Newb with Questions About Reolink Cameras

[mat200]

Got our EmpireTech 1/1.8 4mp bullet cam (IPC-B54IR-ASE-3.6mm) up and running with the DMSS iOS app in our backyard yesterday. According to Andy it has the same sensor/optics/firmware as the turret version (IPC-T54IR-AS 3.6mm). Only difference is the bullet has one IR LED vs. two on the turret.

Very pleased to see the DMSS iOS app allowed immediate access to the alert video clips. It also kept a running log of alerts. Both capabilities were completely missing from the Reolink iOS app. However, so, so disappointed with the DMSS playback function---it was slow, erratic, and cumbersome--almost unusable at 8x (max). Playback in the Reolink iOS app is very stable, even at 16x (max).

Gotta say the 1/1.8 EmpireTech cam's night-vision is damn good. Our backyard only has small, downward facing 2w landscaping lights. It was enough lighting to provide good visual coverage of our entire backyard without the IR LED on (did notice motion coverage decreased without the IR LED active). If you turned off the IR LED on the RLC810a Reolink, all you could see were the landscaping lights!

The daytime resolution and general picture/color quality of the 4mp EmpireTech was a noticeable step down from a 8mp RLC810a Reolink that was mounted in the same location. I tried adjusting brightness, hue, saturation, etc. on the EmpireTech, but not a lot of improvement. Biggest disappointment was the EmpireTech's lack of daytime dynamic range (with WDR turned on and off). In the late afternoon as the sun went down, the shaded areas looked great, while some of the sunny areas were overexposed and totally washed out. In the morning, there was significant lens flare as the rising sun hit the camera lens from the side. Disappointing to see these issues in the EmpireTech. The Reolink had none of these issues.

The EmpireTech's nighttime motion detection was noticeably better than the Reolink. During the daytime the Reolink's human/vehicle/pet detection feature was reasonably accurate and reliable. Night-time bugs created a lot of false alerts on the Reolinks. The EmpireTech not so much. I adjusted the motion sensitivity on both cameras during testing.

For our needs, there's a lot of significant pros and cons going on here. I was so hoping the $170 1/1.8 4mp EmpireTech AND DMSS iOS app were going to be a clear winner vs. the $80 8mp RLC810/$100 8mp RLC811a Reolinks AND Reolink iOS app. Didn't quite work out that way. As an aside, have-to-say the RLC811a is an especially feature-rich cam for $100. Optical zoom, acceptable night-vision (with low/moderate ambient lighting) even with the IR LED/spotlight off, and outstanding 8mp/4k daytime picture quality.

As a homeowner, with no crime issues, that just wants to monitor their yard with 5-6 POE cams, POE switch, micro SD's, and a reasonably capable iOS app, there seems to be no clear path to happiness. Tough choices ahead.

Hi @HomeWPoe

Do check out the DORI section of the cliff notes .. effective pixels do matter.

There's a lot of nuances to this, and often we try to keep things simple enough for new folks to get a better understanding. In the old days of photography, people would take classes to get to a base level of photography .. getting into f-stops, lens quality, exposures, film quality, shudder speeds, et al ..

From a basic POV for starters, just look at pixels, think of effective pixels ( a reason some of us prefer H264 vs H265 and maximizing bitrates yet lowering fps to better balance storage consumption .. that is picking 15fps or less instead of 30 fps )

a 8MP camera vs a 4MP camera ( same ratio H FOV vs V FOV ) .. will give you about 1.5x range if the pixels are equally effectively captured. The 4MP 1/1.8" sensor cameras do really well in lower light conditions vs 8MP 1/2.8" cameras as the "effective pixels" captured by a properly tuned 4MP 1/1.8" camera will far out do the 8MP 1/2.8" ( where the effectiveness of image capture suffers in low light ) .. now under the best light conditions we typically see the 8MP 1/2.8" outperform the 4MP 1/1.8" camera .. what does this mean ?

MORE CAMERAS .. basically, one camera is not going to do well enough to get what those new to this topic image. As a result new folks will often find they want to add a couple more cameras to cover some of the more critical views ( often the driveway / street juncture or front door )
This is why I personally always recommend running an extra cat5e/6 cable to each position during the cable pulling work, as normally it is about the same amount of work - and the price difference often just the extra cabling.

 

[HomeWPoe]

Again you are thinking too logically lol.

As I have said, just because it works for one manufacturer doesn't mean it works for another...

As @Teken has said, the SD card recording process for Dahua is crap. Most use SD card as redundant recording and not as sole recording. As a result Dahua hasn't invested into that as say reolink has.

Heck, which browser you use dictates the speed you can transfer off the card (10mbps versus 100mbps)....so you don't think the whacked out programming in DMSS isn't the result of your issue trying to play off an SD card...

Downloading videos files directly from camera via web interface (using other than IE) seems to be capped at *TEN* Mbps

UPDATE: If you use Internet Explorer (with plug-in), you can download videos via the web interface at the full 100Mbps speed - note you can't go any faster because the NIC doesn't support GigE. Thanks @TheOtherMike for pointing that out on the data sheet (maybe for 2024, Dahua will go with...

ipcamtalk.com

had.

Even the codec you are using could be an issue. If you are using H265 try H264.

I guess it is time for you to give BI another try lol

As far as backlight, have you tried HLC, SSA, etc.

In my post I concluded the hardware was up to the task based on the Reolink iOS app's excellent playback performance. Be default that puts the onus on the DMSS iOS app. That's why I posted "The playback problem appears to be with the DMSS iOS app itself." Are we not saying the same thing?

I don't believe I tried H264. Thanks, I'll give that a try.

Can you clarify HLC, SSA? Odds are DMSS/EmpireTech calls it something else.

As for giving BI another try, do you feel the BI iOS app would provide the alert/playback functionality we're looking for? If not, which iOS apps besides Pushover are necessary to attain this functionality? Thanks.

 

[looney2ns]

If you are having lens flare, that is a mounting/positioning problem. Not the fault of the camera.

No matter how you try to justify the Reolink garbage, when it becomes important to get a face capture, the EmpireTech cam will blow it out of the water, IF you have taken the time to properly adjust the camera to the field of view as per @wittaj instructions.

Friends don't let friends buy Reolink crap.

 

[HomeWPoe]

Hi @HomeWPoe

Do check out the DORI section of the cliff notes .. effective pixels do matter.

There's a lot of nuances to this, and often we try to keep things simple enough for new folks to get a better understanding. In the old days of photography, people would take classes to get to a base level of photography .. getting into f-stops, lens quality, exposures, film quality, shudder speeds, et al ..

From a basic POV for starters, just look at pixels, think of effective pixels ( a reason some of us prefer H264 vs H265 and maximizing bitrates yet lowering fps to better balance storage consumption .. that is picking 15fps or less instead of 30 fps )

a 8MP camera vs a 4MP camera ( same ratio H FOV vs V FOV ) .. will give you about 1.5x range if the pixels are equally effectively captured. The 4MP 1/1.8" sensor cameras do really well in lower light conditions vs 8MP 1/2.8" cameras as the "effective pixels" captured by a properly tuned 4MP 1/1.8" camera will far out do the 8MP 1/2.8" ( where the effectiveness of image capture suffers in low light ) .. now under the best light conditions we typically see the 8MP 1/2.8" outperform the 4MP 1/1.8" camera .. what does this mean ?

MORE CAMERAS .. basically, one camera is not going to do well enough to get what those new to this topic image. As a result new folks will often find they want to add a couple more cameras to cover some of the more critical views ( often the driveway / street juncture or front door )
This is why I personally always recommend running an extra cat5e/6 cable to each position during the cable pulling work, as normally it is about the same amount of work - and the price difference often just the extra cabling.

@mat200, thank you for the explanation. I've read the DORI section. Can't say I'm fluent on its contents, but it is one of the reasons I could somewhat appreciate how/why the 1/1.8 4mp EmpireTech cam came recommended by a number of folks here on ipcamtalk.

Have to say, low light capability was not the main objective for purchasing this EmpireTech cam. A functional iOS app was. Going with the EmpireTech cam gave us an opportunity to try the DMSS iOS app. Unfortunately, (as I stated in my post earlier today), both the cam and app fell short in a few areas. If I can improve the DMSS iOS app's playback performance, we may be good moving forward with more EmpireTech cams.

 

[wittaj]

In my post I concluded the hardware was up to the task based on the Reolink iOS app's excellent playback performance. Be default that puts the onus on the DMSS iOS app. That's why I posted "The playback problem appears to be with the DMSS iOS app itself." Are we not saying the same thing?

I don't believe I tried H264. Thanks, I'll give that a try.

Can you clarify HLC, SSA? Odds are DMSS/EmpireTech calls it something else.

As for giving BI another try, do you feel the BI iOS app would provide the alert/playback functionality we're looking for? If not, which iOS apps besides Pushover are necessary to attain this functionality? Thanks.

No we are not saying the same thing.

You keep concluding the Dahua camera with an SD card is up to the task just because the Reolink camera is up to the task, without understanding the underlying conditions that caused each to perform the way they did? Two totally different manufactures and totally different end client expectations. I don't expect a Hyundai to perform as well in speed and handling and acceleration as a Ferrari, although I would expect the Hyundai to do better in snow LOL.

Remember Dahua's market are businesses, not us homeowners. Reolink's market is homeowners. We are just fortunate enough to be able to get our hands on these better Dahua OEM cameras because we favor better quality images.

You must not believe me that everyone says a shortcoming of Dahua cams is how it deals with SD cards. That isn't a big market for them so they haven't focused on optimizing it. Many people buy a reolink without an NVR. so they have optimized their app experience for that user. Most buy Dahua with an NVR or use another VMS system like BI and don't rely on SD cards as the sole recording.

Maybe it is because Reolink is known to play with bitrate and iframes and FPS and other parameters that could have compressed the video a lot to not be as demanding as the video from the Dahua cam and that could be the reason for the differences in performance. So it may seem like it can handle it, but it has been compromised quality and you don't realize it. And maybe that compromised quality is fine for you based on the size of the phone. If it is then run the EmpireTech at lower quality to get that same performance as the Reolink.

Maybe the lackluster performance you are experiencing is because you are using the SD card instead of an NVR...I don't recall @bigredfish complaining about scrubbing video at 8x with his NVR (he runs 30FPS and higher bitrates also)....Mine doesn't. But if I watch playback from the SD card thru the camera GUI it can get wonky.

For kicks, make the Empiretech H264, 10FPS, 40 iframes and a bit rate of 1,000 and see how the playback performs. I bet it improves....

So maybe lowering the requirements for the Empiretech cameras give you what you need from DMSS or maybe adding an NVR gives you what you need for DMSS?

Regarding the other Backlight conditions, no EmpireTech doesn't call it something else. You have WDR, BLC, HLC, SSA to chose from, along with the changing of shutter speeds and other parameters. These settings are all done within the camera GUI, not DMSS.

But as someone else suggested, show us the images you are having trouble with so we can better assess and offer suggestions.




Also keep in mind that DMSS with one camera and no NVR is one thing, add 4 or 5 cameras and see if you still like it.

When I was using it with one camera, it was fine, but trying to manage 4 cameras with it and it got over complicated and time consuming and slow and an NVR was needed.

Maybe BI with the iOS app is what/all you need. At the end of the day your requests are simple - be able to get an alert and play back an alert immediately and scrub at Xtimes without bugging out.

The BI app is fine with that. Most of us don't use it because we isolate our systems from the internet, so the only way to get an image with the BI app is to port forward, use stunnel, or leave a VPN on all the time on our phones.

So most of use the Pushover app instead to get push notices with an image and then if the image is something we need to watch video, we then VPN back into our system to watch the video.

If you are not overly concerned with security, then using the BI app with port forwarding or Stunnel may be the ticket!

And we find the free UI3 browser feature of BI to be way more user friendly.

But some swear by the BI app and say it is great for their needs, maybe you would fall in this camp!

 

[HomeWPoe]

No we are not saying the same thing.

You keep concluding the Dahua camera with an SD card is up to the task just because the Reolink camera is up to the task, without understanding the underlying conditions that caused it to perform the way it did? Two totally different manufactures and totally different end client expectations. I don't expect a Hyundai to perform as well in speed and handling and acceleration as a Ferrari, although I would expect the Hyundai to do better in snow LOL.

Remember Dahua's market are businesses, not us homeowners. Reolink's market is homeowners. We are just fortunate enough to be able to get our hands on these better Dahua OEM cameras because we favor better quality images.

You must not believe me that everyone says a shortcoming of Dahua cams is how it deals with SD cards. That isn't a big market for them so they haven't focused on optimizing it. Many people buy a reolink without an NVR. so they have optimized their app experience for that user. Most buy Dahua with an NVR or use another VMS system like BI and don't rely on SD cards as the sole recording.

Maybe it is because Reolink is known to play with bitrate and iframes and other parameters that could have compressed the video a lot to not be as demanding as the video from the Dahua cam and that could be the reason for the differences in performance. So it may seem like it can handle it, but it has been compromised quality and you don't realize it. And maybe that compromised quality is fine for you. If it is then run the EmpireTech at lower quality to get that same performance as the Reolink.

Maybe the lackluster performance you are experiencing is because you are using the SD card instead of an NVR...I don't recall @bigredfish complaining about scrubbing video at 8x with his NVR....

For kicks, make the Empiretech H264, 10FPS, and a bit rate of 1,000 and see how the playback performs. I bet it improves....

Regarding the other Backlight conditions, no EmpireTech doesn't call it something else. You have WDR, BLC, HLC, SSA to chose from, along with the changing of shutter speeds and other parameters. These settings are all done within the camera GUI, not DMSS.

But as someone else suggested, show us the images you are having trouble with so we can better assess and offer suggestions.

View attachment 202330


Also keep in mind that DMSS with one camera and no NVR is one thing, add 4 or 5 cameras and see if you still like it.

When I was using it with one camera, it was fine, but trying to manage 4 cameras with it and it got over complicated and time consuming and slow and an NVR was needed.

Maybe BI with the iOS app is what/all you need. At the end of the day your requests are simple - be able to get an alert and play back an alert immediately and scrub at Xtimes without bugging out.

The BI app is fine with that. Most of us don't use it because we isolate our systems from the internet, so the only way to get an image with the BI app is to port forward, use stunnel, or leave a VPN on all the time on our phones.

So most of use the Pushover app instead to get push notices with an image and then if the image is something we need to watch video, we then VPN back into our system to watch the video.

If you are not overly concerned with security, then using the BI app with port forwarding or Stunnel may be the ticket!

And we find the free UI3 browser feature of BI to be way more user friendly.

But some swear by the BI app and say it is great for their needs, maybe you would fall in this camp!

@wittaj, wow! Thank you for taking the time to make this post! It really put our options, moving forward, into much clearer focus. I don't think we'll be in a position to make any good, knowledgeable decisions on any of this until we have a better grasp on security risks, costs, hardware/software compatibility and ease of access/operation, first. In the meantime, we may just go with crap POE cams to give us time to experiment and learn. We'll write off the cost of the crap cams as tuition. Thanks again!

 

[wittaj]

Just try not to make it boxes worth of cheap crap cams

Many of us have been there! If I added up how much I spent on all these cameras sitting in boxes I would cry LOL.

So you have a great Empiretech camera and a cheapo Reolink camera. Those are two great ones to play with.

Try to do as I suggested and get the empiretech camera to be setup similar to what reolink firmware will do and then give it a "fair" comparison.

Then yeah, another topic I touched on was network security.

Right now you probably scanned a QR code in order to be able to get those alerts. That does potentially open up your network, but everyone has a different risk adversity to that. Certainly better than port-forwarding, but the true security conscious here will host their own VPN.

Some have used P2P with the QR code for years without issue, but we also see people come here that exposed their camera to the internet and it was hacked.

 

[Teken]

As it relates to using a Micro SD Card in any device. The first thing to know and understand is they like everything else come in various quality and performance.

Each manufacturer has decided to use similar words / phrases as it relates to higher quality.

You’ll often see the phrase High Endurance, Maximum Endurance, Pro Endurance, Industrial.

If one focuses on just these two things you have already reduced the number of choices / brands / models.

But, have also increased the operational service life of the Micro SD Card when compared to none High Endurance memory.

You can Google the different types of memory and their topology as it relates to 3D NAND, TLC, MLC, etc.

As stated by the other members how a manufacturer handles the use of Micro SD Cards varies a great deal.

The spectrum is from utter shit to fantastic.

If we ignore how the hardware deals with the Micro SD Card for a moment. It’s important to impress upon you and everyone else Micro SD Cards are at the very bottom of the memory longevity scale.

As such, they are destined to fail, not if, but when!

As of this writing Industrial Micro SD Cards offer the longest service life. They also offer the longest warranty because they use all the best in class controllers, memory, error correction, wear levelling, etc.

They also cost 2-3 times as much for a similar capacity!

All of the high endurance / industrial memory do (NOT) offer the same read / write speeds when compared to the so called high performance cards.

The end result as it relates to video is lower performance . . .

Now, if you add in the discussion about how a camera vendor handles Micro SD Cards. You’ll see the performance varies quite a lot just by using the different types.

That is simply relating to viewing said video. Not how it performs with R/W, Formatting, Error correction, Encryption, Locked, etc.

Dahua probably leads the market in accepting the highest capacity memory vs others. This is offset by the fact they have the industry worst reliability handling memory!

Which equates to a Shit rating from me and millions of others having to endure years of the same hardware problem!

Regardless, the use of the Micro SD Card is intended to provide and offer a measure of redundancy, fail over, and resiliency.

It is (NOT) intended to be the only method to record and store video data. There should be at the minimum three levels from onsite to offsite storage whether you use the so called Cloud Power.

Video data should be Micro SD -> NVR / DVR -> NAS -> Offsite (Cloud / NAS / JBOD).

Lastly, it goes without saying everything that relates to your network & WiFi impact directly as to how you perceive any streaming content!

All the best video hardware, memory, will not resolve a weak network performing system. Everything is connected and each play a major role in the so called user experience.

There are literally million dollar systems that can’t stream video security video.

Why?!?

Network hardware restrictions, network security, network bandwidth limits, etc. So as Joe Public you need to step up your game in all directions if you want to find that elusive unicorn

 

[HomeWPoe]

@Teken, thanks for clarifying all the factors that go into choosing and using micro SD cards. I assumed they were more reliable. We used micro SD cards in a bunch of 720p/1080p wifi cams (continuous recording) for many, many years and never had a failure. I'm guessing this was due to the lower data rate, the manufacture's use of good code, and/or luck?

On another topic, the night-time advantages of the 1/1.8 4mp EmpireTech are finally starting to sink-in. Did some testing last night with various amounts of movement---very impressive. The same movement with the Reolinks is a total wash out---literally.

After viewing 8mp daytime vids over the last few weeks, it's hard to get the added clarity out of our heads. What's the consensus on the 1/1.2 8mp EmpireTech IPC-COLOR4K-T and IPC-COLOR4K-X cams? Outside of higher cost, size, weight and rez, is there any other upside/downside with these cams vs. the 1/1.8 4mp EmpireTechs (bullet and turret versions)?

 

[wittaj]

Most of have found the 4K/X and 4K/T to be great overview cams to get color. But are tough for IDENTIFY unless you are in the focus zone of about 15 feet. Anything shorter or longer will be soft/out of focus.

The turret version especially deals with a very narrow window of focus. It really shows that these cameras are not infinity focus.

Plus you need enough light as those cameras cannot see infrared. Or be willing to run the white light all night.

Most of us will go with the 54IR as it can see infrared and works in more situations and doesn't suffer from the shallow focus depth.

Remember we want cameras that perform in your worse condition, which for most will be at night with little light.

Ot do as some do and have the reolink up just for daytime lol. But again we are not entering photography contests lol and the 54IR provides sufficient daytime quality for most of us!

 

[bigredfish]

The 4K models with built in white LEDs can't see and dont have IR
They have some focus issues, ie the focus range tends to be a bit soft close and far due to the big sensor not having the same DOF we are used to with the 4MP 1/1.8 combo
The newest FW is horribly oversharp. Takes some work to tame it

 

[Teken]

@Teken, thanks for clarifying all the factors that go into choosing and using micro SD cards. I assumed they were more reliable. We used micro SD cards in a bunch of 720p/1080p wifi cams (continuous recording) for many, many years and never had a failure. I'm guessing this was due to the lower data rate, the manufacture's use of good code, and/or luck?

On another topic, the night-time advantages of the 1/1.8 4mp EmpireTech are finally starting to sink-in. Did some testing last night with various amounts of movement---very impressive. The same movement with the Reolinks is a total wash out---literally.

After viewing 8mp daytime vids over the last few weeks, it's hard to get the added clarity out of our heads. What's the consensus on the 1/1.2 8mp EmpireTech IPC-COLOR4K-T and IPC-COLOR4K-X cams? Outside of higher cost, size, weight and rez, is there any other upside/downside with these cams vs. the 1/1.8 4mp EmpireTechs (bullet and turret versions)?

To be fair to Dahua the Micro SD Card issue varies with model and firmware installed. The vast majority of people either don’t use a card and simply record to whatever.

To simply defining the hardware to record based on some user defined criteria.

The obvious downside is the memory card won’t have the same video data if and when required!

Each camera performs differently so your personal experience should guide you. The important factors are to understand there are limitations and trade offs in everything.

Based on your replies running a hybrid setup may very well be the break fix solution. Until you feel the need to get more from the system.

As with life it’s always a slow progression to do better . . .

Nobodies made with money so you must balance the cost vs performance ratio that makes sense to you.

Keep playing around to see what the pros and cons are with various settings. Read the material (links) from the other members and keep asking questions.

At some point you’ll decide where that balance is vs this will be (Yet Another) hobby / passion that is truly a time suck / adventure.

Rock On . . .

 

[HomeWPoe]

@wittaj, @bigredfish, thanks! On paper, the 4k/x/t's seem like the cat's meow. The cons, inaccurate focus (and no IR), are disappointing. Nice to discover this now, rather than later. Again, thanks.

@Teken, you nailed my perspective on this journey perfectly. Passion (and money) are key--lol! Both define your roadmap.

Been viewing the 1/1.8 4mp EmpireTech vid alerts from last night and today. Becoming more and more impressed with this cam's night and motion detection capabilities. I've looked at the DORI specs for the 2.8 and 3.6 versions of this cam. Is this also a good indicator of how responsive the cam's firmware motion triggers will be? Not referencing specific distances, just relative motion detection performance between the 2.8 and 3.6. Coverage is nice, but don't want to give up too much motion detection.