Jump to content

Medium Format Rumors


Patrick FR

Recommended Posts

I would prefer the 44x33 to a bigger MF sensor... a bigger sensor would make the camera too big for my interest...

 

A 44x33 Fuji mirrorless can in effect, be their version of the full frame DSLR... same size camera and with a bigger sensor that is a wider differentiation from APS-C. 

 

If Fuji comes out with such a system, it is the lenses that I would be buying into. So that would really decide it for me. Because of the speed with which the sensors have been improving, camera bodies do not retain their value. With film, an old camera still took just as good of a photo as a new camera, so an old, well built camera body retained its value. At some point, and I think we are close to it, the sensors will be good enough that it will not matter so much whether the next one is better. There is a point of diminishing returns and I think we are almost there.

 

I've thought of digital camera bodies as 'disposable' because within a few years the new models have been so much better. $6000+ for a soon to be obsolete camera is rather costly. A sensor generation equal to the new X-Pro2 but in 44x33 at say 70MP would be very impressive and maybe is enough that it will not feel 'disposable'. The point where many users feel that way is when it makes sense (IMO) for Fuji to come out with the MFD.

Link to post
Share on other sites

I don't know where the idea that the 51Mp MF Sony CMOS sensor used in the Pentax 645Z is "too old to be used in a 2017 camera" came from. Excuse me if I missunderstood this.

 

It's not even 2 years old (announced April '14). Fuji used the "old" 16mp APS-C sensor for ages; giving it the X-Trans treatment plus newer processors and firmware of course. That sensor was almost two years in the market when Fuji first put it in the X-Pro1.

 

Speaking of the "small" (sic!) 2 year old cropped MF sensor, if used with similar optics, results seem to be notably better than anything from FF, including newer cameras such as the a7Rii and Canon 5DSr. If we are to trust test shots from Imaging Resource and DPReview, it excibits resolution, color and DR advantages over them, while being almost free of false color and moire. Noise wise it is at least on par with the high-megapixel FF sensors.

 

We can imagine what an X-Trans version would be like, paired with Fuji optics.

 

(Note: there is no DxOMark test of this sensor, which is curious. In fact, there is a rumor circulating that DxO leaked, at some point, a score of 101 for it, but later removed the reference. This has almost become a conspiracy theory among Pentaxians, who claim that DxO burried the actual benchmarks. Take this with a grain of salt)

 

In regards to actual surface, this sensor is ~66% larger than a 36x24 sensor. As a format my opinion is this is ideal to build a new system upon. As previous commenters noted, the point is building a system and I particularly agree with gdanmitchell: Fuji can just "skip" FF and offer a complete high end professional system with definite advantages. Furthermore, future versions of this 44x33 MF sensor can definitely offer higher resolution as well as other image quality improvements.

 

Apart from size considerations (after all, we all expect the Fuji MF to be similar to a GW690 or Mamiya 7 in format), this sensor shall obviously offer cost advantages, both directly and because of smaller lenses. In line with the argument that Fuji is actually going after high-end FF system cameras, perhaps a basic system with 3 lenses could be made available for about (or even less than) $10,000. This is in the same range as the top FF cameras plus premium glass.

 

Link to post
Share on other sites

The Phase One IQ250 was the first to use the 44 x 33mm 50MP CMOS sensor. As this article dated January 2014 shows, it was released in January of that year:

 

http://www.pdnonline.com/gear/Phase-One-Unveils-Fi-10189.shtml

 

So even well before Fuji officially announce their medium format camera, this sensor is already over two years old. Not "almost" two years old, but over two years old. By 2017, when it is rumoured the Fuji MFD model might be released, it will be three years old. That one year may not seem to be much, but in terms of the pace of sensor development, it is highly significant.

 

The problem is that six months ago, last August, Sony released a 42.4MP BSI small format sensor. One more iteration of Sony small format sensors, and they will equal or surpass the Sony 50MP cropped MFD sensor in terms of performance and resolution, something that can be expected to happen some time in 2016. Come 2017, small format sensors would have long since matched and surpassed the performance of the by then increasingly aging Sony 50MP cropped MFD sensor. It would be only too easy for Sony to release something like an a9/a9R/a99II/a99RII with +50MP BSI sensors some time between 2016-2017, at which point who on earth would buy a more expensive 50MP MFD Fuji camera?

 

At the same time, Canon will by 2017 have got set to release a camera with a small format sensor with more than 50MP resolution, and have publically stated last year that they plan to push resolutions up to 100MP within the next couple of years. Canon should soon start to catch up with Sony in the dynamic range and ISO performance of their 50+MP sensors. They know they have gotten behind Sony, and have made announcements saying they are going to invest substantially in R&D. They may even leapfrog Sony again, resulting in their small format sensors performing similarly to or better than the cropped MFD sensors from Sony. If Fuji announce a 50MP MFD model only to be overshadowed by Canon announcing a 100MP model not long after it, the Fuji model could end up looking rather silly, and risks turning into a commercial liability for the company.

 

What is "aging" the Sony 50MP cropped MFD sensor is not the mere ticking of the clock, but the rapid pace of small format sensor development, which proceeds faster than the pace of MFD sensor development. While the 50MP MFD sensor may not have been entirely surpassed as yet, come 2017 it will have been, with the situation rapidly deteriorating for this aging sensor in the market moving ahead of 2017, because Sony tends to invest the latest sensor technology more into their small frame sensors.

 

In conclusion: it is difficult to imagine Fuji releasing a premium priced MFD camera in 2017 with a three year old cropped non-BSI MFD sensor with "only" 50MP resolution, with the intention of keeping this model in production for several years ahead, because small format sensor technology is pressing ahead so rapidly.

Edited by Sator-Photography
Link to post
Share on other sites

The next thing that should be kept in mind about 44 x 33mm sensors is that the claim made by Phase One about the cropped MFD format sensor being over 60% larger than a full frame small format sensor is a highly misleading marketing ploy. Pentax should be commended for not playing this outrageous game, but there sadly remains a difference between meaningful real world sizes and exaggerated marketing sizes. A cropped MFD sensor may be 60% larger by surface area, but resolution is a linear function, which means we must compare format sizes in terms of linear size difference. The true increase in linear size of a cropped MFD sensor over FF small format is more like 20%.

 

In addition to not strictly speaking being medium format at all, cropped MFD sensors represent too small an increase in linear size to be meaningfully differentiated from FF small format sensors. Given that sensor development is driven by smaller formats, the pace of development of small formats tends to additionally outstrip that of MFD, thus further tending to rapidly negate this 20% increase in real world linear size difference, as MFD sensor development lags behind.

 

Another way of putting it is that the improvement in IQ from newer smaller format sensors may be such that there may be ever dwindling yield from increasing sensor size in order to get cost-effective and cost-meaningful improvements in IQ, unless MFD sensors can come down in price while keeping pace in their technological development with smaller format sensors. I am sure Fuji are aware of this and are furiously having board room debates over it, just like we are.

Edited by Sator-Photography
Link to post
Share on other sites

I shall repeat my argument about the 16mp APS-C sensor: this chip was not new when Fuji first started using it and they continued putting it in cameras for 4 years. Why? Because it was a bloody good sensor, that's why! :D Newer iterations of the X-Trans technology helped in keeping it competitive practically until today. The X-Pro2 is the first time Fuji uses a "brand new" sensor in a top camera. Is this a new trend? It remains to be seen.

 

I understand the worries about sensor technology surpassing what the company is currently committed in. Especially since Fuji has an history of keeping the same sensor for as long as they can. But I'm trying to be realistic here.

 

And objectively speaking, sensor technology in itself, hasn't improved overall image quality from our cameras as much as we like to think, during the last 8 years or so.

 

If you think about it, than old 16mp sensor was used in Fuji, Sony, Nikon and Pentax cameras for ages. Various versions of the 24mp sensor are also in use until today. In the world of FF, the same is true about the 24mp Sony sensor. Not to speak about Canon sensors at all...  In all cases what I see is modest improvements, generation after generation, with more substantial improvements in processors and "peripheral" components (e.g. elimination of the AA filter certainly resulted in a serious improvement in a number of cases).

 

I believe we shall only begin to see real advancements only with a new revolutionary sensor technology, not merely evolutionary steps. The organic sensor probably being such a case. BSI also seems a good step forwards, although it seems to be seriously important more for smaller sensors.

 

Back to the existing 50mp MF sensor: I totally trust that, should Fuji put a current technology X-Trans version of this sensor in their MF camera, it would virtually blow away anything south of the top of the line 100mp Phase One. Which means, for 99.9% of the population, everything. Coupled with Fuji lenses and general imaging know-how, this could keep it on top for a couple of years. But my point is, the 44x33 "format" will most probably keep improving. Sony themselves hinted that they might use such a sensor in a future compact (RX) MF camera. Digital backs will obviously use future 44x33 sensors for years to come. And, speaking of Fuji, it goes without saying that they would seek to implement their own R&D (i.e. organic sensor) in a (somewhat distant) future version.

 

Sure, I'd be delighted to see the latest technology used in my favorite cameras (although I might not be able to afford one). But I wouldn't mind Fuji using present day, tested technology either. I'm confident they'll be committed in supporting and improving it for as long as it takes.

Link to post
Share on other sites

(Note: there is no DxOMark test of this sensor, which is curious. In fact, there is a rumor circulating that DxO leaked, at some point, a score of 101 for it, but later removed the reference. This has almost become a conspiracy theory among Pentaxians, who claim that DxO burried the actual benchmarks. Take this with a grain of salt)

here you go: Pentax 645Z

Too bad we only got this within a comparison. I would have liked to see the FullSNR and other measurements

 

In conclusion: it is difficult to imagine Fuji releasing a premium priced MFD camera in 2017 with a three year old cropped non-BSI MFD sensor with "only" 50MP resolution, with the intention of keeping this model in production for several years ahead, because small format sensor technology is pressing ahead so rapidly.

 

What are you expecting - a BSI medium format sensor?

And why would Fuji need that, since they are just starting and won't cripple their mount. 

Overall medium format will stay ahead of full frame for quite some time, as full frame stays ahead of APS-C. There just isn't that much that technology can do anymore. The light itself is the main source of noise. 

 

So the medium format camera will deliver better images. The questions are: whom is it aimed at and what exactly is it capable of. 

Link to post
Share on other sites

There are two separate questions here - 44x33mm vs. larger sensor and the 50 MP sensor vs. something new with more pixels, and more importantly, newer technology. My vote is for 44x33 (anything larger means much larger lenses and higher prices), but NOT for the 50 MP sensor.

 

The 16 MP sensor that became the X-Trans was released in the Nikon D7000 in September 2010 (with availability somewhat later - DPReview's review came out in December, and I'm not sure how findable they were that holiday season), The X-Pro 1 was released in January 2012 - sure, there were parts of three years involved, but the sensor was 14 months old (if you count from real availability of the D7000). Fuji did keep the same sensor forever, but they introduced it when it was relatively new (and even more so with the X-Trans III sensor - that's a brand-new sensor, although at a resolution that's been around). In fact, the reason why a three year old sensor is unappealing to many is that Fuji keeps sensors for years.

 

I'm speculating that switching sensors is a bigger deal in an X-Trans camera than a Bayer camera. Sony may even make a standard Bayer design that they sell with the sensor, so the camera manufacturer doesn't have to source another part at all (and MANY cameras can share that part). Since X-Trans is unique, Fuji has to make or order a fairly complex part (yes, it's a glass plate with splotches of color on it, but they're splotches of color a few microns across, and that may not be easy to do - I'm not sure how you print color on glass at over 6000 dpi, especially with edge precision much higher than that). They can only make enough for their own needs. My suspicion is that the expense is largely in setting up a run of a particular filter, not in making each individual filter (so they kept churning out 16 MP X-Trans arrays for years).

 

If they go with the 50 MP sensor, I'd suspect that sensor will still be around 5 years later, when it's 8 years old. Even now, the 50 MP sensor underperforms the A7rII sensor in some ways (and may actually underperform the 24 MP X-Trans III sensor per pixel in some measurements - overall performance will be better because it's a larger, higher resolution sensor). By five years from now, sensors that outperform that one will be commonplace, and they will include not only full-frame sensors but also APS-C sensors (imagine if the 16 MP X-Trans had fallen behind Micro 43 and 1" sensors in performance during its long run - it never did, performing slightly better than the newest 20 MP Micro 43 sensor, and much better than any 1" sensor).

 

I agree with EyesUnclouded that the GX-Pro 1 should be 44x33mm, but I'd like to see a new sensor... Call it a 44x33mm shrink of the Phase One 100MP sensor, an enlarged A7rII sensor or a huge enlargement of the X-Trans III sensor - they all produce relatively similar results. Depending on which pixel pitch you use, you wind up with somewhere between a 67 mp (shrink of the 100 MP sensor) and a 96 MP (enlargement of X-Trans III) sensor using copper wiring, possibly BSI (A7rII) and possibly with 16-bit output (100 MP). I don't care whether it's 67, 72 or 96 MP, but I'd like to see Fuji start off with a sensor that uses the newest technological generation. This should be easy for Sony to make, and would also appeal to Phase One (who are trying to move to CMOS, but only have an underpixeled 50 MP back and a VERY expensive 100MP back), Hasselblad and Pentax. Pentax's newer lenses all depend on 44x33mm, so they have no interest in the 100 MP Phase One sensor (not to mention the price). Most of Hasselblad's lenses would work with the larger sensor (all except for a couple of wide angles), but they seem to be trying to undercut Phase One, and may not like the price of the 100 MP sensor.

 

Since Sony has the technology from three other sensors, I can't see that sensor being much more expensive than the 50 MP sensor that shares technology with an older generation of smaller-format sensors. By 5 years from now, the 50 MP sensor may actually be more expensive than a newer alternative, as Sony transitions all of their sensor manufacturing to copper. Right now, Sony makes a mix of old and new sensor types (ignoring sensors smaller than APS-C) - they have three sizes of current-generation sensor (24 MP APS-C X-Trans III and A6300, 42.4 MP A7rII and 100 MP Phase One - there may also be a Super 35 version in a movie camera or two, and there are certainly 1" versions ), but ALSO make at least four or five older-generation sensors (20 MP APS-C in the A5100, 24 MP APS-C in the A6000 and other cameras, both 24 and 36 MP full-frame, and the 50 MP 44x33mm sensor). Five years from now, will any of those still be in production? If the 50 MP sensor is an "orphan", its price may go up while Fuji still wants to use it.

Link to post
Share on other sites

The next thing that should be kept in mind about 44 x 33mm sensors is that the claim made by Phase One about the cropped MFD format sensor being over 60% larger than a full frame small format sensor is a highly misleading marketing ploy. Pentax should be commended for not playing this outrageous game, but there sadly remains a difference between meaningful real world sizes and exaggerated marketing sizes. A cropped MFD sensor may be 60% larger by surface area, but resolution is a linear function, which means we must compare format sizes in terms of linear size difference. The true increase in linear size of a cropped MFD sensor over FF small format is more like 20%.

 

In addition to not strictly speaking being medium format at all, cropped MFD sensors represent too small an increase in linear size to be meaningfully differentiated from FF small format sensors. Given that sensor development is driven by smaller formats, the pace of development of small formats tends to additionally outstrip that of MFD, thus further tending to rapidly negate this 20% increase in real world linear size difference, as MFD sensor development lags behind.

 

Another way of putting it is that the improvement in IQ from newer smaller format sensors may be such that there may be ever dwindling yield from increasing sensor size in order to get cost-effective and cost-meaningful improvements in IQ, unless MFD sensors can come down in price while keeping pace in their technological development with smaller format sensors. I am sure Fuji are aware of this and are furiously having board room debates over it, just like we are.

 

You over-simplify in a few ways.

 

The size comparison is tricky, especially since one format uses a 4:3 aspect ratio and the other uses 3:2. If you prefer the 3:2 aspect ratio and would crop the mini-MF images to get it, the comparison looks less favorable than if you prefer 4:3 and currently crop your 3:2 images — which is what I do. From the former perspective moving to 33mm x 44mm means that you would have to "throw away pixels," and that makes the option seem less appealing. However, from my point of view, I am currently throwing away pixels with FF or cropped APS-C and I would get to keep all of them with 33mm x 44mm. (From that point of view, considering the actual used pixels in the two formats the crop factor of FF compared to MF looks like 1.9x!)

 

Sensor pixel resolution is not the only issue, so even thinking that "50MP is 50MP" is not quite the whole story. The system resolution of a larger format sensor is better than that of a smaller sensor system and can benefit more from excellent quality lenses. The effects of diffraction blur come on at different apertures as well on the two formats. There are differences in dynamic range and noise response, too, along with depth of field.

 

While all of those are potential benefits of the larger format, there are also some minuses. For example, coverage of a given angle-of-view will require a longer — and likely heavier and more expensive — lens. Certainly types of lenses are less readily available. The entire system is larger and heavier, and it is probably pricey.

 

The question will really come down to the preferences and needs of individual photographers. For some, the 1.5x cropped sensor Fujifilm format is ideal. For others a mini-MF system could be ideal.

 

Dan

Edited by gdanmitchell
Link to post
Share on other sites

The question will really come down to the preferences and needs of individual photographers. For some, the 1.5x cropped sensor Fujifilm format is ideal. For others a mini-MF system could be ideal.

 

 

Exactly right. Not to mention potential Fuji MF users would be a very unique group of people: not the usual MF crowd (but many would be tempted perhaps), but also high-end mirrorless and DSLR pros, people in more diverse photographic disciplines than "MF-proper" in general.

 

This is what I'm looking forward: a unique system, such as the X-System was unique in many ways.

 

 

here you go: Pentax 645Z

Too bad we only got this within a comparison. I would have liked to see the FullSNR and other measurements

 

 

I hadn't seen the exact numbers, thanks mate.

Link to post
Share on other sites

  • 2 weeks later...

Here is something interesting:

 

http://pentaxrumors.com/2016/03/17/the-next-pentax-645-medium-format-camera-will-most-liekly-have-a-100mp-sony-sensor/

 

Just as I predicated, the Pentax 645 system is full frame ready, which is why the body is quite big. It looks like the next iteration of the Pentax 645 will have the Sony 100MP full frame MFD sensor in it. 

 

My own preference is for full frame, but I do see there is an equally cogent argument to be made in favour of a dedicated cropped MFD system, with lenses specifically dedicated to this sensor size just as in the X-system. The pros would include:

 

1. Smaller body especially if it is mirrorless, and hence more portable. Less of a studio camera and more practical. 

2. Smaller lenses (lighter and less expensive), or else faster lenses (e.g. f/1.8) more competitive with full frame small format

3. Smaller sensors are catching up with larger formats in their performance making it increasingly unnecessary to resort to brute increases in sensor size to improve IQ. This will become more marked after the release of the organic sensor, and in years to come. What would be the point of full frame MFD once smaller format sensors all reach the 100-120MP theoretical limit of current lens resolution?

4. A dedicated cropped medium format system is a good complement to the cropped small format X-system

5. May appeal to a wider market due to portability and relative affordability

 

If Fuji go for a cropped MFD system, they should wait until Sony upgrade their 44x33 sensor to around 80MP. Sony sensors will likely fairly soon bump up their sensor resolution across the board so that no sensor format has the same/higher maximum resolution than the sensor size one step above it. So if Sony sensors increases their full frame small format sensor resolution to around 50MP, they would have to also bump up the resolution of the 44x33 sensor to around 80MP. I am sure Sony cameras are itching to bump up the resolution of their full frame small format cameras up to around 50-60MP, but Sony sensors will only release such a sensor after concurrently bumping the resolution of the format just above it (i.e. 44x33) up to around 80MP. If Fuji do release a dedicated cropped medium format system, they should release it straight after Sony upgrades the 44x33 cropped MFD sensor. They should not use the 51MP version currently used in the Pentax 645Z, as this is likely imminently about to be rendered obsolete (in both performance and resolution) by an impending upgrade to the full frame small format sensor for the next generation of Sony full frame small format cameras.

 

Cons of a cropped MFD format system are:

 

1. Insufficiently differentiated from full frame small format (especially given Sony's trend towards incorporating newer innovations such as BSI and copper wiring sooner into small format sensors than into MFD sensors—clearly Sony invests more R&D into small format sensor development)

2. Might look like the full frame Pentax 645's sad little brother

3. Arguably not really medium format. Once Pentax brings out full frame 645s their advertising department would never stop reminding us of this fact.

4. Neither fish nor fowl. Too small to be genuine MFD, too big to be 135 format. It will fall between the cracks in the marketplace. Is 44x33 a format with a future?

 

The thing that will decide the way Fuji go are probably practical ones:

 

1. Whether Panasonic are willing to cooperate in manufacturing an organic 44x33 sensor in future (Panasonic have a relationship with Leica whose S System is a 44x33 cropped MFD system so maybe Leica might come aboard as a future sensor buying client; although once again the S System may also be full frame ready like the Pentax 645 system)

2. What market research dictates as to relative demands for a cropped vs full frame mirrorless MFD system (this is probably the most decisive factor of them all)

 

Either way, irrespective of whether Fuji choose to make a cropped or full frame MFD system, we all know that they make extremely thoughtful, measured and insightful choices when determining the critical foundation structural characteristics of a system, characteristics that will define the boundaries of the system for decades to come. Whatever they choose to create, it will likely be something quite remarkable. 

Edited by Sator-Photography
Link to post
Share on other sites

Just more thoughts about how things might look in around 36 months from now.

 

Canon: are bringing out the 5Ds III with a 120MP sensor, similar in price to the 5Ds Mark I. They are bringing out high resolution glass advertised as 120MP ready. The rumours mills are full of talk of how Canon will keep the resolution at a maximum of 120MP but concentrate on other performance factors such as dynamic range, and a new Canon layered sensor technology similar to the Foveon is also talked about.

 

Pentax: due to a fall in price of MFD sensor manufacture, they have just announced another price drop in their Pentax 645 100MP full frame MFD camera to keep pricing competitive with the Canon 120MP 5Ds III. By now they have released a couple more MFD lenses, and their 645 lens line-up is looking increasingly more mature. The rumours mills are full of talk of a Pentax 645 with the new 120MP Sony full frame BSI and copper wired MFD sensor with remarkable high ISO performance and dynamic range which Phase One has recently started using. Pentax also offer more affordable entry level 80MP cropped sensor MFD bodies alongside their premium full frame flagship, successors to the 645Z that look tempting.

 

Sony: the a9II and/or a99RII are released with a 60MP full frame small format BSI copper wiring sensor plus other much touted new features. It can capture usable images in the near dark, the a99RII especially has an exceptional frame rate, and they can capture +4K video. The dynamic range is cutting edge. The lens lineup for both A mount and E mount are looking increasingly mature, making Sony the new Minolta. Sony displace Nikon from second place in the market, and now set their sights on beating Canon. There is internet discussion about whether Sony will follow Canon in producing a 100-120MP full frame small format sensor. 

 

Sigma: have now increasingly started to make genuine inroads in the implementation of the Foveon sensor. A small format Foveon camera now stands better comparison in performance and ergonomics with other industry leaders.

 

Panasonic: release an organic sensor M4/3 camera. The IQ for both still and videos is remarkable for such a small body, and this rejuvenates interest in the M4/3 platform. The rumour mills chatter about the Fuji-Panasonic consortium selling organic sensors to Leica.

 

Fuji: the X-Pro3 is released with a 36MP organic sensor, and astonishingly high signal/noise ratios. It has remarkable high ISO performance, and dynamic range that almost stands comparison with some Sony MFD sensors. Scenario A: Fuji also announce a competitively priced GX-Pro1 along with three lenses for it, with the 44x33 Sony 51MP X-trans sensor, but its performance in other aspects such as dynamic range and ISO speed are eclipsed by the new a9II/a99RII, which also have a nominally higher resolution (Sony fanboys laugh, but diehard Fujsters insist that the X-trans sensor makes it more like a 70-80MP or so sensor with IQ competitive in price/performance to entry level cropped sensor 80MP Pentax 44x33 models). Scenario B: Fuji announce the GX-Pro1 with a Sony 80MP 44x33 X-trans sensor, which is now about 1-2 years old (with claims to the X-trans sensor giving it a resolution comparable to 100-120MP MFD sensors). Scenario C: the Fuji rumour mills are full of credibly sourced leaks stating that they will soon release the GX-Pro1 with an organic 44x33 sensor soon after the X-Pro3, making the body dramatically more compact than Pentax MFD models, but with remarkable sensor performance, and there is discussion about a full frame variant coming later.

Edited by Sator-Photography
Link to post
Share on other sites

Just more thoughts about how things might look in around 36 months from now.

 

Canon: are bringing out the 5Ds III with a 120MP sensor, similar in price to the 5Ds Mark I. They are bringing out high resolution glass advertised as 120MP ready. The rumours mills are full of talk of how Canon will keep the resolution at a maximum of 120MP but concentrate on other performance factors such as dynamic range, and a new Canon layered sensor technology similar to the Foveon is also talked about.

 

Pentax: due to a fall in price of MFD sensor manufacture, they have just announced another price drop in their Pentax 645 100MP full frame MFD camera to keep pricing competitive with the Canon 120MP 5Ds III. By now they have released a couple more MFD lenses, and their 645 lens line-up is looking increasingly more mature. The rumours mills are full of talk of a Pentax 645 with the new 120MP Sony full frame BSI and copper wired MFD sensor with remarkable high ISO performance and dynamic range which Phase One has recently started using. Pentax also offer more affordable entry level 80MP cropped sensor MFD bodies alongside their premium full frame flagship, successors to the 645Z that look tempting.

 

Sony: the a9II and/or a99RII are released with a 60MP full frame small format BSI copper wiring sensor plus other much touted new features. It can capture usable images in the near dark, the a99RII especially has an exceptional frame rate, and they can capture +4K video. The dynamic range is cutting edge. The lens lineup for both A mount and E mount are looking increasingly mature, making Sony the new Minolta. Sony displace Nikon from second place in the market, and now set their sights on beating Canon. There is internet discussion about whether Sony will follow Canon in producing a 100-120MP full frame small format sensor. 

 

Sigma: have now increasingly started to make genuine inroads in the implementation of the Foveon sensor. A small format Foveon camera now stands better comparison in performance and ergonomics with other industry leaders.

 

Panasonic: release an organic sensor M4/3 camera. The IQ for both still and videos is remarkable for such a small body, and this rejuvenates interest in the M4/3 platform. The rumour mills chatter about the Fuji-Panasonic consortium selling organic sensors to Leica.

 

Fuji: the X-Pro3 is released with a 36MP organic sensor, and astonishingly high signal/noise ratios. It has remarkable high ISO performance, and dynamic range that almost stands comparison with some Sony MFD sensors. Scenario A: Fuji also announce a competitively priced GX-Pro1 along with three lenses for it, with the 44x33 Sony 51MP X-trans sensor, but its performance in other aspects such as dynamic range and ISO speed are eclipsed by the new a9II/a99RII, which also have a nominally higher resolution (Sony fanboys laugh, but diehard Fujsters insist that the X-trans sensor makes it more like a 70-80MP or so sensor with IQ competitive in price/performance to entry level cropped sensor 80MP Pentax 44x33 models). Scenario B: Fuji announce the GX-Pro1 with a Sony 80MP 44x33 X-trans sensor, which is now about 1-2 years old (with claims to the X-trans sensor giving it a resolution comparable to 100-120MP MFD sensors). Scenario C: the Fuji rumour mills are full of credibly sourced leaks stating that they will soon release the GX-Pro1 with an organic 44x33 sensor soon after the X-Pro3, making the body dramatically more compact than Pentax MFD models, but with remarkable sensor performance, and there is discussion about a full frame variant coming later.

 

Oh, you optimist. Dream on... ;-)

Link to post
Share on other sites

  • 3 months later...

I am hoping that Fuji will release three primes with their system, and that one of these primes will be a 24mm f/2.8 or f/3.2 wide angle lens. The reason is that it will throw down the gauntlet to Hasselblad to come up with a similar wide angle lens for the X1D. The problem is that Hasselblad will probably be unable to do so, just as Sony will be unable to come up with an answer to the rumoured Fuji 8-16mm f/2.8 zoom lens. It will highlight the inadequacies of the Hasselblad system right from launch.

 

Everyone is all gushing over how thin and small the Hasselblad X1D body is. I have seen pictures showing how their medium format body is comparable in size to a Fuji X-Pro or a micro 4/3 body. Rather than celebrate, there is good reason to be horrified that they have likely severely crippled their system from birth. It is thought that the Hasselblad X1D has a flange distance of around 20mm (or possibly even less). Compare this with other systems:

 

Sony FE (full frame): 18mm

Canon EF-M (APS-C): 18mm

Fuji X (APS-C): 17.7mm

Micro 4/3 (MFT): 19.25mm

 

The Hasselblad has a flange distance more like a micro 4/3 mount or APS-C mount. Yet it has to cover a sensor that is dramatically larger:

 

sensor1-Hasselblad-X1D_zpsuyfgfs2l.png

 

This means that the angle of light incidence at the corners of the sensor will become too steep, resulting in degradation of image quality. 

 

Once again, I will point out this diagram:

 

flange-focal-distance-and-angle-of-incid

 

A full frame DSLR has a flange distance around 44mm (top). An APS-C mirrorless mount (bottom) should have a flange distance of around 18mm, which results in an angle of light incidence similar to the full frame DSLR. The middle diagram shows what happens when you maintain an APS-C flange distance of 18mm, but increase the sensor size from ASP-C to full frame, as Sony have done: the angle of incidence of light at the sensor corners becomes excessively steep. 

 

The Hasselblad X1D with a flange distance more like a micro 4/3 mount will have this same problem of an excessively steep angle of light incidence at the corners. 

 

This degradation of image quality from this steep angle of incidence in the corners becomes more marked with ultra wide angle lenses. It is questionable whether the Hasselblad X1D is even capable of supporting the development of a 24mm f/2.8 or f/3.2 wide angle lens, as the drop in corner IQ starts to become unacceptable. It is a silly move because having a shorter flange distance does little to nothing to help reduce the size of medium format lenses. In fact, on fast lenses, it may cause the lens size to increase to overcome the drop in corner acutance. Also comparing a SLR lens vs a mirrorless lens, if the lens design is similar, the length from the front of the lens to the sensor has to be the same—if you shorten the flange distance, this makes the lens longer. You also end up with a ridiculously oversized medium format lens imbalanced against a puny body. 

 

What Hasselblad have done is similar to Sony, and this likely reflects the fact that Hasselblad have prioritised compactness over image quality. Fuji have repeatedly raised the issue of angle of light incidence degrading image quality, as well as repeatedly stating that they prefer to prioritise image quality rather than make makeshift compromises. I hope Fuji do not go down the same track as Hasselblad and instead come out with a 24mm f/2.8 prime lens to showcase the ability of their system to support lens development across all focal lengths. That would mean a proportional increase in the flange distance appropriate to a medium format mount, but it would emphatically highlight the fact that the mount is intended for long term development rather than the short lived advertising value of having an emaciated smaller body.

 

After all, the smallness of the Hasseblad X1D will not last long. Virtually all mirrorless bodies have sequentially grown in size since their initial appearance. Hasselblad is merely a couple of generations behind in the game. The X1D only shoots at 2.3 fps and has one focus point! If you add dual processors to overcome that issue, the body has to be bigger to accommodate that and to allow for better heat sinking as well as to accommodate for a larger battery to power it. Next add a tilt screen, an optical viewfinder, or IBIS and you can resoundingly kiss goodbye to its current diminutive stature. 

 

On the subject of IBIS, Fuji have said that they did not design the X mount to accommodate for IBIS, and thus refuse to add it retrospectively as it would degrade image quality. They also say that if you want uncompromised IQ with an IBIS mount system it necessitates an increase in the mount and lens size to allow room for the sensor to move. I hope that Fuji do make the proportions of their mirrorless medium format mount larger to accommodate IBIS in future (even if it isn't present at launch). The reason is that in future both full frame and medium format sensors are going to see a rapid increase in resolution up to 120MP. It's all very well a medium format mirrorless camera being more compact to carry around with you, but without IBIS it will be so sensitive to minute handshake that it would end up a tripod only studio model. It might not be a big problem yet with a 51MP sensor but it isn't going to be long before it becomes a major headache. Again, I would prefer a future proof system designed for the long-term over gross compromises of flashy short-term advertising value only. 

 

As for any Hasselblad fanboys who are going to get themselves all upset and insist that the X1D can support a full range of focal lengths without problems to them I say this: good! Let Hasselblad prove me wrong by producing a 24mm f/2.8 lens immediately. I am only too happy to have my grave concerns about their system decisively put to rest.

Link to post
Share on other sites

I think that there are a lot of questions about the Hasselblad system. Hasselblad gets a lot of credibility based on their history of being a leader in MF film and then making some fine MF digital bodies. There is lots of optimism that someone can produce a MF digital system that is priced in a range not too far above the most expensive full frame systems, but when folks think of that they are thinking of the whole package, not just a sensor. 

 

The Hasselblad has a very limited focus system, and Hasselblad suggests that it will be nearly impossible to build zoom lenses for the camera. (Check their FB page for more interesting information.) Most people desiring MF are thinking of tripod-based work, but Hasselblad seems to want to suggest this is some kind of handheld camera, perhaps even something that you might use for things like street photography.

Lots of us are very interested in this camera and in what sort of system might evolve around it, but there are a ton of questions yet to be answered.

Link to post
Share on other sites

Second Thoughts on the Hasselblad X1D


 


I am beginning to think that this new system with a flange distance more like an micro 4/3 mount is deliberately built so that it compromises on acutance in exchange for portability. The design team might well have decided to prioritise compactness and decided to accept the limitations it would impose on image quality.


 


It means that they are probably going to only issue slowish primes. The reason is that the steep angle of light incidence compromises corner IQ. This may only be of the order of some percentage, but this becomes proportionately more marked as the maximum aperture is increased. This is probably the primary reason why Sony FE mount lenses seem to mostly be limited to f/1.8, where high-end primes would normally have a maximum aperture of f/1.4. Examples:


 


Sony-Zeiss 55mm f/1.8


Batis 85mm f/1.8


 


The moment the maximum aperture is increased to f/1.4, the lens size suddenly increases disproportionately e.g. the Sony-Zeiss 35mm f/1.4 and the Sony GM 85mm f/1.4. The reason is that the f/1.4 aperture exposes the corner acutance issues from the excessively short flange distance, and to overcome this they have to increase the lens size. However, despite being so huge and expensive, the MTF plots only just manage to equal the average performance of an equivalent DSLR f/1.4 lens. 


 


The same thing is clearly happening with the Hasselblad X1D system. You can see it when you compare the aperture speeds with lenses from the Leica S system (also with a cropped sensor). 


 


Hasselblad 45mm f/3.5


Leica 45mm f/2.8


 


Hasselblad 90mm f/3.2


Leica 100mm f/2.0


 


Hasselblad 30mm f/3.5


Leica 30mm f/2.0


 


The Hasselblad lenses are significantly slower.


 


If challenged as to why their lenses are so slow, Hasselblad would reply that it is to improve the portability of the system. And that would be a correct and honest answer in some respects. However, the more in-depth answer is that they have traded acutance for compactness, and that if they did increase the maximum aperture it would expose the corner problems.


 


That means if you shoot medium format for ultimate image quality with shallow depth of field then this system isn't for you. If optimal IQ is what you want, you still need to choose their more high-end H system. That is probably exactly Hasselblad's intention: product differentiation. I think that is a fair enough a strategy, but you need to take this into account before buying into their system. 


 


Fuji, on the other hand, needn't worry about differentiating their medium format system from a pre-existing system since they have ceased production of medium format film cameras. Fuji need not make these compromises that Hasselblad have made, and should be able to focus on producing a system focused on optimal image quality and balanced overall performance.


Link to post
Share on other sites

Any thoughts about shutter type? If Fuji lens will not have leaf shutter it would make system much cheaper and lenses can be potentially faster but probably less attractive for some professionals.

Edited by flam
Link to post
Share on other sites

I could only find one shot of the back of a Hasselblad lens and it is flatly lit, so no universal truths are implied. However, it looks in this picture that the shallow flange distance is overcome by keeping the optics well forward in the mount. The rear element seems to be some distance from the actual bayonet.

 

_DSC0913.jpeg

Link to post
Share on other sites

I guess what you are saying is that if you increase the distance of the point of convergence from the sensor that you can reduce the angle of corner incidence on the sensor corners. It may mean that the flange distance is increased inside of the lens by placing the optic forward within the lens itself. There is no official word as to what the X1D flange distance is.

 

There are some who claim there are "easy fixes" to the problem of corner angle of incidence from a short flange distance. While there are likely simple ways to partially offset the problem, I am exceedingly sceptical of these quick fixes being total solutions that renders this a non-issue. Manufacturers show every evidence of continuing to regard this a very serious problem indeed. Sony, Canon, Toshiba, and Konica-Minolta have all gone as far as to have patented curved sensor designs to reduce the corner angle of incidence. If they all thought the problem could be so readily solved with a minor tweak like increasing the point of convergence, why go to all of the trouble of funding such R&D? Fuji too have continued to state in interviews that they view degradation of image quality from a steep angle of corner light incidence to be a significant ongoing technical challenge.

 

There is also evidence that Leica take the problem so seriously that they have foregone Sony to instead source sensors from their Belgium sensor maker CMOSIS. CMOSIS makes sensors designed to reduce the degradation of IQ from a steep angle of light incidence. Why is that relevant to Leica? Simple: because rangefinders are "mirrorless" cameras. The Leica M mount has a 27.8mm flange distance, much shorter than a DSLR. Reference source:

 

http://www.the.me/the-leica-m-max-sensor-explained/?PageSpeed=noscript

 

It may have also given Leica the confidence to make a full frame mirrorless SL mount with a 19mm flange distance (some have suggested that telecentric lens designs help too). 

 

In short, it is hardly looking like the issue of steep corner angle of light incidence is going away that easily as is occasionally claimed by Sony fanboys religiously defending the APS-C flange distance of the full frame FE mount. They then bury their heads in the sand after triumphantly proclaiming that the problem no longer exists.

 

While Sony have started a fashion towards reducing the flange distance on mirrorless mounts to the point that we are seeing full frame and cropped medium format mounts being introduced with flange distances as short as the micro 4/3 mount, I suspect that no consensus amongst optical engineer exists as to whether this is optimal practice. While there are pros to a short flange distance too, I suspect that when Fuji introduces their cropped medium format mirrorless mount, and Olympus introduces their full frame mirrorless mount, that they may well buck this trend and proportionately increase the flange distance.

 

In short, unlike DSLR mounts, which have been around since the 1930s, new mirrorless mounts are all still unproven experimental mounts in a field for which the optimal optical formulas remain uncertain. When we buy into a new mirrorless mount we are all crowdfunding research without guarantee of a favourable outcome. Some of these experiments are probably going to end in various degrees of failure, and we cannot assume that every new mirrorless mount is going to survive. It is particularly concerning when things look rushed e.g. turning an APS-C into a FF mount, or rushing headlong into the full frame/medium format mirrorless market without first cutting one's mustard gaining experience in M4/3 and APS-C mirrorless. That's why I cringed when I read one exuberant comment about how since the Hasselblad X1D costs sub-10K you should "just buy it" because "it's only money". The flipside of that is, of course, that a fool and his money are soon parted.

Edited by Sator-Photography
Link to post
Share on other sites

I can only think of one reason that moving the rear element forward would be worse than a natively longer flange distance, and that's simply the amount of leverage being applied to the mount, and if we're only talking about a difference of a few centimeters, it's still practically irrelevant.

There's nothing wrong with having a short flange distance, but that doesn't mean a 20mm Pancake lens on a Hasselblad will actually work.

In practice they may as well have gone with a 30-40mm flange distance, but if the lens designers think that they can pull off something that works in the sub 30mm range at some point then it dosn't make sense to design the mount to prevent that entirely.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Similar Content

  • Posts

    • Looking more like the next Fuji camera we see will be an X-T50 but it is still a rumor and no specs have been published.
    • Hey guys, The shutter on/off button switch may be loose on mine. After turning camera on and pressing the shutter/holding close to the area, the camera will turn off and say 'sensor cleaning'. This doesn't happen if I'm shooting via touch screen at all. Everything else is functional. Anyone else experience this before? Would love a much cheaper fix since Fuji Canada just quoted me $700 CAD to fix it, and considering everything else is functional except that part I'm not even sure why MPMB Main Board parts is being replaced😕 I got no explanation from them either.
    • As far as I know the firmware is not country specific. Are you sure that the filename has not been changed ( I am told this can happen with mac os). That's the only thing I can think of.
    • My x-t5 does not exhibit the focusing switch behaviour as you report it, so that is very strange and indicative of a fault. It does not matter whether the flash is attached or not. Once you set the camera for your studio flash, say 1/250th at f5.6, the camera, which is showing you what you will get at that exposure without the flash, will show a black screen unless the ambient light is brighter than what you would typically get indoors. That is why, as Jerry says, you have to set preview exp/wb to off. I have set a button for this.
    • I connected to FRAME.IO a while back and it works fine, but the camera wouldn't connect to the internet all of a sudden today and would get stuck on the reset screen, including initializing and even switching USB Connection mode. Is anyone else experiencing the same thing?
×
×
  • Create New...