6K Gemini?

[Christoffer Glans]

So, the question was put out in the reveal thread, but I wanted to ask in general why this isn't viable? I mean, it's even bigger than s35 than Dragon 6K and it gives the benefit of downscaling to 4K while having the low-light performance (might even be better in 6K) of the current Gemini.

Is it as simple as the plate used for batching-baking sensors unable to produce enough sensors to make the price viable? I.e, the Dragon 6K and Gemini fits as four sensors on one plate and if going bigger it requires like two or even just one for each plate, increasing the price dramatically?

If so, that's a logical reason, but if not, why isn't a 6K Gemini a better way to go?

 

[Fabricio Morato]

I have no insight what so ever, but a few thoughts:

- this sensor tech was custom made for a client, then changed a little bit it to better fit the needs for film making. So there might not have been further investment to a 6k size..
- 5k is s35ish size, 6k would be FF and red had just released it's new VV sensor, which competes with other FF sensor from other manufactures. So they might be waiting down the line to make it bigger..
- and if they go bigger they might be looking towards 8k, large format size, which could compete more directly with the Alexa 65.. so 5k and 8k Gemini and Monstro 8k in the "middle".

I might be totally off though.. pure speculation.

 

[Christoffer Glans]

- 5k is s35ish size, 6k would be FF and red had just released it's new VV sensor, which competes with other FF sensor from other manufactures. So they might be waiting down the line to make it bigger..

Not sure if 6K would be considered FF though? If 6K Dragon is at the same size as Gemini 5K and an 8K Dragon is FF, then isn't a Dragon equivalent closer to a 7K step (7K on a Dragon 8K sensor or Monstro)?

 

[David Collard]

Wouldn't it be nice to have a 6K camera with 7-8 micron pixels as well as a sensor that is first and foremost designed for 2x anamorphic?

 

[Fabricio Morato]

My math could be off but it was said that Gemini pixel size is 6micron. Red 6k is 6144 * 0.006 = 36.8cm.. little more than fullframe..

 

[Misha Engel]

My math could be off but it was said that Gemini pixel size is 6micron. Red 6k is 6144 * 0.006 = 36.8cm.. little more than fullframe..

From my understanding FullFrame has more to do with the image circle. A FF lens should at least cover a 36x24 mm, which gives an image circle of sqrt(36^2+24^2) = sqrt(1872) = 43.267 ~ 43.3 mm.

 

[Fabricio Morato]

Yes, but then Red has only to make it tall enough. The 6k width matches FF.

But again all speculation here. Don't know how much sense it makes for Red to make a 6k Gemini.

 

[Misha Engel]

Don't know how much sense it makes for Red to make a 6k Gemini.

1.5 is the sweetspot for optimal De-Bayer.

ARRI alexa/amira 2880 > 1920 is the most famous one (Full HD).
For Red DCI this would mean 4096 x 1.5 => 6144 x 3240.

But since they have monstro for the high-end I don't see it comming.

 

[Fabricio Morato]

Yep... This is more or less what I was thinking... that they already have a lot of sensors in their lineup. Different Ks and sizes.
Adding an extra 6k gemini, when they have a 5k gemini, 6k dragon, 8k helium, 8k monstro.. what's the business move?
that's why I was speculating an 8k Gemini... around 50x26mm.. or stitching 3 geminis, ala alexa 65, 54x30mm hehe ok I'll stop now

 

[Christoffer Glans]

Yep... This is more or less what I was thinking... that they already have a lot of sensors in their lineup. Different Ks and sizes.
Adding an extra 6k gemini, when they have a 5k gemini, 6k dragon, 8k helium, 8k monstro.. what's the business move?
that's why I was speculating an 8k Gemini... around 50x26mm.. or stitching 3 geminis, ala alexa 65, 54x30mm hehe ok I'll stop now

The best move I think is to go 6K and abandon the 5K model, if it's possible to bake enough sensors to support 6K Geminis. I mean, you get everything a 5K Gemini does within that system so it's evolutionary. They abandoned the 6K Magnesium a few months after it was released, so it's not the wildest idea to just change the concept to a 6K Gemini. The thing is that a 6K Gemini is a perfect upgrade for anyone with a 6K Dragon. It's a superior sensor, but not 8K which isn't what everyone wants. However, the benifits of the downsampling from 6K is notable compared to 5K down to 4K. Another reason is that 5K doesn't give much room for reframing and stabilizing without loosing the downsampling quality. Finchers productions usually do a 5K crop of the 6K image to be able to do this and then downsample to 4K mastering. The only reason I think a 6K can't be done is that the "cake" in which sensors are batched can't support enough sensors to keep the price of the Gemini camera down.

I would love if we could have the question answered by Red, since if we know the process of making the sensor doesn't support it, there's no reason to proceed with the idea and better to focus on an 8K version at the price closer to the Monstro. But if it's possible to make 6K versions with just a slight increase in pricepoint for it, it will become the camera to appeal to those who aren't jumping on board Helium or Monstro, mainly 6K Dragon owners. I see much more potential in a 6k Gemini than a 5K. Personally, the only reason I'm not keen on changing up my MG to a Gemini is that I loose 1K. It's actually a big deal.

 

[Samir Patel]

I want a 6K Gemini.

 

[Nick Morrison]

I have to assume Gemini may play a big part of RED's future.

It hit's an amazing sweetspot of resolution/data/performance at that pixel size.

A 6K Gemini sensor would be full frame.

An 8K Gemini would be almost 50mm across - that's MASSIVE.

A 4K Gemini would basically be s35.

I can see them phasing out the Dragon sensor pretty soon, and dropping it into different tier cameras:

RAVEN Gemini: 4K
Epic W Gemini: 5K
Weapon Gemini: 6K or 8K

Just musing here.

 

[Blair S. Paulsen]

I have to wonder if there are enough potential Gemini 6K buyers who wouldn't be fine with the 5K version to justify the NRE. There are certainly some, as evidenced by this thread, but how many? Perhaps RED could do a pre-order (with deposit) offer to judge financial viability? For example: if less than X number of people sign up within 30 days then everyone gets their deposit back, if that number is eclipsed then it gets made, simple.

FWIW, based on my observation of RDC over the last 10+ years, they likely have new sensor tech under internal development that is more evolved that anything currently shipping - including Gemini. I'd further wager that lessons learned in the Gemini program will improve the sensors to come. Just my .02.

Cheers - #19

 

[Nick Morrison]

I have to wonder if there are enough potential Gemini 6K buyers who wouldn't be fine with the 5K version to justify the NRE. There are certainly some, as evidenced by this thread, but how many? Perhaps RED could do a pre-order (with deposit) offer to judge financial viability? For example: if less than X number of people sign up within 30 days then everyone gets their deposit back, if that number is eclipsed then it gets made, simple.

FWIW, based on my observation of RDC over the last 10+ years, they likely have new sensor tech under internal development that is more evolved that anything currently shipping - including Gemini. I'd further wager that lessons learned in the Gemini program will improve the sensors to come. Just my .02.

Cheers - #19

6K Gemini would be a direct competitor at Venice, so yeah I think it's definitely viable. If it's less $$$ than Monstro, watch out.

Imagine if 6K Gemini is $30-40K???

Game over.

RED wins.

 

[Blair S. Paulsen]

One thing we don't know, is how long it would take RED to have QC'd 6K Gemini sensors ready to ship in quantity. If that could be done quickly enough then perhaps Nick's assessment is correct. If it would take a year or more, which I suspect, most of the folks hot for Gemini will already have a 5K version in hand.

The other issue is the sheer number of available lenses designed for motion capture that only cover S35, versus the pool of FF lenses with the operational characteristics required for narrative production. Moreover, if you're looking to really exploit the big format look, wouldn't you just go Monstro? Or ARRI LF, Sony Venice, etc. Gemini at 5K with a 6 micron pitch will vignette on some S35 glass (especially wides) but, based on Phil's extensive notes, most DP's favorite lens sets could still be used with occasional windowing to 4.5K.

Cheers - #19

 

[Misha Engel]

6K Gemini would be a direct competitor at Venice, so yeah I think it's definitely viable. If it's less $$$ than Monstro, watch out.

Imagine if 6K Gemini is $30-40K???

Game over.

RED wins.

There are more brands than Sony and RED and the winning brand at the moment is another one and they are pretty good in winning.

 

[Nick Morrison]

There are more brands than Sony and RED and the winning brand at the moment is another one and they are pretty good in winning.

Yes, but Arri will always have market share. However to your point, with the Mini going up in price recently, RED has a real opportunity to make some noise right now.

 

[Mike Quill]

This entire conversation is silly. You guys are advocating for yet another flavor of a Red sensor? LMAO!!! Red released way too many convoluted sensors, muddied the waters, and still doesn't make an image as desirable as an Alexa Classic.

 

[Hrvoje Simic]

I'd be surprised if FF 6K Gemini was not considered.

 

[Jake Wilganowski]

So is the the dynamic range and highlight roll off in gemeni in standard mode equivalent to dragon with equivalent OLPF? I know it clips a stop sooner in low light mode. But the range, clipping, and rolloff, more similar to helium or dragon? I personally am a bigger fan of dsmc2 dragon than helium. And I've shot a lot of red.