Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma ..

In article <230220082115581764%rag@nospam.techline.com>, Mr. Strat
<rag RemoveThis @nospam.techline.com> wrote:

> In article <47c09995$0$26122$9a6e19ea@news.newshosting.com>, Michael
> Tuthill <tuthill RemoveThis @nospamshaw.ca> wrote:
>
> > Well here's a scoop for you. Both my D200 and my SD14 will shoot in
> > jpeg mode. Don't use it on either though.
>
> I thought the Sigma only did RAW. But, since I've never seen one in the
> wild, I'm just going on what I've heard.

sd9/10 were raw only. sd14 offers both raw *or* jpeg, not both at the
same time.
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

nospam wrote:
> In article
> <b6c535dd-f2ba-47cf-a55e-249a6c037498.RemoveThis@z17g2000hsg.googlegroups.com>,
> acl <achilleaslazarides.RemoveThis@yahoo.co.uk> wrote:
>
>> On Feb 22, 11:38 pm, "Peter Stavrakoglou" <nto....RemoveThis@optonline.net>
>> wrote:
>>
>>> Sigma does reverse engineer the protocol. Re-chipping is necessary because
>>> Canon changes the protocol from time-to-time.
>> OK but then how come old canon lenses don't need rechipping if the
>> protocol's changed?
>
> canon knows what's in their old lenses so they can make a change that
> won't cause a problem with them, or they know how to work around any
> problems that might occur.
>
> sigma doesn't have that luxury. they only know about what they reverse
> engineered, and perhaps they made a few assumptions that turned out to
> be incorrect.
>
> even with published specs, things don't always work out. just look at
> web browsers and how they each render web pages a little differently.

Do Tamron or others pay for full specs?
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

David J. Littleboy wrote:
> acl wrote:

>> Anyway, one way or another, there are compatibility problems, so I
>> don't know.
>
> (The following is all "to the best of my knowledge" based on not having
> noticed any complaints of problems on the Canon forums in a long time.)
>
> Well, no. There _have been_ compatibility problems. Past tense.
>
> As I understand it, there haven't been compatibility problems for a long
> time, and the only lenses that had problems were ones that have been long
> since discontinued.
>
> Have I missed more recent problems? What's the most recent Sigma lens that
> has failed to work on a newer body?

My Sigma 12-24 has an update available for using on a D200, some one
particular metering mode or something, I read about it & it wasn't
anything that effected me so I didn't send it in.

> And even the claim that Sigma doesn't license the Canon interface sounds
> like an urban myth.
>
> Of course in this case the Sigma's excessive CA makes it a lens to be
> avoided. Especially since the Tamron is so much better and cheaper. (The
> similar Canon lens is almost three times the price (list price in Japan) of
> the Tamron.)
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

On 2008-02-23 22:15:58 -0700, "Mr. Strat" <rag.RemoveThis@nospam.techline.com> said:

> In article <47c09995$0$26122$9a6e19ea@news.newshosting.com>, Michael
> Tuthill <tuthill.RemoveThis@nospamshaw.ca> wrote:
>
>> Well here's a scoop for you. Both my D200 and my SD14 will shoot in
>> jpeg mode. Don't use it on either though.
>
> I thought the Sigma only did RAW. But, since I've never seen one in the
> wild, I'm just going on what I've heard.

They added jpeg to the SD14.

>
>> I'm sure you do have some very nice prints. Pretty well every dslr on
>> the market takes nice shots these days. As far as the megapixel count
>> goes, would you not concur that the bayer method of counting is kinda
>> stretching things? If you have a 10 mp rig does it have 10 mp red, 10
>> mp green and 10 mp blue photosites? For that matter does it even have
>> 5 mp red, 5 mp green and 5 mp of blue?
>
> And the Foveon in the new Sigma is 14mp? I don't think so. Keep trying.
> Sooner or later you'll see that you pissed away your money on that
> piece of junk.

Well it's my money and it does take pictures that I, and a few other
people, are happy with. In some situations I prefer it to my D200. My
only real gripe about it is that it's relatively slow.
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

On 2008-02-23 22:32:08 -0700, nospam <nospam.DeleteThis@nospam.invalid> said:

> In article <47c0e702$0$26254$9a6e19ea@news.newshosting.com>, Michael
> Tuthill <tuthill.DeleteThis@nospamshaw.ca> wrote:
>
>>>> Do you have a problem with how the bayer sensor sometimes creates mush
>>>> when it can't handle a particular situation or due to overenthusiastic
>>>> noise reduction?
>>>
>>> how have you determined that the bayer *sensor* is responsible for the
>>> 'mush' you claim exists, and not something else?
>>
>> Well, whatever it is I see it on bayer shots and not typically on
>> foveon shots.
>
> that may be, but it could be the result of a number of factors, such as
> not focusing properly, shallow depth of field, or due to higher levels
> of sharpening in sigma images.
>
>> I can produce mush easily myself by being "overenthusiastic" with the
>> noise reduction. It's pretty well established that noise reduction can
>> lead to detail loss.
>
> noise reduction affects any image. it's not exclusively bayer that
> suffers from overenthusiastic noise reduction, and it can be adjusted
> or turned off.

Most of the mush is probably due to interpolation errors or the
anti-aliaslng filter I suppose. The interesting question is why I see
mush more frequently on shots captured by bayer sensors?

At any rate, IMO, bayer and foveon sensors both have strengths and
weaknesses and neither is perfect.
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

Mulperi wrote:
> Tamron SP 180mm F/3.5 Di Macro
> Sigma EX 180mm F/3.5 DG Macro
>
> I have found those which are not so different in paper but what do you
> think? Yes I know that the Tamron is cheaper.

I'd generally say "Tamron", but I do know that the full frame v. of the
Sigma 180 macro turns out sharp, high contrast images and I'd have faith
in Sigma on this one a well.

Cheers,
Alan


--
-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
-- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
-- [SI] gallery & rulz: http://www.pbase.com/shootin
-- e-meil: Remove FreeLunch.
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

In article <47c193f8$0$21007$9a6e19ea@news.newshosting.com>, Michael
Tuthill <tuthill RemoveThis @nospamshaw.ca> wrote:

> Well it's my money and it does take pictures that I, and a few other
> people, are happy with. In some situations I prefer it to my D200. My
> only real gripe about it is that it's relatively slow.

That would be *very few* other people, since nobody's buying the POS.
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

In article <47c19576$0$26229$9a6e19ea@news.newshosting.com>, Michael
Tuthill <tuthill.TakeThisOut@nospamshaw.ca> wrote:

> At any rate, IMO, bayer and foveon sensors both have strengths and
> weaknesses and neither is perfect.

Foveon isn't even in the ballpark.
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

"Michael Tuthill" <tuthill RemoveThis @nospamshaw.ca> wrote:
>
> Most of the mush is probably due to interpolation errors or the
> anti-aliaslng filter I suppose.

The "mush" is the right thing to do. When the sensor is not capable of
resolving detail in the image projected on the sensor, the Stigma/Foveon
approach is to render the above-Nyquist-frequency detail as
below-Nyquist-frequency detail. In other words, as aliasing artifacts.

If you look at the test charts on Dpreview, at frequencies above the Nyquist
frequency, the Sigma cameras render a 9-line image as 7 or 5 lines, whereas
the Bayer cameras render them as smooth gray.

Which is to say, the images from the Sigma cameras are completely bogus.

The Sigma fans apparently don't care that the images produced are unrelated
to the subject matter.

> At any rate, IMO, bayer and foveon sensors both have strengths and
> weaknesses and neither is perfect.

The mathematics of discrete sampling tells us that correct imaging within
the capabilities of the sensor requires that detail above the Nyquist
frequency be removed before it hits the sensor. If the Sigma cameras had the
mathematically required low-pass filters, then the images they'd produce
would be indistinguishable from Bayer cameras with the same pixel count.

The funny thing is that Bayer is better at color resolution (relative to
luminance resolution) than the human eye. So in an honest comparison, only a
Foveon-equipped robot could tell the difference.

David J. Littleboy
Tokyo, Japan
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

David J. Littleboy <davidjl.TakeThisOut@gol.com> wrote:

> "Michael Tuthill" <tuthill.TakeThisOut@nospamshaw.ca> wrote:

>> Most of the mush is probably due to interpolation errors or the
>> anti-aliaslng filter I suppose.

> The "mush" is the right thing to do. When the sensor is not capable of
> resolving detail in the image projected on the sensor, the Stigma/Foveon
> approach is to render the above-Nyquist-frequency detail as
> below-Nyquist-frequency detail. In other words, as aliasing artifacts.

> If you look at the test charts on Dpreview, at frequencies above the Nyquist
> frequency, the Sigma cameras render a 9-line image as 7 or 5 lines, whereas
> the Bayer cameras render them as smooth gray.

> Which is to say, the images from the Sigma cameras are completely bogus.

Not at all. We have a situation, for example some very fine striping,
in which the lens is capable of resolving more detail than the
sensor. The stripes can't be accurately shown because because we don't
have enough pixels. Only some kind of approximation to the striping
can be shown. One approach is to refuse to show anything if the detail
can't be accurately shown, i.e. to show stripes to fine to be shown as
a featureless plain surface. The other approach is to show the stripes
as well as can be managed, e.g. if 9 stripes can't be shown, but 5
can, then at least show some in order to show that it is a striped
surface.

Both approaches are bogus in that they are inaccurate. They simply
choose different kinds of bogisity, one bogusly showing a striped
surface as featureless in order to avoid being wrong about the number
of stripes, the other avoiding the false claim of a featureless
surface, but showing fewer larger stripes than were really present.

> The Sigma fans apparently don't care that the images produced are unrelated
> to the subject matter.

Not unrelated, they get the fact of striping right, but the number
wrong, whereas the Bayer anti-aliasts congratulate themselves on not
making false claims about the number of stripes by making the false
claim that it's a featureless surface.

This is rather like the choice a portrait painter has in representing
the hair of the subject, when his pens and brushes are too thick to
accureately represent the fineness of a human hair. Some will simply
show the hair as a wash of colour, while others will suggest the
detailed hairiness by sketching in some hairs which are rather too
thick to be accurate.

Your argument would claim the flat colour washers were accurately
representing the hiar within the scope of their tecnology, whereas the
sketchers of hairiness are doing something bogus which is unrelated to
the subject matter.

I'd like to suggest that you are exaggerating in order to make a bogus
claim of superior accuracy for the anti-aliast approach

>> At any rate, IMO, bayer and foveon sensors both have strengths and
>> weaknesses and neither is perfect.

> The mathematics of discrete sampling tells us that correct imaging within
> the capabilities of the sensor requires that detail above the Nyquist
> frequency be removed before it hits the sensor. If the Sigma cameras had the
> mathematically required low-pass filters, then the images they'd produce
> would be indistinguishable from Bayer cameras with the same pixel count.

Your "requires" in that argument hides an aesthetic assumption with
which many of our finest portrait painters have disagreed.

> The funny thing is that Bayer is better at color resolution (relative to
> luminance resolution) than the human eye. So in an honest comparison, only a
> Foveon-equipped robot could tell the difference.

The best methods of image compression exploit the idiosyncrasies of
the human eye and brain by selectively removing the details we find
hard to notice and preserving those which we easily notice. In effect
the Bayer sensor is that kind of image compression accomplished at the
sensor design level. It works because the human eye and brain, faced
with similar problems of limited processor power faced with large
highly detailed images, already employ a similar kind of compression.

--
Chris Malcolm cam.TakeThisOut@infirmatics.ed.ac.uk DoD #205
IPAB, Informatics, JCMB, King's Buildings, Edinburgh, EH9 3JZ, UK
[http://www.dai.ed.ac.uk/homes/cam/]
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

"David J. Littleboy" <davidjl DeleteThis @gol.com> wrote in message
news:0MqdnZ4-PtrOvF_anZ2dnUVZ_j2dnZ2d@giganews.com...
>
> "Michael Tuthill" <tuthill DeleteThis @nospamshaw.ca> wrote:
>>
>> Most of the mush is probably due to interpolation errors or the
>> anti-aliaslng filter I suppose.
>
> The "mush" is the right thing to do. When the sensor is not capable of
> resolving detail in the image projected on the sensor, the Stigma/Foveon
> approach is to render the above-Nyquist-frequency detail as
> below-Nyquist-frequency detail. In other words, as aliasing artifacts.
>
> If you look at the test charts on Dpreview, at frequencies above the
> Nyquist frequency, the Sigma cameras render a 9-line image as 7 or 5
> lines, whereas the Bayer cameras render them as smooth gray.

Unless my math is wrong, 5 or 7 lines is a lot closer to 9 lines than one
smooth gray one is.
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

"Pete Stavrakoglou" <ntotrr.TakeThisOut@optonline.net> wrote:
> "David J. Littleboy" <davidjl.TakeThisOut@gol.com> wrote:
>> "Michael Tuthill" <tuthill.TakeThisOut@nospamshaw.ca> wrote:
>>>
>>> Most of the mush is probably due to interpolation errors or the
>>> anti-aliaslng filter I suppose.
>>
>> The "mush" is the right thing to do. When the sensor is not capable of
>> resolving detail in the image projected on the sensor, the Stigma/Foveon
>> approach is to render the above-Nyquist-frequency detail as
>> below-Nyquist-frequency detail. In other words, as aliasing artifacts.
>>
>> If you look at the test charts on Dpreview, at frequencies above the
>> Nyquist frequency, the Sigma cameras render a 9-line image as 7 or 5
>> lines, whereas the Bayer cameras render them as smooth gray.
>
> Unless my math is wrong, 5 or 7 lines is a lot closer to 9 lines than one
> smooth gray one is.

Seven randomly incorrect lines is 7 more mistakes than no incorrect lines.
And they really are randomly incorrect: the test pattern has equal-width
lines, but the patterns you get are varying in width and contrast (and this
starts at about 2/3 Nyquist, where correctly antialiased systems are still
correctly resolving the pattern as 9 equal-width lines).

If you want to resolve something, the sensible thing would be to put more
pixels on it, not accept lies.

David J. Littleboy
Tokyo, Japan
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma .. 

David J. Littleboy <davidjl.RemoveThis@gol.com> wrote:

> "Pete Stavrakoglou" <ntotrr.RemoveThis@optonline.net> wrote:
>> "David J. Littleboy" <davidjl.RemoveThis@gol.com> wrote:
>>> "Michael Tuthill" <tuthill.RemoveThis@nospamshaw.ca> wrote:
>>>>
>>>> Most of the mush is probably due to interpolation errors or the
>>>> anti-aliaslng filter I suppose.
>>>
>>> The "mush" is the right thing to do. When the sensor is not capable of
>>> resolving detail in the image projected on the sensor, the Stigma/Foveon
>>> approach is to render the above-Nyquist-frequency detail as
>>> below-Nyquist-frequency detail. In other words, as aliasing artifacts.
>>>
>>> If you look at the test charts on Dpreview, at frequencies above the
>>> Nyquist frequency, the Sigma cameras render a 9-line image as 7 or 5
>>> lines, whereas the Bayer cameras render them as smooth gray.
>>
>> Unless my math is wrong, 5 or 7 lines is a lot closer to 9 lines than one
>> smooth gray one is.

> Seven randomly incorrect lines is 7 more mistakes than no incorrect lines.
> And they really are randomly incorrect: the test pattern has equal-width
> lines, but the patterns you get are varying in width and contrast (and this
> starts at about 2/3 Nyquist, where correctly antialiased systems are still
> correctly resolving the pattern as 9 equal-width lines).

> If you want to resolve something, the sensible thing would be to put more
> pixels on it, not accept lies.

However many pixels you have, there will always be the problem of how
to handle detail as you approach the sampling frequency. Do you dump
all possible aliasing artefacts at the cost of losing some accurate
detail in cases where aliasing artefacts are not obtrusive? Do you go
far all possible detail and live with the aliasing artefacts produced
in some cases? Do you go somewhere between those two extremes?

It's not just a Bayer/Foveon thing. It's a question of compromise
between the evil of losing detail and the evil of aliasing
artefacts. Different camera makers all with Bayer sensors adopt
different compromises.

Let's get away from the special case of regular stripes and consider
wisps of hair in a headshot. Let's suppose there are nine wisps, and
the anti-aliast picks up none of them whereas the aliast high-res
camera picks up seven of them, getting their position and shape right,
but of course misrepresenting their thickness in a "random" fashion.

Which was more accurate?

--
Chris Malcolm cam.RemoveThis@infirmatics.ed.ac.uk DoD #205
IPAB, Informatics, JCMB, King's Buildings, Edinburgh, EH9 3JZ, UK
[http://www.dai.ed.ac.uk/homes/cam/]
 >> Stay informed about: Which one to buy: Tamron SP 180mm F/3.5 Di Macro or Sigma ..