Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

how to know, 8 bit or 10 bit

Tom.Wong

Tom.Wong

Well-known member
Joined
Jul 21, 2009
Messages
1,975
Reaction score
0
Points
0
Location
New York/New Jersey
Website
www.thewongcut.com
Was wondering if anybody had a clever way of figuring out whether the signal coming out of a sdi is 10 bit or 8 bit. I tried looking up the whitepapers on the device i'm trying to capture from, but nothing is telling me whether it's a 10 bit sdi or 8 bit sdi. And it's really bugging the hell out of me not knowing. Any help would be great.
 
Tom.Wong said:
Was wondering if anybody had a clever way of figuring out whether the signal coming out of a sdi is 10 bit or 8 bit. I tried looking up the whitepapers on the device i'm trying to capture from, but nothing is telling me whether it's a 10 bit sdi or 8 bit sdi. And it's really bugging the hell out of me not knowing. Any help would be great.
Click to expand...

You can construct artifisial BW gradients that travel across the whole screen but include so few unique values in 8bit space that they will 'always band'.

EDIT I was asuming the piece of equipent could be sent a signal... didn't think it might be a camera
 
I think you'd have to hook it up to a scope and read out the code values. You could feed a proper gradient in, and check at all the code values are represented in the output.

Graeme
 
can i just print it out, out of a crappy printer and it will work all the same? or do I need some kind of over priced chart that will make me punch myself in the skull everytime i look at it?
 
yeah easiest thing would be to run it through a scope and analyze that thing.

and hahaha. if that's your philosophy, i'd call it 10 or 8 and be done.
 
Tom.Wong said:
Was wondering if anybody had a clever way of figuring out whether the signal coming out of a sdi is 10 bit or 8 bit. I tried looking up the whitepapers on the device i'm trying to capture from, but nothing is telling me whether it's a 10 bit sdi or 8 bit sdi. And it's really bugging the hell out of me not knowing. Any help would be great.
Click to expand...

Maybe you should name the device, there is an off chance someone here might know.
 
Craig W. Bickerstaff said:
Maybe you should name the device, there is an off chance someone here might know.
Click to expand...

it's from a foreign camera, so i'd rather not pollute the boards with that. just need some proper advice on how to check what the sdi signal is, bit depth wise. and it's good for future knowledge too.

so from the advice I take, i need a dsc chart, and a hardware scope to check my values.

means i gotta save the pennies, been saving for a ultrascope setup, and dsc charts aren't exactly cheap either. but it's good to know that it's actually a simple process, but not really the cheapest... if anybody has a dirt cheap easy way that would be great too...
 
Light a smooth white card at an angle to create a gradient. That should allow you to see a range of code values. Instead of a scope, why not capture it over HD-SDI and look at the code values in software, or stress the image in Photoshop to see how readily posterization occurs?

Graeme
 
Just creative problem solving!

For measuring code values, Flanders Scientific updated their monitor firmware with new measurement features at my request and that's what I've been using.

Graeme
 
Graeme Nattress said:
Light a smooth white card at an angle to create a gradient. That should allow you to see a range of code values. Instead of a scope, why not capture it over HD-SDI and look at the code values in software, or stress the image in Photoshop to see how readily posterization occurs?

Graeme
Click to expand...

Sir you truly are the best. If the signal is just 8 bit, the banding will occur no matter what bit depth i capture at correct? ( always gotta make sure ) cause i know the values stay the same but stretching it across a higher bit depth, i dont know how that effects the results
 
i would almost think that you would see hints of banding without processing if it is 8bit. or if the slightest adjustment in values bring them out. otherwise, congratulations on 10 bits!
 
Graeme Nattress said:
Just creative problem solving!

For measuring code values, Flanders Scientific updated their monitor firmware with new measurement features at my request and that's what I've been using.

Graeme
Click to expand...

be sure to let the boards know if you ever get one of those sony OLED 28 thousand dollar monsters ins :)

my inner and out geek lit up when i read about those. 178 degree viewing angle!

and just a big thanks to everyone, gonna do my tests tomorrow, hope the results are good.
 
You'd have to capture the HD-SDI 10bit to ensure that the capture process doesn't induce banding. You need a smooth card - may even de-focus slightly so no grain from the paper will dither the signal, and well lit too as it's going to be the mid-highs that show it. Another alternative is a nice blue sky, but we won't be seeing one of them here for quite a while :-)

Graeme
 
right right. but at the end of the day some point, the numbers aren't going to matter as much as how good your images are going to look. wot wot.
 
I'll pollute the boards a bit... maybe one of these perpetrators actually is your answer.

The XDCAM EX cameras (all of them from EX1 up) have a 10 bit SDI output

The AF100 has 8 bit output. Still the best way to make pictures with that camera is to record the HD-SDI- because the in camera codec is miserable.

The XL-H1 has 10 bit output ... but the signal is kind of crappy. I'd call it 9-bit.

Other than that I endorse the card method and have used it a few times myself.
 
Don't vendors of SDI capture gear provide simple bit scopes? Don't own so I don't know. IIRC such tools were quite common in the audio world when they transitioned to 24 bit material.

Pointing a camera at a real world target in order to figure out its output bit depth reminds me of those funny stories from support people who get emailed PDFed scans of screenshot printouts on a regular basis :)
 
Wolfgang Woehl said:
Don't vendors of SDI capture gear provide simple bit scopes? Don't own so I don't know. IIRC such tools were quite common in the audio world when they transitioned to 24 bit material.

Pointing a camera at a real world target in order to figure out its output bit depth reminds me of those funny stories from support people who get emailed PDFed scans of screenshot printouts on a regular basis :)
Click to expand...

No they don't.

Typically you only see that on higher end dedicated scopes.

I'd be very pleased to see another solution... it seems it would be an easy piece of software for any SDI equipped capture card to implement- or any NLE using such a board for that matter.
 
Graeme Nattress said:
Just creative problem solving!

For measuring code values, Flanders Scientific updated their monitor firmware with new measurement features at my request and that's what I've been using.

Graeme
Click to expand...

Graeme, which of their monitors have this firmware? Or is it all of them?
 
I think they all do - but I'd ask them direct. They've been very helpful with my questions and requests.

Graeme
 
You must log in or register to reply here.
Back
Top