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 :)

Forged CF Red Weapon #000027 has been delivered to Kosmos/ dino

I think I can make out some smearing on the EPIC camera that is not on Weapon. If this is the case, this is WONDEFUL! Thank you for taking the time to do this for us Dino, I certainly appreciate it!

Its not actually where I would expect the smear, so its hard to tell. I have attached a cutout to show people where I am looking.

Have not had heart burn, I mean eaten at Pinks, in a long time! :) Actually I ate at the Vegas Pinks during NAB, though I am not really sure that counts.
 

Attachments

  • EPIC_Smear.jpg
    EPIC_Smear.jpg
    19.4 KB · Views: 0
Stacey, I agree. That is what I saw too, which was very similar to what you pointed out in your example.

Now Jarred has so graciously met me at RSH and gave me a loaner Lo light Olpf. I have installed it into the dragon, I did it in the parking lot of the palladium because it was shaded and no one was around, thus allowing me to focus on installing the OLPF and not worry about the homeless. Now I have a better rig to test with, which is what I will do.
 
If you can find a Xyla 21 test chart from DSC Labs, it makes it easy to repro the smearing. The first, brightest, chip causes the smear to occur. Worst case, I can ship you my Xyla 21 and you can test and then ship back when finished, if you want. Then you can also measure the EPIC and Weapon using the same technique. Its a backlit chart, so all you need is a dark room and a stand. You can measure one chip at a time or all the chips at once or cap off the chips you don't want in the way.

The thing about the smearing is that it does not always happen when out shooting. Its strange for sure. I think Phil Holland spent some time trying to figure out under which circumstances it occurs so that he can avoid it. I also know Phil wanted to borrow my Xyla chart. Maybe you can give the chart to him after you are done testing and he can ship back. Either way, I will pay to get to you if someone will pay to return it. :)

You can PM me if interested.

Here is a thread Gavin started when we were comparing the MX sensor to Dragon using the Xyla.
 
Stacey, I will want to borrow your Xyla when I return from China. Heading home on the 16th.
 
here is the answer to that question...

which is NO SMEAR ON WEAPON....

this is the same exact shot from two separate cameras on the same head

Y500_C032_0809HE.0000010.tif

this is the Dragon with skin tone at 4000 ISO ( which i would never do in real life btw) DC2, RG4. 32 MM Master prime, 180 shutter no grade, straight out of RCX tiff Snapshot


Z500_C033_0809IY.0000042.tif

weapon 6kHD low light OLPF, , 4000 ISO, DC2, RG4, 27 MM Masterprime, 180 shutter - no grade, straight out of RCX tiff Snapshot

Cannot make the images load, unfortunately.
 
here is the answer to that question...

which is NO SMEAR ON WEAPON....

this is the same exact shot from two separate cameras on the same head

Y500_C032_0809HE.0000010.tif

this is the Dragon with skin tone at 4000 ISO ( which i would never do in real life btw) DC2, RG4. 32 MM Master prime, 180 shutter no grade, straight out of RCX tiff Snapshot


Z500_C033_0809IY.0000042.tif

weapon 6kHD low light OLPF, , 4000 ISO, DC2, RG4, 27 MM Masterprime, 180 shutter - no grade, straight out of RCX tiff Snapshot

Images? R3D's plz
 
Dino, if you want, I can ship you the Xyla this week and then when you are done with it, you can pass on to Phil when he returns from China.
Stacey, very cool of you! And to me this exemplifies the spirit of this forum. And I know Dino and Phil will put that chart to good use.
 
If you can find a Xyla 21 test chart from DSC Labs, it makes it easy to repro the smearing. The first, brightest, chip causes the smear to occur. Worst case, I can ship you my Xyla 21 and you can test and then ship back when finished, if you want. Then you can also measure the EPIC and Weapon using the same technique. Its a backlit chart, so all you need is a dark room and a stand. You can measure one chip at a time or all the chips at once or cap off the chips you don't want in the way.

The thing about the smearing is that it does not always happen when out shooting. Its strange for sure. I think Phil Holland spent some time trying to figure out under which circumstances it occurs so that he can avoid it. I also know Phil wanted to borrow my Xyla chart. Maybe you can give the chart to him after you are done testing and he can ship back. Either way, I will pay to get to you if someone will pay to return it. :)

You can PM me if interested.

Here is a thread Gavin started when we were comparing the MX sensor to Dragon using the Xyla.


Thank you for the offer, fortunately I am working tuesday-friday on real jobs, so i will not be able to use it, that is why i am pushing so hard now because the next chance i get might not be for a couple of weeks.

just got back to the lab after being out...time to download and play a bit then sleep and hit it again tomorrow.

thank you
 
If Weapon has the same sensor as Dragon . . . what would be the difference between the cameras that would make a change in the CMOS smear? If there is a difference, why would that not be available to Dragon owners as a retrofit / upgrade / update?
 
If Weapon has the same sensor as Dragon . . . what would be the difference between the cameras that would make a change in the CMOS smear? If there is a difference, why would that not be available to Dragon owners as a retrofit / upgrade / update?

I wish you're right about that. I can see that a few parts of the design seems to be different, including processing, though.

A de-smear in RCX would be cool if it's done through processing.
But if it's done in camera before recording/encoding, that might not be plausible.
It is for sure an issue has been kept warm on since the release of the Dragon, and no matter CMOS characteristics, I would guess that was pretty high on the engineering teams priorities when re-building a camera bottom-up based on the Dragon sensor.
Alongside faster internals (which hardly is just turning the knob up a notch, as all have to play nice together), and ergonomics.

It's kinda hard to say how much change is needed while the sensor is still "the same". (or v2/v3/v4). But from the pictures released so far, it doesn't seem different enough to be a "new" sensor. If they didn't do any improvement they could while at it, it would be kinda strange, though.

See Alexa.
That is supposedly the same sensor as in 2011, but I'd be hard pressed not saying it has evolved a fair bit since first release.
Most of those evolvements have come as upgrades, and they haven't all been software.
(which BTW makes the "4 year old sensor shooting 2k baked prores" argument not so relevant as it may first seem when comparing the platforms. I wouldn't directly compare the Alexas released today with those first released. Not only because they don't only shoot prores.)
 
Last edited:
boot times for my cameras..

dragon x @00031 = 35 seconds
weapon #000027 = 47 seconds

i am sure they will get the weapon down a few more seconds as the firmware evolves.




If Weapon has the same sensor as Dragon . . . what would be the difference between the cameras that would make a change in the CMOS smear? If there is a difference, why would that not be available to Dragon owners as a retrofit / upgrade / update?


regarding the smear....u can clearly see it on the dragon as stacey pointed out , and not at all on the weapon. not sure how else to prove it as they both have very similar lens es 25mm master prime and 32 mm master prime, they are mounted less than an inch from each other on the same horizontal plane, same stop, same iso, same color temp...what am i missing...i see it with my own eyes ...here are the r3d's trimmed...the two tiffs are there and the RDC with the r3d;s which have been trimmed

http://weapon8k.com/weapon000027/
 
27 is a really meaningful number, i've done entire musical compositions related to the power of 9's.. good on you!

Ladies and Gentlemen:


This afternoon I met our spiritual and logistical leader Brent Carter at RSH in Hollywood,CA where he handed me a very stylish pelican-like case containing lots of laser-cut foam padding and one Forged Carbon Fibre Weapon serial number #000027. Now this will require a little rebranding as I jumped the line a little and now I am 4 ahead of 31, but not to worry I'll find a way to make the connection for my customers and followers. The camera will be put through its paces tonight and tomorrow. stay tuned!!!

Thank jarred, Jim and brent as well as chris and michael, and of course all of my customers whom have kept me on the leading edge.

You can go on my Instagram to see a pic. @dinogeorgopoulos

Dino
 
boot times for my cameras..

dragon x @00031 = 35 seconds
weapon #000027 = 47 seconds

i am sure they will get the weapon down a few more seconds as the firmware evolves.







regarding the smear....u can clearly see it on the dragon as stacey pointed out , and not at all on the weapon. not sure how else to prove it as they both have very similar lens es 25mm master prime and 32 mm master prime, they are mounted less than an inch from each other on the same horizontal plane, same stop, same iso, same color temp...what am i missing...i see it with my own eyes ...here are the r3d's trimmed...the two tiffs are there and the RDC with the r3d;s which have been trimmed

http://weapon8k.com/weapon000027/

Great!

Thx Dino
 
Downloaded the TIFF files, tried to match levels and balance colors. Ran a noise reduction pass on the Weapon still and applied the same correction to Dragon. Weapon is cleaner in this scenario and like others have mentioned does not have the CMOS smear that we all hated.
W-D.jpg


Thanks Dino for running around and doing all these first looks. It's greatly appreciated!
 
Looking nice. Curious. When I tried to bring in the Red files the images were corrupted. I am on latest RC 35, is there something I am missing or doing wrong?
 
Looking nice. Curious. When I tried to bring in the Red files the images were corrupted. I am on latest RC 35, is there something I am missing or doing wrong?
Sometimes that happens with the trims when they are uploaded I'll try it in a little bit...that was they last thing I did before I slept...just gettin going.
 
Back
Top