No, but it might have DPAF.
Is DPRAW enough information to build a depth map with? To drag this conversation back to reality, given the frame rates the R5 is supposedly able to crunch, DPRAW could be a lot more usable on this camera.
Upvote
0
Canon announces more mind blowing specs for the Canon EOS R5
- Thread starter Canon Rumors Guy
- Start date
Is DPRAW enough information to build a depth map with? To drag this conversation back to reality, given the frame rates the R5 is supposedly able to crunch, DPRAW could be a lot more usable on this camera.
I do shoot raw. In fact, I only shoot raw.
But I'm forced to pass files around in post (DXO photolab) to print as jpgs (whose quality isn't as good as I'd like) or tiffs (that bloat up the storage too much). When I make huge panoramas in Ptgui I can use psb format to pass them to Affinity Photo which handles the huge files. There may be better ways to do things with Adobe software, but when they forced everyone to pay high monthly fees for their Creative Cloud platform I decided it was time to wash my hands of them and seek other post software. That's just my decision, and I mean no disrespect for those still using it.
What I'd like is for the camera to take raws, and have all the post processing and uploading be in heif format. And if the camera does some special processing, like multi-picture blending together as a software way to do long exposure neutral density filtering, it could save a single heif file which could be in an even higher pixel bit-depth & more noise-reduced image than the raw format allows.
Last edited:
Upvote
0
Agreed. If it can shoot 8k concurrently with DPAF, there is a LOT of data throughput capacity. If DPAF can tell a lens where to focus, then, yes there ought to be enough info to at least build a rudimentary depth map.
Upvote
0
Well, if I had access to the R&D that has been claimed, I would probably have access to enough technology, possibly additional small cellphone-like lenses & 3D sensing software, to create a fast accurate depth map. Or maybe they just quickly scan the whole range of focus of the main lens to do it if they want the max. resolution, otherwise use whatever depth resolution they get from the DP or QP(quad pixel) etc sensor.
Last edited:
Upvote
0
I hear you, bro. To each his own. For people who can take advantage of it, 8K is handy. For those who don't need it, obviously this camera isn't for them.
Upvote
0
I'm going to give Canon the benefit of the doubt because they've consistently undersold their tech specs for many years now, even though their products consistently do great work. This is why people also berate Canon for not having enough bells and whistles like the other brands. So the fact that Canon dares to publish these numbers so loudly, I think it's a show of confidence.
Upvote
0
Or it IS for them due to other things it does have. Especially if they are adult enough to know they can just ignore a feature they have no use for. (*couigh* *cough*)
(For what it's worth, I do a fair amount of casual video...for example, photographing an animal at the zoo then taking video if it happens to be moving. I don't expect to ever sell the videos or anything like that; they're basically for my own enjoyment. And for that, 1920x1080 is just fine. I guess one could call it 2K.)
Upvote
0
---
Binary Coded Decimal I was using in a LOT of Turbo Pascal and Turbo Prolog work where I was dealing with the equivalent of multi-thousand-bit sized integer, fixed point and floating point values for mostly Defence-related A.I. and Deep Aerospace Applications. BCD is actually STILL NEEDED for work at sub-atomic scales and macrocosmic scale for scientific visualization purposes when DOWN or UP-SCALING extremely small or extremely large numbers and then converting the BCDs to fit within 128-bit integer and real number values. We put BCD in because there are a LOT of values used in modern science that STILL does NOT FIT in even a 128-bit value!
We mostly use it for Plasmadynamic simulations and Quantum-Chromodynamic physics exploration where we are exploring (and Manipulating!) the very underlying fabric of SPACE AND TIME!!! --- SOMEONE has to design that fancy HUUUUUUGE-number processing code that calculates the initial or resulting energy values and superposition/pathways of virtual particles being created out of "nothing" as membranes hit each other or strings collide within their 21 dimensional non-Euclidean space! -- THAT WOULD BE ME!
We also don't want to accidentally nucleate a Vacuum Meta-instability Event and destroy the present universe via an expanding bubble of no-longer-false-vacuum, so we need to ACCURATELY calculate the precision of a number of Tera-Electron-Volt limit variables necessary to guard the Earth and this Universe from ACCIDENTAL TOTAL ANNIHILATION!
In terms of validating the NAME of our company, actually YOU WILL NOT BE ABLE TO !!!! We are completely under the radar and only a Revenue Canada investigation (i.e. only CRA aka the Canadian Government knows!) would be able to dig out the name of the owner(s).
In terms of 575 TeraFLOPS 128 bits wide GaAs superchip, THAT IS OUR SMALL ONE --- We're almost finished on the TWO THz (Two TeraHertz) version which will be a FULL 19.2 PetaFLOPS at 128-bits wide on a combined GaAs/GaN on Ceramic base substrate PER CHIP !!!
I never worked on Prodigy or Black Ops games and neither am I a Marcus Fenix Gears of War fan! BUT I am a low level systems coder with a WHOLE LOT of experience with Realtime 4K/8K/16k/32k/64k/128k resolution Video Capture/Compress/Encrypt/Transmit/Display systems and Realtime Vision Recognition Aerospace applications and hardware!
In terms of finding more about us, I can tell you that we've hid a full size 60 Ghz GaAs 119 ExaFLOP supercomputer (the world's FASTEST by the way!) in an underground facility by the PNE in East Vancouver from even various state-level agencies! They STILL haven't been able to find it EVEN AFTER stationing some bored multi-country lackeys in the Acura MDX, Mercs, Bimmers and Chevy in the parking lot of my Day Job (which has utterly NOTHING to do with my UNPAID off-hours "work" at my friend's facilities!) to follow me home and elsewhere!
They can't even find our hangar which I have EXPLICITY STATED ONLINE is at YVR in Richmond! There is an SSTO craft sitting there in almost PLAIN SIGHT yet they STILL cannot find or see it! So ANYONE ELSE trying to find out where and who we are would have some SIGNIFICANT difficulty! AND .... we have NOTHING to do with SWALA Aerospace or John Hollaway. We are entirely Canadian-based with offices across the country who does highly-specific scientific inquiry and specialty research and development jobs for certain clients that don't like being named!
Anyways, we are introducing some AWESOME imaging software products first that take video and stills processing to the next level, so you will be able to find out ALL ABOUT US in a short little while! Sooooo, I respectfully decline a tour of our factory. (it also happens to be my friend's factory NOT MINE, so I have no say in who gets in or out --- I don't even "officially" work there --- I just do "technical favours" for him!)
Complicated Isn't It?!
P.S. It's all REAL !!!
V
Last edited:
Upvote
0
++++ We are completely under the radar and only a Revenue Canada investigation (i.e. only CRA aka the Canadian Government knows!) would be able to dig out the name of the owner(s).
A.M.: Harry is the official Company nominated CR forum member responsible for the interstellar public relationships and also development of all the above top secret Space Command software development projects...
++++ It's all REAL !!!
A.M.: Who would have thought...
A.M.: Harry is the official Company nominated CR forum member responsible for the interstellar public relationships and also development of all the above top secret Space Command software development projects...
++++ It's all REAL !!!
A.M.: Who would have thought...
Upvote
0
Alpha Channel is actually used as a DISTANCE channel during recording where the 16-bits represents millimetres, centimetres, metres or other user-specified measurement units taken from an infrared scanning device similar to (but NOT the same!) as a Microsoft Kinect 3D scanning system built into the camera and lens system itself OR we take the distance information from an external long-range (multi-KM+) RF or Millimetre wave 3D-XYZ scanning devices attached directly to the camera. This allows VFX personnel to KNOW WHERE in 3D-XYZ space an actor or object is during filming of a scene and then MATCH that position and movement during insertion of VFX (Visual Effects) elements such as CGI cars, persons, buildings, etc.
We also use it for collision avoidance in aerospace applications and for realtime 4K/8K video object tracking. My software is capable of REALTIME automatic 3D-XYZ SOBEL edge detection, tracking, pixel-to-vector curve/line conversion and automated object recognition (and subsequent targeting/fire-control!) at up to 65,000+ SEPARATE OBJECTS PER SECOND based upon a current 200,000+ object library of comparison 3D vector objects representing planes, trains, ships, cars, trucks, buildings, ordnance, terrain, submarines, spacecraft and even people!
AND if you want to, you could also use it for plain-old ALPHA TRANSPARENCY CHANNEL purposes to composite digital objects and sets into a real-world actors and sets based upon user-specified distances or as simple image layering.
THAT is why we record RGBA or YCbCrA at up to 16 bits per channel! The more bits you set, the more distance accuracy you get!
V
Upvote
0
The 400 mm by 400mm CMOS image array YES does have a separate 3D scanning system (which is Multi-GHz RADAR when you're at FL3000 and above!)
The upcoming 50.3 megapixel MF camera system has a kind-of-like a Microsoft Kinect but to avoid being subject to various MS patents, we've had to make modifications to the scanning array by going a little more old-school from the early-to-mid 1990's era BUT it is full enclosed within the new MF camera itself and it automatically merges its 3D-XYZ depth scanning data with the RGBA/YCbCrA Alpha Channel which can be up to 16-bits in size which is quite-accurate for near-field distancing applications even for use as a general 3D object scanner within 3D modeling and animation programs!
V
Upvote
0
To all who don't need/want video: The thing is this: I think we e.g. only have good EVFs now and 20 pictures per second and DPAF and superior AF in general BECAUSE video became part of cameras. Video didn't work with optical viewfinders so they had to become EVFs. For faster video frame rates sensors and processing engines were updated significantly and now also the photos-only-users benefit from 20 pictures per second, 30 p raw bursts, enhanced AF and so forth. I doubt we'd have these possibilities as photographers already if there was no integration of video in photo cameras.
Upvote
0
---
This discussion is being played out here on the R5 thread BECAUSE new super large sensor Medium Format camera, large sensor super-smartphone and 128-bit desktop supercomputing technology is about to be released showcasing that Canon, Sony, Fuji, Nikon, Panasonic, Arri, Hasselblad, Phase One, Olympus and even Pentax ARE NOT the only games in town!
Someone much bigger and more aggressive is coming out to play and THEIR upcoming new products will UTTERLY SHOCK YOU ALL into a stone cold draw-dropping stupor !!!
Don't get me wrong! This new Canon R5 is a FANTASTIC PRODUCT for it's upcoming body-only price point of probably around $3600 US! If you're a Youtube vLogger or a New York/London/Hong Kong/Berlin/Tokyo/Vancouver/Moscow hipster documentary, studio or street photographer or a debutante film school grad, then THIS Canon R5 is the camera you want for your 24 fps 8k cinema videos and/or your Studio-quality 36 inch Prints!
HOWEVER, what is coming out fairly soon now, are the SUPER PRO-LEVEL CAMERA AND COMPUTER SYSTEMS that will take video and still photography to THE NEXT LEVEL !!!! What is coming out is a NEW WORLD of Video-centric AR (Augmented Reality), VR (Virtual Reality) and Premium-level Stills photography that will SURPASS ALL that is currently offered by Arri Alexa and Sony Venice for video and that from Phase One/Hasselblad/Fuji for medium format stills!
What is coming IS THE BIG KAHUNA BURGER of Combined Video/Stills MF cameras and large sensor super-smartphones AND I can definitely tell you that this IS a tasty burger!
And YOU WILL wash all of it down with this tasty beverage of very alcoholic red fruit drink cider!
AND YOU WILL VAROCIOUSLY ENJOY PAYING DEARLY FOR IT ALL !!!
Thank You and Good Night!
V
Last edited:
Upvote
0
Upvote
0
