Reading mode: Light Dark

A few pluses for Sigma Sd9

Go to last post
Mike Chaney said:
For example,
since the processing time required to save a final image (not raw)
will be greatly reduced due to not needing complex color
interpolation, you could substitute other much-needed things in its
place.
Click to expand...
Nope. Don't buy it. The processing time just isn't that long on modern cameras. No modern bayer camera is significantly slower processing Bayer/Jpeg than raw. At worst it can be pipelined with the very slow CF write time so that the I/O size reduction triumphs over any compute time.
Mike Chaney said:
One reason that
manufacturers continue to claim that they can't support a real
color management engine inside the camera is due to the processing
time required to take the raw image and run it through ICC profiles
in addition to having to interpolate the colors from the Bayer
design.
Click to expand...
Their processing pipeline is highly optimized for the current transformations. So, yes, this would increase time significantly. But it does not follow that it would be any faster with X3.
Mike Chaney said:
If some of these things are true, it might enable photographers
to finally depend on JPEG format images.
Click to expand...
Of course the huge, gaping hole in this theory is that the SD-9 does not support JPEG in camera. It's not clear when or if it ever will. Foveon touts their raw format as the proper format. The lack of JPEG support on the SD-9 has been a source of puzzlement since it's announcement.
Mike Chaney said:
The complex "stuff" you have to do to a Bayer raw image is
time intensive enough that you really need to do it on a
computer... outside the camera, or your save times may be too long
to be usable.
Click to expand...
Huh? What camera are you using where Bayer/JPEG is a performance issue? Now I can see arguing that due to the need to keep in-camera performance high, the in-camera routines are not optimal. That you can do a better job with more time and compute resources (although a PC is not the best signal processor architecture). That you can exploit the human eye to make the final adjustments. Of course, the same arguments apply to off-camera bayer processing as well. On the PC, X3 may be faster than Bayer.
Mike Chaney said:
but not having to do the
complex interpolation associated with the Bayer design should give
camera manufacturers some more flexibility to do other things.
Click to expand...
Again, the unstated and unproven assumption is that there is no complex processing needed for the Foveon sensor. But we still have no description and no numbers yet. Phil won't comment because the SW is still beta. When this part is proven and efficient and fast in-camera processing of X3 is available, perhaps some of this might be possible.

--
Erik
 
JoePhoto said:
I just cannot believe so many people want to stick with "old"
technology and just cannot accept that something "better" may be
possible.
Click to expand...
Actually my problem with most of the posts on Foveon stuff is that much of it makes little sense technically. Lots of BS floating around...

E.g. smaller file size for higher quality images. No. The Foveon RAW files that the camera actually produces are 6 to 8 megabytes apiece. This is quite large compared to RAW and JPEG output from current 6 megapixel cameras.

For JPEGed output, sharper images inherently require more space. ("Sharper" generally means "has more high frequency components.") If equivalent size images from Bayer filter based cameras (e.g. downsampled D60 output) are compared to Foveon images, both JPEGed at the same level, the Bayer images should be smaller. Of course this is an experiment to be done.

The Foveon output looks great, but it is not blowing the doors off high quality output from 6 megapixel cameras, much less the output from the DCS-14n and the EOS-1Ds. At the likely $1500 price-point though, the SD9 is looking like an image quality bargain. (I haven't handled the camera so I have no idea how it fares from that point of view.)

I have not changed my attitude that Foveon X3 is a very significant technology and I expect it to be somewhat quite prevalent in 10 years. How it ramps up in the shorter term, I don't know. I expect the Foveon F7 (chip used in the Sigma SD9) to be more relvant to the prosumer market place than in the higher-end market of pro DSLRs. The SD9 is somewhat of a bridge product and hopefully it will do well for Sigma and Foveon. The F7 in an Olympus E20 type product would be really quite something. X3 technology in a D60 form factor or even better, full-frame 24mmx36mm would be truly amazing, but I don't expect to see that in the next year.

-Z-
 
Erik Magnuson said:
Mike Chaney said:
For example,
since the processing time required to save a final image (not raw)
will be greatly reduced due to not needing complex color
interpolation, you could substitute other much-needed things in its
place.
Click to expand...
Nope. Don't buy it. The processing time just isn't that long on
modern cameras. No modern bayer camera is significantly slower
processing Bayer/Jpeg than raw. At worst it can be pipelined with
the very slow CF write time so that the I/O size reduction triumphs
over any compute time.
Click to expand...
Plus the Foveon has to pull three times as much data off the imager. One simply can't ignore such a large factor in pointless hypothesizing about the design space.

-Z-
 
You have to take MTF ratings with a grain of salt. Unless they can collect data from a group of lenses, we're getting a distorted and likely unrepresentative lens test. What if they get a bad lens for their test? What if they get a spectacular one? I'm not saying not to read the MTF tests, but just don't expect them to be the last word.

That said, the high-end Sigma lenses get very good reviews on my site. For a long time the Sigma 105 macro was the highest rated prime in the reviews. Although I am pretty much addicted to expensive Canon glass, I would happily recommend the good Sigma lenses to people - especially someone interested in the SD9. I would not let Sigma lenses be the reason you decided not to buy the SD9. If you are willing to pay for the good stuff, they've got it. Whether it's as good as the best Nikon or Canon is another issue - and maybe an unimportant one.

John Shafer
http://www.PCPhotoREVIEW.com
http://www.PhotographyREVIEW.com
 
All the current DSLR's have their own limitatons. Time will tell how well the SD9 holds up ( it is a first for Sigma after all ). But the iamge quality on Phil's site is simply incredible given the limitations and the great price point, especially if street price drops another $200 USD or so. I originally thoght NO WAY to considering the SD9 due to the many limitations such as ISO400 and the Sigma lens collection ( not due to quality just my like for Tokina glass ). But, nothing is static and if the SD9 sells well who knows if Tamron or Tokina will start making SA mounts as an option. I agree it is not a pro camera. Also admit Sigma needs some faster glass in mid range, such as a 50 mm f1.4 or 1.8, an 85 mm f1.4 or 1..8 and so on. But then again it would be nice if the D60 had lightning fast AF and if the D100 could make jpgs lok like the S2, and if the S2 had a faster flash sync and.... each has limitations. The SD9 has perhaps more than most but at same tiem WOW to the iamge quality :) And the simplicity to getting there. Dont have to be a Photoshop wizard to get nice prints out of camera :)Before I get the post process flame ( I own Potoshop and like it and ont afraid to use it Just do not always want to HAVE to use it ) PLEASE remember each person has their own needs and wants and those needs are valid for them but not for everyone. Is why we have options... just pick the DSLR which best fits your needs and have fun. besides the more on the market means supply may soon catch up to demand and more price drops.. hmmm Maybe an SD9 for Christmas lol

PS I wrote Sigma regarding the rumour about remounting the lenses from SA to Nikon or Canon.. will post reply if I ever get one!
David Barkin said:
Your point is well taken. I like the Sigma lenses, I don't think
the quality is that bad at all - But you have summed up some of the
problems, not with Foveon technology but with this particuler
camera.

Even I, a non-professional, need low light lenses. 400ISO, another
problem.

I've heard talk of adapters for lenses (proably vapor ware) and
I've heard that Sigma will modify the mounts for Canon or Nikon for
a stiff fee, $400 (also just a rumor). So, while I am absolutely
sold on the Foveon technolgy. I really don't think that existing
6Mp sensors are even comparable to this Foveon chip, but as you
point out you have to use this sensor with the camera.

If any of the above reumors prove true I will seriously consider it
despite the limited ISO. More likely I will wait. One thing I can
say decisively is that I can't imagine purchasing another bayer
sensor camera.

Dave
Justin Case said:
David Barkin said:
I own a 28 1.8. There are quite a few more. As far as quality I'm
not qualified to answer.
Click to expand...
Hi David. Just a few nights ago, I shot a concert with an 85 1.4. I
was at
ISO 1600, shooting hand held by necessity given the venue. Wide
open at 1.4, at
ISO 1600, the best I could get was 1/30th second. This is normally
below where I would shoot with a lens of the focal length, but I
managed to get some good shots. With anything less than a 1.4 lens,
I would not have been able to shoot at all. Shooting with little
available light happens in Wedding and Photo Journalistic
situations on a regular basis. Both Canon and Nikon make 1.4 or
faster lenses.

Canon even has a 1.0, and a 1.2. Sigma does not, and they have very
few at 1.8. This is my point about people positioning the camera
as a tool for professionals. It isn't going to fit the bill for
many. It sounds like it will be a wonderful tool for advanced
amateurs, and even a few Pros may be tempted to use it. But this
limitation is what prompted my opinion that Foveon wasted an
opportunity giving exclusive rights to Sigma. Just think of what
would have happened had they housed it in an EOS-1D or F5 body!

So sad.
Click to expand...
Click to expand...
 
Speed of file conversion will be an issue.. if it takes more than 30-45 seconds per image it could come into play for some people. Time will tell.
Erik Magnuson said:
Mike Chaney said:
For example,
since the processing time required to save a final image (not raw)
will be greatly reduced due to not needing complex color
interpolation, you could substitute other much-needed things in its
place.
Click to expand...
Nope. Don't buy it. The processing time just isn't that long on
modern cameras. No modern bayer camera is significantly slower
processing Bayer/Jpeg than raw. At worst it can be pipelined with
the very slow CF write time so that the I/O size reduction triumphs
over any compute time.
Member said:
One reason that
manufacturers continue to claim that they can't support a real
color management engine inside the camera is due to the processing
time required to take the raw image and run it through ICC profiles
in addition to having to interpolate the colors from the Bayer
design.
Click to expand...
Their processing pipeline is highly optimized for the current
transformations. So, yes, this would increase time significantly.
But it does not follow that it would be any faster with X3.
Member said:
If some of these things are true, it might enable photographers
to finally depend on JPEG format images.
Click to expand...
Of course the huge, gaping hole in this theory is that the SD-9
does not support JPEG in camera. It's not clear when or if it ever
will. Foveon touts their raw format as the proper format. The lack
of JPEG support on the SD-9 has been a source of puzzlement since
it's announcement.
Member said:
The complex "stuff" you have to do to a Bayer raw image is
time intensive enough that you really need to do it on a
computer... outside the camera, or your save times may be too long
to be usable.
Click to expand...
Huh? What camera are you using where Bayer/JPEG is a performance
issue? Now I can see arguing that due to the need to keep in-camera
performance high, the in-camera routines are not optimal. That you
can do a better job with more time and compute resources (although
a PC is not the best signal processor architecture). That you can
exploit the human eye to make the final adjustments. Of course, the
same arguments apply to off-camera bayer processing as well. On
the PC, X3 may be faster than Bayer.
Member said:
but not having to do the
complex interpolation associated with the Bayer design should give
camera manufacturers some more flexibility to do other things.
Click to expand...
Again, the unstated and unproven assumption is that there is no
complex processing needed for the Foveon sensor. But we still have
no description and no numbers yet. Phil won't comment because the
SW is still beta. When this part is proven and efficient and fast
in-camera processing of X3 is available, perhaps some of this might
be possible.

--
Erik
Click to expand...
 
JDavis37 said:
Speed of file conversion will be an issue.. if it takes more than
30-45 seconds per image it could come into play for some people.
Click to expand...
Sure. The point I wanted to make is that PC processing time != camera processing time. Even if the Sigma PC conversion is faster than Bayer conversion on the PC, that does not mean that it will be faster on-camera which is a completely different processing environment.

--
Erik
 
Z (is real) said:
E.g. smaller file size for higher quality images. No. The Foveon
RAW files that the camera actually produces are 6 to 8 megabytes
apiece. This is quite large compared to RAW and JPEG output from
current 6 megapixel cameras.
Click to expand...
RAW file output: SD9 = 6-8 MB, D60 = 7.4 MB, D100 = 9.5 MB , S2 = 12.8 MB.
JPEG output: SD9 = 1.3 MB. D60 = 2.5 MB, D100 = 2.5 MB, S2 = 2.2 MB

Other then the RAW being close to the D60 and a little smaller then the D100, I really failed to see how the SD9 is "quite large compared to RAW and JPEG output from current 6 megapixel cameras" can be made as a general statement.

--
jc
Sony F707
http://www.reefkeepers.org/gallery/f707
http://www.reeftec.com/gallery
 
Mike Chaney said:
Most importantly, freeing up the time needed to interpolate the
Bayer design might finally mean that manufacturers will be able
to produce images in known color spaces!
Click to expand...
Mike Chaney said:
Will the Foveon technology finally
open the door for manufacturers to start doing real color
management in cameras?
Click to expand...
You know, what is the color space of the X3 chip? I dont think read anything that said anything about this other then "real color".

--
jc
Sony F707
http://www.reefkeepers.org/gallery/f707
http://www.reeftec.com/gallery
 
Jimmy Chen said:
Jimmy Chen said:
E.g. smaller file size for higher quality images. No. The Foveon
RAW files that the camera actually produces are 6 to 8 megabytes
apiece. This is quite large compared to RAW and JPEG output from
current 6 megapixel cameras.
Click to expand...
RAW file output: SD9 = 6-8 MB, D60 = 7.4 MB, D100 = 9.5 MB , S2 =
12.8 MB.
Click to expand...
Hmmm. I was going on expected scaling of the D60 from D30 numbers which would have placed it in the 6 to 7 meg range. So even then "quite large compared to..." is not warranted. However the D100 and S2 numbers look to me like their compression is not very good. (Are you sure those are RAW and not some sort of TIFF?)
Jimmy Chen said:
JPEG output: SD9 = 1.3 MB. D60 = 2.5 MB, D100 = 2.5 MB, S2 = 2.2 MB
Click to expand...
You have to resize the D60 etc. images to equivalent dimensions for that comparison. There is more information in the SD9 image per pixel so it should not compress as well. More importantly, I definitely cannot see the images being smaller given an equivalent JPEG compressor. (Who knows, maybe Foveon's conversion software has better JPEG compression. Though that is a very mature technology at this point.)

-Z-
 
Jimmy Chen said:
You know, what is the color space of the X3 chip? I dont think read
anything that said anything about this other then "real color".
Click to expand...
The chip does not have a color space that you would recognize. Each of its sensor regions has a sensitivity curve against the spectrum of incoming "light." Among other things, these overlap, but it is well correlated to a real color space and a transform can be constructed to convert the data, but the transform is specific to the Foveon chip. (This is true for all sensors by the way. And it is one of the reasons RAW files for different cameras are tricky to convert to normal images. For Bayer filter sensors, one can choose the dues in the filters to change the transform for specific optimizations.)

It is sort of a miracle of the physical world that the depth/frequency transfer curves for silicon can be well correlated to what we percieve as color.

-Z-
 
Z (is real) said:
JoePhoto said:
I just cannot believe so many people want to stick with "old"
technology and just cannot accept that something "better" may be
possible.
Click to expand...
Actually my problem with most of the posts on Foveon stuff is that
much of it makes little sense technically. Lots of BS floating
around...

E.g. smaller file size for higher quality images. No. The Foveon
RAW files that the camera actually produces are 6 to 8 megabytes
apiece. This is quite large compared to RAW and JPEG output from
current 6 megapixel cameras.

For JPEGed output, sharper images inherently require more space.
("Sharper" generally means "has more high frequency components.")
If equivalent size images from Bayer filter based cameras (e.g.
downsampled D60 output) are compared to Foveon images, both JPEGed
at the same level, the Bayer images should be smaller. Of course
this is an experiment to be done.

The Foveon output looks great, but it is not blowing the doors off
high quality output from 6 megapixel cameras, much less the output
from the DCS-14n and the EOS-1Ds. At the likely $1500 price-point
though, the SD9 is looking like an image quality bargain. (I
haven't handled the camera so I have no idea how it fares from that
point of view.)

I have not changed my attitude that Foveon X3 is a very significant
technology and I expect it to be somewhat quite prevalent in 10
years. How it ramps up in the shorter term, I don't know. I expect
the Foveon F7 (chip used in the Sigma SD9) to be more relvant to
the prosumer market place than in the higher-end market of pro
DSLRs. The SD9 is somewhat of a bridge product and hopefully it
will do well for Sigma and Foveon. The F7 in an Olympus E20 type
product would be really quite something. X3 technology in a D60
form factor or even better, full-frame 24mmx36mm would be truly
amazing, but I don't expect to see that in the next year.

-Z-
Click to expand...
Why dont you go download a sample from Phil's gallery before posting. Sd9 jpg took approx 1.4mb whereas a jpeg from D60 took average 2.2mb.

Sharper does not nessecary means more data. Each pixel from the Sd9 image is exactly what the camera sees. Whereas each pixel from bayer-type camera is a result from the calculation and estimation of the camera. Therefore you can have an image with 6 crappy megapixel which doesnt even have to look like anything and still has a larger size than an image with 1 megapixel but each pixel has a correct RGB value.

Check ur fact before criticize ppl
 
nvn007 said:
Z (is real) said:
JoePhoto said:
I just cannot believe so many people want to stick with "old"
technology and just cannot accept that something "better" may be
possible.
Click to expand...
Actually my problem with most of the posts on Foveon stuff is that
much of it makes little sense technically. Lots of BS floating
around...

E.g. smaller file size for higher quality images. No. The Foveon
RAW files that the camera actually produces are 6 to 8 megabytes
apiece. This is quite large compared to RAW and JPEG output from
current 6 megapixel cameras.

For JPEGed output, sharper images inherently require more space.
("Sharper" generally means "has more high frequency components.")
If equivalent size images from Bayer filter based cameras (e.g.
downsampled D60 output) are compared to Foveon images, both JPEGed
at the same level, the Bayer images should be smaller. Of course
this is an experiment to be done.

The Foveon output looks great, but it is not blowing the doors off
high quality output from 6 megapixel cameras, much less the output
from the DCS-14n and the EOS-1Ds. At the likely $1500 price-point
though, the SD9 is looking like an image quality bargain. (I
haven't handled the camera so I have no idea how it fares from that
point of view.)

I have not changed my attitude that Foveon X3 is a very significant
technology and I expect it to be somewhat quite prevalent in 10
years. How it ramps up in the shorter term, I don't know. I expect
the Foveon F7 (chip used in the Sigma SD9) to be more relvant to
the prosumer market place than in the higher-end market of pro
DSLRs. The SD9 is somewhat of a bridge product and hopefully it
will do well for Sigma and Foveon. The F7 in an Olympus E20 type
product would be really quite something. X3 technology in a D60
form factor or even better, full-frame 24mmx36mm would be truly
amazing, but I don't expect to see that in the next year.

-Z-
Click to expand...
Why dont you go download a sample from Phil's gallery before
posting. Sd9 jpg took approx 1.4mb whereas a jpeg from D60 took
average 2.2mb.

Sharper does not nessecary means more data. Each pixel from the Sd9
image is exactly what the camera sees. Whereas each pixel from
bayer-type camera is a result from the calculation and estimation
of the camera. Therefore you can have an image with 6 crappy
megapixel which doesnt even have to look like anything and still
has a larger size than an image with 1 megapixel but each pixel has
a correct RGB value.

Check ur fact before criticize ppl
Click to expand...
Just checked out Phil's spec. Each D100's Raw image is about 10mb compare to 6mb from Foveon. Each D100's Tiff is about 17mb compare to 10mb from the Foveon. Canon's Raw might be smaller because they have their own proprietary format (which I love with my measily 128mb CF card and my G2)).

You are full of it man
 
Jimmy Chen said:
RAW file output: SD9 = 6-8 MB, D60 = 7.4 MB, D100 = 9.5 MB , S2 =
12.8 MB.
Click to expand...
Of course the D60 number includes about 1.4MB for the embedded 3 MP JPEG (a feature not available on the SD-9 at any price.) The D100 is w/o compression - but compression increases the processing time immensely. (Of course, we don't know the SD-9 processing time....) The S2 is totally uncompressed and padded out to 16 bits as well. Easy to decode, but not compact.

In fact, one thing bothers me. The SD-9 compression works TOO well. If the output really is 3.5 X 3 X 1.5 (e.g. 12 bits) = 18 MB uncompressed. But they are getting 3:1 lossless on photo data. This is WAY better than expected. PNG and compressed TIF get are almost never more than 2:1 and more typically 1.5:1.

Possible explanations:

1. The compression really is that good; It takes advantage of the properties of X3 files to do a lot better than standard algorithms.

2. There is not really 12 bits of real data for all channels, so the initial size is wrong. Even if padded to 12 bits, the extra bit would be more compressable if not random.

3. The compression is not lossless, but low loss. I think the D100 compression is like this, using a logarithmic-type encoding.

--
Erik
 
Member said:
Of course the D60 number includes about 1.4MB for the embedded 3 MP
JPEG (a feature not available on the SD-9 at any price.)
Click to expand...
And where did you read this information? Phil very clearly stated in his review that they are two files.

Image processing sequence:

Record data as it comes off the CMOS sensor, unprocessed data (approx. 9.3 MB per shot)
1. Store this unprocessed data in the SDRAM buffer
2. Process this data into image files (JPEG or compressed RAW)

3. Buffer these converted image files (JPEG approx. 2.5 MB or RAW approx. 7.0 MB)
4. Write JPEG / RAW image files to CF card

--
jc
Sony F707
http://www.reefkeepers.org/gallery/f707
http://www.reeftec.com/gallery
 
Jimmy Chen said:
Jimmy Chen said:
Of course the D60 number includes about 1.4MB for the embedded 3 MP
JPEG (a feature not available on the SD-9 at any price.)
Click to expand...
And where did you read this information? Phil very clearly stated
in his review that they are two files.
Click to expand...
Phil doesn't mention it. But the Imaging-Resource review does: ( http://www.imaging-resource.com/PRODS/D60/D60A10.HTM )
Jimmy Chen said:
NOTE that the D60's RAW format now automatically includes an embedded medium/fine JPEG as well, which several of Canon's various software packages can extract with a new button called "Extract JPEG."
Click to expand...
The 1d writes them as separate files. The S2 RAW also includes a JPEG, but it's much smaller: only 1.3 MP or about 350k.

--
Erik
 
nvn007 said:
Why dont you go download a sample from Phil's gallery before
posting. Sd9 jpg took approx 1.4mb whereas a jpeg from D60 took
average 2.2mb.
Click to expand...
First off, you are being rude. This is not cool.

The D60 image is significantly larger in terms of number of pixels. Resize it down to the same size as the Foveon output and compare JPEGed sizes. The Foveon may look better, but it will not be smaller. You can also resize the Foveon image up. This should not affect its JPEG size much, but I wouldn't expect it to look very good. If you use something like Genuine Fractals, the JPEG size will increase. (Not sure how much.)
nvn007 said:
Sharper does not nessecary means more data.
Click to expand...
Actually, for JPEG in scene with detail (i.e. where shaper makes a difference) it does.
nvn007 said:
Check ur fact before criticize ppl
Click to expand...
The images may look better, but the explanation for how it works with JPEG, given equivalent image dimensions, makes no sense. If the JPEG files are actually smaller, it is a very interesting situation, but it likely has to do with superior compression algorithms or some such. Or perhaps the noise properties of the two systems are different. It is not true though that sharper images JPEG better than less sharp ones.

-Z-
 
Z (is real) said:
nvn007 said:
Why dont you go download a sample from Phil's gallery before
posting. Sd9 jpg took approx 1.4mb whereas a jpeg from D60 took
average 2.2mb.
Click to expand...
First off, you are being rude. This is not cool.

The D60 image is significantly larger in terms of number of pixels.
Resize it down to the same size as the Foveon output and compare
JPEGed sizes. The Foveon may look better, but it will not be
smaller. You can also resize the Foveon image up. This should not
affect its JPEG size much, but I wouldn't expect it to look very
good. If you use something like Genuine Fractals, the JPEG size
will increase. (Not sure how much.)
Z (is real) said:
Sharper does not nessecary means more data.
Click to expand...
Actually, for JPEG in scene with detail (i.e. where shaper makes a
difference) it does.
Z (is real) said:
Check ur fact before criticize ppl
Click to expand...
The images may look better, but the explanation for how it works
with JPEG, given equivalent image dimensions, makes no sense. If
the JPEG files are actually smaller, it is a very interesting
situation, but it likely has to do with superior compression
algorithms or some such. Or perhaps the noise properties of the two
systems are different. It is not true though that sharper images
JPEG better than less sharp ones.

-Z-
Click to expand...
I might seem rude because you said ppl give "BS" information without leginimate proof. As for raw image, Sd9 photo its as big as its suppose to be. Compare 3.5 mega pixel to 6.1 megapixel and 6mb to 10mb from D100. A simple ratio calculation will show both have comparable size. We not discussing Jpeg compression method here. At this high level of technology we do not expect to any camera to have inferior method of jpeg compression, that's just like shooting yourself in the foot(Although to my understanding SD9 can only capture Raw and convert it to Jpeg at the PC). No matter how good the compression is, remember the rule "Garbage in=Garbage-out", and that's what we are say here, foveon capture a better image period. How much better? At this point it is still being debating.

If you can dish out be ready to take it too.

Sorry you have to see the my ugly side
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top