Hi, did they send it with a small nick in the frame? Or before you sent it to them there was already a nick?I contacted HTC support, put in a repair ticket, sent it in, and had it back by end of the week. No charge, even though it is out of warranty period. Had mine back within about a week. Took them less than a couple of hours to do the repair.
And yes, I did get my original phone back. It had the screen protector and a small nick in the metal frame
Mine is stone cold and it still happens. Since it didn't happen prior to the 4.2.2 update, it definitely isn't a hardware issue.
So I was playing around a bit just to see what would make the pics better. I believe heat is certainly a contributing factor. I have an Otter Box Defender case which protects it well but, it may be part of my problem. Once I took the case off, the purple tint was virtually gone. Even pointed into a dark room, the screen stayed dark. The pics were okay but I found you really need the flash to be on to get a clear pic. My ISO was set to 200 and saturation was set to -1. Not sure if this will help anyone but I thought I would share.
Sent from my HTCONE using xda app-developers app
Just to add my 2 cents, so it’s not worth much. I tried an experiment bylocking the AE and AF on an object or color in a given light level, and then testing on a dark flat black surface. First I did it on a (low-light) red item, and got a fantastic blue noise. then i tried a low-light white object, locked that and got indigo noise. I then used a blue surface in bright light and had pure dark, no noise. When i remove the AE and AF lock, in the facedown all dark mode, it shifts to red noise.
So, the pattern is (as I see it) is all that the sensor chip does is collect the data,what we see is what the software renders from the sensor. White balance and ISOlevel have a huge factor in what color noise, and when. If you have your ISOthrough the roof and no real info hitting the sensor, the software willdecrease the software based tolerance for noise by increasing the ISO until allyou have is noise. After that, the noise color is dependent on the color tempof what the auto WB sees, and will gradually work its way to"daylight" or "cloudy" equivalency. You can test this bygetting to the point where you have red noise, and then manually changing thewhite balance through the selections. As far as a fix, the sensor is doing itsjob. The software would have to be tweaked to not get so noisy in an ultra-darksituation, and use an algorithm to remove appropriate color noise in high ISOsituations based on the detected color temp (WB).