Alan Roberts BBC report on the AF101

[Gary Nattrass]

Link to the report is here: http://thebrownings.name/WHP034/pdf/...onic_AF101.pdf

[David Parks]

Interesting. far short of a ringing endorsement. Oh well. Good thing I'm not producing a project for BBC!!

So far so good for me. We're using the AF100 out here NASA/JSC on the Engineering Science Contract with really good results for shooting in low light. (Shooting in flight similators last week which is like shooting in a dark movie theatre). So there you go. Nothing is perfect. :)

[Allan Black]

Thanks Gary, seems like the AF100/1 might have been somewhat rushed to market with development in some areas but not in others. Witness a software update before some first customers got their machines.

Alan Roberts might have rushed his report too, pages 12 and 13 are not finished.

Cheers.

[Andy Tejral]

Its strange to me that it says they were production models but didn't have serial numbers... Were they really pre-production models? Seems like a huge oversight, but...

[Olof Ekbergh]

This kind of confirms my feelings about the AF100.

I still think it is a great camera and I am now using it in production. But I found that my EX3 and EX1R are much sharper, I spent a lot of time trying to make the AF100 look as sharp, and I never really could (lots of different lenses and lots of trying settings). The less than 800 (680) vertical resolution seems consistent with my findings.

There are also some really strange things that happen with banding, but I have not had a problem with aliasing at all.

I find that I can create beautiful images with the AF, just always a little soft. This is not a bad thing, you just have to be ready for it. The DSLRs are way to (artificially) sharp for my taste, this can be dialed down and fixed in post.

At this time I think in this price class the AF100 is the best you can do for image quality and SDOF, with reasonable working ergonomics and features. It is a big step above the DSLRs. And I bet that Pany will improve some of the codec problems with future FW updates.

I don't think we have to be evangelical about this camera just because we bought it, but we do have to work around the short comings and appreciate the innovative design features. I think it is the first in a whole new class of camera and a very good one at that.

Remember the BBC compares this with cameras costing many times more. This is really a very affordable camera. Personally I will always rather use the AF100 than an EXcam and a SDOF adapter, and in most cases a DSLR, I have both a 5DmII and a 7D. Most of all I love being able to use my old, and new Canon lenses with the coming Birger adapter (really great for remote focus pulling on jibs and Stedicam). And the Voightlander 25mm f.95 is almost worth getting the AF100 by itself, a wonderful lens (all I want is to now de-click its iris).

The bottom line is I like this camera a lot for many reasons, it has some problems but so does every camera. I am glad I have it and I will use it until something better comes along that makes sense for me, but for now this is my SDOF camera.

[John Mercer]

I don't think there should be any surprises here. You really do get what you pay for. These cameras are 95% aimed at the prosummer market (which is huge and lucrative) and the current trends that drive it. But I don't think anyone should get hung up on resolution, it looks perfectly sharp to me, too sharp infact. I think someone needs to come up with some good settings for it more urgently. As Alan pointed out his filmlike settings are only a starting point. My main concern with it is a quite video look to what I've seen.

In any case it is an interesting product that offers some of the advantages of the DSLRs with improvements and proper video ergonomics. People will start making great things with it I'm sure and that's all that matters. Alan is critically testing it within a severe broadcast (BBC) context and against the standards of much more expensive equipment. He was quite positive about the colour and motion aspects of the camera though, however he did conclude it definitely does not skip lines but there was little or no OLPF at work, which maybe answers some of the conundrums questioned over this model. (This in theory should be easy and cheap to rectify in future versions though.)

But again I would say not to worry, go out and make good stuff with it (and all cameras). That's the only way to test them in the end.

[David Heath]

Originally Posted by Olof Ekbergh

I found that my EX3 and EX1R are much sharper, I spent a lot of time trying to make the AF100 look as sharp, and I never really could ……….. The less than 800 (680) vertical resolution seems consistent with my findings.

I’m afraid even that understates the issue.

I reported tests I’d seen on the AF101 several months ago which gave figures for basic resolution and aliasing very close to what Alan is now confirming (to within 2% accuracy).

Alans report makes clear that the alias circles are centred on 2350 (hor) and 1322 (vert). Those must be twice the Nyquist frequencies. (By definition.) This shows the effective system resolution is symmetrical, and must be based on a figure half that – 1175x661. Anything greater than 661 TVlines/ph must therefore be aliasing.

In terms of actual resolution, it would be wrong to claim even 661 TVl/ph – the actual figure must be somewhat under the Nyquist limit. About 10% less is a generally accepted figure, so more like 610, maybe up to 630 – certainly not as high as 680. (In practice, it tails off, so a precise figure isn’t very meaningful.)

Effectively, this means that in resolution terms it’s a long way short of what would be expected for a 720p camera – let alone for 1080. And low fundamental resolution tends to mean detail enhancement levels have to be set higher. That tends to mean a more “video” - edgy - look to images.

Originally Posted by Olof Ekbergh

At this time I think in this price class the AF100 is the best you can do for image quality and SDOF, with reasonable working ergonomics and features. It is a big step above the DSLRs.

Yes, I agree. But “at this time” is likely to mean a very short period. We’re expecting to see the large format NXCAM on March 23rd, and initial reports are that it will use the (much better) F3 sensor, yet still be around the same price as the AF101. The question is not really “is the 101 good or bad?” – that’s highly subjective – but more a question of “will the new NXCAM offer me more for about the same money?”

Originally Posted by Olof Ekbergh

Remember the BBC compares this with cameras costing many times more. This is really a very affordable camera.

Originally Posted by John Mercer

Alan is critically testing it within a severe broadcast (BBC) context and against the standards of much more expensive equipment.

The BBC may test a large variety of cameras, but this is likely to have been head to head with other cameras in similar price brackets. Obvious likely models are the EXs from Sony, and the Canon XF305 – all of which the BBC use extensively. (And are fully approved for BBC use, at with a nanoFlash in the case of the EX.) By the time you take lenses into account for the 101, and something like a nanoFlash for the 101 and the EX cameras, they are all around the same price.

Yet the EX and the XF305 give a resolution to match the 1080 system, the AF101 isn’t that good even by 720 standards.

Since part of the current broadcast expectation is decent 1080 resolution, I don’t ever expect to see the AF101 get full BBC/EBU approval (even with external recorder) in the way the XF305 and EX with nanoFlash have. The big question will have to be how the new NXCAM fares.

[David Heath]

Hmmm. Before you point it out Olof, I now note that Alan himself claims "up to 1210x680, which is not good, given the strength of the aliases" - I assume that is where you derived the 680 figure from? But there seem to be inconsistencies in what he says.

From the report, the most significant point is :

The horizontal and vertical aliases have a centre frequency of 1.224 times the system frequencies.

which is verifibale from the zone plate images. The alias circles are centred on 1.224x1920 and 1.224x1080 (2350 and 1322). But he then goes on to say:

This implies that there is an underlying resolution 1.224 times that of 1920x1080, or 2350x1322

And this cannot be true. Alias circle centres show *TWICE* the underlying resolution, not the underlying resolution itself. The underlying resolution must therefore be 1175x661, so in no way is 1210x680 possible without giving rise to aliasing.

I also note he says:

However, to achieve a smaller depth of field in this camera, relative to, say, a ½” camera, then the lens must be opened by at least 1.5 stops; using an F/2.8 lens on this camera, wide open, will give the same depth of field as on a ½” camera with a lens opened to F/1.6.

That's not true. The area dimensions of the AF101 sensor are 8x that of a 1/2" chipped camera, which means a 3 stop difference, not 1.5.

Hence, compared to a 1/2" camera at f1.6, the same dof will be obtained with an AF101 at about f4.8 - not f2.8.

[John Mercer]

Originally Posted by David Heath

The BBC may test a large variety of cameras, but this is likely to have been head to head with other cameras in similar price brackets. Obvious likely models are the EXs from Sony, and the Canon XF305 – all of which the BBC use extensively. (And are fully approved for BBC use, at with a nanoFlash in the case of the EX.) By the time you take lenses into account for the 101, and something like a nanoFlash for the 101 and the EX cameras, they are all around the same price.

This I cannot agree with. I quote Alan himself about this camera: "I suspect the official position will be that it's an HD camera, but not top grade. So, it'll get approval for use in film-style shoots but not as the main camera. I guess it's footage will be classed in the 'non-HD' quota, including archive and SD footage. But, I don't make the rules or the decisions, I only measure and report (and get paid for so doing)."

The BBC does not use EXs and X305s extensively. They are approved for a certain kind of use but the BBC would much prefer you to use 2/3" broadcast quality cameras for major programming. His job is to test and measure cameras against that standard and there is never usually a question or a need on the majority of mainstream programming to have to use anything less. Whilst I agree he can knowingly compare similar cameras from this category the requirement will always be to determine whether the BBC can accept a particuler model into the fold set against the highest standards they would ideally prefer. They are realistic to the likely increased use of such cameras but not overly enthusiastic either.

[David Heath]

John, I think the most revealing bit of what you quote from Alan is the phrase "I guess it's footage will be classed in the 'non-HD' quota, including archive and SD footage."

In other words, he's expecting them to look at it in the same light as an SD camera. Use it if you must, but...... That's tantamount to saying it will not get full approval for unrestricted HD use - as such as the XF305 has. (Let alone more expensive cameras.)

Equally, what about the new large format NXCAM? If - *IF* - it offers much better quality than the AF101 at about the same price, then why or earth would the BBC (or anyone else) use an AF101 in preference to that? That will be especially true if the NXCAM DOES get full broadcast approval, at least if used with an external coder.

[John Mercer]

In other words, he's expecting them to look at it in the same light as an SD camera. Use it if you must, but...... That's tantamount to saying it will not get full approval for unrestricted HD use - as such as the XF305 has. (Let alone more expensive cameras.)

Precisely David. I would just like to also clarify, because I've done a lot of work for the BBC, that when they say unrestricted use that does not necessarily mean they are happy for main stream programming all to be shot on these cameras (like the X305) merely that HD from them can usually be classed as true HD (unlike say the Z1). Again the vast majority of programmes are shot on full blown 2/3" broadcast cameras.

[David Heath]

Originally Posted by John Mercer

......when they say unrestricted use that does not necessarily mean they are happy for main stream programming all to be shot on these cameras (like the X305) merely that HD from them can usually be classed as true HD

No argument with that at all - but if you're saying that they will effectively tolerate for some programming a lower end (albeit approved) camera like the XF305, even if not 100% happy about it, where does that leave the AF101?

It seems extremely unlikely the AF101 will even get the level of approval of the XF305, so surely they can be expected to be even less likely to agree to it's use than the XF305?

[John Mercer]

The XF305 has a good codec (50mbps 4:2:2) and proved surprisingly high quality for it's category in Alan Roberts tests. It is capable of producing footage in the right hands that would cut into broadcast material without too much issue particularly for the encoding to air. Still you are going to normally use 2/3" broadcast cameras in day to day main stream progrmming where there is a pro cameraman.

From what I can gather from his test results for the AF101 there is an issue with the way the camera downconverts its megapixel cmos to 1920 x 1080 that is less than perfect. It doesn't pixel skip as was argued here but its OLPF is either less than perfect or non-existent and so the useful resolution is curtailed because of resultant alias artifacts. Combined with a 4:2:0 AVCHD codec of 21mbps (can be fixed with Nanoflash say but still leaves the 1st problem) it doesn't meet the required standards. Remember that the encoding concatenations through post to air in either Mpeg2 or Mpg4 (BBC's current HD) dictates here not how it looks on Vimeo. But also as he says he doesn't write the rules.

I return to what I said earlier, it does not surprise me as you get what you pay for (even with the XF305 they are not giving it you in a shoulder mounted interchangeable lens package). But this will not stop people using it, even for broadcast, and even at the BBC, the dam has opened and I expect to see many fine productions shot with these cams so I don't think people should worry too much.

[Brian Drysdale]

I thought a key element of the XF 305 was the combination with the built in lens, so it was an overall package that created a camera that the BBC are happy with.

[John Mercer]

I thought a key element of the XF 305 was the combination with the built in lens, so it was an overall package that created a camera that the BBC are happy with.

That's what I meant Brian by 'surprisingly good quality'. When used by DV Directors and APs, or when used as 2nd multiple cameras, say on something like Top Gear, or less accessible current affairs and news, it produces footage that is considered true HD (unlike the Z1 which is still widely used). But where you shoot drama or studio, or high end documentary then the main camera is and will most of the time be a 2/3" broadcast camera. Something like the XF305 (or any other fixed lens camera) can never be a realistic substitute for these.