The only thing I can personally see the 5D2 standing to loose would be the video, and that certainly wouldn't make it $600-$800 cheaper.
Not video, because that is Live View, and Live View is now a requirement (for me, too). All cameras with Live View already have video--it's a software implementation...a few people have hacked their 40D cameras and turned it on. So to actually remove video would be to remove Live View.
You do realize that it was sold at the sub $2000 price two years before the 5DII was announced.
I never saw a 5D advertised with a true price under $2000 anywhere. I'm not talking about 3-card-monte marketing gimmicks. I'm talking about a genuine price.
Neither did I. Stores here in Toronto were selling them at $2500 CAD right up to the very end IIRC. Again, it's hard to speculate if the 5D was actually profitable at $1700 unless you really know their true costs.
The full frame dream is alive for those who are willing to use the original 5D, and buy it used. It's a superb deal at only $1200. At some point in the distant future, the trickle down technology from the original 5D, 5DmkII and 1Ds series will eventually be found in cheaper cameras, but that's still a ways off.
In the meantime, I would just enjoy using the cameras available from Canon. They're all quite good tools!
Yohan Pamudji wrote: dancam wroteEverybody talking about stripping out features on the 5D2 to make a cheaper FF makes no sense! The only thing I can personally see the 5D2 standing to loose would be the video, and that certainly wouldn't make it $600-$800 cheaper. What else would anybody give up just to get FF?For real. In terms of functionality the 5DII doesn't have much to strip out or downgrade to make it a cheaper camera. Downgrade the AF? Weather sealing? Viewfinder? Framerate? None of those are particularly stellar as it is, so would such a camera sell very well? Depends on the price I guess, but I agree that there's not much room for the 5DII to move downward in features.
If another "affordable" FF were to hit the market the more reasonable direction would be up. Upgrade the features I listed above, sell it for $3000 and move the 5DII downward in price....Show more →
agreed the 5D2 just basically is the rebel of FF, the body specs are worse than xxD, nevermind 1 series
tsangc wrote: Earlier quoteYou do realize that it was sold at the sub $2000 price two years before the 5DII was announced.Earlier quoteI never saw a 5D advertised with a true price under $2000 anywhere. I'm not talking about 3-card-monte marketing gimmicks. I'm talking about a genuine price.Neither did I. Stores here in Toronto were selling them at $2500 CAD right up to the very end IIRC. Again, it's hard to speculate if the 5D was actually profitable at $1700 unless you really know their true costs.
It's not hard to find cached pages pointing out the fact that f J&R was selling 5D+24-105L kit for 2749 flat with free shipping. The really price conscious would have realized that Circuit City was selling that kit for $2700+tax to anyone with a AAA card. Those less price conscious consumers were paying 2799 instead.
I'm sure many would consider that a 3-card monty trick though, since they were sticking you with a lens that was widely panned as being crap on the 5D.
Daniel Heineck wrote:
The present 1ds3's process became mature enough (i.e. high enough yield) to handle moving the 5Dii to it as well, simplifying the manufacturing lines and reducing costs universally. The process will likely repeat itself on a new 1Ds4, which will be a low-yield FF sensor (i.e. VERY expensive at onset), whose process will mature over some lifetime (2 years?) to make it inexpensive enough to move the 5D platform over to it.
That's how this business works. There is a market driven nature, but be sure that a 24x36 chip is not going to be easy in comparison to a crop frame sensor, ever.
Well this is how this business worked anyway. That is, of course, until they ran out of runway on the upgrade cycle and DSLR sales started tanking. Look at what has changed in the DSLR market. 1) unit volume is down in a big way. 2) Micro 4/3 and other options are presenting a substantial alternative for the DSLR's normal growth market: P&S shooters looking for the next step in quality. 3) The market has exhausted it's runway in terms of upgrade cycles. The days of releasing an upgrade after 18 months and having consumers ready to buy it are over.
The DSLR market is no longer a growth area and commodity marketing strategies are going to play out. Price competition is going to become more important and top end technologies are going to reach consumer level products. What we are getting ready to watch is going to almost be an instant replay of both the P&S and flat panel tv markets in the past few years.
Will FF remain a niche, high end feature? Maybe, but it doesn't look like Sony wants to let that happen. Further, Nikon has no other business to fall back on, so they are likely to follow Sony's lead. After all, they started the race to the bottom with the D40.
nads wrote: Daniel Heineck wroteThe present 1ds3's process became mature enough (i.e. high enough yield) to handle moving the 5Dii to it as well, simplifying the manufacturing lines and reducing costs universally. The process will likely repeat itself on a new 1Ds4, which will be a low-yield FF sensor (i.e. VERY expensive at onset), whose process will mature over some lifetime (2 years?) to make it inexpensive enough to move the 5D platform over to it.
That's how this business works. There is a market driven nature, but be sure that a 24x36 chip is not going to be easy in comparison to a crop frame sensor, ever.
DanielWell this is how this business worked anyway. That is, of course, until they ran out of runway on the upgrade cycle and DSLR sales started tanking. Look at what has changed in the DSLR market. 1) unit volume is down in a big way. 2) Micro 4/3 and other options are presenting a substantial alternative for the DSLR's normal growth market: P&S shooters looking for the next step in quality. 3) The market has exhausted it's runway in terms of upgrade cycles. The days of releasing an upgrade after 18 months and having consumers ready to buy it are over.
The DSLR market is no longer a growth area and commodity marketing strategies are going to play out. Price competition is going to become more important and top end technologies are going to reach consumer level products. What we are getting ready to watch is going to almost be an instant replay of both the P&S and flat panel tv markets in the past few years.
Will FF remain a niche, high end feature? Maybe, but it doesn't look like Sony wants to let that happen. Further, Nikon has no other business to fall back on, so they are likely to follow Sony's lead. After all, they started the race to the bottom with the D40. ...Show more →
You miss one thing - there is significant differentiation in terms of technology and approaches between Nikon and Canon. The others are not players at this point anymore (in real terms).
That is a different problem - with essentially two competitors you don't get commodity pricing. If anything it goes the other way.
That won't happen though, you will see somewhat stable pricing with introductions at traditional price points (exactly what is happening now) and there will be differentiation in features because they have, and own, different technologies.
There is no way in that you will see commodity pricing on DSLRs like with P&S.
nads wrote: Daniel Heineck wroteThe present 1ds3's process became mature enough (i.e. high enough yield) to handle moving the 5Dii to it as well, simplifying the manufacturing lines and reducing costs universally. The process will likely repeat itself on a new 1Ds4, which will be a low-yield FF sensor (i.e. VERY expensive at onset), whose process will mature over some lifetime (2 years?) to make it inexpensive enough to move the 5D platform over to it.
That's how this business works. There is a market driven nature, but be sure that a 24x36 chip is not going to be easy in comparison to a crop frame sensor, ever.
DanielWell this is how this business worked anyway. That is, of course, until they ran out of runway on the upgrade cycle and DSLR sales started tanking. Look at what has changed in the DSLR market. 1) unit volume is down in a big way. 2) Micro 4/3 and other options are presenting a substantial alternative for the DSLR's normal growth market: P&S shooters looking for the next step in quality. 3) The market has exhausted it's runway in terms of upgrade cycles. The days of releasing an upgrade after 18 months and having consumers ready to buy it are over.
The DSLR market is no longer a growth area and commodity marketing strategies are going to play out. Price competition is going to become more important and top end technologies are going to reach consumer level products. What we are getting ready to watch is going to almost be an instant replay of both the P&S and flat panel tv markets in the past few years.
Will FF remain a niche, high end feature? Maybe, but it doesn't look like Sony wants to let that happen. Further, Nikon has no other business to fall back on, so they are likely to follow Sony's lead. After all, they started the race to the bottom with the D40. ...Show more →
Nads,
May I ask what field you work in? Just curious, honestly.
I'll make this very simple. The economics of a DSLR--especially a full frame--are vastly different to a PnS. Electronics/DSP-wise, most components are shared between a PnS and a DSLR, as evidenced by the Digic chips or what have you. Power supply, lcd displays, etc--all the same. You have the housing of a DSLR and the mirror/shutter mechanism which is a lot larger and thus more expensive, but nowhere near enough to differentiate between the cost of a PnS and a DSLR.
The sensor makes up a much much much larger percentage of the manufacturing cost of a DSLR than it does in a PnS. Over time, the percentage that the sensor costs versus the total is likely to only increase. Just look at the link I posted earlier for yield versus die size at a given technology node. This give reason for the lower cost of 4/3 and the obscene cost of a MFDB.
Assuming 100% yield (ha!), per unit area costs for semiconductor fabrication have not gone down over the years, just chips have been miniturized, lowering individual chip costs. One cannot miniturize a image sensor, or it is no longer a full frame, aps-c, etc.
Thus, at best, we'll see a slight decrease in chip cost in the future, assuming that yield increases. Whomever wants to put out a FF chip at a very low cost is trying to make a market share grab at a loss, or severely reducing their margins. We may see some amazing new FF sensors in the future as Canon/Sony/Nikon all push forwards in their processing technology, but I doubt we'll see a material drop in cost of sensor.
Increasing wafer size does however decrease cost, which is why we've seen the LCD market explode. Not only have their fab processes improved across technology nodes, the sheet of glass they start with has grown tremendously. I forget the specification for Gen 8, 9, and 10 lcd fabs, but have a dig and see how large of glass they use. Also look at their design rules and have a good laugh how relaxed they are.
The chip market hasn't seen that level of wafer size growth--a LOT of electronics is still made on 200mm wafers, with a greater and greater percentage of 300mm taking the market. Will we see another wafer size jump? I'm guessing not soon, but I don't have that level of industry knowledge. It's a different economic model than LCD's.
Hope that helps with a snapshot into FF versus APS-C, 4/3, 1/1.8, etc sizes.
tsangc:
Stores here in Toronto were selling them at $2500 CAD right up to the very end IIRC
Not true. Even Henry's and Blacks had it for C$2200. The lowest Canadian price (without rebates) was C$2000 (Nov 08). In January 2008, it was C$2340 with C$700 rebate)
I believe that it is very costly to make a full-frame sensor based on the existing technology, however, I also believe that with innovation cost can be driven down. If demand is high enough, new technology will appear to satisfy it.
During early-80's, personal computers were still expansive. During mid-80's, a computer that came with a 40 megabytes hard-drive was considered rather high-end for home-use. Nowadays, a digital photo can take way over 10 megabytes. With old technology, a high-end hard drive can possibly store 3 nowadays photos. What I wanted to say is that we cannot measure economic growth with old or existing technology.
Saying the cost of full-frame sensor is very difficult to driven down is like saying vehicle mpg (miles per gallon) is difficult to reduce. With old technology, mpg is of course difficult to reduce. That's why we now have hybrid technologies. In the future, our cars may be powered by hydrogen fuel-cell.
twilighter wrote:
I believe that it is very costly to make a full-frame sensor based on the existing technology, however, I also believe that with innovation cost can be driven down. If demand is high enough, new technology will appear to satisfy it.
During early-80's, personal computers were still expansive. During mid-80's, a computer that came with a 40 megabytes hard-drive was considered rather high-end for home-use. Nowadays, a digital photo can take way over 10 megabytes. With old technology, a high-end hard drive can possibly store 3 nowadays photos. What I wanted to say is that we cannot measure economic growth with old or existing technology.
Saying the cost of full-frame sensor is very difficult to driven down is like saying vehicle mpg (miles per gallon) is difficult to reduce. With old technology, mpg is of course difficult to reduce. That's why we now have hybrid technologies. In the future, our cars may be powered by hydrogen fuel-cell.
This is what people who do not understand the economics of semiconductors always say.
Read Daniel's post.
What your comment misses is that through all that innovation the cost effective maximum size of semiconductor chip to produce has not changed much (or very slowly). HOWEVER, we can do a LOT more in the same size chip.
The problem with sensors is that they are space constrained - they can't get bigger or smaller. What you are asking to do, and implying will work through "innovation" is that we will be able to project full frame performance onto a APS-C sensor size. That obviously wouldn't work because it would no longer be full frame.
Try this. You buy silicon by the acre. You have to buy the same acreage for a full frame sensor regardless of how much stuff you put in that space. The price of that same acreage hasn't changed much over time - at least no where even close to the rate of innovation and change in the IC industry. Therefore, as the cost of the acreage has always been about what it is, don't expect that to change anytime soon (it really hasn't changed much in 20 years).
Does that make more sense?
Those old style PCs were made with the largest chips of the day (and were roughly - in broad terms) the same size as they are now - they just had lots less in them. Now, that same PC has the same size chips but there is a lot less of them BUT there is a ton more in them. The same thing that made the PC cheap won't work here. The PC benefited by putting more stuff in less area. The cost is determined by the area. In cameras, the sensor size can't change so it doesn't matter how much you put in it, it still costs the same.
well Canon did in the past bring back a body you had the F-1 and in the 70s
the new F-1 in the 80s...correct me if I am wrong..but why not do the same with the original 5D but call a new one 3D...yep I said it again 3D
Not true. Even Henry's and Blacks had it for C$2200. The lowest Canadian price (without rebates) was C$2000 (Nov 08). In January 2008, it was C$2340 with C$700 rebate)
I stand corrected.
I do however think it's useless to speculate if producing a FF camera is profitable at $1700 USD or $1000 USD unless someone actually knows real cost numbers and projected/past revenue in each market segment. There's people here who seem to know a fair bit about semiconductors and they're saying the cost of the imager chip is not going down.
So while we can guess and hope all we want (I for one, would love a 5D2), the informed opinion here appears to say a sub $1k FF DSLR is impossible for the near future. But hey, I will be the first one to line up for that $1K 7D when it comes out.
I can assure you that improvement in the cost of Si manufacturing is an area of intense research. The problem though is how to make large ingots of ultra pure single crystal silicon. In this basic physics is the limitation. Conventional manufacturing techniques do still have some room for optimisation (there is a lot of wastage when the wafers are cut, for example), but a step change in costs requires a new manufacturing technology to come online. It is quite possible that we may change to a new substrate before this happens.
JohnJ80 wrote:
Try this. You buy silicon by the acre. You have to buy the same acreage for a full frame sensor regardless of how much stuff you put in that space. The price of that same acreage hasn't changed much over time - at least no where even close to the rate of innovation and change in the IC industry. Therefore, as the cost of the acreage has always been about what it is, don't expect that to change anytime soon (it really hasn't changed much in 20 years).
15Bit--I think we're going to see a lot of innovation before we see a new wafer size. The cost of moving to 300mm from 200mm was staggering for most major fabs, it's a very long-term investment to make a switch.
I agree on wafer cuts,etc eating a bunch of area, along with all the diagnostic "stuff" that is put on the wafer. There is room, but historically speaking, the cost per unit area hasn't shrunk too fast. There's no doubt that people are trying their hardest to bring aerial costs down, but the progress is slow.
Thin-films on the other hand could be pretty interesting, but you need a pretty deep well for decent photon absorption. Multi-layer MOCVD or MBE-grown superlattice converter cells with the color filter built into each pixel (via bandgap engineering) would be awesome. Or just a non-silicon foveon-based sensor technology is exciting. But those are way down the line.
It's truly fun stuff, and I'm glad I'm in this field. Challenging yet stimulating.
15Bit wrote:
I can assure you that improvement in the cost of Si manufacturing is an area of intense research. The problem though is how to make large ingots of ultra pure single crystal silicon. In this basic physics is the limitation. Conventional manufacturing techniques do still have some room for optimisation (there is a lot of wastage when the wafers are cut, for example), but a step change in costs requires a new manufacturing technology to come online. It is quite possible that we may change to a new substrate before this happens.
Absolutely. We need a waiver on the laws of physics to make this work right now. It's a fundamental challenge in the industry.
Watch out for Sony. They will lead the way to the low cost FF system.
A low cost anything would be a new first for sony. They forbid retailers to sell at a discount, and generally set their products at the top of the price range.
Still, there is always hope that they will see the light and use the tremendous manufacturing capability that they have to compete on price and quality.
all I can say is that there were plenty of threads around these parts "proving" that there could never be a sub $1000 APS DSLR camera produced... then there were threads were all the resident experts "proved" that a FF DSLR with sensor based IS could not be made... etc. etc...