Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Scarlet 2/3" DR?

Being the sensor an M-X, I think it will be similar to the one in the Epic, if not exactly the same. Maybe I'm wrong and the smaller size affects the DR. Pure speculation here ;)
 
If Sony's new 2/3" midrange cine-alta camera, the PDW-F800 can hit 13-14 stops of DR, I expect Scarlet can and will come pretty close to the same.
 
Yea, if it can hit 13.5 stops like Epic... it will be something to behold, considering the price.

Everyone is always looking for the holy grail of the "film look" and DR is such a huge part of that. Far more important than super shallow DOF.
 
Last edited:
I would be quite happy if they hit their original target of 11-1/2+ stops. That would still be 4 to 5 stops better than what I am using now.
 
Yeah we all wish it could be equal to the 5K M-X, that'd be amazing.

But the photosites are 1/3 the size (~9.2 per cm2 vs ~3 per cm2) so "equal to" [not just in DR but in high ISO] I don't believe is possible and even "very close" would be a huge accomplishment.

But Jim said in a post once that it would in fact be "very close" and he doesn't steer us wrong so I'm optimistic!
 
Yeah we all wish it could be equal to the 5K M-X, that'd be amazing.

But the photosites are 1/3 the size (~9.2 per cm2 vs ~3 per cm2) so "equal to" [not just in DR but in high ISO] I don't believe is possible and even "very close" would be a huge accomplishment.

But Jim said in a post once that it would in fact be "very close" and he doesn't steer us wrong so I'm optimistic!

Semiconductor design plays a far more important role than pixel size in DR - take a look at Mysterium vs Mysterium X, for example. The pixels are exactly the same size but MX has far better noise characteristics and better DR - just because it's an even better design than the older sensor.
 
Semiconductor design plays a far more important role than pixel size in DR - take a look at Mysterium vs Mysterium X, for example. The pixels are exactly the same size but MX has far better noise characteristics and better DR - just because it's an even better design than the older sensor.

Wafer design is certainly important, but let's not underestimate pixel size given the current state of the art. Until the next big idea in sensor technology is mastered, the difference would be even greater if the newer sensor had bigger pixels.
That's why 4/3rds can always be trumped by APS-C, and so on up the scale via full-frame 35mm to MF.......all other things being equal (and of course they aren't always), bigger is usually better:).
 
http://reduser.net/forum/showthread.php?t=30076&highlight=myth+busted

For pixel sizes between 3 and 7 microns, there is little difference in inherent noise and DR everything else being equal. The overall size of the sensor has more impact on light gathering and S/N ratio than pixel size.

Yes I've seen that before but, as in many scientific fields, there are differing views so the 'myth' isn't busted but open to interpretation. Sensor size and wafer design of course play their part, and it's from advances in the design and implementation of both sensor and the associated signal processing that we've seen the progress of the last ten years.

Another practical example, similar to M v MX, is seen in the Nikon D3 to D3s progression. Both have full-frame 12mpx 35mm sensors with a pixel size of 8.4 microns, the D3s however benefits from newer technology and demonstrates improvements in DR and sensitivity.
Their D3x sibling by comparison also has a full-frame 35mm sensor, but with 24mpx and a pixel size of 5.9 microns. Whilst it has, as might be expected, higher resolution, it also has lower DR and sensitivity (as might be expected by some:)).

Of course in the real world it's the visible output quality that matters and from the comments here, RED have done a great job in making the transition from M to MX. I'd still be (pleasantly) surprised though if the 2/3rds MX was quite as good as the S35.
 
Innovations in chip design and silicon fab processes over the last 3 years have increased S/N ratios for all kinds of sensors by about 10dB, not just camera sensors.

It is interesting that going back an regrading some build 17 R1 M sensor files in the new Redcine X I see an effective 1-1/2 stop improvement in DR just from software improvements alone over the last year. Red's software team is doing some remarkable work.
 
Yes, I agree, they certainly are, and hopefully there's more to come too. And then there's Monstro of course:).
 
Wasn't larger pixel size one of the sited explanations for the superior DR of Alexa over the Epic?
 
Wasn't larger pixel size one of the sited explanations for the superior DR of Alexa over the Epic?

True, but the actual difference in DR is very slight. Both are right around 13.5 stops. Just goes to show that simple explanations aren't always completely accurate.
 
I thought the charts showed a pretty significant difference, at least looking at it with my unsophisticated eyes.
 
No, only about 1/3-1/2 stop difference at most and pretty much only in the highlights because of their 'HDR' type scenario. The interesting thing that at the other end the Alexa showed a lot more ugly noise than the MX so it's a kind of win/lose situation with the Alexa sensor...slight more in the top but less in the low.
And don't forget RED are doing a HDR mode which could see a couple of extra stops 'DR' put into the image which would kill off the Alexa in this regard. They have already done it and only slightly have an advantage in the highlights.
 
Also, from my understanding and until i am corrected i am wrong...a larger pixel size can result in more noise when pushed...hence more with the Alexa in the low end. MX squeezes a lot more pixels into a similar size sensor!
 
Also, from my understanding and until i am corrected i am wrong...a larger pixel size can result in more noise when pushed...hence more with the Alexa in the low end. MX squeezes a lot more pixels into a similar size sensor!

Usually, certainly in stills sensors to date, more (therefore smaller) pixels on the same sized sensor results in increased resolution but reduced sensitivity (narrower ISO range and more noise kicking in at a lower max ISO).

The Nikon comparison above is one example. Comparing the D3s (full-frame 12mpx) to Canon's 5D MkII/1Ds MkIII (full-frame 21mpx) is another.
Of course wafer design and processing firmware play an important role too, as seen in the D3s to the older D3 comparison, and also in RED M v the more recent RED MX.

I stress that it's today's technology that is being compared in those examples, who knows what's just around the corner. I'm also not a cinematographer and there are no doubt differing priorities to take into account when designing sensors and electronics for cinema cameras, but the underlying principles remain the same.
 
Last edited:
Back
Top