Worse: ZKTeco 8" Worse Temperature Results Than 5" Tested

By Derek Ward, Published Jun 16, 2020, 03:15pm EDT (Research)

***** ****** ***'* *** ******** ******'* *********+ *" **** *******, ****** *** *" ******* "***********" the *" *******, *** **** ******* shows **** *** *" *** *** same ******** ** ***** ** *** 5" *** ********* ***********.

IPVM Image

** ****** *** ****** ******+, *********:

  • *** **** ** ******* ** *** 5" ****** ******-*+?
  • ******'* ******** ******** ** ** ** thermometer.
  • **** ****** ** ****** *** ******** have ** *********** ************?
  • **** *** ****** ****** **** ** easier ** *** **** *** ***** views?

********, ** *** ********* ***** ** Dahua *** *********'* *********** ******* **** a ******** ** *****, *********, *** ZKTeco.

Executive ******* - *" ** *"

***** ********** ******* ***** **** *** ZKTeco ******+, ** ********* ***** **** the ******-*+ ********* ******** **** ***********, with ******* ****** ** *** ******** and ***** ******* ** *** ***. Several ************ **** **** ** ****** it *** ********* *********, *** ***** miss *********, **** *** ****** ******* low ** ~*°-*°* ******** ** ** Extech ** ***********.

******** *** ******** ****** ****** **** the **** *** **** ********, **** errors **** **** ******** ***** *** far **** (**** **" *****, ** our *******, ******** ******** ***** ***** to *** ****), *** ***** *** tall ** *** ***** (*** **** has ****** **** ****** ******* ** a **" ****** *****, *.*., ******* over *' ** ***** *' *").

** *** ******** ****, *** *** 8" ****, ** ** ****** ** read *** **** ** *** ****** for **** ******** *** ****** (****** shown ***** ************* ****). ****, *** 8" *** ******** ****** ********** (***** in ******** ******** *****) *** ******** larger ******* *** ***** *** ***** (not ****** ***** *** ***** **** was ***********).

*******

*** ****** ******+ ***** *** ~$*,*** USD ******, ~$*** **** ********* **** the ******-*+ (~$*,*** ***). *** ******+ is **** ********* **** ********** *********** tablets **** ***** (***-********-**) *** ********* (DS-K1T671TM-3XF), ***** **** **** *** ~$*,*** USD ******.

Missing ******** **** ************

*** ****** ******+ ****** ******** **** temperatures *****, **** ********* ************ ** forehead, ***** ******* *** ******. ************ of **** ~***° ** ******** ** our ******** ****** ** ************, **** ZKTeco ******* ** ~*°** ****.

IPVM Image

** ********** *** ****** ******** ***** during ******* *** *** ******'* ************, and *** ** ********* ** *** missed ******** **** ************ ** ******** subjects *** ** ********* ***** ** the ****.

Bad ***** / ****** ******* *********** ********

** *** *****, *** ******* ******** from *** ****** *** * *********** impact ** ******* *********** ************. ****** technical ******** *********** **" *********** ********, but ***** ************* ****** **** (~**.**"), even ******.

**** *** *********** ***** ******** ** 18" ** *** ******, **** ***** may **** *******, **** *** ****** too ***** ** ********** ** ***** personal *****, ****** **** ***** ** standing ******* **** **** ***********, ******* results. ** *** ********* **********, *** subject ****** ~**" **** **** *** reader, *** **** ** ~**" **** at *** ******'* *** ********, **** a ********** ~*.*° **********.

IPVM Image

** *** ********** *****, ** **** a ******* ******** ~**" *** ~**" away **** *** ******. ***** *** close ***** *** **** ** *** reader, **** *** ********* **** **** from *** ****** *** **** ***** personal *****. *** ********* **** *** reader **** ***** **** *** *******'* nose ** *** ******.

IPVM Image

Hardware *********** *** ************

*** ******* ******** *************** ******* *** 8" *** *" ****** ****** ******* screen **** ** ***** ******* ******* capacity *** ***** *** *****. *** SF1008+ ** **** ** ***** **,*** faces *** *,*** ***** ****** *** SF1005-V+ ******* *** *,*** ***** *** 3,000 *****.

IPVM Image

******* ********** ** *** ******+ **** a ****-**** ********** ** ***, ******* the ******-*+ ** ******** **** * Dual **** ***, ******** ** *****'* seen * ******* ********** ** *********** measurement ** **** ******. *** ********* hardware "**********" ** *** ******+ **** the ******-*+ *** *** *********.

IPVM Image

************, *** ******* ****** ****** *** identical, **** *** ************* **** (**** for *** ******-*+, **** *** *** SF1008+).

IPVM Image

** *** ***** *****, ** ******* the ******** ************ ** *** *" ZKTeco ******+, ** **** ** **** it *** ****-**-**** ******* *** *" ZKTeco ******-*+.

**** **** *** *" ***** **** not **** * *********** *******, *** the ****** ** *** **** ** fixed ** *** ******, ****** *** 5" *****, ***** ******** ********** ** manually ******* ***** ** *** ********* they **** ** ***.

Screen **********

**** *** **** **** *** ***** setup ***** *** ****** ** **** with *** *" ******+ **** *** smaller *" ******-*+. **** *** ****** model, ******* ******** *********** ************ *** instructions ** ****** **** ~**" **** than **** *** *" *****.

IPVM Image

************, *** ***** *** ******** **** of *** *" ***** ** ****** to ******** *** ****, ** ***** and **** *** ******, ****** ** less ****** ***** **** *********** **** an ****** ** *************.

IPVM Image

Size **********

*** *" ******+ ** **** ****** and ***** **** *** *" ******-*+. The *" ***** ** ******* ** size ** ***** **** ****** ********* on *** ******.

IPVM Image

Made ** *****

*** ****** ******+ *** ****** ******-*+ were **** ** *****.

IPVM Image

Firmware ******* ******

******* *** ********* ***** *** ****** SF1008+ **** ******** ******* ******-**-****.*.**. **** firmware *** ******** ** ** ** ZKTeco ****** *** *******.

Comments (7)

I don't think the 2012 and 2014 numbers are manufacturing dates. It does not seem likely that these boards have been sitting around for 6-8 years, just waiting for a use.

Agree: 1
Disagree
Informative
Unhelpful
Funny

I would say it's more likely that the date in the box is the manufacture date:

V1.0-200213 - V1.0-YYMMDD (Feb 13, 2020)

Agree
Disagree
Informative
Unhelpful
Funny

I don't think it's fair to stand 24" away from the device (when the maximum is 18", or within 15.75" for highest accuracy) for a temperature read only to say that results are off by 2-3 degrees. Why stand further away than the device's stated capability? People know its limitations prior to purchasing it.

Agree
Disagree
Informative
Unhelpful
Funny

Raymond, good question!

Having used the device, naturally, I stood significantly farther than 18" away. To me (and others), it feels odd to stand 18". Some people may feel more natural to stand 15" or even closer.

Secondly, the device gives a reading at 24" away so as an ordinary user, the expectation is that the reading is good. If ZKTeco wants to ensure readings consistent with their specs, I would encourage them to not give a reading when people are farther away.

Third, I just tried this when opening the refrigerator. I naturally (without thinking) stood ~24" from the door and it makes physically sense since arms are in the 30" range.

People know its limitations prior to purchasing it.

ZKTeco does list this on its specification sheet:

Body-temperature detection accurate up to 0.6 F from up to 18 inches away

And that is good disclosure. It's still a competitive usability issue, e.g. see the Dahua temp station which worked at significantly longer distances.

Raymond, do you think longer-range reads are something customers want?

Agree
Disagree
Informative
Unhelpful
Funny

If you think it feels odd to stand within 18 inches of the device, then buy another device! But don't buy the device, stand further than 18 inches, and then test it and report on why it doesn't work outside its stated capabilities. That does not make any sense whatsoever.

Yes, I believe that longer-distance temperature reads are important in certain cases, but not all. For example, a lot of companies who have purchased the ZK units from us want the optional payroll integration so that each temperature scan constitutes a clock-in. That is more important to some companies since they can ditch their antiquated T&A machines and save time on payroll.

For others who are not primarily scanning employees, the Dahua DHI-ASI7213X-T1 kiosk/terminal would be more ideal since subjects will not have to walk up to the device but rather past the device. It improves efficiency in addition to scan range. But Dahua did state that although the temperature range is 6 ft, accuracy is increased when subjects stand closer to the terminal. I leave it up to IPVM to test the Dahua DHI-ASI7213X-T1 at varying distances WITHIN its stated specifications.

(Both the SF1008+ and SF1005-V+ can do facial recognition at up to 8 ft away, but palm-vein ID and temperature detection are limited to 18 inches.)

Agree: 1
Disagree
Informative
Unhelpful
Funny

If you think it feels odd to stand within 18 inches of the device, then buy another device! But don't buy the device, stand further than 18 inches, and then test it and report on why it doesn't work outside its stated capabilities. That does not make any sense whatsoever.

It makes total sense. First, we reported how it worked at 18 inches, so it's not as if we ignored it. But we also reported at further distances that we suspect many people will naturally gravitate towards, especially since the ZKTeco continues to generate readings (albeit, worse incorrect ones) at longer distances.

Returning to your earlier comment:

People know its limitations prior to purchasing it.

To the contary, for virtually everyone buying these products right now, this is their first time using such a device for temperature readings, ergo even if they carefully read the 18" spec, they likely do not have any idea the practical meaning of that since they have never experienced it.

Agree
Disagree
Informative
Unhelpful
Funny

I would like to see more specifics on the temp comparison testing. A grid that shows actual test results. Headers like Device type, Distance from device, Thermometer temp, device temp with ALL the test results listed. I did not read your report the same way that Raymond read it and a table with the actual data would provide clarity on these questions. It would also allow comparisons between devices and distances vs. accuracy.

This question of distance from device matters as does how the operator controls for this. Since this is an access control device there is likely to be no operator. If I put a piece of tape on the floor at 18" and say stand here people will put their toes on the line. Since the human foot is ~9 to 12 inches long the face itself might be 26" or more from the device.

If the device is allowing a read from >18" and admitting the user, that is an issue with the device. If a max. distance of 18" is a requirement then the device needs to refuse to read at a distance greater than 18". Remember this is not an operator mediated read so the device itself MUST control the parameters of when it will and will not grant access.

Agree
Disagree
Informative
Unhelpful
Funny
Subscribe to IPVM Research to read the full report.
Why do I need to subscribe?
The IPVM Research Service includes products tests and shootouts plus competitive and financial analysis, helping decision-makers better evaluate purchasing, partnering, developing, and/or competing against companies in physical security.
Already have an account?
Loading Related Reports