I maintain, issue can be seen, except that nobody actually saw it before testing. After that of course you take care and then you notice. Not saying there is no issue, or that it's not good to fix it. I just think there's no need to make it appear so critical or to make a psychodrama whereas for actual end users it remained unnoticed so far. I'd far prefer usb3 full speed.
USB 3 speed issues were already fixed in firmware 1.4.6. I agree that this bug was extremely important, but there are other bugs to take into account and they should not be understated; this color issue is not a minor bug if you use your player as a mini computer too, not only as a media player.
boblo, to be able to see difference in a color, you must look simultaneously on two same TVs, with same calibration, displaying the same image from the the two boxes. If you know one is the reference then you can tell (or even not tell if the difference is small) if there's difference. If I show you on a non-calibrated TV (how many users in here have TV calibrated by ISF certified calibrator?) two reds, two greens, two blues, how will your mind know which one is the right one? To your eyes (and mind) even the wrong may be better.
It's very likely you are right, but if other competitors like Minix and nVidia do not have this issue, why a high-end media player have to have this problem? That's the question and that's the reason to fix it. Remember that @mirror wanted a proof that this issue is inexistent on other competing chips. Well, here it is.
We do not disagree with that and wesk05 gave to mirror the feedback he wanted to work on a fix. What I am saying is that the issue is a lot less important that you may think.
@nikos_a: It depends on the use of the player and how demanding you are. As you've noticed, I'm a very demanding user. For me it's unacceptable to have this issue on a 230€ device (X10), because for much less money there are players without this issue (as the very good Minix U9H). But they're not the same as the stunning X10 (input recording, UHD BD playback, a good display, internal bay, almost flawless auto frame rate switching...) And that's the point (at least, for me). On the other hand, I think people who do not want to fix this issue (not you, I know), or another one in the future which 'lacks importance' (for them) are not doing this community any favour. All fixes, for important or non-important bugs, should be welcome. @mirror: Thanks. You would make my day if Zidoo Team implement DVD folder structure playback (VIDEO_TS/AUDIO_TS folders) as Egreat already do...
Just to inform you how completely wrong you are - TV Panels get sold in different price brackets partly based of their color performance. The error margin the Zidoo messup adds, is large enough to send a panel right into the bottom tier, no questions asked. Its color performance is worse than its expected from the cheapest TV out there. This makes Zidoo boxes the worst Android Media players on the market just by applying the same logic. That it only affected ALL non video content, isnt a redeeming quality either. (Although someone tried to pull that PR stunt...) Thank you very much @wesk05 for testing other Android Boxes and conforming, that only Zidoo managed to mess up color performance in that "special" way - so we can actually recommend alternatives until Zidoo has been confirmed to have managed to get their act together. Also thanks @mirror for the acknowledgement of the issue - although it only came after you were somewhat sure you could fix it in a firmware update.
its a little scary seeing how we had 2 threads ( 1 locked twice than deleted ) till we actually got the problem acknowledged
Hi all: Please help us to test the follow version.It solved the GUI color issue. http://ota-cloudfront.zidoo.tv/X9S/OTA/OTA-ZIDOO_X9S-v1.4.8-201707181945.zip
You should have fixed some other little things too, like "unknown audio format" message when playing MKV with HD audio or other easy fixes. At least, it will make your clients happy for the time being. Then you can take more time fixing other serious stuffs for the next release.
C'mon, they use the version number 1.4.8. If it is just one fix, why can't they just come up with APK file, click and install. For me, I will do clean install every times I deploy new firmware. I have to go through a hassle to do it. So for this one, I will pass.
Just installed the experimental firmware and tested its color reproduction. (Calman with the Mobile Forge App on the X9S.) Issue is NOT resolved, but worsened - new (unbelievably bad) issues are introduced. 1. Neither YCbCr 422 nor YCbCr 444 output a limited range signal anymore, when manually selected. They all output 0-255. Thats a new issue and its really bad. 2. The only way to have the X9S with the new firmware output a limited 16-235 signal now, is to set the color space to auto, and then reboot. Thats a new issue and its really bad. 3. At which point this happens (and I had to do a triple take to believe it, double checked my TV settings, did two restarts and it is reproduceable): You have managed to leave the rec 709 color space entirely - with the X9S set to 1080p60 - which shouldnt be possible at all. I dont know how you managed to do it - but you did. This is not a fix, this just worsened he issue. 4. As soon as you even switch ANY video setting in the menu (even lets say from 12bit to 10bit) the entire signal reverts to 0-255 (full RGB, pc level) again, without any way to change it back. Unless you reboot again (with it set to auto). Your task was to fix the video level output (16-235 = YCbCr 422 and YCbCr 444) for the UI and app interface on the X9s, what you did with the patch is to prevent it from being selected at all with any setting other than auto, and then messed up the video levels even further (beyond what I thought was even possible, because the TV (LG 2016 OLED) now thinks this is a HDR signal, even though 1080p60 is selected.). Let me put this as plain as possible - TAKE SOME REAL MONEY, throw it at someone who can wrap his head around this, fix it. Test it. Then release it. Your competitors managed to get colors right, so get your act together and follow their lead. As of now - I have no confidence in Zidoo producing a TV box that outputs correct colors. Period. (I hope they can prove me wrong.) Also, as a customer it is not my job to do the testing for you. Pay someone to do it for you, before you release a firmware publicly. Its in your own interest to do so.
@n_p: I'm grateful for your contribution and efforts, but you should be understanding with technical staff because they are trying to do their best and with your irritation we won't win anything (you included). At least, you and me want to have this issue fixed, right? So keep calm and wait for @mirror 's response. Thank you.
I stick this issue out with you guys, as long as Zidoo is officially proclaiming a fix. I really cant help, other than keep saying that they should hire someone with a basic understanding of color spaces and signal levels in the video realm. This isnt that hard - we are talking about extremely basic stuff here, tbat everyone who has ever picked up a colorimeter (a thing to calibrate a monitor) can do for them. I can repeat, that if the Mobile Forge app from the Google Playstore doesnt work for them, they should look into getting the App from the Amazon App Store instead. The app at startup looks for the Amazon store to be installed and a user thats logged in, thats allowed to use/own it - so in the worst case scenario, you'd have to buy a a Fire TV Stick and download it from there (if you cant download it from a cellphone, because of licensing limitations.), then install the Amazon store on the X9S. Having Mobile Forge installed, allows you to do those readings automated and within seconds. Before today I'd even have said, that it would be a better idea to get in contact with @wesk05 and ask him, how he does his bit level readings, because, those are very easy to interpret. Either you hit them, or you dont - but then today - I was presented with a X9S that "auto" configered itself to a HDR (bt2020) color space, even though 1080p60 was selected as an output resolution. And thats just unheard of. This isnt supposed to happen, ever. And as soon as the box switched to a full signal level (0-255) output, it was back to a pretty decent (as in correct, but without further (more intensive) tests) rec709 color performance. If you mess up on that level, looking at bit values alone will not help, because you have severely messed up the devices signal handshake - so as things get procedurally worse, I'm actually getting more hesitant to push the issue, because with the last firmware - at least the limited (16-235)/full (0-255) select (YCbCr 422 and 444 vs Full RGB) worked. Now thats broken as well. I dont want people out there to have a worse device from when Zidoo "proclaimed" they were "fixing stuff" as a result of pointing out where they arent able to do what would be standard compliant. Also - every post like this, from someone like me is a mid level PR disaster for them - so, for their own sakes, they should fix it on their own, and then present the (hopefully correct) result. Its not that no one could have figured out, that this firmware doesnt do at all, what it was supposed to. If your black level doesnt change switching between YCbCr and Full RGB, you messed up. Every nerdy teenager whose ever owned a Playstation could have told them that.
