Welcome, Autel Pilots!
Join our free Autel drone community today!
Join Us

Evo 3 coming in 2022 (maybe). What will this mean for the Evo 2?

This is the downfall of just about every drone..... I have my eyes on an expensive drone from a small company. I believe in their equipment but am afraid they will go out of business and leave me with a multi-thousand dollar brick.
Any drone company selling >249g drones without the ability to rapidly iterate new models with Standard Remote ID built in (that can't be switched off) will be out of business in the US come September 2022, per the FAA Final Rule. X-Dynamics was cagey in answering the question whether RID is possible with current hardware of their Evolve2, saying only that they would be compliant with RID. (Given their timeline of 2 years to revamp the seriously flawed Evolve1, compliant might just mean they have an eight month window to sell current product in the US before they pull it from the shelves, and fold their tent. Same thing could happen to any other drone manufacturer, though I'm betting that Autel--in the #2 market-share spot-- will weather this)
 
  • Like
Reactions: Saladshooter
What Sony, Google, and Apple have demonstrated --what Autel is beginning to understand, as well, with the Nano+ and Lite+, perhaps?-- is that computational photography can allow smaller sensors to punch above their size/weight class. A 4:3 sensor is not so much bigger than a 1" sensor that it can automatically be assumed to be better. A tried-and-true, well-sorted 1" sensor can be better with computational AI noise reduction and stacking than a 4:3 without. The airframe and controller are sexy beasts on the X-Dynamics Evolve 2 but I've yet to see any footage from it that dazzles me compared to what is being done with the 1" Sony IMX383 sensors. (DJI Mavic 3 is better executed and does seem to show some improvements, though not obviously better in night shots.) Provided Autel follows through with putting a 4:3 camera in the AE3 they'd probably hope for it to be a knockout punch for the M3. The longer Autel waits, maybe the easier that becomes, while the sensor manufacturer (Sony) iterates and improves its own implementation strategies. Autel merely by using CFExpress rather than DJI's proprietary SSD, could be that knockout punch. This IMHO is the weakest link on the DJI M3 for professional use-- that it's going to be a major inconvenience to attach via cable and carry an external device just for the purpose of offloading that internal card (aka "get it in the can") after every shoot so as to not risk losing the footage with a subsequent fly away, rather than to simply pop out a card and pop in another. What was DJI thinking? Not to mention that CFExpress might allow the end user who already has a bunch in their pro-video arsenal to shave a grand or more off the price, instantly
 
Last edited:
Just remember this when you set your pricing for work. Most people leave off the recurring cost of drone and battery replacement when they calculate their cost to set their prices. This is also why those $50 jobs don't make sense to take, unless you can't get anything else. I figure that every job I do costs me at least $6 in replacement battery costs, since I get about 65-70 recharges out of a battery and I end up using at least part of two. You have to account for drone replacement too but that requires a guess on the number of flights you will get out of one and the time frame for replacement (plus of course, mileage, insurance, taxes...the usual suspects).
 
  • Like
Reactions: gschulzuio
What Sony, Google, and Apple have demonstrated --what Autel is beginning to understand, as well, with the Nano+ and Lite+, perhaps?-- is that computational photography can allow smaller sensors to punch above their size/weight class. A 4:3 sensor is not so much bigger than a 1" sensor that it can automatically be assumed to be better. A tried-and-true, well-sorted 1" sensor can be better with computational AI noise reduction and stacking than a 4:3 without. The airframe and controller are sexy beasts on the X-Dynamics Evolve 2 but I've yet to see any footage from it that dazzles me compared to what is being done with the 1" Sony IMX383 sensors. (DJI Mavic 3 is better executed and does seem to show some improvements, though not obviously better in night shots.) Provided Autel follows through with putting a 4:3 camera in the AE3 they'd probably hope for it to be a knockout punch for the M3. The longer Autel waits, maybe the easier that becomes, while the sensor manufacturer (Sony) iterates and improves its own implementation strategies. Autel merely by using CFExpress rather than DJI's proprietary SSD, could be that knockout punch. This IMHO is the weakest link on the DJI M3 for professional use-- that it's going to be a major inconvenience to attach via cable and carry an external device just for the purpose of offloading that internal card (aka "get it in the can") after every shoot so as to not risk losing the footage with a subsequent fly away, rather than to simply pop out a card and pop in another. What was DJI thinking? Not to mention that CFExpress might allow the end user who already has a bunch in their pro-video arsenal to shave a grand or more off the price, instantly
In line with what you say here, the biggest thing Autel could do to improve the video image quality in the Evo 3 would be to increase the data rate so they can decrease the amount of compression. CF Express (Type B, please) would certainly help that cause.
 
  • Like
Reactions: Saladshooter
In line with what you say here, the biggest thing Autel could do to improve the video image quality in the Evo 3 would be to increase the data rate so they can decrease the amount of compression. CF Express (Type B, please) would certainly help that cause.
There's a caveat to computational photography: where is the computation happening in the signal chain? The economics of making a sub-$2K drone require mass-market appeal as the primary target. While it is nice to have the ability to get RAW video codecs out, that means also whole bunch of time-intensive post-processing-- more than the average YouTube influencer or fire-department or real-estate shooter has time for. Hence, jpg-based MP4's and .MOV rule the roost, and they've proven more than adequate. Which means there has to be on-board processing. MicroSD has proven to be up to the task for this case use, and they're ever so small and light. I do agree that the 120 Mbps (aka 15 MB/s) bus that the Evo II now has is somewhat limiting, but it's light finger-flexing for a MicroSD to write that speed already-- any legit V30 card has a minimum write speed double this bus speed. But if Autel wants to future-proof their Enterprise line, CF Express would help that cause and also differentiate them from DJI
 
  • Like
Reactions: MikeReidPhotography
There's a caveat to computational photography: where is the computation happening in the signal chain? The economics of making a sub-$2K drone require mass-market appeal as the primary target. While it is nice to have the ability to get RAW video codecs out, that means also whole bunch of time-intensive post-processing-- more than the average YouTube influencer or fire-department or real-estate shooter has time for. Hence, jpg-based MP4's and .MOV rule the roost, and they've proven more than adequate. Which means there has to be on-board processing. MicroSD has proven to be up to the task for this case use, and they're ever so small and light. I do agree that the 120 Mbps (aka 15 MB/s) bus that the Evo II now has is somewhat limiting, but it's light finger-flexing for a MicroSD to write that speed already-- any legit V30 card has a minimum write speed double this bus speed. But if Autel wants to future-proof their Enterprise line, CF Express would help that cause and also differentiate them from DJI
I don't disagree with anything you say here. Except maybe that 100Mbit mp4s are "more than adequate." They're adequate, but not really more than that. And you're right, none of the influencers are really concerned about quality at all, so it's a non-issue for them. And for 1080 delivery, shooting in 10-bit 4K gives you enough of the benefits of downscaling for the EVO to deliver great material that cuts with the footage out of a Red just fine. But if you need to deliver in 4K, it's a different story. There are just too many compression artifacts showing up. You don't have to get to crazy-high data rates to eliminate those, and you don't even have to record raw. But getting up into the 300-400 Mbit range makes a world of difference. For one thing, it would allow 10-bit 6K recording, which would make a lot of difference in 4K delivery.
 
I've assumed the Evo3 will need a much faster bus than 120 Mbps to make CF Express useful. Perhap Evo3 gets a different main processor/CPU. Heat dissipation issues and battery life issues may require a clean sheet approach to the airframe. If it's so radical it's no longer modular with existing interchangeable gimbal mount, I'd hope they just diverge the Evo3 from the Evo 2/ Enterprise lines since they're already hitting a sweet spot for small folding drones there (ones that don't aspire to be alternatives to >$30-50K Hollywood cinema hexacopters or octocopters. I expect DJI and Xdynamics are going to fall flat on their faces within that sector). Beyond marketing hype it's hard to see where 4:3 cameras buy enough of a video improvement considering the crop factor needed to develop a mass-market appeal for a successful whole new product launch around-- especially not if the price is 50-100% higher (and it takes a post-processing professional cinematographer tack). Does a 4:3 sensor with the RYYB magic of the Nano+ Huwei sensor exist yet? It's straight-out-of-the-camera shots that will make or break such a mass-market product launch. With that camera, Autel is wisely piggybacking what's already being done in the smartphone sector (tech that may not be applicable to 4:3 sensors, maybe having color science that does not dovetail with RED or Blackmagic or Arri cameras)
 
I've assumed the Evo3 will need a much faster bus than 120 Mbps to make CF Express useful. Perhap Evo3 gets a different main processor/CPU. Heat dissipation issues and battery life issues may require a clean sheet approach to the airframe. If it's so radical it's no longer modular with existing interchangeable gimbal mount, I'd hope they just diverge the Evo3 from the Evo 2/ Enterprise lines since they're already hitting a sweet spot for small folding drones there (ones that don't aspire to be alternatives to >$30-50K Hollywood cinema hexacopters or octocopters. I expect DJI and Xdynamics are going to fall flat on their faces within that sector). Beyond marketing hype it's hard to see where 4:3 cameras buy enough of a video improvement considering the crop factor needed to develop a mass-market appeal for a successful whole new product launch around-- especially not if the price is 50-100% higher (and it takes a post-processing professional cinematographer tack). Does a 4:3 sensor with the RYYB magic of the Nano+ Huwei sensor exist yet? It's straight-out-of-the-camera shots that will make or break such a mass-market product launch. With that camera, Autel is wisely piggybacking what's already being done in the smartphone sector (tech that may not be applicable to 4:3 sensors, maybe having color science that does not dovetail with RED or Blackmagic or Arri cameras)
I’m not sure who a 4x3 sensor would be for?

I’m primarily a video shooter, so I live in 16x9. Or occasionally in dci 4K (4096 x 2160) which is even wider.

Yes, I’d like to see a wider bus that would allow a less-compressed data stream, but even ProRes at 6K will be under 300 Mbits. It’s not something that would require a complete re-design of the airframe.

As for cfexpress-B cards—I like them because they are very robust and proven, but I don’t really care. I’d just like to have a data stream and camera that gives me great footage for 4K delivery.
 
  • Like
Reactions: Brock Inspector
I think he meant a micro four thirds, MFT, M4/3, etc, sensor.
I don't use Micro Four Thirds or MFT or M4/3 terminology because it evokes a trade name for a lens mount consortium, nowadays partially defunct. What the format name derives from is the aspect ratio, and is why I use the expression as a ratio. I've heard some suggest recently that it's a four-thirds of an inch sensor. Which is as ridiculously overblown as calling a Sony IMX383 a 1" sensor, harking back to the Vidicon tube standards and outside diameter of a glass vacuum tube of the 1950's. Sure would rather we were naming these small sensors for the actual measured diagonal in millimeters and the true lens focal lengths. "Lens focal length equivalent" really leaves me cold. Save me the effin' math by putting the actual focal length in the specs, Autel. There is no such animal as equivalent focal length, only angle of view. \Rant off
 
Last edited:
As for cfexpress-B cards—I like them because they are very robust and proven, but I don’t really care. I’d just like to have a data stream and camera that gives me great footage for 4K delivery.
I don't find the Evo II deficient for it's weight class and niche. But as a pro stills shooter for 4 decades there are several things that could be improved upon or refined, yet.
One is the extraction of the MicroSD is slightly fussy because the damn things are so tiny, and because the hinge for the door is a delicate polyurethane strap that I miraculously haven't broken yet (I've had a secondary strip of gaffer's tape over the door that helps with securing it and opening it, since day one). I like that every Nikon Pro DSLR I've ever had is really robust re the card slot and extraction. Interchangeable cards with either the larger SD form-factor or CFExpress would be preferable. Simply for the better ergonomics of not having to pick the drone up off the landing pad, finding a soft spot that won't scratch the upper OA sensors while on it's back to get the MicroSD out, and not dropping the thing in some environment where I'd never find it, adds more drama than needed for lack of professional implementation.

Another annoying thing is ruining shots due to sun flare, partly due to no provision for a lens shade or baffle. I've jerry-rigged foam and tape ones, but forward flight or even hovering facing into a stiff breeze can knock the gimbal asunder or cause it to flop down with such a hood attached. Flare is a common enough problem with ultra wide angle lenses, but the FOV is not so extreme here. Alternately, nano-particle multicoated glass lens elements that don't flare so readily might help.
 
  • Like
Reactions: AirStreamPictures
I don't find the Evo II deficient for it's weight class and niche. But as a pro stills shooter for 4 decades there are several things that could be improved upon or refined, yet.
One is the extraction of the MicroSD is slightly fussy because the damn things are so tiny, and because the hinge for the door is a delicate polyurethane strap that I miraculously haven't broken yet (I've had a secondary strip of gaffer's tape over the door that helps with securing it and opening it, since day one). I like that every Nikon Pro DSLR I've ever had is really robust re the card slot and extraction. Interchangeable cards with either the larger SD form-factor or CFExpress would be preferable. Simply for the better ergonomics of not having to pick the drone up off the landing pad, finding a soft spot that won't scratch the upper OA sensors while on it's back to get the MicroSD out, and not dropping the thing in some environment where I'd never find it, adds more drama than needed for lack of professional implementation.

Another annoying thing is ruining shots due to sun flare, partly due to no provision for a lens shade or baffle. I've jerry-rigged foam and tape ones, but forward flight or even hovering facing into a stiff breeze can knock the gimbal asunder or cause it to flop down with such a hood attached. Flare is a common enough problem with ultra wide angle lenses, but the FOV is not so extreme here. Alternately, nano-particle multicoated glass lens elements that don't flare so readily might help.
Any hood attached should have a system to be mounted on drone and not on gimbal...probably can be made using a 3D printer and stitch to drone front arms and the locker where the gimbal cover is entering (in front)...Unfortunately i don't own a 3D printer...
 
I don't find the Evo II deficient for it's weight class and niche. But as a pro stills shooter for 4 decades there are several things that could be improved upon or refined, yet.
One is the extraction of the MicroSD is slightly fussy because the damn things are so tiny, and because the hinge for the door is a delicate polyurethane strap that I miraculously haven't broken yet (I've had a secondary strip of gaffer's tape over the door that helps with securing it and opening it, since day one). I like that every Nikon Pro DSLR I've ever had is really robust re the card slot and extraction. Interchangeable cards with either the larger SD form-factor or CFExpress would be preferable. Simply for the better ergonomics of not having to pick the drone up off the landing pad, finding a soft spot that won't scratch the upper OA sensors while on it's back to get the MicroSD out, and not dropping the thing in some environment where I'd never find it, adds more drama than needed for lack of professional implementation.

Another annoying thing is ruining shots due to sun flare, partly due to no provision for a lens shade or baffle. I've jerry-rigged foam and tape ones, but forward flight or even hovering facing into a stiff breeze can knock the gimbal asunder or cause it to flop down with such a hood attached. Flare is a common enough problem with ultra wide angle lenses, but the FOV is not so extreme here. Alternately, nano-particle multicoated glass lens elements that don't flare so readily might help.
I'm a Nikon shooter for stills too, and I totally agree with you on their choices of media. Always solid. The Evo II is not too compact to support a bigger, more robust card format.

And I agree on flare as well. I'd even be happy if the flares just weren't so damn ugly!
 
And what do you think the odds are the Evo 3 will be released bug free and even remotely when they say it will be? Evo 2 is just getting around to be useable.
This month will make 2 years since I've owned my Evo 2. I haven't had any problems with it yet!! Not one!! Everytime I updated the firmware it updated without incident. I even sent it in for an inspection like 4 months ago to make sure nothing is wearing because I've flown it so much. The only thing Autel did was give me new props. And they gave them to me for free. I have been pleased 100% hands down by this Autel drone. I would by and can't wait to buy another one! It's a solid drone that's easy to fly and steady in the air. I have a mavic 2 pro as well. And honestly flying with Autel is much much much less Cumberson, Than when I pull out the DJI. Their app is more user friendly than that of DJI and when your up in The air I feel like the Autel handles better and takes better snap decisions if you need to make a quick stop or vear around something like a tree. Everytime I have to make a snap decision on the DJI. It jerks around so much it looks like it's going to fall out of the sky. Thankfully you don't have to make those decisions very often. But when you do. I always worry the DJI is going to crash just by the way it looks like it handled the action in the air.. Actually, I've never had problems with either drone, but I worry less flying the Autel drone . I can't wait for the Evo 3
 
If they have managed to get the Evo2 and Evo2 Enterprise into industrial clients then they would be crazy to end support too quick... especially battery availability. Industrial clients tend to not welcome change and their capex demands maximizing operational lifespan.
 
  • Like
Reactions: Waldo Pepper
This month will make 2 years since I've owned my Evo 2. I haven't had any problems with it yet!! Not one!! Everytime I updated the firmware it updated without incident. I even sent it in for an inspection like 4 months ago to make sure nothing is wearing because I've flown it so much. The only thing Autel did was give me new props. And they gave them to me for free. I have been pleased 100% hands down by this Autel drone. I would by and can't wait to buy another one! It's a solid drone that's easy to fly and steady in the air. I have a mavic 2 pro as well. And honestly flying with Autel is much much much less Cumberson, Than when I pull out the DJI. Their app is more user friendly than that of DJI and when your up in The air I feel like the Autel handles better and takes better snap decisions if you need to make a quick stop or vear around something like a tree. Everytime I have to make a snap decision on the DJI. It jerks around so much it looks like it's going to fall out of the sky. Thankfully you don't have to make those decisions very often. But when you do. I always worry the DJI is going to crash just by the way it looks like it handled the action in the air.. Actually, I've never had problems with either drone, but I worry less flying the Autel drone . I can't wait for the Evo 3

It is rare these days to see positive things said about Autel products, possibly because their latest offerings are so problematic, but I agree 100%, I couldn't imagine flying any other done at the moment.

I will admit that you are braver than I am; I will probably never update the FW, every problem that my EVO has is easy to work around, I am not going to risk a FW update introducing a problem that I can't fix.
 
  • Like
Reactions: Waldo Pepper

Latest threads

Members online

No members online now.

Forum statistics

Threads
11,228
Messages
102,653
Members
9,818
Latest member
redwingaerials