Q: about the Phase One P30+

[Alex MacPherson]

For those of you that have used one: What is the highest practical ISO for this back?
The specs say 1600 but we all know the specs tend to be optimistic. I have the P30
which lists 800 .. but I would never consider using it at 800.

Thanks in advance.

[TMARK]

Depends on your use.  Commercial use, 400, and depending on your lighting, some noise in shadows but it cleans up well. 

[Alex MacPherson]

More suitable for available natural light than the P30 (non-plus)?

I guess what I am really asking is the P30+ much of a jump from the P30.
I love the performance of the P30 in studio conditions... and am considering streamlining
my equipment and selling the DSLR.
I wouldn't give up the the Phase because the studio images are outstanding.

[JV]

I use a P30+ on a Hasselblad H1. 
I would say 100 is excellent, 200 is very good, 400 is usable, 800 and 1600 are in my opinion not very usable.

[Doug Peterson]

Depends on...

Software
ISO1600 in Capture One 6 with fine tuned noise reduction settings will be significantly better than Capture One 3.7.9 (no big surprise there).

Firmware
Early firmware like 2.9.8 was not as good at ISO800 as modern firmware like 5.1.2.

This is similar to Hasselblad where they actually felt comfortable adding an entire stop in the ISO setting menu because they were able to improve (through firmware) the high ISO performance enough to justify it.

I would not say the improvement on a P30+ from early firmware to modern firmware would justify a full stop added to the menu setting, but it's something like half a stop better regarding noise at ISO800 and ISO1600 from oldest to newest firmware - doesn't sound like a lot, but when you're already pushing the max of the system it's a very welcome improvement.

All backs can be updated to the modern firmware, but occasionally earlier digital backs (e.g. ones delivered with 2.9.8 ) will not take the new firmware easily and may require a trip to the factory or a dealer to have the firmware updated, which if it is out of warranty would carry a sizable service charge (free of course if in-warranty). One (very selfish) suggestion to avoid this hassle is of course to buy from a dealer who will make sure you receive a back with the latest firmware and/or add a warranty to the back.

Reproduction Size
If for a magazine page a P30+ is 31 megapixels with minimal cropping for the page size.
If for a magazine page a 5DII is 22 megapixels with a crop down to around 18 for the page size.
So the same noise at 100% would reproduce nicer from the P30+ than the 5D Mark II.

I'm not saying the P30+ will always hold it's own at very high ISOs against a D3 or 5DII etc. Only that many people judge noise at 100% pixel size on the screen and forget that a higher resolution file will print with smaller pixels for any given print size. Just like a noisy ISO800 film emulsion looked better when exposed and printed from an RZ67 than when exposed and printed from a 35mm SLR (same amount of grain per square inch of film, but smaller grain when printed at the same size).

Lighting type
Lighting with studio strobes or other near-daylite light sources exposes all three channels (red/blue/green) equally. Mixed lighting (e.g. warm light - in other words reddish light) can underexpose one channel while overexposing another. So rebalancing the raw file to more neutral coloring is effectively pushing the underexposed channel and results in higher noise in that channel.

blah blah blah... ISO1600 in the studio with 6500k light is not the same as ISO1600 under god-awful street lights. So when you do testing make sure to use it YOUR shooting style in your typical environment.

Personal Taste
I've met people who call ISO3200 on a 5D mark 2 "clean". I do not. I am not saying they are wrong - simply that our tastes and standards are obviously different. Hands on testing, in a scenario realistic to your needs, is the only way you can fully answer this question.

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration
Phase One Partner of the Year
Leaf, Leica, Cambo, Arca Swiss, Canon, Apple, Profoto, Broncolor, Eizo & More

National: 877.217.9870  |  Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One 6 at 10% off

[ziocan]

I would not mind using the P30 at 800iso. Of course the files would need some noise reduction with Noise ninja or similar plugs in and it would never be as crisp and clear as at 100 iso, but the file would still be cleaner and look "different" from a DSLR file at the same iso.
31mp at 800 iso with a bit of PP, can still print a spread on a magazine without showing grain at all.

[Alex MacPherson]

So the answer seems to be that there isn't a lot of improvement from the P30 in the area of high iso performance to justify the expense of upgrading.

I have a 1dsMkiii... which sits in my case a great deal now that I shoot with my 'Blad and P30.

I will go try some high iso stuff tomorrow and run it through C1 v6. Is there a tutorial for the noise reduction tool?

[SecondFocus]

I think it would depend on your subject matter. I used a P30+ with guys working out in a gym and it was just fine at 800 for what I needed.

[ondebanks]

Regarding firmware - the only thing that matters in this regard is whether changing ISO on the back sends the data into a different readout mode (with real analog gain being increased upstream of the AtoD converter, reducing overall readout noise).

If the P30/P30+ is built like the P45+ (another Kodak sensor of the same generation in another PhaseOne back of the same generation), the firmware does not do this. All ISOs above 100 in the P45+ are "faked" by downstream digital multiplication (DxOmark result: "only the values for the lowest two ISO settings (50 and 100) are real; the other ISO values are achieved by applying a digital gain during RAW conversion". With this type of ISO manipulation, you gain absolutely nothing in noise reduction, and you potentially (depending on how the RAW conversion is done) lose a stop of highlights for every stop you increase the ISO above 100. Losing a stop of highlights with each stop of ISO increase over base ISO is of course universal in digital cameras/backs where true ISO gains are used, but you at least get lower readnoise (and less significantly, lower quantization noise), so there is an improvement in shadow detail, and that's why people do it. With the P45+, you gain nothing in shadow detail, and (as I said, depending on how the RAW conversion is done) you might still throw away the highlights!

Setting the camera to anything above ISO 100 is therefore meaningless. If you want to use, say, ISO 800 for creative reasons like faster shutter speeds, you don't need to do it in the camera; an equivalent strategy is to set the camera/back to ISO 100, exposure compensation to -3 stops, and recreate the ISO 800 intensities in in post processing while retaining your highlights.

[As an aside: I wonder if faked high ISOs with highlight retention might in part explain the continuing reputation of MFDBs for having remarkable dynamic range. Shoot a real-ISO DSLR alongside a fake-ISO MFDB at the same higher-than-base ISO, and the DSLR will clip at some highlight level, while the MFDB may not if its RAW conversion software is smart enough. But the shadow detail will tell another story.]

So anyway, the P45+ has a real ISO range of 50-100...don't be distracted by the specs which say 50-800. Putting it another way: since ISO 200, 400 and 800 are not real, why stop there? Why not further double and quadruple the digital multiplication and say it goes to ISO 1600 and 3200? What PhaseOne mean is that they think that there's enough shadow detail in a 3-stop underexposed ISO 100 image that you can multiply all the pixel intensities by 8 and still get an acceptable "ISO 800" image. But that's entirely a subjective call, which is not underpinned by any basis in the hardware/firmware.

This is not just a PhaseOne issue either, BTW. Hasselblad does exactly the same with their H3DII-39 and H3DII-50. It's particluarly interesting to compare the P45+ with the H3DII-39, since they use the same Kodak sensor. The P45+ nudges ahead of the H3DII-39 in DxOmark mainly because its firmware takes it to a real ISO of 100 (vs. the Hassy's only real ISO of 50), and this is reflected in the "Sports (low light ISO)" scores of 622 vs 532, the only substantial score difference between the two, and in Hasselblad's decision to declare the max ISO spec as ISO 400, rather than PhaseOne's call of ISO 800 (this is the same sensor, remember!). This comparison shows how real ISO matters, up to a point determined by the underlying sensor noise floor.

[John R Smith]

This is really important stuff. If this is all true (and my own tests with the 39MP Kodak sensor seem to bear it out), then all the ISO settings above base are just nonsense. You might just as well shoot at 50 and give it plus 3 EV in Lightroom for 400 ISO equivalent.

John

[yaya]

This is really important stuff. If this is all true (and my own tests with the 39MP Kodak sensor seem to bear it out), then all the ISO settings above base are just nonsense. You might just as well shoot at 50 and give it plus 3 EV in Lightroom for 400 ISO equivalent.

John

It is true for many cameras but not for all of them. However if you shoot everything 4 stops under you won't be able to judge anything on screen, neither from the preview nor from the histogram...

Also Lightroom does not always read all the data that is tagged to the raw file, hence why sometimes high iso images come in and look under exposed.

Yair

[Guy Mancuso]

I have used ISO 800 under tungsten lighting many times which is the worst for noise and processed with C1 it does a very nice job. Not sure i would go past a 16x 20 with it but for smaller output very nice like doing PR type work. For advertising and full resolution output above like a 16 x20  I would probably hold at 400. Now these are my preferences and experiences with the P30+. Some folks may have different threshold points and different needs. Once again i would NEVER use anything but C1 on a Phase file PERIOD.

[fredjeang]

Setting the camera to anything above ISO 100 is therefore meaningless. If you want to use, say, ISO 800 for creative reasons like faster shutter speeds, you don't need to do it in the camera; an equivalent strategy is to set the camera/back to ISO 100, exposure compensation to -3 stops, and recreate the ISO 800 intensities in in post processing while retaining your highlights.

I'm happy to hear that, and as John said it confirms my own experienced. I wanted to post something exactly on that but didn't dare because I wasn't sure enough.
And IMO it is not just the 2 brands involved.

Recently I did a video testing in high isos with the 5D under horrible light conditions and obtained exactly the same result you are mentioning pushing the isos not in camera but in software. In other words, the original files obtained by setting to very high isos from the beginning where by far worsed than shooting at 200 with e.c and then recreate the desired iso in Premiere. I'm not engineer, but this is weired! 

[tho_mas]

In other words, the original files obtained by setting to very high isos from the beginning where by far worsed than shooting at 200 with e.c and then recreate the desired iso in Premiere. I'm not engineer, but this is weired!  

this is indeed weird as with the video from the 5D2 you are not working with RAW files. Or do you???... to my knowledge the 5D2 only provides 2 ("semiprofessional") codecs; with reduced bit depth, of course.

As to my non-plus P45 I can reinforce that underexposed shooting at base ISO and pushing exposure in Capture One gives better results than shooting at the respective ISO. It's not a world of a difference but visible.
The same goes for my P21+.
But Yair is also right... shooting this way is quite limiting as you can't control the histogram. It's doable under controlled conditions, of course.

[eronald]

Ondebanks - please explain a bit?

The P45+ nudges ahead of the H3DII-39 in DxOmark mainly because its firmware takes it to a real ISO of 100 (vs. the Hassy's only real ISO of 50), and this is reflected in the "Sports (low light ISO)" scores of 622 vs 532, the only substantial score difference between the two, and in Hasselblad's decision to declare the max ISO spec as ISO 400, rather than PhaseOne's call of ISO 800 (this is the same sensor, remember!). This comparison shows how real ISO matters, up to a point determined by the underlying sensor noise floor.

[fredjeang]

this is indeed weird as with the video from the 5D2 you are not working with RAW files. Or do you???... to my knowledge the 5D2 only provides 2 ("semiprofessional") codecs; with reduced bit depth, of course.

That's correct, that is what I wrote it's weird. Don't have any explaination but there might be one. As I said I don't have the technical competences in that matter to understand all the parameters involved.

[JV]

All backs can be updated to the modern firmware, but occasionally earlier digital backs (e.g. ones delivered with 2.9.8 ) will not take the new firmware easily and may require a trip to the factory or a dealer to have the firmware updated, which if it is out of warranty would carry a sizable service charge (free of course if in-warranty). One (very selfish) suggestion to avoid this hassle is of course to buy from a dealer who will make sure you receive a back with the latest firmware and/or add a warranty to the back.

Quite frankly I cannot understand why Phase One cannot deliver an upgrade that will work for sure, I am running the older 2.9.8 firmware version and I would love to upgrade if somebody could tell me whether the upgrade will work or not but I keep getting the same reply as above, it might or it might not, it doesn't reflect very well on the reliability of Phase One...

[bcooter]

Quite frankly I cannot understand why Phase One cannot deliver an upgrade that will work for sure, I am running the older 2.9.8 firmware version and I would love to upgrade if somebody could tell me whether the upgrade will work or not but I keep getting the same reply as above, it might or it might not, it doesn't reflect very well on the reliability of Phase One...

I don't believe it reflects on reliability, it just seems like Phase has a less than transparent process.

Of all the good things Phase does, the warranty, repair price system probably builds more bad will than anything, because it comes across as a forced fear buy.

Either pay us $2,000 (I guess it's $2,000 I can't find the price)  for the warranty or pay us $2,500 to change the watch battery on the back.  

It's the same with  "file a support claim then we'll talk".  That's not what one wants to hear when your on set or actually anytime.

Nobody really minds paying a fair price for repairs, but it gives the impression that phase subsidizes their dealers heavy lifting, off the backs (no pun intended)  of their users.

Case in point.  Try to find out virtually anything on the phase forum that pertains to price and the answer always is, "call your local dealer".

Try to find out how long it takes to swap mounts . . . "call your local dealer", etc. etc.

A good dealer has worth, but not everyone needs the same level of support and most of us learned how to make a phone call or send an e-mail.  Why the added layer?

Obviously someone new to digital capture will probably call the dealer daily, someone seasoned only needs to talk to a dealer when something goes down, but both pay the same warranty price.  

On my two Phase backs I've had two warranties.  One was value added, the second standard 1 year.  Neither needed repair but I've called the dealer only twice to ask about issues and one time was told "I dunno maybe firewire cord", the second time we sent files and listed the issues and never got a response.

I believe Hasselblad has a logical get a quote, then pay for the repair system and Canon with CPS allows you to select the type of service you require.

Phase makes good, reliable equipment, but the day I see a price for a $2,500 watch battery is the day I buy a Hasselblad, if just out of principle.

IMO

BC

[bcooter]

In regards to a p30 or p30+,  it's your decision, but . . .

The file is virtually the same and the original p30 has a much better lcd, especially in bright light.  Actually if you put the lcd into a light you can see details that the plus back's lcd loses.

In regards to noise and higher iso's the trick is to process out two images and blend them.  One to hold detail, one to smooth the darker midtones and shadows.

This image was produced that way . . . actually we shot this image with both the p30+ and a 1ds3 at 800 iso and out of the can there wasn't much difference in noise, except the noise was just a different shape, but about the same.



This was also shot with continuous light  (tungsten and hmi) and fairly low amounts of illumination.

IMO

BC

[michele]

Try to develop the files at 800 and 1600 iso with adobe camera raw 6.0 or earler with the default settings... it's a different world from capture one. the big problem could be the colors of the image... capture one is better, but with high iso sucks...