Worlds First Review - Dahua - IPC-Color4K-X / DH-IPC-HFW5849T1-ASE-LED - Full Color 4K Camera

[Wildcat_1]

All, quick update. I have both day and night testing completed and am in the process of editing all these together. Currently encoding the day 'Dahua vs Hikvision 4K' clips and will upload them then will move onto the night.

 

[Wildcat_1]

All

As promised here is the first of the comparison videos. This one is focused on the day time testing. We’d all expect 4K performance to be great in full sun and it certainly doesn’t disappoint but there are key differences that I see here and point some out in the video. Unfortunately not being able to use exposure/shutter ranges on the Hikvision does limit you for sure (day & night). Either way I’ve set the cameras identical (based on settings) and that includes using fixed exposure/shutter on each, reducing fps to match the 20 of the Hikvision + adjusting each picture image setting to be identical (by the numbers). For full disclosure, both cams white balances the scene differently but for fairness left these as they were, would normally dial in the image for the scene but wanted to keep settings (again by the numbers) inline with each other.

Some of the image related things I see in the testing (and even more apparent on pro broadcast monitors) are:

• Sharpness on the Hikvision out of the box at default (50) is WAY TOO SHARP
• Regardless of sharpness setting (I adjust in the video, look at timestamps) you don’t actually gain much detail from this increased sharpness. Specifically if you look at facial details, the wall, pergola etc even though you see the added noise that image processing (sharpness in this case) is adding, you’re not benefiting much in terms of 'real detail' from that on the Hikvision
• Even when taking into account focal length, I also noted that again when you zoom into the detail areas that the image is more flat and muted on the Hikvision which speaks to it’s algorithms around Auto IRIS + its AGC which as I mentioned in my previous Dahua videos, if not implemented correctly (by any manufacturer) can lead to these issues. Something I had to work on with Dahua if you'll recall for the latest FW being run on it. This can certainly be improved in FW should Hikvision wish to do so as well and I am always willing to provide the details on what should be tweaked in the codebase itself
• Most concerning for me (when dialing in an image and looking at overall quality) is that the Hikvision has a ‘pulsing’ noise pattern to it which as I mentioned in posts over the years is unfortunately the worst type of noise and hardest to mitigate against. You’ll notice that its evident on a 1 sec interval (check the grass etc) and appears both day and night regardless of setting. Certain Dahua cams have had this over the years so they are by no means exempt but recently and on the 4K here plus other pro models you notice this for the most part has been fixed and replaced with what I refer to as the ‘crawl’. This is much easier to mitigate through dialing in and other steps as I’ve shown before. You’ll notice this same ‘crawl’ here (look at the pergola supports) but again its not as distracting or impactful as the pulsing IMO
• The Hikvision also lost target more (even though I was still moving around) and therefore separates video files more easily than the Dahua regardless of sensitivity. I’ve noticed this on the day and night testing so wanted to call it out. Certainly not a deal breaker but something to be aware of

Anyways, just a few things I noticed so far. Timestamps as always are in the YouTube link. Be sure to go to YouTube to watch this and select full 4K. Lastly, unfortunately the cams weren’t clock synced so I manually frame synced the side by sides for you

I’ll post the night video’s as soon as I’ve finished compiling them

As always, let me know with any questions


VIDEO LINK



Also, here are some zoomed images to show some of the detail difference at the sharpness levels:


Dahua 4K @ default 50 sharpness




Hikvision 4K @ default 50 sharpness



Hikvision 4K @ default 30 sharpness

 

[ljw2k]

Thanks for the Daytime clips WC and my thoughts are the Hikvision wins hands down no doubt as to me more detail and dynamic range with much better colour than the saturated Dahua. One thing I did notice is the Dahua seams to be sharper close range looking at the wooden pogonia and the string lights but further away it looks like mush where the Hikvision still looks sharp looking at the pale green tree and the pattern on the houses across your street but close up the Hikvision looks OOF looking at the wooden Pogonia and string lights.

I also wanted to comment on how the test was set up with the same settings in each camera and would it not be better to have the settings looking the best they can on each camera rather than the same on each.
I am looking forward to the night time clips as I think this is what matters to most people.

Dahua lens is focussed for close range
Hikvision Lens is focused for long range

Excellent work once again very detailed and thanks for sharing.

 

[EMPIRETECANDY]

Thanks for wildcat's testing. From my young eyes, the wall from dahua is more to the real color of the wall. dh Pic is a little better than hik
Look forward to watching the nighttime comparing.

 

[sebastiantombs]

Maybe it's me and my tired eyes, but I still see "noise", or jaggedness, appearing in the Hik video when he's at the farthest point away and it shows even with slight motion and that's after the final tuning. Prior to that it was annoying to me. Overall, I think the Dahua is "smoother" video. In terms of color registration the Dahua is more vivid but I have to defer to WC and his assessment of which one is more "true to life".

 

[ljw2k]

Thanks for wildcat's testing. From my young eyes, the wall from dahua is more to the real color of the wall. dh Pic is a little better than hik

I looked back at some of WC reviews of past Dahua camera's and that patio area and wall is not the right White Balance and this is comparing it to the 12mp Model 71242 but the Hikvision is very close in terms of WB not looking at pixels or sharpness here.

I think the Hikvision focus is further out where the Dahua is closer which is where it should be. I know a few members here have re focused their Hikvisions with excellent results compared to out the box performance.

"But" WC did mention that the White Balance was different but the Hikvisions shows the most accurate compared to other Reviews and clips of the same scene.

 

[Parley]

Thanks for the Daytime clips WC and my thoughts are the Hikvision wins hands down no doubt as to me more detail and dynamic range with much better colour than the saturated Dahua. One thing I did notice is the Dahua seams to be sharper close range looking at the wooden pogonia and the string lights but further away it looks like mush where the Hikvision still looks sharp looking at the pale green tree and the pattern on the houses across your street but close up the Hikvision looks OOF looking at the wooden Pogonia and string lights.

I also wanted to comment on how the test was set up with the same settings in each camera and would it not be better to have the settings looking the best they can on each camera rather than the same on each.
I am looking forward to the night time clips as I think this is what matters to most people.

Dahua lens is focussed for close range
Hikvision Lens is focused for long range

Excellent work once again very detailed and thanks for sharing.

View attachment 94118View attachment 94119

I agree with you. The Dahua is sharper closer in. Just look at the lightbulbs. However the Hikvision is sharper further away. Look at the wood chips in the planter and the trees in the background. Thank you Wildcat-1 for the excellent review. Looking forward to the night comparison.


Edit. I have bought 4 of the Hikvisions. Two of them were out of focus from the start. I had to take them apart and focus them myself.

 

[Crows]

Alot of natural light difference though in the first pic ( 71242 cam ) than the other 2 . Must of been overcast ? No shadows can be seen anywhere even by foreground trees , very bright in the other 2 pictures throwing lots of shadows.

Thanks for the excellent review WC , looking forward for this one ....love my 5442 LED .

 

[Wildcat_1]

Thanks all for the kind words, I do appreciate it.

Some additional commentary on my findings so far. As I mentioned above, both cams on default WB are not spot on. The brick has a hue of yellow (makeup of the brick) in reality and definitely glows more towards that hue in the sun therefore the Dahua is closer but at default still a tad too warm. On the other hand when you compare the Hikvision, it is too cool. The other thing is that this was a very hot, very sunny day (I know you’re amazed I stayed cool in the hoodie and hat AND mask ) so there was an even stronger golden hue from the sun that the Hikvision did not adjust to. The 71242 image (as @Crows pointed out) was on a more overcast afternoon and later in the day for reference so expected to look a little cooler. In a pro white balance setup you do want white properly represented BUT taking into account the correct balance of other colors in the scene. Take a look at the Youtube thumbnail as a quick reference and you can see the white on the hoodie is represented on the Dahua better than the Hikvision on the right BUT again both don’t balance correctly for the entire scene on defaults as I mentioned.

With that said, the color balance doesn’t bother me as much because that can be corrected on both cams AND if you want a truly perfect scene then you can custom balance as we would with pro broadcast cams. Simple to do but very effective.

The biggest concerns I have with any camera I test are image quality, noise, adjustment options (help mitigate other issues we’ve spoken about) and any inherent defects. Noise can be mitigated if you dial in correctly (as I’ve shown before) and it’s the ‘right’ noise. As I mentioned above, the pulsing noise seen on the Hikvision is the worst kind unfortunately. Older Dahuas had this (early 4Ks exhibited it and the 5MPs) and it’s not good because it’s distracting and deteriorates the image more than a crawling noise. This can’t be fixed in cam as generally it’s down to the sensor itself and in the case of pulsing noise can also be exacerbated further by electronic interference produced by surrounding components. Therefore I’m some cases a redesign is needed in other cases algorithms can be tweaked to reduce it BUT the pulsing will still be there with algorithm tweaks. While you expect noise on any camera it should generally be more prominent in dark scenes and at night. The fact that this is so evident during the day on the Hikvision is a concern for me personally.

The AGC and overall Auto Iris algorithm on the Hikvision definitely needs work. Lets not forget that Dahua’s needed that too but when I reached out to their dev team they were able to take my feedback, the code changes I recommended and put those into the latest FW. The other changes I recommended I’m told will be in the next FW release and I’ll test it as soon as I get it. I’ll reach out to Hikvision and see if they are interested in working with me on fixing those with my input.

Coming back to some of the questions / comments. Agree with LJW2K that you can always pull a cam apart or re-calibrate to refocus BUT let’s be honest the point is not for an end user to do that so whether it’s Hikvision as in this case or Dahua or Arecont, Axis etc that’s not a mark of quality leaving the factory without focus on point (if it truly is just calibration related) and generally you would send back. I do hope that Hikvision aren’t increasing the sharpness to the over sharpened extent (out of the box) to make up for softness of the lens or other issues. May not be the case at all but its a concern if others have seen this issue too and something that should be flagged for the manufacturer to address. Maybe a bad batch etc but something that should be looked at IMO by an cam manufacturer. @Parley doesnt have time to continue to refocus cameras, the mans busy and his time is worth a lot of $$$$$$$$ .Either way, in comparison to the 2 cams here, on the focus front the Dahua does give you details in the sweet spots you need it as even from a DORI perspective Recognize (not Identify) is around 79’ and just before the fence line as I show is 62’. So its not perfect but its close. Of course when we get a Dahua or Hikvision in vari focal that will be ideal as you will generally see sharpness through the focal range but hey that’s for another year and with that comes chromatic aberration

As I said though, even taking into account focal length and any factory calibration, in my testing this current Hik is still a more muted and flat image (regardless of saturation and color balance) with more details lost overall which is not helped by the noise. The Hikvision in my opinion is still a good camera and can be a phenomenal cam BUT does need some work. Let’s see if they’re willing to work with me.

On the testing, absolutely could dial in both cams to be their ultimate best BUT that was not the point of this testing. The point was to make it as fair as possible which is why I even did the extra test of reducing some of the harsh sharpness out of the box on Hikvision while I left the Dahua at default. Remember the Dahua has the ability of exposure ranges, gamma control, higher fps, wider WDR etc but using those would not have made this a like for like comparison. Hope that helps explain the approach to the testing setup on this.

Either way it’s good to see these 4K cams starting to mature and turn into serious contenders, hope the video's so far have helped.

Now, lets move onto the night stuff next and see what that brings too

 

[Crows]

as a quick reference and you can see the white on the hoodie is represented on the Dahua better than the Hikvision on the right BUT again both don’t balance correctly for the entire scene on defaults as I mentioned.

WC...You need 1 of these to do your testing in ....the good old TV test pattern. ( Oops now I'm giving away my age ).

 

[CCTVCam]

Great video. I find the Hik the winner.

I don't have time to critique properly atm but the differences I'm seeing are:

1. Different saturation and color temp - Dahua wamer, Hik cold

2. Different depth of field explained by a later finding

3. I'm not seeing the Hik as too sharp. That's how I like a picture - able to see every blade of grass and that isn't down to sharpening in my opinion but probably compression levels. Personally, I don't see an excess of sharpening - this would manifest itself in artacts and at a quick glance I'm not seeing them

4. More detail in the Hik especially at range

5 Slightly better clarity to the overall image although no detail lost as a result in favour of Dahua. Hik has a slight mist effect. Maybe the lens quality or maybe the settings as the Hik is flatter

 

[Wildcat_1]

All

Just so everyone knows what to expect from the night testing. I once again will break it down into 2 distinct lighting scenarios:

1 - Illuminators Only (but with a twist, more explanation below)
2 - With supplemental lighting - Again use the 100w bulb + 5% dim approach

Again to keep it fair I adjust the settings to match each other (Dahua and Hik) and utilize fixed exposures since the Hik doesnt support ranges. Therefore I test against 1/60, 1/120 and I throw just a couple of 1/250's in to push them both.

Now when I say Illuminators but with a twist, what I tested against is each camera against its own and then the competitors illuminators. i.e. Dahua & Hik with Hik's illuminators on only and then reverse it (Dahua & Hik with Dahua's illuminators on). I do this for each test so you will see that too. That way this not only shows what each cam can do with its own on-board illuminator but how it does with the competitors, thought it would be interesting especially.

I'll edit these together as soon as I can and again build out the side by sides for easy comparison after the full screen caps. I'm just as interested to see how these both do at night as you are as until I pull these into the editor and put on broadcast monitors you really don't appreciate any differences. That's the problem with YouTube is that no matter what pro codec you keep the master's in and then compress at high quality, YouTube comes in with its crazy compression and muddies the water

Either way, this will be coming as soon as I can finish it. Stay tuned !

 

[Kevin Doe]

I really appreciate your time and detailed comparisons between these two cameras. I'm with a few other though. I don't understand why it is fair to match settings. Most advanced end users will dial in either camera they use, so I think it would be more of a fair comparison to dial both cameras in to their ideal settings and let each show the best of what they can do.

 

[ljw2k]

I'm hoping the Dahua does better at night but looking at post#2 we already have the best possible night time clip the Dahua can achieve.

For me also like CCTVcam from the clips viewed the Hikvision is the best daytime camera out the 2 and can be much better with a simple re focus. I have this camera in the 2.8mm and the focus is spot on throughout the frame and again can we see QC issues from hikvision as Parley had to re focussed his Hikvision and the rewards are very much clear to see.

Another factor is the price will it be around the same price as the Hikvision?

 

[Maconi]

I agree that it seems like the biggest difference is the focal range. Which is odd since they're both the same size sensor (1/1.2") and both aperture f1.0. Meaning the only difference should be the lens size (4mm vs 3.6mm). You wouldn't think 0.4mm would change the focal distance THAT much. The Hikvision seems to be better focused IMO (if you want crystal clear focus to infinity like the spec sheets claim then the focus has to be a bit further away from the camera with such a low aperture). The Dahua's focus needs to be moved out a bit to achieve clearer infinity focus. Having said that the Hikvision does seem to be slightly out-of-focus itself. Fixing that would require opening it up, removing the glue from the ring, and adjusting it a bit though.

 

[Wildcat_1]

Thats the beauty with testing these the way I do, get to pull them apart (figuratively and physically) and find the pro's and cons In this case as I said above, both cams need work. Dahua needed it as well (as I mentioned up front) out of the box for FW, luckily we've already seen great improvements there through the first iteration of the FW with the fixes to bugs I raised, feedback provided etc. My hope is we see this next FW release incorporate more of those that I've provided through testing. With the Hik as I said, this is a good cam too BUT in my testing to be phenomenal it too needs to have work done. It is not great out of the box (this 4mm version) and does suffer lens issues that others have discussed too and need to be fixed. This is not something that should be put on an end user to do. The noise that I show is also an issue, pulsing noise and the issues it brings is not good and needs to be corrected. The addition of further illuminators on the Hikvision (add 2 more for near/far control), exposure/shutter ranges, more granular control over things like gamma and as LJW2K mentioned before, an overhaul of the menu. If we're all honest these need to be done and would only benefit all.

Therefore if we all look at this from an unbiased perspective (which is the purpose of any of these reviews) both cameras need work. The key is which manufacturer(s) a) listen and b) most importantly then go fix, because the ones that do enable all of us to get better products in our hands while the companies benefit from increased sales. Those manufacturers that don't, in turn don't deserve the sales because we should not put up with subpar products from factory especially in the critical area of security when they don't meet expectations OR stated performance claims (we all know this last one on 'claims' has to be interprested somewhat loosely but you get my point). This is incredibly important whether putting these in residential, commercial, dark site or global installations where it's not only key to get the right product (working as it should) but the support of fixes in a timely manner as each product will have issues at some point.

Lets not forget, mixed systems are ok too That's why my hope is that in me putting the time, effort and energy into these reviews, surfacing issues in cams / other product ranges across manufacturers, then most importantly providing critical, data backed feedback, that we all benefit from improved FW, redesigned HW (where needed) and ultimately a better end product.

WC

 

[CCTVCam]

Great video. I find the Hik the winner.

I don't have time to critique properly atm but the differences I'm seeing are:

1. Different saturation and color temp - Dahua wamer, Hik cold

2. Different depth of field explained by a later finding

3. I'm not seeing the Hik as too sharp. That's how I like a picture - able to see every blade of grass and that isn't down to sharpening in my opinion but probably compression levels. Personally, I don't see an excess of sharpening - this would manifest itself in artacts and at a quick glance I'm not seeing them

4. More detail in the Hik especially at range

5 Slightly better clarity to the overall image although no detail lost as a result in favour of Dahua. Hik has a slight mist effect. Maybe the lens quality or maybe the settings as the Hik is flatter

Ok bit more of a critique of the Daytime performance, may not be as fluid or detailed as I would have liked as 12 hours have passed since I wrote the above / saw the video and so intial thoughts lost:

1. The Dahua looks more dynamic but this is almost certainly due to greater saturation and higher contrast to the picture. A look at the pictures side by side show the Dahua to have deeper reds and greens and more punchy colours and darker shadows. All signs of increased saturation and contrast. However, there is nothing here that cannot be adjusted in the settings The Dahua could be de-saturated or the the Hik saturated to be equally as punchy on colour and saturation. The warm Dahua picture and Cold Hik are also just preference that can be adjusted with sliders. So nothing to choose between them here as I see it. Dahua chose to go warm and punchy. Hik cold and lower colour / contrast. Personal preferences. Want more colour, adjust the saturation and contrast. Want warmer or cooler, adjust the colout temp. Both cameras are probably going to be indistinguishable for colour after tweeking.

2. Depth of field, Wildcat's observation (valid) that the Dahau is sharper close up. One of his valid critisisms is that the Dahua captures a sharp image of him close to the camers whereas the Hik does not.

However, I still believe from my obeservations that the Hik is sharper and more detailed and the better camera potentially for daytime therefore.

The striking difference in close upsharpness can be explained by looking at the background - these cameras have similar focal lengths so should have a similar DOF with similar apertures. Look at the background and the Dahua is blurred, even the wall is starting to blur when Wild is close to the camera. Conversely, looking at the Hik, the background is sharp not only at the wall but across the grass and back to the trees. Yet Wild is slightly off focus close to the camera:



My interpretation here is the Dahua and Hik are focused at different distances. The Dahua close to the camera and the Hik a little further back. This explains the dof sharpness differences and also means the Hik could be adjusted by Hik or after market by the user to provide a sharper picture close to the camera as per the Dahua at the expense of DOF should the user require it.

I believe my suspicisions may be confirmed by my 2nd screen capture, where Wild being a little further away from both cameras, suddenly appears sharp, arguably more so than on the Dahua:



As for the Hik being too sharp, I'm not seeing it. I am not observing obvious aretfacts. I can see more detail in the Hik picture and like seeing every blade of grass and the leaves. However, I doubt the latter is due to sharpening alone but also maybe compression. It would be interesting to see the out of camera rate (as opposed to the BI / DVR rate) to see if there's a difference in file size.

One final point, I do not agree the Dahua picture is more detailed, look at the paving and some of the areas highlighted below. I believe the Hik is retreiving more detail (sat and contrast could have effects here two and could also explain a slightly misty look to the Hik despite the detail).

For me the Hik is probably the better camera still for day use. It retieves more detail and is sharper overall. It's downside is sharpness with a subject close to the camera, but this could conceivably be adjusted away by re-focussing the lens closer at the expense of some DOF. How much of the detail matters to CCTV over hollywood is another matter for debate but I see little to choose between them with Hik having the edge from my obversations although this is always subjective and as we have seen Wild prefers the Dahua.

Some detail areas for me where the Hik wins are the Trees, Grass, Wall, Patio slabs, furniture rug, wire mesh around the planters, Wild's shoes in the 2nd picture. The Dahua wins on the boarding of the house although here 2 points: 1. The Dahua is focused at this distance, I believe the HIk is not. 2. The greater contrast level on the Dahua here, probably helps pick ou the grain. Looking at the pic closely, I believe the wood detail is probably still there in the Hik but would benefit from greater contrast.

Only my observations and opinions and not intended to take away from what is a great review by Wild. Many thanks to him for providing the material for debate.

 

[Wildcat_1]

Spacer For Night Video

 

[Wildcat_1]

Hey cctvcam you know you're always welcome on my posts brother ! Appreciate your kind words.

As I mentioned above, both cams need work for sure, the key is the manufacturers stepping up to fix the issues. While great that the Hik focuses to infinity as we've discussed here and you and I agree, does not mean that it does not need work in close focal distance which is clearly off on this Hik model (maybe bad batch or indicative of the 4mm, either way needs fixing). That's not something that any end user/purchaser should be doing hence raising it here (and I intend to do the same with Hik) for awareness and ultimate fix. On the Dahua front there needs to be work on the far end of the focal length. There absolutely would be a trade off on both cams for DOF but that goes with the territory on any lens build and there is absolutely room for a middle ground there in my opinion where both cams could benefit from improvement. Will be interesting to see their varifocal counterparts when released albeit have to watch for chromatic aberation.

On the color balance side as I stated (and you and I agree) this is not a big concern at all just a difference thats worth pointing out. Its the same as pro broadcast cameras I own, Sony, Canon, Panasonic they all shift one way or the other but as I mentioned above, its easy to correct and worst case you custom balance it. Reason for not doing it here is for like for like out the box test.

The pulsing noise that is present on the Hik is an issue that needs to be corrected and to be fair, shouldn't be there in today's current crop of cams from ANY manufacturer. Its easy to fix with design and build around the SOC, hard to mitigate when its present (not been fixed up front) and should be taken care of regardless of manufacturer.

There is definitely over sharpening and potentially compression issues with the Hik out of the box before adjustments (potentially a combination as you mention). Even more visible when you look at a broadcast monitor. Now as we've debated, we may beg to differ whether one or the other (or both) makes more impact but either way an 'over processing' (maybe a more general way of putting it) is present out of the box. This was similar to the Dahua 5442 vari's when first released. You mention file size which funnily enough was one of the first things I looked at as I was interested too I was surprised that straight out of the camera the sizes were very similar. The only difference being that the Hik 'chopped' files more frequently for target acquisition than Dahua but when you calculated the total for a cap they were extremely close. Therefore there could be tweaks under the hood to the algorithm (in 1 vs other) but end files size is close.

Either way, as I said above, the key is getting manufacturers to take the feedback, fix and or make ultimately better products. Then we all win regardless of product or manufacturer I'll keep everyone updated on what/if I hear back from Hik.

Thanks again for the kind words, while you guys keep loving the reviews and finding them useful, I'll keep doing them

If I have time after the out of the box tests I may also do a full dial in of both cams if people are interested. I may even separate out a thread for that one to make it cleaner.

 

[Wildcat_1]

Just to set a quick expectation. I’m going to take a day or 2 off to enjoy the holiday as well so will continue editing the night video together and look to post early next week. Stay tuned and for those celebrating the 4th July have a happy and safe one with your family, friends and loved ones !