Noise Battle of the EM1X vs Em1 Mk3
ฝัง
- เผยแพร่เมื่อ 5 ส.ค. 2024
- This video is a battle between the Em1x and Em1 Mk3. These two cameras are at the top of the Olympus product portfolio. Many often call the Em1 Mk3 a miniaturized version of the Em1x, however my own experience with the camera has demonstrated that it is something more.
Now with one has more noise? Let's see the tests.
Bellow are results pulled from ASTAP. If you want to try this software for yourself it is free. Here is a link to get you started. www.hnsky.org/astap.htm
EM1X 240s
x̄ : 238.13 (sigma-clip iterations=5)
x̃ : 239
Mo : 236
σ : 12.66 (sigma-clip iterations=5)
σ_2: 10.97
mad: 9
m : 132
M : 4095 (169 x)
≥64E3 : 0
Vignetting [Mo corners/Mo]: -6%
Bit depth data: 12
EM1 Mk3 240s
x̄ : 220.34 (sigma-clip iterations=5)
x̃ : 222
Mo : 220
σ : 12.96 (sigma-clip iterations=5)
σ_2: 11.91
mad: 10
m : 125
M : 4095 (210 x)
≥64E3 : 0
Vignetting [Mo corners/Mo]: -4%
Bit depth data: 12
Legend:
x̄ = mean background | x̃ = median background | Mo = mode or most common pixel value or peak histogram, so the best estimate for the background mean value | σ = standard deviation background using mean and sigma clipping| σ_2 = standard deviation background using values below Mo only | mad = median absolute deviation | m = minimum value image | M = maximum value image | ≥64E3 = number of values equal or above 64000
Music: Bensound.com - วิทยาศาสตร์และเทคโนโลยี
Dude this is an amazing test! Thanks so much for taking the time to do this!
Glad you liked it! It was fun to make.
I am thankful for this. I hope I can learn more from you.
Extremely interesting thank you
Excellent video
I'm just impressed that you have the E-M1 Mkll, Mklll, & X -- That's a collection of excellent gear! :-)
I also have two Em1 Mk1s. I will probably show the entire family when I do my Em1X review.
@@TheNarrowbandChannel That's funny, I actually have two gripped E-M1 Mkl's and only lack the E-M1 Mklll of your collection, so I'm right behind you. But then, I do still have a functioning Olympus E-300 with battery grip too! :-)
i just got the M1X less than half year, i liked this camera, i hope you doing more videos about M1X and astrophotography looks very interesting, i do wildlife but i will give it a try for sure, subscribed
Thank you. I do intend to be using my Em1x a lot. Especially once they do the 3.0 firmware update. It I supposed to add Starry Sky AF.
Thanks Benjamin, another great video, need an addition to my Mkii so this might swing me to the X . Cheers
Plus the X prices has some down a bit. That is why I picked one up. But now that I have one I wish I had grabbed one two years ago.
@@TheNarrowbandChannel Came here from the Alpaco Film's channel. Really interesting comparison. I come from a wildlife shooting background but all I can say is 'what took you so long' 😅. I have had mine since just after they were released and I love it to bits.
Apart from the ubiquitous shots of the moon I've not done any astro but I'm certainly thinking about it. I had great views of the Saturn/Jupiter conjunction last year through my field scope. Would be great to try and capture those as a start. I'm luck enough to have the 150-400mm so it could be fun to see what I get with that once the air cools a bit (tried it on the moon back in the summer when I got it but the heat distortion was horrid, assuming that's what it was).
Anyway enough random waffle. The important thing is I've subbed.
@@Raysnature Thanks Ray. You will have to check out my moon videos. Someday I will get my hands on the 150-400. Olympus has said that they would let me barrow one. I just have to get time to make sure I can do a proper review of it. I think this winter I will do it.
Ya I love the X. And I must say that once I got it I tole myself that I should have bought it the day it came out two years ago. Would have been worth paying the 3000$ it cost back then too. Ill do a review on it soon too. I have lots of thoughts about the X.
@@TheNarrowbandChannel That's great, if you get the chance do, it's a fantastic bit of kit.
Well done!
Thank you!
Enjoyed it thank you
Benjamin, I was wondering if you have had a chance to try and see if adding time between the exposures would give the Mark III a way to dissipate heat enough over time to keep up better with the M1X? I haven't done the tests you did, but when I have done long exposure sessions of many frames, it does seem that at around 4 seconds between 60 second exposures, the temperature might stay more constantly lower than with 1 seconds between. I think it might make a bigger difference with the really long 4 minute exposures, but I might have to test this myself. Thanks, it's very interesting what you figured out.
The Em1x will cool of faster. In some of the other stuff that I have done I can see that in temperature swings that it has. However for astrophotography I just do a 1-5sec gap. It is worth it to just capture more data rather than let the camera cool down. Its sort of a lesser of two evils choice sort of thing.
I have an Olympus EM1 Mk3 and only use a Mac. I would like to get into astrophotography using my Olympus. What stacker do you recommend? I noticed you mentioned Deep Sky Stacker. I was interested in Starry Stacker but it does not support RAW files and Mac does not support the Mk3 raw files. LR and Bridge will convert to TIFF but according to Starry but they also apply lens corrections automatically and that apparently introduces artifacts that mess up Starry. So what is the best approach for my situation? Thanks.
You are the man. Mac user here too. So I used to use DSS a hacked version that ran on Mac. Still have it but I have moved on to ASTAP. It is WAY better. I have a tutorial on how to use it on my channel. Did only a week or two ago. Check it out.
I also have a tutorial on the DSS hack version and I think I did that last year.
Nice video mate. I have little idea of what you’re talking about but your video is immensely interesting. I shoot an E-M1x + 40-150 pro and love it. Cheers from Down Under.
PS. Your pick for best all round Olympus lens for astrophotography?
My pick of the best lens would be the 75mm f1.8 for all around cost, size and deep sky capturing power.
@@TheNarrowbandChannel thx mate
On the cooler idea, can you even turn the lcd off if it's flipped out though? I got the em5 II recently and it's terrible in not allowing you to turn it off, with it flipped it's either it or the evf always active
You might want to go to Rob Treks channel for that. He might have a solution. He did a video on the EVF and LCD screen recently.
I will say though that as far as I can remember there is no way to turn both off. Even when you use Olympus Capture. Bit off a shame really. Would be nice to turn both off to save battery.
It might not be a handicap; I will try and find it, but Rob Trek may confirm, the EVF is reported to use battery power quite a bit faster than the LCD; and especially for some of the settings we use in astro, the LCD is -7 in brightness, it uses less energy. The EVF apparently is reading more pixels from the sensor and requires more power than the LCD.
Would really appreciate your comparison of the E-M1x (which I bought partly because of your review) and the OM-1.
Ill probably do it after the OM1 full review.
That will be this fall.
Great video buddy. That was awesome
Glad you liked it. I am sure someday you will want an Em1x for fox and other wildlife. Especially when they ad animal eye focus.
Good experiment.
And something I've looked at on MFT as well with my radiometrologist hat on.
Now something buried in the small print of the E-M5ii manual, and only stated the once, is that picture quality may degrade when shooting long video or long exposures. And it does, the picture gets a magenta tinge. Now when taking pictures of winter pansies (violets), part of my series of experiments, which unsurprisingly have violet tramlines to guide insects, Olympus cameras (every one I own) cannot see violet and Panasonic can, even though they both have the same Panasonic designed built by Sony sensor. The difference lies in the camera 'engine', and the greater emphasis Panasonic set on shooting movies. Panasonic do not tinge magenta even after prolonged shooting. I surmise there is something in the Olympus TruePic engine to kill the magenta tinge form sensor heating. The entire frame of the GH4 for example is built as a heat sink for the sensor. The relatively cheap-used Lumix G7 also shoots prolonged video well without overheating; &h considered one of if not the best vlog camera. The Lumix G9 even though it has body stabilisation still has good cooling. The E-M1X has obviously been given some form of heat sink but it eats two of the big Panasonic batteries (yes, they're made by Panasonic), so it obviously draws a lot of current &h heat. Like Olympus, all Panasonic also interval to 60 seconds and will AF on stars (certainly Jupiter) whether they are billed to have 'Starlight AF' or not (G7, GH4, G9). Otoh the E-M5ii is small and light like the G7 but has better haptics and more functions - it is better for hanging on an OTA
I have been posting updates on the progress on the OMD Astrophotoraphy group on FB. Another member did the first build using mail ordered parts.
00000
I would be interested in the Peltier cooler.... sell for a good price for you... is not easy to solve many problems... I did a mod that almost frost the camera... the main problem was the vibrations from the fans.... the best are the Noctuas... and you can add some shock absorbers too.... I also had to rise the rpm of the fans... so be sure to use a fan with roller ball bearings...
Thank you. What size fan did. you use?
@@TheNarrowbandChannel You have to choose one that almost covers the all the heatsink area.... you can put one that push the air and other that sucks the air in either side of the heatsink.... and that way you may lower the rpm to avoid vibrations.
Why not just use a cooling spray (ice)?
Interesting idea. The cameras are weather sealed but my telescopes are not.
Subed!
For astro, yes is the best.
But in high iso, em1 III keep colors better.
I have an idea for you to cooling, use a small fountain if the lens is waterproof lol
They are splash proof!
On a side note. At the observatory we have an older SBIG camera that was nitrogen cooled. It is quite old though. Many older cameras were in fact water cooled. You had a tube going down to a bucket of water.
Interesting video even if I don't do astro photography, a little too complicated for my ability to undertake such a complicated path.
thanks for the explanations
Thanks
Would be interesting to put a TEC (thermoelectric cooler) on the back where the LCD closes. Of course the TEC requires it's own heatsink and/or heatpipes plus the power supply for it so not realistic though in astrophotography could be useful. Large telescopes such as in observatories use CCD sensors and often liquid nitrogen. lol
Very systematic and well done presentation.
We put Peltier pumps on germanium detectors to see very very small signals but that seems a little OTT for a MFT camera which have other means of reducing noise. Astro cameras do not have such fancy firmware and construction, so they sometimes need fans. Fans often introduce their own noise, there is heat but also EMF spitting out all over the place from the motor, magnetic fields whizzing round.
Astronomy is often a winter sport and it can get so cold the lithium batteries pack up.
Thank you very much... 👍👌👍🙏🙏👍 Amazing and Fantastic Test... 🙌🙌👌👌 Then I'm happy to acquire an Olympus E-M1X ... , I see that it is cooler and produces less noise than the E-M1 Mark III ... , in addition I also want to photograph motor sports and animals... and record some video... 👍👌👌👍
Health and Greetings from Spain... "SALUDOS... 👋🏻👋🏻🙋🏻♂️😃 "
I share you enthusiasm, I only picked up an Em1x a few weeks ago and it is my favorite camera ever. The viewfinder especially is wonderful.
@@TheNarrowbandChannel Terrific... 👍👍🏻🙌🙌🙆🏻♀️🙆🏻♀️
GPS is reason I picked E-M1X
Something i was surprised to see absent in the OM1
Was that a hot water storage cylinder in the shed with them?🤣
Air compressor. Or my old Air compressor. I should get rid of it.
I just watched this video, sorry for being late to the party.
I'm not sure about the conclusion.
x-bar is the mean background pixel value, and x-tilde is the median background.
For a dark frame, a zero value would be perfect. Higher values are from thermal and read-out noise. The E-M1mk3 has a lower value of x-bar, x-tile, Mo, so if you don't do dark-frame subtraction, the baseline (the blacks) are lower in the mk3.
The standard deviation (sigma) in that background noise is SLIGHTLY (0.3 pixel count) higher in the mk3 versus the 1X. But that is pretty trivial.
I guess you are calling that "noise" because that slightly higher variation in the background will be left in the image during dark-frame subtraction. But again the mk3 has overall less background signal overall.
I think this is pretty close to a tie.
I thought that too but when you stack about 30 dark frames to make an average and then subtract that the sigma becomes the most important factor. I know when I was first doing this test I thought the Mk3s darker background was more important but several of the more experience astronomers that I talked while making this video, all said that the sigma value was what defines the victor. And they all gave the X the nod.
I suspect that the darker background could be changes that Olympus made to the raw codec. Camera companies raw files are not truly raw and they do have some noise reduct made in them no matter how many things you turn off in the camera. Makes me wish I could get them to add a FIT file option that would be truly untouched.
Great comment! and insightful thoughts. Thanks.
@@TheNarrowbandChannel Thanks much for the follow up. That does make sense.
I used to have a 1970's photo article about 'hypering' film to do astrophotography. You dip the film in liquid nitrogen which enhanced sensitivity way above the rated ASA.
So . . . maybe just dip your OMD-EM cameras in liquid nitrogen . . . ;)
Lol that might have some drastic results. Yes I remember those days. I think NASA does use Nitrogen to cool some of their cameras to -100C on land based observatories.
I was having anxiety with the cameras just sitting there exposed 😂
Olympus dust removal system is pretty awesome.
So I have never feared having the sensor exposed.
@@TheNarrowbandChannel - it's not flawless, however; some dust particles ( although very small ) do stick and you won't see dust spots on test shots because of the thick ( 4mm ) glass infront of the actual sensor which has the IR and UV filters , which smoothes-out but spreads-out, the effect of a dust spot into a regionalised region of lower-contrast ( not detectable with traditional dust spot algorithms ). if you use a loupe which can focus 19.5 mm down i.e. onto the surface of the 4mm glass filter in front of the actual sensor then you may very probably see dust spots or even hair which the sswf anti-dust movement hasn't been able to shake off, either due to electrostatic charges on the dust particle or due to the frequency of the sswf being too dissimilar to the larger piece of hair etc. whose resonant frequency is orders of magnitude smaller than the 30khz vibration frequency of the sswf. just my thoughts.
In electronics we normally say, noise doubles each 10°C, because the leakage current of semiconductors and the dark current of photodiodes doubles each 10°.
In radiometrology it is signal to noise. Where we might only be collecting a few photons their signal has to be statistically above the noise. Many methods used to keep electronic noise down; cryogenic cooling is but one, gating the noise out is but another. There's not that much difference between a gamma detector and a digital camera really, both detect photons. Our MFT LMOS is not CMOS, it is biassed and sweeps stray electrons/dark current off. Being a smaller sensor also makes it electronically quieter. Olympus OMD are good cameras, I use two of them, but I have some reservations that Lumix may be lower noise at long exposures because they stay cooler (the GH4 barely gets warm, the E-M5ii can fry eggs), I have five of those, three of which are worth comparing.
Now all I need are some dark skies and something worth taking a photo of ;)
Peter great comment. So with noise in an image and how it affects it, it is a function on time and temperature. Obviously a short 1/1000s exposure would yield very little thermal noise as it would have little to no time to accumulate. However when doing longer exposures such as is done in astrophotography, the noise builds up from the thermal signal.
As a rule of thumb for exposers over a minute once you hit 20C thermal signal doubles about every degree of change. And the it continues to increase as you go up in temperature. It's a function of a logarithmic curve if you will.
@@TheNarrowbandChannel I am not a scientist, but I understand it this way: In an ideal photodiode one electron corresponds to one photon. So one false leakage electron corresponds to one false (noise) photon. Therefore electronic noise should be proportional to leakage current. Of course there are other noise sources as infrared photons, I do not know much about the latter.
@@hpeterh I think Infrared and heat react similarly. But I would have to give that some thought too.
INFORMATIVE- BUT, with Topaz Denoise and DXO raw, they are two excellent programs to eliminate noise, so while a good idea, I feel software is the way to go.
Topaz is pretty powerful. Just remember though if you can reduce noise before you even do Topaz Denoise then you can upshot that much farther.
great video, brilliant test...thank you...
Thank you!