XPro-1 Depth of Field Question

Started 10 months ago | Discussions
Jim Evidon
Contributing MemberPosts: 503Gear list
Like?
XPro-1 Depth of Field Question
10 months ago

In the XPro-1 viewfinder, the depth of field is shown as very narrow. DP review commented on this in it's review and said that is equated to a Circle of Confusion of .005, whereas typical APSC sensors have a CofC of approx .022.

I downloaded into my iPad, a D of F calculator which also shows the XPro-1 having a CofC of .005 but it shows all other DPSC sensor cameras have approximately .022; just as said by DPReview. It may be that the application administrator simply calculated the .005 number from looking at the viewfinder of the camera.

So, here is my question. What is the CofC for the XPro-1, and if it is different from the typical APSC camera, why so? If it is actually .005 instead of .022, that would yield an inconvenient and practically unusable depth of field for a hand held camera. Is there a design flaw or anomaly? I posed the question to Fuji tech support this afternoon and am awaiting an answer

What has been the experience of XPro-1 users?

Any and all comments are welcome.

-- hide signature --

Jim

 Jim Evidon's gear list:Jim Evidon's gear list
Canon PowerShot G15 Fujifilm XF1 Leica M9 Fujifilm X-Pro1 Olympus OM-D E-M5 +2 more
mjl699
Regular MemberPosts: 309Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to Jim Evidon, 10 months ago

Jim Evidon wrote:

In the XPro-1 viewfinder, the depth of field is shown as very narrow. DP review commented on this in it's review and said that is equated to a Circle of Confusion of .005, whereas typical APSC sensors have a CofC of approx .022.

I downloaded into my iPad, a D of F calculator which also shows the XPro-1 having a CofC of .005 but it shows all other DPSC sensor cameras have approximately .022; just as said by DPReview. It may be that the application administrator simply calculated the .005 number from looking at the viewfinder of the camera.

So, here is my question. What is the CofC for the XPro-1, and if it is different from the typical APSC camera, why so? If it is actually .005 instead of .022, that would yield an inconvenient and practically unusable depth of field for a hand held camera. Is there a design flaw or anomaly? I posed the question to Fuji tech support this afternoon and am awaiting an answer

What has been the experience of XPro-1 users?

Any and all comments are welcome.

-- hide signature --

Jim

DOF on the XP1 is the same as with any other APS-C sensor at the same aperture. Fuji's DOF as shown in the viewfinder is narrow, but at 100% my impression has been that it is roughtly accurate. Almost everything is OOF if not smack-bang in the focus plane.

Perhaps it is more noticable with the XP1 since, without an AA filter, less mathematical sharpening is applied which will affect apparent sharpness (and maybe apparent DOF? I stand to be corrected on this)?

J.

-- hide signature --

mjl599

 mjl699's gear list:mjl699's gear list
Canon EOS 400D Panasonic Lumix DMC-GX1 Fujifilm X-Pro1 Canon EF 50mm f/1.4 USM Canon EF 28-135mm f/3.5-5.6 IS USM +5 more
Reply   Reply with quote   Complain
Mescalamba
Contributing MemberPosts: 826
Like?
Re: XPro-1 Depth of Field Question
In reply to Jim Evidon, 10 months ago

It has nothing to do with real DOF, its just calculation done internally to show "accurate" DOF scale in hybrid viewfinder.

Which contrary to what DPreview thinks might be actually right. Thing is that more mpix you have and less AA you got (or in X Pro-1 case, none) every fault is visible sooner. D800E users know that now. Even more of that applies to medium format users.

So no, X Pro 1 has same DOF as anything else in APS-C realm, but diffraction will be visible bit sooner and small errors in AF/MF will be visible more too. Thats of course valid only if you are using some RAW converter that gives sharp results (there are some these days).

Reply   Reply with quote   Complain
Mescalamba
Contributing MemberPosts: 826
Like?
Re: XPro-1 Depth of Field Question
In reply to mjl699, 10 months ago

mjl699 wrote:

Jim Evidon wrote:

In the XPro-1 viewfinder, the depth of field is shown as very narrow. DP review commented on this in it's review and said that is equated to a Circle of Confusion of .005, whereas typical APSC sensors have a CofC of approx .022.

I downloaded into my iPad, a D of F calculator which also shows the XPro-1 having a CofC of .005 but it shows all other DPSC sensor cameras have approximately .022; just as said by DPReview. It may be that the application administrator simply calculated the .005 number from looking at the viewfinder of the camera.

So, here is my question. What is the CofC for the XPro-1, and if it is different from the typical APSC camera, why so? If it is actually .005 instead of .022, that would yield an inconvenient and practically unusable depth of field for a hand held camera. Is there a design flaw or anomaly? I posed the question to Fuji tech support this afternoon and am awaiting an answer

What has been the experience of XPro-1 users?

Any and all comments are welcome.

-- hide signature --

Jim

DOF on the XP1 is the same as with any other APS-C sensor at the same aperture. Fuji's DOF as shown in the viewfinder is narrow, but at 100% my impression has been that it is roughtly accurate. Almost everything is OOF if not smack-bang in the focus plane.

Perhaps it is more noticable with the XP1 since, without an AA filter, less mathematical sharpening is applied which will affect apparent sharpness (and maybe apparent DOF? I stand to be corrected on this)?

J.

-- hide signature --

mjl599

Its exactly like you said, due no AA filter.

Reply   Reply with quote   Complain
Dietr
Junior MemberPosts: 47Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to Jim Evidon, 10 months ago

The Aspherical lenses that Fuji has designed are typically very sharp. What is in focus is generally bang! in focus.  The out of focus areas are therefore a little more noticeable in my opinion.  I have a Nikon 50 1.2  that I have used for years and always thought that it was reasonably sharp and that the DOF was shallow.  But mounted on the XP-1 I have been surprised by how shallow the DOF is, it is even more shallow then I realized using it on Nikon DLSR's.  It is also a VERY sharp lens, I had no idea until I mounted it on this camera.

 Dietr's gear list:Dietr's gear list
Fujifilm X-Pro1
Reply   Reply with quote   Complain
Adrian Tung
Senior MemberPosts: 2,785Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to Jim Evidon, 10 months ago

Despite a lot of theory and mathematics surrounding the subject of DOF, but it is a surprisingly subjective aspect. A lot of DOF calculations have to make assumptions about the enlargement factor (the larger you print, the shallower the apparent DOF) and about the range of "acceptable" sharpness (and what's acceptable for one person may be soft for another).

Even the FAQ on dofmaster.com recommends using values 1 stop more than required:

"How accurate are depth of field calculations?

DOF calculations are remarkably good estimates of the subjective depth of field in photographs.  However, you shouldn't expect to get highly accurate results from the calculations.

To ensure that you get enough depth of field in a photograph, stop down an extra stop.  For example, calculate the depth of field at f/11, but set the f-stop on the lens to f/16."

Therefore in my opinion, having the camera being more conservative than the normal assumptions is usually a good thing.

 Adrian Tung's gear list:Adrian Tung's gear list
Leica D-LUX 4 Canon EOS 10D Canon EOS 5D Mark III Panasonic Lumix DMC-GX7 Canon EF 35mm f/2.0 +5 more
Reply   Reply with quote   Complain
dark13star
Contributing MemberPosts: 745Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to Adrian Tung, 10 months ago

DOFMaster still uses .020 for Fuji, as they do for the other APSC sensor I checked: http://dofmaster.com/digital_coc.html

I have an iPhone app called TrueDOF which allows you to put in your intended print size and it then varies the output. That's the way I learned it a long time ago and I think that might be more concern than the detail on the sensor. Crank the app up for a big print and the depth of field gets really small, as it should, since we can see so much more.

Rich

 dark13star's gear list:dark13star's gear list
Fujifilm X100S Fujifilm X-E1 Nikon D600 Fujifilm X-E2 Fujifilm X-T1 +10 more
Reply   Reply with quote   Complain
Jim Evidon
Contributing MemberPosts: 503Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to dark13star, 10 months ago

From a practical standpoint, i agree with most of your comments except for the ones that refer to the fact that the Fuji has no anti-aliasing filter.  C of C and DOF are the result of mathematical equations that do not take the media (film or sensor) quality into consideration. It would seem that from the equation standpoint alone, the only variables  that apply are format size (film or sensor), focal length and lens opening (F stop). See the formulae below as copied from:

http://www.dofmaster.com/equations.html

It is always true that stopping down improves the depth of field until you hit the circle of confusion (C of C) at which point the sharpness starts to decrease.

Now the formula doesn't take into account real world situations such as the quality of the media, the quality of the glass, etc. The formula is only a starting point from which individual camera/lens combinations are better or worse than the formula predicts.

Therefore, using that standard formula, the XPro-1 should be no different than any other APSC camera. So why does Fuji use .005, a number that is four times more critical than standard? It's good to be conservative in order to assure better results, but I was simply wondering if Fuji is simply being arbitrary and super conservative, or is there something else going on with the camera that caused them to use such a critically small C of C.

Hyperfocal distance:

hyperfocal distance equation

Near distance of acceptable sharpness:

near distance equation

Far distance of acceptable sharpness:

far distance equation

where:

H

is the hyperfocal distance, mm

f

is the lens focal length, mm

s

is the focus distance

Dn

is the near distance for acceptable sharpness

Df

is the far distance for acceptable sharpness

N

is the f-number

c

is the circle of confusion, mm

f-number is calculated by the definition N = 2i/2 , where i = 1, 2, 3,... for f/1.4, f/2, f/2.8,...

-- hide signature --

Jim

 Jim Evidon's gear list:Jim Evidon's gear list
Canon PowerShot G15 Fujifilm XF1 Leica M9 Fujifilm X-Pro1 Olympus OM-D E-M5 +2 more
Reply   Reply with quote   Complain
Jim Evidon
Contributing MemberPosts: 503Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to Jim Evidon, 10 months ago

Sorry, but the formula which i copied and pasted did not reproduce. Please go to the DOFmaster website to see the formulae.

-- hide signature --

Jim

 Jim Evidon's gear list:Jim Evidon's gear list
Canon PowerShot G15 Fujifilm XF1 Leica M9 Fujifilm X-Pro1 Olympus OM-D E-M5 +2 more
Reply   Reply with quote   Complain
nixda
Senior MemberPosts: 1,876Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to Jim Evidon, 10 months ago

Jim Evidon wrote:

From a practical standpoint, i agree with most of your comments except for the ones that refer to the fact that the Fuji has no anti-aliasing filter.  C of C and DOF are the result of mathematical equations that do not take the media (film or sensor) quality into consideration. It would seem that from the equation standpoint alone, the only variables  that apply are format size (film or sensor), focal length and lens opening (F stop). See the formulae below as copied from:

The print medium (or any other type of display) comes into play through the 'acceptable sharpness'. Acceptable sharpness also depends on the individual, so CoC can vary quite a bit between people, applications, and desired end results.

Indeed, Fuji is quite conservative in its DoF calculation. I have suggested in the past to make it possible for the user to set the CoC in the camera. That way, one can adjust the DoF scale to one's liking. I highly doubt that Fuji will provide such a feature, though... In the meantime, one has to resort to tables or cell-phone apps.

 nixda's gear list:nixda's gear list
Fujifilm X-E1 Fujifilm XF 14mm F2.8 R Fujifilm XF 18-55mm F2.8-4 R LM OIS Fujifilm XF 55-200mm F3.5-4.8 R LM OIS Fujifilm XF 27mm F2.8
Reply   Reply with quote   Complain
CraigArnold
Regular MemberPosts: 484Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to Jim Evidon, 10 months ago

No Jim, that's not correct.

You're not reading the site correctly.

Your desired CoC depends on sensor/film size, viewing distance, print size and visual acuity of the viewer.

http://en.wikipedia.org/wiki/Circle_of_confusion

Based on those variables you choose a CoC that suits you. After you have chosen the CoC you can then vary aperture, focal lenth and subject distance to estimate DOF.

There are standard CoC values that were traditionally chosen for 35mm film, enlarged to 8x12, for a viewer 1m from the print of average visual acuity. -> this is the value you get from the standard dropdown list on DOFMaster.

In the world of pixel peeping and large prints, Fuji have chosen a pretty conservative CoC which suggests that people will be printing larger and viewing closer (or looking at 1:1 zooms in photoshop, etc.)

Many of us have long felt that it would be very good for Fuji to allow in the menu settings a range of CoC choices which would then be reflected in the visible DOF scales.

For myself, I find the Fuji scales slightly too conservative, but the standard scales too relaxed. I would prefer to choose a setting in the middle.

-- hide signature --

Blog ------------------------ http://craigspics.net/?tag=blog
X100 Blog ----------------- http://craigspics.net/?cat=6
X100 Quickstart Guide -- http://craigspics.net/?page_id=1345

 CraigArnold's gear list:CraigArnold's gear list
Fujifilm FinePix X100 Sony Alpha 7R Sony FE 35mm F2.8 Sony FE 55mm F1.8
Reply   Reply with quote   Complain
CraigArnold
Regular MemberPosts: 484Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to CraigArnold, 10 months ago

Erratum: I think the standard values are for 25-30cm not 1m as I wrote above. Anyway, you get the idea.

-- hide signature --

Blog ------------------------ http://craigspics.net/?tag=blog
X100 Blog ----------------- http://craigspics.net/?cat=6
X100 Quickstart Guide -- http://craigspics.net/?page_id=1345

 CraigArnold's gear list:CraigArnold's gear list
Fujifilm FinePix X100 Sony Alpha 7R Sony FE 35mm F2.8 Sony FE 55mm F1.8
Reply   Reply with quote   Complain
dark13star
Contributing MemberPosts: 745Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to CraigArnold, 10 months ago

Of course we are assuming that Fuji chose the value of .005 purposefully. It could just be an error, since that value is appropriate for some Finepix cameras. During my career in the tech industry I've seen many such errors get defined as features and the requests for change can end up as feature requests instead of bug fixes because the value wasn't in the requirements document. I'm not saying this is the situation. We don't know, but it is definitely possible.

It could have originated when a programmer started a new code project based on an old one.

Rich

 dark13star's gear list:dark13star's gear list
Fujifilm X100S Fujifilm X-E1 Nikon D600 Fujifilm X-E2 Fujifilm X-T1 +10 more
Reply   Reply with quote   Complain
CraigArnold
Regular MemberPosts: 484Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to dark13star, 10 months ago

dark13star wrote:

Of course we are assuming that Fuji chose the value of .005 purposefully. It could just be an error, since that value is appropriate for some Finepix cameras. During my career in the tech industry I've seen many such errors get defined as features and the requests for change can end up as feature requests instead of bug fixes because the value wasn't in the requirements document. I'm not saying this is the situation. We don't know, but it is definitely possible.

It could have originated when a programmer started a new code project based on an old one.

Rich

I suppose that's possible, but I think fairly unlikely as they've had nearly 3 years and 4 cameras to change it, and they certainly do change the firmware fairly often.

I'm not sure they ever had another camera with DOF scale before the X-series.

Stranger things have happened I suppose.

-- hide signature --

Blog ------------------------ http://craigspics.net/?tag=blog
X100 Blog ----------------- http://craigspics.net/?cat=6
X100 Quickstart Guide -- http://craigspics.net/?page_id=1345

 CraigArnold's gear list:CraigArnold's gear list
Fujifilm FinePix X100 Sony Alpha 7R Sony FE 35mm F2.8 Sony FE 55mm F1.8
Reply   Reply with quote   Complain
dark13star
Contributing MemberPosts: 745Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to CraigArnold, 10 months ago

I'm not sure they ever had another camera with DOF scale before the X-series.

Stranger things have happened I suppose.

They sure have. I agree that it was probably a conscious choice, but I've learned to never just assume so. The process from requirements to shipping is incredibly complex and has so many opportunities for things to go wrong that I've stopped assuming every "feature" is by design

Rich

 dark13star's gear list:dark13star's gear list
Fujifilm X100S Fujifilm X-E1 Nikon D600 Fujifilm X-E2 Fujifilm X-T1 +10 more
Reply   Reply with quote   Complain
Jim Evidon
Contributing MemberPosts: 503Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to CraigArnold, 10 months ago

Craig,

You said:  "...No Jim, that's not correct.

You're not reading the site correctly.

Your desired CoC depends on sensor/film size, viewing distance, print size and visual acuity of the viewer.

http://en.wikipedia.org/wiki/Circle_of_confusion

Based on those variables you choose a CoC that suits you. After you have chosen the CoC you can then vary aperture, focal length and subject distance to estimate DOF...."

*******************

I am aware of that and I suppose I should have included the variable factors as well. I was just simplifying and I should not have done so.

I raised this entire matter only because Fuji did something that no other manufacturer has done; i.e., display the depth of field in the finder. When I saw how narrow the display was, I raised the question, to wit, is there something unique about the Fuji XPro, did Fuji make an error, or are they just being super conservative?

I called Fuji-US this morning and spoke to their Pro camera adviser. First the good news. He has been watching this discussion on DP Review with interest and was glad to talk to me. Now the bad news. He has no answer and directed me to e-mail Fuji technical in Tokyo and assured me that they do pay attention as shown by their often prompt response to customer complaints and/or suggestions by issuing upgrades. So, I did so. I really don't expect a prompt response, but I hope I get some answer some time.

In the meantime, I got an answer from my inquiry to Montana Media, the producers of the iPad app. called " Simple DOF ", a rather clever application. He advised that an update was, or will be made to the application to show the C of C for the Fuji X series as .02 instead of .005. So, either by testing, or operating on faith, he is treating the Fuji X as no different than other APSC format cameras.

When I get my XPro-1 on Monday, I'll assume that the DOF is no different than other APSC cameras and then after shooting, Ill check the results. Not very scientific, but practical.

-- hide signature --

Jim

 Jim Evidon's gear list:Jim Evidon's gear list
Canon PowerShot G15 Fujifilm XF1 Leica M9 Fujifilm X-Pro1 Olympus OM-D E-M5 +2 more
Reply   Reply with quote   Complain
dark13star
Contributing MemberPosts: 745Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to Jim Evidon, 10 months ago

Thanks for that Jim. I've been impressed by Fuji's willingness to update firmware. It will be interesting to see how this turns out.

In my very limited time with the Fuji, I'm not finding that their CoC is any different than other APSC sensors I've used. The issue here is that I don't have a 35mm equivalent lens and that is the focal length where I have practically memorized DOF and would really notice the difference. Maybe after my X100S arrives.

Rich

 dark13star's gear list:dark13star's gear list
Fujifilm X100S Fujifilm X-E1 Nikon D600 Fujifilm X-E2 Fujifilm X-T1 +10 more
Reply   Reply with quote   Complain
Jim Evidon
Contributing MemberPosts: 503Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to dark13star, 10 months ago

That's good to know, Rich.

I'll have a better idea when mine arrives on Monday.

-- hide signature --

Jim

 Jim Evidon's gear list:Jim Evidon's gear list
Canon PowerShot G15 Fujifilm XF1 Leica M9 Fujifilm X-Pro1 Olympus OM-D E-M5 +2 more
Reply   Reply with quote   Complain
CraigArnold
Regular MemberPosts: 484Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to Jim Evidon, 10 months ago

Jim Evidon wrote:

Craig,

You said:  "...No Jim, that's not correct.

You're not reading the site correctly.

Your desired CoC depends on sensor/film size, viewing distance, print size and visual acuity of the viewer.

http://en.wikipedia.org/wiki/Circle_of_confusion

Based on those variables you choose a CoC that suits you. After you have chosen the CoC you can then vary aperture, focal length and subject distance to estimate DOF...."

*******************

I am aware of that and I suppose I should have included the variable factors as well. I was just simplifying and I should not have done so.

I raised this entire matter only because Fuji did something that no other manufacturer has done; i.e., display the depth of field in the finder. When I saw how narrow the display was, I raised the question, to wit, is there something unique about the Fuji XPro, did Fuji make an error, or are they just being super conservative?

I called Fuji-US this morning and spoke to their Pro camera adviser. First the good news. He has been watching this discussion on DP Review with interest and was glad to talk to me. Now the bad news. He has no answer and directed me to e-mail Fuji technical in Tokyo and assured me that they do pay attention as shown by their often prompt response to customer complaints and/or suggestions by issuing upgrades. So, I did so. I really don't expect a prompt response, but I hope I get some answer some time.

In the meantime, I got an answer from my inquiry to Montana Media, the producers of the iPad app. called " Simple DOF ", a rather clever application. He advised that an update was, or will be made to the application to show the C of C for the Fuji X series as .02 instead of .005. So, either by testing, or operating on faith, he is treating the Fuji X as no different than other APSC format cameras.

When I get my XPro-1 on Monday, I'll assume that the DOF is no different than other APSC cameras and then after shooting, Ill check the results. Not very scientific, but practical.

-- hide signature --

Jim

Well, that's all good news; but forgive me for not holding my breath. This has been the case since the X100 was released. The question has certainly been raised with Fuji in Japan before now. There were some threads around since before the X-Pro1 was released where one of the forum members met with some Fuji execs and engineers in Japan and posed it as a question.

Maybe if we keep on asking.

I suspect that they worry about opening a can of worms and confusion. But at the least they might have a menu option to allow you to choose from say a bunch of CoC settings - call them:

  • Small web images (0.03)
  • Traditional (0.02)
  • Relaxed (0.01)
  • Fuji recommended (current 0.005)
  • Pixel Peeping Lunatics (0.0025)

Actually there's even a chance they could do something to allow a trade-off between AF speed and accuracy along the same lines.

It seems to be a shared meme on many forums that although Fuji AF might be slower than some NEX cameras it's more accurate. They might let the user choose between faster and less accurate or slower and more accurate - and include the currently set CoC in the calculations.

After all if working at f16 on a bright day with a wide angle lens, you might as well just go HF and leave it at that.

Oh yes, and one more thing. How about being able to assign the F button in MF mode to focus the camera at the hyperfocal distance according to the currently set Aperture, FL and CoC selection!!! That would be amazing for street/documentary photography.

-- hide signature --

Blog ------------------------ http://craigspics.net/?tag=blog
X100 Blog ----------------- http://craigspics.net/?cat=6
X100 Quickstart Guide -- http://craigspics.net/?page_id=1345

 CraigArnold's gear list:CraigArnold's gear list
Fujifilm FinePix X100 Sony Alpha 7R Sony FE 35mm F2.8 Sony FE 55mm F1.8
Reply   Reply with quote   Complain
Jim Evidon
Contributing MemberPosts: 503Gear list
Like?
Re: XPro-1 Depth of Field Question
In reply to CraigArnold, 10 months ago

Craig,

I like your idea for a DOF menu, but that would be confusing for the user who does not really understand what DOF is all about. All I was really after was whether there is something unique about the Fuji's DOF, or is the indicator in the viewfinder simply the product of an over conservative mind?

Frankly, I have never seen a camera that displays DOF range in the viewfinder. Old time camera users,  me included, enjoyed using the very easy DOF scales on the lens barrel of the better film cameras. Those scales, however, were designed for FF cameras and those scales are unusable on those lenses when fitted to smaller format cameras.

So, I suppose Fuji thought they were doing us a favor by bringing back the DOF scale in the viewfinder,and perhaps with some clarification on their part they may be doing a good thing.

In the meantime, conventional wisdom by current Fiji X users seem to indicate that the DOF and circle of confusion for these cameras and lenses are little different from the competition. So, I'll just ignore the very narrow range shown in the viewfinder and treat it as a decoration, while I continue to resort to external tables and/or calculators.

In the meantime, should any of you have any additional/new facts to share beyond the mostly helpful information already shared, please come forward.

Jim

 Jim Evidon's gear list:Jim Evidon's gear list
Canon PowerShot G15 Fujifilm XF1 Leica M9 Fujifilm X-Pro1 Olympus OM-D E-M5 +2 more
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads