Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

EPIC and Alexa...

Status
Not open for further replies.
Here's an interesting look at this issue from Matt Jeppsen of Fresh DV:

http://provideocoalition.com/index.php/freshdv/story/epic_vs_alexa_dynamic_range/

This is not scientific by any means, but it appears that both cameras have tremendous dynamic range, with Alexa performing slight better in the highlights and Epic performing better in the shadows.

The bottom line is, chemical film is about to be usurped.
 
George A. said:
EPIC definitely has the overall advantage over Alexa and I'll buy the EPIC, no doubt.
Click to expand...
Even if the roles were reversed and Alexa had all the advancements, I think most people would still buy the RED because of price. But to have the advantage of price along with all the other advantages... well, let's just say a lot of us got lucky.
 
Epic has one tremendous advantage in being able to shoot around 100 frames per second at full RAW. To me, that is simply HUGE.
 
Tom Lowe said:
Epic has one tremendous advantage in being able to shoot around 100 frames per second at full RAW. To me, that is simply HUGE.
Click to expand...
I agree. RAW!
 
visualnary_red_speed2.jpg
 
Last edited:
Tom Lowe said:
Here's an interesting look at this issue from Matt Jeppsen of Fresh DV:

http://provideocoalition.com/index.php/freshdv/story/epic_vs_alexa_dynamic_range/

This is not scientific by any means, but it appears that both cameras have tremendous dynamic range, with Alexa performing slight better in the highlights and Epic performing better in the shadows.

The bottom line is, chemical film is about to be usurped.
Click to expand...

I can always get minimum one stop more out of R1-MX image with a set of pretty good and effective helper plug-ins that means only with my special way of doing things to help everybody later to easy achieve this in post...

If ever I get one proper side by side shot from MX and Alexa I'll show that here.

ARRIRAW plug in for FCP was installed long ago on my desktop and waiting to be finally in use just for that sort of tests...
 
Tom Lowe said:
Here's an interesting look at this issue from Matt Jeppsen of Fresh DV:

http://provideocoalition.com/index.php/freshdv/story/epic_vs_alexa_dynamic_range/

This is not scientific by any means, but it appears that both cameras have tremendous dynamic range, with Alexa performing slight better in the highlights and Epic performing better in the shadows.

The bottom line is, chemical film is about to be usurped.
Click to expand...

Maybe I am missing something here but how could one camera be better in shadows and the other better in highlights? Can't we make them exactly equal by changing exposure?
 
Tom Lowe said:
The Arri can shoot RAW, I believe, but is maxed out at 60fps, as far as I know, and does not have enough resolution for a clean 4K RGB finish.
Click to expand...
Yeah, Should have said REDCODE RAW!
 
Martin, I hope that "Magic Mirror" of yours is correct:001_unsure:
 
Rob Ruffo said:
Maybe I am missing something here but how could one camera be better in shadows and the other better in highlights? Can't we make them exactly equal by changing exposure?
Click to expand...

My comments that Red handled shadow detail better, while Alexa did better with highlight detail, are simply a commentary on the DR charts that Red posted.

This is what Red gave us to examine, we have to assume that all other things are equal in the test. And that being the case, on these charts, Alexa showed more latitude available in the highlights, and was clearly noisier in the shadows. Red didn't show quite the same range in the highlights, but was visually cleaner in the shadows. It may be that you can shift Epic's exposure range to match Alexa's highlights, but in theory that would also shift the shadow range. That's kinda the point of a DR chart like this, right? A sensor can only resolve so many stops either direction. And it looks like Red techs exposed the camera perfectly to show it's range.

Interestingly, it cuts the other way too...given how I was able to recover a full stop of Alexa's highlights with a harsh image curve, I wonder if Alexa's exposure range can be shifted down in a future test. Would love to see the Alexa highlights trail off as they do on my tweaked Red highlights chart. On this test we run out of chart before the Alexa highlights are fully clipped (once tweaked). And if the exposure range was shifted down and additional highlights were recoverable, we'd theoretically also see "more shadow range" out of Alexa (as the scale would be shifted towards the shadows).
 
I have not been following closely how Alexa achieves gains in highlights, and I could be talking totally out of my ass here, but I wonder if the deal might be this: Alexa is essentially a regular camera with a regular sensor that also takes a second, separate pass at highlights and protects them, somehow. Whereas Red MX sensor might be a whole different kind of beast altogether, with some kind of secret sauce built physically into the sensor. This might explain why MX is extremely low in noise on the floor but can not achieve quite the almost HDR-like highlights of Alexa?

I know that both of these cameras smoke DLSR RAWs in terms of dynamic range, so something is going on here.

BTW, it has been rumored that Canon's forthcoming 1Ds4 will boast a big leap in DR, so it seems like the DSLR RAW stills cameras might onto this as well.
 
Jannard said:
...If we motivate other companies to chase perfection, then we have done something good for the industry. I just did not see a rapid push to elevate digital 5 years ago and that is the only reason we stepped in.

Jim
Click to expand...

dear jim, i wish it would motivate other companies too. but what i fear is that a certain know-how might be lost in this process when other companies will start to have to compete at your very low and attractive prices just to stay in business.
now i understand that's the rules of the free market. in the automotive industry cheaper and cheaper cars now crowd the market place but there will always be buyers for beautifully engineered, very expensive cars pushing the technology, like an aston martin DB9, although a tata nano does a good job bringing you from place A to B.
i'm not sure the same will happen in the camera market as it is primarily a tool used by professionals not a loved and fetishized object..although sanjin really loves his red one :)

now RED really pushes the envelope on sensor technology, electronics engineering, compression technology in a spectacular way, but mechanically (and forum users will roast me for saying this) it's more of a tata nano car - although a very robust nano.
(red one: hard disk directly bolted to the mags, wtf?, camera handles that break, cooling system, wobbly base plate (i know it's a feature :) )etc etc)

or for the epic: after a heated debate about thermal expansion you said (bored by the lengthy discussion), okay we'll make a a stainless steel mount, a stainless steel hard front, a stainless steel body for those who pay (i can try to find this post if you want). the discussion then died and we now have a stainless steel mount option, but not a thermal stable hard front design. (the housing can stay in aluminum if the hard front and mount are made out of CNS)

now i look at 15-30 year old cameras from aaton and arri and they are mechanical marvels! the viewfinder on the 535 is a such a beauty. the LTR7's viewfinder is almost 40 years old..yet the cnc machining is perfect.
but the consensus here is..optical viewfinders are so passé, you don't look at what you record.
well i disagree..with raw image recording you don't look at what you record on an 8 bit EV anyways. i prefer a optical viewfinder and have a monitor to control my electronic image and view clipping areas and false color imaging.
but i'm one of the very few. so maybe after you see how few are actually interested you scrap the plans for the optical viewfinder you once announced.

and who knows maybe aaton and arri will in their next version of the cameras (after penelope and alexa) also scrap these plans in order to stay competitive.

what i really would love to see is a high end epic, a DB9 epic if you want, with your electronics and a 235's mechanics and accessories, so the knowledge of the past 150 years of camera making wont be lost - but i know i'm dreaming.

sorry for the long post and i want to state i have the highest respect for what you pulled off - i mean after all you could have spent the rest of your life surfing, taking flights on virgin galactic or watching sunsets..but instead you chose to enter and change a market that is one of the toughest to satisfy...chapeau monsieur jannard!
 
the tata nano is very humble example.. maybe we should replace with a Camry like..
in the end RED is also having lots advantages/price values over DB9.. as fuel consumption.. fuel is expensive these days plus it polutes the air.. :sifone:
 
mjeppsen said:
My comments that Red handled shadow detail better, while Alexa did better with highlight detail, are simply a commentary on the DR charts that Red posted.

This is what Red gave us to examine, we have to assume that all other things are equal in the test. And that being the case, on these charts, Alexa showed more latitude available in the highlights, and was clearly noisier in the shadows. Red didn't show quite the same range in the highlights, but was visually cleaner in the shadows. It may be that you can shift Epic's exposure range to match Alexa's highlights, but in theory that would also shift the shadow range. That's kinda the point of a DR chart like this, right? A sensor can only resolve so many stops either direction. And it looks like Red techs exposed the camera perfectly to show it's range.

Interestingly, it cuts the other way too...given how I was able to recover a full stop of Alexa's highlights with a harsh image curve, I wonder if Alexa's exposure range can be shifted down in a future test. Would love to see the Alexa highlights trail off as they do on my tweaked Red highlights chart. On this test we run out of chart before the Alexa highlights are fully clipped (once tweaked). And if the exposure range was shifted down and additional highlights were recoverable, we'd theoretically also see "more shadow range" out of Alexa (as the scale would be shifted towards the shadows).
Click to expand...

Redcode is raw sensor data, any shifting is happening afterwards. Thats why you can apply the new colour sciences to old footage. The range that is captured is the range that is captured. Alexa is using what Arri call a dual gain path, seperating shadows and highlights and processing them, before recombining them. Red do a lot of magic during development of the image in post.
 
well the DB9 is maybe a bad example..but i see this trend everywhere. we have embroidery and sewing machines that cost new 16'000$ a piece, but pfaff and melco nearly went bankrupt because of cheap chinese knock-offs that are good enough for 90% of the tasks (i'm not saying red is a cheap chinese knock-off)
or high-end european cnc machining centers selling less and less because of cheaper japanese and now taiwanese or worse chinese machines. yet they aren't built to the same standards.
it's a trend i see everywhere and i find it a bit sad.
at the very low price of an epic or an even lower price of a scarlet some solutions just can't be implemented in order to stay in that price range (even at relative high volumes) and hence a certain 'savoir faire' might get lost over time. i know i sound nostalgic.
 
Irmeli R said:
well the DB9 is maybe a bad example..but i see this trend everywhere. we have embroidery and sewing machines that cost new 16'000$ a piece, but pfaff and melco nearly went bankrupt because of cheap chinese knock-offs that are good enough for 90% of the tasks (i'm not saying red is a cheap chinese knock-off)
or high-end european cnc machining centers selling less and less because of cheaper japanese and now taiwanese or worse chinese machines. yet they aren't built to the same standards.
it's a trend i see everywhere and i find it a bit sad.
at the very low price of an epic or an even lower price of a scarlet some solutions just can't be implemented in order to stay in that price range (even at relative high volumes) and hence a certain 'savoir faire' might get lost over time. i know i sound nostalgic.
Click to expand...

"Knockoffs" are typically just copies of more expensive options as you stated. However, when it comes to Red they are not copying any other products. In fact, quite the opposite, they are leading the way by pushing new technology further and the more traditional well known brands are playing catch up. So it is sort of like Red is the DB9 and everyone else is trying to catch up.
 
from the electronics engineering standpoint, yes, absolutely. mechanically - no. (optically we'll have to see if a OV will ever be released)
 
I think Red's philosophy is scaring somehow the market here in Brazil, they just can't buy the idea of a new revolutionary way of seeing the digital cinema market... I say that because there is a lot of people from here waiting in lines for Alexa and LOT'S OF PEOPLE ANXIOUS to shoot Alexa... Why?!?!?!?!?! people that never traded Film to Red now are open to Alexa?! unbelievable!!! This people scares ME out! and one more thing... everything about Alexa scares me out, size, quality, PRICE!! and rawless c'mon what could we do to get those people mind? Alexa and it's crazy effects on the market are going to be good for us Red owners?



thanks!!!!

Felipe Abreu

"On this crazy roulette I still keep all my chips on the Red!"
 
Status
Not open for further replies.
Back
Top