# A naive MPS problem of idmrg

**URL:** https://itensor.discourse.group/t/a-naive-mps-problem-of-idmrg/510
**Category:** ITensor C++ Questions
**Tags:** dmrg
**Created:** [November 10, 2022, 12:58pm UTC](https://itensor.discourse.group/t/a-naive-mps-problem-of-idmrg/510 "2022-11-10T12:58:59Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![kevinh](https://yyz2.discourse-cdn.com/free1/user_avatar/itensor.discourse.group/kevinh/32/139_2.png) [@kevinh](https://itensor.discourse.group/u/kevinh)
#### Post date: [November 10, 2022, 12:58pm UTC](https://itensor.discourse.group/t/a-naive-mps-problem-of-idmrg/510/1 "2022-11-10T12:58:59Z")

</div>

Hi,ITensor

I have confirmed that I need to use infinite dmrg.Because idmrg is simpler than vumps, I have studied the relevant parts of the ITensor idmrg c++ version, and now I understand some of the principles, and will continue to learn it. However, I still have a naive question related to the code.

I am using the example in idmrg/sample/run.cc, I found that after executing idmrg, try to output

> Print(psi) and Print(psi(0))

I found the **ord** of site in **psi** = **3** , but when output **psi(0)** alone , **ord=2** , which confuses me a lot.

If this is the result of the principle based on idmrg, please forgive me for only learning part of it for the time being. Many thanks to all ITensor developers, thank you for your dedication!