Pages: [1]   Go Down

Author Topic: Sinar 54M performance  (Read 10981 times)

torger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3267
Sinar 54M performance
« on: February 22, 2012, 10:57:14 am »

Looking at discontinued second hand backs, Sinar 54M comes up as good deal concerning price/performance compared to other discontinued backs from other manufacturers. However, I cannot find a raw file (sti) to get a rough idea how it performs and test it with my post-processing tools. For some reason it seems to be a bit sensitive for manufacturers to provide raw files (there are exceptions though) so I could not find any at Sinar's web site.

Is someone using a Sinar 54M that could provide a sti file to look at? Perhaps it is taboo asking for raw files, not sure :), but I would be glad to get one to look at...
Logged

PdF

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 297
Re: Sinar 54M performance
« Reply #1 on: February 22, 2012, 11:10:58 am »

Using the latest versions of CaptureShop (now 6.1), you save your files directly in DNG. No export problem, either with Bridge, Camera Raw, Photoshop or else. Simply choose a favorite!

PdF
Logged
PdF

torger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3267
Re: Sinar 54M performance
« Reply #2 on: February 22, 2012, 11:29:05 am »

That's good to hear! The other worry I have though (I always worry) is that old sensor tech might not perform as well as I'd expect/want. I've followed the pro digital second hand market for some time, and the general pattern is when something gets down to a price range similar to 35mm digital it is beyond obsolete. The Sinar 54M is really close to that price range now which makes me a juuuust a little bit suspicious about it.

It seems like the Sinar 54M was considerably less expensive than other 22 megapixel backs at the time it was available, and also stated lower dynamic range, which also makes me wonder why. So it would be great to get a look at an actual file, normal exposure time, base ISO, to see if there are any obvious problems. The things I care / not care about:

  • Usuable dynamic range - low noise levels and no pattern noise please. I've heard something about that it needs "black calibration" for a few minutes which sounds kind of worrying
  • Blooming issues - is it possible to shoot into the sun? Rather important
  • Long exposure performance largly irrelevant (a bonus if it works though)
  • Base ISO only, higher ISO performance irrelevant

I need to find out some stuff around the tethering too, but that may be easier to find via some browsing. I does seem like it may have a bit more cumbersome workflow than usual [=used to these days] concerning calibration...
« Last Edit: February 22, 2012, 01:21:50 pm by torger »
Logged

ondebanks

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 858
Re: Sinar 54M performance
« Reply #3 on: February 23, 2012, 07:41:44 am »

That's good to hear! The other worry I have though (I always worry) is that old sensor tech might not perform as well as I'd expect/want. I've followed the pro digital second hand market for some time, and the general pattern is when something gets down to a price range similar to 35mm digital it is beyond obsolete. The Sinar 54M is really close to that price range now which makes me a juuuust a little bit suspicious about it.

It seems like the Sinar 54M was considerably less expensive than other 22 megapixel backs at the time it was available, and also stated lower dynamic range, which also makes me wonder why. So it would be great to get a look at an actual file, normal exposure time, base ISO, to see if there are any obvious problems. The things I care / not care about:

  • Usuable dynamic range - low noise levels and no pattern noise please. I've heard something about that it needs "black calibration" for a few minutes which sounds kind of worrying
  • Blooming issues - is it possible to shoot into the sun? Rather important
  • Long exposure performance largly irrelevant (a bonus if it works though)
  • Base ISO only, higher ISO performance irrelevant

I need to find out some stuff around the tethering too, but that may be easier to find via some browsing. I does seem like it may have a bit more cumbersome workflow than usual [=used to these days] concerning calibration...

The 54M uses the same Kodak KAF-22000 sensor as the Phase One H25 & P25/P25+, Hasselblad CF22, and Imacon [V]132C. So they can all serve as references for expected performance. The H25 might be the closest comparison, as it too was a tethered-only back - no LCD, no battery, no card.

I have no first hand experience with the 54M, but would expect the following:
  • Usuable dynamic range - low noise levels and no pattern noise please. I've heard something about that it needs "black calibration" for a few minutes which sounds kind of worrying. Unless Sinar chose to read the sensor out faster than others, the readout noise should be the same as others, and hence the dynamic range (it is curious, as you point out, that they spec "greater than 11 stops" whereas others spec "12 stops"). The sensor has never been reported to exhibit noticeable pattern noise. The "black calibration" sounds like a bias frame, which would remove any residual pattern noise. Taking and reading out a bias frame should only take a couple of seconds though! Perhaps the process takes a few minutes because it takes one at every exposure time within the 1/4000 to 4 second normal time range, that might be used in a session? - just my guess. Of course once the exposure time becomes appreciable, it's no longer a bias frame, it's a dark frame.

  • Blooming issues - is it possible to shoot into the sun? Rather important. Kodak 9-micron sensors like this one have stronger anti-blooming protection than Dalsa 9-micron sensors (see my detailed analysis on getDPI), but it's still not enough in my experience (KAF-16802, Kodak DCS 645M) to prevent blooming when shooting with the sun in the frame (other than close to sunset/sunrise, or with thick haze/thin cloud to attenuate the direct light of the solar disk).

    In my two examples below, which I happened to shoot last weekend, the one which bloomed was actually taken 13 minutes later than the one which didn't; the sun was lower but unscreened at that moment by any cloud. These were taken with the Mamiya 24/4 ULD fisheye, so of course it was not possible to use a grad ND on the sky/sun.

  • Long exposure performance largly irrelevant (a bonus if it works though). Sinar limit this back to 32 seconds max, and most other backs from most other manufacturers also have a 30/60 second limit. But there are good 30-second backs and not-so-good 30-second backs. Sinar have always been amongst the best in the market for keeping their backs cool, and the 54M has regulated, active cooling (peltier with a dewpoint sensor), so long exposures should be very good unless the ambient temperature is very high.

  • Base ISO only, higher ISO performance irrelevant. At base ISO (25), you'll be very happy with the signal-to-noise. Sinar conservatively spec the base ISO at 25, whereas other manufacturers using this sensor spec it at 50. In practice, this means you are exposing to the right by +1 stop more; gaining 1 stop more shadow detail and losing 1 stop highlight detail, and improving signal-to-noise at all subject tones. But of course you could get the same result by applying a constant +1 stop exposure compensation at ISO 50 on any other back with the same sensor, since with MFD, unlike CMOS DSLRs, there's no variation of readout noise with ISO. Sinar's conservatism extends to max ISO as well: just ISO 200 with the 54M, versus ISO 400 or 800 with everyone else. You could shoot it with -1 stop exposure compensation to "make" it an ISO 400 back - the same argument applies.

Ray
Logged

TMARK

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1841
Re: Sinar 54M performance
« Reply #4 on: February 23, 2012, 10:10:07 am »

I used these often back in the day, in a studio, to shoot beauty and fashion.  They are SLOW (think 5 seconds between shots), shoot tethered only, but produce beautiful files.  Check out tis site:  www.sarasilver.com.  Look at her early beauty work.  Not sure what Sara shoots today, but back in the day ut was a 54M on a 503CW Blad, tethered to a G4 Powerbook, running lightroom and Capture shop.  The low prices reflect the (lack of) speed, no LCD, tethered only operation, not the IQ.  I switched to Leaf and Phase before exiting the MFDB market, but only because I like to shoot handheld with a screen.

The 54M files are as flexible as the Aptus 22 and P25 files in terms of range and post.  ISO is better on the Aptus, and the P25+ is a different generation.  The Phase H25 is closer in terms of post, but I found the 54M files to be MUCH better (subjective, I know) in terms of color.

Good luck with this.
Logged

TH_Alpa

  • Full Member
  • ***
  • Offline Offline
  • Posts: 214
Re: Sinar 54M performance
« Reply #5 on: February 23, 2012, 06:21:32 pm »

Ray (and for "Torger"),

Just to confirm most of what you are writing and giving some more information, because I have used this back for years:

The 54M uses the same Kodak KAF-22000 sensor as the Phase One H25 & P25/P25+, Hasselblad CF22, and Imacon [V]132C. So they can all serve as references for expected performance. The H25 might be the closest comparison, as it too was a tethered-only back - no LCD, no battery, no card.

I have no first hand experience with the 54M, but would expect the following:
  • Usuable dynamic range - low noise levels and no pattern noise please. I've heard something about that it needs "black calibration" for a few minutes which sounds kind of worrying. Unless Sinar chose to read the sensor out faster than others, the readout noise should be the same as others, and hence the dynamic range (it is curious, as you point out, that they spec "greater than 11 stops" whereas others spec "12 stops"). The sensor has never been reported to exhibit noticeable pattern noise. The "black calibration" sounds like a bias frame, which would remove any residual pattern noise. Taking and reading out a bias frame should only take a couple of seconds though! Perhaps the process takes a few minutes because it takes one at every exposure time within the 1/4000 to 4 second normal time range, that might be used in a session? - just my guess. Of course once the exposure time becomes appreciable, it's no longer a bias frame, it's a dark frame.
The SB 54M as well as the SB 54 H (Same sensor) do effectively make a black calibration (or "dark frame") starting exposures longer than 1/2", just before taking the shot. It takes a few seconds and the back is ready, this to remove noise from the heating of the sensor with "long" exposures. It doesn't take minutes like mentioned.

  • Blooming issues - is it possible to shoot into the sun? Rather important. Kodak 9-micron sensors like this one have stronger anti-blooming protection than Dalsa 9-micron sensors (see my detailed analysis on getDPI), but it's still not enough in my experience (KAF-16802, Kodak DCS 645M) to prevent blooming when shooting with the sun in the frame (other than close to sunset/sunrise, or with thick haze/thin cloud to attenuate the direct light of the solar disk).
Absolutely correct.

  • Long exposure performance largly irrelevant (a bonus if it works though). Sinar limit this back to 32 seconds max, and most other backs from most other manufacturers also have a 30/60 second limit. But there are good 30-second backs and not-so-good 30-second backs. Sinar have always been amongst the best in the market for keeping their backs cool, and the 54M has regulated, active cooling (peltier with a dewpoint sensor), so long exposures should be very good unless the ambient temperature is very high.
Absolutely correct, the performance and IQ at 32" is outstanding when the ambient temperature is not too high (> 30 degrees C). The 54M though doesn't have an active cooling (Peltier element) built-in, the cooling happens with the specially built casing.

  • Base ISO only, higher ISO performance irrelevant. At base ISO (25), you'll be very happy with the signal-to-noise. Sinar conservatively spec the base ISO at 25, whereas other manufacturers using this sensor spec it at 50. In practice, this means you are exposing to the right by +1 stop more; gaining 1 stop more shadow detail and losing 1 stop highlight detail, and improving signal-to-noise at all subject tones. But of course you could get the same result by applying a constant +1 stop exposure compensation at ISO 50 on any other back with the same sensor, since with MFD, unlike CMOS DSLRs, there's no variation of readout noise with ISO. Sinar's conservatism extends to max ISO as well: just ISO 200 with the 54M, versus ISO 400 or 800 with everyone else. You could shoot it with -1 stop exposure compensation to "make" it an ISO 400 back - the same argument applies.

Ray
Absolutely correct.

Best regards
Thierry[/list]
Logged

torger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3267
Re: Sinar 54M performance
« Reply #6 on: February 24, 2012, 02:46:38 am »

Thank you all for the helpful replies.

Sensor performance probably not be a problem then. I'm used to Canon sensors so I'm not exactly hard to impress :). I've got my paws on a set of P25+ raw files, and while not as clean as the best CMOS sensors (i e Sony Exmor) it had only slight pattern noise and is much better than say Canon 5Dmk2 and won't give any problems even if pushing a couple of stops. I guess the Sinarback performs about the same. I also need to take into account that these old backs mounted on a large format camera with a mechanical shutter cannot be exposed as precisely as a DSLR with ETTR-trial-error-loop, so one probably loses a stop or so in DR due to that. The twice as large sensor area should give a little back though.

H25 backs have started to appear too at around the same prices. So I'll keep my eyes open for those as well. When costing only around $2500, what kind of deals you can get on adapter plates and sync cables together with the back becomes significant.
« Last Edit: February 24, 2012, 02:53:41 am by torger »
Logged
Pages: [1]   Go Up