Jump to content

Megapixels


fsprow

Recommended Posts

Advertisement (gone after registration)

I've always found that virtually all of my work's value is not fine detail and doubted that more MP behind say 50 added any content of importance.  While this is I believe still true, I was astounded when I sat down with a Hasselblad rep and looked at some images, processed in Hasselblad Phocus, from the H6D-100C camera, which I am ordering one of for studio and non-moving subjects.  I believe they have image examples on their website.  Currently I have a SWC and H5D Hasselblad. 

 

Wow -- the difference between the 50 and 100 MP images was startling, not just in fine detail but overall impact. Of course in either case the medium format (sensor is larger in the 100 MP version) makes a difference too.

 

I'm sure at some point the lenses will be limiting (the Hasselblad HC lenses are roughly comparable in quality to Leica's I have found on my M240), but to my surprise we are not there yet. I will do more comparisons of my own, including versus my Nikon 810, when I get the camera.

 

Regards

 

 

Link to post
Share on other sites

I've always found that virtually all of my work's value is not fine detail and doubted that more MP behind say 50 added any content of importance.  While this is I believe still true, I was astounded when I sat down with a Hasselblad rep and looked at some images, processed in Hasselblad Phocus, from the H6D-100C camera, which I am ordering one of for studio and non-moving subjects.  I believe they have image examples on their website.  Currently I have a SWC and H5D Hasselblad. 

 

Wow -- the difference between the 50 and 100 MP images was startling, not just in fine detail but overall impact. Of course in either case the medium format (sensor is larger in the 100 MP version) makes a difference too.

 

I'm sure at some point the lenses will be limiting (the Hasselblad HC lenses are roughly comparable in quality to Leica's I have found on my M240), but to my surprise we are not there yet. I will do more comparisons of my own, including versus my Nikon 810, when I get the camera.

 

Regards

 

 

We're in rough agreement. I've considered 4000x6000 pixels (24 Mpixel) on 35mm format to be a near ideal balance point for the format and available lens technology (considering size, weight, and speed) since I did some evaluations and estimations in 2001. The step up to 50 Mpixels on this format is less important to me; I have always felt the next important step up has to be to a larger format as well as more megapixels to truly exploit the potential. That's where my interest in the X1D and larger-sensored, higher megapixel cameras, of course with suitable lenses, begins. 

 

I also have an SWC and 500CM system. The quality step I'm thinking of above was the quality step that going from 35mm film to 6x6 film made. I am constantly tempted these days to get the CFV-50c back for these cameras, but what I really really want is a back with the full native format (56x56 mm dimensions) and 80 to 120 Mpixels. It may not be available at a price I can afford in my lifetime, but I keep hoping.  :)

  • Like 2
Link to post
Share on other sites

Advertisement (gone after registration)

The sensor on the 100C has been enlarged to 53 x 40mm.  Getting close!

I can hardly wait to look at the Hasselblad images compared to Velvia scans from my Nikon 9000ED scanner.

 

Since this is an SL -X1D forum I'll mention that I'm saving $$$ for the X1D as well after the hit from the H6D!

Link to post
Share on other sites

"Just another 13mm, please!" :) 53x53 would be fine.

 

I am still on the fence about an X1D. The 30mm lens is getting close to what I'd like, but I'd rather a lens closer to 21-22mm so that it becomes a digital SWC when set to square crop mode. A 33 Mpixel square camera with that field of view would suit me well.

Link to post
Share on other sites

My first day with the SL and 90-280 APO suggests this lens would be better with a higher-resolution sensor.  Much higher.  It will take a lot more pixels to over-sample enough to avoid aliasing and color moire in fine feather detail.

I'm trying to understand what you mean, Doug. Are you saying that the 90-280 is so sharp that its resolution is limited (or even degraded) by the SL's 24 sensor? It's just not clear to me... ?

Link to post
Share on other sites

Doug, a lens cannot be limited by the sensor and the other way around. A lens will always get a better result on a better sensor, regardless of  lens resolution, a sensor will always improve its output by a better lens. it is not a weakest link situation.

Link to post
Share on other sites

??

 

presumably these issues are a result of the feathering being at a specific spatial resolution relative to the sensor resolution ..... which in turn is dictated by the distance from the subject, or lens magnification of the subject, or the sensor pixel pitch ...... so altering any one of the three factors will avoid moire and aliasing. more pixels is only one solution. 

 

I look forward to the doubling of linear resolution ...... 96mpx ..... that would make a significant quantum jump in image quality. Unfortunately the technology required that goes with the sensor to make a really good, fast, compact camera seems to lag well behind the development of the sensors themselves.....

Edited by thighslapper
Link to post
Share on other sites

Doug, a lens cannot be limited by the sensor and the other way around. A lens will always get a better result on a better sensor, regardless of lens resolution, a sensor will always improve its output by a better lens. it is not a weakest link situation.

Perhaps exemplified by the Leica S, where its best-in-class lenses make a "mere" 36mp S007 punch very, very far above its mexapixel weight, with a pretty remarkable quantity of both resolution AND smooth rendering that I find less "digital" than almost any other digital files I've seen. Pure quality glass. AF issues aside. I'd still like to see how far that sensor could see an increase in megapixels (e.g. for landscapes), however, and keep a similar acuity.

 

For what it's worth, I've played with samples off 100mp cameras. To me, they have a rough feel in terms of resolution of drum-scanned 5x4" film, but I prefer the smoother and less nervous feel of the film image.

Edited by Jon Warwick
Link to post
Share on other sites

Megapixels aren't what they used to be.

 

Well. Pixels aren't what they used to be. Pretty much everything that defines a group of pixels -- hue, saturation, brightness, transition contrast, field homogeneity -- changes from each camera to the next. And having millions of them doesn't make things any clearer.

 

So, I think we should stop talking about megapixels, and start talking about cameras as though they each have an inherent film stock. Resolution only matters in the context of contrast, color, and tone. Not to undermine the importance of resolution, just to emphasize the importance of ... everything else.

 

I suspect the H6D-100C comes with a very interesting sensor. I doubt hosting 100 million sensels is its most interesting aspect. Let's share digital negatives and discuss what we see after processing and printing. Yes?

 

Cheers,

Jon

Link to post
Share on other sites

I suggest you re-think this.

 

Actually, this has been shown to be true, to an extent. While a better lens is alway more ideal just adding resolution to the sensor does lead to more image resolution even in lower resolving lenses. But not to the same extent that a better optic does.

 

While I for one would definitely appreciate more resolution in both the SL and S systems (I keep both an A7R2 and pentax 645Z) it comes with the condition that technique also has to be at those standards. IS will help but over the resolution currently available I find I need to have far better technique, or less coffee, to get the benifits of that resolution. My A7R2 and 645Z are almost always tied down to a tripod and even with the IBIS on the Sony I can often see a difference doing that.

 

I consider 24MP about the resolution I can handhold reliably at 1/F without IS.

 

If leica made a 40+ MP version of the SL I would get one but I'd still keep my 24MP body. AND I wouldn't  but a second $11K body unless it had more resolution. I don't want two 24MP SL bodies. I'll use something else as a backup. So leica are losing at least one sale, with me. Two if they made a 50MP mirrorless S body.

 

Gordon

  • Like 2
Link to post
Share on other sites

The resolution is limited by the sensor.

 

 

The resolution of the total system is certainly limited by the number of pixels available on the sensor, but that's true no matter what lens you put in front of it and whether any specific lens has optical resolution that exceeds the sensor's acutance. 

 

I think what you are saying, whether intentionally or not, is that the SL90-280 lens has more optical resolution than the SL sensor can image. But how this translates to the negative impression implicit in your first statement ... 

 

My first day with the SL and 90-280 APO suggests this lens would be better with a higher-resolution sensor.  Much higher.  It will take a lot more pixels to over-sample enough to avoid aliasing and color moire in fine feather detail.

 

... is a question mark. If it would "take a lot more pixels to over-sample enough to avoid aliasing and color moire in fine feather detail," what's different about the SL90-280 compared to, say, the APO 280mm f/4? Would it not also 'take a lot more pixels to oversample enough to avoid aliasing...' etc for that lens?

 

If what you meant to say was "My first day with the SL and 90-280 APO showed this lens to produce very high sharpness and resolution. It would be even better with a higher-resolution sensor" then your statement would be clear. 

  • Like 1
Link to post
Share on other sites

I suggest you re-think this.

Thought experiment: The Apo-Telyt 280/4.0 will give a better result on the DMR sensor than the Telyt 280 4.8, despite the latter lens outresolving the DMR sensor by a fair margin.

Reason: Image quality is not determined by sensor resolution as such (this is of more interest for detail rendering at high magnification) but by edge contrast, which is of far higher frequency than pixel size. Thus even a low MP sensor can demand a high-resolving lens in order to obtain a high per-pixel acuity.

Link to post
Share on other sites

Thought experiment: The Apo-Telyt 280/4.0 will give a better result on the DMR sensor than the Telyt 280 4.8, despite the latter lens outresolving the DMR sensor by a fair margin.

Reason: Image quality is not determined by sensor resolution as such (this is of more interest for detail rendering at high magnification) but by edge contrast, which is of far higher frequency than pixel size. Thus even a low MP sensor can demand a high-resolving lens in order to obtain a high per-pixel acuity.

Yes, but. If you try and project a raster on a sensor and the spatial frequencies are too close, there will be aliasing. Hence the request for oversampling to avoid aliasing for that particular bandwidth.

  • Like 1
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...