👓 Eachine EV300D FPV Goggles - bit.ly/2vyCd3I 💸 Get $35 off with coupon code BGPREEV300D Or BG333D Or BGOTC21 💸 7% off with code AlbertKimRC 💸 $259.99 price with code BGPAC238 for new accounts 👉 Video on my current favorite goggle - th-cam.com/video/YNRlTqqeTak/w-d-xo.html
This is interesting. I believe one big flaw with the stopwatch experiment is that the times on the stopwatch are not accurate. This is because the phone screen is most likely 60hz. This means that the stopwatch will proceed in increments of about 17ms for each frame.
U nailed it imo, lcos panels can not provide the refresh rate needed for high end fpv goggs inherently due to how they work. Firmware may be able to scrape down, but it will never be able to match other display tech. And I can feel 10-20 ms difference of latency flying high rates freestyle w crossfire for sure.
I have never seen such screens but they were not laggy because of the refresh rate. The entire world of TV runs at 24 frames (>40ms) it slowly changed to 30fps. So no, it's not laggy or a slideshow, we watched all the action movies in 24 fps @@sheeren6188
This is just me guessing. First the phone stopwatch test probably has to do with the fact that most phone screens are like 60hz, there are some specific android phones that will do 90hz or 120hz. So that means there is a lag time between each number change already of 16.67 ms. Also Using a website to run the stopwatch most likely powered by javascript isn't the best way to do it either. If they are manipulating the DOM to draw the characters/numbers I imagine there is some latency in that as well. I would do something more like a drop test of a long item with white/black squares and numbers on each and then compare the two screens instead. That gives you a sense of motion more so than relying on another screen. The light/black works well, but then I wonder how much of it is the backlight nature of lcos (probably using 3 LEDs to create the color meaning their brightness levels need to be in sync with the display instead of OLED where you can turn the individual pixel off quickly.
If it is close, the price factor is the deciding factor. Still cheaper than the hdo3. I still fly with the ev200d, they rock for there price. So, ev300d, great for the price. Thanks for the update and the effort!👊👊👍👍
Well ... i am a regular fpv pilot. Not much extras in my flight - not top notch, not bottom line. But at around 200 bucks there is simply no better one.
1. An aspect is known, ev200 which has the same screen technology, doesn't have such an issue... so still exist some hopes 2. And if the reason is its screen technology, maybe it has different refresh rates depending on how much information has changed since previous frames So... ...in the case of stopwatch, the whole image was the same, changes were just in the region of running digits, so there's not so much to redraw on the screen ...but in the case with turning on and off the light, much more information has come to the screen for being refreshed, resulting a bigger delay It's the same how works the image encoding, pictures that have more different elements weigh a lot more memory than a photo of one coloured wall But the good part is that day and night do not change so many times during a several minutes of flight, so usually the luminosity is the same... but here you will say "Well, but what about diving from light room, to a dark one?.."... Try and see, if there is some difference, but anyway you won't fly in completely dark room, so there will not be a need to completely darken the screen, phenomenon that might be the the center point here
Ok. Upgraded mine. What's worse is that the image quality is now really quite horrible. Flickering/jumping image and very noisy. Albert - did you manage to compare image quality? On mine it's a lot worse, especially inside / at night. Latency is worse than the previous version ~60ms on PAL (didn't bother testing NTSC). This is up from 30ms that I measured with the previous version, but down from 120ms I had with the original firmware. And compares to 30ms I had with a cheap Eachine fpv monitor. (Measuring with a stopwatch is ok if: you take multiple pictures and average and use it to compare to other goggles/screens using the same technique. There may be an offset between my measurements and someone else's but the absolute different between goggles will be the same.)
If you look at the phone based timer test, it seems that you could describe the difference between frames of that video with much less information than what would be needed to manage the difference between frames of a light to dark example. The phone timer test image data is far more uniform, with the changing areas of the video frame to frame being much smaller. The light to dark test comes closer to requiring a full refresh on every frame, which seems like it might be causing the Eachines to struggle especially if the LCOS screens have a more elaborate refresh system.
Thank you Albert for staying on top of this. My biggest concern at this point is not the latency but the source from which the goggles are purchased. The latency problem may have improved but customer service may still be the same (okay but difficult) as it relates to a certain online retailer. Looks like a great goggle but my money is on Fatshark and Skyzone! At least for now.
Do you know if the ev200d goggles use the same screen panel technology ?(i.e. LCOS) and they appear not to have this latency problem. If so, maybe there is hope that the EV300D can be fixed with only a firmware upgrade. Let's hope. :)
I measured the latency with an oscilloscope, there is a thread on rcgroup where i show the results. In short: stock fw, 96ms latency, screen ok Update v1: 20 to 28ms of latency, slow refresh on the screen sometimes. Update v1.1: 40 to 56ms of latency, flickering screen. I havent hdo2 or any other goggles so i cant compare directly the results except for my old vrd2pro, the latency is between 10 to 15 ms
I tested both PAL and NTSC... see the video link in the description for the test comparison. I noticed the latency was a little better than before the update.
I would give up if i were you Albert as a reviewer it shouldn't be your job to fix a poorly made product. seems that they just cant be bothered to test this stuff themselves. why do they send out products like this that clearly have issues. goggles that have latency drones that dont fly out of the box with software thats 12 months old. really confusing to me
There is no latency to the naked eye on a running displayed counter, its a constant display to which the camera is recording? Maybe I’m missing something obvious there, but that confused me a bit!
FPV camera pointed at screen with counter so it shows the latency in the screen on the goggles. You can just see the difference in the numbers to determine the latency. But this is also making a bunch of other assumptions that are way beyond the scope of this video and will take a whole video series to explain further. In a nutshell, this is complicated stuff and not that simple.
Albert Kim - Yes sir, I understand that, but you mentioned the unknown of the latency of the counter display itself, which to my thinking, which is flawed many times, there wouldn’t be? At least nothing those goggles are going to pick up, as that would be the scan of the actual display itself?
![[Full Episode] The Restaurant War Thailand ศึกพ่อค้าซ่าแม่ค้าแซ่บ Episode 2 | 29 ก.ย. 67](http://i.ytimg.com/vi/r3lfLNH6Ke4/mqdefault.jpg)
👓 Eachine EV300D FPV Goggles - bit.ly/2vyCd3I
💸 Get $35 off with coupon code BGPREEV300D Or BG333D Or BGOTC21
💸 7% off with code AlbertKimRC
💸 $259.99 price with code BGPAC238 for new accounts
👉 Video on my current favorite goggle - th-cam.com/video/YNRlTqqeTak/w-d-xo.html
This is interesting. I believe one big flaw with the stopwatch experiment is that the times on the stopwatch are not accurate. This is because the phone screen is most likely 60hz. This means that the stopwatch will proceed in increments of about 17ms for each frame.
U nailed it imo, lcos panels can not provide the refresh rate needed for high end fpv goggs inherently due to how they work. Firmware may be able to scrape down, but it will never be able to match other display tech. And I can feel 10-20 ms difference of latency flying high rates freestyle w crossfire for sure.
You certainly can not feel 10ms difference in latency. You are not superman
@@haka8702 Oh yes you. I remember 30ms screens being used at grocery shops and they were always laggy and only good for slideshows.
I have never seen such screens but they were not laggy because of the refresh rate. The entire world of TV runs at 24 frames (>40ms) it slowly changed to 30fps. So no, it's not laggy or a slideshow, we watched all the action movies in 24 fps
@@sheeren6188
This is just me guessing. First the phone stopwatch test probably has to do with the fact that most phone screens are like 60hz, there are some specific android phones that will do 90hz or 120hz. So that means there is a lag time between each number change already of 16.67 ms. Also Using a website to run the stopwatch most likely powered by javascript isn't the best way to do it either. If they are manipulating the DOM to draw the characters/numbers I imagine there is some latency in that as well. I would do something more like a drop test of a long item with white/black squares and numbers on each and then compare the two screens instead. That gives you a sense of motion more so than relying on another screen. The light/black works well, but then I wonder how much of it is the backlight nature of lcos (probably using 3 LEDs to create the color meaning their brightness levels need to be in sync with the display instead of OLED where you can turn the individual pixel off quickly.
I use the ev200d goggles and have no problem with latency but if I can improve my goggles, Im in.
Thank Yoy very much.
If it is close, the price factor is the deciding factor. Still cheaper than the hdo3. I still fly with the ev200d, they rock for there price. So, ev300d, great for the price. Thanks for the update and the effort!👊👊👍👍
Well ... i am a regular fpv pilot. Not much extras in my flight - not top notch, not bottom line. But at around 200 bucks there is simply no better one.
They shouldn't release beta products the way they do.
They completely ruined the good reputation gained with EV200D...
That's too bad
1. An aspect is known, ev200 which has the same screen technology, doesn't have such an issue... so still exist some hopes
2. And if the reason is its screen technology, maybe it has different refresh rates depending on how much information has changed since previous frames
So...
...in the case of stopwatch, the whole image was the same, changes were just in the region of running digits, so there's not so much to redraw on the screen
...but in the case with turning on and off the light, much more information has come to the screen for being refreshed, resulting a bigger delay
It's the same how works the image encoding, pictures that have more different elements weigh a lot more memory than a photo of one coloured wall
But the good part is that day and night do not change so many times during a several minutes of flight, so usually the luminosity is the same... but here you will say "Well, but what about diving from light room, to a dark one?.."... Try and see, if there is some difference, but anyway you won't fly in completely dark room, so there will not be a need to completely darken the screen, phenomenon that might be the the center point here
Great job, Albert! Thanks a lot! 😊
Ok. Upgraded mine.
What's worse is that the image quality is now really quite horrible. Flickering/jumping image and very noisy. Albert - did you manage to compare image quality? On mine it's a lot worse, especially inside / at night.
Latency is worse than the previous version ~60ms on PAL (didn't bother testing NTSC). This is up from 30ms that I measured with the previous version, but down from 120ms I had with the original firmware. And compares to 30ms I had with a cheap Eachine fpv monitor.
(Measuring with a stopwatch is ok if: you take multiple pictures and average and use it to compare to other goggles/screens using the same technique. There may be an offset between my measurements and someone else's but the absolute different between goggles will be the same.)
I haven't tested image quality. Will do that after the latency is fixed.
any update on that ?
If you look at the phone based timer test, it seems that you could describe the difference between frames of that video with much less information than what would be needed to manage the difference between frames of a light to dark example. The phone timer test image data is far more uniform, with the changing areas of the video frame to frame being much smaller.
The light to dark test comes closer to requiring a full refresh on every frame, which seems like it might be causing the Eachines to struggle especially if the LCOS screens have a more elaborate refresh system.
Thank you Albert for staying on top of this. My biggest concern at this point is not the latency but the source from which the goggles are purchased. The latency problem may have improved but customer service may still be the same (okay but difficult) as it relates to a certain online retailer. Looks like a great goggle but my money is on Fatshark and Skyzone! At least for now.
Can you use rapid fire with these
Do you know if the ev200d goggles use the same screen panel technology ?(i.e. LCOS) and they appear not to have this latency problem. If so, maybe there is hope that the EV300D can be fixed with only a firmware upgrade. Let's hope. :)
I'm not sure... some say it does, others disagree...
I measured the latency with an oscilloscope, there is a thread on rcgroup where i show the results. In short: stock fw, 96ms latency, screen ok
Update v1: 20 to 28ms of latency, slow refresh on the screen sometimes.
Update v1.1: 40 to 56ms of latency, flickering screen.
I havent hdo2 or any other goggles so i cant compare directly the results except for my old vrd2pro, the latency is between 10 to 15 ms
you need to test NTSC and PAL.
It seems the latency differs by nearly a magnitude between those cam formats
so Ive been stocking your channel for your review of the Xflip oled goggles ... i heard a rumor that you got a set..????lol
Where did you hear this?
Hi Albert thanks for the update. Where you in PAL or NTSC? Also did you notice latency when using the goggles after update
I tested both PAL and NTSC... see the video link in the description for the test comparison. I noticed the latency was a little better than before the update.
Ok so if we used rapid fire module u think will the latency will change or that has nothing to do with it?
Module has nothing to do with this problem
I would give up if i were you Albert as a reviewer it shouldn't be your job to fix a poorly made product. seems that they just cant be bothered to test this stuff themselves. why do they send out products like this that clearly have issues. goggles that have latency drones that dont fly out of the box with software thats 12 months old. really confusing to me
There is no latency to the naked eye on a running displayed counter, its a constant display to which the camera is recording? Maybe I’m missing something obvious there, but that confused me a bit!
FPV camera pointed at screen with counter so it shows the latency in the screen on the goggles. You can just see the difference in the numbers to determine the latency. But this is also making a bunch of other assumptions that are way beyond the scope of this video and will take a whole video series to explain further. In a nutshell, this is complicated stuff and not that simple.
Albert Kim - Yes sir, I understand that, but you mentioned the unknown of the latency of the counter display itself, which to my thinking, which is flawed many times, there wouldn’t be? At least nothing those goggles are going to pick up, as that would be the scan of the actual display itself?