big enough step up from my current DSLM setup to justify it. Some of the descriptions about the shutter, flash, long exposures etc. also make me feel unsure about taking such a huge step ... This requires further research.
 

jng

Well-known member
I routinely use the IQ4's frame averaging to reduce the effects of convection/atmospheric turbulence in distant landscapes and cityscapes. When I'm out with the X2D, I use the intervalometer function to capture some number of images and then align and blend in Photoshop. The number of frames is generally limited by the number that my first-generation M1 MacBook Pro can reasonably handle - somewhere between 12-20 frames is usually within the limits of my patience (or how full my wine glass is when I start). I employ a ND filter if I need to stretch the total effective exposure time to smooth water, clouds, etc.

And I agree, frame averaging - whether in camera or in post - is a very effective noise reduction technique.

John
 

rdeloe

Well-known member
In principle, I might be a candidate ... I have some experience with TS lenses and since last year with an Arca-Swiss F-Line. The biggest advantage for me would be the non-existent FFD of a CFV digital back for wide-angle in architecture (I'm an architect myself and also shoot for other offices) With DSLM / DSLR, of course, a certain willingness to compromise is always required, as we have already discussed in detail here in the 35mm lens thread, for example.
Not so much the resolution, but BSI is of course an argument against a CFV 50. However, if I follow the posts here, the solution is not yet clear to me, as the bigger investment will be the lenses. And all in all, I don't know if that would be a big enough step up from my current DSLM setup to justify it. Some of the descriptions about the shutter, flash, long exposures etc. also make me feel unsure about taking such a huge step ... This requires further research.
I hear you. I'd be an obvious candidate for one of these, but in my own personal calculus, the benefits do not outweigh the many costs on the functionality side.

Having said that, I am beyond thrilled that Hasselblad has brought another option to market that will be exactly what a lot of people need.
 

MGrayson

Subscriber and Workshop Member
I’m in total agreement, it’s certainly lighter weight than JPEG creation. The only trick is you’d need to convert from integer numbers to floating point to make sure you don’t overflow. (unless they sum into a 32bit int I guess then rescale at the end? Several ways to do it).

The camera probably does more math to make the EVF/LCD look good.
Of course, I've been stupid. You only need one division at the end if you just add a few bits to the buffer size (and doubling is as easy as anything smaller) so 32 bit accumulation. That's why I'm not a professional coder....
 

diggles

Well-known member
In principle, I might be a candidate ... I have some experience with TS lenses and since last year with an Arca-Swiss F-Line. The biggest advantage for me would be the non-existent FFD of a CFV digital back for wide-angle in architecture (I'm an architect myself and also shoot for other offices) With DSLM / DSLR, of course, a certain willingness to compromise is always required, as we have already discussed in detail here in the 35mm lens thread, for example.
Not so much the resolution, but BSI is of course an argument against a CFV 50. However, if I follow the posts here, the solution is not yet clear to me, as the bigger investment will be the lenses. And all in all, I don't know if that would be a big enough step up from my current DSLM setup to justify it. Some of the descriptions about the shutter, flash, long exposures etc. also make me feel unsure about taking such a huge step ... This requires further research.
The CFV 100 C on a technical camera will be a very capable tool for architectural photography. Since your primary use is for architecture, lenses are the key.

With Schneider APO Digitar lenses anything shorter than 60mm will require an LCC. For the times you do need a wider angle lens like the 43XL or 35XL then you will need to take a LCC reference image to compensate for color cast. Not a big deal really because Phocus has LCC functionality, it is called Scene Calibration. So this is easily fixable.

The difficulty with Schneider APO Digitar lenses is that they can be very difficult to find.

With many of the latest Rodenstock lenses you will have to deal with distortion. When shooting at an angle it is not objectionable to me, but if you like to make 1 point perspective images then it may bother you. The issue will be fixing the distortion. I am not 100% sure, but I do not think Phocus has lens distortion correction with camera movement settings built into their software like Capture One does.

The problem here is that Capture One does not accept Hasselblad files so you may have to fix distortion issues manually using Photoshop.

A solution could be to start with the easier to find Schneider lenses like the 35XL, 47XL, 72L, 90n, and 120n. All of these are 'reasonably' priced, sharp, and should do well with the new Hasselblad back. Then you can be on the lookout for the more sought after Schneider lenses like the 43XL, 60XL, and 120ASPH to 'upgrade' when the opportunity comes around.

Hope that helps.
 

jng

Well-known member
In principle, I might be a candidate ... I have some experience with TS lenses and since last year with an Arca-Swiss F-Line. The biggest advantage for me would be the non-existent FFD of a CFV digital back for wide-angle in architecture (I'm an architect myself and also shoot for other offices) With DSLM / DSLR, of course, a certain willingness to compromise is always required, as we have already discussed in detail here in the 35mm lens thread, for example.
Not so much the resolution, but BSI is of course an argument against a CFV 50. However, if I follow the posts here, the solution is not yet clear to me, as the bigger investment will be the lenses. And all in all, I don't know if that would be a big enough step up from my current DSLM setup to justify it. Some of the descriptions about the shutter, flash, long exposures etc. also make me feel unsure about taking such a huge step ... This requires further research.
The CFV 100 C on a technical camera will be a very capable tool for architectural photography. Since your primary use is for architecture, lenses are the key.

With Schneider APO Digitar lenses anything shorter than 60mm will require an LCC. For the times you do need a wider angle lens like the 43XL or 35XL then you will need to take a LCC reference image to compensate for color cast. Not a big deal really because Phocus has LCC functionality, it is called Scene Calibration. So this is easily fixable.

The difficulty with Schneider APO Digitar lenses is that they can be very difficult to find.

With many of the latest Rodenstock lenses you will have to deal with distortion. When shooting at an angle it is not objectionable to me, but if you like to make 1 point perspective images then it may bother you. The issue will be fixing the distortion. I am not 100% sure, but I do not think Phocus has lens distortion correction with camera movement settings built into their software like Capture One does.

The problem here is that Capture One does not accept Hasselblad files so you may have to fix distortion issues manually using Photoshop.

A solution could be to start with the easier to find Schneider lenses like the 35XL, 47XL, 72L, 90n, and 120n. All of these are 'reasonably' priced, sharp, and should do well with the new Hasselblad back. Then you can be on the lookout for the more sought after Schneider lenses like the 43XL, 60XL, and 120ASPH to 'upgrade' when the opportunity comes around.

Hope that helps.
There are some good lenses on the longer end that aren't too rare and shouldn't break the bank. For example, the Hasselblad-Zeiss 3.5/100 Planar is exceptionally sharp and distortion free, and also provides a generous image circle. In my experience, it performs well when shifted simultaneously 10mm laterally and 10mm vertically on the larger 54x40mm sensor, with only a hint of purple fringing at the corners that's easily corrected in post.

I'll let others speak to some of the Mamiya and Pentax lenses.

John
 

mristuccia

Well-known member
The colors out of the box (processed through Phocus) I see with the Hasselblad files from recent years are as good as any color I've ever seen from any capture device (from my own perspective). By good, I don't mean so much that they are more accurate (though they may be). It is the relatively gentle handling of the colors and the tones, the lack of over saturation in certain colors that I see with almost every other digital capture device and then have to tamp down a bit. I would say the colors are more realistic, rather than more accurate (though ... that kind of sounds the same, I know, but it's not). I suppose for some who prefer contrasty, punchy looking files, this could be a negative. But I like the idea of a modest and realistic starting point, and then amping up if you feel the need or the desire versus starting amped up and trying to tone down. Everyone's eyes are different, but ... I have seen files from many many many many different capture devices with my eyes.


Steve Hendrix/CI
I'd use the word "honest", like audiophiles use to say for the best high-end audio gears.
IMHO, with such high level of technology and amount of money involved (yes, for me 8K is a big amount of money) the law of diminishing return applies, so even the smallest advantage Phocus can give in terms of "color honesty" matters a lot.
 

hcubell

Well-known member
In viewing one of the YT videos about the new 907x/100C, I was very amused to see a couple of features on the 907x/100C that have been notably missing on the the X2D. First, you can plug the Hasselblad remote cable release into a USB-C port on the bottom of the new digital back. Second, the accessory handgrip for the 907x has a.....JOYSTICK for moving the AF point around.
 

buildbot

Well-known member
Of course, I've been stupid. You only need one division at the end if you just add a few bits to the buffer size (and doubling is as easy as anything smaller) so 32 bit accumulation. That's why I'm not a professional coder....
Not at all, that's a numerical thing a lot of programmers would miss probably. Thinking about numerical precision is almost literally half my job, but I am also not a software developer/engineer (currently, I have been).

I'd use the word "honest", like audiophiles use to say for the best high-end audio gears.
IMHO, with such high level of technology and amount of money involved (yes, for me 8K is a big amount of money) the law of diminishing return applies, so even the smallest advantage Phocus can give in terms of "color honesty" matters a lot.
This dovetails nicely into my above point - Phocus has the experts who know how to extract every last bit of information from their capture chain (sensor on up!), and I agree, it's like using Lightroom for a IQ series back. Like you can, but you are missing that extra little bit of integration and that's (IMO) silly when you are paying 8-50K! Or running Windows on a MacBook. Of course, each to their own! if you like Lightroom or Windows more, awesome :)

In viewing one of the YT videos about the new 907x/100C, I was very amused to see a couple of features on the 907x/100C that have been notably missing on the the X2D. First, you can plug the Hasselblad remote cable release into a USB-C port on the bottom of the new digital back. Second, the accessory handgrip for the 907x has a.....JOYSTICK for moving the AF point around.
I wonder if that release will work for the X2D as is? Maybe with a firmware update? I imagine it's literally just a USB HID device which they coded the firmware to respond to as a shutter activation.
 

Paul Spinnler

Well-known member
The CFV 100 C on a technical camera will be a very capable tool for architectural photography. Since your primary use is for architecture, lenses are the key.

With Schneider APO Digitar lenses anything shorter than 60mm will require an LCC. For the times you do need a wider angle lens like the 43XL or 35XL then you will need to take a LCC reference image to compensate for color cast. Not a big deal really because Phocus has LCC functionality, it is called Scene Calibration. So this is easily fixable.

The difficulty with Schneider APO Digitar lenses is that they can be very difficult to find.

With many of the latest Rodenstock lenses you will have to deal with distortion. When shooting at an angle it is not objectionable to me, but if you like to make 1 point perspective images then it may bother you. The issue will be fixing the distortion. I am not 100% sure, but I do not think Phocus has lens distortion correction with camera movement settings built into their software like Capture One does.

The problem here is that Capture One does not accept Hasselblad files so you may have to fix distortion issues manually using Photoshop.

A solution could be to start with the easier to find Schneider lenses like the 35XL, 47XL, 72L, 90n, and 120n. All of these are 'reasonably' priced, sharp, and should do well with the new Hasselblad back. Then you can be on the lookout for the more sought after Schneider lenses like the 43XL, 60XL, and 120ASPH to 'upgrade' when the opportunity comes around.

Hope that helps.
Don’t forget the 28XL :)
 

corvus

Active member
The CFV 100 C on a technical camera will be a very capable tool for architectural photography. Since your primary use is for architecture, lenses are the key.

With Schneider APO Digitar lenses anything shorter than 60mm will require an LCC. For the times you do need a wider angle lens like the 43XL or 35XL then you will need to take a LCC reference image to compensate for color cast. Not a big deal really because Phocus has LCC functionality, it is called Scene Calibration. So this is easily fixable.

The difficulty with Schneider APO Digitar lenses is that they can be very difficult to find.

With many of the latest Rodenstock lenses you will have to deal with distortion. When shooting at an angle it is not objectionable to me, but if you like to make 1 point perspective images then it may bother you. The issue will be fixing the distortion. I am not 100% sure, but I do not think Phocus has lens distortion correction with camera movement settings built into their software like Capture One does.

The problem here is that Capture One does not accept Hasselblad files so you may have to fix distortion issues manually using Photoshop.

A solution could be to start with the easier to find Schneider lenses like the 35XL, 47XL, 72L, 90n, and 120n. All of these are 'reasonably' priced, sharp, and should do well with the new Hasselblad back. Then you can be on the lookout for the more sought after Schneider lenses like the 43XL, 60XL, and 120ASPH to 'upgrade' when the opportunity comes around.

Hope that helps.
Thank you, Warren! That helps me. The software is of course also an issue. I find the correction methods in c1 very helpful - my smallest symmetrical lens is currently a 58mm, which I already correct with lcc when shifting. The distortion is practically zero. For smaller focal lengths, I use converted shift lenses. here, of course, distortion is the main issue. It's a pity that HB and C1 obviously don't correspond. I'm already looking for XL lenses, but let's see...
 

PeterA

Well-known member
No really.
Leica M11 has the same pixel pitch as CFV 100C, so it has the same camera shake issue. Leica M11 owners are shooting handheld "en masse."
You get the rangefinder experience and all the blurry photos are the bonus icing on top.!!
 
Top