# How to decrease Z error value? **URL:** \ **Category:** PIX4Dmapper Questions/Troubleshooting\ **Created:** [April 1, 2019, 8:11am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150 "2019-04-01T08:11:19Z")\ **Posts on this page:** 12\ **Page:** 1
**Author:** ![LS\_Survey\_Consultant](https://avatars.discourse-cdn.com/v4/letter/l/b3f665/32.png) [@LS\_Survey\_Consultant](https://community.pix4d.com/u/LS_Survey_Consultant)\ **Post date:** [April 1, 2019, 8:11am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/1 "2019-04-01T08:11:19Z")
    Hello,   As you can see, for GCP10, the error for z value is high compare to other gcp where I expected  below 5cm error. Is there any idea what causes this to happen and any solution for this to improve the one  gcp(gcp10) value. Thanks! ---
**Author:** ![Beata](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/beata/32/187_2.png) [@Beata](https://community.pix4d.com/u/Beata)\ **Post date:** [April 2, 2019, 1:04pm UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/2 "2019-04-02T13:04:53Z")
Hello, Indeed the 10th GCP characterises much bigger error than the rest of the points. Did you measure this GCP at the edge of the mapping area? Can you send us the screenshot of the GCPs location so we could analyse it better? As points that have been marked on a high zoom level are taken more into account than points that have been marked on a low zoom level I would recommend to zoom out them a little and unmark the GCP on the pictures like 0297 and 0298 - the target is a bit overexposed and can cause some problems. Can you give it a try and tell me what the result is? Best!   ---
**Author:** ![LS\_Survey\_Consultant](https://avatars.discourse-cdn.com/v4/letter/l/b3f665/32.png) [@LS\_Survey\_Consultant](https://community.pix4d.com/u/LS_Survey_Consultant)\ **Post date:** [April 3, 2019, 5:57am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/3 "2019-04-03T05:57:26Z")
Hi, Does that means that it is not necessary to marked gcp on every image where that one particular gcp are visible on the image? What i mean is for example like image 0297 where the marker seems to be overexposed, so its not necessary to marked that image with gcp?   Thanks! ---
**Author:** ![LS\_Survey\_Consultant](https://avatars.discourse-cdn.com/v4/letter/l/b3f665/32.png) [@LS\_Survey\_Consultant](https://community.pix4d.com/u/LS_Survey_Consultant)\ **Post date:** [April 3, 2019, 7:36am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/4 "2019-04-03T07:36:51Z")
  This is the screenshot of my gcp distribution. ---
**Author:** ![Beata](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/beata/32/187_2.png) [@Beata](https://community.pix4d.com/u/Beata)\ **Post date:** [April 4, 2019, 12:44pm UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/5 "2019-04-04T12:44:28Z")
Yes, it is true that it is better not to mark GCPs on the images that are under/overexposed or blurry because it increases the RMS error. Therefore could you give it a try and unmark the GCP on overexposed pictures (please check all of 26 images that you have for point) and let me know what is the result?  Best! ---
**Author:** ![LS\_Survey\_Consultant](https://avatars.discourse-cdn.com/v4/letter/l/b3f665/32.png) [@LS\_Survey\_Consultant](https://community.pix4d.com/u/LS_Survey_Consultant)\ **Post date:** [April 5, 2019, 12:55am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/6 "2019-04-05T00:55:00Z")
  Hi there, After I remove GCPs on blurry image, result on rms error getting worst :'(. ---
**Author:** ![Beata](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/beata/32/187_2.png) [@Beata](https://community.pix4d.com/u/Beata)\ **Post date:** [April 5, 2019, 12:22pm UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/7 "2019-04-05T12:22:39Z")
Hi! Thank you for trying. Please upload your images, .p4d file, GCP and marks to our [OneDrive](https://pix4d365-my.sharepoint.com/:f:/g/personal/beata_ciesielska_pix4d_com/En8FioeDZ71AjCtnQL7wdM4BE-Xz5x9-PkraFkvijkVKMg?e=rag7Bb). I think the best way would be if we process your project from our end.  Best ---
**Author:** ![LS\_Survey\_Consultant](https://avatars.discourse-cdn.com/v4/letter/l/b3f665/32.png) [@LS\_Survey\_Consultant](https://community.pix4d.com/u/LS_Survey_Consultant)\ **Post date:** [April 8, 2019, 2:08am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/8 "2019-04-08T02:08:27Z")
Hi,   Can check your drive. Have been sent.    THanks! ---
**Author:** ![Beata](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/beata/32/187_2.png) [@Beata](https://community.pix4d.com/u/Beata)\ **Post date:** [April 10, 2019, 9:55am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/9 "2019-04-10T09:55:56Z")
Hi, I’m downloading your dataset. I’ll come to you as soon as I have some updates. Best! ---
**Author:** ![Beata](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/beata/32/187_2.png) [@Beata](https://community.pix4d.com/u/Beata)\ **Post date:** [April 11, 2019, 11:18am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/10 "2019-04-11T11:18:47Z")
Hi, I didn’t mark the 10th GCP in 8 images ( you can see the labels in the screenshot below - I zoomed out everything that you can see the whole view) and the Z error is 0.005 [m] now. I would recommend you remark your GCPs avoiding adding them on blurry/overexposed images. I believe that it will help you obtain even better results. Moreover, be careful with the zoom - don’t zoom in too much unless you are certain that the measurement is accurate.  Best! ---
**Author:** ![LS\_Survey\_Consultant](https://avatars.discourse-cdn.com/v4/letter/l/b3f665/32.png) [@LS\_Survey\_Consultant](https://community.pix4d.com/u/LS_Survey_Consultant)\ **Post date:** [April 18, 2019, 12:39am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/11 "2019-04-18T00:39:31Z")
Hi, I still having the same problem. Looks like the computed position have big difference value to the initial position (ray cloud editor). In my opinion this may cause the error. Am I right? ---
**Author:** ![Daniel\_Rico\_Cordero](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/daniel_rico_cordero/32/187_2.png) [@Daniel\_Rico\_Cordero](https://community.pix4d.com/u/Daniel_Rico_Cordero)\ **Post date:** [April 23, 2019, 10:21am UTC](https://community.pix4d.com/t/how-to-decrease-z-error-value/8150/12 "2019-04-23T10:21:05Z")
Hello again, I processed your project using “All prior” and “Geometrically Verified Matching” and everything fits now. I also removed the marks in the blurry images. I even used the problematic point as Check and the residual is very low. You can download the result in a folder called “result” in the same link where you uploaded your project. Regards.