Technicolor teams up with Canon for DSLR's

[pulpfiction007]

IDk what you two ^^^ are talking about. The fact the Technicolor has changed the color codec from REC709 to LogC is a huge deal. It is the difference in a useable color space, all while increasing the dynamic range to the full potential of the sensor... I aplaudee Canon for this... and thank them greatly for doing it for free... like you said, Canon has no obligation to do anything for us, we already have purchased their product. Your already SOLD. Yet they still try, and they still push... to make you, me and everyone else as happy as they can without making their next products obsolete...

You are right about the Technicolor pairing being a great thing - as were the firmware upgrades for the 5DMK2 and the addition of new framerates/resolutions to the next releases. All I am saying is that new options are becoming available that will pose a challenge for Canon to continue holding on to it's dominance in this range of large sensor cameras. Like I said before they don't have to do anything, and I personally don't have the attitude "why isn't Canon giving us this or that?" I'm just analyzing the whole playing field and I think Canon as a company would benefit big time if the next generation of cameras has-
1. The 4:2:2 XF300/XF100 coedec (or similar)
2. Minimalized moire/aliasing
3. Technicolor or other cinema picture profiles.
That's it....I firmly believe the 5DMK3 will have these things, but I might be in the minority in thinking that.

 

[Perry Wilson]

You are right about the Technicolor pairing being a great thing - as were the firmware upgrades for the 5DMK2 and the addition of new framerates/resolutions to the next releases. All I am saying is that new options are becoming available that will pose a challenge for Canon to continue holding on to it's dominance in this range of large sensor cameras. Like I said before they don't have to do anything, and I personally don't have the attitude "why isn't Canon giving us this or that?" I'm just analyzing the whole playing field and I think Canon as a company would benefit big time if the next generation of cameras has-
1. The 4:2:2 XF300/XF100 coedec (or similar)
2. Minimalized moire/aliasing
3. Technicolor or other cinema picture profiles.
That's it....I firmly believe the 5DMK3 will have these things, but I might be in the minority in thinking that.

No I think you are correct in saying that the Mk3 will definitely be ground breaking, I have no doubts that it will carry the XF 4:2:2 or perhaps even better 4:4:4... but im not counting my chickens. sorry i didnt mean to make it a personal attack lol, I am just tired of rerading the same complaint post about comanies I.E. Canon/RED, who owe people sh*t. These companies are just that companies... they make products that we buy regardless. But I do believe that canon understands that they need to step up to the plate with new tech, and specs. I belive the best bet for canon would be to keep the same codec that they have implimented in their current dslr's... focus on the cameras true meaning which is stills... and for us film makers... All we need is 8bit 4:2:2 or 4:4:4 uncompressed HDMI output. This is technology that canon already has in their camcorders and could be easily moved over to dslr. With the invent of all these amazing and cheap external recorders, especially the Atomos Ninja, there is huge4 possibility in keeping the dslr what it needs to be for photographers, give the CF recording possibilities that every soccer mom dreams about, and lastly make true output fior the pro market and EXPECT them to outsource for external capture. This is the best bet and this is what will keep me buying these products. I am a through and through Canon guy always have been, always will be... for god sakes I have and aperature guide tattooed on my arm with the words "Nifty Fifty" and my 50 f/1.8MK1 serial number under it on my fore arm... Canon will make me proud as they always do... and I can almost promise that they will do so at Photokina

 

[pulpfiction007]

Perry...I'm with you on peoples expectations of these companies. I'm pretty amazed with what one can do at such a low price point shooting what's on the market right now. Even my $675 HF S200 can put out incredible pictures. I didn't take your post as a personal attack either - Perhaps I was being over dramatic in my statements about Canon and it's advertising - I'm just actually pulling for them to keep going and not be too conservative with the next gen. I love Canon products and have owned a 50D, an HFS 200 and I'm looking to get a 60D as well. I hope we are both right about the 5DMK3!

Actually the big 3 Panasonic, Sony and Canon have released some great stuff this year. Competition is good.

 

[Nektonic]

You are right about the Technicolor pairing being a great thing - as were the firmware upgrades for the 5DMK2 and the addition of new framerates/resolutions to the next releases. All I am saying is that new options are becoming available that will pose a challenge for Canon to continue holding on to it's dominance in this range of large sensor cameras. Like I said before they don't have to do anything, and I personally don't have the attitude "why isn't Canon giving us this or that?" I'm just analyzing the whole playing field and I think Canon as a company would benefit big time if the next generation of cameras has-
1. The 4:2:2 XF300/XF100 coedec (or similar)
2. Minimalized moire/aliasing
3. Technicolor or other cinema picture profiles.
That's it....I firmly believe the 5DMK3 will have these things, but I might be in the minority in thinking that.

Don't forget less jello from the rolling shutter.

I am all for reducing or eliminating the moire and aliasing that these cameras have, but I find the jello effect to be more of a pain in rear.

I am glad they are starting to address some of these issues though. The crushed blacks and blown highlights of the camera even under optimally tweaked settings being improved with this Technicolor scene file is gonna help for sure.

 

[Perry Wilson]

Don't forget less jello from the rolling shutter.

I am all for reducing or eliminating the moire and aliasing that these cameras have, but I find the jello effect to be more of a pain in rear.

I am glad they are starting to address some of these issues though. The crushed blacks and blown highlights of the camera even under optimally tweaked settings being improved with this Technicolor scene file is gonna help for sure.

I dont know which camera you are shooting on, but I have no issues with the "jello" rolling shutter on my 5D and 7D. Its all about a balanced rig... if you are shooting handheld you are asking for trouble... I think you need to work on counterbalance and weight... remember that yes though it is nice having a small light DSLR for video... but the heavier your rig the less vibrations affect it... My shoulder mounted rig weighs in at about 20/25lbs loaded up... perfect balance... and I can run with it without issues with rolling shutter... and sometimes you can use the rolling shutter to add dramatic effect to scenes... its all about the work arounds... everycamera has them, learn them use them, and avoid them when need be... but yes this is a good time to be a videographer... look at the new Sony NEX fs100 that camera is amazing... 4:4:4 8bit uncompressed output... Canon WILL do something big soon... they have too... in the mean time, love what you have, use it to the best of its abilities... and remember dont break your Canon right now... parts are insanely limited...

 

[Tzedekh]

I just hope Canon finds a way to do a full-resolution sensor -- that would eliminate the artifacts associated with line skipping and allow them to optimize the OLPF for the full resolution, which means decent stills and video out of the camera, as well a >4K video image. Maybe they could use something like the CMOSIS CMV12000, a 4K, 12-bit, global-shutter sensor, but increase it to 6K (to yield a 4K debayered image) and 16 bits.

 

[Danielvilliers]

I just hope Canon finds a way to do a full-resolution sensor -- that would eliminate the artifacts associated with line skipping and allow them to optimize the OLPF for the full resolution, which means decent stills and video out of the camera, as well a >4K video image. Maybe they could use something like the CMOSIS CMV12000, a 4K, 12-bit, global-shutter sensor, but increase it to 6K (to yield a 4K debayered image) and 16 bits.

I don't see this happening in the next 5 to 10 years for a camera costing 3K. Just the amount of data would be gigantic, more so in such small body camera. The GH2 has proven that you could have a high resolution true HD dslr video without moire and aliasing. Now just give us the same with uncompressed 10 bit hdmi output and it would be game on again for a next HDSLR salvo. Better readout from the sensor for less jello and the Technicolor Log for maximum dynamic range. Just iron out the problem of the first models and solid uncompressed output would be enough to hold a big part of the hdslr and at large large sensor video camera. One thing I have learn for all these manufacturers is always be prepared to be disappointed. Most of the time after a revolution they just go into an evolutionary update of their product line. I almost forgot the gh2 also does not heat.

 

[HKO]

That sensor is dope!!!

 

[Tzedekh]

I don't see this happening in the next 5 to 10 years for a camera costing 3K. Just the amount of data would be gigantic, more so in such small body camera.

Probably less than 5 years. Right now the CMV12000 would cost about $600 per unit in large quantities. And the improvements in specs I've suggested would probably be doable within 2 years, and might cost less than $300, given evolutionary march of fabrication technology. Red's doing it for $12k, but two years ago thought they could do it for about half that, and that's with 2008-2010 tech in relatively low volumes. With 2012-2013 tech in high volumes, the price should be much, much lower. As for the amount of data, Red Raw is compressed, and a comparable codec would be able to run at 42-200 MBps. Blackmagic's HyperDeck Shuttle can handle data rates over 180 MBps.

Three or so years ago, probably no one would've predicted how HDSLRs would take off or evolve. Who knows, maybe a company like Canon, who has no high-end video camera line to protect, will fight against their traditionally conservative nature and go for broke.

 

[Danielvilliers]

Canon will never buy a sensor from another company. In terms of resolution it is only a 8 megapixel, so a no no to hybrid dslr camera because of photo. I do think that if need be they could produce such a sensor (They already have shown all types of experimental sensor), but people have to be realistic if someone thinks that we will get 4k (8 megapixel) camera in a less than $ 3000 camera. The small dslr form factor is a not good also when you think that many of the high-end camera are actively cooled. What would be revolutionary would be a limited (in time) 2k-4k raw mode (canon has already s-raw). Good for movie makers because the take are short 99% of the time.

Now another false belief is because Canon has no high-end camera so it can challenge the Sony and Pana. What many forget is that Canon is one of the very few broadcast camera lens manufacturer. In fact they are imbedded with Sony and Panasonic professional camera division. That is why you never saw a camera higher than the prosumer range and they tend to come very late. They don't want to endanger their $ 10 000+ high-end lens market. Look at what they launched at nab!!! Can Canon do a 4k camera, sure they can (in a video camera body I mean for a no time limit one). Will they, I don't think so, for most of the engineering and marketing reason I listed above. For the last decade I have been in the digital photography and I say it again be prepared to be disappointed. With time it will come, but be prepared for more evolution than revolution.

I hope someone does a video head with one of those large sensors someday. In the video world it is more about niche high-end market than the mass market. There are two high speed $ 30 000+ camera that have been released at nab. They will sell more in the hundred. With their know how they could have produce a $ 4000/5000 video camera that would have sold in many order of magnitude more!!!

 

[Tzedekh]

Canon will never buy a sensor from another company.

Supposedly the Canon Powershot G11 had a Sony sensor. Not a particularly good example, but it does demonstrate that one should never say never, particularly if buying another company's sensor is expedient and the other company isn't a direct competitor (and CMOSIS isn't).

In terms of resolution it is only a 8 megapixel, so a no no to hybrid dslr camera because of photo.

I agree, and that's why I suggested a 6K (or 24-MP) sensor.

I do think that if need be they could produce such a sensor (They already have shown all types of experimental sensor), but people have to be realistic if someone thinks that we will get 4k (8 megapixel) camera in a less than $ 3000 camera.

With it's only slightly lower resolution 4K2K technology, it looks like JVC is aiming to bring 4K-ish video tech to the consumer market, although in a video-camera form factor and a smaller sensor size. The price point hasn't been established, but JVC, unlike RED in 2008, understands that $3,000 is hardly consumer level.

The small dslr form factor is a not good also when you think that many of the high-end camera are actively cooled.

Currently true, but with the addition of HD video to many mid- and higher-level DSLRs, I'm sure the manufacturers are working on it.

Now another false belief is because Canon has no high-end camera so it can challenge the Sony and Pana. What many forget is that Canon is one of the very few broadcast camera lens manufacturer. In fact they are imbedded with Sony and Panasonic professional camera division. That is why you never saw a camera higher than the prosumer range and they tend to come very late. They don't want to endanger their $ 10 000+ high-end lens market.

I think that's a false belief. Canon, like any mega-corporation, has divisions that don't necessarily coordinate well with one another, particularly if there's little overlap. Besides, you're conflating camera tech with lens tech. If anything, lenses have gotten more expensive, especially because of ever increasing demands made of them (speed, resolution, range, etc.). Besides, Canon's lens business largely consists of the broadcast/ENG market, which seems impervious to the HDSLR "effect." Barring a major technologic breakthrough that enables fast, high-quality zooms to be produced on the cheap, I seriously doubt that Canon's lens manufacture will be endangered. Look at Red -- if anything, its cameras have created a demand that is driving up lens prices, both new and used.

Look at what they launched at nab!!!

Canon typically doesn't announce new cameras at NAB.

Can Canon do a 4k camera, sure they can (in a video camera body I mean for a no time limit one). Will they, I don't think so, for most of the engineering and marketing reason I listed above. For the last decade I have been in the digital photography and I say it again be prepared to be disappointed. With time it will come, but be prepared for more evolution than revolution.

I hope someone does a video head with one of those large sensors someday. In the video world it is more about niche high-end market than the mass market.

Not if JVC has anything to do with it!

 

[Barry_Green]

Canon will never buy a sensor from another company.

This is simply not true. Did you know that the sensors in the Canon XL1/XL1s were actually bought from Panasonic -- they were the same chips as used in the AG-EZ1.

These companies share technology and parts all the time. The lens in the VX1000 was made by Canon, the LCD in the HVX200 was made by Sony, and the chips in the Canon XL1 were made by Panasonic.

 

[miko meloni]

Hi guys,
I'm not sure this is off topic,
but someone somewhere write about Canon's Curves by Light Illusion?
http://www.lightillusion.com/canoncurves.htm
Inside the link it's possible to find a very good article of the very accurate Alfonso Parra,
with some images.

I think it's impressive!
Waiting for something like that for Panasonic's shooters (GH1,2,AF100)!

 

[Nektonic]

At this point there is no way to know exactly what Canon will do, besides observing the fact that they are doing their usual waiting around to see what Sony and Panasonic are doing. The video camera market is so different and changes so quickly now. These aren't the old PD150 vs. DVX100 vs. XL1s/XL2 days. With all the recent shakeups over the past 4 years (Red One, Scarlet, video DSLR's, and now the AF100/F3/FS100), I think it might be harder to predict what Canon will do to compete, at least on the affordable digital HD cinema camera front. It would be great if they improved the video capabilities of their DSLR line before releasing an AF100 and FS100 competitor.

For all we know Canon could:

A) Not do much of anything on the digital cinema front besides add some improvements to their DSLR's
B) Release a competitor to the F3, or lower cost competitors to the AF100 and FS100
C) Continue what they are doing and stick with the XF 422 codec 1/3 sensor cameras like the XF300 and XF100.
D) Release something completely out of left field that will take us all by surprise

Who knows at this point? It is fun to try and predict though.

This is simply not true. Did you know that the sensors in the Canon XL1/XL1s were actually bought from Panasonic -- they were the same chips as used in the AG-EZ1.

These companies share technology and parts all the time. The lens in the VX1000 was made by Canon, the LCD in the HVX200 was made by Sony, and the chips in the Canon XL1 were made by Panasonic.

Barry, you are blowing my mind here. Those cameras are mutants!

 

[Danielvilliers]

This is simply not true. Did you know that the sensors in the Canon XL1/XL1s were actually bought from Panasonic -- they were the same chips as used in the AG-EZ1.

These companies share technology and parts all the time. The lens in the VX1000 was made by Canon, the LCD in the HVX200 was made by Sony, and the chips in the Canon XL1 were made by Panasonic.

I was talking about their photo division as we were talking about the DSLR. Since Canon has transited from film to digital, they have not used any sensor from any other manufacturer to my knowledge on the contrary of most of the other manufacturers (most of them have use sony sensor). On the rest of the camera like the lcd screen, I am sure that their have been a lot of parts that were bought from other manufacturers.

 

[Tzedekh]

I was talking about their photo division as we were talking about the DSLR. Since Canon has transited from film to digital, they have not used any sensor from any other manufacturer to my knowledge on the contrary of most of the other manufacturers (most of them have use sony sensor). On the rest of the camera like the lcd screen, I am sure that their have been a lot of parts that were bought from other manufacturers.

With respect to the photo division, see my comment above about the Powershot G11.

 

[Danielvilliers]

With respect to the photo division, see my comment above about the Powershot G11.

You said supposedly that it was a Sony so you are not sure. Even if it is the case it is a a market where almost everyone buys the sensor. In the market where it really counts, they use there own sensor and r&d. Now it might happen someday but it will be very difficult in there core imaging technology (dslr).

For years I have always dreamed about what I would call the good enough camera. That is an affordable large sensor hd camera with at least a 10 bit output and good dynamic range. Something that would approach the tools I have been using in photography (considering that moving pictures don't need the res of photos etc). A camera that would be good enough that it will be your creativity the limiting factor and not the tool. In photography for the last 3 to 5 years the tools have reached such a level that it is the photographer and not the camera which is the limit (even if a lot of people won't admit it). In today's world you have the Nikon d7000 body for a little more than $ 1000 with 16 megapixel and 14 stop of dynamic range. Yes 14 stops of dynamic range with 16 megapixel in an apsc sensor!!! I am just waiting for a successor to the D300 and D700 to see if I stay in the apsc or go full frame (it is becoming a long wait Nikon). I even bought a Canon 7d for video. For me it took some time to reach that level, and as Nikon user I had some waiting and disappointments before the Nikon line matured.( mainly compared to Canon who had a good 5 year head-start)

Now if you think that Canon will give go and buy a sensor from another manufacturer to give you 4k raw 16 bit video. I say better you prepare to be disappointed for a long time. For one, I was talking about the technological challengers that would be enormous and then about pure marketing logic. As I told you perhaps they might implement a limited time or burst mode video mode but I don't think we will have a Red Epic Canon camera for a foreseeable future. If tomorrow they come with a solid (a la gh2) video implementation with a 10 bit hdmi out, the Techicolor log to get 12+ stop DR (better sound would be good also) I will be more than happy. It will be a solid evolution and no one will have any excuse if they can't do some good images out of it.

Look at was for me the tipping point camera the Sony FS-100. It is everything I have dreamed, but it has only 8 bit output. I am still on the fence to buy it, I will have to see the different test to judge if the 8 bit disappointment warrants it. If it had 10 bit output, it would have sealed the deal. When I said to some about the 10 bit output, they told me if I wanted an F3 for $ 6000. In a world where you can get a 10 bit prores recorder for less than $ 1000 and an uncopressed 10 bit one for $ 350, some people are still defending the manufacturers. For now I will wait and see if it better than the next Canon to make my choice.

 

[HHL]

http://www.lightillusion.com/zippdf/canon_gamma_curves.pdf

Hey Barry! Thoughts?

 

[dust'n the callipygous]

so what does this mean in terms of dynamic range? will we realistically see an extra stop or more? dxo lists the 5D and 7D as having 11.7 and 11.6 stops. i think zacuto's test found them somewhere around 10 in video mode. i'd certainly welcome an extra stop of latitude. did anybody see a demonstration at NAB?

 

[pulpfiction007]

so what does this mean in terms of dynamic range? will we realistically see an extra stop or more? dxo lists the 5D and 7D as having 11.7 and 11.6 stops. i think zacuto's test found them somewhere around 10 in video mode. i'd certainly welcome an extra stop of latitude. did anybody see a demonstration at NAB?

Dxo measures the DR of the sensor's capability, centered around raw stills AFAIK. Zacuto's tests are centered around video mode where the sensor loses a lot of it's pure capability- in more ways than one - which we've all seen. If the video of a 7D looked like a 7D still we'd all be happy campers! One of the guys on the site did a video utilizing stills shot at at max rate....I forget which Canon model he used, so let's say 6fps. It was incredible, not like time lapse either...it was very different and you could see just how good these sensors are before they get mauled by the method of deriving 1080 from them.