Cliff Totten

IHoly crap...this could be a true "Batman & Robin" dynamic duo combination. If both sides are very serious about this then they could actually be a serious threat to Panasonic, Canon and Sony. The only BIG question is....who is going to fab their sensors? Im betting Sony would tell these guys "no". Actually,...does anybody know who supplies RED sensors today? They have done an awesome job of hiding this so far. Is it Teledyne? Dasla?

Well? The patents could be complicated. Maybe Sony only needed a "portion" of Canon's patent. They both share a basic concept of wafeform phase cancelation analysis. I can only guess that this basic concept can create a few different highways for taking it in different directions. I dont know how old that rumor really is. It might have already been a year old before it was told to me at NAB. If Canon is doing on-sensor column A/D conversion and Sony hasnt sued them yet? Hmm...that adds a bit more stregenth to the rumor. This patented function is supposedly a key part of Exmor. Who knows for sure. These moves are supposed to be "hush hush" and the few people that DO know are under NDAs...or maybe its nothing? I think that today, most of us believe that Dual Pixel and PDAF are supposedly different but seem to have almost identicle overall performance. On the cell phone side, Sony sensors DO advertise "Dual Pixel" AF. They are not marketed as Sony "PDAF". If these crossovers were NOT done with an agreement than both sides could have grounds for patent infringements. Hmmm....

This is only a rumor I heard at NAB. Fact: Sony makes the dual pixel sensors on Samsung smart phones. Rumor: Sony phase detection AF adopts Canon patents that they use on Canon's dual pixel system. Rumor has it that Sony and Canon TRADED patent rights to each other. Sony gave Canon "Exmor," on sensor, column A/D conversion patents for the rights for usage of Canon's dual pixel patent. Sony call theirs "PDAF" for Sony branded cameras but its based on that Canon patent. About a year after this deal was rumored to be done, Sony released the A6300 which was their first gen PDAF. Take that for whatever you think its worth. It may be nothing but it could be something.....

Does anybody believe that it wasnt technically possible to put this sensor inside an IBIS mount? I suspect it "could" be true but there are other marketing and camera seperation tactics possibly at play here too. Does anybody "really" believe that Panasonic could not give the G9 longer video recording limit? Or do you think that it really was optimized for dispersion of photography heat vs videographiy heat. I would argue that the GH5's ability to handle heat generated by 10bit 4k and a 400mbps CODEC writing to two cards at once with unlimited recording is a FAR more challenging heat problem than photographs...,yet its done perfectly well. Then again, I am nobody to be questioning Panasonic on their public statements or marketing choices. Whatever it is, its not right to question them.

Yeah....Panasonic specifically denied at NAB that they were purchasing a Sony sensor. Again, Chip Works let the genie out if the bottle on Panny. Its impossible for electronics companies to hide from Chip Works. They reverse engineer with extrordinary microscopic detail. They map the entire foundry of sensitive chips like image sensors and sell all that data to competitors for a heavy price. Anyway....all Panasonic can do is hope Chip Works (aka Tech Insites) doesnt target the GH5-S this time arround. They already made a liar out of Panny on the GH5. We will see. I will also poke arround Sony at NAB again this year. They were the ones that let the cat out of the bag first.

We know for a fact that the GH5 IS a Sony IMX272. Although, it not verified, we have every indication this S model is an IMX294. Even the 14bit is another give away. Im certain that when Tech Insights (Chip Works) gets this into their lab, de-solders it, X-Rays it, inspects it under their electron microscope and blue prints the entire sensors schematics, they will say its another Sony IMX sensor. (IMX294 or variant) I could be wrong but i wasnt wrong on the GH5 either... :-) (although I must admit that I was tipped off about it by a Sony engineer at the Sony NAB booth last year abour that GH5 sensor)

This DPReview database tool really does allow us to strip away the processing and get under the hood of the GH5-S and "really" see what this sensor is taking in. For days we have seen this sensor compared to an A7S-II chip with pixel wells (photo sites) TWICE the size of the GH5-S. For days we have been thinking "how is this possible...I feel like Im seeing heavy ammounts of noise reduction" So yeah,...that is the answer. After inpecting the raw sensor output, its very easy to see the A7S-II's actual signal to noise ratio significantly beats the GH5-S at every ISO/gain level. Even at a low 200 ISO. Mystery solved. Its very hard to beat the laws of "physics". I do give the GH5-S all the credit in the world though. I think Panasonic and Sony really did something very nice together on this camera. This in-camera processing is also pretty intense and is very well executed. CT

I remember seeing this IMX294 STARVIS sensor spec sheet last summer. I remember thinking; "Wow, this would make a perfect sensor for a video camera...and it does 14bit readout?..." Little did I know then, what we all know now. Yup, its got a 14bit readout. You can bet that the Panasonic DVX200 will be the next camera with this sensor. That is going to be a SWEET ENG camera. (if Sony allows it to be sold for that purpose, that is)

Yeah...from digging around the web, I'm seeing people say that the GH5 and GH5-S have the same dynamic range too. This is interesting because if Panasonic is using the STARVIS IMX294 sensor, that sensor has been measured at 13.5 stops in astronomy cameras that its used in today. This makes you wonder if Panasonic chose not to allow it to have 13.5 stops to protect the EVA-1's 14 stop advantage. If that did happen, than I dont blame Panasonic. The GH5-S already already outperforms the EVA-1 in noise and high ISO. (I dont think the EVA-1 uses a Sony sensor) Has anybody placed their GH5-S demo on DR charts yet?

I think we need to be careful about wanting to limit the voices of idiots on the web. Everybody, no matter how despicable we think they are, has a right to speak their mind. I think we would all be going down a VERY slippery slope if we started censoring people that we think that dont have the morale values that we have. The guy is a total ass. But he should have that right to be an ass. We certainly have the right to ignore him. Censorship is a very, VERY dangerous thing and its potential for abuse is massive once you start approving ways to justify doing it. Censorship is FAR more dangerous to society than this dick-head ever will be.

Based on what the Sony A9 is doing today. The two phase systems perform very closely now.

They both use signal phase cancelation to determine AF points. This is just a rumor, BTW. I dont know how true it is.

As shocking as some people would find this,...it is absolutely correct. Sony Semiconductor company is a fantastic money maker for the mother company. It FAR outweighs Sony Imaging company (camera company) in revenue. It is Sony Semiconductor's ultimate dream goal to have its sensors, readout, scaling engines and patents inside 100% of every single camera in planet Earth. Unfortunately, Sony's camera company has no say or bearing in this goal. They are literally left to "deal with it" and fend for themselves. Its kinda sad but its "business" and the best way for the mother company to survive and grow. It could be expected that Sony Imaging could order special designs and and certainly purchase sensors at a very deep discount to help them stay profitable. Big question....who owns the Sony pattent for their phase detection AF? Im guessing Semiconductor owns it. Rumor has that Sony traded pattents with Canon. Canon gave them dual pixel phase detection and Sony gave Canon their on sensor column AD conversion pattent. (Part of EXMOR) Sony cant call it "dual pixel" but it uses Canon's pattented process. And suddenly Sony had a damn good AF system that is almost identicle to Canon today. About a year after I heard that rumor, the Sony A6300 came out.....hmmmm.

I have heard soooo many people have this idea: "Sony Semiconductor can help Panasonic in MFT because its a completely different market than APS-C or full frame" Different market because of its sensor size? Haha...yeah right. The GH5 and GH5-S ABSOLUTELY does compete with Sony in sales. I dont believe that buyers are "loyal" to sensor size in that way. I believe people will go to where the "features" are first and the "end result" they can get. As a die-hard, self admitted, Sony zombie, I sold off all my APSC-C gear and moved quickly to the GH5. (I did keep my A7S-II and all my FF lenses...im not completely crazy) Anyhoo,...the GH5 is a Sony Semiconductor/Panasonic monster and Im betting the GH5-S is going to be another for Sony Imaging Comany to fear.

Thinking about the GH5-S. I really think that Panasonic and Sony Semiconductor really do team up beautifully together. With Sony's sensors and scaling engine teamed up with Panasonic's color science and feature marketing boldness, they make the GH5 one sweet camera. I can only imagine what this "dynamic duo" will come up with in a possible GH5-S. This a combo that really makes the consumers the big winner. The only loser could be Sony Imaging Company! Anybody that thinks that the GH5's MFT sensor does not compete with Sony's APS-C or full frame models....is lying to themselves.

Sensor "A" and Sensor "B"....Sensor "A" has 4micron photo well and "B" has an 8micron. Both take in 100 (low light) photons and both produce the same voltage from those 100 photons. Let's say 1,000 photons will fully saturate a 4micorn well. Those same 1,000 photons might only fill "half" of the 8micron photosite well. The two of them have the same "low light" ability when they only have the same number of few photons collected but they both produce very different voltages when the light is good or strong. The smaller one will top out fast while the larger can keep producing higher and higher voltages before it saturates. So dynamic range in bright light is a big difference. But when light is so "low" that non of the pixels saturate, they will both produce the same low light performance. Big question....If an 8micron photosite takes in 100 photons....what can TWO photosites that are 4microns each do with 50 photons and 50 photons they both collected? Can you bin or "add" those two voltage readings "together" to equal the same voltage collected from ONE 8micron photosite with 100 photons? How does pixel binning and summing actually work? Can you "really" play with and manipulate the light collection and voltage reading math like this? We are dealing with small solar panels here. In very low light situations, a Micro 4/3 surface area only has exposure to a certain number of photons and can only generate voltages based on that limited value. How you "slice and dice" and "allocate into buckets" that FIXED NUMBER of low light photons....doesn't change the amount of overall light you took in on that read cycle. You are either just bunching them up or spreading them out......but the overall photon count number doesn't change in "low light"?

Wow,...yeah, I dont know of any image sensor to ever do a stunt like this. (mixuture of long and short exposure times) Could this Sony IMX294 produce an image with motion blur in the shadows but with sharp highlights....all in the same image? Obviously, this timing trick will see into the dark or "low light" only as well as the slowest collected pixels can take in light. In theory, a similar sensor without this quad timing cluster (single timing) would see into the shadows to the exact same extent. It would only be missing the highlight protection feature that this multi-timing would bring. Does that make sense? At 1/30 of a second exposure, both, the single time exposure and the dual time exposure will collect the same amount of photons in low light. The benefit only comes to the dual time sensor in the highlights? So,..I can see this being a high dynamic range sensor but not really a "low light" beast. I dunno....im just guessing out loud.

The only thing with having 2 pixels with different ISO is that "gain" never increases dynamic range or signal to noise ratio....only light/photons do this. Dynamic range happens at the photosite level and measured by its pixel well saturation amount. Photosites that saturate (100% full) very quickly will have lower dynamic range. While photosites that can take in allot more photons without saturating will have more dynamic range. Signal "gain" doesnt improve the photosites saturation point. Dynamic range is determined before gain ever enters the picture. Remember, raw sensor data has no actual "gain" applied to it yet. Gain/ISO is a "post processing" tool to increase brightness of an already fully exposed sensor signal. In other words, "exposure" is completed and finished before signal gain is applied later down the road. What this IMX294 sensor appears to be doing is "time" shifting each pair of pixels in the cluster. One set is exposed longer and takes in MORE light and saturates quickly but the other set has a shorter exposure time, taking in LESS light and protecting saturation. The "more light" value will help your shadows and the "less light" values protect the highlights. In the cluster....2 pixels will saturate easy while the other will saturate less! The question is....how in the Hell do you merge these different values into ONE high dynamic range pixel value? My guess? Every value below middle grey comes from the slower exposed pixels while every value above middle grey comes from the faster exposed pixels?? I would also guess that these quad pixel clusters will produce a much "softer" 4k image over the current GH5's 5k oversampled image?? This is going to happen anytime you bin, sum or average chunks of pixels together. So, if you set your camera for 1/30 second shutter speed, in your image, the darker pixels are drawn from the slower exposure photosites but the brighter pixels have maybe a 1/40 or 1/50 exposure time? Damn....thats a whicked concept!

Andrew, do you want to take a shot at trying to explain how this IMX294 quad bayer circuit works? We have a cluster of 4 photosites, 2 have one exposure time while the other 2 have a shorter exposure time. One set digs deeper into the shadows, while the other set protects highlights. Pixel "size" is not their trick...exposure "time" is their trick! How do you average or bin those 4 values into one pixel ourtput high dynamic range number? This is going to be a major selling point of this GH5-S. ( if its using a IMX294)

I have been told that these organic sensers require ALLOT of power to drive. They supposedly also get very hot and will probably only show up in a large, fan cooled camcorder body. Who knows for sure? Maybe they have been able to conquer the heat and power problems and are now able to get them into a small passive cooled body under low wattage??? CT It's my guess that the GH5 and GH5-S will not match other in video features. The GH5 offers a BEAUTIFUL 6k readout and 5k to 4k (Sony style) full pixel readout and scaling engine. The GH5 offers enormous resolving power in 4k. The GH5-S will be better in low light and could offer 13+ stops of dynamic range...but at the cost of that georgous resolutuon the GH5 has. I think they will both produce two different "types" of images. I will own BOTH for different reasons! ;-)

Does anybody know how to read these Starvis IMX294 sensitivity numbers? Sensitivity (F5.6) Typ. 1900 mV 1/30s accumulation (HCG) Saturation signal Min. 970 mV Tj = 60 °C (LCG) Can anybody pull a signal to noise ratio out of this, like "60db" or anything like that? How do these numbers stack up against other known performing sensor numbers?

Wolf33d. - You could be right that this "S" model uses the same Sony IMX272 sensor that the current GH5 does today. However, the idea of Panasonic using the same. Sony style A7r/s marketing model clone goes out the window. G9 - Fast photo centric model GH5 - Great video/photo hybrid model GH5-S - Heavily optimized, for video only. With the same pixel count sensor, there is no room to improve the pixel well (photosite) saturation ability. These pixel wells will simply fill up and clip at the same amount as they do today. The only way to increase dynamic range is to lower the density of sensor and get those photosite wells larger to hold more photons. This kills it's photo worthiness completely but could allow for significant dynamic range improvements. I strongly suspect that they will buy Sony again for this "S" camera. Plus, this Starvis sensor seems to promise ALL the improved video specs that a GH5-S could ask for. (Minus 6k readout and 5k ovesampling) An 11mp video only sensor with 4.63 micron photosites will almost make you forget you are shooting MFT. This new quad Bayer pixel layout is freaking me out. How do you expose a cluster of 4 pixels with 2 different speeds and create ONE pixel value from that? Two pixels see into shadows deeper while the other two protect highlights better. That is some crazy whichcraft! Maybe this is why Luke mentioned "best V-Log ever" If this crazy idea DOES lead to impressive dynamic range, then yes,...that WOULD make for great VLog! CT

Yeah, its true that this is not a good way to measure photosites. CamSpec has the IMX284 APS-C A6500 sensor has having a pixel pitch of 3.9 microns. The IMX294 is 4.63microns so they are larger than the A6500 but not as much as my sensor size comparison has them. Yes, we dont know for sure what sensor the GH5-S will have. They purchased a Sony IMX272 for the current GH5 so they certainly have no trouble buying from Sony. Its certainly possible they could do it all over again on this "S" model. I'm also convinced they purchased Sony's 6k/5k readout to 4k scaling engine. As far as I know, Panasonic up until this GH5 has only ever pulled off a 1:1 4k deBayer process. They have never oversampled before. I could be wrong but it just smells like Sony to me. It also makes sense as Sony made the sensor for that very reason and we all know how adept Sony is at oversampling readout as they have been doing it for years now. The biggest question for me is; "What in the hell is this QUAD pixel layout with dual exposure process?" In a 4 photosite cluster, we have TWO different read and dump cycle times??? In theory, 2 pixels could have a 1/30 second exposure (to see deep into shadows) while the other two could have 1/60 exposure times to protect the highlights? How do you bin those four values into one "pixel"? Do you take the 1/30 for everything "under" middle grey....and take the 1/60 for everything "over" middle grey and then blend the readings? Does Sony count each 4 pixel cluster as 4 "seperate" sized photosites or join the four into one fat multi-exposure photosite output? I think this binning process is some kind of vudoo! Panasonic/Fuji "Organic" CMOS? I doubt it. I have read that these organic sensors get extremely hot and use enormous amounts of power. I asked a well-known Sony engineer at NAB about this organic CMOS tech and his exact quote was; "Hahaha!,...let me tell ya'... Building the Spruce Goose is ONE thing...actually FLYING it is a completely different story!" They can play with that in a lab all they want but actually making a "real" working camera with it is one hell of a task. CT

The IMX294 is 10.7. But,...Im still not getting this "quad pixel" structure that it has and how that affects "actual" pixel count. Of course, we are still not sure that the GH5-S is actually using this sensor. It seems like they will but we wont know 100% until ChipWorks (TechInsights) tears the GH5-S down and X-rays it. If Panasonic actually uses this chip and explains they are using a "quad pixel" array, than it's a 100% guarantee it's this Sony chip. However, Panasonic WILL go to great lengths to not discuss or give any "hints" that its a Sony chip. Panasonic is VERY against the idea if us all knowing who made the sensors they use. I had 3 Panasonic reps at NAB tell me "No, it's OUR chip in the GH5" ChipWorks (TechInsights) actually does detailed analysis reports on many popular electronic devises. They de-solder chips, X-ray them, do cross section imaging and closely inspect with electron microscopes,...all the parts on a mainboard. They literally break down the foundry and manufacturing process of these sensors and sell the reports to companies that want to know how their competition builds their boards and who their suppliers are. They even map out the entire "inside" logic and schematic of a chip! VERY expensive analytic data! I cant afford it, nor would I even want to pay for it. However, the outside of the report does easily identify the sensor. Anyhoo,...they absolutely nailed the GH5 with a Sony EXMOR IMX272 image sensor. So, sorry Panny and Sony,...you guys got outed,...dirty style! CT

Sorry folks, I have adjusted the sensor size numbers- here is the corrected ones... ;-) Guys,...this should "sill" blow your minds: Sony A6500 vs. (potential) "GH5-S" Physical sensor size and pixel count; - Sony A6300/6500 (Sony IMX284) 24.3mp APS-C (23.6 x 15.6mm) = 368 - GH5-S - (Sony IMX294) 11mp Micro 4/3 (17.3 x 13mm) = 225 ______________________________________________________ Average photosite (pixel) size; - Sony A6300/6500 = 368/24.3mp = 15.1 - GH5-S = 225/10.7mp = 21.0 Compared together = 21.0/15.1 = 1.39x LARGER pixels on this "GH5-S"! This Sony STARVIS IMX294 sensor "should" easily out perform the A6300/6500 in low light conditions. This does not even take into account the "quad pixel+ dual shutter speed" HDR modes that nobody understands yet. This only thing that we are going to lose is the lovely 5k oversample that we all love in the GH5 today :-( There....I think that's more accurate now! ;-) WOW!!!!