Potential future of F3xx

[David Heath]

Originally Posted by Alister Chapman

Lots of people seem too hung up on just the bit rate. Simply saying that you'd rather have 5Mbps over 35Mbps doesn't guarantee better picture quality.

Yes, but the difference is that with a 50Mbs mode, you can walk up to any broadcaster and know that your equipment meets a current EBU tested and approved spec, at least as far as the compression goes.

Also, the whole subject should be seen far less in terms of first generation picture quality than how the images will stand up to the final broadcast chain, and especially the final transmission compression. There's lots of evidence that differences invisible to eye on a first generation can influence the final result.

But most importantly, enabling the 50Mbs mode doesn't involve any extra engineering effort - certainly for an SxS camera - and shouldn't be any more expensive to manufacture. The same can't be said of a difference between 1/2" and 2/3" chips. (Desirable though 2/3" wouls indeed be.)

I suspect that the only reason the EX1/3 are 35Mbs is that to have enabled the 50Mbs mode would have just given them too many advantages over the 330/350. If we're now talking about a 3xx replacement, it just doesn't make any sense to NOT make them 50Mbs.

As far as colour space goes, then a lot depends on format. 4.2.0 may indeed be fine for progressive material, but for interlace there are a lot of advantages to going to a system with equal vertical lum/chrom resolution. Since most 1080p/25 material is likely to transmitted 1080psf/25, that may well be an added advantage to 4.2.2 even for that.

[Simon Wyndham]

There's lots of evidence that differences invisible to eye on a first generation can influence the final result.

That is very true. However do not forget that Discovery stress tested the 35Mb/s codec right through the broadcast chain to determine how well suited it was for programming when the 350 first arrived on the scene, and they approved it for unrestricted "Silver" level production. As Alister mentioned the actual compression ratio is no different to the 50Mb/s, although the latter is CBR rather than VBR.

Regarding the colour space, it was interesting to see Adam Wilts test of the EX when he compared a 4:2:2 SDI capture with the 4:2:0 recorded one in progressive mode. There was no visible difference. ProVideo Coalition.com: Camera Log by Adam Wilt | Founder | Pro Cameras, HDV Camera, HD Camera, Sony, Panasonic, JVC, RED, Video Camera Reviews But as David pointed out in interlaced modes it is a different story altogether.

I think that my line is that with production budgets being cut all the time some of these lower end cameras really should be made acceptable. As Discovery have proved the 35Mb/s codec holds up very well through the broadcast chain for certain levels of programming. Eventually high def will be used for hum drum programmes, and in fact already is. Look at "A Place In The Sun" on Channel 4! In that programme they appear to be using low spec HDV cameras much of the time. Given that 25Mb/s DV on 2/3" cameras has been fine until now for news and current affairs, and even programmes such as "Coast" on the BBC I do not see why they couldn't use a low cost 35Mb/s 2/3" HD camera in the same sorts of applications.

One thing that isn't clear is what the EBU's testing was based upon? For example were they testing it with only higher end programming in mind? Mind you, will anyone listen to the EBU recommendations anyway? After all this is the same organisation that has decided upon 720p as their main HD recommendation.

[Alister Chapman]

Originally Posted by David Heath

Also, the whole subject should be seen far less in terms of first generation picture quality than how the images will stand up to the final broadcast chain, and especially the final transmission compression. There's lots of evidence that differences invisible to eye on a first generation can influence the final result.

Agreed, but as both 35Mbps and 50Mbps XDCAM have the same amount of compression the generation losses will be the same.

I'm not going to argue the fact that if someone specifies a particular format then that is what you should use, but you are more than likely going to be delivering on HDCAM SR or HDCAM and it is my opinion that in most cases you would not be able to tell from the technical quality of the master whether you used a 700, EX3 or 750 HDCAM.

The BBC, Discovery, Nat Geo and many others have used lots of my 4:2:0 XDCAM HD and XDCAM EX footage in all kinds of shows. They have never complained about the quality. Most of the time the only thing they care about is how I physically deliver it to them. Often this is HDCAM which is a shame as I can see the resolution drop when I dub EX footage to HDCAM. The more switched on facilities companies are requesting delivery in the native format.

Will Sony give us a 4:2:2 F35* replacement? I doubt it. If you are producing programmes for the BBC on a regular basis then I don't really see why you wouldn't get a PDW-700.

For Discovery the EX3 and F35* is perfectly acceptable.

[David Heath]

Originally Posted by Simon Wyndham

Regarding the colour space, it was interesting to see Adam Wilts test of the EX when he compared a 4:2:2 SDI capture with the 4:2:0 recorded one in progressive mode. There was no visible difference.

"No visible difference" is not quite what he said - "All 4:2:0 does in progressive mode is equalize the color subsampling on H & V directions..... It’s not much worse looking than 4:2:2." But I agree that for progressive 4:2:0 shouldn't be considered too much of a problem. I'd rather have 4:2:0 full raster than subsampled 4:2:2, for example!

But it still leaves the issue of interlaced material, and 1080i is likely to be important for a while to come.

I think that my line is that with production budgets being cut all the time some of these lower end cameras really should be made acceptable. As Discovery have proved the 35Mb/s codec holds up very well through the broadcast chain for certain levels of programming.

I don't want anything I'm saying to be taken as "35Mbs is rubbish and shouldn't be even considered for broadcast" but rather as ways in which the "potential future" could (should?)develop, to refer to the thread title. And if 50Mbs can displace 35Mbs *for no extra cost* in new equipment, surely that can only be a good thing? I don't just mean in F3xx replacements, but in EX1/3 MkIIs. There's no reason a 50Mbs EX3 need cost any more than a 35Mbs EX3.

One thing that isn't clear is what the EBU's testing was based upon? For example were they testing it with only higher end programming in mind? Mind you, will anyone listen to the EBU recommendations anyway? After all this is the same organisation that has decided upon 720p as their main HD recommendation.

I think it was to try to bring some science to the debate. At the moment it seems that many parties have their goalposts set all over the place, often for a variety of vested interests. I think what they're trying to do is to establish where goalposts should be - eventually.

There is one huge change coming that needs to be recognised. Whereas in the past higher quality recording methods meant higher costs, that is no longer true with solid state recording. (Except in so far as higher bitrates mean more storage capacity.) If 50Mbs can be coded and recorded with the same hardware costs and 35Mbs, the question becomes not "why to do it?" but "why NOT to do it"!?

[Mike Marriage]

Originally Posted by David Heath

If 50Mbs can be coded and recorded with the same hardware costs and 35Mbs, the question becomes not "why to do it?" but "why NOT to do it"!?

I'm in agreement on that David, but if people want a lower cost 2/3" HD camera, Sony are going to want to protect the PDW700 and HDCAM range. However, they may be happy to make a 2/3" SxS 35Mbit 4:2:0 camera. I suppose a DVCAM priced PDW700 may be possible with SxS instead of XDCAM but I'd guess they'd want to limit the product further than that to protect the higher end market - but then again, I hope not! Personally I don't think I would consider a 1/2" camera for my main camera, mainly due to lens choice/ compatibility & DOF control.

Many companies have deliberately crippled a lower cost product. I heard that one of the early IBM's could be bought in "home" spec or "office/pro" spec for 5 times the price. Both were identical apart from the settings of hidden dip switches!

[Alister Chapman]

I think the 700 has hit the limit of what can be recorded on the Professional Disc system which may be why there is currently no Slo-Mo. Certainly read speeds seem to be max'd out at around 90Mbps and you need some overhead when recording to allow for TOC updates, proxies, knocks and bumps. I would guess that the next XDCAM HD 4:2:2 camera will record to SxS to allow for over-cranking without having to restrict the data rate. So a low cost full size SxS based camera could impinge on any top end XDCAM HD developments.

This is an interesting discussion and I too would love to see a F350 replacement that was 4:2:2, 50Mbps with 2/3" sensors. But I just don't see it happening, not because it can't be done but to protect the high end. Maybe in a couple of years time when Scarlet come to market they will have no choice, but at the moment, no.

If I had to give up one of the above features it would be the 4:2:2 @ 50Mbps. We all know how good the EX3 looks. A full size 2/3" 35Mbps 4:2:0 camera with HDSDi would be a really useful piece of kit. If you really need 50Mbps (or more) you bolt on a Nano-Flash, problem solved.

Just a side note: F35* family cameras in HQ mode have a compression ratio of 16:1.
EX3 HQ is 21:1
PDW700 is 20:1

Yet the F35* are IMHO the worst looking of the lot, shame because based on tose numbers it should do much better over multiple generations than any of the others. However when you take into account noise and scaling issues the story can be quite different.

In terms of the same product being sold at different prices just consider some of the cameras we are discussing. The F355 is 50% more than the F335 but the only difference is the firmware.

[Eugene Kosarovich]

Exactly, and the price difference versus features I need is why I got a F335 instead of a F355.

It's very interesting hearing all the viewpoints here. For myself, as with any of us, my wishes for the next F3xx are based on my needs for it. I think I'm coming toward this from the opposite of most of you.

The vast majority of you are very high end and broadcast. For me, the F335 is the highest end camera I've ever used in my life. Myself, and a friend of mine in this area, both bought our F335s early in 2008 to replace our DSR-300As. Even though my friend works for our local NBC affiliate, the F335 is a higher end camera than his station uses in his normal work, their local production is still SD.

For both of us, the F335 gives us each the highest production values in our respective fields in our local area. My friend in wedding and bar/bat mitzvah event work and myself in performing arts and corporate work.

As always, it all comes down to who the particular camera lines are really marketed for.

Though we certainly understand the advantages of 2/3" imagers, for my friend and I, all our current lens investments are in the 1/2" world. I just checked and apparently there are now 2/3" HD lenses down in the "affordable" range, but for those that have already bought into 1/2" HD lenses, well... Also, there's the issue of our lenses being useful in potential future EX cameras, which is a nice capability to keep.

Why didn't we go with the EX line now?

We needed the record time of dual-layer Professional Discs. I was so happy when Sony surprised us with the F335 after telling me at NAB 2007 there would be a F355, but no F335. This same reason is why a 50Mb/s data rate isn't important to me, for the work that I do, I can't afford to give up all that recording time per disc.

We also needed the current affordability and direct archive capability of the Professional Discs. For our uses, it is not unusual to have dozens of Professional Discs around waiting on editing, we simply couldn't do that with SxS media cost. And the continuing lack of a simple and reliable archive method for the SxS footage via the U1 drive to Professional Disc is a major weakness for SxS for any use by us. SxS just doesn't have the workflow for our needs.

The other important issues were wanting a real shoulder mount camcorder and more than two audio channels. The EX line currently just doesn't fit these needs.

The only alternative would have been going with the S270, but we wanted to step up in quality.

So, we went with the F335, and we are happy we did, our only wishes are better low light capability and the full raster quality of the EX line imagers.

Ah well, just over three months and we'll see what's announced. Meanwhile the F335 is making me money.

[David Heath]

Originally Posted by Eugene Kosarovich

Why didn't we go with the EX line now?

We needed the record time of dual-layer Professional Discs. .....We also needed the current affordability and direct archive capability of the Professional Discs. For our uses, it is not unusual to have dozens of Professional Discs around waiting on editing, we simply couldn't do that with SxS media cost.

Not with SxS media cost - but the current usage of cheap SDHC cards via adaptors does mean that EX cameras can now begin to rival disc cameras for media costs. It wouldn't be feasible to have dozens of SxS cards hanging around, but it would be feasible to have dozens of SDHC cards.

I'm assuming all those SDHC options weren't an option when you committed to buy, but if they had been, do you think you'd have come to the same decision?

[Eugene Kosarovich]

Correct, the SDHC options weren't there yet. And it is amazing, 10:1 cheaper for 32GB and 20:1 cheaper for 16GB in SDHC versus SxS. Amazing.

So that would have made the decision harder, but for the other reasons I mentioned, I think both my friend and I would have still went with the F335.

There's also the all important durability. I trust Professional Discs completely, Alister's tests were very convincing. As for tape, once you've seen a full half second lost due to one dropout in HDV, you tend to stop trusting tape. As for hard drives, I've had too many fail on me, whether they're used a lot in a computer or barely used on a shelf for long term storage. As for solid state media, well, just cause it doesn't have moving parts, doesn't make me trust it more. With all the little connections internally in it, I'm not sure it would have stood up to Alister's tests. (Expansion and contraction of all those solder joints.) Plus, I've personally had a USB flash drive corrupt before. It was fine once I reformatted it, but that wouldn't help me if something important had been on it.

[Alister Chapman]

I've frozen SxS cards into lumps of ice without any issues.

[David Heath]

Originally Posted by Eugene Kosarovich

There's also the all important durability.

The only tale I've heard (first hand) of solid state media failing was a friend using a CF card in a DSLR when it died. The first thing he'll say is that it was a very cheap non-branded card, and since then he has always used branded cards.

Professional still photographers have been using flash cards for quite a number of years now, and I haven't heard any reliability issues from anybody. I've also heard first hand from one photographer who left a CF card in his jeans pocket when he put it in the washing machine. Not only is the card still in current use, but the data on it was fully intact afterwards.

So you seem to be able to freeze them and put them through a wash cycle.... anybody think of any other tests!?!

[Eugene Kosarovich]

OK, that's good to hear. It's not like I have any doubt that solid state is the future, it's just currently it still hasn't gotten to the price/capacity to beat Professional Disc for my needs yet.

So accepting that SxS cards are durable, what about all the SDHC cards being suggested? I mean, they can't be as good as SxS at 1/10th or 1/20th the cost in every way, right?

[David Heath]

Originally Posted by Eugene Kosarovich

So accepting that SxS cards are durable, what about all the SDHC cards being suggested? I mean, they can't be as good as SxS at 1/10th or 1/20th the cost in every way, right?

No, their maximum data transfer speeds are slower. But since they are still more than fast enough to cope with writing rates in normal usage, the only real drawback may be that you won't be able to download from them as fast. But the huge price difference means that's a price that a great many people seem happy to accept! :-) That also means more cards can be owned, and it becomes less imperitive to download just to free up more card space for further shooting.

Only other issue may be that they are more "fiddly" than true SxS cards, but there is a possibility of an ExpressCard memory with "normal" flash specs and price, but in an SxS like form factor. (See the Verbatim thread in the EX forum.)

What isn't clear is how long the data will last on the cards in normal storage. Years certainly, but the very long term may not be as good as for tape or disc. As an acquisition medium that may not matter much for most people, but it's worth bearing in mind.

[Brian Bang Jensen]

Just a comment from a one man company, who mostly does cooperate work.
I have been using the F 330 for nearly 3 years now.
In my world I won’t trade for an EX camera. One of the main things is archiving of raw material. Yes you can set up system for archiving, but as it is for now, I can just put the XDCAM disk on the shelf… The form factor is also an issue in my opinion.
The only thing, I can se, in a real world situation, is that the EX is more light sensitive.
I have given up takes because of low light situations, but if it should make a difference, the camera should be at least two stops more light sensitive.
I have been looking into the step up to a 700, but what I really would like, is a more light sensitive 3xx with a 1920x1080 sensor.

[Trell Mitchell]

what I really would like, is a more light sensitive 3xx with a 1920x1080 sensor.[/QUOTE]

I totally agree with you Brian! In addtion, I would like to see a LCD monitor similar to the EX-1, or a OLED viewfinder.