AF speed D700 slower than D3? how come ?

Started Aug 7, 2009 | Discussions
Scott A. Flaherty
Scott A. Flaherty Regular Member • Posts: 398
Re: But what you got may be not be what you think you got.

I agree with you completely. I see little if any difference in AF speed between the D3 and the D700.

-- hide signature --

Scott A. Flaherty
----------------------------------
http://www.clanoflaherty.com/
----------------------------------

'No man, however great, is known to everybody and no man, however solitary, is known to nobody.'
Thomas Moore (1779-1852)

 Scott A. Flaherty's gear list:Scott A. Flaherty's gear list
Nikon D700 Nikon D800 Nikon AF-S Nikkor 14-24mm f/2.8G ED Nikon AF-S Nikkor 24-70mm f/2.8G ED Nikon AF-S Nikkor 70-200mm f/2.8G ED VR II +6 more
ChicagoNikonGuy Regular Member • Posts: 208
Re: AF speed D700 slower than D3? how come ?

JK5700 wrote:

If the D700 was identical to the D3 apart from size, why would anyone buy a D3? There has to be some justification for the price difference.

Nikon pro bodies (D2H, D2X, D3) have always been more responsive than the next level down (D100, D200, D300). The processors are faster, the drive motors more powerfull, the shutter mechanisms faster and more robust, etc.

I have both the D3 and the D300 and notice a significant difference in the speed of aquision, focussing and tracking; especially when shooting birds in flight. But then I expected it because of the price differential.

IMHO, the D700 is a D300 with a larger sensor, not a scaled down D3.
--
http://kennekam.blogspot.com/
http://www.pbase.com/kennekam

According to Nikon the D700 and D3 have the same sensor, the same processor running at the same speed and the same shutter mechanism (as I understand it).

I know that the D3 shutter is spec'd at 300,000 activations and the D700's is spec'd at 150,000. This could be due to differences in quality control and/or librication and/or calibration.

The only optical difference between the two cameras is that the D700 has a 95% view finder.

I don't know if the mirror is the same. This could effect focusing speed but I don't really know.

Additionally the D3 uses the EN-EL4A battery and the D700 without the MB-D10 uses an EN-EL3 battery. The EN-EL4A battery is capable of delivering more current. Theoretically this could allow the focusing motor to go faster. But this is just a guess.

Try doing a controlled test using the same subject and same distance using a D3, a D700 w/o MB-D10 and a D700 with MB-D10 & EN-EL4A.

Also, is the lens that you are "testing" with using the internal focusing motor or does it have it's own. AF-S lenses from Nikon have the focusing motor in the lens.

The D3 is capable of faster shooting speeds then the D700 (9 FPS in FX and 11 FPS in DX for the D3 and 8 FPS in FX for the D700). I don't believe that this means that the D3 focuses faster but it could explain it.

Firmware Release 1.01 for the D700 help increase focusing speed I believe.

The D700 is built as a high end professional camera. The D300 is intended as a high end "pro-sumer" camera.

If the D3 truly performs better than the D700 and you need that better performance and the nearly $2000.00 price difference is also worth it to you, then by the D3. For my money the D700 is a more versatile solution at two thirds the price.

Steve Bingham
Steve Bingham Forum Pro • Posts: 25,818
D700 and D3 use the same processor

The D3 and D700 processors are the same - according to the Nikon designer. This processor is very different from the D300. Here is a quote from Imagimg Resource test. Dpreview said the same thing in their test. Why? Because they were quoting Nikon. Ya think these guys know what they are talking about?

"The Nikon D700's Expeed processor is also the same as the Nikon D3, if not a newer version . . . "

That said, different focal lengths, different lenses, amount of available light, different focus distances, etc will all affect focus speeds. Do you think a casual user could really "test" a difference. Come on. I can't imagine any casual user (or even average pro) that good, that scientific, or has the equipment to do the exotic timing necessary!

So I take Nikon's word for it.

Steve Bingham
http://www.dustylens.com
http://www.ghost-town-photography.com

 Steve Bingham's gear list:Steve Bingham's gear list
Nikon D810 Nikon D7200 Nikon AF Nikkor 85mm f/1.8D Nikon AF-S DX Nikkor 18-140mm F3.5-5.6G ED VR Nikon AF-S DX Nikkor 16-80mm F2.8-4E ED VR +22 more
Steve Bingham
Steve Bingham Forum Pro • Posts: 25,818
Re: test numbers

em_dee_aitch wrote:

More numbers...

http://www.imaging-resource.com/PRODS/D3/D3A.HTM

http://www.imaging-resource.com/PRODS/D700/D700A.HTM

Compare the "shutter lag, prefocused" with the "shutter lag, full auto focus." The prefocused number is not much different, because it's effectively cheating, but this is the number that marketing depts will quote for obvious reason. The functional number that really matters is the "full auto focus" number, because that is your systemic lag. Notice that the D700 takes over twice as long. Again, this is huge. If you think the D700 is as good, you're kidding yourself.

Maybe. Maybe not. Consider the test variables:

1- The first test was done by Michael McNamara (the same guy who picked my digital photos to be the first published by Popular Photography - along with Eric Meola, December 1994, Digital Comes of Age). Michael's test was done in Jan 23, 2008. The second test was done by Philip Ryan - using a DIFFERENT lens. Michael left Pop Photo before the 2nd test.

2- The first test was done at ISO 100. The second test at ISO 160.

3- While the focusing test read .29, .31, .38 seconds for the first three light levels with the D3 the D700 test recorded .35, .36, .36. depending on light levels. We are talking thousandths of a second. This variable is well within different lens samples of the same lens.

In short, you really can't take these times to be exact unless the variables were better controlled. In actual use, I could never detect the difference of 6 thousandths of a second focusing speed, much less 4 thousanths, or even 2 thousandths better in the case of the 3rd light level.

Lastly, Nikon says the focusing systems and processors are identical.

The D3 is a better camera in many respects, but focusing speed would NOT be my deciding factor. Of course we differ - and that's fine. :^)

-- hide signature --

David Hill
http://www.austinweddingphotographer.com
Austin, Texas
Likes Canon for panoramas and the 24/1.4.
Likes Nikon for wedding and everything else.

 Steve Bingham's gear list:Steve Bingham's gear list
Nikon D810 Nikon D7200 Nikon AF Nikkor 85mm f/1.8D Nikon AF-S DX Nikkor 18-140mm F3.5-5.6G ED VR Nikon AF-S DX Nikkor 16-80mm F2.8-4E ED VR +22 more
Marianne Oelund Veteran Member • Posts: 7,360
There isn't a single processor

Steve Bingham wrote:

"The Nikon D700's Expeed processor is also the same as the Nikon D3, if not a newer version . . . "

Expeed refers to the image processing. Other processors handle AF functions.

That said, different focal lengths, different lenses, amount of available light, different focus distances, etc will all affect focus speeds. Do you think a casual user could really "test" a difference. Come on. I can't imagine any casual user (or even average pro) that good, that scientific, or has the equipment to do the exotic timing necessary!

Yes, it is a huge and complex topic. I've already made some fascinating discoveries, taking an engineering approach to investigate the details. I'll be starting a thread soon, to disclose those, so keep an eye out for it.
Hint: Lots of surprises in store for everyone!

Marianne Oelund Veteran Member • Posts: 7,360
Two different subjects

photoforfun wrote:

friend of my own D300/D700, telling me..... AF speed & performance of D700 are clone of D300.......the whole DSLR body also clone of D300, only CMOS is FF.

Your friend is wrong, I owned D300/D3 and D700. D300 is different than D700, the fact that D300 slows enormously down when shooting 14 bit is the best evidence of the difference in processor speed, power battery pack doesn't help to change that.

Image processing is separate from AF processing. The previous poster's comment about AF speed & performance being the same between D300 and D700 is not disproven by Expeed differences. This is why I'm going to twist someone's arm to borrow a D700 soon and put it through some objective tests.

Also, the D300's 14-bit slowdown could be nothing more than sensor-related, i.e., it's not necessarily evidence for slower image processing.

Marianne Oelund Veteran Member • Posts: 7,360
I have the same reservations

NickMJr wrote:

I don't believe the D3 has 2 EXSPEED processors. I've heard that mention on photography forum. Its not even mentioned anywhere of their website or the D3 product broucher. In fact the broucher also shows a picture of the mother board only shows one chip with EXSPEED stamped on it.

Canon put considerable marketing emphasis into the fact that the 1D Mk III has dual Digic processors. Not a peep from Nikon about dual Expeed processors in the D3, and yes, the main board photos I've seen do appear to show a single processor.

Marianne Oelund Veteran Member • Posts: 7,360
You're on the right track

Steve Bingham wrote:

em_dee_aitch wrote:

Again, this is huge. If you think the D700 is as good, you're kidding yourself.

Maybe. Maybe not. Consider the test variables:

1- The first test was done by Michael McNamara (the same guy who picked my digital photos to be the first published by Popular Photography - along with Eric Meola, December 1994, Digital Comes of Age). Michael's test was done in Jan 23, 2008. The second test was done by Philip Ryan - using a DIFFERENT lens. Michael left Pop Photo before the 2nd test.

Correct. Changing the lens (or even the focal length) invalidates the comparison.

2- The first test was done at ISO 100. The second test at ISO 160.

I'm not terribly bothered about this, due to the separation between AF and image processing.

3- While the focusing test read .29, .31, .38 seconds for the first three light levels with the D3 the D700 test recorded .35, .36, .36. depending on light levels. We are talking thousandths of a second. This variable is well within different lens samples of the same lens.

Variability is a fact of life with AF, and studying it accurately requires a statistical approach. You can't simply sit down, take a few measurements and consider the job finished. A proper investigation takes considerable patience, and I absolutely do not consider the differences between numbers quoted above to be proof of anything.

In short, you really can't take these times to be exact unless the variables were better controlled. In actual use, I could never detect the difference of 6 thousandths of a second focusing speed, much less 4 thousanths, or even 2 thousandths better in the case of the 3rd light level.

Well, they're hundredths, not thousandths, but your point is still valid. The numbers can vary wildly from run to run, and this is why we see people making outrageous claims from field use experience, such as "camera A is only 30-60% as fast as camera B" when they may actually be very close.

Gabriele Sartori Veteran Member • Posts: 4,298
100% agree, pointless number

Marianne Oelund wrote:

These extremely limited tests do not begin to scratch the surface of the AF speed topic. The situation can change completely, just by mounting a different lens on the camera, or even by just turning the zoom ring!

The quoted test results apply only to the particular lens used for the test. Attempting to extrapolate them to other lens cases is pointless.

Agree and there is more to it. The power system in the D700 is inherently weaker. This is proven by the 5fps Vs 8fps of the D3. There is a very high probability that the AFS motor is influenced as well. A 1:1 test would need to take into account the extra battery grip. With the grip the FPS becomes 8 and the AFS speed will probably be just like the D3. Without testing this way there is no test, so:
1) Same lens
2) exact same conditions
3) extra grip and battery

Than we can talk.

-- hide signature --

Regards
Gabriele
California, CA

 Gabriele Sartori's gear list:Gabriele Sartori's gear list
Leica D-Lux (Typ 109) Nikon 1 J1 Nikon D800E Panasonic Lumix DMC-GH3 Leica M Typ 240 +2 more
Leonard Shepherd
Leonard Shepherd Forum Pro • Posts: 15,937
Re: D700 and D3 use the same processor

Steve Bingham wrote:

"The Nikon D700's Expeed processor is also the same as the Nikon D3,

Yes - but - Expeed does the noise and signal processing.

We are discussing AF speed - where Nikon say the AF module is the same as in the D3, but do not say if the processor is the same.

The consensus at sites like naturescapes.net seems to be the D3 has a speed advantage.
--
Leonard Shepherd

Practicing and thinking can do more for good photography than buying or consuming.

 Leonard Shepherd's gear list:Leonard Shepherd's gear list
Nikon D810 Nikon D7200 Nikon D500 Nikon D850 Nikon Z7 +24 more
Steve Bingham
Steve Bingham Forum Pro • Posts: 25,818
Re: D700 and D3 use the same processor

Leonard Shepherd wrote:

Steve Bingham wrote:

"The Nikon D700's Expeed processor is also the same as the Nikon D3,

Yes - but - Expeed does the noise and signal processing.

We are discussing AF speed - where Nikon say the AF module is the same as in the D3, but do not say if the processor is the same.

The consensus at sites like naturescapes.net seems to be the D3 has a speed advantage.

In the focusing speed test in Popular Photography of the D3 and the D700 they seem pretty close - although the test is flawed due to less than exacting controls.

1- Michael McNamara did the D3 test in Jan of 2008. Michael was the one who selected my digital images to be the first published in Popular Photography, along with Eric Meola's (Digital Comes of Age, November 1994 issue). Later in Sept 14 of 2008, Philip Ryan tested the D700 as Michael had left a couple of months earlier. He used a different lens. As you know, focus speed is dependent on many variables such as lens, f stop, available light, focal length, focus distance tested, etc. Even a different lens sample can make a small difference. In the first three light settings Pop Photo measured .29, .31, and .38 for the D3. Then with the D700 they measured .35, .36, and .36. These mere thousandths of a second are not really statistically significant considering the rather inconsistent controls.

2- Also consider one camera was tested at ISO 160, the other ISO 100.

I just don't see how the average pro could detect mere thousandths of a second difference in focusing speed - especially considering all the variables such as lens used, focal length used, focusing distance, light available, type of light, subject matter, etc, etc, etc. I think we have a prestige principal working here. If it cost more, it has to be better, faster, etc.

In any case, I have no problem with our disagreeing. :^)

-- hide signature --

Leonard Shepherd

Practicing and thinking can do more for good photography than buying or consuming.

 Steve Bingham's gear list:Steve Bingham's gear list
Nikon D810 Nikon D7200 Nikon AF Nikkor 85mm f/1.8D Nikon AF-S DX Nikkor 18-140mm F3.5-5.6G ED VR Nikon AF-S DX Nikkor 16-80mm F2.8-4E ED VR +22 more
William Nguyen Junior Member • Posts: 42
To improve AF speed, try this

In custom menu a4 (focus tracking with lock on), change to OFF from default setting which is medium (page 287, user's manual.) Before changing the setting, my D700 had to think a little before it locked and clicked. Now it clicked right away.

Marianne Oelund Veteran Member • Posts: 7,360
What AF lock-on does, and does not affect

William Nguyen wrote:

In custom menu a4 (focus tracking with lock on), change to OFF from default setting which is medium (page 287, user's manual.) Before changing the setting, my D700 had to think a little before it locked and clicked. Now it clicked right away.

This setting changes the camera's response delay when you repoint to a new subject, i.e., if focus was already achieved and you are continuing to hold the shutter button half-pressed. Because of imperfections in the AF system, use of shorter lock-on times may also improve tracking of moving subjects, even though this is not the intent of the design.

For initial focus acquisition, that is, starting with the shutter button not depressed at all, this setting does not increase or decrease the camera's time to focus after the shutter button is pressed. If it seems to do so after just a few tries, you need to run about 100 attempts to establish a statistical basis, because delays in focus initiation can be random and unpredictable; the AF system does inconsistent and inexplicable things at times. Such is the way of nonlinear systems and fuzzy logic.

Marianne Oelund Veteran Member • Posts: 7,360
And I would add

Gabriele Sartori wrote:

Agree and there is more to it. The power system in the D700 is inherently weaker. This is proven by the 5fps Vs 8fps of the D3. There is a very high probability that the AFS motor is influenced as well. A 1:1 test would need to take into account the extra battery grip. With the grip the FPS becomes 8 and the AFS speed will probably be just like the D3. Without testing this way there is no test, so:
1) Same lens
2) exact same conditions
3) extra grip and battery

I would add:
4) large number of repetitions of each measurement
The AF system is fickle and inconsistent, so a statistical approach is required.

P.S.: You can strike #3. I've already proven that battery voltage has no effect on AF drive speed. This applies to both AF-S lenses, and the camera's built-in AF motor.

em_dee_aitch Veteran Member • Posts: 3,675
Re: And I would add

I will be very interested to see the results of Marianne's research.

As to tests being flawed, that's great. I've been assuming all along they're flawed, but I've also assumed that a publication like Pop Photo would at least take a statistical approach and average many attempts and not just do one and call it a day.

What is interesting to me is that pretty much all tests, flawed or not, seem to align with the opinions of pro shooters from various disciplines who think the D3 feels faster than the D700.

Steve- those times you were quoting were hundredths of a second, not thousands, btw. And a tenth of a second is easy to feel, so 6 hundredths, being over half a tenth, is also pretty easy to feel at the trigger. If you don't think a human can easily learn to feel a hundredth of a second, ask a musician or a baseball player.

-- hide signature --

David Hill
http://www.austinweddingphotographer.com
Austin, Texas
Likes Canon for panoramas and the 24/1.4.
Likes Nikon for wedding and everything else.

em_dee_aitch Veteran Member • Posts: 3,675
Re: AF speed D700 slower than D3? how come ?

ChicagoNikonGuy wrote:

According to Nikon the D700 and D3 have the same sensor, the same processor running at the same speed and the same shutter mechanism (as I understand it).

The quote which I saw posted earlier in this thread said the same AF sensor and CMOS. And if they have the same main "processor," what about additional stuff in the D3? Haven't Ken Rockwell (I know, credibility questions :)) and Thom and others gone on about extra chips that were in the D3 but not the D3x? Are these extra chips in the D700? I would think not if they're a cost issue. Are they helping the D3 perform better? And where is the quote on shutter being the same? I haven't seen that one. The fact that the rating is different and they sound different could mean they are different. If it were really the same, why not rate the D700 at 300,000 as well? It's not like that's going to be the one detail that pushes someone over the edge between the two.

I know that the D3 shutter is spec'd at 300,000 activations and the D700's is spec'd at 150,000. This could be due to differences in quality control and/or librication and/or calibration.

The only optical difference between the two cameras is that the D700 has a 95% view finder.

I don't know if the mirror is the same. This could effect focusing speed but I don't really know.

This point is already so well speculated upon that I wish you would at least claim to know, just to make your paragraph more interesting.

Additionally the D3 uses the EN-EL4A battery and the D700 without the MB-D10 uses an EN-EL3 battery. The EN-EL4A battery is capable of delivering more current. Theoretically this could allow the focusing motor to go faster. But this is just a guess.

Try doing a controlled test using the same subject and same distance using a D3, a D700 w/o MB-D10 and a D700 with MB-D10 & EN-EL4A.

Ya, we've all already covered that.

Also, is the lens that you are "testing" with using the internal focusing motor or does it have it's own. AF-S lenses from Nikon have the focusing motor in the lens.

I think you're really wrong here. Even with AF-S, what you're testing is the way the camera drives the lens and the firmware in the camera, and how that particular camera processes the AF data and how fast among other things. If you assume that all cameras have the same response curves and firmware, that would be quite a gigantic assumption indeed.

The D700 is built as a high end professional camera. The D300 is intended as a high end "pro-sumer" camera.

Well, to be technically correct, the D700 and D300 have pretty much identical build quality feature sets and even use the same accessory grip. The D700 is elevated only by its FX sensor and improved joy pad. The D700 certainly is not a "high end professional" model but rather simultaneously a "low end professional" or "highest end amateur" model, like the 5D and 5Dm2 (except having better overall feature sets than them, with exception of megapixels in latter case obviously).

If the D3 truly performs better than the D700 and you need that better performance and the nearly $2000.00 price difference is also worth it to you, then by the D3. For my money the D700 is a more versatile solution at two thirds the price.

Once you add in the MBD10 grip and a real battery, the price difference falls to something like 1200-1300, depending on where you shop, I think. You also have to remember that for some people the ability to not have a silly pop-up flash in the way all the time is worth at least a couple hundred bucks in itself. The D3 is still well worth the money if you like a serious, best-in-class camera.

-- hide signature --

David Hill
http://www.austinweddingphotographer.com
Austin, Texas
Likes Canon for panoramas and the 24/1.4.
Likes Nikon for wedding and everything else.

em_dee_aitch Veteran Member • Posts: 3,675
Re: AF speed D700 slower than D3? how come ?

p.s. -- adding to that last thought, the enhanced balance and solidity of a D3 type body, as compared to the non-solidity of using an accessory grip, is also worth hundreds of dollars in enhanced experience. If you use brackets and other types of mounts, the cameras with grips will flex at the grip.

-- hide signature --

David Hill
http://www.austinweddingphotographer.com
Austin, Texas
Likes Canon for panoramas and the 24/1.4.
Likes Nikon for wedding and everything else.

K3nKen Regular Member • Posts: 252
Re: AF speed D700 slower than D3? how come ?

James,
Do you have both cameras? Could you do a test.
Put a lens cap on and switch the AF to CAF.
Also, please turn off the AF assist illumination for the 700.

See if the lens searches from near to infinity and back faster on the D3. I suspect it might because the D3 is operating on a higher voltage.

Please let us know. I suspect there might be a difference but imperceptible for most photogs.

James Kei wrote:

I owned/used D3 since April 2008, now just brought D700 as lightweight version as for travel....

I read a lot about D3/D700 spec & comparsion review, so pitty much conclude they are same in 99%.

Start using D700 at indoor, normal / low lighting...

I use 28-70 afs lens, the same lens I have been using it with D3 for yr and half.

1st feeling;
woo...... this D700 body AF system seems slower response than D3.
I use AF 51 points dynamic mode on both body.

BUT the spec said same ; ? really ?

• 51 focus points (15 cross-type sensors)
• Multi-CAM 3500FX
• AF working range: -1 to +19 EV (ISO 100, normal temperature)
• Contrast Detect in Live View (Tripod) mode

1 more diss-like about D700;

My Capture NX 1.3.5 come from D3 box package does not read D700 raw, dam....

does that mean I have to spend extra $$ for Capture NX 2 ? but the spec said, they are the same image senior of 12 MP FX, why need difference software to read RAW .........

Does anyone has same feeling when you switching D3/D700 in use ?

OR am I over sensertive ?

many thanks

Marianne Oelund Veteran Member • Posts: 7,360
Requires more than one test

K3nKen wrote:

James,
Do you have both cameras? Could you do a test.
Put a lens cap on and switch the AF to CAF.
Also, please turn off the AF assist illumination for the 700.

See if the lens searches from near to infinity and back faster on the D3. I suspect it might because the D3 is operating on a higher voltage.

Battery voltage does not determine AF drive speed.

Please let us know. I suspect there might be a difference but imperceptible for most photogs.

A single test will not address the overall issue, because results will depend on the lens used, as well as the zoom setting. Differences can be much more than merely perceptible - and they are not always in favor of the D3, either!

K3nKen Regular Member • Posts: 252
Re: Requires more than one test

Marianne Oelund wrote:

K3nKen wrote:

James,
Do you have both cameras? Could you do a test.
Put a lens cap on and switch the AF to CAF.
Also, please turn off the AF assist illumination for the 700.

See if the lens searches from near to infinity and back faster on the D3. I suspect it might because the D3 is operating on a higher voltage.

Battery voltage does not determine AF drive speed.

Please let us know. I suspect there might be a difference but imperceptible for most photogs.

A single test will not address the overall issue, because results will depend on the lens used, as well as the zoom setting. Differences can be much more than merely perceptible - and they are not always in favor of the D3, either!

I believe there's not much difference too. The Nikon F5 and F100 used the same CAM 1300 autofocus as well and most photogs have the opinion that they're identical in performance.

Keyboard shortcuts:
FForum MMy threads