-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Who make D3x/s's sensors ?
- Thread starter Jackie Q Acomson
- Start date
Two at a time, by hand, in my cellar.
"That's not how Active D-Lighting works. Your "Nikon source" is wrong. There's nothing at the sensor level that's changed when Active D-Lighting is turned on or off. It's essentially a custom re-linearization of the data."
My sources are the Nikon engineers who did the work on Active D-lighting . Yo uare correct about what Active D-lightign is doing but it happens on the CMOS.
My sources are the Nikon engineers who did the work on Active D-lighting . Yo uare correct about what Active D-lightign is doing but it happens on the CMOS.
so I leve the field to him to worry about these sorts of things and wil go back to making photos rather than argue with him.
Luke most of what you say, the 'name calling' seems to be a figment of your imagination.
If I can summarise the 'evidence' and 'facts' that you have presented:
1. The D700 firmware update contains copyright strings for NEC and Renesas.
Your logic draws you to the conclusion (if I remember right) that the firmware was produced by NEC and the chip produced by Renesas, and also that this is uniquely true of the D700.
Have I put forward your proposition correctly?
If so, could you please enlarge on the process of 'logic' which leads from that evidence to those conclusions, particularly since we now know that the D700 also contains the 'Nikon' copyright, like every other firmware update.
Rickard Hansson
Senior Member
Who cares as long as it delivers...
--
Rickard Hansson
Sweden
--
Rickard Hansson
Sweden
bikinchris
Forum Pro
Since all of the Sony (and many other) sensors are made with Nikon steppers, then all of those companiers who use Nikon steppers are selling Nikon sensors. And that includes every single camera maker!
--
I am an expert at contradicting myself. Just wait a while. It will be evident.
Chris, Broussard, LA
--
I am an expert at contradicting myself. Just wait a while. It will be evident.
Chris, Broussard, LA
Luke Kaven
Veteran Member
You meant "Brad"... I'm Luke
Thank you, PixelPepper and Thom for contributing. Your posts made very interesting reading.
There's interesting things in there such as finding out that Canon uses ARM and Nikon uses the SPARC. I've always wanted a SPARC workstation and now I get one in my D700. Its also interesting to know in the firmware debate that Fujitsu's RTOS is based on micro-itron so its really cool to own something thats based on tron. Does anyone know what's the OS that's running on Canon's ARM chips?
The other thing is that the Canon 5D mk II's sensor package is remarkably similar to the D3/D3s and therefore I conclude that the Nikon D3/D3s/D700 family of sensors are made in Canon fabs or use Canon IP and thats why Nikon is so secretive.. Kidding Kidding! I do not know that. It just stuck me that they were so similar.
Could you also comment on how the Canon 1D mk IV's sensor compares with the Nikon D3s from a technical point of view such as read noise and architecture and share info on Panasonic's nMos sensors if you have any. It would be really interesting reading.
There's interesting things in there such as finding out that Canon uses ARM and Nikon uses the SPARC. I've always wanted a SPARC workstation and now I get one in my D700. Its also interesting to know in the firmware debate that Fujitsu's RTOS is based on micro-itron so its really cool to own something thats based on tron. Does anyone know what's the OS that's running on Canon's ARM chips?
The other thing is that the Canon 5D mk II's sensor package is remarkably similar to the D3/D3s and therefore I conclude that the Nikon D3/D3s/D700 family of sensors are made in Canon fabs or use Canon IP and thats why Nikon is so secretive.. Kidding Kidding! I do not know that. It just stuck me that they were so similar.
Could you also comment on how the Canon 1D mk IV's sensor compares with the Nikon D3s from a technical point of view such as read noise and architecture and share info on Panasonic's nMos sensors if you have any. It would be really interesting reading.
IMO it seems clear that all aspects except the fact that the connects are on 2 sides is different. You need to know how a package like that is made and what components it is made of. Then you can understand that those are very much different.
--
Osku
Unfortunately, pixelpepper is no longer with us. I've taken up the conch, as it were.
I'm not sure this is right. Looking at the firmware copyright notices, they mention the Fujitsu FR series, which is not SPARC. None of the FR has enough throughput to do this job, but I would guess that the FR-V very long instruction word processor does (it was designed for image processing). The FR is also used in Pentax cameras, and the new Leicas have an FR/FR-V on one chip ( http://www.electropages.com/viewArticle.aspx?intArticle=11618 ). IMO, it's very likely that the chip labelled EXPEED is very similar.
DIGIC and DIGICII VxWorks, DIGIC III and IV, DryOS (see http://chdk.wikia.com/wiki/For_Developers - a host of interesting information about Canon internals, if only there was a NHDK)
The most striking thing is all the output lands being at the end of the chip. They both also lack the complex circuitry round the edges that the D3x/A900 chip has for its digital output. The choice of a 'CCD style' analog output is also similar. Note pixelpepper's hunch that both had roots in Mitsubishi design practice.
Not enough yet on the 1DIV. From Marianne Oelund's measurements, the D3s seems to be something of a breakthrough in terms of quantum efficiency, more than 30% above the D3, which was already the best in the business. It's read noise seems to have been about halved, as well, which would put it on a par with the 5DII and 50D.
SimeonPilgrim
Member
- Messages
- 38
- Reaction score
- 4
Just to be a thread vampire....
I've just loaded the D700 firmare into my dissembler, and can confirm the B firmware is a Fujitsu FR processor, just like all the other Nikon DSLR's.
For those playing at home, split the D700_0103.bin @0x80002 and load into RAM space at 0x40000
I've just loaded the D700 firmare into my dissembler, and can confirm the B firmware is a Fujitsu FR processor, just like all the other Nikon DSLR's.
For those playing at home, split the D700_0103.bin @0x80002 and load into RAM space at 0x40000
Similar threads
Keyboard shortcuts
- f
- Forum