Phase One IQ3 100mp. Shipping now. Full frame CMOS. HDMI.

[Jorgen Udvang]

This is so far out of reach for me that I won't even speculate over what I could sell to finance it. However, this is exactly what medium format needs. Real sensor development both with regards to size, resolution, DR and all the rest. Hopefully, enough photographers can afford this beast to keep MF alive and weel. Congratulations to all photographers!

 

[ptomsu]

This is finally the MFDB which many have been waiting for - real MF-FF - KUDOS to Phase One!

Having said that this is built as a money earning machine, especially with that price tag. So better you can earn that money within the first 2 years with the projects you realize with that back, or you need to be very wealthy to afford it.

I am definitely not in this game anymore, but still interested in results and the further development of digital MF photography.

 

[miska]

Do you think there will ever be a CMOS FF medium format chip with fewer pixels ?
Phase One's pricing policy by the Mpix makes the 100 Mpix version just crazy expensive. Or perhaps Hassi will make a CFV at a more "reasonable" price (20k-30k Euros ?).

 

[ErikKaffehr]

Hi,

No, I don't think so. Why not? Because it would need another sensor design, meaning a lot of extra effort.

I would guess that Phase One will make an "affordable" version with some less features like the IQ-250/IQ-150 pairing.

Best regards
Erik

Do you think there will ever be a CMOS FF medium format chip with fewer pixels ?
Phase One's pricing policy by the Mpix makes the 100 Mpix version just crazy expensive. Or perhaps Hassi will make a CFV at a more "reasonable" price (20k-30k Euros ?).

 

[ErikKaffehr]

Hi Anders,

Bart van der Wolf calculated the DR based on the raw sample he got from DigitalTransitions and found that DR was 13.65 (on the pixel level), but he did not know if it was 14 or 16 bit mode. Anyway 13.65 would put it on the same level as Nikon D810 and quite a bit above the A7rII (12.69 if I remember correctly). But Nikon and Pentax used to make better use of Sony's sensors than Sony themselves, or at least so it seems. Bart's figure seem pretty credible to me. Pulling another 1.4 EV out of the hat just doesn't seem very likely.

http://forum.luminous-landscape.com/index.php?topic=106969.msg880768#msg880768

I guess the great news is that Phase One now has a state of the art CMOS sensor with excellent live view that fills the 645 format. Should make view camera users happy, if it works with the wide angle lenses they have.

Now, getting back to the DxO scaling of DR makes a lot of sense. MFD users of course always mistrusted DxO, but that may change once DxO presents data for the new backs. If they do, that is.

Best regards
Erik

BSI is not all good, there are some challenges with it too related to manufacturing, electrical crosstalk, pixel non-uniformity etc. The A7r-II (BSI) seems to have (very) slightly worse DR than the older A7r. I think the main advantage of BSI so far is good high ISO, not better DR at base ISO.

I don't think this new sensor is a BSI sensor, but it's just a pure guess. We'll find out at some point. I don't think it's actual 15 stops either (DxOMark per pixel definition, but indeed downsampled 8MP definition they will probably do it), something also time will tell.

It's surely 100 MP and 645 fullframe and all the usual CMOS advantages though, and I think it's so revolutionary that I would not be surprised if the IQ3 is the last range with CCD sensors in them, because who's going to want to buy a new 80MP CCD now with this new king of the hill?

 

[ErikKaffehr]

Hi,

Yes great news, I think. But a lot of questions unanswered, like how it works with wide angles. The 15 bit of DR speaks a bit against BSI as it is said that BSI is a bit problematic with regard to DR. Indeed the Sony A7r has a bit better DR than the A7rII. BSI gives about 1/3 of a stop improvement in ISO, so it is not such a great step.

Bart measured DR on the image he got from DigitalTransaction: http://forum.luminous-landscape.com/index.php?topic=106969.msg880768#msg880768

He got 13.65EV, in line with the Nikon D810 sensor. A state of the art figure.

The figure that they use may be the scaled value DxO-mark is using. DxO has two values, screen and print. Print is based on 8MP.

So you divide 100/8 -> 12.5. Take square root of 12.5 -> 3.5. Take log(3.5) / log(2) and get 1.8 EV. You add that 1.8 EV to Bart's mesaured data and end up with a normalised DR of 15.4 EV. No doubt that would put the IQ3 100MP on top of DxO-mark, well ahead Pentax 645Z at 101 points.

On the other hand, I don't think it really matters. Very difficult to utilise that DR is.

The great news I think that there is a full frame solution for real time viewing for technical camera users. Hopefully with peaking which is a godsend when finding optimal tilt.

I think it is a great solution and would buy immediately if I could find the necessary change!

Best regards
Erik

I think 15 stops of DR with clean higher ISO's makes BSI a pretty good bet.

Any way you slice it, this is a(nother) breakthrough piece of imaging equipment -- kudos to Phase!

 

[torger]

Do you think there will ever be a CMOS FF medium format chip with fewer pixels ?
Phase One's pricing policy by the Mpix makes the 100 Mpix version just crazy expensive. Or perhaps Hassi will make a CFV at a more "reasonable" price (20k-30k Euros ?).

No I don't think so, because it won't be a product that sells well. Megapixel is king, and you can easily see that on the second hand market, when Canon launched their 50MP cameras and Sony with 42MP the 33-40MP CCD backs suddenly got considerably harder to sell.

While there is a small group of users that have MFD due to MFD features like a large viewfinder, leaf shutters, tech camera movements or whatever, the number one reason people upgrade from 135 digital to MFD digital is to get better image quality and then you want more pixels. The users that don't care that much about resolution or the latest DR numbers are often using older gear.

And indeed if you're prepared to buy second hand you can get away quite "cheap", and I'm sure the second hand market will improve a lot for higher end CCDs with this new release. In the short term the higher end CCDs are protected by the ultra-high price of this new CMOS, but I'm sure there will be more reasonably priced products within a year or so.

Phase One is the brand to have if you want the latest highest end image quality first. If you have the patience to lag a year there's more economical alternatives.

 

[Hulyss Bowman]

Do you think there will ever be a CMOS FF medium format chip with fewer pixels ?
Phase One's pricing policy by the Mpix makes the 100 Mpix version just crazy expensive. Or perhaps Hassi will make a CFV at a more "reasonable" price (20k-30k Euros ?).

I am sensitive about this subject. Your wish is very actual and legitimate imho. I'd rather want a 16bit 22MP "FF" CMOS back around the price of a D4 or 1Dx. Just the back; so between 5.5 to 7 K€.

At this price they would sell a lot, even 22 MP backs. The "look" of a larger sensor is often more important than the pixel number...

 

[gerald.d]

Phase One's pricing policy by the Mpix makes the 100 Mpix version just crazy expensive.

Back - Year of Introduction - Price

P65+ - 2009 - $40k (not sure if that included the value added warranty)
IQ180 - 2011 - $44K
IQ280 - 2013 - $44K
IQ3 80 - 2015 - $44K
IQ3 100 - 2016 - $44K

Seems like good value when you look at it like that...

 

[Hulyss Bowman]

No I don't think so, because it won't be a product that sells well. Megapixel is king, and you can easily see that on the second hand market, when Canon launched their 50MP cameras and Sony with 42MP the 33-40MP CCD backs suddenly got considerably harder to sell.

While there is a small group of users that have MFD due to MFD features like a large viewfinder, leaf shutters, tech camera movements or whatever, the number one reason people upgrade from 135 digital to MFD digital is to get better image quality and then you want more pixels. The users that don't care that much about resolution or the latest DR numbers are often using older gear.

And indeed if you're prepared to buy second hand you can get away quite "cheap", and I'm sure the second hand market will improve a lot for higher end CCDs with this new release. In the short term the higher end CCDs are protected by the ultra-high price of this new CMOS, but I'm sure there will be more reasonably priced products within a year or so.

Phase One is the brand to have if you want the latest highest end image quality first. If you have the patience to lag a year there's more economical alternatives.

The old backs got harder to sell because they are USED and old tech compared to a modern 35 mm camera. To use a DM22 today it a bit complicated... A new low pix count CMOS modern back with tremendous video possibility can be a major seller because MF video is a market too.

 

[torger]

The old backs were used even when the 135 cameras where 24 megapixel tops, no the problem is that pixel count and DR numbers are not competitive any longer, and pixel count and sharpness is the number one reason that sells MFD, not "look". There are exceptions of course (I'm personally one of them), but they're too few to make business.

A low MP digital back just won't fly, and for the record it will probably not be much cheaper to manufacture either, but rather the opposite as they would need to develop some big pixel tech specifically for that rather than relying on their 135 pixel tech and just enlarge the surface.

To make it cheaper make it smaller size. 44x33mm will be the cheap alternative now when 54x41mm is here at last.

This is a megapixel race wether we like it or not, and it will continue to be that for quite some time...

I think low pixel count is a bad idea due to aliasing issues, however I think lens look/weight/cost vs resolving power is a good tradeoff to make, but also there we see that there's only one parameter that really counts for the big business -- resolving power.

 

[MrSmith]

the sensor looks great for those looking for huge file sizes, and presuming it’s just a bigger A7rII chip the image quality is already known to be very good.
kudos to Sony for producing a 645 sensor, shame they didn’t go full frame but that’s unlikely to ever happen due to the last 10 years of lens design based on a crop format.
look forward to seeing the hassleblad and pentax ‘value’ versions that will undoubtably follow as competition is always good.

photographically/fiscally it’s a non-starter for my business as it would not increase revenue or bring in any new clients but i’m sure in the right circles it will be a big seller. :thumbup:

 

[GrahamWelland]

photographically/fiscally it’s a non-starter for my business as it would not increase revenue or bring in any new clients but i’m sure in the right circles it will be a big seller. :thumbup:

Sometimes, if you can afford it, there's nothing to justify about getting the best. That way the only limiting factor is you and there is absolutely no opportunity to blame the equipment. Heck it's why some people drive Porsches/BMWs/Cadillacs/Lexus's vs Fords & Chevys; they all get between A and B. For a working pro however, real world economics of business vs hobby/obsession/work distraction come into play and that's a completely different ROI justification. (And why you might drive a panel van or F350 extra cab vs a Mercedes SUV).

If I can fund it, an IQ3 100 is definitely on my radar at some point even if reluctantly not immediately. (and as I believe I may have posted before, I'm still in the honeymoon phase with my IQ150 after nearly a year).

 

[Kolor-Pikker]

I think Pentax's latest 645 lenses do not cover Full Frame MF with the image circle so I don't know if Pentax would get it. That being said, I think Sony could go for it and reintroduce Zeiss MF lenses that we have missed so much from Contax.

The only Pentax lenses which do not cover full 645 are the DA 25mm and 28-45mm, while every other lens does, including the D-FA 25/4, 35/3.5, 55/2.8, and 90/2.8 lenses. It would be interesting to see which focal length the 28-45 will cover out to, probably around 32mm.

 

[torger]

The "best" is not necessarily going to be the IQ3 100 depending on how you shoot. We're still waiting for that 32HR test. What if it doesn't shift well? Use SLR wide angles and do software perspective correct? Sounds boring to me.

If it does shift well, better than the IQ250, then it will be the greatest news for the tech cam genre in years. Considering the amount of tech cam shooters on this forum this should be the number one thing to test.

 

[ondebanks]

The happy stuff:

I've been waiting for this day since 2002!

Yes, seriously, that long. Canon's first low-noise CMOS SLR, the EOS-D60 of 2002, set a new standard for clean sensitivity (high ISO + long exposure) that full-frame medium format had not caught up with - until now. The XF 100MP is way out of my price league, but still excellent news for us lovers of medium format systems, and the Mamiya 645 in particular.

We are finally back to the norms of the film days: when cameras all had the same image quality on a per-square-mm-of-capture-area basis, and medium format was king purely because it was bigger. [OK, I probably said the same thing when the 50MP CMOS backs came out, but they were cropped-medium-format; e.g. the 24mm fisheye cannot do its full 180-degree fisheye thing on them. This is a really big step forward].


The critical stuff:

Why is it, yet again, that Phase One cannot supply its dealers and potential customers with the full pertinent technical details of a product that is shipping? Here we all are, Doug included, stabbing in the dark as to whether:

1. this is a BSI sensor or not?
2. the claimed 15 MP DR is per pixel or not?
3. it behaves well with tech-cam lenses or not?
4. it takes a mandatory dark frame in long exposures or not?
5. the full signal chain including ADC is 16 bits or not?

Ray

 

[voidshatter]

but they were cropped-medium-format; e.g. the 24mm fisheye cannot do its full 180-degree fisheye thing on them.

Unfortunately for single exposure the 44x33 crop sensor + 23HR combo is wider than the fullframe sensor + 32HR combo, assuming that the 32HR works with it at all. If you mount a 23HR on a fullframe sensor then you cannot shift. Sometimes it's not the bigger the wider. A good example would be the Voigtlander FE-mount 10mm fullframe rectilinear lens, wider than 11-24mm on a 5DSR or a 17mm TSE on an IQ3 100MP.

(But I do agree that the IQ3 100 MP is the new king!)

it takes a mandatory dark frame in long exposures or not?

I can't think why they would disable the Aerial option in the IQ3 100MP. You could have disabled the darkframe NR ever since the first IQ series I believe (but only CMOS would get least penalized IQ.)

 

[Quentin_Bargate]

This thread was started essentially as a Phase One marketing release, complete with front page supporting publicity, given that Doug is a Phase One dealer. Nothing wrong with that so long as you are aware of the situation, and its clearly a great new product. However, the problem with MFD and the reason I no longer use it in the field is sheer impracticality. By the time you earn enough to buy the kit, your body is too old to carry it around!

Then when one gets in to stitching, how many photographers would really find the results from, say, a Sony A7RII 3 or 4 image stitch to be an inadequate, second rate solution? Not many I would think.

Similar or better quality to a 100mp back has always been available if you use an 8x10 field camera and film. The reason most of us don't use it is sheer inconvenience. I see inconvenience (as well as price) as still being significant impediments MF digital cannot overcome, regardless of the pixel count.

Still, happy to test out the kit if anyone will lend me a back and camera

 

[voidshatter]

If I can fund it, an IQ3 100 is definitely on my radar at some point even if reluctantly not immediately. (and as I believe I may have posted before, I'm still in the honeymoon phase with my IQ150 after nearly a year).

Oh no! If the thieves took your IQ150, then we could see the wrath of the Dante strikes sooner! :grin: Come on! While the IQ3 100MP drops value, the IQ150 would also keep dropping value, making the upgrade cost about a constant over time; however if you upgrade sooner then you would have more time to enjoy the new monster before it becomes old ("old, but not obsolete") :grin::grin::grin:

I have heard rumors that the 135 format will reach 100 MP in about 2-3 year's time from Sony's sensor department. Time waits no one! Time is money! Don't let the new technology depreciate before you can enjoy it! Life is soooo short!!!

 

[Hulyss Bowman]

This thread was started essentially as a Phase One marketing release, complete with front page supporting publicity, given that Doug is a Phase One dealer. Nothing wrong with that so long as you are aware of the situation, and its clearly a great new product. However, the problem with MFD and the reason I no longer use it in the field is sheer impracticality. By the time you earn enough to buy the kit, your body is too old to carry it around!

Then when one gets in to stitching, how many photographers would really find the results from, say, a Sony A7RII 3 or 4 image stitch to be an inadequate, second rate solution? Not many I would think.

Similar or better quality to a 100mp back has always been available if you use an 8x10 field camera and film. The reason most of us don't use it is sheer inconvenience. I see inconvenience (as well as price) as still being significant impediments MF digital cannot overcome, regardless of the pixel count.

Still, happy to test out the kit if anyone will lend me a back and camera

For renting it should be ok, at the end. Owning such gear on field is just over the top. Meanwhile, for 300$, one can find easily a TEXAS leica with a 6x9 sensor inside and produce out of the box superior results in many circumstances

Sorry to hijack the commercial thread